a.
. r.
‘93: 294:. a ..
".53 {I ":3

:-
0. “#1 .03»! 4A
.. :4
ran-"V.“fiftxvaflfluuw‘I
. l . .. . .
..,....Lv...t=
E 4.; an
.I .
fr
.1...
1.... A.
7.5.5..

,3 a...
3... I131!

1.33.;

z 53... It.

1....

It: I
.it! ..

.
2ft...“

. 2:1...

(,5 3:21Q
2...}.

 

 

jaw.
3.

.

my

.9‘$1..

‘1.

15k. n ‘3

t.

O ‘a... .
.6 9 1
L13...

.1
L .2.-
it:

i:
14.31!
, ,i‘?

..
nu

.2.
II kt...“

..
x
.

L.
i-
1%.,w

.u-

x his MUw

2.

‘nnrril-in JV

. i, ‘v g
. 3.9.LKFJ

.
art-.9 on

 

32 LIBRAIW
it) 5 } Michigan State
- University

This is to certify that the
dissertation entitled

SYNTHETIC STUDIES TOWARD THE TOTAL SYNTHESIS
OF FOSTRIECIN AND SOME ANALOGS

presented by

Glenn Walton Phillips

has been accepted towards fulfillment
of the requirements for the

Ph.D degree in Chemistry

 

 

(AW D. low/z

VV Major Professor’s Signafire’
§// 0 /0 4

Date

MSU is an Affirmative Action/Equal Opportunity Institution

 

 

- ‘—_-—-.-.--—.—-

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/07 p'lClRC/DateDue.indd-p.1

 

 

SYNTHETIC STUDIES TOWARD THE TOTAL SYNTHESIS OF FOSTRIECIN AND
SOME ANALOGS

By

Glenn Walton Phillips

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

2006

ABSTRACT

SYNTHETIC STUDIES TOWARD THE TOTAL SYNTHESIS OF
FOSTRIECIN AND SOME ANALOGS

By

Glenn Walton Phillips

The development of a novel aldol reaction between 2-aIkynaIs and methyl [(4R,
SS)-l,S-dimethyl-4-phenyl-2—imidazolidinone] methylene tetracarbonyl chromium (O)
and its enantiomer, has provided a unique approach to the total synthesis of fostriecin; an
antitumour agent. The synthetic strategy outlined for this natural product is a convergent
one and involves a lactone, diene and triol fragment. All three fragments have been
successfully prepared in high yields and a formal synthesis of fostriecin has also been

accomplished.

ACKNOWLEDGEMENTS

I would like to begin my acknowledgements by thanking God for the privilege I
was afforded to attend Michigan State University Organic Chemistry program and the
ability to complete its degree requirements. I have been blessed to have a patient,
encouraging and creative advisor who has guided me tremendously throughout my
graduate career. Thank you professor Wulff for not only being a source of inspiration, but
a tennis partner and friend. I would also like to thank my committee members, professor
Babak Borhan, professor Robert Maleczka, and professor Milton Smith who established
high standards for me to ascertain to as a chemist and have assisted me in reaching these
standards. I would also like to thank my fellow “Wulves,” particularly Manish Rawat,
Huang Jie and Dr. Su Yu for helping me through my first year at MSU. Dr. Jones and Mr.
Khun - my undergraduate professors — I thank you for your encouragement. Additionally,
I would like to thank Mapitso Molefe, Chryssolua Vassiliou, Monsteratt Rabago-Smith
and Edith Onyeozili for being great classmates and a source of encouragement.

It was quickly apparent that one needs a source of friends outside the chemistry
community to keep one in the ‘real world.’ My Bethel Seventh-day Adventist church
family has provided me with this and I thank them. Colville Heskey, Calvin Grant,
Lawrence Leathers, Tamesha Harewood, Saara Daniel, Joelle Hall, Chelauna Davidson to
mention a few have kept me sane over the weekends. I thank the Huntes, Bissons,
Richardons, Mary, Keren, Pam, Olivia, Claire (fellow AUC alumna); Stacey-Ellen, Neil
T, Justin, and Teddy for keeping a smile on my face during the latter part of my degree.

My family in Barbados - aunty Verna, grand-grand, granddaddy, uncles Oliver

and Rodney - I thank for their support (financial especially). Nanny, Aunty G, Micky and

iii

 

Tiffani and Slyvester for their occasional calls and food; uncle Randall, aunty Pauline,
Paul and Reesa for their food and hospitality during my vacations.

Finally, I especially want to thank four very important people. First my father,
Walton Phillips and my mother, Karlene Phillips for their love, encouragement, prayers
and support during my six and a half tenure here at MSU and indeed all my life. If
anyone deserves to be honored for any of my achievements, it is they. I have learned
over these few years how tremendously blessed I am to have those two as my parents.
My brother, Karl Phillips for being there whenever I needed him for anything and my
girlfriend, Alva Ferdinand for her support, typing skills, washing and cleaning skills and

TLC (not Thin Layer Chromatography).

iv

TABLE OF CONTENTS

LIST OF TABLES .......................................................................................................... x

LIST OF FIGURES ........................................................................................................ xi

CHAPTER 1

INTRODUCTION TO FOSTRIECIN (CI-920) AND

ITS SYNTHETIC APPROACHES .................................................................................. 1
The Discovery of Fostriecin ......................................................................................... l
The Biological Activity of Fostriecin ........................................................................... 2
Structural and Stereochemical Determination .............................................................. 8
Introduction to the Synthetic Approaches to Fostriecin .............................................. 11
General Approaches used to Prepare the Lactone Fragment ....................................... 12
Grubb’s Ring Closing Metathesis .............................................................................. 13
The Heteroatom Diels-Alder Reaction ....................................................................... 14
The Acid Laconization Method ................................................................................. 14
General Approaches used to Prepare the Diene Fragment As a Precursor
to the Triene Unit. ..................................................................................................... 14
Heathcock’s Method to Prepare the Diene Fragment ................................................. 17
Hatakeyama’s Approach to the Diene Fragment ........................................................ 17
Falck’s Preparation of the Diene Fragment ................................................................ 17
Just Approach to the Triene Assembly ....................................................................... 18
Trost Approach to the Triene Fragment ..................................................................... 18
Synthetic Approaches ................................................................................................ 19
Just’s Synthetic Approach ......................................................................................... 19
Bog'er’s Synthetic Approach24 ................................................................................... 22

Cossy’s Synthetic Approach33 ................................................................................... 25

Jacobsen’s Synthetic Approach25 ............................................................................... 27
Falck’s Synthetic Approach26 .................................................................................... 3O
Imanishi Synthetic Approach27 .................................................................................. 31
Kobayashi’s Synthetic Approach”54 ......................................................................... 33
Hatakeyama’s Synthetic Approach28 ......................................................................... 37
Shibasaki’s Synthetic Approach29 .............................................................................. 39
Brown’s Synthetic Approach3S .................................................................................. 43
Trost’s Synthetic Approach31 ..................................................................................... 44
Our Synthetic Approach ............................................................................................ 47

CHAPTER 2

THE SYNTHESIS OF THE LACTONE AND DIENE FRAGMENT S AND A NOVEL

ALDOL REACTION .................................................................................................... 49
The Lactone Fragment ............................................................................................... 50
Alternative Preparation of S-Glycidol ........................................................................ 52
The Diene Fragment 4a ............................................................................................. 54
The Diene Fragments 4b and 4d ................................................................................ 56
Rationale For The Preparation of Diene Fragment 4d ................................................ 59
A Novel Aldol Reaction ............................................................................................ 61
Other Factors that Influence the Asymmetric Aldol Reaction ..................................... 65
Temperature and Concentration ................................................................................. 65
Type of Base Used .................................................................................................... 66
The Effect of Additives ............................................................................................. 67

vi

Analysis of the Factors Affecting the Aldol Reaction of Carbene Complex 162a ....68

Rationale For The Reversal of Diastereoselctivity in Alkynals ................................... 69

CHAPTER 3

SYNTHESIS OF THE TRIOL FRAGMENT ................................................................ 72
First Generation Synthesis of the Trio] Fragment ....................................................... 72
Second Generation Synthesis of the Trio] Fragment ................................................... 74
Third Generation Synthesis of the Triol Fragment ..................................................... 79
Preparation of TES Protected Triol Fragment 3b ....................................................... 83
Problems Encountered in Modification - Preparation of Phosponate 3b ..................... 85
Preparation of the Methyl Ester 200 and Completion of Phosponate 3b ..................... 88
Preparation of Phosponate 3c ..................................................................................... 9O
Accidental Preparation of Phosponate 3c ................................................................... 92
Preparation of Phosponate 3d .................................................................................... 93

CHAPTER 4

ASSEMBLY OF FRAGMENT S AND THE

INHERENT PREDICAMENTS .................................................................................... 94
The Homer-Wadsworth-Emmons Reaction ............................................................... 95
The Methylation Step ................................................................................................ 98
Determining the Stereochemistry at C8 .................................................................... 106
Alkyne Deprotection and Silyl Migration ................................................................ 111
Alkyne Deprotection and Michael Addition ............................................................. 114
Model Palladium Cross Coupling ............................................................................ 115
Palladium Cross Coupling ....................................................................................... 1 l7
Attempted Alkyne Reduction and Methyl Addition of 234 ...................................... 123
Conclusion .............................................................................................................. 125

vii

IAPTER 5

[E FORMAL TOTAL SYNTHESIS OF FOSTRIECIN .......................................... 127
Preparation of Triene 239 ........................................................................................ 130
Methylation of Ketone 239 ................................................................ _ ...................... 132
Possible Causes for the Erosion of Selectivity ......................................................... 135
Acetal Removal and TBS Migration ........................................................................ 139
TBS Migration ........................................................................................................ 141
Migration Anomaly ................................................................................................. 145
Formal Synthesis ..................................................................................................... 152
Attempts to Prepare Lactone .................................................................................... 153
Conclusions ............................................................................................................. 154
[AFTER 6

LPERIMENT AL PROCEDURES ............................................................................ 156
Experimental Data for Chapter 2 ............................................................................. 156
Experimental data for Chapter 3. ............................................................................. 182
Experimental data for Chapter 4. ............................................................................. 213
Experimental data for Chapter 5. ............................................................................. 237
FERENCE .............................................................................................................. 256

viii

LIST OF TABLES

Table III-1 Oxidative Deprotection Screening Reactions ............................................... 88

LIST OF FIGURES

Figure [—1 Planar Structures of Fostriecin and Related Compounds .................................. 1

Figure [-2 Protein Phosphatasea Selectivity and Cytotoxic Activityb (ICSO, yM) of

Fostriecin Derivatives ...................................................................................................... 6
Figure [—3 Structural Determination from Spectral Data ................................................... 8
Figure 1-4 Boger’s Determination of the C9 and C 11 Stereochemistry ............................... 9
Figure I-5 Boger’s Determination of the Cg/C9 Relative Stereochemistry ....................... 10
Figure [-6 Boger’s Determination of C 11 Absolute Stereochemistry ............................... 10
Figure [-7 Identification of Diene Fragment ................................................................... 12
Figure 1-8 General Approaches to the Lactone Fragment ............................................... 13
Figure [-9 General Approaches to the Diene Fragment .................................................. 16
Figure I-lO Dephosphorylated Isomer of Fostriecin Synthesized by Just ........................ 19
Figure [-11 Just Synthesis of Fostriecin Isomer le ......................................................... 21
Figure I—12 Boger’s Retrosynthetic Analysis24 ............................................................... 23
Figure [-13 Boger’s Synthesis of the Lactone Fragment24 .............................................. 23
Figure [-14 Boger’s Synthesis of the C7—Cl8 Fragment24 ................................................ 24
Figure 1-15 Boger’s Completetion of Fostriecin24 .......................................................... 25
Figure [-16 Cossy Synthesis of the QC [2 Fragment33 ................................................... 26
Figure [-17 Jacobsen’s Retrosynthetic Analysis25 .......................................................... 27
Figure I-l8 Jacobsen Hydrolytic Kinetic Resolution of Epoxyketone 6225 ..................... 28
Figure [—19 Jacobsen Asymmetric Hetero-Diels-Alder Reaction25 ................................. 29

Figure [-20 Jacobsen Synthetic Analysis Continued25 .................................................... 29

Figure 1-21 Falck Synthetic Analysis26 .......................................................................... 31
Figure I-221manishi’s Retrosynthetic Analysis of Fostriecin27 ...................................... 32
Figure [-23 Kobayashi’s Retrosynthetic Analysis54 ........................................................ 34
Figure [-24 Optimizing Conditions for the Sharpless AD Reaction54 ............................. 35
Figure 1-25 Kobayashi’s Formal Synthesis30 .................................................................. 36
Figure 1-26 Kobayashi’s Formal Synthesis Continued3O ................................................. 37
Figure 1-27 Hatakeyama’s Retrosynthetic Analysis28 ..................................................... 38
Figure [—28 Hatakeyama’s Synthetic Approach28 ........................................................... 39
Figure I-29 Shibasaki’s Retrosynthetic Analysis29 ......................................................... 40
Figure 1-30 Shibasaki’s Synthetic Approach29 ............................................................... 42
Figure I-31 Brown’s Synthetic Approach of the C ,-C 11 Subunit35 .................................. 43
Figure [-32 Trost’s Retrosynthetic Analysis31 ................................................................ 44
Figure I-33 Trost’s Synthetic Approach of the Cr'Cri Subunit31 .................................... 46
Figure 1-34 Our Retrosynthetic Analysis of Fostriecin ................................................... 48
Figure II-l Our Retrosynthetic Analysis of Fostriecin .................................................... 50
Figure II-2 Synthesis of the Lactone Fragment .............................................................. 51
Figure II-3 Jacobsen’s HKR of Gycidyl Derivatives90 ................................................... 53
Figure [14 Preparation of TBDPS Protected R-Gycidyl Ether ....................................... 54
Figure II-5 Synthesis of the Diene Fragment 4a ............................................................. 55

xi

Figure II-6 Attempted Synthesis of the Diene Fragment 4g Via Corey-Fuchs Reaction38

...................................................................................................................................... 57
Figure II-7 Synthesis of the Diene Fragments 4b and 4d ................................................ 58
Figure II-8 Synthesis of the Diene Fragment 4c ............................................................. 59
Figure II-9 Palladium Cross-Coupling With Dienes 4a and 4b ....................................... 61
Figure II-10 Asymmetric Aldol Reactions Using a Chiral Imidazolidinone Fischer
Carbene Complex"9 ....................................................................................................... 62
Figure II-ll Asymmetric Aldol Reactions of 2-Alkynals Using a Chiral Imidazolidinone
Fischer Carbene Complex .............................................................................................. 63
Figure II—12 Hypothesized Transition State of Aldol Reaction ....................................... 64
Figure II-13 Preventing Rotation Around the C3-N2 Bond In Chiral Auxiliaries ............ 64
Figure II-14 The Effects of Temperature and Concentration on the

Asymmetric Aldol Reaction .......................................................................................... 66
Figure II-l6 The Effects of Additives on the Asymmetric Aldol Reaction ..................... 68
Figure II-l7 Hypothesized Transition States of Aldol Reactions .................................... 70

Figure II-18 Asymmetric Aldol Reactions of 2-Alkynal Cobal Complexes with a Chiral

Imidazolidinone Fischer Carbene Complex ................................................................... 71
Figure III-1 First Generation Retrosynthesis of Triol Fragment ..................................... 74
Figure III-2 Second Generation Retrosynthesis of Triol Fragment ................................. 76

Figure III-3 Addition of Dithiane, Reduction, and Acetonide

Formation-757677 ............................................................................................................ 78
Figure III-4 Protection, Oxidation and Phosphonate Addition ........................................ 79
Figure III-5 Selective Protection of DioI78’79 .................................................................. 80
Figure III-6 Preparation of Phosphonate 209 ................................................................. 81
Figure III-7 New Approach to the Phosphonate 3a ........................................................ 82

xii

Figure III-8 Improved AcyI Anion Equivalent Addition ................................................ 83

Figure III-9 Boger Non-Chelation Controlled Methylation Conditions“ ........................ 84
Figure III-10 Retrosynthetic Analysis of Triol Fragment 3b ........................................... 85
Figure III-11 Improved Selective Protection of Diol Fragment ...................................... 87
Figure III-12 Oxidative Deprotection and Triol Fragment Completion ........................... 89
Figure III-13 Oxidative Deprotection and Triol Fragment 3c Completion ...................... 92
Figure III-14 Preparation of Triol Fragment 3d .............................................................. 93
Figure IV~1 Our Retrosynthetic Analysis of Fostriecin .................................................. 95
Figure IV-2 Attempts at Homer-Wadsworth-Emmons Coupling .................................... 97
Figure IV-3 Homer-Wadsworth-Emmons Coupling of Triol Fragments 3 ..................... 98
Figure IV-4 Strategies Used for Constructing the C8 Chiral Center .............................. 101
Figure IV-5 Attempts at Methylation ........................................................................... 102
Figure IV-6 Model Methylation Reactions ................................................................... 104
Figure IV -7 Methylation of Ketone 211 with AlMe3 .................................................... 105
Figure IV-8 Predicted Model for C8 Methylation ......................................................... 107
Figure IV-9 Methyl Addition to Ketones 211, 212 and 213 .......................................... 110
Figure IV-lO TMS Removal and Silyl Migration of Alkyne 221 .................................. 111
Figure IV-ll Comparing H9 and H” of Alcohol 226 to Alcohol 244 ........................... 113
Figure IV-12 Comparing H9 and H“ of Alcohol 222 to Alcohol 223/224 .................... 114
Figure IV-13 TMS Removal from ketone 212 ............................................................. 115
Figure IV-14 Model Study for Diene Triol Coupling ................................................... 117
Figure IV-15 Plan for Completion of the Total Synthesis of Fostriecin ........................ 118

xiii

Figure IV-l6 Attempts at Diene Triol Coupling ........................................................... 120

Figure IV-17 An Attempt at Alkyne Reduction of Iodide 229 ...................................... 121
Figure IV-18 Attempts at the Palladium Cross Coupling ............................................. 123
Figure IV-19 Attempts at Methyl Addition and Reduction of 234 ................................ 125
Figure V-l The New Retrosynthetic Analysis of Fostriecin ......................................... 128
Figure V-2 Retrosynthetic Analysis of Triol Fragment 3e ............................................ 129
Figure V-3 Projected Formal Synthesis of Fostriecin ................................................... 130
Figure V—4 Synthesis of Compound 239 ...................................................................... 132
Figure V-5 Structures of Ketones 234 and 239 Compared ........................................... 133
Figure V-6 Diastereoselectivities of Methyl Addition to Ketones 211, 212,213 and 239

.................................................................................................................................... 134
Figure V-7 Predicted model for C8 Methylation ........................................................... 138
Figure V-8 Acetal Removal With HC1 and Ang03 ..................................................... 140
Figure V—9 Acetal Removal With PCC ........................................................................ 140
Figure V-10 Selective TBS Protection ......................................................................... 141
Figure V-ll Migration of TBS Followed by Lactone Preparation ................................ 143
Figure V-12 Screening Conditions for TBS Migration on Alcohol 240 ........................ 145
Figure V—l3 1H NMR of TBS Migrated Products 224 and 242 ..................................... 147
Figure V-l4 Comparing Secondary Alcohols 224, 247 and 242 ................................... 149
Figure V-15 Partial HMBC and HMQC Analysis of Alkyne 247 ................................. 152
Figure V-l6 Formal Synthesis of Fostriecin ................................................................ 153
Figure V-17 Fostriecin Analogs ................................................................................... 155

xiv

CHAPTER 1

INTRODUCTION TO F OSTRIECIN (CI-920) AND

ITS SYNTHETIC APPROACHES

The Discovery of Fostriecin

In 1984 several articles were published describing CI-920 as a structurally novel
antitumor compound, that was first isolated from a fermentation broth of ATCC 31906
fostreus subspecies of bacteria streptomyces pulveraceus.l Initial screenings of the
fermentation beer isolates showed strong in vitro activity against murine leukemia with
ID50 versus L1210 cells of 0.073 pg/ mL. This high level of antitumor activity incited a
more detailed investigation of this extract. Upon careful characterization three

compounds, fostriecin (CI-920), and two others numbered PD 113,270 and PD 113,271
were found (Figure H).2 The maximum yield of fostriecin that could be obtained per mL

of fermentation beer was 400 p g.

Figure [-1 Planar Structures of Fostriecin and Related Compounds

1-Fostriecin R = H, R' = OH

1a-PD 113. 270 R = H, R' = H

 

1b-PD 113.271 R = OH. R' = OH

1,18,1b

The Biological Activity of Fostriecin

The explanation for the current synthetic interest of fostriecin lies in its biological
activity. It displays in vitro activity against a plethora of tumor cell lines including lung,

breast, and ovarian cancer and displays effacious in vivo activity against lymphoid
leukernias.3’4 This novel phosphate ester has also been investigated in a phase one clinical
trial at the National Cancer Institute, but was halted due to concerns about stability and
purity.5

In 1988, fostriecin was found to inhibit in vitro purified samples of topoisomerase

II (ICSO = 40 pM). Based on this observation it was immediately hypothesized that
fostriecin had a mode of action analogous to that of etoposide,6 doxorubicin7 and

amsacrine,8 leading topoisomerase II inhibitors at the time of fostriecin’s discovery.

Classical t0poisomerase II inhibitors induce irreversible DNA strand cleavage by

stabilizing the interaction between topoisomerase II and double-stranded DNA,
inadvertently trapping the enzyme-DNA complex.9 Etoposide and 4’-(9-

acidinylamino)methanesulfon-m-anisidide (m-AMSA) are examples of this type of topo
II isomerase inhibitor. The other type of topo II isomerase inhibitors prevent the enzyme
from binding to DNA or block additional steps in the enzymes catalytic cycle. Amsacrine
and surarnin are examples of this type of inhibitor. The mechanism of such inhibitors has
not been established as well as the classical topo II isomerase inhibitors. The cytotoxic
effect brought about by these inhibitors is as a result of a protein-associated DNA strand
cleavage. The activity of fostriecin is weak by comparison to these other topoisomerases,

which is inconsistent with the mechanism proposed, since such high levels of antitumor

activity were recorded initially. Further evidence that this hypothesis was incorrect was

provided by Fostrina’s group in 1992, when they discovered that fostriecin does not
inhibit topoisomerase II in mamalian cellular extracts.lo
This anomaly is remedied by another one of fostriecin’s biological characteristics,

its ability to inhibit protein phosphatases 1, 2A, and 4 (IC50= 45 pM, 1.5 nM and 3.0 nM,

11.12.13.14.15.

respectively). 16‘” With respect to this property, fostriecin has the highest

selectivity for inhibition of protein phosphatase 1 (PP1) known to date. Compounds
possessing this characteristic have the ability to block the mitotic entry check point
preceding mitosis.l3 This phenomenon is also known as G2 arrest, and is the point in cell
division where damaged DNA is replaced or its synthesis is completed on entering
mitosis.18 The 62 arrest hypothesis is based on the observation that fostriecin exerts its
cytotoxic effects at low concentrations (0.5-0.15 nM) in Chinese hamster ovary (CHO)
cells. At this level PP2A is completely inhibited but not PP]. The existence of PPl

indicates that there is cell damage and the cell cycle will not proceed to the M phase.‘2

Another school of thought suggests that fostriecin induces cells to enter mitosis
prematurely, the opposite of G2 arrest. Characteristics of cells that have entered into
prophase are chromosome condensation, separation of spindle poles and formation of
asters. Entry into this phase is regulated by the maturation promoting factor (MPF)

4CdC2

complex which consists of cyclin B and Ser/T hr kinase p3 . When this complex is

activated it is thought to stimulate normal chromosome condensation. When 375 pM of

fostriecin was administered to baby hamster ovary (BHK) cells in the G2 phase,

premature mitosis resulted. This was confirmed by the presence of condensed
chromatids, separation of spindle poles and aster formation in the cells examined.13

The method of transport into tumor cells is via a reduced folate carrier system,
which also serves to enhance its selective antitumor properties. In addition, recently it has
been found that this unique property as a potent and selective inhibitor of protein

phosphatase 2A (PP2A) was showntto limit myocardial infarct size and protect

cardiomycytes during ischemia.19

Most recently, a structure activity relationship (SAR) study revealed that the

protein phosphatase selectivity is probably due to fostriecin’s outfit-unsaturated lactone.20

These results were obtained when fostriecin’s structure was compared to the
pharmacophore for nonselective PPl inhibition and its binding was modeled to PP2A
utilizing a homology model derived from PPl X—ray structures. The comparative model
revealed that the pharmacophore present in fostriecin includes: (1) a phosphate that binds
the metal ions in the active site; (2) a methyl group in close proximity to the phosphonate
acid proposed to mimic the substrate phosphothreonine methyl group and (3) an extended
hydrophobic segment thought to mimic the substrate hydrophobic residues. The feature
that did not correspond to the phamacophore was fostriecin’s unsaturated lactone. To test
this hypothesis Boger and co-workers synthesized fourteen derivatives seen in Figure I-2
below and examined their protein phosphatase selective inhibition as well as their
cytotoxic activity on L1210 cells. The first five 1, 1h, 5, 6 and 7 were designed to
examine the importance of the phosphate, while the latter ten compounds 8-17 were
designed to test the unsaturated lactone. As may be observed from the table in Figure [-2,

the presence of the unsaturated lactone is responsible for an approximate ZOO-fold

increase in PP2A inhibition. From the model it was suggested that the serine at residue
269 (C2698) is the nucleophile that assist in the active Site binding via a Michael addition
to the lactone. Supporting this hypothesis was that when the serine was replaced by a
phenylalanine the resulting mutant was much less active to fostriecin. Other results
obtained confirmed that the presence of the phosphate was even more crucial to the

phosphatase inhibition than the a,B-unsaturated lactone, as dephosphorylated fostriecin

resulted in a 105-fold loss in PP2A inhibition.

Figure [-2 Protein Phosphatase‘ Selectivity and Cytotoxic Activityb (ICso, 11M) of

Fostriecin Derivatives

 
  
 
 
  

1, R ' PO3H2
1h, R 3 H
5. R ' P(O)(OMe)OH

6,R=H
7,R-Ac

   

 

 
    
 
 
 
  
 

9,RIH
10,R=SEt

R‘"

0P03H2

11,R=H,R1=H
12,R=SEt,R1=H
13,R=0Me. R1=H
14,R=H,R1=Et
15,R=OMe.R1=OMe

Me "’OH

OH

 

 

 

 

Compound PPZA PPl PPS L1210 L12 10/C1-920
1 0.001 (: 0.0007) 50(i 10)c 70(i 33) 0.3 35
lb 350 (t 100) >100 >100 20 35
5 2.9(i1.5) > 100° >100 15 > 50
6 3.2 (i1.1) >100 >100 15 35
7 >100 >100 >100 >100 >100
8 73 (i 9) >100 >100 >25 >25
9 0.21 (i 0.05) > 100° > 100 > 50 > 50
10 0.5 (t 0.4) > 100° > 100 3 > 25
11 >50 >100 >100 >50 >50
12 8.8 (i 2.5) > 100° > 100 > 25 > 25
13 >100 >100 >100 > 25 > 25
14 1.7 (i 0.2) >100 >100 >25 >25
15 2.0 (i 2.8) > 100 > 100 > 25 > 25
16 2.1 (3: 0.6) 2 100°.d 140 (: 50)d >100 > 100
17 0.19 (i 0.02) 2100d 2 100d 40 60

 

‘ Assays were conducted with native PPZA (rabbit muscle), thPlor and thPS catalytic subunits as

detailed.b L1210/CI-920 is a cell line resistant to l by virtue of an impaired folate transporter
required to import 1. ° Also assayed with native PP] (rabbit muscle) with identical results.
dEnzyme inhibition at 100 uM = 40—50%.

 

Structural and Stereochemical Determination

Although the 2—dimensional structure of fostriecin was first published in 1983, it

would be fourteen years before the absolute configuration of all four stereocenters would

be known. In 1985 Hokanson and French determined several stereochemical assignments

of the molecule via proton and carbon-l3 experiments, particularly the lactone and triene

functionalities (Figure Ie3).21 A periodate cleavage was used to separate the lactone

moiety from the rest of the molecule following the removal of the C9 phosphate

monoester via an alkaline phosphatase. The C5 stereocenter was determined to be R by an

independent synthesis of the lactone fragment by comparing its optical rotation to that of

the lactone derived from the natural product.

Figure [-3 Structural Determination from Spectral Data

 

IR, Proton NMR —' l

"‘o

 

 

I 8:9

 

Me
1

Vinylic signals (Proton NMR Decoupling)

OH

/

 

11—

 

— OH

 

 

A“ 0PO3HN3

°~\

Alkaline phosphatase. Periodate

In 1997 Boger’s group completed the absolute stereochemical assignment of

fostriecin, reaffirming Hokanson and French’s partial analysis and assigning the C8, C9

and CH stereocenters.i Extensive NMR, experiments and chemical degradation were the

techniques they used to solve the absolute Stereochemistry.

Figure [-4 Boger’s Determination of the C, and Cu Stereochemistry

 

Hm
H10A : 0 ><c113 61.39
H1013 0 CH3 51.35
”U
£19190 .Lttlzl
9 10.5.1.9
10A 10.5.3.7
103 9.6.19
11 9.6.3.7

The relative stereochemistry of C9 and C11 was determined to be trans, by

preparing the acetonide derived dephosphorylated fostriecin. Proton, carbon-l3 and 2D

proton-proton NOESY NMR experiments all confirmed a twist-boat conformation

characteristic of the 1,3—arlti diol acetonides (Figure 1-4).23
The relative stereochemistry of C8 and C9 was determined by converting fostriecin

to a five-membered cyclic phosphate diester. 31P NMR and 2D proton-proton ROESY

NMR confirmed a 1,2-syn relationship (Figure I-5).

Figure [-5 Boger’s Determination of the C8/C, Relative Stereochemistry

 

NOE

The absolute stereochemistry of the C 11 stereocenter was used to confirm chirality

at C8 and C9. Benzyl protected 1,2,4-butanetriol chemically derived from the

dephosphorylated natural product was matched by chiral HPLC to a synthetic sample,

prepared from commercially available R-1,2,4-butanetriol (Figure 1-6). This confirmed
the C11 chiral center to be R and fostriecin’s complete stereochemical assignment to be

5R, 8R, 9R, 11R.

Figure [-6 Boger’s Determination of C11 Absolute Stereochemistry

1) TBDPSCI. imid., DMF
2) N8I04, CHgoH. H20; NaBH4
3) BzCl, Eth, DMAP,CH2012

 

'

 

 

oe
08’ z 1)O3.CH3OH;NaBH4 03’ 08‘ oe
__ — _ E Z

11 OTBDPS 2) 3ch Et3N, DMAP 11

Assigned R by chiral HPLC matching to
material derived from oommerically
available R-1,2,4-butanetriol.

10

Introduction to the Synthetic Approaches to Fostriecin

With the knowledge of fostriecin’s biological activity in mind, a profusion of
syntheses, formal syntheses and partial syntheses have been reported to date with the vast

majority being published in the last four years. Thus far, there have been five total

24.25.25 27.28 29.303 1

syntheses; three formal syntheses; the synthesis of a dephosphorylated

isomer of the natural product,32 and two partial synthetic analyses — one of a C8 epimer —

reported.”34 Both classical and modern organic chemistry have been explored to a large

extent. Some key reactions employed are the Wittig and Horner-Wadsworth-Emmons

36.94

(HWE)74'93 olefination, Sharpless asymmetric dihydroxylations, Felkin and non-

Felkin additions, an asymmetric Diels-Alder reaction,” asymmetric hydrogenationf’9

diimide reductions,29 Sonogashira,55 Stille,40 and Suzuki couplings,”52 Grubb’s ring-

closing metathesis (RCM),45 Swem, Dess-Martin and N-morphorline oxide-tetrapropyl

ammonium peruthanate (N MO-TPAP) oxidations. In this chapter we shall explore briefly
some general methods used to prepare the lactone and triene moieties (Figure [-7) and
then examine these eleven approaches in a chronological fashion. This chapter will

culminate with a brief look at our retrosynthetic analysis.

11

Figure [-7 Identification of Diene Fragment

 

OP
or3 / /
R /
\ J
Y
Triene Unlt

 

0 H

OH Y / OP
/0 RV + M
X

Lactone Fragment 0'1"“ Fragment

General Approaches used to Prepare the Lactone Fragment

Even though there have been eleven publications involving synthetic routes to
fostriecin, only three different methods have been used to prepare the lactone fragment:
Grubb’s ring-closing metathesis, a heteroatom Diels-Alder approach and acid mediated

lactonization.

12

Figure I-8 General Approaches to the Lactone Fragment

Grubb's Ring Closing Metathesis
Cossy. Falck, Hatakeyama. Trost. Kobayashi, Shibasaki, Brown

 

 

 

 

OBn '/ Lactone fragment \ O
/ 0 . OH
\ H %

TIPS R
Heteroatom Dials-Alder Lactonization
Jacobsen Boger. Just, Imanishi

Grubb’s Ring Closing Metathesis

Of the three ways explored to make this moiety, the Grubb’s RCM is by far the

most popular. Cossy,33 Falck,26 Hatakeyama,28 Trost,“ Kobayashi,30 Shibasaki,29 and

Brown34 all use this technique. As can be seen in Figure I-16, the protection of a

secondary homoallyic alcohol with acroyloyl group provided the RCM precursor 55 in
high yield and the subsequent metathesis also results in high yields of lactone 56. This
method is most attractive because it avoids having to protect the lactone as an acetal,

which is prevalent in all the other approaches.

13

The Heteroatom Diels-Alder Reaction

The heteroatom Diels-Alder reaction (Figure [-8) has only been exploited on one

occasion out of the eleven approaches mentioned to construct this fragment. In
Jacobsen’s synthesis25 of fostriecin a chromium based asymmetric catalyst was used to

give a greater than a 99%ee and 65% yield of the benzyl acetal precursor after
crystallization. The reaction shown in Figure [-19 gives the initial heteroatom Diels-Alder
adduct with 89%ee in 90% yield and a 95:5 disatereomeric ratio. Subsequent removal of
the TIPS group, epimerization with toluene-sulfonic acid and recrystallization gave 64

with very high enantiomeric purity.

The Acid Laconization Method

The remaining three publications used acid lactonization to construct the
6—1actone ring. As was mentioned earlier, the protection of the lactone as an acetal was
essential to prevent decomposition or low yields. Boger22 and Just32 reported obtaining
poor yields when attempting to do a Wittig reaction on the lactone aldehyde 2 seen in
Figure [-8. The other key feature seen in this method is the introduction of the double
bond using selenium chemistry. Our approach adopts the acid lactonization technique but
differs at this point, having the double bond already intact prior to cyclization. Chapter 2

will discuss in detail the preparation of this moiety.

General Approaches used to Prepare the Diene Fragment As a Precursor to the

Triene Unit.

Even though there have been many approaches to fostriecin, at many of the

various pivotal points in these syntheses, synthetic strategies have overlapped. Preparing

14

the triene (Figure I-7) unit of fostriecin is an archetypal example of this overlap. With

eleven synthetic approaches published, only Sonogashira,55 Stille,40 Suzuki-Miyauram'52

and a Hiyama type coupling31 are employed to construct the triene unit. In this section we

will examine how the diene precursors necessary for these couplings were prepared.

15

Figure [-9 General Approaches to the Diene Fragment

Jacobsen, Imanishi, Shibasaki

1) TBDPSCI, imid

 

 

 

SnBu3
2) KHMDS. Bu3SnC| 1) CpZZrHCI W
A /
HOW - BPSO/V\ BPSO / /
l8 \ 19 SnBu 4"
3 48% (to Alcohol)
Heathcock did the
deprotection in situ
Hatakeyama
iigfiigiort S"B"3 iigreeflcnco Et SnBu3
3 "' 2
20 H0 21 3) Dibal-H. TBDPSCI 4d

67% overall

F... W

 

 

O
1) TBDPSCI. imid / 1) [Rh(cod)C|12 ‘3’0
HO/\//\\ 4' apso \ 4
\ \ Cy3P,E13N BPSO / /
18 22 H Pinacolborane 4e
63% overall
Trost
1) Nat, AcOH 1) Dibal-H. CHzClz

 

 

 

E1020 : EtOZC’ "‘ ‘l # l/a=\/0H
o o
23 2A 2) Me0\ iii 41’
Meo’ \/U\o/ 33% overall
3) DibaI-H. CHZCIZ

JUSI o

Meow Br

1) TBS, imid / G
reso/\/\ =
\\ H (PPh3)4Pd,Cu.

t-butylamine

 

 

1)NaBH4.(NiOAc)2 R —— _ _ ores
ores 4.

 

27

 

16

Heathcock’s Method to Prepare the Diene Fragment

Jacobsen,25 Imanishi,27 and Shibasaki29 all adopted a method published by
Heathcock for the synthesis of diene 4d from 2—pentene-4—yn-1-ol (Figure 19) which was
used in the synthesis of myaxlamide Am Heathcock actually employed TBS protection
of the primary alcohol was employed and the alkynal stannane was prepared with

KHMDS and Bu3SnCl. Jacobsen,25 Imanishi,27 and Shibasaki,29 however, used a TBDPS

group for the initial protection step (see Figure [-9) The proceeding steps after this
protection were, however, identical. A hydrozirconation reduction gave the
corresponding diene. Heathcock’s final product was the deprotected alcohol in 48%

yield. No yield was given by these authors for the TBDPS protected precursor.

Hatakeyama’s Approach to the Diene Fragment
Hatakeyama, also prepared diene fragment 4d using Heathcock’s procedure but in
addition to that approach he presented an alternative strategy to prepare diene 4d.28

Starting with 2-propyn-1-ol, a LiAlH4 reduction followed by addition of tributyltin
triflate gave the vinyl tin reagent. A Swem oxidation to the corresponding aldehyde
followed by a Wittig reaction to a phosphate ester completed the carbon chain with
correct stereoselectivity. A DIBAL reduction and TBDPSCI protection gave the desired

diene fragment 4d in 67% yield overall.

Falck’s Preparation of the Diene Fragment

Falck’s method of choice to assemble the triene moiety was via a Suzuki-Miyaura

coupling of the boronic ester “.2652 The starting material used and the protection Step

17

were identical to that reported by Jacobsen,25 Imanishi27 and Shibasaki.29 The last Step

differed, however, with a rhodium-mediated trans addition of pinacolborane to the
terminal acetylene 22 to form the vinyl borane 4c. The overall yield for this approach was

63%.

Just Approach to the Triene Assembly
In Just’s approach an alkyne reduction was used to unmask the central olefin of
the triene after the coupling had taken place.32 A Sonogashira55 coupling of the TBDPS

protected 2E—penten-4—yne-l-ol to a vinyl bromide provided the triene precursor 26. The
reduction of this alkyne turned out to be one of the most challenging reactions of the

synthesis. The author resorted to a nickel boron (NiB) catalyst with one equivalent of
hydrogen, a system reported by Brown.35 To their disappointment this reduction gave

only a small amount of desired product 27 along with several over reduced products and

some un-reacted starting material.

Trost Approach to the Triene Fragment

The Trost approach was unique in the sense that the diene moiey 41' is a vinyl
halide instead of an organometallic reagent.31 All the other preparations of this fragment
involved diene units that were either a tin or a boron organometallic compound (see
Figure I-9). This diene fragment 41' was prepared from ethyl propiolate. The vinyl iodide
24 was prepared by treatment of ethyl propiolate with NaI and AcOH in 77% yield.
Reduction to the aldehyde was achieved with DIBAL-H. A HWE reaction with trimethyl
phosphonoacetate and subsequent DIBAL reduction gave the (2E, 4Z)-5-Iodopenta-2,4~

dien-l-ol 4f in 33% overall yield.

18

Synthetic Approaches

J ust’s Synthetic Approach

The first attempt at the total synthesis of fostriecin was by Just and O’Connor in
1988.32 It was attempted without knowledge of its absolute configuration, which would
only be determined nine years later by Boger and co-workers.22 Of the eight possible

diastereomers, they choose to prepare the 5R, SR, 95, 11R diastereomer (Figure HO) and
found it to be non-identical to the natural product. Their work narrowed the number of

possibilities to just seven.

Figure 1-10 Dephosphorylated Isomer of Fostriecin Synthesized by Just

0
O OH OH
I 5 _
/ 9 11 — — OH
Me 8 OH 19

Their approach to the dephosphorylated fostriecin diastereomer 1e utilized 1,2-0-
isopropylidene-D-glucofuranose as a chiral starting reagent. The C5, C9, and CH
stereogenic centers were set in place by this choice of starting material. A few

transformations led this synthetic team to a diethyl dithioacetal 28 and a very similar

dithioacetal methyl ester 31. The acetal 28 was used to make the central portion of the
molecule, setting stereocenters C9 and C”, (Figure LI 1) and the ester 31 was used to

prepare the lactone 2 (Figure l-ll) with the C5 stereocenter. In the preparation of the
lactone 2, the acid catalyzed lactonization gave low yields and the lactone aldehyde 2
74,93

proved to be very unstable on silica gel. A Homer-Wadsworth-Emmons (HWE)

reaction between 30 and 2 connected the lactone to the rest of the molecule. The triene

l9

unit was introduced in intermediate 30 by conversion of the dithioacetal 28 to a cis vinyl

bromide by mercury deprotection of the thioacetal group and a Wittig reaction with
bromomethylene triphenyl phosphorane. Sonogashira55 coupling of that bromide to a
tertiary butyl silyl (TBS) protected enynol provided 30. The last stereogenic center C8,

was constructed by asymmetric methylation of the ketone 34, which gave a 98:2 ratio of

alcohol diastereomers in favor of the correct 8R isomer.

20

Figure [-11 Just Synthesis of Fostriecin Isomer 1e

The Central portion

  
  

 

 

0 HO
SEt 4 steps M 3 ste S
o=<1Wt _ H. ; ;_., ——p—»
0 a 5 SB _____. = = —_,
OH OH
28 29
o
X 5 g Q
o o — ores
6 30a x = OCH;,
30 X = (OCH3)20PCH2
The Lactone Aldehyde
OH
SEt /
W p-TsOH H MsCl H
MeOZC a a SEt SEt Et N O 0 SB
OH 6H 0 0 3
31 32 SEt 33 SEt
8T2 / H
5120,1120 O O H Decomposes on silica gel

2 o

The Methylation Step

0

  

 

  

cu-Izm2

"vb

HOS Me

35
60%, 98 : 2 selectivity

 

A setback in the synthesis occurred at this point. When the hydrogenation of 35
using Lindlar’s catalyst was attempted Just and O’Connor obtained a mixture of

overreduced products. Having a low supply of compound 35, Just and O’Connor decided

21

to carry out this transformation at an earlier stage in the synthesis. Methyl ester 303 was

available in near gram quantities, so hydrogenation was attempted on that substrate.
Brown’s NiB catalyst system with 1 equivalent of H2 provided the best results.18° The
reaction however was still not clean, several products of overreduction and some starting

material were also isolated. A yield for this step was not reported. The ensuing steps

worked smoothly to give the 5R, SR, 95, 11R diastereomer of fostriecin.

Boger’s Synthetic Approach“

Since the Boger group was the first group to tackle the stereochemical
determination22 and complete the total synthesis of natural fostriecin,24 they were also the

first to encounter many of the problems indigenous to this molecule. One key theme
which maybe seen throughout this chapter is the use of convergent syntheses instead of a

linear one, as a tool to combat the stability issues mentioned in the following chapter.

Boger’s retrosynthetic analysis shows three main fragments the C1—C6 unit leading
to the lactone moiety; the C8-C12 unit leading to the C8—C9 syn and the C9-Cll anti

arrangements in the center portion and the C ,6-C ,8 stannane used in the assembly of the
triene fragment (Figure I-12).

5-Hexenoic acid was the starting material employed to make the lactone fragment
(Figure I-l3). A Sharpless AD36'94 on the olefin constructs the C5 chiral center in diol 42
with 92 %ee and 98 %ee after crystallization. After an acid catalyzed lactonization, the
internal olefin was introduced using selenium chemistry. The aldehyde lactone as
observed by Just and O’Connor’s is very unstable. Boger solved this problem by

converting it to its isopropyl lactol (Figure I-l3).

22

Figure [-12 Boger’s Retrosynthetic Analysis24

 

n W0 Bu3$n\/\/0R
0 OR \

38 39 4O
C1-Cs C8-C12 C16’C18

Figure [-13 Boger’s Synthesis of the Lactone Fragment24

 

 

 

W protection PMBOOCWOH TFA
H006 Sharpless AD é PhSeBr
41 42 OH H202
It) DiBAl-H; PPTS; i-PrOH O
5 >
O o ."H/OR NMO'TPAP /LO 05 '."I¢O
43 2

Synthesis of the C7-C18 fragment commenced with a two-step conversion of D-

glutamic acid to an optically active lactone 44 incorporating the nascent C9 chiral center

23

(Figure l~l4). This was converted to the corresponding dihyrofuran before the C11 alcohol

was introduced by Sharpless AD. A subsequent TBS protection of C11 gave 45. Boger
then used a stepwise approach to assemble the sensitive Z,Z,E-triene. Condensation with
a Still- Gennari phosphonate gave the methyl ester 47 and installed the first Z olefin.37

Conversion of the aldehyde derived from this ester to a cis vinyl bromide was achieved

using Corey-Fuchs two-step procedure and a tributyl tin hydride palladium reduction

(Bu3SnH—Pd(PPh3)4).38‘39 The last olefin would be constructed using a Stille coupling“).

of the vinyl bromide and the vinyl stannane 4041 shown in the retrosynthetic analysis.

Figure 1-14 Boger’s Synthesis of the C7-C18 Fragment24

 

ores
fl 1)DIBAIHMCI fl FCHcofi o
' o s 3 2 \
9 O ; 9 OH + / /
PO 0 ”Shaw” AD po 0 F30H200 \/u\o
44 3) resort 4s 4s

Stilll-Gennari phosphonate

1)HWE QTESOTBS COzMe 1)DlBAL-H,Dess-Martin OTESOTBS / Br
2) T530" POW 2) Corey-Fuchs AGO /

 

 

47 3) Buaan‘(Ph3P)4Pd 4a
1)BU3Snv/\/OTBDPSM QTESOTBS / / OTBDPS
. ‘ EtO\ ’ /

2) DIBAL-H. Doss-Martin EtO’fi 9 11
3) (EtO)2POCH2Li o o 49
4) Bess-Martin

74.93

A Horner-Wadsworth-Emmons was used to couple the isopropyl lactol 2 to
the C7-C18 fragment and a methylation of the C8 ketone with a McLi/CeCl3 slurry set the

last stereocenter (Figure MS).42 The latter step only gave a 3:1 ratio of diastereomers in

favor of the needed 8R isomer, and a 20:1 ratio of 1,2 versus 1,4 products. Separation was

accomplished at a later stage in the synthesis. Boger selectively removed the triethyl silyl

24

(T ES) protecting group on C9 and installed the phosphonate first before doing a global
desilylation. PC13 followed by p-methoxybenzyl alcohol (PMBOH) and subsequent
phosphite oxidation with HzOz-HZO was used to introduce the phosphate ester at €9.43

Global desilylation was the last step (Figure I-lS).

Figure I- 15 Boger’s Completetion of Fostriecin24

QTESOTBS / / 0TBDPS

 

||
00

BO

2) MeLl, CeCI3

/
1) t-BuOk. toluene
Et0\ A
AOQH'U¢O + /P 9 11 / V
2

49

0TBDPS 1) Selective TES Deprotection
and TBS Protection
2 'd ' th I
)A92003 ox: atlon to e actone : Fostrieci n
3) PCI3; PMBOH: H202 1
4) Global desilylation

 

 

Cossy’s Synthetic Approach33

A partial synthesis of fostriecin was reported by Janine Cossy and co-workers at

the Organic Chemistry Laboratory Association in Paris.33 Despite the fact that it was just
a partial synthesis, (only the C1-C12 fragment) some interesting chemical applications
were employed. Using S-glycidol as starting material preset the C 11 stereocenter. A linear
sequence of six steps led to the preparation of the C8 and C9 stereocenters, which were

introduced by a Sharpless AD reaction.”94 This method was used to establish the C5 and

25

Cll chiral centers in Boger’s synthesis but was used here to set the two stereocenters C8

and C9 simultaneously (Figure l-16).
Another interesting application was the use of an allyltitanium complex to
construct the CS stereogenic center.44 This reaction not only accomplishes this, but leads

to the lactone in only two additional steps. Protecting the alcohol resulting from allyl

addition with acryoyl chloride, set up the two terminal olefins for a Grubbs’ metathesis
reaction,45 which proceeded with an 86% yield. This was the first example of this type of

lactonization used on route to fostriecin.

Figure [-16 Cossy Synthesis of the C1-C1; Fragment33

 

 

H 0 3.0 6 steps PMPO 9H 0 Sharpless A04
\/11\J W051
51 52

9H 0H 0 OR OR o
? 4 steps ? 9
map 9 __ PMPO
O 11 -. OEt 11 \ H
HO 8 _—’ R0 8'“
53 54

 

 

 

1) WT! or lie“
"'Ru-
Cl’ I —\
PCygph
2) acryloyl chloride (Grubb's Metathesis)

 

26

Jacobsen’s Synthetic Approach25

Shortly after Boger’s and Cossy’s publications, Jacobsen and Chavez achieved a
second total synthesis of fostriecin.25 Their approach was especially interesting because

all four stereocenters in the natural product were established differently and none

utilizing the chemical methods used by Just, Boger, or Cossy.

Figure [-17 Jacobsen’s Retrosynthetic Analysis25

Fostreicin
1

5 W
o 0“" / M \ \ 0TBDPS
4d

%
OBn o 60 61 TMS
/
i + “K
\ TIPS
58 59

The C5 stereocenter was established via an asymmetric hetero-Diels-Alder

reaction catalyzed by a chromium salen complex developed in the Jacobsen laboratory.’6

High yields, enantiomeric excess (ee’s) and diastereomeric ratios (dr) were obtained

(Figure I—19). The acetylene unit on the protected lactol after

hydrozirconation\transmetalation47 acts as a nucleophile, adding by chelation control to a

27

chiral epoxy ketone. This addition sets the C8 stereocenter with greater than 30:1

diastereoselectivity (Figure I—20). The C9 stereogenic center was also prepared in a unique

fashion. A [(salen)COJ-catalyzed hydrolytic kenetic resolution (HKR) reaction was used

to prepare enantioenriched R-epoxy ketone, this technique was also developed in
Jacobsen’s laboratory (Figure I-l8).48

The last chiral center was constructed using Noyori’s transfer hydrogenation
methodology.49 The reaction proceeded with a 25:1 diastereomeric ratio. The sensitive
triene unit was completed by a Stille50 coupling of a vinyl iodide 69 to the Z,E-stannane

4d (Figure I-17) to give the fostriecin core. The phosphonate was installed by a method

developed by Evans, which was used in Boger’s synthetic approach.

Figure I- 18 Jacobsen Hydrolytic Kinetic Resolution of Epoxyketone 15225

 

_N\ /N_
/C°\
f-BU O O t-Bu OH
O _...0 OH
W t-Bu t-Bu Y\l + \n/l\/
° HOAc (4 mol%). o2 balloon , 0 0
62 so 63

28

Figure I- 19 Jacobsen Asymmetric Hetero-Diels-Alder Reaction25

Me

I
N o
Ctfi \Cr/
03" o 0’ \Cl fj
/
‘l’ H)\ 1) 36 h, RT, _ BnO O

 

\
\
\ TIPS 2) TBAF, THF \H
58 59 3) TSOH 64
4) Recrystallization 99 %ee
65%

(After recrystallization)

Figure 1-20 Jacobsen Synthetic Analysis Continued25

1) [Cp22r(H)Cl]. MeZZn

/ _..O /
0 + Mew ~ ,0
i-PrO O 2) TES protection 7 i-PrO 0 [M

 

 

 

 

 

\ O
65 H so 66 Me OTES
Ts Me
Ph ,1)
\
r
1 DithianeAddition N
l / QPMBO P” I
2) Oxidation .. / 7 H i-Pr
. : O O "I 3' \ >
3) Cg-Protectlon Me 'OTES \ TMS
4) PPTS/Aceone/H20 57
/ PMP H
n 9 O 1) TBS-Protection
-.,’ /
o o ' “ores \\ 2) AgNOg. le
Me TMS 3) BBQ
68
4) NBSH.TEA
3—Steps . .
Fostnecm

 

29

Falck’s Synthetic Approach“

Reddy and Falck reported the third complete synthesis which had very few steps
that would render their strategy unique.26 Two of their key steps RCM45 and Sharpless

dihyroxylation,36’94 are identical to Cossy’s approach”. A third step, allylation of an

aldehyde was the same but a different catalyst was used. Reddy and Falck’s26 synthesis

began with allylation of 70. Allylation of the aldehyde 70 with (+)-

B-methoxydiisopinocamphenyl borane and allyl magnesium bromide of the aldehyde 70
sets the CH stereocenter with approximately 98 % ee (Figure I-21).5 1 Later the same

method was used to generate the C5 chiral center in intermediate 75 which occurred with

the same level of induction. This approach to setting the C5 center is closely related to
Cossy’s approach with the difference being that Cossy’s synthesis33 required the chiral

allyl titanium complex (Figure I-16). Considering this last step, it should come as no
surprise that the identical method used to form the lactone in Cossy’s synthetic efforts
was applied here, the Grubbs’ ring closing metathesis.45 The other two chiral centers

were also generated as seen before by Cossy and co-workers,33 via a Sharpless AD.36‘94

A Suzuki-Miyaura cross coupling52 was the strategy utilized by this group to

construct the Z,E,E— triene moiety, which completed the synthesis of the fostriecin core.

30

Figure 1-21 Falck Synthetic Analysis“

 

 

 

 

 

 

OBPS
_ 1) (+)IDC2BOM6 1)Oso,. NalO4
allylmagnesium bromide \ 2) EtOZCC(Me)PPh3
7° 2) BPSSiCl 51
caps OR OBPS
1)Sharpless Et02C 8 ’ 4-steps
Etozc / 11 \ ”=9 11%
M \ 2) Acetal Me 'OR
e 72 Protection 73
\i
\
one a on oeps a 1) (+)lpczBOMe 0 9R oeps
— r : / 8 ’ Br
/ a, 9 11 allylmagnesium bromide / .99 11 —
M° OR 74 2) acryloyl chloride 75 M8 CR
/ l
,B\ OBPS
O O

1) (Suzuki-Miyaura) F t“ .
t as necun

l

 

2) Phosphonate Installation

 

Imanishi Synthetic Approach27

In March 2002, sixteen days after Falck’s publication,27 the Imanishi group

published yet another total synthesis of fostriecin. Like Falck’s synthesis many steps are

reminiscent of those seen in previous syntheses (Figure 1-22). A Homer-Wadsworth-

Emmons

7493 reaction establishes the C6- C7 olefin joining the lactone to the center portion

of the molecule, and at the other end a Stille50 coupling of a cis vinyl iodide to a Z,E-
stannane. The C8 and C9 stereocenters were prepared via a Sharpless AD.36‘94 A R-

Binapthol aluminum hydride (BINAl-H) reduction53 of 78 was used to construct the C5

31

chiral center, with a 20:1 diastereoselectivity. The alcohol resulting from this

transformation would complete the acid lactonization in high yield, following the
approach used earlier by Boger. The Cll stereocenter was obtained using R-malic acid as

a starting substrate, which was not used as a starting material in any of the earlier

synthetic approaches or since.

Figure I-22 Imanishi’s Retrosynthetic Analysis of Fostriecin27

FOSTRIECIN

o l

0 OH OH
| 5 s + /Woreops
/ ' /
a 8.. 9 11 _§:/I K

Me "OH SnBu3
77 fl 4d
0 OP OP
BO 0 a ’ I
2 5 / 9 11 ‘—
78 Me OP
0 H OH OH
+ OHC ’ =
Et02C\/\/g\/PO(OMe)2 8 9 11 CHO + IHC PPh3
19 Me OH 80 81
OH
HO C
2 11 (3021"l
R-MalicAcid

32

Kobayashi’s Synthetic Approach‘w’54

Shortly after lmanishi’s synthesis was published, Kobayashi published his
retrosynthetic analysis for dephosphorylated fostriecin which is shown in Figure I-23.

The article however delineated the synthesis of the C3-C12 fragment of fostriecin as a

mixture of isomers 82.54 The C8 and C9 stereocenters were introduced by a Sharpless
asymmetric dihydroxylation (Figure I-23) and careful optimization with various dienes of

type 84 (Figure 1-24). Suzuki,30 Stille50 and Sonogashira55 coupling reactions were also
utilized in this synthesis in the construction of the C3-C12 fragment of fostriecin 82. Only
the C8 and C9 chiral centers were explicitly defined (via a Sharpless AD) the C5 and the

C 1 1 centers were present as a mixture of isomers. The author alluded to the fact that these

chiral centers could be obtained from commercially available starting materials, so an

asymmetric synthesis of fostriecin would be possible with this strategy.

33

Figure 1-23 Kobayashi’s Retrosynthetic Analysis“

R10 on3 0R4 x 023%;— —
, _ / / +
Dephosporylated Fostnecm a: R20 Me OR 5(1 0R6
3
82 Me Me 83
Sharpless AD
AD-mix-B
0R. R10 0R4
X 8
7M 4:: R20 / / A one carbon elongation
0R 5
85 5 84 0R5

R10 rho/1A
X=I, Zn. Sn
R20 Q or R20 / x

86 87

34

Figure 1-24 Optimizing Conditions for the Sharpless AD Reaction54

R10 0R3 0R4
/
R10 0R4 R20 M 0R3 88b

e

 

   

 

 

 

 

 

 

 

/ a SharplessAD 4' 0R5
R20 / AD-mix-B
6
R20 88a
0R3 0R5
ratio yield
Entry R. R2 R4 R5 (88az88b) PM
52
1 PMB EE TBS PMB 131 (83% conversion)
<42
2 TBS EE TBS PMB 131 (80% conversion)
_ <20
3 TBS — TBS PMB 1‘1 (complex mixture)
4 PMB THP MOM TBS 1:10 93
5 PMB EE - ' res 1;>17 85
6 PMB res - “ EE 1135 55

 

All reactions were carried out at room temperature for 2 days.
a - No hydroxyl group was present at that position. just a Hydrogen atom.

In September of the same year, Kobayashi and Wang published a full paper with
its contents outlining a formal synthesis of the natural product.30 The key intermediate
targeted is the vinyl iodide 69 (Figure 1—26) which was also an intermediate in
Jacobsen’s, Imanishi’s,25 Shibasaki’s,27 and Hatakeyama’s28 syntheses. Intermediate 69

(Figure I-26) was shown by others to couple to diene fragment 4f to give the fostriecin

core (Figure 1-20). Despite the extensive experimentation with Sharpless dihydroxylation

of various dienes seen in Kobayashi’s earlier work54 (Figure I-24), he resorted to a kinetic

35

resolution via Sharpless asymmetric expoxidationm‘94 to install the chiral centers. The

scheme below outlines this approach.

Figure 1-25 Kobayashi’s Formal Synthesis3o

0” Kenetic resolution 9” 5 steps

$002E1 via 4' YVCOZEt T v
89 Sharpless Asymmetic 9“

 

 

 

 

 

 

 

rac- Epoxidation 49%. 98% 9°
9PMB OPMB
? Sharpless Asymmetic i 0 4 steps
91 Epoxidation 92 ——-—-r
23 I 1
PMBO 0TBS
0 °' \
93 \
OH
OHCWSiMe NMQC'. THF W
3 A .
94 / 95 / SlMe3
Sharpless
Asymmetic
Epoxidation
OH
/ 96 S'Mea + / 97 / sue,
>99% 66 >99% 66
a) Bu38nH, LDA Sharpless
2’ 2 Epoxidation
OH OH
W E 9""
/ 99 / I / 98 SIMe3
1. Mitsinobu
c)TESCl 2. NaOH
OTES OH
/\/'\/\ Steps a,b,c W
/ / ' / i SiMe3
101 100

36

Figure 1-26 Kobayashi’s Formal Synthesis Continued"o

 

 

ores / PMBQ ores
W 1) n-BuLi, MgBr =- 1) HF
/ / I ., :

 

_ TESO / Q 2) c)

101 PMBQ 0TBS 6“ 102 /

o ‘ /\g/
\\

3) resort

1 PMS? 0735 1)Grubb's RCM I] OH 0TBS
o 0 “M 2)NlS.AgN03 o o ”'-/ T — I

103 ores 69 ""
3) 000. CH2C|2IHZO 0753

4) Diimide Reduction

Hatakeyama’s Synthetic Approach28

Other than the USA, publications on synthetic efforts on fostriecin have come

primarily from one other country, Japan. Thus far we have examined Imanishi’327 and

Kobayashi’s30'54

syntheses of fostriecin, but more recently two other syntheses surfaced
from this country. Hatakeyama’s synthesis28 was the last total synthesis of fostriecin to

date. Even more recentl, a formal total synthesis was reported by Shibasaki.29

37

Figure 1-27 Hatakeyama’s Retrosynthetic Analysis28

  
  

llR,X=I,Y=H
llR.X=H.Y=I
llS.X=l.Y=H
11S.X=H.Y=l

RC M

Asymmetric

Allylation Me OR \—/ AsymMCU'lC
Dihydroxylation

Hatakeyama’s approach is delinated in the (Figure 1-27).28 A key ynone
intermediate 105 can be manipulated to prepare fostriecin as well as a number of E, Z
isomers and C11 diastereomers. This key intermediate ynone is prepared from
dihydrofuran 106, which when subjected to Aldisson’s procedure for stannylation
(providing perfect E selectivity),98 para-methoxy-benzyl protection, iodination and Heck
coupling99 gave the E, E-diene aldehyde 108. This aldehyde was converted to a
secondary alcohol by selective nucleophilic addition of a propenyl boron reagent in 77%

ee, and formation of the lactone moiety by Grubb’s ring closing metathesis.45 This

lactone was then subjected to a Sharpless dihydroxylation,36’94 bis TES protection, a

38

selective PMB deprotection and a Dess-Martin oxidation to give the lactone aldehyde 111
(see Figure 1-28). Aldehyde 111 was converted to the key intermediate ynone 112 via the
addition of ethynyl Grignard, followed by a Dess-Martin oxidation. In order to complete
the total synthesis, the terminal alkyne was converted to a cis vinyl iodide using NaI,
AcOH in acetone in a 10:1 ratio, reduced to the secondary alcohol in 84% de and coupled

to the Z—stannane 4d prepared by Jacobsen”, Imanishi27 and Shibasaki29 (see Figure 1-9).

A few hydroxyl group protection and deprotection steps provided the natural product.

Figure 1-28 Hatakeyama’s Synthetic Approach28

 

 

 

 

 

O l) r-BuLi. (Bu3Sn)2Cu(CN)le 0PMB l) 12. CH2C12 OPMB
0 2) Mel [3°35an 2) MCHO Och
106 3) PMBCl. NaH 107 PleAch. ch03 108
Bu4NCl
ngBr l
”M - IpczBOMe g 0mg 1) Acryoyl Chloride _ fj 0PMB
30% H202. 3M NaOH HO / / 2) Grubb's RCM 0 ya / /
109 110
77% ee (from Brown's Allylation)
Enantiomers separated by HPLC
l) Sharpless Dihydroxylation b QTES O I) HCCMgBr I] QTES 0
-. 7 CeCl .THF 7
100% cc t O O "1W" _—3_. O 0 HM
2) TESO”. 2.6-lutidine ’OTES 2) Bess—Martin ’OTES
3) DDQ 111 112
4) Bess-Martin

Shibasaki’s Synthetic Approach”

A couple months after Hatakeyama’s publication,28 Shibasaki and co-workers

from the University of Tokyo-Hongo published yet another formal synthesis of

fostriecin.29 Their approach coincided with that of Jacobsen,25 Imanishi,27 Kobayashi30

and Hatakeyama28 at the cis-vinyl iodide Stille coupling precursor 69 (Figure I-29) and

39

thus constitutes a formal synthesis. The key features of this approach included a Noyori

reduction,53 a direct catalytic asymmetric Aldol reaction, a catalytic asymmetric

allylation, and a catalytic asymmetric cyanosilylation of a ketone.

Figure 1-29 Shibasaki’s Retrosynthetic Analysis29

 

Intermediate Prepared by Jacobsen. lmamanishi, Hatakeyama, Kobayashi

Noyori Reduction

Iodination
Diimide Reduction

 

/ Direct Catalytic\ o

Asymmetric

  

 

 

Aldol Reaction \\
Me ."OMOM
. Catalytic
Gala? Asymmetric
Allylation
TMSCN
H OTIPS Catalytic Asymmetric 4.
8 Cyanosilyation of Ketone
o / . :- ENC/W
116 Me OMOM 117 o

The asymmetric cyanosilylation of 117 was achieved in 85% ee using the titanium
catalyst shown in Figure I-30. The resulting (R)-ketone cyanohydrin 118 was converted

to a diol which was selectively protected with a TIPS and a MOM group, respectively.

40

Removal of the benzyl group from the primary allylic alcohol at the other end of the
molecule followed by oxidation provided the a,B—unsaturated aldehyde intermediate 116.
At this point a catalytic asymmetric allylation using 20 mol% AgF-(R)-p-tol-BINAP
complex was achieved in 80% yield with a 28:1 diastereomeric ratio. Lactonization using
the Grubb’s ring closing metathesis technique was then applied as seen in previous
syntheses. The resulting lactone 120 was easily converted to the aldehyde 121 which was
the precursor for yet another catalytic asymmetric reaction. Using 6.5 equivalents of a
TMS protected 2-but-3-ynone and an (S)-Lanthanide Lithium BINOL complex catalyzes
this enantioselective aldol reaction proceeded to give 122 in 65% yield with a 3.6:1 ratio
of diastereomers. Conversion of this mixture to the corresponding acetonide followed by
a Noyori reduction53 gave a 49% yield of pure desired propargyl alcohol in a 97:3
diastereomeric ratio. Conversion of the TMS protected acetylene to the alkynal iodide
followed by diimide reduction gives the vinyl iodide 123 which was easily converted to
the desired intermediate 69 by acetonide removal and selective TES protection. While

preparing this thesis, Shibasaki and co-workers published a total synthesis of the C8

epimer of fostriecin.29 This synthesis retained the same main features described here only
varying at the cyanosilylation of ketone 117 (Figure I-30). The variation was using a

gadolinium catalyst complex to obtain the (S)-stereoisomer at C8 of 118, instead of the

titanium catalyst seen in Figure I-30.29

41

Figure 1.30 Shibasaki’s Synthetic Approach29

 

TMSCN 2 uiv . 5 mol% cat (R) CN 7 Ste 5
BnoW ( eq ) : BUG/W __E._.

117 o 1311‘ o 118 Measo/‘ZOTMS
Ph’ ll 99
O

    
  

‘\

0,.

'P \

- /
mo \o 300

\ .
H OTIPS \/\SIOMe3
8 1)A9F(2°m°'°’°) _ OTIPS 1)Actyoylcmoride_
0 /Me ""0M0M (R)-p-tol-BINAP (20 mol %) HO / 2) Grubb's RCM
"5 28: 1 dr 119 "’OMOM

 

 

O

/ OTIPS / 0 /ll\
1)» 1) 3111:. N83, THF _ I] (6 ooh ms
0 0 o” / . O O "I a, H =
bMOM
121

 

 

12° "’OMOM 2)D°SS'M3“‘" 10 mo|%cat

 

2) Noyori Reduction

TMS 3) TBSOTt, 2.6-lutidine 97 3
4) NIS. A9N03, Acetone Me "0 (pure after column)
5) NBSH, Et3N. THF. i—PrOH 123

    

1) 1M HCI. MeOl-l
2) TBSOTf, 2.6-Iutidine _

 

3) resort. 2.6-lutidine
4) 1M HCI, THF, CH3CN

 

42

Brown’s Synthetic Approach35

Herbert C. Brown and co-workers joined the fostriecin bandwagon with their

publication in August of 2003 of the Cl—Cll subunit 130 of 8-epi-fostriecin, Shown in

Figure 1-31.35 The key step in the synthesis of 130 is a chelation controlled addition of a

Grignard to an (it-oxygenated ketone. As can be seen in Figure 1-31 below, cis-Z-butene-
1,4-diol was employed as the Starting material. After mono-protection, the resulting
alcohol was oxidized to give the trans—aldehyde 125 which upon alkoxyallylboration with
(-)-B-y-methoxyethoxymethoxyallyldiisopinocampheylborane gave the homoallylic
alcohol 126, in > 98% de and 94% ee. A Dess-Martin oxidation follwed by a methyl
Grignard addition gave the anti tertiary alcohol 127 in 90% de. A few selective
deprotection and protection steps leads to the Grubb’s ring closing metathesis45 which
gives the lactone 130. The formation of the lactone via RCM has been seen in earlier

synthetic approaches.

Figure 1-31 Brown’s Synthetic Approach of the Cl-Cll Subunit35

 

 

 

 

 

 

cnzon MOMEM _
[ 1)TBSCl,lmidazolc 0TBS l) scc-BuLi.(-)-lpc-_.BOM¢ 0TBS 9MB“
4’ O t . /
/
cnzon 2) FCC. NaOAc / ne,,oa,.1~non. n20, / i
124 125 98% dc. 94%cc 126 0”
l) Dess-Martin 0TBS OMEM l) TBSOTf. zolutioine on OMEM
2) MeMgBr. 90% dc / .- ' / 2) AcOH. THF. H20 / p I /
Ma 127OH 3) DcssMam Me 0TBS
l) Bess-Martin OH QMEM l) Acryoyl Acid
4 . / _
2) (=)-poBAll.NaOH. H302 / / DCC,DMAP
87% dc 129 M‘ 0TBS 2) Grubb's RCM

 

43

Trost’s Synthetic Approach31

The last publication investigating fostriecin as a synthetic target described the
efforts of Barry Trost and co-workers.31 They used a dinuclear asymmetric zinc complex
in an aldol reaction, a chelation controlled Grignard addition and a palladium cross-
coupling reaction between an alkenyl silane and a vinyl iodide as key reactions. The
synthesis was a formal one, with dephosphorylated fostriecin 11 being the target. The

retrosynthetic analysis is outlined in Figure I-32 below.

Figure I-32 Trost’s Retrosynthetic Analysis31

 

O 133
\O +
OEtO OEt Q
134 135 BDMS

Trost’s formal synthesis began with ynone 135, which was derived from an
addition of BDMS protected ethynyl magnesium bromide to the Weinreb’s amide of
acetic acid. Ynone 135 was subjected to the Zn-catalyzed direct aldol reaction conditions

developed in Trost’s group to give the desired adduct in 99% ee and 73% yield.

Reduction of the ketone under Noyori’s53 ruthenium-catalyzed transfer hydrogenation

followed by selective TBS protection and acetal removal gave intermediate a-hydroxy
ketone 139. After a 3,4-dimethoxybenzyl (DMB) protection of the secondary alcohol, the
vinyl magnesium species was added in a chelation-controlled fashion to give tertiary
alcohol 140 as a single diastereomer in 75% yield. Removal of the TES-group and
acryoyl chloride addition set the stage for the lactone by Grubb’s RCM45 (Figure L8).
This precursor was then subjected to a diimide reduction following a DMB deprotection.
The resulting alkenyl silane was coupled to the vinyl iodide 41‘ (Figure 1-33) in 54% yield
with simultaneous deprotection of all the silyl groups furunishing dephosphorylated

fostriecin Ii.

45

Figure 1-33 Trost’s Synthetic Approach of the C,-C,, Subunit31

 

 

OTES
1 + I BOMe
51020 —— 1) NaI’ACOH> OW' H )-p02 : H\/'\/\
2) DIBAL-H Allylmagnesium / l
3) 31:3.0512

(955 5:2) 2) TESCI. imidazole

1) 2-ethoxy-2-(ethylperoxy)propanal

O O
/U\ 1) BDMS _" MgBr, THF /U\ 3 mol % cat 136; 6 mol % EtZZn
N/ Av \ :

 

 

 

 

 

 

l \ BDMS 67%,99%ee
W" 135 Ph Ph
Ph; I ,OH no. I gPit
N OH N
NRu:-I
OH 0 OH ores
. 1) 1 mol% \ll/\/l\
\ \
Etc 05‘ \ BDMS PrOH. 88°/ 10 1 d \ BDMS
137 " rt. ° ’ 139
2) TBSCI, lm, DMF
3) CSA, Acetone, 77%
ODMB
1) DMBOCHZCI. TBAI (14 mol %) 1) ”Ft M90” MeCN- 91%
DIEA,DMF, 97% _ ores o ores 2) Acryloylchloride, DIEA, 99% =
2) MgBrz, THF, rt, then 139 / / .. ' \ 3)TES.D|EA. 011201239700
added to 20H \ BDMS 4) Grubbsl (10 mol /o), 934
0101, i-PngCl. s-BuLi, 14o

THE—78 °C. 75% >20:1 dr

1) 000, CH2Cl23H20, 940/0
2) NaHC‘Og. MeOH,138:. 72%
3) Pd2(dba)3. CHCI3 (5 mol %)

TBAF (4 eq slow addition) 41‘
BDMS THF 54%

  

0TBS

  

 

 

138a

Our Synthetic Approach

At the time our synthetic strategy was planned, only Just and O’Connor’s
synthesis of the dephosphorylated fostriecin isomer 1e had been published32 (Figure I-9).

Just’s attempt proved to be a valuable asset, and was instrumental in our development of

a feasible and practical synthetic approach. The Horner-Wadsworth-Emmons olefination
used to connect the lactone to the center portion of the molecule and the Sonogashira55

coupling used to form the triene moiety, were both tools that were adopted from Just’s
approach. Some challenges they encountered such as the unstable lactone aldehyde and a
sensitive acetylene reduction forced us to design a strategy that would avoid these
problems.

As time progressed and as more syntheses were published a few changes in our
approach were encured, but the basic strategy remained the same. The following scheme
shows our retrosynthetic approach for this molecule and involves the union of lactone 2,

phosphate ester 3, and diene 4 (Figure I-34). High E-selectivity may be achieved from the
Homer-Wadsworth-Emmons olefination74'93 between 2 and 3, while the Sonogashira55

coupling of the deprotected acetylene to the vinyl iodide should complete the fostriecin
core. A detailed examination of the synthesis of each fragment and their assembly will be

given in the following chapters.

47

Figure 1-34 Our Retrosynthetic Analysis of Fostriecin

OH

Pd-Catalyzed Coupling

Homer-Wadsworth-Emmons
Olefination \j

 

 

 

 

 

 

 

 

 

 

 

 

 

O 0 OP
MeO\ ”,OMe
P / OP
I O Q TMS M
0 OP |
2 H 0 3 4
Lactone fragment Triol fragment Diene fragment

48

CHAPTER 2

THE SYNTHESIS OF THE LACTONE AND DIENE

FRAGMENTS AND A NOVEL ALDOL REACTION

As was outlined in chapter one, our synthetic approach to fostriecin involves the
preparation of the three key intermediates, a lactone, a triol and a diene fragment. In this
chapter we will examine how the synthesis of the lactone and the trio] fragments have

been achieved, and look at a novel aldol reaction which is the key step in the triol
fragment synthesis. The lactone synthesis was first developed by Mark Parisi56 and then
modified by Su Yu.57 The synthesis of the diene fragment was developed by Mark Parisi
and the aldol reaction of imidazolidinone carbene complexes with 2-alkynals was

developed by Dr. Kenneth Wilson.58

49

Figure II-l Our Retrosynthetic Analysis of Fostriecin

OH

Pd-Catalyzed Coupling

//
/K/

    

Homer-Wadsworth-Emmons ‘
Olefination \j

 

 

 

 

 

 

 

 

 

 

 

 

'.’ OPO3HN3

Me 'OH
1

O 0 OP
II
I O MeO\P/OM6 MOP
TMS
0 OP l
2 H 0 3 4
Lactone fragment Triol fragment Diene fragment

The Lactone Fragment

The lactone fragment possesses one of the four stereocenters found in fostriecin
which would ultimately become C5. This prompted the design of a route using a chiral
starting reagent, to set that C5 stereocenter. Using commerically available S-glycidol, a
mono-protection of the primary alcohol with tertiary butyl diphenyl silyl chloride
(T BDPSCI)59 initiated the six-step sequence shown in Figure II-2. Nucleophilic ring
opening of epoxide 142 with the anion of ethyl propiolate gave alcohol 143 in 75%
yield.60 The anion of ethyl propiolate is not stable above —78 OC and this is the first time
that it has been alkylated with an epoxide. This alkynol was then reduced to the cis-
alkene 144,61 and the six-membered ring lactone formed by acid catalysis in an overall
yield of 42% for the five steps.62 The oxidation step was reserved for the next stage of the

synthesis as the aldehyde obtained from oxidation is very unstable, and must be made in

50

situ. In his 1997 paper that established the stereochemistry of the natural product, Boger

used a Swem oxidation to obtain this lactone in situ which was coupled with a stabilized
Wittig reagent.22 They only obtained a 52% yield for this transformation. Later, in his
total synthesis of fostriecin, he prepared the lactone in its isopropyl lactol form, to

counteract this low yield.63'64 This methodology was adopted and the isopropyl lactol 2a

was obtained in 74% yield in three steps from the lactone 145.

Figure lI-2 Synthesis of the Lactone Fragment

 

 

 

 

 

 

Heco Et
0 NaH,TBDPSCI o 2
W011 = Womops A, // 0TBDPS
51 85% 142 n-BuLi. BF3-0Et2 E020 143 OH
75%
0
H2, Pd/BaSO4 COzEt p-TsOH o
EtOAc. 1 3"“ Woraops reflux, 4h > I 0TBDPS
92% 144 (an 73% “5
o
Swem oxidation I O
or V H
NMO—TPAP
2 o

 

[ Lactone fragment ]

0 0k 0k

fi/ Dibal-H o TBAF 0
0TBDPS PPTs. i-PrOH I 0TBDPS NMO-TPAP I /o

145 90% 146 82% 2a
(over two steps) _
[ Modified ]

 

 

 

Lactone fragment

 

51

Alternative Preparation of S -Glycidol

As was mentioned in the previous section S-glycidol was choosen as the chiral
starting reagent. This compound could be bought from the Aldrich chemical company at
a price of $64.20 for 5 grams. Interestingly, racemic glycidol could be obtained from the
same company for $88.40 for 500 grams, a factor of about 40 times cheaper. Inspired by
this drastic difference in price, we set out to prepare S-glycidol or a derivative of S-
glycidol from its racemic mixture, instead of purchasing the pure chiral material. A
technique developed by Jacobsen, namely the hydrolytic kinetic resolution of epoxides

provided a solution for the cost efficient preparation of epoxide 142.90 Jacobsen has

shown that this method also works for glycidols and some examples from his work are
shown in Figure II-3. Very small catalyst loadings (0.5-2.0 mol %) of 1,2-
cyclohexadiamino-N,N’-bis(3,S-di-t-butylsalicyclidene) cobalt (II) (Co"-Salen) are
required to give >99% ee with a variety of substrates. The cost of this catalyst is only
$23.00 per gram from Strem Chemicals. In the only example reported by Jacobsen of a
silyl derivative of glycidol, the TBS ether gave a 48% yield and >99% ee, with a 0.5

mol% catalyst loading (see figure II-3).

52

all

Figure II-3 Jacobsen’s HKR of Gycidyl Derivatives90

0 0.5-2 mol% Catalyst ((R,R)—160) 0
OR ' [>50/OR

THF,0 Oc - r.t 0.55 equiv H20

 

Substrate Yield ee
157 R = H 19% (+oligomers) >99%
158a R =TBS 48% >99%
158b R =Bn 47% >99%
158C R =CO(CH2)ZCH3 44% >99%
HQ”H
_N\ ,N_
Co
/ \
t-Bu O O t-Bu
t-Bu t-Bu
160

Jacobsen's Catalyst

High performance liquid chromatography (HPLC) was used to monitor the
progress of the resolution, but because TBS protected glycidol is not UV active, the
product had to be derivatized by ring opening with 2-napthalenethiol prior to its
subjection to the chiral column. Using TBDPS as a protecting group would allow us to
monitor the progress of the reaction without derivatization, providing that the correct
conditions for separation could be determined. After a few days of searching, the optimal
condition that would separate the two enantiomers of TBDPS protected glycidol 158d
were found using a chiracel-OD column with pure hexanes as the eluent. As can be seen
in Figure II-4 the results were comparable to those obtained by Jacobsen for the TBS

analog 158a. The epoxide 142 could be obtained in 43% yield and greater than 99% ee.

53

In addition the catalyst could be recycled using a protocol described by Jacobsen and in a
second run the epoxide 142 was obtained in the same yield and 96% ee. Despite the lost
of half the starting glycidol 157, this method is much more cost efficient than purchasing

the chiral material, especially if a large scale synthesis of this fragment is desired.

Figure II-4 Preparation of TBDPS Protected R-Gycidyl Ether

HOH

 

 

Hokoreops
159
o l o 2 1°/ tal 1
WOH NaH,TBDPSC : WOTBDPS mo oCa yst( 60) :
157 85% 158d “*3 ”2° 0
43% ; >99%ee Woraops
142
HHQH
_N‘ ,N_
/C°\
t-Bu O O t-Bu
t-Bu t-Bu
160
Jacobsen's Catalyst
The Diene Fragment 4a

This fragment was the least difficult to prepare but as reported in Chapter one, it
is also the part of the triene unit in fostriecin that is suspected to be responsible for its
instability. A late stage coupling of the acetylene of the triol fragment 3 (Figure II-l) to
the Z,E—iododiene 4a minimizes the exposure of this sensitive portion of fostriecin to

many transformations. If this fragment was to be installed too early, these transformations

54

might produce undesired products.56 The synthesis of the diene fragment 4a is outlined in

Figure ll—S.
Figure II-S Synthesis of the Diene Fragment 4a
n-B L', THF PDC
HOAOH —L~ TBSOAOH
124 TBS-Cl (1 6(1) 147 CH2CI2
94% 75%

TESO/W0 [ICHzPPh3]I. NaHMDS A TBSOW

84%

 

9:1 Z/E
Exam Fragment 4a

 

 

Our synthesis of 4a commences with the tertiary butyl silyl (TBS)

monoprotection of cis-2-butene-1,4-diol using Marshall’s protocol.65 The unprotected

alcohol group in 147 was then oxidized with pyridinium dichromate (PDC) to form the

a,B-unsaturated aldehyde 148 with complete isomerization of the double bond to the
desired trans stereochemistry.89 The final step was achieved using Stork’s procedure for
the synthesis of cis iodo-alkenes.66 A 9:1 ratio of E22 isomers was obtained and these

isomers of 4a were easy to separate. The overall yield for these three steps for the
mixture of isomers was 59%. It is important to note that compound 4a was prepared
immediately before use. Vinyl iodide 43 is light sensitive and cannot be stored for any

period, otherwise decomposition of the products results.

55

The Diene Fragments 4b and 4d

The synthetic route used to prepare diene fragment 4a was a short one, but it
involves at least two major problems: the E/Z selectivity of its formation and its
instability to light. Only a 9:1 ratio of cis : trans isomers of 4a was obtained, which
means that not only is 10% of the material not used, a separation is required. Compound
4a’s sensitivity only complicated matters because it had to be prepared, purified and used
while being meticulously protected from light. A more feasible fragment that should be
less prone to this stability issue would be the equivalent vinyl bromide. It has been well

established that vinyl bromides are more stable alternatives to vinyl iodides when being
handled in the laboratory.67 The vinyl bromide most likely could be stored and would not

have to be used as soon as it was prepared.
The selectivity issue on the other hand could only be addressed if a different

chemical protocol was employed since Stork reported that inferior selectivities are

obtained when Ph3P=CHBr was used instead of Ph3P=CHL66 Using Corey-Fuchs38

procedure on aldehyde 148 followed by selective reduction68 of the vinyl dibromide may

give higher selectivity for vinyl bromide 4g. However when Xuejun Lui applied this
procedure to aldehyde 148 an undesired product was obtained which was devoid of the
TBS group. The product was tentatively assigned as tribromide 161 based on the proton

NMR spectrum (see Figure II-6).

56

Figure II-6 Attempted Synthesis of the Diene Fragment 4g Via Corey-Fuchs

 

Reaction38
Br Br
/ O CBr4. PPha W Zn-Cu, AcOH W
W ------------- 9- --------------- 0-
TBSO THE 0 0c TBSO Br THF/MeOH T880
148 148a 49
_ n—BuLi. THF _ PDC
HOAOH ————' TBSOAOH ————>
124 TBS-Cl (1 eq) “7 0112012
94% 75%
B
/ o CBr,, PPh3 W
W > /
TBSO THE 0 00 Br Br
148 161

55%

Failure to convert 148 to 148a was successfully combated by changing the TBS
protecting group to a TBDPS group in the first step of the synthesis. As outlined in
Figure II-7 the sequence of reactions proceeded smoothly to give 84% yield of diene
fragment 4b as a single Z-isomer by NMR analysis. Vinyl stannane 4d was also prepared

to provide an alternative to coupling 3 to the diene fragment.

57

Figure II-7 Synthesis of the Diene Fragments 4b and 4d

 

 

 

 

 

 

 

n—BuLi, THF _ PCC, NaOAc
HOAOH = TBDPSOAOH :
124 TBDPS-0K1 69) 149 0112012
94% 100%
C Br
0 Br,, PPh3 W ZnoCu, AcOH
TBDPSOW r TBDPSO / / Br - ~0R-
THF 00c THF/MGOH
150 ' 151
91% 98%
BI’ SUBU3
Bugan, Pd(PPh3)4 W Bu SnCl, t-BuLi W
A, TBDPSO / / 3 = TBDPSO / /
Benzene 4b Ether, -78 0c 4d
95% 63% Diene Fragment
53% Overall

 

 

Another minor change that improved the overall yield of fragment 4b was the use
of a combination of pyridinium chlorochromate (FCC) and sodium acetate (NaOAc),
instead of pyridinium dichromate PDC as the oxidant for the second step. This change
greatly simplified the purification process since filtration over a plug of silica gel gave
the product 150 that was pure enough to be used for the next step and pure enough to be

completely characterized.
Extending the Corey-Fuchs38 reduction protocol to vinyl iodide 4c via diiodide

157 was troublesome. Preparation of the vinyl diiodide 157 using Corey-Fuchs protocol
was unreliable, with the optimal yield being 36%. In addition, neither of the two methods

successfully employed to do the selective reduction on the dibromide (Figure Il-7)

worked on the vinyl diiodie 157. Both the tri-butyl tin hydride39 and the Zn-Cu68 couple

58

reduction methods gave decomposed products. Diene fragment 4c could however be

obtained in a 72% yield with a 5:1 ratio of cis : trans isomers using Stork’s protocol.66

Figure II-8 Synthesis of the Diene Fragment 4c

 

 

 

n-BuLi, THF _ PDC
HOAOH - TBDPSOAOH
94% 60%
c 1
I , PPh Zn-Cu AcOH
O ‘ 3 /\/\/k '
TBDPSOW r TBDPSO / / l T '15 """" 7 'OR'
150 THF, 0 00 157 H lMeOH
36%

 

Bu3$nH,Pd(PPh3) W Diene Fragment4c
------------------- 4- » TBDPSO / /

Benzene

 

 

 

Stork's Protocol
I
o [ICHzPPh3]I.NaHMDS W
TBDPSOW o ‘ TBDPSO / /
150 HMPA,THF,-78 0 4c
72% 5:1Z/E
[Diene Fragment 4c]

 

Rationale For The Preparation of Diene Fragment 4d

At an earlier stage in the development of our strategy to fostriecin, some model
reactions were carried out with vinyl iodide 43 and 3-butyn-2-ol (and its TBS derivative)
to access whether a I’d-cross coupling reaction with a propargyl alcohol was feasible.

And to determine if the alkyne in a trans, cis-dienyne of the type 153 or 155 could be

59

0a

.5
~

f0!

selectively reduced to the cis-alkene without any over reduction of the diene unit. Both
substrates gave positive results as can be seen in Figure Il-9.

Unfortunately, later in the synthetic scheme when diene fragment 4b (Figure Il-7)
was coupled to the core 213 (Figure II-9), reaction times of up to six days were necessary
to obtain good yields. The synthetic route was modified, and the modification included a

change from the vinyl bromide diene fragment 4b to a tributyl tin derivative 4d (Figure
II-7).92 The extra step can be seen in Figure lI-7, was achieved in 63% yield

(unoptimized). Preparing the stannane was not difficult but its purification was a hassle.

A common side product was the reduced stannane, which was in abundance if the silica
gel column was not buffered with triethyl amine (Et3N).

Even though preparing the vinyl stannane 4d requires an extra step lowering the
overall yield, there were some advantages to using this as the diene fragment. First it is a
known compound making its characterization and the characterization of any unstable
intermediates less compulsive;
secondly it increases the weight of this fragment making small scale reactions easier to
run; and last it is much more stable than its halide counterparts, being able to be stored

for months without any sign of decomposition.

Figure 11-9 Palladium Cross-Coupling With Dienes 4a and 4b

 

 

 

 

HO
| OH
W 5%”0'2‘PP"3’2
TBSO / / + //152 pyrrolidine A _' //153
4a H 87% ‘-
anCu-Ag T330
2Z1M80H/H20 ; _ —
80% 154 OH
HO TBSO
TB so // TBSOTf. NEt3 2 TBSO __ //
— 100% ——
153 _ 155
mew/,9 TBSO __
2:1 MeOH/Hzo V —‘ — 0TBS
600/ 156
0

Pd(dppf)Cl2, 4b

Pyrollidine. 6d
96%

   

A Novel Aldol Reaction

The synthesis of the trio] fragment will be discussed in rigorous detail in the
following chapter, but the impetus for its construction, a novel aldol reaction will be

discussed here.

61

Figure 11- 10 Asymmetric Aldol Reactions Using a Chiral Imidazolidinone Fischer

Carbene Complex69

 

 

 

 

 

 

0—’Cr(CO)4 0 O OH 0 O OH
1) n-BuLi ;
Meg , ~
NJLN CH3 2) R CHO Av Me\NJLNMR' + Me\NJLN/u\/\R.
\—/ 3) HOAc/Ce'V L1 .\—/. .‘
M P“ M5 l’h Md 1311
162a 163a 184b
temperature time ratio yield
R (°C) (min) (163a: 164b) (%)
- 10 2 91 :9 83
n-Pr -30 30 89:11 87
- 30 30 87:13 85 "
-10 10 91 :9 83
i-Pr
- 30 30 95:5 88
Ph -78 to -30 30 98:2 60
* anion generated with LDA

In 1994 Wulff, Shi, and Wilson published the use of a chiral imidazolidinone

Fischer carbene complex developed in our group as a chiral a-unsubstituted acetate

enolate synthon for asymmetric aldol reactions.69 As can be seen in Figure Il-10,

excellent yields and diastereoselectivities were observed when the enolate anion of

complex 162a was reacted with a variety of alkyl and aryl aldehydes.71 These

encouraging results prompted Dr. Wilson to expand the scope of this reaction to 2-

alkynals.70 He found that the desired propargylic alcohols were prepared in good yields

and diastereoselectivities, however the stereoinduction observed in these products was

reversed (Figure IL] 1). This observation was confirmed by X-ray crystallography on

166d.58

62

Figure Il-ll Asymmetric Aldol Reactions of 2-Alkynals Using a Chiral

Imidazolidinone Fischer Carbene Complex

 

 

 

JCJ>\—'Cr(CO). JOL O OH
Me\ O M x
N N/lkCHg + 1. LDA, -95 0C: e N N/U\/\
H H \\ 2. HOAc/Ce'V H R
e Ph Me Ph
162D 165 166
Entry R Diastereoselectivity (antizsyn) yield (%)

a CHcha 13287 57

b Ph 17:83 75

C TMS 9:91 58

d TBS 9:91 59

9 TIPS 9:91 45

 

Before one can attempt to explain the anomaly of a change in diastereoselectivity
when alkynals of type 165 are used a clear understanding of the scope of the reaction is
essential. The chiral auxillary on the carbene complex has three main features that ensure
high diastereoselectivity. First the phenyl and methyl groups on the imidazolidinone
provides facial selectivity by steric interactions with the incoming aldehyde. The
aldehyde will approach from the less sterically hindered face of the enolate. Secondly the
bulky ligands on the chromium provide an even more hindered environment. Transition
state I was proposed to account for the observed stereoselectivity with aliphatic and aryl
aldehydes (Figure II-12). The model has substituent R’ of the aldehyde in between the
two hydrogens of the enolate carbon. This model predicts that as the size of R’ increases,

the stereoselectivity should increase. This expection is realized in the data shown in

63

Figure lI-lO. The selectivity increases from 53 : 42 acetaldehyde (R’ = Me) to 98 : 2 with

benzaldehyde (R’ = Ph).

Figure 11.12 Hypothesized Transition State of Aldol Reaction

1’?
$358
Monomeric Enolate
The third feature that ensures high diastereoselectivity is the chelation of the
imidazolidinone oxygen to chromium. Without this feature there would be free rotation
around the nitrogen-carbene carbon bond of complex 162a in Figure II-13. This chelation
is necessary to set the orientation of the chiral auxiliary spacially. In other oxazolidinone
and imidazolidinone chiral auxiliaries of the type 162c there is free rotation around the
amide bond. Rotation around the amide bond in these systems can be prevented by

adding a Lewis acid or chelating transition metal to the system. The beauty of the carbene

complex 162a is that a chelation controlled conformation about the C,-N2 bond is built

in.89

Figure II-13 Preventing Rotation Around the C3-N2 Bond In Chiral Auxiliaries

O—~Cr(00)4
Me. JKZ/lk O O
N N 1 CH3 /\/U\2)K
\ / R2 \ 1 N 0
Me“; 71311 -'\—/
162a R" 1320

Other Factors that Influence the Asymmetric Aldol Reaction

A number of factors were examined by Wulff and Shi in an attempt to obtain a
greater understanding of the scope of the asymmetric aldol reaction between alkyl
aldehydes and complex 162a. Already discussed is the effect of the size of the R’ group
(Figure II-IO), but other conditions such as temperature, concentration, type of base used

and the effect of additives also investigated and their findings are summarized.

Temperature and Concentration

To examine the effect of temperature and concentration on the asymmetric aldol
reaction, butanal was chosen as the alkyl aldehyde. As can be seen in Figure II-14 from
entries 1 through 3 as the temperature is lowered from -10 0C to —78 OC there is an
erosion of diastereoselectivity. An anti : syn ratio of 93 : 7 was observed at —10 OC and an
anti : syn ratio of 55:45 at —78 0C. In addition, at —95 0C (entry 5) a reversal of selectivity
is observed with the anti : syn ratio of aldol adduct 167 to aldol adduct 168 being 28 : 72.
Entries 3 and 4 examined the effect of concentration. A 10 fold decrease in concentration

results in a change in selectivity from 55 : 45 to 73 : 27, favoring the anti product 167.

65

 

 

Figure II-14 The Effects of Temperature and Concentration on the

Asymmetric Aldol Reaction

0 ' CNCOM 1) n-BuLi

0 0 OH 0 0 9H
M M . ’
e‘NJLNJkCHa 2) n-PrCHO _ Me\N)kN)J\/'\k+ e NJLNW
‘-, 3) HOAche'V a, 5' '9
‘ ph Me Ph
167 168

 

 

 

 

 

 

 

M5 162aPh Me
R We?“ (51:... 1:2,":

1 n-Pr -10 93:7 33

2 n-Pr ~50 84:16 33

3 n-Pr - 78 55:45 85

4 n-Pr‘ -78 73:27 85

5 n-Pr -95 28:72 60

" This reaction was performed with the enolate concentration at 0.007 M.
All others in table-were carried out at 0.07 M.

Type of Base Used

The choice of base used in the asymmetric aldol reaction between complex 162a
and butanal affected the selectivities dramatically. In Figure II-15 entries 2 and 3 show
that using sodium or potassium instead of a lithium based base results in almost complete
erosion of selectivity. Both entries 2 and 3 gave almost equal amounts of the anti and syn
products, 167 and 168, while in entry 1 where LiN(T MS) is used as the base, a 90 : 10

anti : syn ratio of products was observed.

 

 

Figure II-lS The Effects of Other Cations on the Asymmetric Aldol
Reaction

0 —'CF(C0)4 1) Base

0 O OH 0 O OH
Met JL /U\ 2) n-PrCHO Me. A Me\ A i
N N CH3 : N N + N N
Ll 3) HOAche'V .\-—/-.
6 ’Ph Me‘ Ph
167 168

 

 

 

 

 

Me Ph M
162a
Entry Base temgeéature (ariztii:tl 3:41) {if/cl?
1 LiN(TMS)2 -30 90:10 76
2 NaN(TMS)2 -30 55:45 61
3 KN(TMS)2 - 30 53:47 58

 

The Effect of Additives

The effect of additives was also studied by Wulff and Shi and some data are
shown in Figure II-16. In the reaction between complex 162a and butanal, both HMPA
and BITMSA improves the selectivity of the reaction, but a more dramatic change occurs

at -78 0C than at -30 0C.

67

Figure II-l6 The Effects of Additives on the Asymmetric Aldol Reaction

0—*Cr(CO)4 1) n-BuLi

0 0 OH 0 0 9H
Me~ )L k 2)Additive Me. A Me\ A ’
N N CH3 _ N N + N N
.\—/~. 3) n-PrCHo .\—/-. .\—/-.
Mé ’Ph Mé Ph
167 166

 

 

 

 

 

 

 

M 1 6281”“ 4) HOAche'V
em Adam W16,“ 1.1231", 1:2,,
1 none -78 55:45 85
2 HMPA (1.3) -78 80:20 61
3 none -30 88:12 88
4 HMPA (2.0) -30 90:10 66
5 BITMSA (3.0) -76 64:36 75

 

Analysis of the Factors Affecting the Aldol Reaction of Carbene Complex 162a

When the data from the experiments described above was complied and analyzed
it suggested that one possibly for the erosion of selectivity is aggregation of the enolates.
The results of Figure II-14, entries 1 and 3 are consistent with the presence of aggregates
at lower temperatures. The monomeric enolate transition state (Figure II-12) would be
expected to give higher selectivity at lower temperature rather than the reverse. Entries 3
and 4 of Figure II-14 also support this hypothesis, when a 10-fold decrease in
concentration occurs selectivity for the formation of aldol adduct 167 is increased. The
aggregation of enolates is known to be disrupted with dilution. Less aggregation would
be expected at 0.007 M (entry 4) than at 0.07 M (entry 3). Thus a greater proportion of
the monomeric enolate would be present and the observation of higher selectivity is
consistent with the aggregation of enolate I (Figure II-12) and with a lower selectivity

from the reaction of the aggregated enolate than with the monomeric enolate.

68

 

 

 

The results from Figure lI-15 also suggest that at —30 0C the sodium and
potassium bases promote the formation of aggregates. The ability of sodium and
potassium to form aggregates at higher temperatures can be expected because sodium and
potassium are bigger and softer cations. However a more conclusive argument may be
reached if these reactions are repeated at —78 0C.

The effect of additives maybe due to disruption of aggregates shown in Figure ll-
16. It is known that lithium aggregates maybe disrupted using bases such as

hexamethylphosphoramine (HMPA) or tetramethylethylenediamine (T MEDA).72 In the

reaction with n-butanal, extensive studies were carried out to determine if aggregates
were involved in this asymmetric aldol reaction. Figure II-16 entries 1-4 suggest that
lithium aggregates are being formed at very low temperatures, because using HMPA at
—78 OC improves the diastereoselectivity dramactically in favor of the anti product, 167.

At -30 OC, however very little change in diastereoselectivity is observed.

Rationale For The Reversal of Diastereoselctivity in Alkynals

Only aldehydes that cannot chelate to the chromium have been discussed so far.
These results might imply that the alkynals ability to chelate to the metal center might not
have an effect on the selectivities observed, but rather, are the results of sterics and
aggregation alone. However, Figure II-16 entry 5 shows that bistrimethylsilylacetylene
(BTMSA) can have a small effect on the selectivity. An analysis of all this data and more
that has not been presented here has been summarized.-"‘56'73

While the mechanism of the reaction is not known in detail, the stereoselectivity in

alkyl aldehydes appear to be dependant on the aggregation state of the enolate where the

69

least aggregated species favor the anti-adduct and the more aggregated form of the
enolate favors the syn-adduct. If the least aggregated form is the monomer, then the
observed stereoselectivity could be accounted for by the open transition state I where the
larger Rl group leads to high anti-selectivity (Figure lI—17). The reversal of selectivity in
the reaction of the alkynals could be accounted for by their reaction with the more
aggregated enolate since these reactions can only be carried out at low temperatures. It is
also possible that the alkynals could react via displacement of the imidazolidinone
oxygen as in transition state 11. The data does not allow for a definitive distinction to be

made at this time.

Figure II-l7 Hypothesized Transition States of Aldol Reactions

Me\N Me O
O=< jPh Pf) 8 N )LN/
OC‘ 9 N OCEaCr ’6 Me
Cf! " H R :
8 J, o
H R' o
I ll
Monomeric Enolate Co-ordinated Alkyne enolate

High selectivities are obtained when the reactions are carried out using a dicobalt
hexacarbonyl complexed 2-alkynal (Figure ll-18).58 The 3R diastereomers are observed
which is the same as seen with the aryl and alkyl substrates. As was discussed above, this
could be due to a change in the mechanism or to steric factors, since the protected alkyne
is much bigger than the 2-alkynals. The diastereoselectivities obtained were higher by
comparison to the unprotected alkynals. With this modification both diastereomers can be

accessed in high yields and selectivities. This discovery is utilized in the early stages of

the trio] fragment synthesis to set the C11 stereogenic center of fostriecin.

7O

Figure II-18 Asymmetric Aldol Reactions of 2-Alkynal Cobal Complexes with a

Chiral Imidazolidinone Fischer Carbene Complex

 

0—’CF(CO)4
Me A NJL O R 1L0A-760c Me ijiNo
‘N CH3 + co. CO
\—/N

C0—CO' C0—C0 2 HOAC/COIV

 

 

.~ ', 1
Me‘ Ph
182a CO 169 CO W166
Entry R Diastereoselectivity (anti:syn) yield (%)

a CH2CH3 88:12 50

b Ph 87:13 59

c TMS > 99.5:O.5 67

d TBS > 99.5:0.5 65

e TIPS >99.5:0.5 48

 

Despite the many experiments carried out so far, the exact mechanism of the aldol
reaction of imidazolidinone carbene complexes is still unknown. There is however some
evidence to suggest that steric interaction, aggregation, and alkyne chelation to the

chromium all could possibly influence the stereochemical outcome of this reaction.

71

CHAPTER 3

SYNTHESIS OF THE TRIOL F RAGMENT

First Generation Synthesis of the Trio] Fragment
The initial synthetic strategy of the triol fragment dates back to 1994 and the
discovery of the asymmetric aldol reactions of imidazolindinone carbene complexes.58 At

this point the absolute configuration of fostriecin was unknown and as a result the initial

and final strategies differ significantly with a few key reactions remaining unaltered.

The lack of knowledge about the stereochemical environment at C8, C9 and C”,
led to the route seen in Figure III-1. The three key reactions being an asymmetric aldol
between a Fischer carbene complex l62b and a 2-alkynal 1165 to construct the C11
stereogenic center; a Homer-Wadswoth-Emmons (HWE)93 olefination to construct either
the E or Z isomer of trisubstituted alkene 172; and a Sharpless asymmetric
dihydroxylation94 on that alkene to give the C8 and C9 stereocenters. The absolute

stereochemistry of the asymmetric aldol depends on the choice of the proper enantiomer

of the imidazolidinone auxillary in the carbene complex and would afford either of the
two Cll epimers which when combined with the HWE93 and Sharpless AD94 could

access any of the eight permutations possible.

72

This synthetic route was abandoned because of disappointing
diastereoselectivities observed in the Sharpless AD94 reaction. A 2:1 ratio with the PHAL
ligand and a 1:1 ratio with the PYR ligand were the best results obtained. Matters became
more complex when it was observed that these diastereomers were inseparable by silica
gel chromatography and that the physical state of the diol is an oil. Derivatization using
9-fluorenone and p-methoxybenzaldehyde failed, so at this point it was decided that

designing an alternative strategy would be the better option.

73

Figure III-1 First Generation Retrosynthesis of Triol Fragment

 

 

 

 

 

 

 

0TBS
O 11 \\ SharplessAD
9 . -.
EtO OH
OH
171
0TBS 0TBS
O 8} 11 \\ TMS {HWE Olefination 3 911 \\ TMS
E10 0 H
172 173
JOL O 0TBS 0—‘Cr(CO)4 O
Aldol Reaction
\N N 9 1 § L j. \N N)k + HJ\

H TMS H TMS
Ph Ph

174 162i) 165

Second Generation Synthesis of the Trio] Fragment

In 1997, Boger and co—workers published the absolute stereochemistry of
fostriecin.22 This discovery occurred in a timely fashion because it was right around that
time that our second generation synthetic efforts were being developed. In the new
approach the HWE93 and Sharpless AD94 would be replaced by an acyl anion addition
and an Evan’s 1,3-anti reduction of a B-hydroxy ketone as key reaction steps as outlined
in Figure 111-2.75

The C9 and C11 stereogenic centers were known to be anti and both possessing an

R configuration. An Evan’s anti-reduction75 of the B- keto alcohol 176 would induce the

correct chirality at the C9 position since the chirality at the C” alcohol would already be

74

established from the novel asymmetric aldol reaction discussed earlier. The conversion of
the Weinreb’s amide 175 to the dithiane adduct 176 was planned utilizing the previous
work of Leibeskind who demonstrated that Weinreb’s amide could be directly alkylated
with 2-lithio-l,3-dithiane.76 The one-step conversion of 166C to 176 by addition of 2-
lithio-1,3-dithiane to l66c failed. In addition the direct conversion of 166C to 175 failed.
The synthesis of 176 was achieved by initial conversion of 166C to the methyl ester and

then transformed to 176 via 175.

75

Figure III-2 Second Generation Retrosynthesis of Triol Fragment

 

 

 

 

 

 

O P
BC 'I OP I’OEt O
‘ 11 LDA. CI’ ‘OEt 11
BIG] 9 \\ L A, 9 §
8 OP TMS ' 8 OP TMS
O U 177
118
O OH
Evan's Reduction 8 9 11 \ Acyl Anion Addition
r 4', S S \ r -
TMS
176

O OH O O

O H
Weinreb's Amidation
Meo‘N 9 11 \\ r :3 \NAN 9 11 \\
i'ae TMS H ms
175 Ph 1666

As shown in Figure III-3, the reduction of 176 with Evan’s procedure gave a

 

 

single diastereomer by proton NMR, which was presumed to be the anti-diol 179. The

anti-stereochemistry was confirmed by Mark Parisi upon derivatization of diol 179 with
2,2-dimethoxypropane to give 180 and subsequent proton and carbon-l3 studies.56 The

Cll proton adjacent to the alkyne is a triplet at 4.70 ppm with a coupling constant of 6.5

Hz. The C9 proton adjacent to the dithiane is a doublet of doublets at 4.35 ppm with
coupling constants of 4.7 and 10.2 Hz. The C10A proton syn to the C9 proton is a doublet
of doublet of doublets at 2.23 ppm, with coupling constants of 2.7 (geminal coupling),
4.2 and 10.2 Hz. The C10,3 proton syn to the C11 proton is obscured by signals from the
dithiane ring, so its coupling constants could not be determined. The observable 10.2 Hz
coupling constant between C9 and C10A is consistent with the twist-boat confirmation,

characteristic of anti diol acetonides. The carbon-l3 N MR spectrum of the acetonide 180

in Figure III-3 provided additional verification of the relative stereochemistry of the two

alcohols.”77 The chemical shift of the acetal carbon is 101.26 ppm, within the range

76

reported by Rychnovsky77 for anti diol acetonides (syn acetonides usually have chemical

shifts near 99.0 ppm), and well outside the 995-1005 ppm range where and assignment

could be ambiguous. The methyl groups on the acetonide are located between 21 and 27
ppm, also well within the range reported by Rychnovsky77 for anti acetonides (syn

acetonides have methyl group shifts at 19.5 and 30.0 ppm).

77

 

Figure III-3 Addition of Dithiane, Reduction, and Acetonide

 

 

 

 

 

 

 

Formation75’76’77
9H 0 9H 0
: IOMe . . . T 8
// 11 9 '1‘ 2-methyI-1,3-dlth1ane, n-BuLi : ¢ 11 g S
TMS 175 Me THF, ~78 00, 2h TMS 176 V
65%
(2H CH
M64NHB(0AC)3 ’ 2.2—Dimethoxypropane, PPTS (20 mol%)
_ // :
1:1 aetone/ acetic acid, TMS S S CHzclz, 65 00 (sealed tube). 24 h
0 00.211 179 V 78%
91%
(94,
i 6
11 9
TMS é S S
180 ‘\/'
TMS 4.70 ppm. t, TMS
\\ H J '-' 6.5 HZ \\
Obscured ——’H130 O><CH3 O><CH3
HA R 11 H 0 CH3 0 CH3
3 CH3
2.23 ppm. ddd. Acetal carbon = 101.26
J = 4.2, 2.7, 10.2 Hz 4.35 ppm. dd, .
J = 4.7. 10.2 Hz Acetal CH3 5 between
23-27 ppm
Proton NMR Analysis Carbon-13 NMR Analysis

With the stereochemistry of the C9 stereocenter verified, the second generation

synthesis of the triol fragment was pursued by Mark Parisi. A tertiary butyl silyl (TBS)
protection of the diol only gave the mono-protected product. The product expected from
the TBS protection of 179 was the bis silyl ether. This was shown by Su Yu to be the
mono-silyl ether 181 correcting an error that had been made in Mark Parisi’s thesis. The

second TBS protection was only achieved after the cerium ammonium nitrate (CAN)

78

oxidation was carried out on dithiane 181 and these two steps gave the ketone 182 in
moderate yields.95 Alkylation of the enolate of 182 with diethyl chloro phosphonate was

found to give the O-alkylated product 183 and not the desired C-alkylated product.

Figure III-4 Protection, Oxidation and Phosphonate Addition

 

 

 

OH OH 9H 0TBS
8 g 8 «I; 11 1) 3:1CH3CN/H20.rt.3h
S S 11 \\ TBS-CLImIdazole _ S S \\ TMS CAN :
V 179 TMS DMF,rt.2h V 181 2) TBSCI. Imid. DMF
76% 31%
o
grssoras 3,03 orasoras
6 ’ LDA. CI’ \OEt i
9 11 § 0 = 9 11 \\
o 182 TMS THF.402°/C. 1" 0%,0 183 TMS
° EtO’ ‘05:

In addition to the above problem of O-alkylation of 182, it was also realized that

the C9 and C11 alcohols would need to be protected with different groups because later in
the synthesis the C9 would have to be phosphorylated selectively. It was found that while

selective protection of the C11 was possible, it proved difficult to protect C9, presumably

due to the presence of the methyl group on the dithiane. So again another strategy was

sought at this point.

Third Generation Synthesis of the Triol Fragment

Fortunately, the third route retains the same key steps. A change of the C9

protecting group was required as well as and an efficient and successful method for

introduction of the phophonate ester.

79

Finding an appropriate protecting group on C9 proved problematic. The 2-methyl-

1,3-dithiane unit in 179 provides a much more sterically hindered environment for the C9

hydroxyl than does the acetylene unit for its neighboring hydroxyl. Hence after mono-

protection of the less hindered Cll by a TBS group, an attempt to introduce a tri-ethyl

silyl (TES) group and a methoxy methyl (MOM) group on the C9 alcohol failed. A TES

and MOM combination was also attempted but proved unsuccessful.

Su Yu, a post doctoral fellow in the Wulff group, utilized a trimethyl ortho ester

formation between the C9 and C11 alcohols of diol 184 and its reductive cleavage as a
solution to this problem (Figure lII-S).78’79 Reductive cleavage of 184 with DIBAL-H
placed the MOM group on the more hindered C9 alcohol, and a subsequent TBS

protection of the Cl 1 alcohol followed smoothly. It is of interest to note that DIBAL-H in

hexanes and dichloromethane provide the desired product but DIBAL-H in

tetrahydrofuran results in only recovery of starting material.

Figure III-5 Selective Protection of Diolm'79

OMe
OH OH O MOMO OH

O
8 ' g 8 .
s s CSA #383 "R B“- :s g 11R
179
K/I 91% |\/l 184 94% V 135

 

 

 

 

R E—TMS
MOMO ores
8: MOMO ores
7330“ 9 11 R NCS/AgN03 8 =
t s s s 9 11 R
EtaN V 186 CHacN/HZO o 137
96% 94%

80

With the problem of selective protection of C9 vs Cll solved, attention was turned

to the problem of C vs O phosphorylation. It was envisioned that O-phosphorylation
could be reduced if the ketone 187 was converted to an alpha bromo ketone and the
Arbuzov’s96 reaction perfomed. Xuejun Lui in our group found that the reaction of the
bromo ketone 189 with triethyl phosphite gave several products with the desired
phosphonate 209 as a minor product (see Figure III-6). Conversion to the triphenyl
phosphonium salt 190 by reaction of the alpha bromo ketone 189 with triphenyl

phosphine also failed. This reaction gave primarily reduction of 189 to 187.

Figure III-6 Preparation of Phosphonate 209

 

 

MOMO 0TBS
MOM? OTBS LDA, CIPO(OEt)2 8 ':
8 ' —. 9 11 R
918711R THF Etc\ 0 188
0 40% Eto—P’
NBS ”
EhN/TMSOTf
60%
MOMO ores
MOM? 0TBS (EtO)3P \\ 8 ; several
Br 9 11 R 110°C EtO/\ products
0 189 OEt O 209
minor product
PPh3
B,— MOM? ores R: : ms
Ph\+ 8
' 1 0
Ph 0

In 2001 Xuejun Lui suggested that the methylene unit alpha to the phosphonate in
the triol fragment could be installed via nucleophilic addition of a phosphonate enolate to
a methyl ester 195 (Figure III-7) to give phosphonate. This meant that instead of using 2-

methyl-1,3—dithiane as an acyl anion equivalent 1,3-dithiane would have to be used. This

81

 

 

reaction sequenced worked smoothly to produce the desired phosphonate 3a in a 7.6%

yield over 15 steps from chromium hexacarbonyl.

Figure III-7 New Approach to the Phosphonate 3a

 

 

0 0” 9” 0” 1.(MeO)30H.CSA. MOM? OTBS
H 8 9 "R Me48H10Ac>3 _ ”MR 2. DIBAL-H A H a 5
S 1:1 acetone/acetic acid S S 3. TBSOTf. NEt3 s s 11 R

S
191 192 193
V 90% V (20;1) 88% over three steps V

 

 

 

 

MOMQ 0TBS MOMO ores
NBS (seq) H '7 poc, DMF 5 (MeO)2POCH2Li
8911R :M908911R T’
CH3CN/H20 o CH30H 72%
94% ‘9‘ 65% 0 195
o OTBS
MeO\l|
R: :——TM$ MeO’P 11 R
9
8 OMOM
0 3a

 

 

 

 

This change in acyl anion equivalents also provided the solution to another

problem encountered in the first generation synthesis, namely the direct conversion of

imidazolidinone 166c to the dithiane derivative 176 (Figure III-2). Xuejun successfully

achieved the related transformation of 166c to 191 in 79% yield as shown in Figure III-8.

Success in this case may be due to the smaller steric demand of 2-lithio—l,3-dithiane vs 2-

methyl-2-lithio-l,3-dithiane. This removed two steps in the preparation of 3a to provide a

 

13 step synthesis in 8.8% overall yield. The conversion of 166c to 191 was later

optimized to 92% increasing the overall yield of 3a to 10.2% from chromium

hexacarbonyl.

82

Figure III-8 Improved Acyl Anion Equivalent Addition

 

0 O OH
MeOMgBr M83Al 0 OH
M
exNJLNMR CHzclz M MeMeONHzHCl A Me /U\/L
‘3:— 95% MeO 9 11 R 90% 'i' 9 11 R
Me1mPh 195 We 175
R: Z—TMS

S n-BuLi
< s> 80%

s 0 OH
<: > + n-BuLi THE-78°C H><ll\/L

s 11 R

9
T’ S S 191
79 % (optimized to 92 %) V

68 % over three steps

 

Preparation of TES Protected Triol Fragment 3b
The first total synthesis of fostiecin (CI-920) was achieved by Boger and co-

workers24 around the same time our triol fragment was completed and being scaled up.

74.93

Boger’s approach was related to ours in that an HWE reaction was used to couple the

lactone fragment to the rest of the molecule by the introduction of the chiral center at C8

via addition of a methyl organo metallic compound to a ketone. It became apparent that
in order to achieve the correct diastereomer in the methylation step a Felkin-Ann non-
chelation control addition would be necessary (Figure III-9). Boger achieved this using a

MeLi-CeCl3 mixture resulting in a 3:1 diastereomeric ratio of C3 epimers, and with a 20:1
ratio of 1,2 versus 1,4 addition products.24‘97 The protecting group used in Boger’s
synthesis was a triethyl silyl group which is known to promote non-chelation controlled
additions. Our triol fragment possessed a MOM group on the C9 alpha to the C8 carbonyl,

this is known to give the chelation controlled diastereomer as the major product. This

meant our synthetic approach had to be modified to avoid this problem (Figure Ill-10).

83

 

Figure III-9 Boger Non-Chelation Controlled Methylation Conditions24

5.9.991

0TBDPS
MSU-'CBCig

THF/Toluene

-78 00, 10 mins
3 : 1 dr

 

wulff

Methylating Reagent

 

Conditions

   

84

 

Figure III-10 Retrosynthetic Analysis of Triol Fragment 3b

 

 

 

 

 

 

 

 

 

 

 

 

 

 

O OTBS OTBS
MeOJI (MeO)2POCH2Li oxidation
MeO’P 11 § . :» 11 \\ r
TMS
8 OTES M60 3 OTES TMS
0 3b 0 200
OTBS Evans reduction 0 OH acyl anion
11 selective protection 9 11 addition
H 69 Q TMS ' i S 3 § TMS . A
0753 oxidative l\/l 191
O 201 deprotection
J0L O OH O—>Cr(CO)4 o
Me\N N 9 11 \ Asymmetric Adol Me‘N N + H \
\ 7 \ . :- \ 7 \
.3 168C TMS ‘. -. 182a TMS
Mé Ph Me‘ ’Ph 165

Problems Encountered in Modification - Preparation of Phosponate 3b

At first glance, it may seem like a trivial task to change the protecting group at the
C9 alcohol in phosphonate 3a (Figure III-7) from a MOM group to a TES group.
However, many challenges were encountered and they were mainly due to the lability of
the TES group.

Since there was a ready supply of the MOM protected intermediate 193, the most

efficient approach to 203 (Figure III-11) would be to develop a protocol for the removal

of this MOM group from 193 and then reprotection with a TES group.81 Unfortunately,

removal of the MOM group with magnesium bromide etherate (MgBr2.OEt2) in 1,4-

butanethiol gave a mixture of undesired compounds. Better luck was obtained when the

ortho ester 184 (Figure III-5) was cleaved to the diol 192 using boron triflouride etherate

[.82

(BF3.OEt2) in mercaptoethano This reaction gave an 86% yield, but at this point there

85

 

was insufficient material to continue to investigate the remaining steps. Thus the
protection protocol for the diol 192 was modified.
With the change from disubstituted dithiane 179 (Figure III-S) to the

monosubstituted dithiane 192 (Figure Ill-11) it was conceived that the slightly less’
hindered environment at C9 might make the required stepwise selective protection
feasible. Using this change to our advantage we protected the C11 alcohol with TBS to
give 202, and subsequently web were in fact able to protect the C9 alcohol with the TES
group to give 203. The results were encouraging with an overall yield of 86% for the two
steps. As can be seen from Figure III-1 l, the conditions required for protection of C9 and
C11 were similar, hence, we decided a one-pot procedure might be convenient. To our

surprise not only was the reaction successful, but a dramatic increase in the overall yield

was observed producing 203 essentially quantitatively.

86

Figure III-ll Improved Selective Protection of Diol Fragment

 

 

 

 

 

0 OH 9H OH OH ores
H 8 Me4BH(OAc)3 H 8 e resort. NEt3 H 8 :
9 11 > 11 = 9 11
S S 191 \\ TMS 1:1 acetone 8 S192 \\ TMS -78 OC-r.t S S \\ TMS
V laceticac1d overnight V 202
90% 86%
1. resort, NEt3
2. TESOTf TESOTf, N33
-78 °C-r.t
" overnight
100%
oresores
H 6 g A
s s Q ‘

1
V 203 TMS

99% for one-pot process
Oxidative removal of dithiane using N—bromosuccinamide (NBS) in acetonitrile

and water was very successful when the MOM protecting group was on the C9 alcohol

193 (Figure Ill—7, see experimental for details).83 However, repeating this protocol with

the TES protected 203 gave a mixture of products and recovered starting material.
Solubility seemed to be a problem. An attempt to solve this problem was made by
substituting acetonitrile with propionitrile in the solvent system. However, this also gave
a mixture of products with at least six spots on a thin layer chromatography (TLC) plate.
There was also an attempt to solve this problem by reversing the order of addition of the
reagents, i.e the NBS solution was added to the protected diol, but to no avail.

At this point a series of reactions were set up to screen various conditions as seen

in Table III-l. Two of the five reactions screened gave a clean crude proton NMR of the

desired product.84 In entry 5, one can see that using the same conditions developed before

for the MOM derivative 193 with the addition of CaCO3 gave a good result. Based on

87

this observation, it appears that a base had to be used in order for this transformation to
be successful. The source of the problem is presumed to be the generation of hydrogen
bromide (HBr) which under the conditions leaves the MOM group unharmed, but results
in the cleavage of the TES group.

Table III-1 Oxidative Deprotection Screening Reactions

 

Solvent

 

Entry Reagents Temperature/ 0C System a Results
1 NaHCO3, Mel 70 CH3CNszO Failed
2 CaCO3, Mel 70 (CH3)2CO:HZO Failed
3 CaCO3, Hg2C12 25 (CH3)2CO:HZO Failed
4 BaCO3, NBS 25 (CH3)2CO;H20 Good NMR
5 CaCO3, NBS 25 CH3CNzH20 Good NMR

 

a- A 9:1 ratio of solvent to H20 was used in each case.

Preparation of the Methyl Ester 200 and Completion of Phosponate 3b

Having overcome the set back of the removal of diathiane in the presence of TES
protected alcohol, the synthesis was continued as planned. A pyridinium dichromate
(PDC) oxidation of the MOM protected aldehyde 194 in methanol (MeOH) and dimethyl
formamide (DMF) proved successful in the preparation of ester 195 (Figure III-7).85 As
with the dithiane deprotection problems were encountered when the PDC oxidation
protocol was applied to the TES protected aldehyde 201. This reaction gave a 44% yield
of a methyl ester similar to 200 but which had lost its TES protecting group.

The solution to this predicament came via a Leibegs Ann. Chemistry 1992

publication by Konig and coworkers.86 Involving an iodine ([2) oxidation in MeOH in the

88

presence of sodium bicarbonate (NaHCO3) they reported a procedure for conversion of
aldehydes to methyl esters.

The stability of the aldehydes over prolonged periods of times is of a general
concern to organic chemists. Anticipating this, a sequential approach from the dithiane
203 to the methyl ester 200 was attempted and found to work. It was found that
purification of the aldehyde 201 via chromatography was unnecessary to obtain good
yields. A simple work-up of 201 with a saturated solution of sodium thiosulfate

(NaZSO3), filtration, extraction with ether and drying was sufficient to proceed to the next

step. The two steps done sequentially gave an 83% yield of 200.

Figure III-12 Oxidative Deprotection and Triol Fragment Completion

 

 

 

OTESOTBS OTESOTBS
H 8 E NBS.CaC03, H 8 F '2, NaHCO;,
9 11 \ t 9 11 \ Av
S S \ \
V 203 TMS CHacN/Hzo o 201 TMS MeOH
9 : 1 83 % over 2 steps
QTESOTBS . ' oresores
M60 8 ' CH3PO(OMe)2, n—BuLI ;
9 11 \\ > MeO\P 8 9 11 \
20° ° MeO’il \
O TMS 82 /o O 0 3b TMS

 

[ Triol Fragment ]

 

With the knowledge that the TES group is unstable in even mild acids, the
conditions for the introduction of the phosphonate were taken into account. This step
should not be a problem since no acid is generated in the reaction. Thus the conditions

developed from our earlier synthetic efforts on the conversion of 195 to 33 (Figure III-7)

were attempted on ester 200.8!7 The reaction was however not complete after 48h. This

89

was suprising since the MOM protected derivative 195 only required 2h. Increasing the
reaction temperature to 25 OC overnight gave 3b in an 82% yield for a total of 14.2%
yield over 10 steps from chromium hexacarbonyl.

While repeating the sequence in Figure III-12, other problems were incurred that
were not seen in the first time through. The n-butyl lithium used must have an accurate
titer otherwise the TES group is lost. In addition when performing silica gel
chromatography on the intermediates a solvent system containing 0.5-1% triethyl amine

should be used a precautionary measure to avoid any loss of the TES group.

Preparation of Phosponate 3c

lnspite of a very successful synthesis of the triol fragment 3b, two
major problems still plagued this synthetic pathway. First using TES as a C9 hydroxyl
protecting group meant that extreme care would have to be taken when performing any
transformations in which mild acid is generated. On two occasions thus far CaCO3 and

NaHCO3 had to be used to prevent deprotection (Figure III-l2). In addition each silica

gel column would have to be pre-treated with Et3N amine beginning at the point that TES

is first introduced as a protecting group up until the point of its removal.

A second and more serious issue was the low selectivity obtained by Boger when

the C8 carbonyl is methylated. As was mentioned previously (Figure III-9) this is a
Felkin-Ann non-chelation addition to the C8 carbonyl where the selectivity should be a

function of the size of the protecting group on the Cg-oxygen. When protected by a TES

group only a 3:] diastereomeric ratio was observed (Figure III-9).

 

We thought that perhaps both these problems could be addressed if a bulkier and
more acid stable silyl alternative was used. Other silyl protecting groups that would have
these characteristics are the triisopropyl silyl (TIPS), tert-butyl dimethyl silyl (TBS), and

tert-butyl diphenyl silyl (T BDPS) groups. The results from the work of Mark Parasi, Su
Yu and Xuejun Lui suggested that trying to protect the C9 hydroxyl group on dithiane

181 with very bulky protecting groups would not work (Figure III-4). However, using
dithiane 192 instead of dithiane 181 might be expected to relieve some steric constraints
(because the methyl group on the or carbon has been replaced by a smaller moiety, a
hydrogen atom). Nonetheless, we reasoned the large TBDPS group would probably still

not be a viable option. The choice between the two other silyl protecting groups was

made simple when Boger’s publication was carefully examined. In his report a C9 and
C11 di-TBS protected intermediate was published. This derivative was only made after

the C8 methylation step and interestingly after the accidental removal of TES from the C9

hydroxyl. We anticipated that the larger TBS group would give better selectivity for the
methylation step and any similar or identical compounds made could be compared to the
intermediates reported in Boger’s publication.

Fortunately, the change of silyl protecting groups from TES to TBS proceeded
uneventfully. As can be seen in Figure III-13 a quantitative yield of 204 is obtained for
the TBS protection of the diol 192 and the two remaining steps gave over 90% yield. This
change improved the yield of the trio] fragment from 14.2% to 19.8% for the ten-step

sequence starting from chromium hexacarbonyl.

91

Figure III-l3 Oxidative Deprotection and Triol Fragment 3c Completion

 

 

 

 

9H OH oresores
H T T
s 6 3 11 \\ resortquL I; 6 g 11 \\ NBS.CaCO;,. :
‘\/’ 192 TMS ~78 °C-r.t V 204 TMS CH3CN/H20
3d; 100% 9.,
I2. NaHCOe QWSOTBS CH3PO(OMe)2. n-BuLi QWSOTBS
A M60 8 '9 11 - MeO\P 8 9
”‘90” 205 \\ 99% MeO’II 3c 1‘ Q
95 % over 2 steps 0 TMS 0 0 TMS

 

[ Triol Fragment ]

 

Accidental Preparation of Phosponate 3c

While preparing triol fragment 3b (Figure III-12), an unexpected transformation

occurred. When dithiane 203 was sequentially treated with n-bromosuccinimide (NBS)

and calcium carbonate (CaCO3) in a 9 : 1 mixture of acetonitrile : water and iodine (12),

sodium bicarbonate (NaHCO3) and methanol (MeOH) (Figure III-12) a 50% yield of a

new product was observed. After rigourous NMR, IR and MS analysis this compound
was identified as methyl ester 205, which has two TBS protecting groups. It became
apparent that in the work-up of this reaction, the TES group was cleaved and a TBS from

another molecule of 204 underwent intermolecular exchange. This transformation could

occur via a direct protection of the C9 hydroxyl by the TBS of another molecule of 204,

or first an intramolecular silyl migration from Cll to C9 followed by a reprotection of CH

from another molecule of dithiane 204.

92

Preparation of Phosponate 3d

While trying to optimize the reaction conditions for the last step in the preparation
of phosphonate 3c (Figure III—13), a phosphonate side product was isolated in 10% yield
which proved to be the alkyne desilylated material 3d. Varying the ratio of n-BuLi :
dimethyl phosphonate led in some instances to the removal of TMS from the acetylene.
In order to couple the diene fragment 4 to the triol fragment 3c this step would be
essential (Figure IV-13). If a one-pot procedure for these two steps could be developed it
would create a more attractive route towards the natural product. Unfortunately further
experimentation on the ratio of phosphonate to n-BuLi did not improve the yield of this
product but resulted in only reduced yields of triol fragment 3c.

The removal of TMS from phosphonate 3c was also attempted with potassium
carbonate (K2C03) in MeOH but this only produced decomposed material. Larger

quantities of this phosphonate fragment 3d could, however, be prepared by performing
the alkyne deprotection prior to phosphonate addition. The first step is high yielding as
can be seen from Figure III—14. However, the dimethyl phosphonate addition to 200

could only be optimized to 67% yield, a result that was not always reproduced.

Figure III-l4 Preparation of Triol Fragment 3d

 

 

QTBSOTBS QTBSOTBS _
M80 8 g 11 \ K2003.M60H ; M30 3 9 11 \ CH3PO(OMe)z.n-BULI A
99"/ 67°/
0 205 \ TMS o o 206 \ H °
grasoras
Me0\ 8 f
P 9 11 \
M ’II \
90 0 3d H

 

O
[ Trio! Fragment]

 

93

CHAPTER 4

ASSEMBLY OF F RAGMENTS AND THE

INHERENT PREDICAMENTS

With the three fragments in hand, it was planned that the lactone and triol

fragments would be assembled first by utilizing a Homer-Wadworths-Emmons (HWE)
reaction.”93 Methyl addition to C8 of ketone 212 (Figure IV -3) followed by deprotection

of the acetylene and a palladium cross coupling to the diene fragment 4b should afford
the fostriecin core. The subsequent steps have been accomplished by the Boger group on

an almost identical compound. There is a Z-olefin in Boger’s intermediate 50 (Figure IV-

4) at C12 versus a triple bond at C12 in compound 212 (Figure IV-3). In addition the silyl

groups on the C9 secondary alcohol differ, a TBS group in our fostriecin core versus a

TES group in Boger’s.

94

Figure IV-l Our Retrosynthetic Analysis of Fostriecin

OH

Pd~Catalyzed Coupling

Homer-Wadsworth-Emmons ' '
Olefination \_/

 

 

 

 

 

 

 

 

 

 

 

 

 

OIPr [ O 0R2
Me0\ “,OMe
I O P % M 0R1
R
o 0R1 R
2a H 0 3 4
Lactone fragment Triol fragment Diene fragment

3a R=TMS,R1=MOM R2=TBS 4a R=|.R1=TBS

3b R = TMS, R1 = TES R2 =TBS 4b R = Br, R1 = TBDPS

3c R = TMS. R1 = TBS R2 =TBS «to R = SnBu3, R1 = TBDPS

3d R: H, R1=TES R2 =TBS

The Homer-Wadsworth-Emmons Reaction

The retrosynthesis shown in Figure IV-l, like the synthesis by Boger24 and the

74.93

synthesis by Just and O’Connor, employs a Homer-Wadsworth-Emmons reaction as a

key reaction necessary to obtain the E-configuration at C5-C6 double bond. At the time
only triol fragments 3a and 3b had been prepared and 3a was predicted to give the wrong
stereochemistry upon methylation of the C8 carbonyl (see section entitled Preparation of
TES Protected Triol Fragment 3b). Phosphonate 3b was chosen as the Homer-

Wadsworth-Emmons precursor. Attempts to repeat Bogers’24 protocol for the Homer-

74.93

Wadsworth-Emmons reaction gave a disappointing 12% yield as the best result

(Figure IV-2, entry 1). In order to test whether the substrate 3b was sensitive to these

95

conditions, the reaction was repeated with the simple phosphonate 209 as the substrate

which gave the E—olefin 210 was obtained in 86% yield. Thus, it is likely that the TMS-
protected acetylene unit in 3b, which is not present in the phosphonate used by Boger,24
is sensitive to potassium tertiary butoxide (t-BuOK) under these conditions. Other bases

were screened as shown in Figure IV-2. A triethylamine—lithium chloride (Et3N-LiCl)

combination provided the best results with a 94 % yield for this step.77’78

96

Figure IV-2 Attempts at Homer-Wadsworth—Emmons Coupling

 

MW
0
o Et0\"|a
0 EtO’ O t-BuOK, toluene
I + \/ = I
/o -78 °C - fit
23 0 209 overnight
86%

 

k 0 ores
o Mao‘i"

 

 

 

 

Base
MeO’ \\ =
O + TMS
I OTES THF or Toluene
/O
0
2a 3b
Conditions Yield (%)
1) i) t-BuOK, toluene, -78 0C - r.t, overnight 12
2) i) LDA, THF, 3b, -78 °C, 45 min, ii) then aldehyde, 45
-78 °C, 45 min, r.t, 2h
3) i) LiCl, THF, 3b, 5 min, ii) 0 °C; Et3N, warm to r.t, 30 min 94

iii) 0°C, aldehyde, warm to r.t, 24 h

The conditions developed above for the coupling of aldehyde 2a and

phosphonate 3b were also successful when applied to phosphonates 3c and 3d. Over

90% yield of exclusively the E-isomers 211, 212, and 213 was obtained (see Figure IV-

3).

Figure IV-3 Homer-Wadsworth-Emmons Coupling of Triol Fragments 3

J\ o ores
o MeO\'FI,

 

 

 

 

 

, Base 0
M90 § A, I OTBS
| O + OR R‘ THF T I n
/0 or o ue e l \\ R
0R ‘
2a 3 0
Conditions Yield (%)
1) i) LiCI, THF, 3b, 5 min, ii) 0 °C; Et3N, warm to r.t, 30 min 211 (94)
iii) 0 °C, aldehyde, warm to r.t, 24 h; R=TES; R1=TMS
2) i) LiCI, THF, 3c, 5 min, ii) 0 °C; Et3N, warm to r.t, 30 min 212 (99)
iii) 0 °C, aldehyde, warm to r.t, 24 h; R=TBS; R1=TMS
3) i) LiCI, THF, 3d, 5 min, ii) 0 °C; EtaN, warm to r.t, 30 min 213 (90)
iii) 0 °C, aldehyde, warm to r.t, 24 h; R=TBS; R1=H
The Methylation Step

Of the many total syntheses and synthetic strategies towards fostriecin that have

been published, there are only four fundamental methods used to establish the proper
stereochemical relationship between the chiral centers at the C8 and C9 carbons (see
Figure IV -4). These four methods are: the addition of a methyl organometallic reagent to

the C8 ketone of 214; the addition of a vinyl organometallic reagent to the methyl ketone

of 60 which becomes C8 in the natural product; a Sharpless AD36 of a trisubstituted olefin
bearing a methyl group as one of the three substituents; and a catalytic asymmetric

cyanosilylation of methyl ketone of €8.29

98

Boger’s24 and .lacobsen’s25 syntheses are both examples of the addition of

organometallic reagents to set the relative stereochemistry at C8 and C9. These
approaches are complimentary in regard to whether the bond being made is part of the

carbon backbone (.Iacobsen’s)25 or not (Boger’s).24 In entry 1 of Figure IV-4 we see that

Boger’s synthesis adds a methyl group to the C8 ketone of compound 214, an

intermediate in which all the other carbons of fostriecin are already in place. In entry 2,

however, the situation is reversed; it is the carbon skeleton that is used as the nucleophile

on methyl ketone 60 to set the C8 and C9 relative stereochemistry.
The approach used by Shibasaki9 is also complimentary to that used by
Jacobsen,25 but not in the same way that Boger’s24 approach is. Both Shibasaki29 and

Jacobsen25 use a methyl ketone as a synthon, but Jacobsen25 adds the carbon skeleton of
an early fostriecin intermediate to the methyl ketone 60 continuing a linear sequence of
events. Shibasaki29 however begins his linear synthesis with this event via an asymmetric

cyanosilylation of methyl ketone 117 (see Figure IV—4). Ketone 117 is used as a lynchpin
in Shibasaki’s synthesis and the cyanosilylation of ketone 117 initiates a series of
transformations that will occur at both ends of tertiary alcohol 118 (see Chapter I Figure
I-29).

The most popular method seen in almost every other synthesis was a Sharpless

AD“. Falck synthesis exemplifies this is entry 326. Sharpless AD36 is attractive because

both the C8 and C9 stereogenic centers are set in one step.

 

_L--a

The approach taken by the Wulff group is similar to Boger’s since the C8 ketone

to be alkylated is already present in the carbon skeleton. The best result Boger obtained

with the addition of a methyl cerium reagent to the ketone 214 was a 3:1 diastereomeric
ratio of C8 epimers and with a 20:1 ratio of 1,2 to 1,4 addition products (Figure IV-4).24
This reagent was prepared by the addition of 18.2 equivalents of MeLi to 18.8
equivalents of anhydrous CeCl3. The CeCl3 had to be dried thoroughly before the lithium

reagent could be added. It was dried under vacuum at 80 0C-90 0C for 2 h, then at 130
00140 OC overnight. THF was added and the slurry stirred for 10 h, before titrating with
t-BuLi which removes any residual moisture (see experimental for details).24 Impassioned
to improve upon this selectivity, it was decided to try a more bulky methylating agent.
Addition of methyl titanium tris(isopropoxide) to ketone 211 resulted in an 80% recovery
of starting material (Figure IV-S). This outcome was not too discouraging because we
knew beforehand that Boger24 had also been unsuccessful at his attempt with this less
reactive reagents on ketone 214 (Figure IV-4). An attempt to reduce the steric bulk by
using dimethyl titanium bis(isopropoxide) was considered. In addition this reagent would
be expected to be a more reactive nucleophile. However, this reaction also resulted in

only recovery of starting material (Figure IV-S).

lOO

 

Figure IV-4 Strategies Used for Constructing the C, Chiral Center

 

   

 

 

 

 

0TBDPS
MeLi-CeCI3
THFrToluene
-78 °C, 10 mins
96% 3:1d.r
o’l\
.9
l 0 Meme/A 1) [Cp22r(H)Cl], 0112012
Q 0 MeZZn, -78 °C, 10 mins
55 H 6° meson, 45% over2 steps 55 M9 OTES
> 30:1
Falck2§
OBPS EtO cc M PPh oops . 9H 03133
O%\/'\ 2 ( e) 3 _ Etozc 8/ AD-me-B > £0sz
\\ benzene Q t-BuOl-I,H20 (1:1) M b’OH \\
215 0 Me 72 e 216
23 C, 12h 4 °C, 48h
73% 3:1d.r
Shiba§aki29

 

 

TMSCN 2 uiv , 5mol°/ cat (8) CN 7 Ste s
300W ( eq ) o : SHOW ___p__..

117 O ph\ O 113 Me "OTMS
93%, 85% ee

  

0“ O
i-Pr0\ ,,O
_ /Tl Ph
I-Pl‘O \

0

Given the unreactivity of the methyl titanium reagents, attention was turned to the

methyl cerium reagent prepared according to Boger’s procedure. The ketone 211 was
chosen and using the conditions outlined by Boger,24 this reaction gave only the recovery

of the starting material in 57% yield. The failure of the addition of the methyl cerium

reagent to ketone 211 was perplexing because the molecule containing the ketone used by

101

the Boger group differed from ours only in the side chain attached to C 1 1. This position is
three carbons removed from the reaction site. The Boger group accomplished the
methylation of ketone 214 after the side chain with the Z,Z,E-trienol at C12 was already
intact (Figure IV-4), while ketone 211 (Figure IV-S) contained a TMS protected
acetylene at C”. At this point a model system was devised using a-tetralone as the model
substrate (Figure IV-6). This reaction failed as well. It was hypothesized that these

failures were probably due to the inadequate preparation of dry CeCl3.

Figure IV-5 Attempts at Methylation

Methylating Reagent

 

 

 

 

I 0 0TBS :
Conditions
I S ms
OTES
O
211
Entry Methylating Reagent Conditions Results
1 ClTI(I-PTO)3, MCLK 16(1) '40 °C,].5 h, add 211, 80 % starting material
warm to r.t, 3 days recovered
2 ClTi(i-Pr0)3. MeLi(2eq) -30 °c,1o mins,add 211 83 % starting material
warm to r.t, 3 days without TES recovered
3 MeLi-CeCl3 -78 °C,10 mins, 0 °C, 57% starting material
10 mins, —78 °C, add 211, r.t recovered

 

102

There is one example of stereoselective methyl addition to a ketone at C8 in
efforts directed to a fostriecin synthesis and this is to be found in the synthesis of the
fostriecin diastereomer 1e from the work published by Just in 1988 (Figure H 1).26 In

their synthetic plan trimethylaluminium (AlMe3) was the methylating reagent they

gainfully employed. A 98:2 diastreomeric ratio in favor of the C8 R isomer 35 was
obtained in 60% yield. The problem was that they obtained the chelation controlled
tertiary alcohol product 35 using AlMe3 (Figure H l). The C9 and CH alcohol groups in
ketone 34 were protected with an acetonide which promotes this type of stereocontrol.
The C9-S stereogenic center of ketone 34 (Figure H 1) gave the correct C8-R
stereochemistry upon methylation. It was thus hypothesized that in order to maintain the
correct stereochemistry at C8, while using C9-R chiral center found in 220, 221 or 222, a

non-chelation controlled approach would be necessary.

103

 

Figure IV-6 Model Methylation Reactions

0

 

 

 

OH
Methylating Reagent
Conditions
218 219
Entry Methylating Reagent Conditions Results

1 MeLi-C603 '78 °C, 10 minst 0 °C, No reaction

10 mins, -78 °C, add 218, r.t
2 AIClMe2 -15 °C, add 218 No reaction

warm to r.t, 3 days
3 AlMe3 -15 °C, add 218 99 % 219

warm to r.t, 3 h

 

Inspired by the work of Just and O’Connor,26 the reaction of trimethyl aluminum
(AlMe3) and dimethylaluminium chloride (AlClMez) with a-tetralone were examined as

a model system (Figure IV-6). The results obtained using AlClMe2 were disappointing
since after three days there was only starting material as indicated by TLC. In contrast the
reaction of a-tetralone with AlMe3 was quite facile giving a 99% yield of 219 in three
hours. When this methodology was applied to our desired substrate ketone 211, a 48%
yield of tertiary alcohol 220 was obtained (Figure IV-7) as a 10:1 mixture of
diastereomers. This ratio of products could be obtained by integrating the hydroxyl
Protons which are singlets at 2.23 ppm (major isomer) and 2.26 ppm (minor isomer). The
Stereochemistry of the major diastereomer is assigned that shown in Figure IV-7 on the

basis of chemical correlation. This reaction is rather sluggish by comparison to the model

104

reaction with a-tetralone 218 since it required three days for a 60 % conversion to 220

(Figure IV-7).

Figure IV-7 Methylation of Ketone 211 with AlMe3

 

 

I 0 ores “”93. ;
CH Cl ,r.t, 3days
I \\ 2 2
TMS 48% yield
OTES 73% yield based on M "90H
0 recovered starting material 9
211 220
dr = 10:1

There were at least three other advantages to using AlMe3 as a methylating
reagent on ketone 211. First, no 1,4 addition product was observed as reported by
Boger.24 He reported a 20:1 ratio of 1,2 versus 1,4 addition products using the MeLi-
CeCl3 system discussed on page 57. This observation is consistent with the results
obtained by Just (Figure H 1), 26 in which compound 34 gave a 98:2 ratio of C8 epimers

with no 1,4 addition product being reported.
Secondly, the diastereomeric ratio of tertiary alcohol 220 was improved from a

3:1 ratio as reported for 50 Boger’s intermediate (Figure IV-3) to a 10:1 ratio in

compound 220. The TES protecting group on C9 in 211 (Figure IV-7) provides a more
hindered environment around the ketone at C8 than does the acetonide which protects the
C8 and C9 alcohols in compound 34 (Figure I-11). This difference in size may prevent

chelation and lead to the correct stereochemistry at C8.

The last, but certainly not the least, advantage was that the TMS group on

acetylene 211 was cleaved during this reaction (Figure IV-7). The product of this reaction

105

was expected to retain the TMS protection of the acetylene in 211. A deprotection of
TMS would have been required prior to the palladium cross coupling reaction as
indicated by model studies for this coupling in Figure IV-13. This result evades that step.

The loss of the TMS group was confirmed by the appearance of an acetylenic proton as a

doublet (J = 2.2 Hz) at 2.41 ppm, in the 1H NMR and the disappearance of the nine

trimethyl protons of TMS at 0.04 ppm.

Determining the Stereochemistry at C8

One of the advantages of using AlMe3 as a methylation reagent mentioned above

was the improvement in diastereoselectivity which was confirmed by proton NMR.
According to Figure IV-8 the non-chelation controlled Felkin-Ahn product is predicted to
be the diastereomer 220. The bulky TES protecting group should prevent chelation
control and thus the desired stereoisomer 220 should be formed. Boger obtained a 3:1
mixture of diastereomers for this step, with the Felkin-Ahn product 50 as the major

diastereomer (Figure IV-8).

106

Figure IV-8 Predicted Model for C8 Methylation

 

| 0 ores “M93: 4
CH Cl ,r.t, 3days
I \\ 2 2 .
TMS 48 0/0 YIGId
OTES 78 % yield based on
0 recovered starting material
211
TBSO/
OTBS , O
ores
/
I
211

 

In Chapter 3 under the section entitled Preparation of Phosphonate 3c a

hypothesis on improving the selectivity of the methylation step was discussed. The

primary conclusion was that bulkier silyl protecting groups on the C9 hydroxyl should
favor non-chelation controlled products and increase the selectivity at C8- Even though

we obtained a 10:1 ratio of product 220 when TES protects the C9 hydroxyl group, it was
predicted that an even higher selectivity should result when the TBS protected derivatives
212 and 213 are reacted with AlMe3 (Figure IV-9). To our chagrin only a 3:1 ratio of
diastereomers was obtained in 48% yield when ketone 212 was reacted with AlMe3 to

give 221. This unexpected result in selectivity could be explained if there is a competing

107

 

steric interaction in ketone 212 which disfavors conformation present in the non-chelated

transition state and favors the conformation that would be present in a chelated
intermediate. The two TBS groups on the hydroxyls of C9 and C11 could be responsible

for the observed selectivity. This steric interaction must also place the TMS protected
acetylene in a more hindered environment, preventing the cleavage of TMS as was
observed in the case of compound 211 the TES protected derivative. Computational
model studies on these intermediates need to be created in order to better understand

these results.

Thus far it was only hypothesized that the correct stereochemistry at C8 was
obtained using the Felkin-Ahn model. A more accurate method of determination would
be to subject ketone 211 to the conditions used by Boger24 enlisting a MeLi—CeCl3

complex as the reagent and comparing the results we would obtain to those obtained by
Boger. Since our predictions about the stereoselectivity so far had failed, resolving this
challenge before any further synthetic steps were attempted seemed imperative. Using the

protocol described by Boger was unsucessful even with the model study (see Figure IV-

6). Using commercially available anhydrous CeCl3 may have been the problem. At the

prompting of Professor Maleczka a method for preparing anhydrous CeCl3 was attempted
instead of using commercially available anhydrous cerium trichloride. Using cerium
trichloride heptahydrate (CeCl3.7H20) and heating it slowly under high vacuum for three

hours from 70 °C - 100 °C, then heating it overnight at 130 °C - 140 °C gave a white
powder. This was different in color to the anhydrous CeCl3 that was bought from Aldrich

Chemicals which was off-white. The powder was cooled to room temperature under

108

 

argon and stirred in dry THF for ten hours as reported by Boger.24 Following the rest of

Boger’s protocol gave no reaction. However, removal of the terI-butyl lithium (t-BuLi)
from the protocol gave the desired compound 221 in 90% yield and with a 3:1 dr. t-BuLi
was used by Boger to remove any traces of water. In the present case which uses 18.2
equivalents of methyl lithium (MeLi) in the reaction (see experimental for details) this
step seemed unnecessary.

At the time of this investigation only ketone 212 had been prepared and thus its
selectivity was determined first. It was suprising to find that it reacted to give a 7:1 ratio
of the respective diastereomers of 221 and in almost quantitative yield. Based on Boger’s

observation (Figure IV-4) this suggested that the major isomer obtained from ketone 212

using AlMe3 was indeed the one predicted and shown in Figure IV-9. Equally important
is that this result indicated that a change from TES to TBS on the C9 hydroxyl facilitates
higher selectivity when the MeLi—CeCl3 complex is used. What supports this hypothesis

even more was that later when ketone 211 was prepared and reacted with the MeLi-CeCl3
complex only a 3:1 ratio of diastereomers were obtained. This diastereoselectivity ratio is
identical to that reported by Boger on compound 21424 (Figure IV-4). Since Boger’s
product of the methylation alcohol 50 (Figure IV-4) has the identical environment around
C8 as compound 217 (Figure IV-6), we compared the spectra of diastereomers 217
(Figure IV-9) to 50 (Figure IV-4). The C9 proton and the hydroxyl proton at C8 were the
only two protons that would provide any useful information about the stereochemistry of
C8 and were thus chosen to do the analysis. Unfortunately the proton at C9 in both

compounds 50 and 217 are multiplets, hence Boger did not report any coupling constants

109

and a comparison could not be made. In addition the hydroxyl proton at C8 is a singlet at

2.83 ppm in tertiary alcohol 50 but a doublet at 2.83 ppm in tertiary alcohol 217 with a

coupling constant of 2.4 Hz. Only the location of these protons could be compared the

proton at C9 in 50 occurs at 3.65 ppm and in 217 at 3.63 ppm. This was not enough

evidence to confirm the stereochemistry at C8.

Figure IV-9 Methyl Addition to Ketones 211, 212 and 213

 

 

 

 

I O OTBS AlMe3 or MeLi-CeCl3
| Q
R1
OR
0
Entry RIR, Methylating Reagent Results
(Yield: d.r.)'

1 211. R=TES. R1= TMS AlMea 220,48 % 10:1 R=TES, R1: H
2 212, R=TBS, R1: TMS AlMe3 221, 48 % 3:1, R=TBS, R1= TMS
3 213, R=TBS, R,= H AIMe3 Experiment not done

4 211, R=TES, R1= TMS MeLi-CeCI3 217, 90 % 3:1, R=TES, R1= TMS
5 212, R=TBS, R1: TMS MeLi-CeCIa 221, 99 % 7:1, R=TBS, R1= TMS
6 213. R=TBS. R1= H MeLi-Gael; 222. 98 % 7:1, R=TBS, R1: H

 

a- Isolated yields. The yield of 220 was 73% based on unrecovered starting material.

110

tf'..'

Alkyne Deprotection and Silyl Migration

The deprotection of the TMS from the protected alkynes was only achieved on

intermediate 211. Alkyne 212 retained its TMS group regardless of the method used to
methylate C8 (Figure IV-9). Therefore a method to remove the TMS group from 221 had

to be developed. The table in Figure IV-10 below illustrates a number of methods
employed to achieve this transformation. All the reagents used gave a mixture of

products, with the desired product 222 being obtained in approximately 60% yield in
each case. A side product with almost the identical Rf value of 222 was also isolated in
approximately 40% yield. This side-product appeared to result from the migration of a
TBS from a protected hydroxyl to the unprotected hydroxyl at C8 in 222. Thus the likely

structure for this side product is either 223 or 224 (Figure-IV-12).

Figure IV-10 TMS Removal and Silyl Migration of Alkyne 221

 

   

 

 

+ x
Entry Deprotection Reagents Conditions Yield 222 + x Ratio 222:X
1 MeOH. K2003, H20 0 °C. 3h 95-100 % 1.5:1
2 MeOH, K2003, H20 0 °C-r.t, overnight 95-100 % 1.5:1
3 AgNO3, KCN, EtOH, H20 0 °C-r.t, 3.5h 95-100 % 1.5:1
4 Amberlyst resin, (CI‘ form) 0 °C-r.t, overnight 95-100 % 1.521

5 Amberlyst resin, (Cl‘ form) 0 °C-70°C, overnight 95-100 % 1.5:1

 

111

After analyzing the proton and carbon NMR of the unknown tertiary alcohol X it
became obvious that a TBS migration had occurred. With three alcohols and two TBS
groups there were only three possible structures. Alcohol 222 was already identified as

our desired substrate, with its proton and carbon NMR spectra closely resembling that of
the starting tertiary alcohol. The protons at C9 and C 1 1 in this structure are a triplet at 3.68

ppm (J = 5.2 Hz) and a multiplet at 4.38-4.52 ppm respectively; while in the starting
tertiary alcohol 221 they are a triplet at 3.67 ppm (J = 5.2 Hz) and a multiplet at 4.36-4.50

ppm respectively. In addition terminal alkyne 222 had been synthesized in an alternative
manner earlier from ketone 213 (see Figure IV-9 entry 6) and its 1H NMR spectrum
matched the major compound isolated in the alkyne deprotection of alcohol 221. Alcohol
222 retained the TBS groups located on C9 and Cl 1 hydroxyls (Figure IV-12). The other
product could only be the secondary alcohol 223 with the C8 and C I 1 hydroxyls protected
(R1 = TBS, R2 = H, R3 = TBS) or the secondary alcohol 224 with the C8 and C9 hydroxyls
protected (R1 = TBS, R2 = TBS, R3 = H) (Figure IV-12).

Of the two remaining possibilities for the unknown alcohol X, 224 the compound

possessing the TBS groups on C8 and C 11 hydroxyls protected would be a very attractive
intermediate since in the natural product the C9 hydroxyl is phosphorylated. The

migration producing 224 would eliminate one step in the synthesis, as this C9 hydroxyl

would no longer have to be selectively deprotected. Another advantage of alcohol 224 is
that several of Boger’s intermediates at the end of his synthesis possessed this
framework; subsequently a formal synthesis could be achieved at an earlier stage of our

synthetic plan and the intermediates could be readily compared. A last, but certainly not

112

the least advantage of gaining access to 224 would be that upon arrival at a formal
synthesis our intermediates would retain a higher selectivity at C8. Boger only obtained a

3:1 ratio in tertiary alcohol 50 after methyl addition to ketone 214 (Figure IV-3) whereas
compound 224 has a diastereomeric ratio of 7: 1.

With great expectancy, the 1H NMR of the unknown compound X was analyzed.
Protons H9 and H“ in the product were assigned as a doublet at 3.68 ppm (J = 9.3 Hz)

and a multiplet at 4.57-4.65 ppm respectively (see Figure IV-12). Disappointingly this

change in ppm values for Hll from 4.38 to 4.57 (0.19 ppm) implied that the TBS was

migrating from the C11 hydroxyl to the C8 hydroxyl. Boger reported two similar
intermediates in his synthetic approach; compound 226 bearing TBS groups on the C9
and C“ hydroxyls and compound 244 bearing TBS groups on the C8 and C 11 hydroxyls24
(Figure IV-ll). In these compounds H9 is reported as a multiplet with only a small

change in the ppm value from 3.75 ppm in 226 to 3.73 ppm in 244 being observed. Hll

showed no change retaining its value of 4.80 ppm in both compounds 226 and 244. This

chemical correlation strongly suggested that an alternative silyl migration was occurring.

Figure IV-ll Comparing H, and H11 of Alcohol 226 to Alcohol 244

   

226
rig-3.75 (m) rig-3.73 (m)

113

A more accurate way of determining the correct structure would be to compare
the free hydroxyl proton’s coupling constant to H9 and H11. If the hydroxy proton was

coupled it would appear as a doublet and thus its coupling partner could be identified as
the structure of X and could be assigned. Unfortunately, the free hydroxyl proton shows
up as a singlet on the 300 MHz Gemini NMR unit so this comparison could not be made.

Thus the structure of X was tentatively assigned as 223 on the basis of the chemical shift

observed for H“. Later however, the unknown alcohol X was proven to be 224.

Figure IV-12 Comparing H, and HH of Alcohol 222 to Alcohol 223/224

   

-OR-
Pig-3.68 (t , .l = 5.2 HZ) Pig-3.68 (d , .l = 9.3 Hz)

Alkyne Deprotection and Michael Addition

Amazingly not only did the alkyne deprotection of the tertiary alcohol 221 gave
unexpected results (Figure IV-lO), deprotection of ketone 212 provided an unanticipated
outcome as well. All the methods of deprotection listed above in Figure IV-IO were tried

on ketone 212, but none gave the desired TMS-deprotected product. All products that

were formed had lost the TMS group, however the a,B-unsaturated olefin protons at C6-
C7 had also disappeared in the 1H NMR spectrum. The product obtained when KZCO3 in

MeOH/H20 was used for the deprotection of ketone 212 was analyzed and upon careful

114

charaterization it was assigned as the Michael addition product 225 (see Figure lV-l3).
The nucleophile in this case was the methoxy anion adding to the 4 position of the (1,8-
unsaturated system. The other conditions listed in Figure IV-10, provided other
nucleophiles such as ethoxide and cyanide that were presumed to account for the other
unknown products by Michael addition as well. These other products were, however, not

characterized.

Figure IV-13 TMS Removal from ketone 212

0k 0k

 

l O OTBS MeOH, K2003, H20 ‘ I O OTBS
0 °C. 3h. 90 % 0M6
I R %
TMS H
ores ores
o o
212 225

Model Palladium Cross Coupling
With less than lO-steps remaining and a limited amount of substrate, it was
decided that a model study of the planned construction of the Clo-C13 triene unit was

essential. Just reported having over-reduced products when reducing the C14 internal

acetylene of the dienyne system present in compound 35 (Chapter 1, Figure H 1).32 Our
projected substrate 234 (Figure IV-18) has a slightly different arrangement of the
dienyne. The alkyne in 234 is at C12 instead of at C14 as in Just’s intermediate 35. We

believed, however, that there was still a strong possibility of getting over-reduced
products if the right reduction protocol was not chosen. Hence the model study outlined

in Figure IV-14 was established to look at this challenge and inpart in Chapter 2.

115

Subjecting the aldehyde 148 to Stork’s protocol64 for the synthesis of cis-iodo-

alkenes, the Z,E-iododiene 4a was obtained in 84% yield with a 9:1 ratio of separable cis
and trans isomers (Chapter 2 Figure II-S). Due to its light sensitivity, the major Z,E-
isomer of 4a was used immediately after purification. In the model study shown in Figure

IV-l4 the palladium cross coupling of 4a with l-butyn-3-ol gave an 87% yield of dieyne
227.79 The subsequent reduction“ proceeded cleanly to give the Z,Z,E-triene 228 with no

evidence of over reduced products or starting material as indicated by its carbon-13
spectrum.

It is known that the Zn/Cu-Ag reduction of acetylenes is influenced
environment around the acetylene.68 The model compound 227 possesses a free

propargyl alcohol (Figure IV-14). This is different from the desired substrate compound
220 (Figure IV-9) which has a TBS protected propargyl alcohol. To ensure that this
difference would not change the outcome of the Zn/Cu-Ag reduction, compound 227 was
protected with TBS giving the TBS protected dienyne 229 in quantitative yield. The
reduction of 229 was successful giving an unoptimized yield of 60% for the conversion

of dienyne 229 to the Z,Z,E-triene 230. No evidence for any over-reduced products could

be found in the crude 1H or 13C NMR.

116

by the

Figure IV-14 Model Study for Diene Triol Coupling

 

 

 

 

 

HO
. OH 0
/\/\/‘ /K 5/oPdCl2(PPh3)2 TBSO
/ / + =
7330 é pyrrolidine — //
4a H 152 87% 153 —
Zn/Cu-Ag TBSO
2:1Me0H/H20 , — _ OH
80% 154
HO
I OH o
W + /‘\ SAPdClz(PPh3)2 TBSO / TBSOTf' N83
T330 é pyrrolidine 7 _ / 100 %
4a H 152 — 153
87%
TBSO New“ TBSO _
reso // 2:1 MeOH/H20 — ores
— _ 60% 156
155
Palladium Cross Coupling

This successful model study for the alkyne reduction was encouraging and gave
us confidence that these steps could be extended to the actual substrates of the type 220
or 222 in high yields. If the alkyne 220 (Figure IV-9) was used the coupling product
would give the fostriecin core 231 shown in Figure IV-lS. A reduction of the internal
alkyne of 231 followed by conversion of the acetal to the lactone, a couple of protecting

and deprotecting steps, and finally the installation of the phosphate group on C9 would

give the natural product fostriecin (Figure IV-lS).

117

Figure IV-15 Plan for Completion of the Total Synthesis of Fostriecin

1) Zn/Cu-Ag
2) PPTS-EtOH
3) aq HCI

4) A92C03
5) TBSOTf

 

 

Me bras 2) HF, CHgCN-Pyr
232 3) NaOH

 

Ketone 211 gave the highest diastereoselectivity for the methyl addition to C8
which involved the use of AlMe3 as the source of the methyl. A 10:1 ratio of

diastereomers were produced concomitant with deprotection of the alkyne to give 220 in

good yields (Figure IV-9). The next best result was a 7:1 diastereoselectivity with ketone
213 (Figure IV-9) using the MeLi-CeCl3 complex to add a methyl nucleophile to C8- The

methyl addition occurred in almost quantitative yield but the preparation of ketone 213
had some low yielding steps (see Chapter III-Preparation of Phosphonate 3d, Figure III-
14). Naturally tertiary alcohol 220 would be the most desirable intermediate to bring
forward and the conditions used for the palladium cross—coupling model study were
applied. The coupling reaction of 220 with the dienyl iodide 4a (not shown) was
attempted under the conditions used in the model study (Figure IV-14). This reaction
failed to give any desired product. In addition the yields of 211 often dropped off. This

drop was due to TES cleavage from the intermediates made (see chapter III-Accidental

118

Preparation of Phosphonate 3c). A second problem was that when AlMe3 used as the

methylating reagent, the results were inconsistent, sometimes giving no desired product.

Given the failure to effect the coupling of 220 and 4a, the possibility of bringing
the analogous TBS protected derivative 222 forward in the synthesis was investigated.
I The coupling of 222 with 4a was attempted under the same conditions, but as with 220,
no evidence for the coupled product was observed even with a reaction time of 6 days
(Figure IV-16 entry 3). At this point it became clear that the conditions developed for the
model study would not work on the actual desired systems. This meant only a few
options were available: (i) use a different alkyne precusor; (ii) use a different diene; or
(iii) change the reaction conditions. In addition the use of the TES protected tertiary
alcohol 220 was ruled out because its synthesis was problematic.

Changing the conditions and/or using a different diene (4b or 4d Figure II-7) were
the easier options since there was already a moderate supply of tertiary alcohol 222
available. Hence a few different ligands for the palladium catalyst were screened as well
as different solvents and both diene fragments 4b and 4d and the results from this
extensive effort are presented in Figure IV-16. However, as the data in Figure IV—16
indicates, none of these variations lead to the formation of the desired coupling product
233. Thus, the only alternative that seemed reasonable at this point was that a different

alkynal substrate would have to be used as the vinyl halide’s coupling partner.

119

Figure IV-l6 Attempts at Diene Triol Coupling

R1
WOTBDPS

MR1=L4bR=Br

........................ -

   

 

 

-. OTBS 4d R = $1181.13
Me ”OH
222 R = H
221 R = TMS
R R1 Coupling Conditions Results
1) H I 5% Pd(PPh3)4; Diene decomposed and alkyne starting material recovered.
Pyrrolidine, r.t; 16h
2) H Br 5% Pd(PPh3)2CI2; Diene and alkyne starting material recovered.
Pyrrolidine, r.t; 16h
3) H Br 30% Pd(PPh3)2C|2; Diene decomposed and a new TBDPS protected alkynol
Pyrrolidine, r.t; 6d was recovered;
4) H Br 30% Pd(PPh3)ZC|2; Diene decomposed and a new TBDPS protected alkynol
Diisopropyl amine, r.t; 6d was recovered.
5) H Br 30% Pd(dppf)C|2: Diene decomposed and a new TBDPS protected alkynol
Pyrrolidine, r.t; 6d was recovered.
5) H Br 20% S-Phos‘“; 10% Diene decomposed and starting alkyne was
Pd(OAc)2, Pyrrolidine, r.t; 6d recovered-
7) H Br 40% Pd(P-lBu3)2CI2‘°5; Both starting materials were not recovered.
Pyrrolidine, r.t; 6d
3) H Br 10% Pd(PPh3)2Cl2; Diene and alkyne starting material recovered.
20% Cul, EtzNH, r.t; 16h
9) TMS Br CuCl (2.2eq), Bu3N; DMI Diene and alkyne decomposed.
r.t -120 °C; Overnight
10) SnBu3 Br —— Alkynal stannane was not made.
11) I 808% 20% PdtPPhah: Alkynal iodide made in 93% yield, but no desired product
75% Cul, DMF, r.t; 2d isolated.

 

120

Figure IV-l7 An Attempt at Alkyne Reduction of Iodide 229

AgN03, NIS. 93% yield
of Iodide 229

Et3N(1.5);NBSH(1.1)
THin-OPr (1:1), r.t; 14h

   

During the screening reactions shown in Figure IV-16 we noticed that one product
isolated resembled starting alkyne 222 but possessed a TBDPS group. It is believed that
an intermolecular silyl migration of the TBDPS group on the primary alcohol of the diene
fragment 4b was migrating to the alkyne 222. Proton NMR of this isolated product
showed peaks at 7.36-7.52 ppm and 7.66-7 .78 ppm in a 6:4 ratio, as well as a singlet at
1.28 integrating to nine protons. This product had lost one of its TBS group but had a
TBDPS group present. The product of this reaction was not fully characterized as there
are three alcohols that these two silyl protecting groups could possibly be protecting.
However, this gave us a pertinent piece of information, which is that in the presence of

excess base, a competing and faster reaction was occurring, namely, the shuffling of silyl
groups. If the tertiary alcohol at C8 of 222 could be protected or a precursor of 222 in
which the methylation step has not yet been accomplished is used, the cross coupling
might be successful. Indeed protecting the C8 alcohol of 222 would increase the number
of linear steps by two, but if an alkyne intermediate could be deprotected earlier then no
extra steps would be required.

As was mentioned previously, repeating the steps to prepare tertiary alcohol 220

gave side products which were the result of intermolecular scrambling of silicon

121

protecting groups which involved exchange of a TES group for 3 TBS group. One such
product was methyl ester 206 (Figure III-14), which was used to prepare the (1,6-
unsaturated ketone 213 (Figure IV-3). When ketone 213 was subjected to the coupling
conditions shown in Figure IV-18, both vinyl bromide 4b and vinyl iodide 4c gave a

compound that was tentatively assigned by 1H NMR as compound 234 in moderate

yields. Later the reaction was optimized to give a 91% yield of compound 234 when 4
equivalents vinyl bromide 4b were used and the mixture refluxed in pyrrolidine for six

days in the presence of 30% Pd(dppf)C12.

122

Figure IV-18 Attempts at the Palladium Cross Coupling

30% PdCI2(PPh3)2
Of

 

R 30% PdCl2(DPPF)
+ W ‘ I ' t t 23‘
\ 0TBDPS lSOpfOpy amine, r.
_ R = I; 47%
“R‘ I. (19“) R=Br; 41%

4b R: Br, (1eq)

 

-OR-
Br 30% PdCl2(DPPF)
, e 234
WOTBDPS Pyrrolicgp; 6d, r.t
4" (499) (freeze-thaw degassed)

 

 

Attempted Alkyne Reduction and Methyl Addition of 234

A quick glance at fostriecin core presented in 234 produced after the palladium

cross-coupling reaction of 213 and 4b indicates that it differs from Boger’s intermediate

ketone 214 (Figure IV-4) at C9 and C 1 1. There is a Z olefin at C 11 in Boger’s intermediate
and an internal alkyne in compound 234. In addition, there is a TES protected C9

hydroxyl in ketone 214, while at the C9 hydroxyl in compound 234 there is a TBS group.
Reduction of the internal alkyne of ketone 234 to the Z-olefin would allow for a direct

comparison of the effect of the C9 hydroxyl protecting group on the diastereoselectivity

of the methyl addition to the C8 ketone. In the preliminary study outlined in Figure IV-9

123

(entries 5 and 6) reaction of the MeLi-CeCl3 combination with ketones 211 and 212, gave
a 3:1 ratio with the TES protected C9 hydroxy ketone and a 7:1 ratio with the TBS

protected C9 hydroxy ketone. These results were encouraging and gave us confidence that

if the reduction of the alkyne in 234 were to occur to give 227, then we might expect to
see a higher diastereoselectivity in methyl addition to 227 than the 3:1 ratio seen by
Boger for the addition to 214 (Figure lV-4). So the Zn-Cu reduction was attempted on
234 with the method outlined in the earlier model study (Figure IV-14). Unfortunately,

no desired Z-alkyne was obtained. The material isolated proved to be a complex mixture.

Reversing the order of reactions with methyl addition to the C8 ketone of 234 first
followed by the Zn-Cu-Ag reduction was also attempted. Using either a MeLi-CeCl3

complex or AlMe3 to perform the methyl addition to C8 of ketone 234 failed to give any

desired product. Starting material and decomposed material were the only entities

recovered after a number of attempts.

124

Figure IV-19 Attempts at Methyl Addition and Reduction of 234

 

 

| 0 ores
l \\ Zn/CU-Ag
OTBS R2 2:1MeOH/H20
o
234
-OR-
| O 0TBS MeLi-CeCI3 (18 eq)
I \\ THF, -78°C
R2
OTBS
o
234
MOTBDPS
R2 = 2".“
Conclusion

The results obtained from the attempts to methylate ketone 234 or reduce its
internal alkyne were not only strange but also very discouraging and eventually led us to
change our synthetic approach. Compound 234 very closely resembles ketone 214

Boger’s intermediate. Ketones 211 and 212 (Figure IV-9) show clearly that whether the

C9 hydroxyl is protected by TES or TBS that methylation with MeLi-CeCl3 is very

feasible. It also shows that having an alkyne at Cll instead of a triene unit should not

prevent this methylation from being successful. Chapter 5 outlines a different approach to
fostriecin but this approach needs to be re-visited. Possible sources of error could be that

compound 234 has not been completely characterized to verify that it is the structure

presented in Figure IV-19 although this is unlikely. The MeLi-CeCl3 needs to be

125

prepared with a new bottle of CeCl3.7H20 since after one year this reagent is reported to

become inactive. '02

126

CHAPTER 5

The Formal Total Synthesis of Fostriecin

In the previous chapter a series of successful transformations were reported in
high yields to provide a structure tentatively assigned 234 and was to serve as an
advanced intermediate in the synthesis of fostriecin (Figure IV-18). A HWE reaction
between aldehyde 2a and phosphonates 3b, 3c and 3d (Figure IV—3); a palladium cross

coupling reaction with ketone 213 and diene 4b (Figure IV-18); and a methyl addition to
the C8 of ketones 211, 212 and 213 with high diastereoselectivity (Figure IV-9) are a few
of the key successful transformations. However, at the end of the chapter two
disappointing results were described with this approach: first the reduction of the C12‘Ci 3
triple bond in dienyne 234 failed to give any desired product and the methylation of the
ketone at C8 of this same compound gave starting material back. As outlined in chapter

one over ten syntheses of, or synthetic approaches towards fostriecin were reported in just
a short period of four years. This myriad of syntheses provided tactical solutions to

inherent challenges found in fostriecin’s construction. One such challenge was obtaining

the Z,Z,E- triene unit (Figure I-7). Inspired by the work of Jacobsen,25 Kobayashi30 and
Shibasaki29 a diimmide reduction was selected as the method of choice to construct the

ClzoCn cis-double bond. In their reports an alkynal iodide was reduced to a cis-vinyl

iodide and this iodide was coupled to stannane 4d (Figure V-l). Our current synthetic

approach did not incorporate an alkynyl or vinyl iodide and would have to be

127

reconstructed to adopt these intermediates. A new retrosynthetic analysis was designed in
which only the triol fragment would have to be changed. Figure V-l below outlines this

approach.

Figure V-l The New Retrosynthetic Analysis of Fostriecin

    

 

 

 

 

 

 

 

 

 

 

 

O
I 0 / / OH
5 / . .
_ k/ Stille Coupling
Homer-Wadsworth-Emmons '
Olefination \J OPO3HNa
Me ’OH
1
r . * i
Q’Pr l O OTBS
? Meo\||,0Me I
I O P — MOTBDPS
O OTBS SnBu3
23 H 0 39 4d
Lactone fragment Triol fragment Diene fragment

128

Figure V-2 Retrosynthetic Analysis of Trial Fragment 3e

 

 

 

 

 

 

 

 

 

 

 

 

 

 

O OTBS OTBS iodination and
MeOJ'l3 _ I (MeO)2POCH2Li 1 | reduction
MeO’ fi *’ '

MeO
OTBS OTBS
o 30 o 236
0TBS Evans reduction 0 OH acyl anion
\ pr°te°fi°n H \ addition
M o \ TMS ‘ 3’ S s \ E
e OTBS oxidative l\/l 191 TMS
O 205 deprotection
oxidation
H
O _.
e‘N NW asymmetric aldol MexN NJ\ + C80""'C‘.C”°Ccc:o
' 9’ — 0‘ o—
H TMS ' \_/ l \
Me‘ Ph Me 'Ph CO CO
166c 162a 169

As shown in Figure V-2, the TMS protected alkyne 205 is the last common
intermediate of the triol fragmens 3d and 3c in the retrosynthetic analysis. TMS protected

alkyne 205 could be converted to an alkynyl iodide 235 (Figure V-3) and this alkynyl
iodide reduced with p—nitrobenzenesulfonylhydrazide (NBSH)100 and Et3N to the cis-

vinyl iodide 236. Exposure of the methyl ester of vinyl iodide 236 to nucleophilic

addition of the dimethyl methyl phosphonate anion should provide triol fragment 3e.

Triol fragment 3e could then be coupled to stannane 4d and a HWE/4'93

reaction of the
product would give ketone 239. Ketone 239 is three steps away from a formal synthesis
of fostriecin. As outlined in Figure V-3, a methyl addition to the C8 ketone of 239
followed by a TBS migration from the C9 oxygen to the C8 tertiary alcohol and finally an

oxidation of an isopropyl acetal to the lactone would give intermediate 232 a compound

which was made by Boger24 (Figure V-3). This intermediate lactone 232 is two steps

129

away from the total synthesis of fostriecin and the ensuing steps were published by Boger

and coworkers“.

Figure V-3 Projected Formal Synthesis of Fostriecin

PPTS (0.25 eq). Ethanol

 

r.t. 3.5h
3:1 ratio of diastereomers

   

Moreops
R2 _

'm

1) PCI3, PMBOH.

2) HF. CH3CN-Pyr
3) NaOH

 

 

Preparation of Triene 239

The retrosynthetic analysis outlined in Figure V-l appears to be straightforward,
but even these minor changes in our strategy presented some challenges. As outline in
Figure V-3 the vinyl iodide 236 could be synthesized from alkyne 205 in 91% over two

steps. However when phosphonate addition to 236 was attempted with the anion of

CH3PO(OMe)2, no desired product was obtained and only decomposition of the starting

130

material was observed. Without dwelling too long on this result the phosphonate addition

was postponed until after the Stille50 coupling between vinyl iodide 236 and stannane 4d.

This step was reminiscent of that seen in Jacobsen’s,25 Hatakeyama’s28 and Shibasaki’s29

syntheses. In addition it was suspected that the conversion of methyl ester triene 237 to
phosphonate 238 in Figure V-4 would be a more facile feat than trying to convert iodide

236 to phosphonate 3e. This suspicion was due to the fact that Boger achieved a similar

transformation on an almost identical compound.”4 Nucleophilic addition was performed

on an aldehyde in his approach (see Chapter l-Figure I-13) and in our approach methyl
ester 237 was the target (Figure V-4). No reaction occurred when THF was used as a
solvent but exchanging this solvent for toluene gave an 88% yield of phosphonate triene
238.

Phosphonate triene 238 was then subjected to the conditions developed for the
HWE reaction with 2a exploited in Chapter 4. A near quantitative yield of 239 was

obtained for this step.

131

Figure V-4 Synthesis of Compound 239

 

 

 

T339 ores SnBu3 Pd(CH3CN)2C|2
MeO ’ I + T
— TBDPSO / / DMF
o 236 4d 76%
MeOTBS? OTBS CH3PO(OMe)2(4.0eq) A
— E, — 0TBDPS n-BuLi(3.75eq)
O overnight, 88%
Tesg OTBS
MeO\ ’ _ Et3N, LiCI, 2a
,P — — 0TBDPS =
M60 3 o 238 THF, 98%

 

Methylation of Ketone 239

Ketone 239 provided an opportunity to compare how a TBS protected hydroxyl at
C9 and a TES hydroxyl at C9 in ketone 214 (Figure lV-4) influence the
diastereoselectivity when the adjacent C8 ketone is reacted with MeLi-CeCl3. This
reaction appeared to be of a fickle nature because with ketones 212 and 213, a 7:1 ratio of
products were obtained, (Figure V-6) while with ketone 211 (T ES protected C9) only a

3:1 ratio was obtained. However, attempts at methyl addition to ketone 234 (see Chapter

4—Alkyne Reduction and Methylation Attempted) gave only recovered starting material.

132

Compound 234 is only different with ketone 239 at C12 where the internal alkyne of 234

is now reduced to a cis-olefin in 239 (Figure V-S).

Figure V-5 Structures of Ketones 234 and 239 Compared

 

Anxiously, methyl addition to 239 was attempted and gave a 98% yield, however
when the diasteroselective ratio was examined only a 1:1 ratio of compounds was
obtained. This result was very disappointing as it was expected that at least a 7:1 ratio
would be obtained based on the results of ketones 212 and 213 versus 211 (see Figure V-
6). In the methyl addition of ketones 212 and 213 the diastereoselectivity is greater than

twice of that obtained for ketone 211. Lowering the temperature of the reaction from -78
°C to -95 °C did not change the selectivity when 18.7 equivalents of CeCl3 and 18.2
equivalents of MeLi was used. The original procedure used by Boger also employs 18.7
equivalents of CeCl3 and 18.2 equivalents of MeLi to get a 3:1 ratio of products at C8.
Even though this is in great excess we decided to double the ratio of each of these starting
materials. When 37.4 equivalents of CeCl3 and 36.4 equivalents of MeLi were used, the

ratio of diastereomers of 240 increased to 3:1, which was the same ratio of C8

diastereomers obtained by Boger on the TES protected C9 ketone 214 (Figure IV-4).24

133

Figure V-6 Diastereoselectivities of Methyl Addition to Ketones 211, 212, 213 and

 

 

 

 

 

 

 

 

 

239
9*
I ' O OTBS AlMe3 or MeLi-CeCI3
I %
R1
OR
0
Entry RIR, Methylating Reagent Results
(Yield: d.r.)
1 211. R=TES. R1: TMS AIMea 220, 78 % 10:1 R=TES, R1: H
2 212, R=TBS, R1= TMS AlMea 221. 48 % 3:1, R=TBS. R1= TMS
3 213, R=TBS, R,= H AlMe3 Experiment not done
4 211, R=TES, R,= TMS MeLi-CeCIa 217, 90 % 3:1, R=TES, R1= TMS
5 212, R=TBS, R1= TMS MeLi-CeCI3 221, 99 % 7:1, R=TBS. R1= TMS
6 213. R=TBS. R1= H MeLi-CeCIa 222, 93 % 7:1. R=TBS. R1= H
MeLi-CeCI3 (18 eq)
THF. -78°C
98% d.r = 1:1
MeLi-CeCI3 (36 eq) 4
THF. -78°C
98% d.r = 3:1
- MOTBDPS

R2

I

134

Possible Causes for the Erosion of Selectivity

Even though these reactions have been unexplored mechanistically, it is believed

that a combination of two factors led to an erosion of selectivity for the methyl addition
to the C8 ketone from a 3:1 ratio in TBS protected alcohol 214 (Figure IV-4) to a 1:1 ratio

in the TBS protected alcohol 239 (Figure V-6). These factors are a steric interaction

between the TBS groups at C9 and C11 of triene 239 and the special orientation of the
triene unit C12'C17-

A steric interaction between the C9 and CH TBS groups is strongly suggested as a
reason for the erosion of selectivity because in ketone 214 (Figure IV-4), where the only
difference from 239 (Figure V—S) is that C9 is TES protected a 3:1 ratio is obtained after
reaction with MeLi-CeCl3. If this is the only difference between the two structures 214

and 239, then the interaction between the two TBS groups must play a role in the erosion

of selectivity observed.

Another piece of evidence that suggests that the interaction between the C9 and
C11 TBS groups is a strained one is that when tertiary alcohols 221 (Figure IV-lO) or 240
(Figure V-12) are reacted under basic conditions there is a TBS migration from one of the
secondary alcohols to the C8 tertiary alcohol. This is unusual since tertiary alcohols are

inherently more sterically encumbered than secondary alcohols. In order to faciliate a

migration of this sort, some competing steric interaction must be present. The C9 and C 11

bis-TBS protected alcohols seem to facilitate this type of migration, indicating that this

arrangement is indeed a sterically encumbered one.

135

The steric argument presented above only accounts for the change in selectivity in
the methyl addition to the C8 ketones in 214 and 239 from 3:1 to 1:1, respectively. The
change in selectivity between ketones 212/213 (Figure IV-9) and 239 (Figure V-6) upon
methyl addition with MeLi-CeCl3 is much more dramatic. When ketones 212/213 which
are devoid of the (Z,E,E)-triene unit are methylated with MeLi-CeCl3 a 7 :1 ratio of C8
epimers in the products 221 and 222 is observed (Figure IV-9). This is a seven times a
greater selectivity at C8 than is observed when ketone 239 is methylated with MeLi-CeCl3
under the same reaction conditions (Figure V-6). The difference between these structures
lie in the nature of the carbon chain attached to C“. In ketone 212 there is a TMS

protected alkyne, and in ketone 213 there is a terminal alkyne and in ketone 239 there is a
(Z,E,E)-triene. At a glance it might not be obvious what the reason for the erosion of
selectivity is. When ketone 239 is drawn on paper the triene unit appears to be in the
plane of the paper, however when a model of 239 is built, one of the more stable

conformers appears to be one whose triene unit partially blocks the si-face of the
carbonyl. The model for C8 methyl addition to 239 shown in Figure V-7. The face from
which the methyl addition needs to occur in order for the correct C8-R stereochemistry to
be obtained is the si-face as shown in 239b (Figure V-7) which is drawn according to the
Felkin-Ahn model.‘°5 With the re-face blocked by a bulky TBS group and the si-face
blocked with a rigid triene unit, the selectivity at C8 would be expected to be low. In the
case of ketones 212 and 213 where there is no triene unit attached at C11 the si-face would

not be blocked hence a higher selectivity for methyl addition to ketones of the type 212

and 213 would be expected.

136

A change in selectivity brought about by increasing the ratio of MeLi-CeCl3 to

ketone 239 is unusual. The fact that even using 18 equivalents of MeLi-CeCl3 complex

seems to be necessary, implies that cerium or lithium may have more than one interaction
with ketones of type 214 (Figure W4) and 239. They are certainly other oxygens present

in ketones 214 and 239, which despite a very sterically uncompromising environment
might still be able to coordinate to cerium in the MeLi-CeCl3 complex. In addition, there

are five olefins that could also possibly coordinate to the cerium. Futhermore, the large
excess of lithium may also play a role in coordination and or aggregation. Any of these

double bonds could cause ketone 239 to have a different orientation spacially than that

depicted in Figure V—7, where only the C8 carbonyl coordinates to the metal. At this point

however this reasoning is highly speculative and does not provide a clear explanation for

the observed results.

137

Figure V-7 Predicted model for C, Methylation

 

OTBS

   

 

These hypotheses involving metal coordination and/or aggregation are supported

by the observation that when ketones 211 and 212 were reacted with AlMe3, they gave

completely different selectivities than those resulting from a MeLi-CeCl3 protocol. As

can be seen in Figure V-6 entry 1, the TES protected a-hydroxy ketone 211 gave a 10:1

138

selectivity when AlMe3 was used and only a 3:1 selectivity when MeLi-CeCl3 complex

was used (entry 4). In addition, a 7:1 selectivity was obtained with the TBS protected (1-
hydroxy ketone 212 (entry 5) upon reaction of the MeLi-CeCl3 complex but when AlMe3

was used the selectivity dropped to 3:1 (entry 2). A detailed mechanistic investigation

needs to be done to better understand the nature of methyl addition using AlMe3 vs.

MeLi-CeCl3 on compounds of the type 211 and 212 and 214 and 239.

Acetal Removal and TBS Migration
After successful methyl addition to ketone 239 was achieved with a 3:1 ratio of

inseparable C8 epimers, oxidation of the isopropyl acetals 240 to the lactones 241 became

the next feat (The major epimer at C8 is indicated in all figures). This step was essential

at this point in our synthetic approach because Boger had shown that after this
transformation the diastereomers could be separated (Figure I-l4). In order to prevent
having to characterize any future intermediates as mixtures, this step was attempted prior
to silyl migration. As can be seen in Figure V-8, using mild acid followed by a

98101

Fetizon oxidation gave only decomposed material. This procedure was used by

Boger on the ethyl acetals 245 (see Figure V-8) and gave a 40% yield for the two steps.

139

Figure V-8 Acetal Removal With HCI and Ag2CO3

 
    

1) aq HCI (0.5 N)
ores ores R2 =

/ 2) Agzco3. benzene

 

Me ”OH
240 R1 = lsopropyl (Wulffs intermediate 0%)
245 R1 = Ethyl (Bogers intermediate 40%)

R2‘ Mr“

 

The removal of the isopropyl group was also attempted using pyridinium para-
toulene sulfonic acid (PPTS) in an acetone/water mixture but upon analysis of the
reaction mixture after the oxidation, no desired product was obtained.

Eventually some success came when isopropyl acetals 240 (3:1 diastereomeric
ratio) were subjected directly to a pyridinium chlorochromate (PCC) sodium acetate

mixture (NaOAc). A 40% yield for the lactones 241 was obtained (characterization was

based on 1H NMR only) and could be separated. Lactones 241 were not very stable and

were used immediately for the next step (Figure V-9).

Figure V-9 Acetal Removal With PCC

PCC. NaOAc ‘
CHZCIZ, 40%

 

   

241

140

TBS Migration

The migration of TBS from the C9 oxygen to the C8 tertiary hydroxyl is essential
because in the natural product the C9 alcohol contains a phosphate group. This migration
was not accomplished before but was suggested by an observation made by Boger on a
related transformation in his synthetic investigations.24 Alcohol 246 could be protected
selectively either at the C8 or C9 alcoholldepending on the temperature at which the
reaction was done (Figure V-lO). At -78 °C hydroxyl protection at C9 was favored but at
-20 °C protection at C8 was observed. It is possible that silyl migration occurs from the C9
oxygen to the C8 alcohol at higher temperatures prior to quenching.

Applying these conditions to lactones 241 gave no desired product but instead just
decomposed materials. Enlisting a variety of bases such as Et3N, imidazole, NaH and

amberlyst A-26 (chloride ion form) all gave no desired product. The products isolated

were almost all devoid of the a,B-unsaturated olefin in the lactone ring.

Figure V-10 Selective TBS Protection

 
  

9H OTBSRZ 1)TBSOTf,2,6-lutidrne : 9R1 OTBSRZ

/ -7e°c or -2ooc

 

    

Me "’OR
246 232 R =TBS, R1 =H ( -20°C)
241 R =H, R1 =TBS (-78°C)

MOTBDPS
R2 "

'm

141

Eventually it was hypothesized that a Michael addition to the lactone of 246 was a
competing reaction to the silyl migration. If this were true, the order of PCC oxidation
and TBS migration would have to be reversed to avoid this problem.

Unfortunately, reversing the order of reactions posed a very crucial problem. PCC

is widely used as an oxidizing agent for the conversion of 1° and 2° alcohols to the

corresponding aldehydes and ketones.102 Performing the migration on isopropyl acetals

240 (Figure V-9) first would mean that the C9 secondary alcohol that is prone to
oxidation would be unprotected. Instead of just cleaving the isopropyl group and
oxidizing the resulting lactol to lactone 232, the C9 alcohol would also be oxidized
(Figure V-ll). This problem could be solved if the HCI hydrolysis followed by

98101

Fetizon oxidation were applied to isopropy acetals 242 (Figure V-l 1).

142

Figure V-ll Migration of TBS Followed by Lactone Preparation

 
    

MeOH,K co ,H o
I gresoresaz 2 3 2

8 / -20°C. 2h, 25 %

Me 09H 11 12 50% based on SM recovered

240

 

PCC (1.5 eq), NaOAc (4 eq) 1) aq HCI (0.5 N)
........................... - -OR- --
2) A92CO3, benzene

 

MOTBDPS
R2 '-

This approach would leave the C9 alcohol of 242 unharmed but would selectively convert

the isopropyl acetals 242 to lactones 232. It was already shown however that with

0

isopropyl acetals 240, HCI hydrolysis followed by Fetizon’sl 1 oxidation failed (Figure

V-8). Contrastingly however, with ethyl acetals 245 this transformation was successful
(Figure V-8). If isopropyl acetals 242 could be converted to ethyl acetals 244 then the
hydrolysis and oxidation would be feasible, and that would constitute formal synthesis.
Thus far, TBS migration was only attempted on lactones 241 and this was
unsuccessful, so before an attempt to exchange the isopropyl group on 240 for an ethyl
group was made, a series of reactions were screened to actualize the silyl migration on

isopropyl acetals 240 (Figure IV-12).

143

As is shown in Figure V-12 there were at least two sets of conditions that gave the
desired compounds 242, NaH in THF and KZCO3 in MeOH and H20. A mixture of at
least five fractions was seen on TLC and the separation of these fractions was
painstakingly difficult. Each preparative TLC plate was buffered with Et3N and all work-

up procedures and isolation done in the dark. Isopropyl acetals 242 could be isolated as a
3:1 diastereomeric mixture in 25% yield. The other products isolated were thought to be
TBS deprotected diols and ClS'Cl6 trans-isomers 243. Only the Cis'Cis trans-isomers
could be isolated long enough to obtain sufficient data for characterization. Separation of
acetals 242 could be achieved here but this separation was postponed until the next step
because Boger reported an easier separation occurred with lactones 232.24 The isopropyl
acetals 242 were used immediately for the next step because these products are very
unstable. Even freezing in benzene under an argon atmosphere in the freezer (-30 °C) was
not enough to keep acetals 242 pure. Interestingly, the isopropyl acetals 240 could be
stored for relatively long periods (a month) under these conditions without significant

decomposition.

144

Figure V-l2 Screening Conditions for TBS Migration on Alcohol 240

 

MOTBDPS
R2 '-

- I

 

 

Entry Bases Conditions

Results

 

1 NaH, 7eq, THF -78 °C. 3h

2 MeOHzHZO, (10:1), K2003 (2.5 eq) -20°C-2h

25%, 40%SM Recovered

25%, 50%SM Recovered

 

3 Et3N (2 eq). CHZCIZ 0 °C-r.t. 3.5h 80% SM recovered

4 lmidazole (1 -1 Seq), THF 0 °C-r.t. overnight 20-80% SM recovered

5 2,6-lutidine (4 - 8 eq). CHZCI2 0 °C-r.t, overnight 20-80% SM recovered
Migration Anomaly

In order to confirm that the migration from the C9 2° alcohol to C8 3° alcohol had

taken place, we compared the spectral data of products 242 to the isopropyl acetal X

isolated earlier (Figure IV-lO). As reported in Chapter 4, this product was assigned as the

alcohol 223 which had its C8 and C9 hydroxyls protected with TBS and its C 11 2° alcohol

unprotected. Structure 223 was supposed to be derived from the migration of TBS group

from the oxygen at C 11 to that at C8. The result obtained from the comparison of spectral

data of 242 with 223 was disturbing initially as the chemical shift changes for 242 were

identical to those described for 223 in Chapter 4. The proton H9 had not shifted and the

145

proton Hll had shifted downfield by 0.18 ppm. Refusing to believe that Boger isolated
one thermodynamic product and our group another, more concrete information was
sought. One breakthrough occurred when running a 1H NMR of the minor isomer of

alcohol 242 on the 500 MHz NMR instrument. On the 300 MHz NMR instrument, the
unprotected hydroxyl group was a singlet but on the 500 MHz NMR, it was a doublet.

This meant that homodecoupled experiment could now be accomplished to determine
which proton (H9 or H11) was coupled to the unprotected hydroxyl. It was surprising but
delightful to find that when the proton assigned as H9 was irridated the J coupling of 2.9
Hz for the hydroxyl proton disappeared giving a broad singlet at 2.71 ppm. When the
alcoholic peak was irradiated the H9 doublet of doublets at 3.60 ppm with coupling

constants of 11.7 and 2.9 Hz sharpened in appearance significantly, and coalesced to a

doublet J = 11.7 Hz. This forced us to re-examine the previous assignment of compound

223. When the 1H NMR spectra of 223 was taken on the 500 MHz NMR instrument,

instead of a doublet for the C9 hydroxyl as in the case with acetal 242, a singlet at 2.75

ppm was observed. Hence no homodecoupled experiment could be done.

146

Figure V-l3 1H NMR of TBS Migrated Products 224 and 242

 

 

 

 

 

 

 

 

 

‘II

A ., j‘ 3
H5 H9“??? , f ; 1'

 

 

! :
9°; 1‘ 1‘ 1‘
.‘ Riv} V V K.“ \N“

2 j. g ‘
who-“0" ‘W’Hl" ‘W‘ W‘ ”V Luna—l MVWI‘N'WV

147

deprotection of the TMS from acetal 221 was quenched after 1 hour instead of allowing
the reaction to run overnight. The 1H NMR of 247 is very similar to that of the compound

assigned 223. Other than the hydroxyl shifts, only an extra nine protons at 0.12 ppm
corresponding to the TMS and the disappearance of the terminal alkyne proton of 223 at
2.35 indicate any major differences. In addition a low resolution mass spectrum

confirmed the molecular mass of 247 (see Chapter 5 experimental for details). A
homodecoupled experiment performed on acetal 247 confirmed that indeed H9 and the

hydroxyl proton were coupled. When the doublet at 8 = 2.89 (J = 2.0 Hz) corresponding

to the hydroxyl proton was irradiated the doublet at 5 = 3.71 (J = 10.7 Hz) corresponding
to H9 gave a sharper appearance. When the situation was reversed and H9 was irradiated
the doublet at 2.89 coalesced to a singlet. The proton Hll remained unchanged
throughout these experiments and when it was irradiated the appearance of the hydroxyl
proton or H9 was unaffected. TBS migrated product 223 in chapter 4 was now reassigned

as secondary alcohol 224 (Figure IV-l2), based on the evidence suggested here.

148

Figure V-l4 Comparing Secondary Alcohols 224, 247 and 242

   

Amberlyst A-26 (Cl' form)

 

0 °C-r.t 1h

 

Hydroxyl H - 2.59 (s) Hydroxyl H - 2.71 (d, J = 2.9 Hz) Hydroxyl H - 2.89 (d. J = 2.0 Hz)

H9-3.68 (d . J = 9.3 Hz) H9 - 3.60 (d. J = 11.7 Hz) H9-3.71 (d . J = 10.7 Hz)

NVOTBDPS
R2 =

m

The assignment of compound 247 was also confirmed by HMBC and HMQC 2D
experiments (Figure V-15). The results obtained from the HMBC only showed partial

correlations but the evidence obtained was enough to confirm the structure of alkyne 247.

Identification of C8 provided the most concrete evidence. This carbon is one of three
quaternary carbons which do not appear on the HMQC spectrum, C8, C12, and C13. Of
these three carbons only C8 can show an HMBC crosspeak to a vinylic proton (H6), C12
and C13 are well out of the three bond coupling range. If the arrangement of alkyne 247 is
as shown with the C9 hydroxyl unprotected, then not only would H6 couple to C8 in the

HMBC experiment but so would the hydroxyl proton H15. When H6 and H,5 were

149

analyzed both showed correlation to a quaternary carbon at 77.30 ppm which was
identified as C8. The arrangement of silyl protecting groups seen in alkyne 2A8 (Figure

V-15) does not allow for three bond coupling to any carbon around 77.30 ppm which will

in turn be coupled to a vinylic hydrogen. Additional evidence to confirm the structure of
alkyne 247 is provided by the methyl group Cl4 attached to C8. The proton Hl4 is a
singlet on the HMBC data shows crosspeaks to two carbons one at 76.60 ppm and the
other at 134.86 ppm. The carbon at 134.86 ppm is C7 corresponding to an sp2 carbon
which is three bonds away from H14. Therefore a 6 value of 76.60 ppm must correspond

to C9 (Figure V-15). The proton we assigned as H9 in compound 247 is a doublet at 3.71

ppm and it was attached to a carbon at 76.60 ppm in the HMQC. Proton H9 was important
because it was shown by homodecoupling experiments to be coupled to the unprotected
hydroxyl proton. These experiments are described in the previous paragraph. Since H9
which is coupled to the free hydroxyl is attached to a carbon at 76.60 and this carbon

shows a crosspeak by HMBC analysis to H14, it can be safetly concluded that alcohol 247
(and not alcohol 248) is the correct structure (Figure V-l4). In addition, Hl4 showed no
crosspeaks to C5 at 66.50 ppm or Cll at 61.20 ppm which were the other two carbons
seen in the HMQC that bore one hydroxyl.

Unfortunately H5, Hg and H11 showed no crosspeaks to any carbons in the HMBC

experiment. Nevertheless the results that were obtained was enough to confirm the
structure of 247. The assignment of 247 also allows for the confirmation of the
assignment of compound X (Figure IV-12) as alkyne 224 and triene 242 (Figure V-14) by

correlation of proton chemical shifts and multiplicities.

150

As was metioned earlier, TBS migration from a secondary to tertiary alcohol to
give compounds such as alkyne 224 and triene 242 was suggested by Boger.24 This
migration was not accomplished before but was suggested because of an observation
made by Boger on a related transformation in his synthetic investigations (Figure V-
10).24 Alcohol 246 could be protected selectively either at the C8 or C9 alcohol depending
on the temperature at which the reaction was done (Figure V-10). At -78 °C hydroxyl
protection at C9 was favored but at -20 °C protection at C8 was observed. As seen in
Figure IV-10 a 15:1 ratio of compound 222 to compound 224 is observed when the TBS
group migrates from the C9 oxygen to C11 oxygen. If this migration is reversible and the

products 222 and 224 are in equilibrium with a ratio of 1.5:] respectively, then upon

subjection of alkyne 224 to the migration protocol a similar outcome would be expected.

In order to confirm Boger’s hypothesis, alkyne 224 with TBS protected hydroxyls at C8

and C11 was subjected to the migration conditions we developed earlier (Figure IV-10).

As expected a 1.5:] ratio of alkynes 222 to 224 was obtained, indicating that indeed the
silyl migration is reversible and in this case that the alkynes 222 and 224 are in

equilibrium.

151

 

Figure V-15 Partial HMBC and HMQC Analysis of Alkyne 247

 

Proton HMQC Attached C HMBC Correlation
H5-4.38-4.48 ppm Cs -66.50 ppm C-, (134.86 ppm)
H6-4.38-4.48 ppm Cs -128.06 ppm Ce (77.30 ppm)
H7-5.64-5.84 ppm 07 -134.86 ppm C5 (66.50 ppm)

H9-3.71 ppm C9 -76.60 ppm

H1o-1.72-1.84 ppm C10 ~39.40 ppm C11 (61.5 ppm-2 bond coupling)
H14-1.30 ppm C14 -29.35 ppm C; (134.9 ppm); C9 (76.60 ppm)
H15-2.90 ppm C3 (77.30 ppm); C10 (39.40 ppm)

 

Formal Synthesis

With this newly found knowledge in hand the exchange of the isopropyl group in
acetal 242 for ethyl groups in acetal 244 was attempted. Ethyl acetal 244 was prepared as

a 3:1 mixture of diastereomers from the 3:1 mixture of diastereomers of isopropyl acetal
242 in 92% yield to provide a formal total synthesis of fostriecin (Figure V-l6). The 1H

NMR and IR spectral data of 244 matched the spectra of an authentic sample (also a 3:1

152

mixture) provided by Boger. The remaining steps for the conversion of 244 to fostriecin

as reported by Boger are shown in Figure V-16.

Figure V-l6 Formal Synthesis of Fostriecin

PPTS (0.25 eq), Ethanol

 

r.t, 3.5h
3:1 ratio of diastereomers

   

MOTBDPS
R2 _

“m

1) PCI3, PMBOH,
Aq HCI ;A92003

2) HF. CH3CN-Pyr
3) NaOH

 

 

Attempts to Prepare Lactone

During the synthesis of 244 it was realized that compounds containing the Z,Z,E
triene intermediate are very unstable and decompose readily even when stored carefully
under argon at cold temperatures. What seemed to be even more unstable were

intermediates which also included the a,[3- unsaturated lactone. When ethyl acetal 244
was hydrolyzed and oxidized to the lactone, the diastereomers could be separated but 1H

NMR showed the product with solvent peaks. When placed under vacuum (0.2 mmHg)

153

and carefully wrapped in foil overnight the 1H NMR showed decomposition with new

peaks at 6 = 3.60 and 4.05 ppm as well as a messy olefinic region. These were all peaks
that were not present a few hours earlier. Much care and precision needs to be taken
when handling this intermediate.

This concern of stability made it very difficult to obtain clean carbon spectra for

intermediates 232 and 244. Other peaks began to develop while the 13C NMRs were

being taken. More material would need to be prepared in order to get clean 13C NMR

spectra carbons to complete the characterization of lactone 232 and ethyl acetal 244.

Conclusions

Our ultimate goal was not just the formal total synthesis or the synthesis of the
natural product fostriecin itself, but also the synthesis of some closely related analogs.
Fostriecin itself is somewhat unstable and when used in the clinic must be stored frozen
in a buffer. Our experience in trying to make fostriecin indicates that the two main
sources of instability are the anti—unsaturated lactone and the Z,Z,E triene moiety. Boger’s
SAR studies discussed in Chapter 1 indicate that indeed the lactone is one of the most
reactive fragments and is essential for such high protein phosphotase selectivity. The
triene moeity, however, he assigns as just being a hydrophobic tail with not much
significance in terms of the molecule’s activity. For this reason we believe that
intermediates of the type 234 and 243 are precursors to equally active analogs while at
the same time would provide much less sensitive alternatives. A general scheme for these

and other proposed intermediates is outlined in Figure V-17 below.

154

Figure V-l7 Fostriecin Analogs

    

3' OPO3HN8
Me ’OH

Derived fromDienyne (R,R,R)-234

 

R ‘= Me‘NAN i H R
\—l
M6 'Ph
0 (— H bond to Ar9214
OHe— No Role
OH /

   
 

Hydrophobic tail in Substrate
' OPOaHNa binding site
Substrate Thr Mimic —-{ M9 0“ b Substrate Phosphate Mimic

 

155

CHAPTER 6
EXPERIMENTAL PROCEDURES

Experimental Data for Chapter 2

o NaH. TBDPSCI O
WOH 90% t WOTBDPS

51 (R)-142

 

'(R)-(Tert-butyldiphenylsilyl)glycidol 142. A 500 mL round bottom flask was
charged with (S)-glycidol (2.00 g, 27.0 mmol) and dissolved in 100 m1. CH2C12.DMAP

(132 mg, 1.08 mmol, 4 mol%) and triethylamine (3.00 g, 29.7 mmol, 4.13 mL), were
added and the flask was placed under argon atmosphere. Tertbutyldiphenylsilyl chloride
(8.9 g, 32.0 mmol, 8.42 mL), was added neat via syringe. The reaction turned cloudy
after 1 hr, and was stirred for 24 hr.

The reaction was quenched by adding water (20 mL), poured into a separatory
funnel, and the organic layer was washed with saturated NH4C1 solution (2 x 20 mL),
water (3 x 40 mL), and brine (1 x 40 mL) and then dried with MgSO4, and concentrated
to a pale yellow oil. The oi1 was purified by simple distillation (140-150 oC/0.2 torr) and
chromatography on silica gel (9:1 pentane/ether, UV visualization — faint spots), which
gave the product at Rf = 0.50 and TBDPS-OH at Rf = 0.2. The product (R)-(7.6 g, 0.051

mmol) was isolated in 90% yield as a thick colorless oil.
1H NMR (400 MHz, CDC13): 6 1.06 (s, 9H), 2.61 (dd, 1H, J =2.7, 5.2 Hz), 2.74

(dd, 1H, J = 4.1, 2.4 Hz), 3.12 (m, 1H), 3.71 (dd, 1H, J = 4.8 11.8 Hz), 3.85 (dd, 1H, J =

156

3.2, 11.8 Hz), 7.37-7.43 (m, 6H), 7.67-7.70 (m, 41-1); 13’C NMR (100 MHz, CDCI3): 6
19.24, 26.75, 44.45, 52.26, 64.31, 127.71, 129.73, 133.30, 135.56, 135.62; IR (neat film

on NaCl): 2959 (m), 2857 (m), 1428 (m), 1113 (s), 702 (s) cm'l; EI mass spectrum m/z
(% rel intensity) 255 M+ -57 (50), 225 (100), 211 (24), 183 (74), 177 (40), 135 (8), 117
(43), 105 (17), 91 (11), 77 (15); bp 140-150 oC/0.2 torr, Rf = 0.5 (9:1 pentane/ether), [ctlD

—3. 13 (c 1.05, c1103).

HQH

 

Hokoreops
o 2 mol% Catalyst (160) (8)459
Woreops ;
43% ; >99%ee Worepps
(R)-142

Preparation of R-Gylcidol Silyl Ether by HKR-(R)-1429°. Pre-catalyst (1S,2.S')-(+)-1,2-
cyclohexanediamino-N,N-bis-(3,5-di-t-butyl salicyclidene) Co (11) 160 (0.7 mg, 0.0012
mmol) and AcOH (0.32 mL, 0.0056 mmol) was added to neat racemic glycidol (17.46 g,

0.056 mol). The reaction flask was open to air, and after 10 mins the orange color turned

to dark brown. The solution was cooled to 0 °C and 0.6 mL of THF and 0.55 mL (0.028
mol) of H20 were added. A septum was then placed on the flask and a steady air-flow
was maintained through the flask and out to a bubbler. The solution was warmed to
ambient temperature over two hours and kept at that temperature for 30 h. All of the THF
was removed via rotary evaporator and the H20 was removed via short path distillation
under vacuum (0.02 mmHg). The product was then distilled over at 150-158 c’C under

vacuum (0.02 mmHg). Column chromatography with 2% EtOAc in pentane of the

157

distilled product gave 43% of R-gylicdol silyl ether 142 with >99% ee. All the spectral
data taken matched that reported above for (R)-142. The precatalyst may be recovered by
dissolving the residue after distillation in CHzCl2 and washing it multiple times (10 X 20
mL) with water. Some diol contaminant still remains but the recovered material can be
reused to give a 43% yield of (R)-l42 which was > 96%ee. Chiracel OD, hexanes,

lmL/min, 254 nm, (S)-142 (34.86 min), (R)-l42 (47.11 min).

 

 

H co Et
korspps 2 t , 0TBDPS
(RH-12 n-BuLi. 81:3 0512 Etozc ($143 OH

75%

aAlkynal Ester R)-l43. A 250 mL round bottom flask was charged with

freshly distilled ethyl propiolate (0.76 g, 7.44 mmol, 0.75 mL), and dissolved in 60 mL

THF at —78 0C. A solution of n-BuLi (2.5 M in hexane, 7.44 mmol, 3.08 mL), was added

via syringe. The pale yellow reaction mixture was stirred for 10 minutes, then BF3.OEt2

(1.09 g, 7.44 mmol, 0.98 mL), was added neat via syringe. The yellow color persisted as
the reaction was stirred for another 5 minutes, then protected glycidol (R)-l42 (2.187 g,
7.0 mmol) was added neat via syringe. The reaction mixture darkened slightly. The
reaction was complete when checked by TLC after 1 h.

The reaction was quenched by adding saturated NH,,CI at —78 0C, then allowing

the mixture to warm to room temperature. The mixture was poured into a separatory
funnel containing 30 mL water and 50 mL ether. The aqueous layer was back-extracted

with 40 mL ether, and the combined organic layers were washed with water (2 x 50 mL)

158

 

and brine (1 x 50 mL), dried with MgSO4 and concentrated to a yellow/orange oil. This

oil was chromatographed on silica gel (5:1 hexane/EtOAc — KMnO4). One fraction at Rf

= 0.26 was collected and concentrated to give the product (R)-l43 (2.15 g, 5.24 mmol) in

75% yield as a pale yellow oil.

1H NMR (400 MHz, c1303): 8 1.07 (s, 9H), 1.30 (t, 3H, J = 7.2 Hz), 2.60 (dd,
2H, J = 2.1, 6.4 Hz), 3.71 (dd, 2H, J = 4.2, 9.8 Hz), 3.90-3.98 (m, 1H), 4.21 (q, 2H, J =
7.1 Hz), 7.38-7.64 (m, 6H), 7.64—7.66 (m, 4H); 13(3 NMR (100 MHz, CDC13): 6 14.02,

19.25, 23.53, 26.82, 61.86, 66.19, 69.68, 84.98, 127.73, 127.86, 129.94, 132.76, 135.52,
153.45; IR (neat film on NaCl): 3700-3100 (w), 2958 (m), 2931 (m), 2858 (m), 2237 (m),

1711 (s), 1428 (m), 1253 (s), 1113 (s), 1073 (m), 702 (s) cm"; EI mass spectrum m/z (%

rel intensity) 365 M+ 45 (18), 353 (26), 309 (15), 275 (91), 241 (84), 223 (26), 209 (65),
199 (95), 181 (100), 163 (58), 135 (30), 105 (26), 77 (20); Rf = 0.26 (5:1 hexane/EtOAc);

[01]D -6.40 (c 1.05, CHC13).

 

H ,Pd/B so co Et
Woraops ’ a ‘ = (AA
EtOZC OH 510A“ 3"“ 5 0TBDPS
(R)-143 920/0 (R)-1“ 6H

‘Alkyne Reduction to Give (in-144.61 A 250 mL round bottom flask was
charged with ester (R)-l43 (2.554 g, 6.22 mmol), and dissolved in 125 mL EtOAc at
room temperature. Lindlar’s catalyst (250 mg, 5% Pd on CaCO3 poisoned with lead,

Aldrich) and six drops of quinoline were added and the mixture was stirred briefly, then

159

 

placed under hydrogen atmosphere via four evacuation/backfill cycles. The reaction was
stirred for 2.5 h, then a small aliquot was removed, filtered, and checked by IR
spectroscopy for complete disappearance of the C-C triple bond. The reaction was
complete, so the catalyst was removed by filtration through Celite and the solution was

concentrated to a pale yellow oil. The oil was chromatographed on silica gel (5:1
hexane/EtOAc — KMnO4). One fraction at Rf = 0.37 was collected and concentrated to

give 2.37 g (5.7 mmol) of the product (R)-l44 as a colorless oil in 92.3% yield.

1H NMR (400 MHz, CDC13): 6 1.07 (s, 9H), 1.26 (t, 3H, J = 7.2 Hz), 2.82 (m,
2H), 3.57 (m, 1H), 3.67 (dd, 1H, J = 4.2, 10.2 Hz), 3.85 (m, 1H), 4.14 (q, 2H, J = 7.2
Hz), 5.87 (dt, 1H, J = 11.6, 1.6 Hz), 6.34 (dt, 1H, J = 11.5, 7.5 Hz), 7.37-7.64 (m, 6H),

7.64-7.67 (m, 4H); 13C NMR (100 MHz, CDC13): 8 14.22, 1925,2685, 32.62, 60.02,

67.64, 71.40, 121.67, 127.78, 129.82, 133.12, 135.53, 145.59, 166.61; IR (neat film on
NaCl): 3700-3400 (w), 2931 (m), 2858 (m), 1719 (s), 1427 (m), 1177 (m), 1113 (s), 702

(5); EI mass spectrum m/z (% rei intensity) 355 (W -57) (22), 309 (100), 289 (7), 277
(16), 241 (58), 223 (29), 199 (78), 181 (22), 163 (61), 139 (23), 105 (18), 77 (13); Rf =

0.37 (5:1 hexane/EtOAc); [011131.33 (c 1.05, CHCI3).

160

0

C028 p-TsOH 0
W 3 I
; 0TBDPS reflux, 4h 0TBDPS

(R)-144 OH 73% (R)-145

 

a'Lactone (R)-145. A 250 mL round bottom flask was charged with reduced ester

(R)-l44 (2.058 g, 5.0 mmol) and dissolved in 150 mL hexane (Optima grade, Fisher).
Solid p-TsOH hydrate (47 mg, 0.25 mmol, 5 mol%) was added, and the reaction was
heated to reflux for 24 h. The reaction was quenched with 20 mL NaHCO3 solution,

poured into a separatory funnel and washed with water (1 x 50 mL) and brine (1 x 50
mL), dried with MgSO4 and concentrated to a yellow/orange oil. The oil was
chromatographed on silica gel (5:1 hexane/EtOAc/KMnO4) giving two fractions, one at
Rf = 0.6 (presumed to be TBDPS-OH but not characterized) and the product (R)-145 at Rf

= 0.20, which was concentrated to a 73% yield of (R)-l45 (1.34 g, 3.65 mmol) as a

colorless 011.

1H NMR (400 MHz, CDC13): 6 1.07 (s, 9H), 2.45 (dt, 1H, J = 10.2, 1.2 Hz), 2.56
(ddt, 1H, J = 18.5, 11.0, 2.7 Hz), 3.84 (d, 2H, J = 4.9 Hz), 4.44-4.54 (m, 1H), 6.05 (dd,
1H, J = 1.1, 9.8 Hz), 6.83-6.91 (m, 1H), 7.40-7.44 (m, 6H), 7.64-7.68 (m, 411); 13C NMR

(100 MHz, CDC13): 6 19.25, 26.77, 64.76, 77.56, 121.26, 127.80, 129.89, 132.96, 135.53,
135.60, 144.79, 163.75; IR (neat film on NaCl): 2957 (w), 2930 (m), 2858 (m), 1732 (s),

1427 (m), 1247 (m), 1247 (m), 1133 (m), 1113 (s), 1048 (m), 703 (5); EI mass spectrum

m/z (% rel intensity) 309 M+ -57 (100), 241 (55), 223 (22), 199 (21), 183 (13), 163 (58),

105 (13), 77 (7); Rf = 0.20 (5:1 hexane/EtOAc), [alD 383° (c 1.00, CHC13).

161

0 0L

fit Dibal-H 1 o
OTBDPS PPTs. i-PrOH I 0TBDPS

(R)-145 90% (R)-146

 

aIsopropyl Lactol (ID-146.6164 A 100 mL RB flask was charged with lactone
(R)-l45 (0.366 g, 1.0 mmol) and dissolved in 10 mL CHZCI2 at —78 0C under argon. A

solution of DIBAL (1.0 M in hexane, 1.25 mL, 1.25 mmol), was added via syringe, and
the reaction was monitored by TLC for disappearance of the starting material. After 2 h,

the reaction was complete. The reaction was quenched at —78 °C with a 5 mL saturated aq
NH4C1 solution, then allowed to warm to room temperature. The reaction mixture was
poured into a separatory funnel containing 10 mL of CH2C12 and 10 mL of aq NH4CI
solution. The aqueous layer was back-extracted with CHZCIZ (2 x 10 mL). The combined
organic layers were washed with NH4C1 solution (1 x 20 mL) and brine (1 x 20 mL),

dried with MgSO4 and concentrated to a very sticky oil. The crude NMR and IR spectra

were satisfactory.
The crude lactol was dissolved in 10 mL of isopropanol and PPT S (0.037 g, 0.15
mol%) were added to the solution. The reaction was stirred at room temperature while

being monitored by TLC. The reaction was complete in 0.75 h. The reaction was
I quenched with 10 mL NaHCO3 solution and poured into a separatory funnel. The
aqueous layer was back-extracted with ether (2 x 10 mL). The combined organic layers
were washed with water (2 x 20 mL) and brine (1 x 20 mL), dried over MgSO4, and

concentrated to a yellow oil. The oil was chromatographed on silica gel (10:1

162

 

hexane/EtOAc — KMnO4), giving two compounds that were inseparable on TLC. The

fractions containing the two compounds were concentrated to a colorless oil. The crude
lHNMR spectra of this oil showed that two diastereomers of the product were present in a
9:1 ratio in 90% yield (0.369 g, 0.90 mmol). The ratio was determined by integration of

the alcohol methine proton.

Characterization data for major isomer of (R)-l46: 1H NMR (400 MHz,
CDC13): 6 1.05 (s, 9H), 1.17 (d, 3H, J = 6.1 Hz), 1.21 (d, 3H, J = 6.2 Hz), 1.98 (m, 2H),
3.63 (dd, 1H, J = 4.8, 10.6 Hz), 3.78 (dd, 1H, J = 4.5, 10.6 Hz), 4.03 (quintet, 1H, J = 4.2
Hz), 4.12 (m, 1H), 5.10 (s, 1H) (minor isomer has signal at 5.16 ppm), 5.72 (m, 1H), 6.00
(m, 1H), 7.37-7.42 (m, 6H), 7.68-7.71 (m, 4H); 13C NMR (100 MHz, CDCl;,): 6 19.21,
21.77, 23.90, 26.78, 27.10, 66.77, 67.01, 68.95, 92.58, 126.13, 127.62, 128.40, 129.60,
133.61, 135.62; IR (neat film on NaCl): 2966-2857 (m), 1472 (m), 1427 (m), 1183 (m),
1106 (s), 1020 (s), 823 (m), 701 (s) cm"; Rf = 0.52 (minor)/O.50 (major) (10:1

hexane/EtOAc).

A .k

TBAF
I ° | °
0TBDPS 91 % OH

(R)-146 (R)-2b

 

“Primary Alcohols 2b.63‘64 A solution of lactol (R)-146 (740 mg, 1.87 mmol) was

dissolved in 10 mL wet THF at room temperature. A solution of tetrabutylammonuim

163

fluoride (1.0 M in THF, 3.73 mmol, 2 equiv.) was added via syringe. The reaction was
followed by TLC (10:1 hexane/EtOAc) to monitor disappearance of the starting material.
The reaction was done after 1.5 h.

The reaction was quenched with aq NaHCO3 solution (10 mL) and diluted with
10 mL ether. This mixture was poured into a separatory funnel, and the aqueous layer
was back extracted with 20 mL ether. The combined organic layers were washed with
water (2 x 10 mL) and brine (1 x 10 mL), dried with MgSO4 and concentrated to a
colorless oil. The oil was chromatographed on silica gel (gradient elution, 5:1

hexane/EtOAc followed by 2:1 hexane/EtOAc), giving a spot at Rf = 0.52 presumed to be

TBDPS-OH (not characterized) and a spot at Rf = 0.12, which was concentrated to give

the product alcohols (R)-2b (293 mg, 1.64 mmol) in 91% yield as a colorless oil.

Characterization data (proton and carbon-13 NMR and IR) matched those

reported by Crimmins et. a1.63 and Boger et. 31.24

 

NMO-TPAP
| ° 4 °
OH 90% I /o

Oxidation to Aldehyde (R)-23.63’64 A solution of the 9:1 mixture of primary
alcohols (R)-2b (50 mg, 0.29 mmol) and N-methylmorpholine N-oxide (51 mg, 0.44
mmol) in 5 mL of anhydrous CH2C12 was treated with activated 4 A° molecular sieves

(0.75 g). After stirring at 25 °C for 1 h, TPAP (3.2 mg, 0.092 mmol) was added and the

164

reaction mixture was stirred at 25 °C for 30 min. Chromatography (5102, 40% EtzO-

hexanes) provided (R)-2a (44.4 mg, 0.26 mmol) in 90% yield after careful evaporization.

The aldehyde (R)-2a was produced as a 9:1 mixture of diastereomers that could be

separated. The minor aldehyde is the cis isomer and has an R,- value of 0.48 and the major

isomer is the trans isomer and has an Rf of 0.42. The cis and trans isomers were
determined by a nOe experiment on acetal (R)-l46. In that experiment only the acetal
proton of the minor isomer showed a nOe to the methine proton at C5.

Characterization data for the major isomer of (R)-28 (proton NMR) matched that

reported by Crimmins et. a1.63 and also that reported by Boger et. 3124.

A n-BuLi, THF A
HO OH = TBSO OH
124 "359 "1 “1) 147
94%

 

(Z)-4-(tert-butyldimethylsilyloxy)-2-buten-l-ol147.65 Cis-2-butene-1,4—diol
(4.401 g, 4.11 mL, 50 mmol) was dissolved in 100 mL THF at 0 °C under argon. A
solution of 2.5 M n-BuLi in hexane (20 mL, 50 mmol) was added via syringe. Insoluble
yellow/white clumps of solid were formed upon addition of the n-BuLi, which were
broken up to give a suspended white solid upon vigorous stirring. The reaction was
stirred for 1 h at 0 °C, then tert-butyl dimethylsilyl chloride (7.54 g, 50 mmol) was added
neat in one portion, and the cold bath was removed. The white suspension disappeared as

the reaction progressed, leaving a transparent yellow solution. Stirring was continued for
3h, then the reaction was quenched by adding 50 mL saturated aqueous NH4C1 solution.
The mixture was diluted with 100 mL ether, poured into a separatory funnel, and

washed with 75 mL water and 50 mL brine and the dried over anhydrous MgSO4, and

165

concentrated to a yellow oil. This oil was distilled under high vacuum (bp 82-88 °C/0.2

torr) to give 147 (9.54 g, 0.47 mmol) as a colorless oil in 94.3%.

1 H NMR (300 MHz, CDCI3) 6 0.07 (s, 6H), 0.89 (s, 9H), 2.32 (broad s, 1H), 4.17
(d, 2H, J = 5.1 Hz), 4.23 (d, 2H, J = 5.5 Hz), 5.56 (m, 2H), 13C NMR (75 MHz, CDC13) 6
-5.30, 18.27, 25.84, 58.69, 59.51, 130.02, 131.15; IR (neat film on NaCl): 3350 (w),

2950-2850 (m), 1472 (m), 1254 (s), 1088 (s), 837 (m), 776 (m) cm]; EI mass spectrum

m/z (% rel intensity) 145 M+ -57 (27), 127 (8), 99(3), 75 (100).

 

a(E)-4-(tert-butyldimethylsilyloxy)-2-butenal148. Alcohol 147 (2.02 g, 10

mmol) was dissolved in 150 mL dry CHZClz. Pyridinium dichromate (5.64, 15 mmol)

was added, the reaction was placed under argon atmosphere and stirred for 20 h. The
reaction was diluted with 150 mL ether and filtered through a 1 inch thick layer of silica

gel to remove brown solids. The orange organic solution was washed with saturated
aqueous CuSO4 solution (2 x 50 mL), water (2 x 100 mL), and brine (1 x 100 mL) and
then dried over MgSO4, filtered through another 1 inch layer of silica gel, and
concentrated to a pale yellow oil. The oil was chromatographed on silica gel (10:1
pentane/ether — UV/KMnO4 visualization) to give 148 as a colorless oil in 80% yield

(1.61 g, 0.81 mmol).

166

1H NMR (300 MHz, CDC13): 8 0.094 (s, 6H), 0.93 (s, 9H), 4.46 (m, 2H), 6.40
(ddt, 1H, J = 15.4, 8.0, 2.1 Hz), 6.90 (dt, 1H, J = 15.5, 3.0 Hz), 9.61 (d, 1H, J = 8 Hz);
13C NMR (75 MHz, c1303); 8 -549, 18.28, 25.76, 62.21, 130.53, 156.46, 198.93; IR
(neat film on NaCl): 2956-2857 (m), 1694 (s), 1255 (s), 1114 (s), 967 (m), 887 (m), 779

(m) cm'l; Rf = 0.22 (10:1 pentane/ether).

 

ICH PPhall, NaHMDS
O I 2 W
TBSOW o *7 TBSO / /
“a HMPA, THE-78 c 4.
64% 9:1Z/E

a'(Z,E)-Iododiene 4a.66 Note: This compound is light-sensitive, and is best
handled in a darkened room and used immediately. A 250 mL round-bottom flask was
charged with ICH2(PPh3)I (8.80 g, 16.6 mmol) and suspended in 60 mL THF. The flask

was wrapped with aluminium foil and cooled to -78 °C. A 1.0 M solution of sodium
bis(trimethylsilyl)amide in THF (16.6 mL, 16.6 mmol) was added, and the solution was
stirred for 15 min, then allowed to warm to room temperature. Freshly distilled HMPA (4
mL) was added and the reaction was briefly stirred, then cooled back down to -78 °C. A
precooled (-78 °C) solution of aldehyde 148 (3.32 g, 16.6 mmol) in 10 mL THF was
added via cannula, and the reaction was stirred for 15 min, then allowed to warm to room
temperature while stirring for 1 hr.

The reaction was quenched by diluting with 50 mL ether, then adding saturated

aqueous NH4C1 solution. The mixture was poured into a separatory funnel, and the layers

were separated. The aqueous layer was extracted with ether (2 x 80 mL), and the

combined organic layers were washed with water (2 x 100 mL) and brine (1 x 80 mL)

167

and then dried over anhydrous MgSO4 and concentrated to a dark brown oil. The oil was
taken up in a mixture of 50 mL pentane and 1 mL ether, leading to formation of a brown
solid precipitate. This precipitate (Ph3P=O) was separated by filtration through a thin
layer of silica gel and the resulting brown solution was chromatographed on silica gel
(50:1 pentane/ether — UV visualization) and concentrated to give 4a as a light orange
liquid in 84% yield (4.12 g, 13.95 mmol). This material consisted of a 9:1 ratio of
cis/trans isomers of 4a as determined by integration of the vinylic proton doublet of
triplets at 5.81 (minor) and doublet of triplets at 6.06 ppm in crude proton NMR

spectrum.

1H NMR (300 MHz, CDC13): 6 0.12 (s, 6H), 0.96 (s, 9H), 4.26 (d, 2H, J = 4.2
Hz), 6.06 (dt, 1H, J = 15.1, 4.6 Hz), 6.24 (d, 1H, J = 7.6 Hz), 6.42-6.56 (m, 1H), 6.75 (dd,

1H, J = 7.3, 10.2 Hz); R, = 0.6 (50:1 pentane/ether).

 

A n-BULi, THF A
HO 0H t TBDPSO OH
94%

“(2)-4-(tert-butyldiphenylsilyloxy)-2-buten-l-ol 149.65 Cis-2-butene-1,4-diol
(2.2 g, 4.11 mL, 25 mmol) was dissolved in 50 mL THF at 0 °C under argon. A solution
of 2.5 M n-BuLi in hexane (10 mL, 25 mmol) was added via syringe. Insoluble
yellow/white clumps of solid were formed upon addition of the n-BuLi, which were
broken up to give a suspended white solid upon vigorous stirring. The reaction was
stirred for 1 h at 0 °C, then tert-butyl diphenylsilyl chloride (6.5 mL, 25 mmol) was

added neat in one portion, and the cold bath was removed. The white suspension

168

disappeared as the reaction progressed, leaving a transparent yellow solution. Stirring

was continued for 15h, then the reaction was quenched by adding 25 mL saturated
aqueous NH4C1 solution.

The mixture was diluted with 50 mL ether, poured into a separatory funnel, and
washed with 25 mL water and 25 mL brine and then dried over anhydrous MgSO4, and

concentrated to give 149 as a a colorless oil in 94% yield (7.66 g, 23.5 mmol). This oil
was purified on a silica gel column (10:1 hexane:EtOAc) but compound 149 could also

be purified by vacuum distillation (bp 244—254 °C/0.2 torr).

1H NMR (300 MHz, CDC13) 6 1.07 (s, 9H), 4.02 (d, 2H, J = 6.0 Hz), 4.29 (d, 2H,
J = 5.8 Hz), 5.59-5.80 (m, 2H), 7.36-7.52 (m, 6H), 7.68-7.80 (m, 4H) 13C NMR (75
MHz, CDC13) 6 19.06, 26.72, 58.65, 60.18, 127.69, 129.74, 129.89, 130.85, 133.35,

135.57; IR (neat film on NaCl): 3352 (w), 3072 (s), 3049 (s), 3026 (s), 2999 (s), 2959(8),

2932 (s), 2891 (s), 2858 (s), 1472 (s), 1427 (s), 1113 (s), 824 (s), cm}; FAB mass
spectrum m/z (% rel intensity) 327 (M+ +1) (5), 309 (15), 269 (30), 199 (75), 152 (30),

122 (100), 93 (100), 75 (40), 57 (25); Anal calcd for ConngZSi: C 73.57; H 8.03.

Found: C 73.79; H 7.95.

169

PCC. NaOAc 0
TBDPSOAOH : TBDPSOW

 

100%
a(E)-4-(tert-butyldiphenylsilyloxy)-2-butenal 150.89 To a suspension of
pyridinium chlorochromate (2.67 g, 12.4 mmol) and sodium acetate (2.7 g, 13.3 mmol)
was added alcohol 149 (2.7 g, 8.3 mmol) dissolved in 20 mL dry CH2C12. The reaction

was placed under argon atmosphere and stirred for 2 h. The reaction was diluted with 70
mL ether and filtered through a 1 inch thick layer of silica gel to remove the brown

solids. The brown solids were washed twice with 70 mL of ether and the combined
organic layers dried over MgSO4. Aldehyde 150 was concentrated and dried, its crude

NMR was satisfactory and the crude product was used for the next step. All of the
following data was taken on unpurified material. Aldehyde 150 (2.69 g, 8.3 mmol) was

obtained in 100% as a white solid. Characterization data (proton and carbon-13 NMR and

IR) matched those reported by Evans et. al.89

1H NMR (300 MHz, CDC13): 8 1.07 (s, 9H), 4.42-4.46 (m, 2H), 6.56 (dd, 1H, J =
15.4, 8.0 Hz), 6.83 (dt, 1H, J = 15.7, 1.0 Hz), 7.35-7.48 (m, 6H), 7.64 (dd, 4H, J = 6.0,
1.7 Hz), 9.59 (d, 1H, J = 8.0 Hz); 13C NMR (75 MHz, CDC13): 8 19.17, 26.65, 62.90,

127.82, 129.93, 130.52, 132.66, 135.37, 155.93, 193.40; IR (neat film on NaCl): 3072 (s),

3052 (s), 2957 (s), 2919 (m), 2851 (m), 1694 (s), 1473 (s), 1429 (s), 1381 (s), 1113 (s),
968 (s), 824 (s) cm]; FAB mass spectrum m/z (% rel intensity) 325 (M+ +1) (21), 309

(30), 267 (20), 239 (35), 199 (98), 197.05 (95), 137 (95), 135 (100), 105 (60), 91 (60), 57

170

 

(55); Anal calcd for C20H24OZSi: C 73.96; H 7.40. Found: C 73.87; H 7.75. Rf = 0.20

(10: 1 pentane/ether).

Br

TBDPSO / TBDPSO / Br
THF,0°C
150 151

Dibromodiene 151. 38 To a solution of carbon tertrabromide at 0°C (1.02 g, 3.08
mmol) in 15 mL of dry CHZCIZ was added triphenyl phosphine (1.62 g, 6.16 mmol). After

5 mins, a solution of aldehyde 150 (500 mg, 1.54 mmol) in 7.5 mL of dry CHZCl2 was

added. The reaction mixture was stirred for 1.5 h, then diluted with 50 mL of hexanes.
The diluted reaction mixture was then filtered through Celite and evaporated to give a

beige solid. Flash chromatography on silica gel (20:1 pentane/ether) gave vinyl

dibromide 151 in 91% (0.672 g, 1.40 mmol) as an off-white oil. Rf = 0.13 (20:1

pentane/ether).

1H NMR (300 MHz, (200,); 8 1.13 (s, 9H), 4.27 (dd, 2H, J = 4.3, 1.7 Hz), 5.97
(dt, 1H, J = 15.1.4.1 Hz), 6.53 (tdd, 1H, J = 10.4.4.7, 1.9 Hz), 6.99 (d, 1H, J = 10.1 Hz)
7.34—7.50 (m, 6H), 7.69-7.76 (m, 4H); 13C NMR (75 MHz, CDC13): 8 19.20, 26.77,
63.58, 90.62, 126.04, 127.73, 129.76, 133.24, 135.45, 136.29, 136.63; IR (neat film on
NaCl): 3071 (s), 2932 (s), 2857 (s), 1472 (s), 1428 (s), 1113 (s), 968 (s), 826 (s) cm];
FAB mass spectrum m/z (% rel intensity) 481 (M-l)+ (2, 81Br), 479 (M-l)+ (4, 8lBr,
79st), 477 (M-1)+ (2, 79Br), 425 (6, 81Br), 423 (12, 8113:, 7931'), 421 (6, 79Br), 343 (2,

81Br), 341 (2, 7931') 327 (7, 8‘Br), 325 (5, 79st), 281 (9, 8'i3r), 279 (9, 79st), 263 (30,

171

81Br) 261 (30, 7931'), 227 (10), 225 (20), 223 (10), 207 (20), 199 (48), 197 (50), 135
(100), 105 (30), 91 (30), 73 (60), 57 (37), 55 (40); Anal calcd for C21H24OSiBr2: C 52.51;

H 5.04. Found: C 52.16; H 5.13.

 

Br Br
Zn-Cu. AcOH W
TBDPSOW Br t TBDPSO / /
151 THF/MeOH 4b

98%
“Activated Metal Reduction to Give diene 4b.68 A 250 mL flask was charged

with zinc dust (25 g, 0.391 mol, 99.9%, 150-325 mesh, Alfa/Aesar) which was then
suspended in 125 mL HPLC grade water and sparged with argon for 15 min. Anhydrous
copper (II) acetate (2.5 g, 13.8 mmol) was added, the flask was capped with a rubber
septum, and the slurry was stirred for 15 minutes. The black suspension of activated
metal was isolated by filtration on a Buchner funnel followed by sequential washings
with HPLC grade water and methanol.

Acetic acid (25 mL) followed by the black solid was immediately added to a
solution of dibromide 151 (240 mg, 0.5 mmol) in 187.5 mL of a 2:1 mixture of
THF/MeOH. The flask was placed under an argon atmosphere and stirred overnight at
0°C. The reaction mixture was filtered through Celite and the black metal filter cake was
rinsed with 125 mL ether into a stirring solution of saturated sodium bicarbonate (50
mL). The resulting mixture was poured into a separatory funnel and the aqueous layer

was extracted with ether (2 x 60 mL). The combined organic layers were washed with

brine (1 x 50 mL), dried over anhydrous MgSO4, and concentrated to a colorless oil. The

oil was purified by chromatography on silica gel (100% hexane, UV/KMnO4

172

visualization) to give the product 4b as an off-white oil in 98% yield (196 mg, 0.49

mmol). This oil was a single isomer by proton NMR. Rf = 0.10 (hexane).

1H NMR (300 MHz, CDCI3): 8 1.06 (s, 9H), 4.24 (d, 2H, J = 13.8, 3.9 Hz), 5.93
(dt, 1H, J = 14.6, 5.8 Hz), 6.22 (d, 1H, J = 6.9 Hz), 6.58-6.78 (m, 2H) 7.32-7.44 (m, 6H),
7.62-7.70 (m, 4H); 13C NMR (75 MHz, CDC13): 8 19.22, 26.77, 63.82, 107.62, 125.12,
127.59, 129.69, 132.01, 135.41, 135.52, 136.47; IR (neat film on NaCl): 3073 (s), 2959
(s), 2932 (s), 2857 (s), 1458 (s), 1428 (s), 1113 (s), 702 (s) cm]; FAB mass spectrum m/z
(% rel intensity) 402 M+ (0.25, 81Br), 400 W (0.25, 7931'), 372 (1.2. 8‘13r), 370 (1.2,
798:), 345 (100. 81Br), 343 (100, 79Br), 315 (14, 81Br), 313 (14, 79130, 265 (80), 263 (95),

261 (90), 199 (84), 187 (28), 181 (32), 143 (64), 135 (16), 77 (12), 65(12).

 

Br Br
Bu3SnH, Pd(PPh3)4 W
TBDPSOWBr 2 TBDPSO / /
151 benzene 4b
95%

Bu3SnH Reduction to Give diene 4b.39 To a stirred solution of dibromide 151
(50 mg, 0.104 mmol) and Pd(PPh3)4 (4.6 mg, 0.004 mmol) in anhydrous benzene (0.7

mL) was added Bu3SnH (0.032 mL, 0.12 mmol) in anhydrous benzene (0.3 mL) under an

argon atmosphere and the mixture was stirred for 1h at room temperature. The mixture

was diluted with hexane (0.7 mL) and washed with water (0.4 mL) and brine (0.4 mL)
and dried over MgSO4. The solvent was removed under reduced pressure and the residue

purified on a silica gel column with hexanes as the eluent. A 95% (40 mg, 0.099 mmol)

173

yield of vinyl bromide 4b was obtained an off-white oil. The lHNMR spectrum of 4b

matched that for 4b obtained in the Zn/Cu reduction of 151 (see above) and it revealed

that 4b formed from this reaction was also formed as a single isomer.

 

B
/ O CBQ, PPh3 A W
TBSOW Br Br
148 THF' 0 0C 181
55%

d'l‘ribromodiene 161.38 To a solution of carbon tertrabromide at 0 °C (8.62 g, 26
mmol) in 120 mL of dry CH2C12 was added triphenyl phosphine (13.84 g, 52 mmol).

After 15 mins, 8 solution of aldehyde 148 (2.6 g, 13 mmol) in 70 mL of dry CHZCIZ was
added. The reaction mixture was stirred for 1 h, then diluted with 150 mL of hexanes.
The diluted reaction mixture was then filtered through Celite and evaporated to give a
beige solid. Flash chromatography on silica gel (98:2 hexanes/EtOAc) gave 2.21 g (7.34

mmol) of an oil tentatively assigned as the vinyl tribromide 161 in 55% yield.

1H NMR (300 MHz, cpc13); 8 4.02 (dd, 2H, J = 7.8, 0.9 Hz), 6.03-6.15 (m, 1H),
6.36 (dd, 1H, J = 14.4, 10.8 Hz), 6.97 (d, 1H, J = 10.2 Hz); 13(2 NMR (75 MHz, CDCl,):

6 31.80, 93.63, 130.80, 132.14, 135.43.

0 014,13th3 W
TBDPSOW THF 0°C = TBDPSO / / l
150 ' 157

36%

 

Diiododiene 157. 38'67Note: This compound is light-sensitive, and is best handled

in a darkened room and used immediately. To a solution of carbon tertraiodide at 0 °C

174

(322 mg, 0.62 mmol) in 3 mL of dry CHZCI2 was added triphenyl phosphine (325 mg,

1.24 mmol). After 5 mins, a solution of aldehyde 150 (100 mg, 0.31 mmol) in 1.5 ml of
dry CHZCI2 was added. The reaction mixture was stirred for 1.5 h, then diluted with 20

mL of hexanes. The diluted reaction mixture was then filtered through Celite and
evaporated to give a beige solid. Flash chromatography on silica gel (20:1 pentane/ether)

gave 36% (65.8 mg, 0.11 mmol) of colorless oil vinyl diiodide 157.

1H NMR (300 MHz, CDCl3): 6 1.08 (s, 9H), 4.20 (dd, 2H, J = 4.1, 1.9 Hz), 5.99
(dt, 1H, J = 15.1, 4.1 Hz), 6.27 (ddt, 1H, J = 9.9, 5.2, 1.9 Hz), 7.30-7.46 (m, 6H), 7.49 (d,
1H, J = 9.9 Hz), 7.62-7.72 (m, 4H); 13C NMR (75 MHZ, CDCI3): 6 19.21, 26.80, 63.44,
94.21, 127.74, 129.76, 131.57, 133.24, 135.48, 136.54, 149.52; IR (neat film on NaCl):

2963 (s), 2924 (s), 2851 (s), 2363 (s), 2336 (s), 1653 (s), 1262 (s), 1098 (s), 1020 (s) cm'

I

lCH PPh I, N HMDS
TBDPSOWO I 2 3] a : TBDPSOW
150 HMPA. THF, -78 00 4c

72% 5:1 Z/E

 

(Z,E)-Iod0diene 4666 Note: This compound is light-sensitive, and is best handled
in a darkened room and used immediately. A 100 mL round-bottom flask was charged
with ICH2(PPh3)I (163.5 mg, 0.308 mmol) and suspended in 5 mL THF. The flask was

wrapped with aluminium foil and cooled to —78 °C. A 1.0 M solution of sodium

bis(trimethylsilyl)amide in THF (0.308 mL, 0.308 mmol) was added, and the solution

175

 

was stirred for 15 min, then allowed to warm to room temperature. Freshly distilled
HMPA (0.31 mL) was added and the reaction was briefly stirred, then cooled back down
to —78 °C. A precooled (-78 °C) solution of aldehyde 150 (100 mg, 0.308 mmol) in 1 mL
THF was added via cannula, and the reaction was stirred for 15 min, then allowed to
warm to room temperature while stirring for 1 hr.

The reaction was quenched by diluting with 10 mL ether and then adding
saturated aqueous NH4CI solution. The mixture was poured into a separatory funnel and

the layers were separated. The aqueous layer was extracted with ether (2 x 8 mL), and the

combined organic layers were washed with water (2 x 10 mL) and brine (1 x 8 mL), dried
over anhydrous MgSO4 and then concentrated to a dark brown oil. The oil was taken up
in 50 mL : 1 mL pentane/ether leading to formation of a brown solid precipitate. This
precipitate (Ph3P=O) was removed by filtration through a thin layer of silica gel and the

brown filtrate was stripped of solvent. The crude product was purified by
chromatography on silica gel (50:1 pentane/ether - UV visualization) to give 4c 8 light
yellow oil in 72% yield (0.100 g, 0.223 mmol). This material had a 5:1 ratio of cis/trans
isomers as determined by integration of the vinylic protons at 5.78 (minor) and 6.02 ppm

(major) in crude proton NMR spectrum.

1H NMR (300 MHz, CDCI3): 8 1.07 (s, 9H), 4.24 (dd, 2H, J = 4.6, 1.7 Hz), 6.02
(dt, 1H, J = 14.9, 4.7 Hz), 6.22 (d, 1H, J = 7.3 Hz), 6.52-6.65 (m, 1H), 6.72 (dd, 1H, J =
10.0, 7.3 Hz), 7.36-7.48 (m, 6H), 7.64-7.70 (m, 4H); 13C NMR (75 MHz, CDC13): 8

19.19, 26.78, 63.70, 81.75, 100.25, 127.66, 129.66, 133.35, 135.49, 137.13, 137.75.

176

Br

 

. \ \
1) t-BuLl (2.4 eq) 0TBDPS
WOTBDPS - 4d
4h 2) Bu3$nCl (1.2 eq) H
~780C, Etzo HWOTBDPS
49

Preparation of Vinyl Stannane 4d:28 To a solution of vinyl bromide 4b (72 mg,

0.186 mmol) in 3 mL of ether at —78 °C was added t-BuLi (1.7 M in pentane, 263 11L,
0.446 mmol). This solution was stirred for 2h after which time Bu3SnCl (60 11L, 0.223
mmol) was added and stirred for an additional 2h. The reaction was quenched with water

(4 mL) and diluted with EtOAc (8 mL). The organic layer was washed with brine (10
mL), dried over NaZSO4 and concentrated. Column chromatography in pure pentane

revealed the presence of two compounds which were separated and identified as 4d and
4e (both colorless oils) which were obtained in 62% (70.6 mg, 0.12 mmol) and 31% (18.6

mg, 0.058 mmol) of yields respectively.

Vinyl Stannane 4d 23. 1H NMR (300 MHz, CDC13): 8 0.87 (t, 9H, J = 7.5 Hz),

0.95 (dd, 6H, J = 7.8, 8.4 Hz), 1.08 (s, 9H), 1.29 (6H, dq, 6H, J = 14.4, 7.2 Hz), 1.45-1.55
(m, 6H), 4.27 (dd, 2H, J = 1.7, 4.5 Hz), 5.80 (dd, 1H, J = 15.0, 4.5 Hz), 6.06 (d + dd, 1H,

J'H_H = 12.9 Hz, J23,” = 63.9 H), 6.29-6.38 (m, 1H), 7.09 (dd, 1H, J = 10.8, 12.6 Hz),
7.35-7.46 (m, 6H), 7.69 (dd, 4H, J = 2.0, 8.7 Hz); Rf = 0.22 (pentane). This 1H NMR

matches that reported previously for this compound.

177

 

Side Product 4e: 1H NMR (300 MHz, CDC13): 8 1.05 (s, 9H), 4.22 (d, 2H, J =
4.7 Hz), 5.04 (d, 1H, J = 10.2 Hz). 5.15 (d, 1H, J = 15.9 Hz), 5.76 (dt, 1H, J = 14.3.4.7

Hz), 6.22-6.40 (m, 2H), 7.28—7.40 (m, 6H), 7.65 (dd, 4H, J = 1.9, 7.8 Hz); 13C NMR (100
MHz, CDC13) 6 19.25, 26.81, 63.97, 116.58, 127.65, 129.63, 130.31, 132.80, 133.64,
135.54, 136.61; Rf = 0.17 (pentane); IR (neat film on NaCl): 3073 (s), 2957 (s), 2930 (s),
2857 (s), 1472 (s), 1428 (s), 1113 (s), 1053 (s), 1005 (s), 823 (s), 741 (s) cm'l; FAB mass
spectrum m/z (% rel intensity) 321 (M-l)+ (20), 307 (15), 265 (68), 199 (76), 197 (48),

187 (36), 137 (62), 136 (50), 135 (100), 121 (20), 105 (20), 91 (20), 75 (20), 67 (52), 57

(8); Anal calcd for C21H2605i: C 78.21; H 8.13. Found C 78.54; 8.59.

 

0—-Cr(CC),
omMo)4 1) AcBr/CH2C12 A MaxN N CH3
(OC)5Cr=< , ( ,
CH3 2) 0
Me‘ )L M6 Ph
N NH (S,S)-162a
L1.
M6 ’Ph

bMethyl [(48, SS) -l,5-dimethy1-4-phenyl-2-imidazolidinone] methylene
tetracarbonyl chromium (0) 162a and its Enantiomer 162b.69
Tetramethylammonium( l -hydroxyethylidene)pentacarbonylchromium (0)69 (3 .0

g, 9.7 mmol) was dissolved in 45 mL CH2C12 under an atmosphere of argon and cooled

to -78 °C. Freshly distilled acetyl bromide (0.72 mL, 9.7 mmol) was then added dropwise

and the remaining solution was stirred for an additional 60 minutes after which (45, SS)-

1,5-dimeth 14 henyl-2-imidazolidinone69 (1.84 g, 9.7 mmol) was added neat to the
y P

178

solution. The mixture was gradually warmed to -55 °C over a 15 minute period and was

stirred at this temperature for 18 hr. The mixture was quickly warmed to room
temperature, washed with NaHCO3 (3 x 75 mL), dried with MgSO4 and concentrated on
a rotary evaporator to remove two-thirds of the solvent. The resulting reddish-brown
solution was loaded onto a silica gel column and the product was eluted with CHZCI2 (Rf

= 0.63) to give complex 162a (2.40 g, 6.31 mmol) as a deep-red solid in 65% yield.

Spectral data for 162a: mp 117 °C (dec.); 1H NMR (500 MHz, CDC13) 6 0.85 (d,
3H, J = 6.7 Hz), 2.78 (s, 3H), 2.94 (s, 3H), 4.40-4.48 (m, 1H), 5.35 (d, 1H, J = 8.5 Hz),
7.08 (br s, 2H), 7.41 (t, 3H, J = 5.5 Hz); 13C NMR (75 MHz, CDCI3) 6 14.84, 28.43,
34.35, 59.98, 61.79, 126.36, 128.35, 129.24, 133.85, 162.32, 215.21, 215.49, 231.62
(2C), 320.87; IR (neat) 2007 (s), 1982 (shoulder, s), 1900 (vs), 1827 (s), 1711 (s), 1355
(m), 1148 (m) cm]; EI mass spectrum m/z (% rel intensity) 380 M+ (10), 244 (25), 230
(15), 220 (100), 203 (40), 132 (40), 118 (30), 108 (95), 80 (100); Anal calcd for
C17H1605N2Cr: C, 53.68; H, 4.24; N, 7.37. Found: C, 53.31; H, 4.24; N, 7.20.

Carbene complex 162b, the enantiomer of complex 162a, was synthesized
according to the above procedure by using the (4S, 5R)-1,5-dimethyl-4-phenyl-2-

imidazolidinone as the chiral auxiliary.

179

 

1) MeLi
165C SIMea

 

//

3Preparation of 2-alkynals. Illustrated with the Preparation of

Trimethylsilylpropynal l65c.91 A solution of (trimethylsilyl)acetylene (15.72 g, 22.6

mL, 0.16 mol) in 120 mL ether was cooled to —78 °C. A solution of methyl lithium (1.6
M in ether, 100 mL, 0.16 mol) was added via cannula. Note: for the preparation of
volatile aldehydes, solutions of n-BuLi in hexane should not be used. The reaction was
stirred for 20 min, then anhydrous dimethylforrnamide (14.04 g, 14.9 mL, 0.192 mol)
was added neat via syringe. The cold bath was removed and the reaction was stirred for
3h while warming to room temperature. The reaction was quenched and hydrolyzed by
pouring the ether solution into a solution of excess dilute aqueous hydrochloric acid at 0

°C (2.5 eq., 0.4 mol, 33 mL 12 M concentrated HCI). The mixture was neutalized to pH 6
by adding saturated aqueous NaHCO3 solution, and poured into a 1 L separatory funnel.
The aqueous layer was back-extracted with ether (4 x 100 mL). The combined organic
layers were dried with MgSO4, filtered through a 2” plug of silica gel to remove red

material, and concentrated on the rotary evaporator without vacuum and the water bath at
40 °C. The remaining ether was removed via short-path distillation at atmospheric
pressure by heating in an oil bath at 65 °C. The product l65c an acrid-smelling liquid was
purified by vacuum transfer (0.2 mm Hg) into a flask cooled to —78 °C in 66.5% yield

(13.4 g, 0.11 mol). The following lHNMR data matches that reported for this

compound.91

180

1H NMR (300 MHz, CDCI3) 8 0.260 (s, 9H), 9.16 (s, 1H).

H

O O SIM63
C02(CO)3 C0,,“ ~. . ”(.00
H \\ ————-———- CO—Co-Co—CO
S‘Mea C0’ \C0
165C 159

cCobalt protected Alkyne 169. To a solution of C02(CO)8 (8.75g, 25 mmol) in

100 mL of ether was added aldehyde l65c (3.0 g, 23.8 mmol) in 20 mL of ether at room
temperature. There was an immediate effervescence and the solution turned dark red. The

solution was concentrated and first chromatographed with hexanes to remove any
inorganic compounds then with CHZCIZ to obtain the desired product 169 in 94% yield
(9.7 g, 23.6 mmol) a deep red solid.

1H NMR (300 MHz, CDC13) 8 0.32 (s, 9H), 10.28 (s, 1H).

a- Data obtained from Mark Parisi’s Thesis;56

b- Data obtained from Yan Shi’s Thesis;71

c- Data obtained from the unpublished results of Kenneth Wilson and W. D.
Wulff.58

d- Data obtained from the unpublished results of XueLui Jun and W. D.

Wulff.76

181

 

Experimental data for Chapter 3.

 

0—-Crco H
A A‘ " 0 R Me A0
_\___/ C0—-"po Co—Co 2. HOAdCelV

M6 Ph CO co

(4S,5R)-162a ‘69 (486,5R,R)-166c

b(4S,5R)-l-((R)-3-hydroxy-5-(trimethylsilyl)pent-4-ynoyl)-3,4-dimethy1—5-
phenylimidazolidin-Z-one. A solution of LDA was prepared by adding 3.67 mL of n-
BuLi (1.6 M in hexanes, 4.5 mmol) to a solution of freshly distilled diisopropylamine
(0.66 mL, 4.74 mmol) in 20 mL of THF at room temperature and stirring for 15 minutes.
A solution of 1.64 g (5.0 mmol) of carbene complex (4S,SS)-162a in 20 mL THF was
added dropwise to the solution of LDA at —78 °C. The resultant yellow-orange solution

was stirred for 5 minutes at —78 0C. A precooled solution (—78 0C) of dicobalt

hexacarbonyl complexed (trimethylsilyl) propynal 169 (2.13 g, 5.16 mmol) in 15 mL
THF was added dropwise via syringe. The dark red reaction mixture was allowed to stir
for 3 h, then quenched by adding acetic acid (0.271 mL, 4.74 mmol) and stirring for 5

minutes. A freshly prepared solution of ceric ammonium nitrate (37.72g, 68.8 mmol) in
20 mL of H20 : MeOH (2 : 1) was added in 4 equal portions, and the cold bath was
removed. Stirring was continued for 15 minutes, and the reaction mixture was extracted

with ether (3 x 30 mL). The combined organic layers were washed with NaHCO3

solution (30 mL), H20 (50 mL), and brine (50 mL), dried with MgSO4 and concentrated

on the rotary evaporator. Purification of the crude product by flash chromatography on
silica gel (1:1 hexanes/EtOAc) affored aldol adduct (4S,SS,R)-l66c (1.43 g, 4.0 mmol) as

a viscous pale yellow oil in a 99.5:0.5 diastereomeric ratio in 80% yield.

182

1H NMR (300 MHz, CDCI3): 8 0.08 (s, 9H), 0.75 (d, 3H, J = 6.6 Hz), 2.78 (s,
3H). 3.28 (dd, 1H, J = 17.4, 3.3 Hz), 3.48 (s, 1H), 3.51 (dd, 1H, J = 17.4, 9.0 Hz), 3.81-
3.92 (m, 1H), 4.68-4.74 (m, 1H), 5.25 (d, 1H, J = 8.7 Hz), 7.07-7.10(m, 2H), 7.25-7.28
(m, 3H); 13C NMR (75 MHz, CDC13): 8 -003, 15.09, 28.29, 43.29, 54.22, 59.43, 59.49,

89.42, 104.82, 126.91, 128.40, 128.74, 136.12, 155.70, 171.02; IR (neat film on NaCl):
3414 (m), 2957 (m), 2169 (w), 1727 (s), 1634 (m), 1413 (m), 1381 (m), 1243 (m), 1056

(m) cm ‘1; CI mass spectrum m/z (% rel intensity) 358 M4’ (62), 343 (35), 285 (63), 189

(100), 175 (48), 132 (46); Rf = 0.36 (1:1 hexane/EtOAc); [a]D -22.81°, (c 0.79, CHCI3).

O OH

aNJOL JK/'\ MeOMgB'
CHzc'z Jk/'\
MeO §
(81496

95 °’° TMS

(S, S, R)-166¢

aMethyl Ester (R)-196. Anhydrous methanol (1.50 g, 1.90 mL, 4.7 mmol) was

added to 60 mL CHZCIZ at 0 0C. A 3.0 M solution of MeMgBr in ether (1.72 mL, 5.2

mmol) was added dropwise via syringe, resulting in the formation of a white precipitate

and vigorous evolution of methane. A solution of aldol adduct (4S,SS,R)-l66c (1.68 g,
4.7 mmol) in 40 mL CHZCIZ at 0 °C was added via cannula, and the reaction was stirred

for 1 hr, at which time the white precipitate had disappeared, and TLC of the reaction

showed no remaining starting material.

183

 

The reaction was quenched by adding 30 mL saturated aqueous NaHCO3 and
stirring. The mixture was poured into a separatory funnel, and the aqueous layer was
extracted with 30 mL CH2C12. The combined organic layers were washed with 40 mL
water and 40 mL brine, dried with MgSO4, and concentrated to a sticky yellow solid. The

solid was washed with 5:1 hexane/EtOAc. The insoluble white solid was carefully

filtered off, and the yellow liquid was chormatographed on silica gel (5:1 hexane/EtOAc,
KMnO4 visualization) to give the product (R)-196 as a yellow oil in 62.3% yield (584

mg, 2.93 mmol). The insoluble white solid is the imidizolidinone chiral auxiliary, which

was recovered in 66% yield.

1H NMR (300 MHz, CDC13): 8 0.17 (s, 9H), 2.75 (d, 2H, J = 6.1 Hz), 2.99 (m,
1H), 3.73 (s, 3H), 4.77 (q, 1H, J = 6.0 Hz); 13C NMR (75 MHz, CDC13): 8 -027, 41.81,

51.93, 59.11, 85.26, 104.26, 171.61; IR (neat film on NaCl): 3500-3400 (m), 2959 (w),

2176 (w), 1742 (w), 1251 (m), 1060 (m), 844 (s) cm“; EI mass spectrum m/z (% rel

intensity) 199 M+ - 1 (11), 185 (100), 153 (36), 143 (83), 127 (47), 111 (76), 99 (55), 89

(73), 75 (68); Rf = 0.26 (5:1 hexane/EtOAc);

Me3Al O OH

 

O OH MeMeONHzHCI = Me,
”90% 9o % '1' \\
(F0496 TMS 0M6 (R)-175 TMS

aWeinreb amide (R)-17S. The aluminium amide reagent was prepared by adding

trimethylaluminum (2.0 M in hexane, 5.25 mL, 10.5 mmol) dropwise via syringe to a

184

stirring suspension of N, O-dimethyl hydroxylamine in 30 mL CHZCIZ at 0 °C. The
colorless solution was stirred for 45 minutes, then added via cannula to a solution of ester
(R)-l96 (957 mg, 4.78 mmol) in 20 mL CH2C12. The cold bath was removed and the

reaction allowed to stir overnight (16 h) at room temperature, during which time the

reaction color turned slightly yellow.

The reaction was quenched with excess aq. NH4C1 solution, added slowly to
avoid excessively rapid gas evolution, and poured into a separatory funnel. The organic
layer was washed with water (2 x 30 mL) and brine (1 x 30 mL), dried over MgSO4, and
concentrated to a pale yellow oil. The oil was chromatographed on silica gel (2:1
hexane/EtOAc), giving one fraction at Rf = 0.29 which was collected and concentrated to

(R)-l75 as a colorless oil in 90% yield (986 mg, 4.3 mmol).

1H NMR (400 MHz, CDC13): 8 0.18 (s, 9H), 2.83-2.90 (m, 2H), 3.21 (s, 3H), 3.72
(s, 3H), 4.81 (d, 1H, J = 5.1 Hz); 13C NMR (100 MHz, CDC13): 8 -032, 31.73, 38.65,
59.22, 61.35, 89.35, 104.91, 172.34; IR (neat film on NaCl): 3600-3200 (m), 2962 (m),

2174 (w), 1645 (s), 1436 (w), 1389 (m), 1250 (s), 1055 (m), 843 (s) cm]; EI mass
spectrum m/z (% rel intensity) 230 M” + 1 (12), 214 (30), 151 (62), 127 (100), 111 (17),
99 (95), 75 (80), 61 (70); R, = 0.29 (2:1 hexane/EtOAc — KMnO4); [(110 24.00 (c 1,

CHCI3); colorless oil. Yield: 986 mg (90%).

185

OHO 9H0

 

OMe
’ 2-methyl-1,3—dithiane, n-BuLi
// '1‘ > ¢ 3 s

55%
aDithiane (R)-l76. 2-Methyl-1,3-dithiane (2.66 g, 2.37 mL, 19.8 mmol, 2.1
equiv.) was dissolved in 50 mL THF at -—78 °C. A solution of n—BuLi (2.5 M in hexane,

7.92 mL, 19.8 mmol, 2.1 equiv.) was added via syringe. The reaction flask was put into a

0 0C cold bath and stirred for 30 minutes. The solution was then added via cannula to a
solution of Weinreb amide (R)-175 (2.16 g, 9.4 mmol) in 50 mL THF at 0 °C. The

reaction was monitored by TLC and done when checked after 1 h. The reaction was
quenched by adding acetic acid (1.13 mL, 19.8 mmol, 2.1 equiv.) neat via syringe and
briefly stirred. The mixture was poured into a separatory funnel containing 80 mL ether
and 80 mL water. The aqueous layer was back—extracted with ether (2 x 30 mL). The

combined organic layers were washed with water (2 x 50 mL) and brine (1 x 50 mL),
dried with MgSO4, and concentrated to a dark brown oil. This oil was chromatographed
on silica gel (5:1 hexane/EtOAc), giving unreacted/excess 2-methyl-1,3-dithiane at Rf =

0.65 and the product at Rf = 0.31, which was collected and concentrated to (R)-l76 as a

pale yellow oil in 55.2% yield (1.57 g, 5.2 mmol).

1H NMR (400 MHz, CDCl3): 6 0.16 (s, 9H), 1.65 (s, 3H), 1.82 (m, 1H), 2.17 (m,
1H), 2.62 (m, 2H), 2.97 (dd, 1H, J = 3.8, 17.3 Hz), 3.06 (tt, 2H, J = 2.7, 14.0 Hz), 3.42
(dd, 1H, J = 7.8, 17.3 Hz), 4.82 (dd, 1H, J = 3.8, 7.8 Hz); 13C NMR (100 MHz, CDC13): 6

-0.23, 23.92, 24.45, 27.90, 28.01, 42.67, 54.66, 59.53, 89.72, 104.82; IR (neat film on

186

 

NaCl): 3600-3200 (m), 2959 (m), 2900 (m), 2173 (w), 1707 (m), 1416 (w), 1250 (m),

844, (s), 760 (m) cm]; EI mass spectrum m/z (% rel intensity) 302 M+ (1), 269 (1), 195
(0.6), 176 (0.8), 133 (100), 111 (12), 59 (22); Rf = 0.31 (5:1 hexane/EtOAc); [(110 11.60

(c 1, CHCI3).

 

9H 0 9H CH
/ ’ MeaNHB(OAc)3 ?
/ S S . . : // s s
1:1aetonelacetl a d.
TMS (R)-175 '\/‘ 00C 2h C Cl TMS (R’R)-179 V

91%

'Diol (R,R)-l79 by Evans Reduction. See preparation of diol (R,R)-192 for the

procedure. This reaction was run on a 2.71 mmol scale. The product (R,R)-l79 was
isolated as fibrous white needles in 91 % yield (741 mg, 2.44 mmol) as a single

diastereomer.

1 H NMR (400 MHz, CDCI3): 6 0.18 (s, 9H), 1.37 (s, 3H), 1.83 (m, 2H), 2.12 (m,
1H), 2.41 (m, 1H), 2.59 (m, 2H), 3.09 (m, 2H), 4.67-4.70 (m, 2H); 13C NMR (100 MHz,
CDCI3): 6 -0.06, 21.29, 24.13, 25.59, 25.79, 35.93, 52.74, 61.82, 68.34, 89.32, 106.65; IR
(neat film on NaCl): 3600-3100 (m), 2895 (w), 2173 (w), 1249 (m), 1058 (m), 842 (s)
cm']; EI mass spectrum m/z (% rel intensity) 304 M+ (3), 164 (3), 133 (100), 99 (4), 73

(9), 59 (14); mp 112-113 °C; Rf = 0.28 (3:1 hexane/EtOAc); [01]D 32.90 (c 1, CHC13). Anal

calcd for C13H24OZSZSi: C 51.27, H 7.94. Found: C 51.32, H 8.05.

187

 

OH OH 9 o
? 2.2-oimethoxypropane, PPTS (20 mol%) *
¢ s s 3 // s 3
ms CHZCIZ, 85 oC (sealed tube), 24 h TMS
(R,R)-179 V 78% (R,R)-180

aAcetal (R,R)-180. A solution of diol (R,R)-l79 (32 mg, 0.105 mmol), freshly
distilled 2,2-dimethoxypropane (55 mg, 0.65 mL, 0.52 mmol), and PPTS (5.2 mg, 0.02
mmol) was dissolved in 1 mL dry CHZCI2 and stirred under argon at room temperature.

The reaction was followed by TLC, but the reaction did not appear to be proceeding after
48 h. The reaction mixture was transferred into a Schlenk flask, which was sealed and

heated to 85 0C for another 24 h. When checked by TLC after this period of heating, the
reaction had gone to completion. The reaction was diluted with 5 mL CHZCIZ‘ washed

with NaHCO3 (I x 5 mL), water (1 x 5 mL, and brine (1 x 5 mL), dried with MgSO4 and

concentrated to a yellow oil. The crude product was chromatographed on silica] gel
(using a 9” disposable pipet as the column) to give the product a pale yellow oil in 78%

yield (28 mg, 0.082 mmol).

1H NMR (400 MHz, CDC13): 8 0.17 (s, 9H), 1.41 (s, 3H), 1.46 (s, 3H), 1.58 (s,
3H), 1.93 (m, 1H), 2.00-2.05 (m, 2H), 2.23 (ddd, 1H, J = 4.2, 2.7, 10.2 Hz), 2.70 (m, 2H),
3.15 (m, 2H), 4.35 (dd, 1H, J = 4.7, 10.2 Hz), 4.70 (t, 1H, J = 6.5 Hz); 13C NMR (100
MHz, CDC13): 8 -021, 23.50, 23.81, 24.89, 26.81, 27.24, 34.05, 50.19, 61.82, 68.32,

74.16, 90.77, 101.26, 106.63; IR (neat film on NaCl): 2936 (m), 2169 (w), 1380 (m),

1249 (s), 1157 (w), 1106 (m), 1064 (w), 908 (m), 855 (s), 843 (s), 760 (m); EI mass

188

 

spectrum m/z (% rel intensity) 344 M+ (3), 286 (5), 271 (9), 211 (23), 153 (40), 133

(100), 109 ('15), 73 (36), 59 (26); R, = 0.55 (10:1 hexane/EtOAc).

OH OH gresoH

S é S S TBS-Cl,imidazole _ TMS // S S
TM DMF rt 2h
R,R-179 ’ ' R,R-181
( 1 K) 76% ( )

 

a'Protected Diol (R,R)-181. See preparation of (R,R)-202 for procedure. Reaction

was run on a 1.13 mmol scale. The product (R,R)-181 was isolated as a white solid in

76% yield (359 mg, 0.86 mmol).

1H NMR (400 MHz, CDC13): 8 0.17 (s, 9H), 0.19 (s, 6H), 0.88 (s, 9H), 1.39 (s,
3H), 1.69 (m, 1H), 1.87 (m, 1H), 2.09 (m, 1H), 2.41 (m, 1H), 2.60 (m, 2H), 3.03 (m, 2H),
4.35 (d, 1H, J = 10.1 Hz), 4.68 (d, 1H, J = 2.7 Hz); l3C NMR (100 MHz, CDC13): 8 —
5.08, 454, —0.21, 18.21, 21.65, 24.37, 25.77, 25.82, 25.89, 38.70, 52.87, 61.21, 67.24,

88.85, 107.31; IR (neat film on NaCl): 2955 (m), 2930 (m), 2856 (w), 2173 (w), 1472

(w), 1250 (m), 1063 (m), 841 (s), 779 (m) cm]; EI mass spectrum m/z (% rel intensity)
418 M+ (8), 361 (6), 285 (22), 255 (16), 201 (16), 153 (22), 133 (100), 107 (8), 73 (72);

mp 66-68 0C; Rf = 0.18 (50:1 hexane/EtOAc); [GJD 80.820 (c 1.05 in CHCI3).

189

 

 

1) 3:1CH30N/H20.rt.3h grasores
TMS é S 8 CAN 5 / l
- /
(R,R)-181 V 2)TBS§:,°/Imld,DMF TMS (R,R)-182 o
O

aKetone (R,R)-182. The dithiane (R,R)-181 (1.67 g, 3.12 mmol) was suspended in

a solution of 60 mL acetonitrile and 20 mL water. Solid cerium (IV) ammonium nitrate
(6.85 g, 12.5 mmol), was added in one portion, and the reaction was stirred for 5 minutes,
at which time the solid white suspension of dithiane had completely disappeared. The
reaction was diluted with 20 mL water and 50 mL ether and poured into a separatory

funnel. The aqueous layer was back-extracted with ether (2 x 25 mL), and the combined

organic layers were washed with NaHCO3 (1 x 50 mL), water (2 x 50 mL), and brine (l x

30 mL), dried with MgSO4 and concentrated to a pale yellow oil. The crude reaction

mixture was chromatographed on silica gel (50:1 hexane/EtOAc) to give the product

(R,R)-l82 as a pale yellow oil in 31% yield (429 mg, 0.97 mmol).

1H NMR (400 MHz, CDC13): 8 0.07 (s, 3H), 0.14 (s, 3H), 0.16 (s, 12H), 0.17 (s,
3H), 0.90 (s, 9H), 0.92 (s, 9H), 1.94 (m, 2H), 2.16 (s, 3H), 4.18 (dd, 1H, J = 5.3, 7.0 Hz),
4.52 (dd, 1H, J = 5.4, 8.0 Hz); 13C NMR (100 MHz, CDC13): 8 -4.86, -4.82, 477, 454,

-0.36, 18.10, 18.21, 25.39, 25.75, 25.83, 43.90, 59.48, 75.55, 89.95, 106.91, 210.68; IR

(neat film on NaCl): 2957 (m), 2930 (m), 2858 (m), 2173 (w), 1720 (m), 1472 (m), 1257
(s), 1092(8), 889 (s), 778 (s) cm"; EI mass spectrum m/z (% rel intensity) 427 M+-15 (2),
385 (28), 311 (3), 259 (49), 253 (43), 241 (80), 221 (9), 147 (31), 133 (11), 115 (14), 73

(100); Rf = 0.24 (50:1 hexane/EtOAc); [a]D 650° (c 1, CHCI3).

190

gresores 9,051 gresores

 

/ LDA. Cl’ ‘05; _
/ THF 0°C 1h 7 7
TMS (R.R)-182 o - ' TMS (Rm-183 0, ,0
42% ,P(
EtO OEt

aPhosphonate (R,R)-183. A solution of LDA was prepared by adding n-BuLi
(2.5 M solution in hexane, 0.4 mL, 1.0 mmol) to a solution of diisopropylamine (0.15
mL, 1.05 mmol, 1.05 equiv.) in 5 mL THF at -78 0C, then warming the reaction to room
temperature for 15 minutes then cooling back down to -78 °C. This solution was added to
a precooled (-78 CC) solution of ketone (R,R)-182 (420 mg, 0.95 mmol) in 5 mL THF.
The reaction was stirred at -78 ‘’C for 5 minutes, then warmed to 0 0C for 15 minutes and
cooled back down to -78 0C. Diethylchlorophosphonate (0.28 mL, 331 mg, 1.92 mmol),

was added neat via syringe, and the reaction was monitored by TLC for disappearance of
starting material. No reaction was observed after 15 minutes, so the reaction was allowed

to warm to room temperature. It was complete after 45 minutes.

The reaction was quenched with aq. NH4C1 solution and diluted with 10 mL water

and 20 mL ether. The reaction was poured into a separatory funnel and the aqueous layer

was back-extracted with ether (2 x 10 mL). The combined organic layers were washed
with water (2 x 20 mL) and brine (1 x 20 mL), dried with MgSO4 and concentrated to a
pale yellow oil. The oil was purified by chromatography on silica gel (10:1
hexane/EtOAc — KMnO4). One fraction at Rf = 0.12 was isolated and concentrated to a

colorless oil in 42% yield (230 mg, 0.40 mmol).

191

1H NMR (400 MHz, CDC13): 6 0.07 (s, 3H), 0.09 (s, 3H), 0.14-0.17 (15 H,

overlapping TMS and TBS singlets), 0.89 (s, 9H), 0.91 (s, 9H), 1.37 (t, 6H, J = 6.5 Hz),
1.87 (m, 1H), 2.04 (m, 1H), 4.17 (m, 4H), 4.22 (m, 1H), 4.46 (m, 1H), 4.75 (t, 1H, J = 1.6
Hz), 4.98 (t, 1H, J = 1.6 Hz); 13C NMR (100 MHz, CDC13): 8 -500, 459, 437, —3.71, -

0.28, 16.03, 18.11, 25.81, 25.87, 45.33, 59.84, 64.34, 69.42, 89.52, 96.19, 107.35,

156.26; IR (neat film on NaCl): 2958 (m), 2930 (m), 2858 (m), 2172 (w), 1659 (w), 1472

(m), 1276 (w), 1251 (m), 1098 (s), 1034(8), 838 (s), 778 (s) cm']; EI mass spectrum m/z
(% rel intensity) 563 M+-15 (6), 521 (95), 424 (4), 397 (13), 367 (12), 315 (9), 267 (7),
211 (27), 183 (11), 155 (35), 109 (6), 75 (100); Rf = 0.12 (10:1 hexane/EtOAc); [01]D

225° (c 1.5, CHCI3).

OH OH 0 o
H ? (MeO)3CH.CSA H

 

s s Q s s Q
|\/| (R,R)—192 TMS V (R,R)-207 TMS

dTrimethyl Ortho Ester Derived (R,R)-207 (from (R,R)-l92). To a solution of
compound (R,R)-l92 (286 mg, 0.98 mmol) in 2 ml of CHZCI2 was added camphor
sulfonic acid (CSA, 5 mg) in one portion, 10 mg of 4A° molecular sieves and trimethyl

ortho ester (208 mg, 2 mmol) dropwise. The reaction was stirred for 48 h at room
temperature. After separation by flash chromatography (10% EtOAc in hexanes),
compound (R,R)-207 was isolated as a colorless oil 98% yield (319.4 mg, 0.96 mmol) in

a 6:1 diastereomeric ratio. Major isomer of (R,R)-207:

192

1H NMR (300 MHz, CDC13): 8 0.18 (s, 9H), 1.74 (dt, 1H, J = 2.10, 13.19, Hz),
1.86-2.04 (m, 1H), 2.04—2.30 (m, 2H), 2.74-2.89 (m, 2H), 2.89-3.02 (m, 2H), 3.48 (s,
3H). 4.08 (d, 1H, J = 5.77 Hz), 4.38 (ddd, 1H, J = 2.20, 5.77, 8.24 Hz), 4.96 (dd, 1H, J =

1.37, 5.49 Hz), 5.66 (s, 1H); 13C NMR (75 MHz, CDC13): 8 -033, 25.73, 29.16, 29.26,

32.50, 49.61, 52.32, 63.74, 74.51, 93.71, 101.27, 108.34. R,- = 0.50 (20% EtOAc in

hexanes)

OMe
0H OH 0X0
? (M80)3CH, CSA ?

r

 

s 3 § 3 s Q
K/I (R,R)-179 TMS V (R,R)-184 TMS

°Trimethyl Ortho Ester (R,R)-184. Procedure same as above, data not reported.

OMe

0A0 M0119 OH
H = DIBAL-H H =
U......? U..R..§

‘DIBAl Reduction Precursor to (R,R)-208. To a solution of the ortho ester
derivative (R,R)-207 (473 mg, 1.42 mmol) in 12 mL of CHzCl2 was added 7.1 mL of 1 M
DIBAL-H (7.1 mmol in hexanes) at —78 °C. After stirring for 1 hour at —78 °C, the

reaction warm up to 0 0C for 10 min. The reaction was quenched by aq. HCI (IN). The
reaction mixture was filtered through Celite and washed with methylene chloride (4 x

100 mL). The combined organic layers were washed aq. NH4C1, and brine (200 mL),

dried over MgSO4 and concentrated on the rotary evaporator. Purification of the crude

193

product by flash chromatography on silica gel (1:1 hexanes/EtOAc) affored 445 mg (1.32

mmol) of MOM mono-protected product (R,R)-208 as a colorless oil in 93% yield.

1H NMR (300 MHz, CDC13): 6 0.14 (s, 9H), 1.85 (m, 1H), 2.08 (m, 3H), 2.86
(m, 4H), 3.07 (d, 1H, J = 6.6 Hz), 3.44 (s, 3H), 4.11 (dt, 1H, J = 4.3, 9.1 Hz), 4.36 (d, 1H,

J = 4.4 Hz), 4.54 (m, 1H), 4.72 (d, 1H, J = 6.9 Hz), 4.78 (d, 1H, J = 6.9 Hz); l3C NMR
(75 MHz, CDCI3): 6 -0.28, 29.98, 30.15, 30.36, 39.77, 52.51, 56.25, 59.35, 77.15, 89.20,

97.19, 105.96. Rf = 0.14 (20% EtOAc in hexanes).

OMe

0 0 MOMQ 0H

? DBAbH ?
s s \\ s s \\
V (R,R)-184 TMS V (R,R)-185 TMS

°(R,R)-185 by DIBAI Reduction of (R,R)-184. Procedure same as above, data

not reported.

MOMQ 0H MOMQ OTBS
H ? resort, NEt3 H i

‘
v

s s \\ s s
K/l (R, R)-208 TMS |\/) (R,R)-1

*rns Protection (mm-193. To a solution of (R,R)-208 (212 mg, 0.63 mmol) in

§
93 TMS

3 mL CHZCI2 at room temperature, NEt3 (0.263 mL, 1.89 mmol) was dropwise added and
then TBSOTf (0.433 mL, 1.89 mmol) also dropwise added. The reaction mixture has
been stirred for 10 min and quenched with brine (100 ml). After extraction with CHZCI2

(3 x 30 ml) of reaction mixture, the combined organic layers were concentrated in vacuo.

194

 

Flash chromatography on silical gel with 10% EtOAc in hexanes gave 274.2 mg (0.61

mmol) of product (R,R)-l93 as colorless oil in 97% yield.

1H NMR (300 MHz, CDC13): 8 0.14 (s, 3H), 0.15 (s, 9H), 0.18 (s, 3H), 0.91 (s,
9H), 1.80-2.20 (m, 4H), 2.89 (m, 4H), 3.45 (s, 3H), 3.99 (dt, 1H, J = 3.6, 8.8 Hz), 4.47 (d,
1H, J = 3.6 Hz), 4.52 (dd, 1H, J = 3.3, 9.9 Hz), 4.73 (s, 2H); 13C NMR (75 MHz, CDC13):
8 -4.86, 404, -030, 18.10, 25.81, 26.26, 30.44, 30.73, 41.41, 53.32, 56.06, 59.76, 77.12,

89.16, 96.97, 107.02. Rf = 0.34 (10% EtOAc in hexanes).

MCMQ 0H MOMQ ores
= resort, NEt3 ?

A

\ \
S S \ \
K/l (R.R)-185 TMS U(R,R)-186 TMS

°TBS Protection (R,R)-186. Procedure same as above, data not reported.

 

H ' NBS(6eq) H =.
s s Q CHgCN/HZC? RR 194%
V (Rm-193 TMS 94% 0‘ ; )‘ TMS

dAldehyde (R,R)-194. A solution of 200 mg (0.448 mmol) of compound (R,R)-

193 in 5 mL acetonitrile was added to a solution of NBS (476 mg, 2.68 mmol) in aqueous

80% acetonitrile at 0 ‘’C, and was stirred for 10 min. The red reaction solution quickly

turned to an orange color. After quenching with saturated aqueous sodium sulfite, the

reaction mixture was extracted with 1:1 hexane-CHZCIZ. The organic phase was washed

195

 

with saturated aqueous NaCl solution. Chromatography on silica gel (20% EtOAc in

hexanes) provided aldehyde (R,R)-l94 (145.8 mg, 0.41 mmol) as a colorless oil in 91%

yield.

1H NMR (300 MHz, CDCI3): 6 0.14 (s, 3H), 0.17 (s, 9H), 0.18 (s, 3H), 0.92 (s,

9H), 1.95 (ddd, 1H, J = 3.9, 9.1, 14.3 Hz), 2.09 (ddd, 1H, J = 3.9, 9.3, 14.3 Hz), 3.44 (s,

3H), 4.12 (ddd, 1H, J = 1.7, 3.9, 16.7 Hz), 4.58 (dd, 1H, J = 3.9, 9.1 Hz), 4.71 (d, 1H, J =
6.9 Hz), 4.74 (d, 1H, J = 6.9 Hz), 9.86 (d, 1H, J = 1.7 Hz); 13C NMR (75 MHz, CDCI3): 8
492, 422, -035, 18.15, 25.78, 39.05, 56.12. 59.02, 79.72, 89.93, 97.23, 106.56.202.11.

Rf = 0.30 (20% EtOAc in hexanes).

 

MCMQ OTBS MOMQ ores
= NCSIAgN03 ?
\ fi- \
s s \ \
K/I (R,Rl-185 TMS CH3;N:2° 0 (R,R)-187 TMS

eKetone (R,R)-187. Data not reported.

 

M0M0 ores
M01119 ores LDA, CIPO(OEt)2 -.
t : \
\
Q TH: Et°\ ,0 (R,RHBB TMS
o (R,R)-187 TMS 40 4 '30-"

dPhosphonate (R,R)-188. Proceedure same as for (R,R)-103. Data not

 

reported.
MCMQ OTBS MOMQ OTBS
\ NBS. EtaN/TMSOTf : Br \\
o (R,R)-187 TMS 60% o (R.R)-189 TMS

dAcyl Bromide (R,R)-189. Data not reported.

196

MOMQ ores MOMQ ores

E10 P \
Br \\ ( )0th Et0> \
0 (RM-189 TMS 11° OEt o (Rm-209 TMS

dPhosphonate (R,R)-209. Data not reported.

 

0 OH H 0 0H
Me\ . . .
[H Q 1.3-Dlthlan6J1-BULI: S S Q TMS
OMe (R)-175 TMS 80% V (R)-191

dDithiane (R,R)-191 from Weinreb’s Amide (R,R)-175. To a solution of 1,3-
dithiane (48 mg, 0.40 mmol) in 50 mL THF was added n-BuLi (250 11L, 0.40 mmol) at

—78 °C. The reaction mixture was warmed up to 0 °C and stirred for 30 minutes, and then

the solution of adduct (R,R)-175 (50 mg, 0.14 mmol) in 30 ml THF was added dropwise.

The reaction was stirred for 30 min and quenched with acetic acid (1 eq). The solution

was diluted with ether (50 mL), washed with aq. NaHCO3 (1 x 20 mL), extracted with
CH2C12 and subjected to column chromatography. Product (R,R)-l91 was obtained in
79% yield (31.7 mg, 0.11 mmol) as a colorless oil after silica gel chromatography (Rf =

0.40, 1:4 EtOAc/hexanes).

1H NMR (300 MHz, CDC13): 8 0.17 (s, 9H), 1.94—2.20 (m, 2H), 2.57 (ddd, 1H, J

= 2.7, 2.7, 5.2 Hz), 2.61 (ddd, 1H, J = 2.74, 2.74, 5.22 Hz), 3.05 (dd, 1 H, J = 4.1, 16.8

Hz), 3.17 (dd, 1 H, J: 7.7, 16.8 Hz), 3.20 (ddd, 1H, J = 3.0, 4.9, 11.3 Hz), 3.25 (ddd, 1H,

J = 3.0.4.9, 11.3 Hz), 4.23 (s, 1H), 4.83 (dd, 1H, J = 4.1, 7.7 Hz); 13C NMR (75 MHz,

197

 

CDC13): 8 —0.30, 24.96, 25.84, 25.88, 46.83, 59.14, 90.01, 104.45.200.65; ”C DEPT
NMR (75 MHz, CDCI3): 8 -0.30 (CH3), 24.96 (CH2), 25.84 (CH2), 25.88 (CH2), 46.83
(CH and CH2). 59.14 (CH), 90.01 (C), 104.45 (C), 200.65 (C). R1. = 0.40 (1:4

EtOAc/hexanes); Anal calcd for C12H200252313 C 49.96, H 6.99. Found: C 49.85, H 6.96.

 

0 0 OH O 0”
Me, NJLN 1,3-Dithiane, n-BuLi H \
Q - S S \
.\-—/ TMS 92% V (R1491 TMS
Me‘ Ph

(4 S, 5R,R)-1 66¢

dDithiane (R,R)-191 from Imidazolidinone. To a solution of 1,3-dithiane (535

mg, 4.47 mmol) in 50 mL THF was added n-BuLi (2.5 M in hexanes, 1.79 mL, 4.47

mmol) at —78 °C. The reaction mixture was warmed to 0 oC and stirred for 1 h. A

solution of imidazolinone adduct (4S,SR,R)-l66c (550 mg, 1.54 mmol) in 30 mL THF

was added dropwise. The reaction mixture was immediately re-cooled to —78 (’C, stirred
overnight and quenched with acetic acid (3.98 mL, 2.65 mmol). The solution was diluted
with ether (50 mL), washed with aq. NaHCO3 (1 x 20 mL), extracted with CHZCI2 and
subjected to column chromatography on silica gel (R) = 0.40, 1:4 EtOAc/hexanes). The

product (R,R)-l91 was obtained in 92% yield (402 mg, 1.42 mmol) as a colorless oil.
Spectral data for the product from this reaction matched that for (R,R)-191 reported

above.

198

 

0 0H 0H 0H

 

H M64BH(OAC)3 H

\ 3 \
S S \ 1:1 acetone 3 S \
K) (4191 ms scream K) <88th ms

90%

dDiol (R ,R )-192 from Evan’s Reduction”. Tetramethylammonium
triacetoxyborohydride (3.32 g, 12.61 mmol) was dissolved in 10 mL acetone and 20 mL
acetic acid at 0 0C and stirred for 30 min. A solution of compound (R)-191 (562 mg, 1.94

mmol) in 10 mL acetone was added. The reaction mixture was stirred for 1 h, quenched
with excess saturated aqueous sodium potassium tartrate solution and diluted with 50 mL
ether. The aqueous layer was neutralized with solid K2C03 and the reaction mixture was
extracted with ether (3 x 50 mL). The combined organic layers were washed with aq.

NaHCO3 solution (50 mL), H20 (50 mL), and brine (50 mL), dried over MgSO4 and

concentrated on a rotary evaporator to give a white solid. Purification of the crude
product by flash chromatography on silica gel (1:1 hexanes/EtOAc) affored 90% yield
(506.34 mg, 1.75 mmol) of diol product (R,R)-l92 (20:1 ratio of antizsyn diastereomers)

as a white solid.

1H NMR (300 MHz, CDC13): 6 0.19 (s, 9H), 1.92—2.16 (m, 3H), 2.31 (ddd, 1H, J
= 2.2, 6.5, 14.3 Hz), 2.71 (ddd, 1H, J = 3.3, 8.0, 14.0 Hz), 2.92-3.02 (m, 2H), 3.80 (d, 1H,
J = 7.4 Hz), 3.90 (d, 1H, J = 6.4 Hz distinguishable proton), 4.45 (ddd, 1H, J = 2.2, 7.4,
9.8 Hz), 4.72 (dd, 1H, J = 3.3, 6.9 Hz); 13C NMR (75 MHz, CDC13): 6 016, 25.40,

26.95, 27.34, 40.21, 50.88, 60.87, 68.94, 89.67, 106.00; IR (neat film on NaCl): 3150-

3610 (w), 2957 (s), 2924 (s), 2901 (s), 2172 (s), 1423 (s), 1277 (s), 1250 (s), 1064 (s),

199

843 (s) cm]; EI mass spectrum m/z (% rel intensity) 290 M+ (8), 149 (10), 121 (17), 120

(36), 1 19 (100), 106 (8), 84 (10), 75 (13), 73 (15); Rf = 0.26 (40% EtOAc in hexanes).

 

9H OH HO OTBS
H = resort. NEH H =
\ =
S S \ -78 °C-rt S S \\
(Rm-192 TMS . ' (R r8202 TMS
ovemlght K/l ’

86%
(R,R)-202 : TBS-Monoprotection of Diol (R,R)-l92. To a cooled solution (—78
0C) of diol (R,R)-l92 (110 mg, 0.38 mmol) in 3.5 mL of CHZCIZ was added NEt3 (191
uL, 1.36 mmol) and TBSOTf (87.3 11L, 0.38 mmol). The solution was stirred overnight at
this temperature and allowed to warm to ambient temperature prior to quenching with

NaHCO3. The organic phase was extracted with CHZCIZ and dried over MgSO4 and

concentrated. Flash chromatography on silica gel (1:9 EtOAc/hexanes, Rf = 0.24) gave

(R,R)-l92 as a colorless oil in 86% yield (132.6 mg, 0.33 mmol).

1H NMR (300 MHz, CDCI3): 6 0.13 (m, 15H), 0.88 (s, 9H), 1.84-2.14 (m, 4H),
2.70-2.80 (m, 2H), 2.80- 2.98 (m, 2H), 3.27 (broad s, 1H), 3.93 (d, 1H, J = 6.3 Hz), 4.28
(t, 1H, J = 7.8 Hz), 4.69 (dd, 1H, J = 6.9, 3.6 Hz); 13C NMR (75 MHz, CDC13): 6 -0.52, -
0.46, 0.00, 18.2, 26.12, 26.20, 28.61, 28.98, 42.12, 53.53, 61.98, 69.97, 90.20, 106.59; IR
(neat film on NaCl): 3340 -3580 (w), 2955 (s), 2928 (s), 2349 (s), 1259 (s), 1095 (s), 841
(s) cm'l; EI mass spectrum m/z (% rel intensity) 404 M“ (15), 386 (13), 285 (18), 255
(14), 241 (28), 221 (13), 201 (42), 179 (10), 147 (49), 133 (28), 119 (69), 84 (30), 73

(100), 59 (20), 47 (13). Rf: 0.24 (1:9 EtOAc/hexanes).

200

HQ OTBS QTESOTBS

 

H TESOTf, NEt3 H
\ = \
S S \ -78 °C-rt S S \
TM ' TMS
V (R,R)-202 S ovemlght V (Rm-203
86%

(R,R)-203: TES Protection of Alcohol (R,R)-202. To a cooled solution (—78 °C)
of mono protected diol (R,R)-202 (100 mg, 0.25 mmol) in 5 mL of CHZCI2 was added
NEt3 (70 pL, 0.5 mmol) and TESOTf (79 pL, 0.35 mmol). The solution was stirred
overnight at this temperature and allowed to warm up to ambient temperature prior to

quenching with aq. NaHCO3. The organic phase was extracted with CHZCI2 and dried on

MgSO4 and concentrated. Flash chromatography on silica gel (Rf = 0.57, 1:9

EtOAc/hexanes) provided (R,R)-203 as a colorless oil in 86% yield (113.4 mg, 0.22

mmol).

1H NMR (300 MHz, CDCI3): 6 0.09-0.16 (m, 15H), 0.64 (dq, 6H, J = 4.15, J =
0.7 Hz), 0.87 (s, 9H), 0.97 (t, 9H, J = 4.9 Hz), 1.60-1.93 (m, 2H), 2.00-2.17 (m, 2H),
2.72-2.92 (m, 4H), 4.10 (quintet, 1H, J = 3.9 Hz), 4.19 (d, 1H, J = 3.4 Hz), 4.46 (dd, 1H,
J = 4.4, 4.9 Hz). IR (neat film on NaCl): 2957 (s), 2930 (s), 2897 (m), 2857 (s), 1250 (s),
1093(8), 839 (s) cm]; EI mass spectrum m/z (% rel intensity) 518 M+ (5), 365 (4), 262
(16), 241 (100), 207(4), 181 (6), 147 (16), 115 (14), 87 (15), 73 (42), 59 (12). R: 0.57
(1:9 EtOAc/hexanes); Anal calcd for C24H5002525133 C 55.60, H 9.65. Found: C 55.31, H

10.0.

201

 

OH OH 1. TBSOTf, NE13 QTESOTBS

 

H \ 2. TESOTf _ H \
S S \ -78 °C-rt S S \

TM - RR 203 TMS
K/i (R,R)-192 S overnight K/l ( r

86%

(R,R)-203: One-Pot Protection of Diol (R,R)-l92. To a cooled solution (-78 °C)

of diol (R,R)-192 (37 mg, 0.127 mmol) in 5 mL of CHZCI2 was added NEt3 (89.1 11L,

0.635 mmol) and TBSOTf (29.2 11L, 0.127 mmol). After all the starting material was
consumed as indicated by TLC, TESOTf (40.2 11L, 0.178 mmol) was added to the
solution. The solution was stirred overnight at this temperature and allowed to warm up

to ambient temperature prior to quenching with aq. NaHCO3. The organic phase was
extracted with CHZCIZ and dried on MgSO4 and concentrated. Flash chromatography on

silica gel (Rf: 0.57, 1:9 EtOAc/Hexanes) gave (R,R)-203 as a colorless oil in 100% yield

(66.8 mg, 0.127 mmol). Spectral data for the product from this reaction matched that for

(R,R)-203 reported above.

 

gresores gresores
H i N88. 080%. H i
\ = \
S S \ \
V (Rm-203 TMS C”;8"1"”2° O (R,R)-201 TMS

Aldehyde (R,R)-201. A solution of protected diol (R,R)-203 (54 mg, 0.104 mmol)

in 1 mL of acetone was added to a solution of NBS (111.3 mg, 0.625 mmol) and CaCO3
(416 mg, 4.16 mmol) in 90% aqueous acetonitrile at 0 °C, and was stirred for 10 min. The

white suspension quickly acquired a yellow coloration. After quenching with saturated

202

aqueous sodium sulfite, the reaction mixture was extracted ether. The organic phase was
washed with saturated aq. NaCl solution. Chromatography on silica gel (20% EtOAc in

hexanes) provided the aldehyde (R,R)-201 as a colorless oil in 91% yield (40.4 mg, 0.095

mmol). A crude 1H NMR indicated that the material was satisfactory and could be used

for the next step without further purification.

lH NMR (300 MHz, CDCl3): 6 0.11 (s, 3H), 0.13 (s, 9H), 0.14 (s, 3H), 0.61 (q,
6H, J = 7.8 Hz), 0.87 (s, 9H), 0.94 (t, 9H, J = 7.8 Hz), 1.86-1.95 (m, 1H), 1.99-2.11 (m,
1H), 4.18 (dt, 1H, J = 5.1, 1.5 Hz), 4.55 (dd, 1H, J = 5.1, 3.0 Hz), 9.61 (dd, 1H, J = 1.2,

0.5 Hz). Rf: 0.68 (1:4 EtOAc/hexanes).

 

MOMQ OTBS MOMQ OTBS

H ‘ \ PDC. DMF : M80 = \
\

O (R,R)‘19‘ TMS CH3OH O (R-R)’195\ TMS

85%

dMethyl Ester (R,R)-l95. To a solution of aldehyde (R,R)-l94 (150 mg, 0.42

mmol) in methanol (100 uL, 25 mmol) and dry dimethylformamide (5 mL) at room
temperature was added pyridinium dichromate (950 mg, 25 mmol) and the reaction
mixture stirred for 40 h. The solution was poured into hexanes (200 mL)/water (50 mL),
filtered over Celite and then the water layer was extracted with hexanes (3 x 50 mL). The
combined hexanes extracts were dried over magnesium sulfate. Removal of the solvent
on a rotary evaporator gave the methyl ester (R,R)-195 as colorless oil in in 85% yield

(139 mg, 0.36 mmol). The crude product was used for the next step.

203

1H NMR (300 MHz, CDC13): 5 0.14 (s, 3H), 0.16 (s, 9H), 0.19 (s, 3H), 0.92 (s,
9H), 2.05-2.12 (m, 2H), 3.40 (s, 3H), 3.75 (s, 3H), 4.26 (dd, 1H, J = 3.8, 9.2 Hz), 4.58

(dd, 1H, J = 3.8, 9.2 Hz), 4.69 (s, 2H); 13C NMR (75 MHz, 030,): 0 -4.86, 4.11, -033,

18.13, 25.78, 41.96, 51.96, 56.28, 59.15, 72.77, 89.49, 96.90, 106.61, 173.10. Rf = 0.45

(10% EtOAc in hexanes).

QTESOTBS QTESOTBS

H '2, NaHCO3 M30
\ = \\
o (R,R)-201\ TMS M60" 0 (R,R)-200 TMS
83 % over 2 steps

 

Methyl Ester (R,R)-200. To a solution of aldehyde (R,R)-201 (22.4 mg, 0.053
mmol) in methanol (1 mL) was added NaHCO3 (17.4 mg, 0.21 mmol) and 12 (39.5 mg
0.312 mmol). The reaction was stirred for 36 h at room temperature and then quenched

slowly with aq. NaS204 at 0 0C slowly. The organic phase was extracted with EtOAc (3 x
10 mL), washed once with aq. Na8204 and (3 x 10 mL) with brine. The combined

organic layers was dried on Na2804 and concentrated. The crude product (R,R)-200 was

isolated in 91% yield (21.8 mg, 0.048 mmol) as a colorless oil and was used for the next

step without any further purification.

‘H NMR (300 MHz, CDCl3): a 0.03 (s, 3H), 0.06 (s, 3H), 0.12 (s, 9H), 0.62 (q,
6H, J = 7.4), 0.88 (s, 9H), 0.93 (t, 9H, J = 2.75 Hz), 1.80-2.00 (m, 1H), 2.00-2.14 (m,

1H), 3.69 (s, 3H), 4.37 (dd, 1H, J = 8.5, 4.4 Hz), 4.53 (dd, 1H, J = 9.1, 4.7 Hz); 13C NMR

204

 

(75 MHz, CDCl3): 6 -4.3, -3.8, -0.33, 5.0, 7.0, 18.6, 26.2, 44.8, 54.0, 59.8, 68.9, 89.9,

108.2, 174.5; EI mass spectrum m/z (% rel intensity) 458 M+ (2), 443 (4), 401 (64), 269

(23), 241 (50), 227 (56), 215 (9), 189 (24), 147 (40), 89 (38), 73 (100).

 

O OTBS
M80 ' = \ P \
\ 72% MeO/ \
(R,Rr195\ ms
0 (Rm-33

dTriol Fragment (R,R)-3a. To a solution of dimethyl methyl phosphonate (37.6
11L, 0.347 mmol) in 2 mL of THF at —78 0C was added n-BuLi (0.23 mL, 0.368 mmol).

After 1 h, a solution of ester (R,R)-195 (73 mg, 0.16 mmol ) in 2 mL of THF was added

and the reaction mixture allowed to warm to ambient temperature. After another hour at

this temperature, the solution was quenched with 5 mL of saturated aq. NH4C1 and
diluted with CH2C12 (30 mL). The aqueous solution was extracted with CHZCI2 (3 x 30
mL) dried on MgSO4, then concentrated down to a yellow oil. Column chromatography

on silica gel with 1:2:17 CH3OH/EtOAc/hexanes gave (R,R)-3a as a yellow oil in 72%

yield (79.2 mg, 0.115 mmol).

1H NMR (300 MHz, CDCI3): 6 0.14 (s, 3H), 0.15 (d, 9H, J = 0.6 Hz), 0.18 (s,

3H), 0.92 (d, 9H, J = 0.6 Hz), 1.98 (m, 2H), 3.24 (m, 2H), 3.37 (d, 3H, J = 0.6 Hz), 3.78
(m, 3H), 3.82 (m, 3H), 4.31 (dd, 1H, J = 9.07, 3.02 Hz), 4.54 (dd, 1H, J = 9.1, 3.6), 4.65

(s, 2H); 13C NMR (75 MHz, c003): 0 4,914.20, -037, 18.10, 25.75, 36.84 (d, J =

205

132.8 Hz), 40.39, 52.98 (m), 56.14, 59.24, 79.50 ((1, J = 0.4 Hz), 89.85, 97.02, 106.40,

202.15. Rf = 0.76. (1:2: 17 CH3OH/EtOAc/hexanes).

 

o ores
QESOTBS (MeO)2POCH2Li 1|
M90 . : M90\ P \
\ 82% M60] \
(R, R)-200\ ms
0 TMS ores
O (RM-3b

Triol Fragment (R,R)-3b. To a solution of dimethyl methyl phosphonate (87.4

11L, 0.81 mmol) in 2 mL of THF at —78 0C was added n-BuLi (0.531 mL, 0.85 mmol).

After 1 h, a solution of ester (R,R)-200 (170 mg, 0.37 mmol ) in 2 mL of THF was added
and the reaction mixture allowed to warm to ambient temperature. The reaction was held
overnight at room temperature. The work-up procedure was identical to that given above
for (R,R)-3a. Phosphonate (R,R)—3b was isolated as a yellow oil in 82% yield (166.6 mg,

0.30 mmol).

1H NMR (300 MHz, CDC13): 6 0.13 (m, 15H), 0.62 (q, 6H, J = 8.3 Hz), 0.93 (m,
18H), 1.80- 1.92 (m, 1H), 1.95-2.50 (m, 1H), 3.11 (dd, 1H, J = 14.9, 7.1 Hz), 3.33 (dd,
1H, J = 14.9, 7.08 Hz), 3.75 (s, 3H), 3.78 (s, 3H), 4.33 (dd, 1H, J = 5.4, 1.5 Hz), 4.48 (dd,
1H, J = 5.6, 2.2 Hz); 13C NMR (75 MHz, CDC13): 6 -4.2, -3.8, 0.0, 5.0, 7.0, 18.6, 26.0,
35.7 (d, J = 115.5 Hz), 43.8, 56.5, 59.9, 75.8, 90.4, 106.9, 204.0. IR (neat film on NaCl):
2957 (s), 2918 (s), 2851 (s), 1726 (s), 1462 (s), 1521 (s), 1035 (s), 841 (s) cm]; El mass
spectrum m/z (% rel intensity) 535 M+ -15 (10), 521 (20), 421 (15), 389 (70), 367 (35),
333 (18), 309(13), 287 (19), 241 (72), 181 (27), 147 (30), 129 (25), 87 (52), 73 (100), 57

(31); Rf = 0.85 (5:2 pentane/ether).

206

9H OH
H ? 1. TBSOTf, NEt3 H

\
8 § -78 004.1. 3d 3 S \

s
l\/[ (R,R)-192 TMS 100% K/l (R.R)-204 TMS

QTBSOTBS

 

(R,R)-204 : Di-TBS Protection of Diol (R,R)-l92. To a cooled solution (—78 °C)

of diol (R,R)-l92 (77 mg, 0.27 mmol) in 2.0 mL of CHzClz was added NEt3 (151.6 11L,

1.08 mmol) and TBSOTf (170 11L, 0.74 mmol). The solution was stirred at this
temperature for 3 days while being allowed to warm up to ambient temperature. After

three days the reaction was quenched with NaHCO3. The organic phase was extracted
with CHZCIZ and dried over MgSO4 and concentrated. Flash chromatography on silica gel

(Rf = 0.80, 5:19 EtOAc/hexanes) gave the protected diol (R,R)-204 in 100% yield (140

mg, 0.27 mmol) as a colorless oil.

1H N MR (300 MHz, CDCI3): 6 0.02 (s, 3H), 0.06 (s, 3H), 010-014 (s, 12H), 0.16
(s, 3H), 0.87 (s, 9H), 0.88 (s, 9H), 1.74-1.94 (m, 2H), 1.96-2.15 (m, 2H), 2.72-2.90 (m,
4H), 4.04-4.12 (m, 1H), 4.20 (d, 1H, J = 3.3 Hz), 4.45 (dd, 1H, J = 9.2, 4.1 Hz); 13C
NMR (75 MHz, CDCl3): 6 -4.86, -4.75, —3.93, -3.31, -0.61, 17.83, 25.50, 25.57, 26.14,
30.10, 30.48, 43.41, 55.09, 59.59, 70.95, 89.17, 106.92, (1 Sp3 C not located); IR (neat

film on NaCl): 2957 (s), 2930 (s), 2897 (s), 2856 (s), 1251 (s), 1093 (s), 839 (s), 777 (5);

EI mass spectrum m/z (% rel intensity) 518 (4), 503 (3), 461 (5), 397 (2), 387 (9), 355

(3), 263 (25), 241 (50), 147 (50), 133 (20), 73 (100), 59 (15). Rf = 0.80 (5:19

207

EtOAc/hexanes); [61D41.0°, (c 2.0, acetone). Anal calcd for (22411500252513: c 55.54, H

9.71. Found: C 55.93, H 9.35.

 

QTBSOTBS 1) NBS CaCO 9TBSOTBs
H T o 3. M90 3
K} (R'RHM TMS 2) 12. Nance, 0 (R,R)-205 ms

Dithiane Removal To Give (R,R)-205. A solution of protected diol (R,R)-204

(200 mg, 0.386 mmol) in 1 mL of acetone was added to a solution of NBS (417 mg, 2.32

mmol) and CaCO3 (1.54 g, 15.4 mol) in aqueous 90% acetonitrile at 0 c’C, and was stirred

for 10 mins. The white suspension quickly turned to a yellow coloration. After quenching
with saturated aqueous sodium sulfite, the reaction mixture was extracted ether. The

organic phase was washed with saturated NaCl solution. The crude aldehyde was
dissolved in methanol (8 mL) and then NaHCO3 (129 mg, 1.54 mmol) and I2 (294 mg,
2.32 mmol) were added. The reaction was stirred for 36 h at room temperature and then
quenched with aq. NaS204 at 0 0C slowly. The organic phase was extracted with EtOAc

(3 x 10 mL), washed once with Na8204 and (3 x 10 mL) with brine. The combined

organic layers were dried over NaZSO4 and concentrated. The crude product (R,R)-205

was isolated as a colorless oil in 95% yield (168 mg, 0.367 mmol) and used in the next

step without any further purification.

208

 

1H NMR (300 MHz, CDC13): 6 0.03 (s, 3H), 0.05 (s, 3H), 0.12 (s, 3H), 0.13 (s,
9H), 0.15 (s, 3H), 0.88 (s, 18H), 1.90-2.16 (m, 2H), 3.68 (s, 3H), 4.41 (dd, 1H, J = 8.5,
3.8 Hz), 4.53 (dd, 1H, J = 8.9, 4.4 Hz); 13C NMR (75 MHz, CDC13): 6 -5.23, -4.82, -
4.44, -3.74, -0.31, 18.17, 18.23, 25.75, 25.87, 44.57, 51.76, 59.41, 68.73, 89.16, 106.96,
173.98; FAB mass spectrum m/z (% rel intensity) 459 M+ +1 (3), 443 (1), 401 (3), 369
(4), 327 (5), 277 (12), 241 (5), 185 (100), 93 (70), 73 (15), 57 (5). Anal calcd for

C22H4604Si3: (3 57.59.11 10.10. Found: C 57.68, H 10.11.

grasorss 0TBS

MeO

 

o
(MeO)2POCH2Li. n-Bu_Li M804
R 205% 99% M80/ \\

0 ( 'R)‘ was 0733

o (R,R)-3c

Triol Fragment (R,R)-3c. To a solution of dimethyl methyl phosphonate (76 1.1L,

0.70 mmol) in 10 mL of THF at —78 ‘,C was added n-BuLi (0.416 mL, 0.67 mmol). After

1 h a solution of ester (R,R)-205 (160 mg, 0.35 mmol ) in 2 mL of THF was added and
the reaction mixture allowed to warm to ambient temperature. The reaction was held

overnight at room temperature. The work-up procedure was identical to that described
above for (R,R)-3a. After purification on silica gel (Rf = 0.81, 5:2 pentane/ether)

phosphonate (R,R)~3c was isolated as a yellow oil in 99% yield (190.5 mg, 0.347 mmol).

1H NMR (300 MHz, CDCI3): 6 0.09 (s, 6H), 0.15 (s, 3H), 0.16 (s, 9H), 0.18 (s,

3H), 0.91 (s, 9H), 0.93 (s, 9H), 1.84-2.10 (m, 2H), 3.13 (dd, 1H, J = 22.0, 15.1 Hz), 3.35

209

(dd, 1H, J = 21.2, 15.4 Hz), 3.78 (s, 3H), 3.82 (s, 3H), 4.35 (dt, 1H, J = 5.2, 1.7 Hz), 4.52
(dt, 1H, J = 5.5, 2.2 Hz); 13C NMR (75 MHz, CDCl3): 6 -4.62, -4.60, 4.29, -3.69, -0.09,
18.43, 26.04, 26.11,. 35.29 (d, J = 80.1 Hz), 43.67, 53.14, 59.78, 75.80, 90.44, 107.04,
203.50, (1 Sp3 C not located); FAB mass spectrum m/z (% rel intensity) 551 (M + H) (8),
535 (8), 493 (15), 419 (28), 361 (10), 295 (15), 287 (22), 241 (12), 73 (100); Rf = 0.81

(5:2 pentane/ether).

 

gresoras QTBSOTBS
MeO . \ K2003. MeOH ; M60 1 \
\ 99% \

o (R.R)-205 TMS o (RRHOG H

Methyl Ester (R,R)-206: To a solution of alkyne (R,R)-205 (40 mg, 0.088 mmol)
in MeOH (2.5 mL) at 0 0C was added K2CO3 (24 mg, 0.176 mmol). The mixture was
stirred 2 h at 0 0C and then filtered through a sintered glass funnel lined with Celite. The

Celite bed was washed with 3 x 10 mL of EtOAc and then the combined organic layers
were concentrated. The crude (R,R)-206 was isolated as a white solid in 99% yield (33.6

mg, 0.087 mmol) and was used in the next step without any further purification.

‘H NMR (300 MHz, CDCl3): o 0.04 (s, 3H), 0.06 (s, 3H), 0.11 (s, 3H), 0.15 (s,
3H), 0.88 (s, 18H), 2.01 (ddd, 1H, J = 13.5, 8.0, 4.8 Hz), 2.12 (ddd, 1H, J = 13.7, 8.3, 4.4
Hz), 2.39 (d, 1H, J = 2.2 Hz), 3.69 (s, 3H), 4.37 (dd, 1H, J = 7.8, 4.3 Hz), 4.56 (ddd, 1H,
J = 8.0, 4.9, 2.0 Hz); 13C NMR (75 MHz, c1303): 6 -5.60, -513, -5.10, 4.32, 17.80,

17.89, 25.39, 25.47, 44.20, 51.42, 58.45, 68.42, 72.79, 84.63, 173.74.

210

97330733 QTBSOTBS

M80 CH3PO(OM8)2, n-Bu L1 ; M60

 

\ o 3P; Q
o (R,R)-206\ H 6m M30 6 0 WW“ H

 

Triol Fragment]

 

 

Trio] Fragment 3d. To a solution of dimethyl methyl phosphonate (28.6 111., 0.264

mmol) in 5 mL of THF at —78 0C was added n-BuLi (151 pL, 0.242 mmol, 1.6 M). After
1 h, a solution of ester (R,R)-206 (34 mg, 0.088 mmol) in 5 mL of THF was added and
the reaction mixture stirred at —78 0C for an additional 2 hours. The reaction was

quenched with 10 mL of saturated aq. NH4CI and then diluted with CHZCIZ (30 mL). The

aqueous solution was extracted with CHZCIZ (3 x 30 mL) dried on MgSO4, then

concentrated to an oil. Column chromatography on silica gel with 6:4 pentane/EtOAc

gave (R,R)-3d a colorless oil in 67% yield (28.2 mg, 0.059 mmol).

1H NMR (300 MHz, CDCl3): o 0.06 (s, 3H), 0.07 (s, 3H), 0.11 (s, 3H), 0.14 (s,
3H), 0.87 (s, 9H), 0.89 (s, 9H), 1.78-2.20 (m, 2H), 2.42 (d, 1H, J = 2.2 Hz), 3.12 (dd, 1H,
J = 22.0, 15.1 Hz), 3.35 (dd, 1H, J = 21.2, 14.8 Hz), 3.74 (s, 3H), 3.78 (s, 3H), 4.33 (1,
1H, J = 6.0 Hz), 4.50 (dt, 1H, J = 8.0, 1.9 Hz); ”C NMR (75 MHz, c1303): 6 4.64, -
4.50, -3.85, 18.34, 18.42, 26.02, 26.06, 26.11, 35.40 (d, J = 80.6 Hz), 43.54, 53.20, 59.14,

74.04, 75.64, 84.96, 203.70, (1 Sp3 C not located). Rf = 0.45 6:4 pentane/EtOAc).
a- Data obtained from Mark Parisi’s Thesis;56
b- Data obtained from Yan Shi’s Thesis;7'

c- Data obtained from unpublished results of Kenneth Wilson and W.D. Wulff;58

d- Data obtained from unpublished results of Xuejun Lui and W.D. Wulff;88

211

e— Data obtained from unpublished results of Su Yu and W.D. Wulff.57

212

Experimental data for Chapter 4.

\
I

—

-
o
a

 

O J\
0J\ EtO - 0
O +
/O

1
p
510 O t-BuOK, toluene O
I \/ ‘ l o
o -78 °c - r.t / \/
(R).Za 209 ovemlght (PO-210 0

96%

HWE Olefination (R)-210. A solution of phosphonate 209 (23.2 mg, 0.104
mmol) and the purified major isomer of aldehyde (R)-2a (30 mg, 0.177 mmol) in 10 mL

of anhydrous toluene at —78 °C was treated dropwise with t-BuOK (0.152 mL, 0.152
mmol, 1.0 M in THF). The reaction mixture was allowed to warm up to 0 0C slowly and

stirred at 0 °C overnight. The reaction was mixture was quenched by addition of 10 mL of

saturated aqueous NaCHO3. The organic layers were combined, dried with N32504,

concentrated and chromatographed with pentane/ether (4:1). Ethyl ester (R)-210 was
obtained as a colorless oil in 96 % yield (24.2 mg, 0.101 mmol).

1H NMR (300 MHz, CDC13): o 1.07 (dd, 6H, J = 6.0, 1.7 Hz), 1.28 (t, 3H, J = 7.1
Hz), 2.04—2.10 (m, 2H), 3.96 (sept, 1H, J = 6.3 Hz), 4.18 (q, 2H, J = 7.2 Hz), 4.58-4.66
(m, 1H), 5.12 (d, 1H, J = 2.5 Hz), 5.68-5.79 (m, 1H), 5.95-6.05 (m, 1H), 6.08 (dd, 1H, J
= 15.9, 1.9 Hz), 6.95 (dd, 1H, J = 15.7, 4.1 Hz); IR (neat film on NaCl): 2916 (m), 2849
(m), 2363 (m), 2338 (m), 1718 (m), 1653 (s), 1558 (s), 1458 (s), 1030 (m); Rf = 0.46 (4:1

pentane/ether).

213

/l\ O OTBS 9*
Q MeOa‘I ‘

 

P .
, M80, \\ LICI, NEt3. THF; l 0 OTBS
O + TMS
| ores 94% \
/0 I \
o TMS
(R)-2a (R,R)-3b OTES
o
(Ramqn

HWE Olefination (R)-211. A solution of LiCl (3.35 mg, 0.0797 mmol) in 0.5 mL
of THF was added to a solution of phosphonate (R,R)-3b (40.4 mg, 0.0736 mmol) in 3

mL of THF at room temperature and stirred for 5 minutes. The solution was then cooled

to 0 °C and Et3N (10.30 (1L, 0.0736 mmol) was added and the solution stirred for 30

minutes at ambient temperature. At this point, the solution was re-cooled to 0 0C and the
purified major isomer of aldehyde (R)-2a (12.5 mg, 0.0736 mmol) was added dropwise in
1 mL of THF. The solution was stirred for 24 h at ambient temperature before being
quenched with H20 (5 mL) and extracted with ether (10 mL). The organic layer was

washed with brine and dried over MgSO4. Column chromatography on silica gel (2:5
ether/pentane) provided ketone (R,R,R)-211 in 94 % yield (41.1 mg, 0.069 mmol) as a

colorless oil.

1H NMR (300 MHz, CDC13): 6 0.04 (s, 9H), 0.13 (s, 6H), 0.58 (q, 6H, J = 8.0
Hz), 0.80-0.96 (m, 18H), 1.16 (d, 3H, J = 6.1 Hz), 1.18 (d, 3H, J = 6.1 Hz), 1.85 (ddd,
1H, J = 4.4, 8.8, 13.6 Hz), 1.96-2.10 (m, 3H), 3.98 (sept, 1H, J = 6.1 Hz), 4.44 (dd, 1H, J
= 3.9, 8.3 Hz), 4.52 (dd, 1H, J = 8.8, 4.4 Hz), 4.58-4.66 (m, 1H), 5.13 (broad s, 1H),
5.70-5.79 (m, 1H), 5.96-6.04 (m, 1H), 6.67 (dd, 1H, J = 13.9, 1.7 Hz), 6.95 (dd, 1H, J =

11.7, 3.9); 13c NMR (75 MHz, CDC13): 6 -4.36, -3.62, 1.24, 5.15, 7.03, 18.45, 22.30,

214

 

24.00, 26.10, 30.13, 44.41, 59.88, 65.53, 70.10, 74.78, 89.96, 93.41, 107.25, 123.60,
126.57, 128.19, 146.27, 200.80; IR (neat film on NaCl): 3045 (s), 2959 (m), 2928 (m),
2857 (S), 2174 (m), 1703 (m), 1632 (S), 1462 (m), 1259 (m), 1096 (m), 1032 (m), 841 (S),
802 (s); EI mass spectrum m/z (% rel intensity) 594 M+ (1), 565 (1), 537 (2), 505 (1), 477
(2.5), 461 (l), 433 (2), 425 (2), 411(5), 403 (2), 271 (9), 241 (100), 161 (7), 87 (15), 73

(39), 59 (6). Rf = 0.85 (5:2 pentane/ether).

OTBS

n.6 Sb
CK}, LiCl NEt3 THF Cfiw/A
(R)- 23 (R, R) -3c

(R, R, R)-21 2

 

HWE Olef'mation-Ketone (R,R,R)-212: A solution of LiCl (16.8 mg, 0.4 mmol)
in 2 mL of THF was added to a solution of phosphonate (R,R)-3c (87 mg, 0.182 mmol) in
5 mL of THF at room temperature and stirred for 5 minutes. The solution was then
cooled to 0 0C, Et3N (35.5 (1L, 0.255 mmol ) was added and the solution stirred for 30

minutes at ambient temperature. At this point, the solution was re-cooled to 0 °C and the

purified major isomer of aldehyde (R)-Za (37.2 mg, 0.218 mmol) was added dropwise in

1 mL of THF. The solution was stirred for 24 h at ambient temperature before being
quenched with H20 (5 mL) and extracted with ether (20 mL). The organic layer was
washed with brine and dried on MgSO4. Column chromatography on silica gel (2:5

ether/pentane) gave ketone (R,R,R)-212 in 99 % yield (107.0 mg, 0.180 mmol) as a

colorless oil.

215

 

1H NMR (300 MHz, CDCl.,): 5 0.03 (s, 6H), 0.12 (s, 12H) 0.16 (s, 3H) 0.88 (s,
18H), 1.15 (d, 3H, J = 6.1 Hz), 1.17 (d, 3H, J = 6.1 Hz), 1.80-1.92 (m, 1H), 1.92-2.18 (m,
3H), 3.97 (sept, 1H, J = 6.3 Hz), 4.38 (dd, 1H, J = 8.2, 4.1 Hz), 4.52 (dd, 1H, J = 8.8, 4.4
Hz), 4.52-4.68 (m, 1H), 5.12 (broad s, 1H), 5.73 (d, 1H, J = 10.4 Hz), 5.94-6.04 (m, 1H),
6.68 (d, 1H, J = 15.7 Hz), 6.94 (dd, 1H, J = 15.5.4.4 Hz); 13C NMR (75 MHz, CDC13): o
4.86, -4.56, 4.45, —3.74, -031, 18.17, 22.00, 23.81, 25.81, 25.90, 29.83, 44.17, 59.53.
65.33, 69.78, 74.71, 89.81, 93.04, 107.02, 123.40, 126.31, 127.92, 145.95, 200.62, (1 sp3
C not located); IR (neat film on NaCl): 3045 (s), 2961 (s), 2928 (s), 2857 (s), 2174 (s),
1700 (s), 1636 (s), 1464 (s), 1401 (s), 1362 (s), 1318 (s), 1260 (s), 1096 (s), 1032 (s), 839
(s), 802 (s), cm“; EI mass spectrum m/z (% rel intensity) 595 (M*+1) (3), 537 (3), 477
(2), 461 (4), 419 (3), 411 (2), 405 (3), 403 (6), 397 (2), 377 (2) 363 (2), 331 (2), 271 (5),

241 (100), 227 (5), 147 (30), 73 (90); Rf = 0.82 (5:2 pentane/ether), [0t]D +6.60 (6 1.0,

C5H12). Anal calcd for C31H5305Si3: C 62.57; H 9.82. Found: C 62.23; H 9.50.

)\ O OTBS Q/I\
9 M60.” '

 

P .

_ Meo’ § 1.101. N513, THFA I O ores

l O + ores H 90°/

/0 ° 1 §

0 H

(R)-2a (R, R)-3d 01133

0
(R,R,R)-213

HWE Olefination-Ketone (R,R,R)-213: A solution of LiCl (3.2 mg, 0.076
mmol) in 0.5 mL of THF was added to a solution of phosphonate (R,R)-3d ( 18 mg, 0.038

mmol) in 1 mL of THF at room temperature and stirred for 5 minutes. The solution was

216

 

then cooled to 0 ‘’C, Et3N (7.4 11L, 0.053 mmol) was added and the solution stirred for 30
minutes at ambient temperature. At this point, the solution was re—cooled to 0 0C and the

purified major isomer of aldehyde (R)-Za (7.7 mg, 0.045 mmol) was added dropwise in

0.5 mL of THF. The solution was stirred for 24 h at ambient temperature before being

quenched with H20 (1 mL) and extracted with ether (5 mL). The organic layer was

washed with brine and dried over MgSO4. Column chromatography on silica gel (2:5

ether/pentane) gave ketone (R,R,R)-213 in 90% yield (17.7 mg, 0.034 mmol) as a

colorless oil.

'H NMR (300 MHz, CDCl3): o 0.04 (s, 6H), 0.12 (s, 3H) 0.16 (s, 3H) 0.88 (s,
9H), 0.89 (s, 9H), 1.16 (d, 3H, J = 5.7 Hz), 1.18 (d, 3H, J = 5.8 Hz), 1.85-2.16 (m, 4H),
2.41 (dd, 1H, J = 0.8, 1.4 Hz), 3.97 (sept, 1H, J = 6.6 Hz), 4.38 (dd, 1H, J = 8.0, 4.4 Hz),
4.53 (td, 1H, J = 6.9, 2.2 Hz), 4.574.64 (m, 1H), 5.12 (d, 1H, J = 2.2 Hz), 5.72 (dd, 1H, J
= 9.9, 1.9 Hz), 5.94-6.04 (m, 1H), 6.69 (d, 1H, J = 16.8 Hz), 6.95 (dd, 1H, J = 15.7, 4.1
Hz); 13c NMR (75 MHz, CDCl3): o 4.85, —4.69, 4.53, -395, 18.18, 22.03, 23.82, 25.80,
25.85, 29.85, 44.21, 59.00, 63.35, 69.84, 73.39, 74.73, 85.05, 93.09, 123.38, 126.33,

127.92, 146.07, 200.59, (1 sp3 C not located); IR (neat film on NaCl): 3312 (s), 2963 (s),
2926 (s), 2855 (s), 1738 (s), 1373 (s), 1260 (s), 1094 (s), 1022 (s), 800 (s), cm’l; FAB

mass spectrum m/z (% rel intensity) 523 (M+ +1) (7), 463 (7), 423 (2), 405 (3), 391 (2),
349 (3), 331 (7), 327 (7), 325 (6), 251 (10), 193 (10), 169 (60), 147 (15), 73 (100);
HRMS calcd for C28H50058i2 m/z 523.3275, meas 523.3276. Rf = 0.75. (5:2

pentane/ether).

217

AlMe3, CHZCIZ

r.t, 3d
73% d.r = 10:1

(based on SM recovered)
(R, R,R)-211 (R,R,R,R)-220

   

Tertiary Alcohol (R,R,R,R)-220: To a solution of ketone (R,R,R)-211 (8 mg,
0.014 mmol) in CH2C12 (3 mL) was added at —15 0C, AlMe3 (2.0 M, 0.056 mmol). The
reaction mixture was warmed to 0 0C and stirred at that temperature for 3 h. After no

change in TLC occurred, the reaction mixture was raised to ambient temperature and

stirred for 3 days. The flask was then recooled to 0 CC and 2 mL of H20 was added

slowly. The organic portion was extracted with CH2C12 (3 x 5 mL), dried over MgSO4

and concentrated. Column chromatography on silica gel gave (R,R,R,R)-220 in 48% yield
(3.6 mg, 0.0067 mmol) as a colorless film. The starting material ketone (R,R,R)-211 was
recovered in 40% yield (3.2 mg, 0.0057 mmol). The yield of (R,R,R,R)-220 based on
starting material recovered is 73%. Tertiary alcohol (R,R,R,R)-220 was isolated as a 10:1
inseparable mixture of diastereomers. The ratio was determined by intergration of the

hydroxyl proton (6 = 2.23 ppm major, 6 = 2.26 ppm minor). The stereochemistry of the

major diastereomer was assumed to be the C8(R) epimer based on the stereochemistry

observed by Just32 and Boger24in a similar addition to a related molecule.

The following spectral data was taken on a 10:1 mixture of diastereomers.

218

Major isomer: 1H NMR (300 MHz, CDC13): o 0.14 (s, 6H), 0.63 (q, 6H, J = 8.0
Hz), 0.83-1.04 (m, 18H), 1.14 (d, 3H, J = 6.0 Hz), 1.19 (d, 3H, J = 6.3 Hz), 1.23 (s, 3H),
1.58-1.70 (m, 1H), 1.92-2.10 (m, 3H), 2.23 (s, 1H), 2.41 (d, 1H, J = 2.2 Hz), 3.67 (t, 1H,
J = 5.5 Hz), 3.98 (sept, 1H, J = 6.3 Hz), 4.38-4.54 (m, 2H), 5.09 (s, 1H), 5.69 (d, 1H, J =
10.4 Hz), 5.74-5.86 (m, 2H), 5.92-6.00 (m, 1H). Rf = 0.60 (4 :l pentane/ether).

Minor isomer: 6 = 2.26 (s, 1H) only distinguishable proton.

9k

 

 

l 0 ores
MeLi-CeCl37H20
| \\ THF, -78°c
TMS
ores 90% d.r=3:1 ores
0 Me OH
(R,R,R)-211 (R,R,R,R)-217

Tertiary Alcohol (R,R,R,R)-217: CeCl3.7H20 (892 mg, 1.7 mmol) was
heated from room temperature to 100 °C overnight under vacuum (0.2 mmHg). At 70 0C
to 100 °C heating was allowed to proceed slowly. The temperature was then raised to 140
0C and kept there for 12 h. At this point the reaction flask was allowed to cool to room

temperature under argon and 2 mL of THF was added to a grayish-white solid and stirred

for 10 h. The solution was then cooled to -78 °C and MeLi (1.49 mL, 1.69 mmol) was
added. This reaction was stirred for 10 min at -78 0C and 10 min at room temperature.
The reaction mixture was beige/orange in color. The flask was then re-cooled to -78 0C

and a single epimer at C1 of ketone (R,R,R)-211 (18 mg, 0.032 mmol) was added in 1 mL

219

of dry THF and stirred at the same temperature for 10 mins. A saturated solution of
NaHCO3 (2 mL) was used to quench the reaction, which was extracted with CHzClz (3 X

5 mL) and chromatographed (4:1 pentane/ether) on silica gel. Alcohol (R,R,R,R)-217
was obtained in 90 % (17.5 mg, 0.029 mmol) as a colorless oil. Tertiary alcohol
(R,R,R,R)-217 was isolated as a 3:1 inseparable mixture of diastereomers. The ratio was

determined by intergration of the hydroxyl proton (6 = 2.31 ppm major, 6 = 2.36 ppm

minor). The stereochemistry of the major diastereomer was assumed to be the C3(R)

epimer based on the stereochemistry observed by Just32 and Boger24 in a similar addition

to a related molecule.

The following spectral data was collected on a 3:1 mixture of diastereomers. The
1H NMR and 13C NMR data for the major isomer were extracted from the spectrum of
the mixture.

Major isomer: 1H NMR (300 MHz, CDCl3): 6 0.04 (s, 9H), 0.10-0.l4, (m, 6H),
0.63 (q, 6H, J = 7.8 Hz), 0.87 (s, 9H), 0.94 (dt, 9H, J = 2.7, 7.8 Hz), 1.15 (d, 3H, J = 6.0
Hz), 1.18-1.26 (m, 6H), 1.58-1.68 (m, 1H), 1.92—2.13 (m, 3H), 2.31 (d, 1H, J = 7.0 Hz),
3.64-3.73 (m, 1H), 3.99 (sept, 1H, J = 5.7 Hz), 4.41-4.48 (m, 2H), 5.10 (broad s, 1H),
5.74 (d, 1H, J = 10.1 Hz), 5.76-5.82 (m, 2H), 5.95-6.02 (m, 1H).

Minor isomer: 6 = 2.36 (d, 1H, J = 6.7 Hz), only distinguishable proton.

13c NMR (75 MHz, CDCl.,): 6 4.38, -3.68, —0.27, 5.38, 6.95, 18.14, 21.97, 23.90,

25.87, 25.98, 29.68, 30.75, 43.10, 60.88, 69.40, 74.89, 75.57, 89.50, 93.03, 107.48,

126.12, 128.51, 129.53, 135.11; IR (neat film on NaCl): 3474 (w), 3045 (s), 2959 (s),

220

2930 (s), 2901 (s), 2858 (s), 2174 (s), 1464 (s), 1383 (s), 1260 (s), 1096 (s), 1030(8), 839
(s), 806 (s), 777 (s) cm"; FAB mass spectrum m/z (% rel intensity) (M+-17) 593 (0.7),
551 (0.4), 493 (0.5), 486 (0.6), 485 (0.6), 461 (0.7), 419 (2), 401 (1), 399 (1), 385 (0.6),
366(1), 341 (l), 327 (3), 325 (4), 311 (l), 309 (l), 295 (1), 281 (7), 267 (4), 241 (70),
147 (30), 73 (100); HRMS calcd for C32H6104Si3 m/z 593.38778, meas 593.3874. Rf =

0.56 (4:1 pentane/ether).

The 1H N MR spectrum of (R,R,R,R)-217 was also taken in CD3CN to compare to
Boger’s tertiary alcohol (R,R,R,R)—50. Major isomer: lHNMR (300 MHz, CD3CN): 6 -
0.03-0.16 (m, 15H), 0.32 (dq, 6H, J = 7.8, 4.4 Hz), 0.54 (s, 9H), 0.62 (t, 9H, J = 7.8 Hz),
0.76 (d, 3H, J = 5.9 Hz), 0.82 (d, 3H, J = 6.1 Hz), 0.86 (s, 3H), 1.12-1.23 (m, 1H), 1.45-
1.59 (m, 3H), 2.83 (d, 1H, J = 2.4 Hz), 3.63-3.72 (m, 1H), 3.89-3.97 (m, 1H), 4.37 (m,
1H), 4.47-4.59 (m, 1H), 5.07 (broad s, 1H), 5.63-5.68 (m, 1H), 5.69-5.90 (m, 2H), 5.92-
5.99 (m, 1H).

Minor isomer: 6 = 2.79 (d, 1H, J = 10 Hz), only distinguishable proton.

221

 

 

I 0 ores
MeLi-CeCl3_7H20
I \\ THF, -78°C
0TBS 99% d.r = 7:1
0
(R, R, R)-21 2 (R,R,R,R)-221

Tertiary Alcohol (R,R,R,R)-221: CeCl3.7H20 (892 mg, 1.7 mmol) was heated
from room temperature to 100 °C overnight under vacuum (0.2 mmHg). At 70 0C to 100
oC heating was allowed to proceed slowly. The temperature was then raised to 140 0C and

kept there for 12 h. At this point the reaction flask was allowed to cool to room
temperature under argon and 2 mL of THF was added to a grayish-white solid and stirred

for 10 h. The solution was then cooled to -78 °C and MeLi (1.49 mL, 1.69 mmol) was
added. This reaction mixture was stirred for 10 min at -78 0C and 10 min at room

temperature. The solution was beige/orange at this point and was then re-cooled to -78 OC
and a single epimer at C1 of ketone (R,R,R)-212 (50 mg, 0.09 mmol) was added in 2 mL

of dry THF and stirred at -78 °C for 10 min. A saturated solution of NaHCO3 (2 mL) was

used to quench the reaction, which was extracted with CHZCl2 (3 X 5 mL) and

chromatographed (4:1 pentane/ether) on silica gel. Alcohol (R,R,R,R)-221 was obtained
in 99 % (54.4 mg, 0.089 mmol) as a colorless oil. Tertiary alcohol (R,R,R,R)-221 was
isolated as a 7:1 inseparable mixture of diastereomers. The ratio was determined by

intergration of the hydroxyl proton (6 = 2.29 ppm major, 6 = 2.35 ppm minor). The

stereochemistry of the major diastereomer was assumed to be the C8(R) epimer based on

222

the stereochemistry observed by Just32 and Boger24 in a similar addition to a related

molecule.

Major isomer: 1H NMR (300 MHz, CDCl3): 6 0.03-0.15 (m, 21H), 0.86 (s, 18H),
1.14 (d, 3H, J = 6.3 Hz), 1.16-1.24 (m, 6H), 1.52-1.70 (m, 1H), 1.92-2.14 (m, 3H), 2.29
(s, 1H), 3.67 (t, 1H, J = 5.2 Hz), 3.98 (sept, 1H, J = 6.0 Hz), 4.36—4.50 (m, 2H), 5.08
(broad s, 1H), 5.69 (d, 1H, J = 7.3 Hz), 5.76-5.82 (m, 2H), 5.92-6.02 (m, 1H).

Minor isomer: 6 = 2.35 (s, 1H) only distinguishable proton.

13C NMR (75 MHz, CDCl3): 6 -4.38, -4.37, -3.85, -3.75, -0.29, 18.18, 18.19,
21.97, 23.89, 25.87, 25.98, 29.67, 30.75, 43.10, 60.87, 66.07, 69.06, 74.96, 75.48, 89.51,
93.02, 107.48, 126.13, 128.49, 129.52, 135.09; IR (neat film on NaCl): 3466 (w), 2963
(s), 2930 (s), 2901 (s), 2859 (s), 2174 (s), 1201 (s), 1095 (s), 1022 (s), 839 (s), 800 (s),
cm'l; FAB mass spectrum m/z (% rel intensity) (M+-1) 609 (0.1) 593 (0.7), 551 (0.4), 493
(0.5), 461 (1), 419(2), 397(1), 349(1), 327(2), 325(1), 309(1), 241 (100), 147 (30), 73

(95); Rf = 0.50 (4:1 pentane/ether).

 

 

I 0 ores
MeLi-CeCl37H20
I Q H THF. -78°C
0TBS 98% d.r = 7:1
0
(R,R,R)-213 (R,R,R,R)—222

Tertiary Alcohol (R,R,R,R)-222: CeCl3.7H20 (132.4 mg, 0.36 mmol) was

heated from room temperature to 100 °C overnight under vacuum (0.2 mmHg). At 70 0C

223

to 100 0C heating was allowed to proceed slowly. The temperature was then raised to 140
0C and kept there for 12 h. At this point the reaction flask was allowed to cool to room

temperature under argon and 1 mL of THF was added to a grayish-white solid and stirred

for 10 h. The solution was then cooled to -78 0C and MeLi (0.216 mL, 0.346 mmol) was
added. This reaction was stirred for 10 min at —78 °C and 10 min at room temperature.

The solution was beige/orange at this point and was then re-cooled to —78 0C and a single
epimer at C1 of ketone (R,R,R)-213 (10.0 mg, 0.019 mmol) was added in 1 mL of dry

THF and stirred at -78 0C for 10 min. A saturated solution of NaHCO3 (1 mL) was used

to quench the reaction, which was extracted with CH2C12 (3 X 5 mL) and

chromatographed (4:1 pentane/ether) on silica gel. Alcohol (R,R,R,R)-222 was obtained
in 98 % (10.0 mg, 0.018 mmol) as a colorless oil. Tertiary alcohol (R,R,R,R)-222 was
isolated as a 7:1 inseparable mixture of diastereomers. The ratio was determined by

intergration of the hydroxyl proton (6 = 2.23 ppm major, 6 = 2.30 ppm minor). The
stereochemistry of the major diastereomer was assumed to be the C8(R) epimer based on
the stereochemistry observed by Just32 and Boger24 in a similar addition to a related

molecule.

Major isomer: 1H NMR (300 MHz, CDCI3): 6 0.07 (s, 3H), 0.11 (s, 3H), 0.13 (s,

3H), 0.15 (s, 3H), 0.88 (s, 18H), 1.15 (d, 3H, J = 6.2 Hz), 1.21 (d, 3H, J = 6.0 Hz), 1.24

(s, 3H), 1.58-1.70 (m, 1H), 1.95-2.08 (m, 3H), 2.23 (s, 1H), 2.41 (d, 1H, J = 2.7 Hz) 3.68

224

 

(t, 1H, J = 4.6 Hz), 3.99 (sept, 1H, J = 6.0 Hz), 4.40-4.52 (m, 2H), 5.09 (broad s, 1H),
5.70 (d, 1H, J = 7.3 Hz), 5.76-5.84 (m, 2H), 5.94-6.02 (m, 1H).

Minor isomer: 6 = 2.30 (s, 1H) only distinguishable proton.

13C NMR (75 MHz, CDCl3): 6 459, -4.30, -3.95, -3.91, 18.14, 21.97, 23.88,
25.83, 25.95, 29.67, 30.78, 43.34, 60.30, 66.04, 69.42, 73.02, 75.01, 75.54, 85.50, 93.03,
126.15, 128.48, 129.61, 134.93, (1 sp3 C not located); IR (neat film on NaCl): 3467 (w),
3312 (s), 2957 (s), 2928 (s), 2857 (s), 1258 (s), 1099 (s), 1022 (s), 839 (s), 800 (s), cm];
FAB mass spectrum m/z (% rel intensity) (M+-17) 521 (10), 479(6), 461 (5), 421 (8), 389
. (8), 347 (12), 327 (59), 267 (15), 169 (100), 147 (30), 129 (22), 115 (60), 97 (25), 75

(70), 73 (99); Rf = 0.34 (4 :1 pentane/ether).

Trimethyl silyl Deprotection of (R,R,R,R)-221

 

 

Me "’01"
MeOH, K2003, H20 (R,R,R,R)-222
0r ; +
AgN03. EtOH. KCN, H20 k
Or 0

Amberlyst. MeOH

 

Me ”OH
(R, R, R, R)-221

 

(R, R, R, R)-224

225

Methanol and Potassium Carbonate: To a solution of a 7:1 diastereomeric

mixture of alkyne (R,R,R,R)—221 (10 mg, 0.018 mmol) in MeOH (1 mL) at 0 0C, was
added K2C03 (5.0 mg, 0.036 mmol) and H20 (0.125 mL). The mixture was stirred 3 h
after which it was quenched with H20 (1 mL), extracted with EtOAc (3 X 5 mL) and

dried on MgSO4. The solution was concentrated on the rotary evaporator and

chromatograghed on silica gel with EtOAc/pentane 1:20. Compound (R,R,R,R)-224 was
obtained in 40% yield (3.8 mg, 0.007 mmol) in a 7:1 ratio as a colorless oil and
compound (R,R,R,R)-222 was obtained in 60% yield (5.8 mg, 0.011 mmol) in a 7:1 ratio

as a colorless oil. Compounds (R,R,R,R)-224 and (R,R,R,R)-222 were isolated in a 2:3
ratio and had an Rf values of 0.30 and 0.34 respectively in 20% ether in pentane. The

spectral data for compound (R,R,R,R)-222 matches that reported above from the methyl

addition to (R,R,R)-213. Compound (R,R,R,R)-224 was assigned as the compound having
the C8 and C11 hydroxyls protected with TBS based on HMQC and HMBC data of a

related compound, (R,R,R,R)-247. Secondary alcohol (R,R,R,R)-224 was isolated as a 7:1
inseparable mixture of diastereomers. The ratio was determined by intergration of the

hydroxyl proton (6 = 2.75 ppm major, 6 = 2.59 ppm minor). The stereochemistry of the

major diastereomer was assumed to be the C8(R) epimer based on the stereochemistry

observed by J ustz'2 and Boger24 in a similar addition to a related molecule.

Silver Nitrate, Ethanol and Potassium Cyanide: To a solution of alkyne (R,R,R)-

221 (10 mg, 0.018 mmol) in EtOH (1 mL) at 0 c’C, was added dropwise a solution of

AgNO3 (7 mg, 0.041 mmol) dissolved in H20 (0.3 mL) and EtOH (0.7 mL). Stirring was

226

 

continued for 1 h and KCN (12 mg, 0.18 mmol) was added neat. The mixture was stirred

for 2.5 h, diluted with ether, washed with H20 (10 mL) and brine (10 mL), and dried on
MgSO4. The solution was concentrated on the rotary evaporator and and the crude oil
chromatographed on silica gel as described above. The yield and ratio of (R,R,R,R)-222

and (R,R,R,R)-224 were the same as listed above in the preparation using MeOH, K2C03

and H20. Secondary alcohol (R,R,R,R)—224 was isolated as a 7:1 inseparable mixture of

diastereomers. The ratio was determined by intergration of the hydroxyl proton (6 = 2.75

ppm major, 6 = 2.59 ppm minor). The stereochemistry of the major diastereomer was

assumed to be the C8(R) epimer based on the stereochemistry observed by Just32 and

Boger24in a similar addition to a related molecule.

Amberlyst A-26 (Chloride [on Form) in Methanol: To a solution of alkyne
(R,R,R)-221 (10 mg, 0.018 mmol) in MeOH (1 mL) was added amberlyst resin A-26 (Cl

ion form, 8.0 mg) which was prewashed with MeOH. The reaction was run overnight and
then filtered and washed sequentially with MeOH (5 mL), EtZO (5 mL) and CH2C12 (5

mL). The combined rinses were concentrated in vacuo and the crude oil chromatographed

on silica gel as described above. The yield and ratio of (R,R,R,R)-222 and (R,R,R,R)-224
were the same as listed above in the preparation using MeOH, K2C03 and H20.

Secondary alcohol (R,R,R,R)-224 was isolated as a 7 :1 inseparable mixture of
diastereomers. The ratio was determined by intergration of the hydroxyl proton (6 = 2.75

ppm major, 6 = 2.59 ppm minor). The stereochemistry of the major diastereomer was

227

 

assumed to be the C8(R) epimer based on the stereochemistry observed by Just32 and

Boger24in a similar addition to a related molecule.
Characterization of ( R,R,R,R)-222 and (R,R,R,R)-224.

1H NMR spectrum obtained for tertiary alkynol (R,R,R,R)-222 matches that

reported above from the methyl addition to (R,R,R)-213.

Major isomer: 1H NMR for tertiary alkynol (R,R,R,R)-224: 6 0.04 (s, 3H), 0.06 (s,
3H), 0.10 (s, 3H), 0.12 (s, 3H), 0.85 (s, 9H), 0.87 (s, 9H), 1.15 (d, 3H, J = 6.0 Hz), 1.20
(d, 3H, J = 6.6 Hz), 1.23 (s, 3H), 1.75-1.89 (m, 1H), 1.98-2.10 (m, 3H), 2.35 (s, 1H), 2.75
(broad s, 1H) 3.68 (d, 1H, J = 9.4 Hz), 3.97 (sept, 1H, J = 6.0 Hz), 4.38-4.48 (m, 1H),
4.55—4.64 (m, 1H), 5.08 (broad s, 1H), 5.65-5.84 (m, 3H), 5.94—6.04 (m, 1H).

Minor isomer: 6 = 2.59 (s, 1H) only distinguishable proton.

13C NMR (75 MHz, CDC13): 6 -5.18, -4.44, -1.98, —1.89, 18.35, 18.43, 22.41,
24.13, 25.93, 26.10, 29.92, 30.84, 39.67, 60.72, 66.42, 70.03, 72.36, 74.93, 85.68, 93.54,
126.48, 128.63, 130.81, 134.82, (1 sp3 C not located); IR (neat film on NaCl): 3468 (w),
3312 (s), 2957 (s), 2930 (s), 2859 (s), 1385 (s), 1254 (s), 1090 (s), 1032 (s), 1005 (s), 838

(s), 800 (s), 777 (s) cm"; R, = 0.30 (4 :1 pentane/ether).

228

MeOH. K2003. H20
0 °C, 3h, 90 %

 

   

(R, R, R)-21 2 (R, R, R)-225

Ketone (R,R,R-225): To a solution of alkyne (R,R,R)-212 (10 mg, 0.018 mmol) in

MeOH (0.5 mL) at 0 °C, was added 1<2c03 (5.02 mg, 0.036 mmol) and H20 (0.125 mL).
The mixture was stirred 3 h after which it was quenched with H20 (1 mL), extracted with

EtOAc (5 mL) and dried over MgSO4. The solution was concentrated on the rotary

evaporator and chromatograghed on silica gel with EtOAc/pentane 2:5. Ketone (R,R,R)-

225 was obtained as a 2:1 ratio of diastereomers in 90% (9.0 mg, 0.016 mmol) as a

colorless oil. It is assumed that the ratio of Michael products at C6 is the only unknown

stereogenic center since 1H NMR shows no epimerization at C9. Ketone (R,R,R,R)-225

was isolated as a 2:1 inseparable mixture of diastereomers. The ratio was determined by
intergration of the acetal proton (6 = 5.04 ppm major, 6 = 5.08 ppm minor). The

stereochemistry of the major diastereomer was not determined.

(Spectra was obtained as a 2:1 mixture): 1H NMR (300 MHz, CDCI3): 6 0.05 (s,

6H), 0.11 (s, 3H), 0.15 (s, 3H), 0.88 (m, 9H), 0.89 (s, 9H), 1.13-1.24 (m, 6H), 1.80-2.15
(m, 4H), 2.40-2.42 (m, 1H), 2.58 (ddd, 1H J = 28.4, 17.2, 3.0 Hz), 2.90 (ddd, 1H, J =
32.3, 17.3, 9.1 Hz), 3.37 (s, 3H), 3.75-3.88 (m, 1H), 3.90-4.04 (m, 2H), 4.26 (dd, 1H, J =

7.3, 5.2 Hz), 4.45-4.58 (m, 1H), 5.04 (s, 1H), 5.58-5.74 (m, 1H), 5.92-6.04 (m, 1H); 13C

229

NMR (75 MHz, CDC13): 6 -4.72, -4.03, 1.00, 18.15, 21.67, 23.84, 25.79, 25.82, 29.69,
39.60, 43.46, 58.93, 67.51, 68.91, 73.43, 74.71, 75.78, 78.80, 84.95, 92.50, 126.00,

128.28, 210.26, (2 sp3 C not located); IR (neat film on NaCl): 3312 (s), 2957 (s), 2930
(s), 2858 (s), 1727 (s), 1472 (s), 1385 (s), 1258 (s), 1098 (s), 1018 (s), 839 (s) cm']; FAB
mass spectrum m/z (% rel intensity) (M+-1) 553 (1), 495 (12), 463 (5), 461 (5), 437 (3),
405 (6), 385 (3), 363 (15), 353 (4), 331 (15), 169 (99), 147 (30), 136 (18), 115 (60), 73

(100); HRMS calcd for C30H5605812 m/z 553.3745, meas 553.3749. Rf = 0.67 (5:2

 

pentane/ether).
HO
1 OH
W /\ 5% Pdc'2(PPh3)2 TBso
/ / + __
T330 H // pyrrolidine — //
4a 152 87% 153 —

aDienyne 153. A 100 mL round-bottom flask was charged with PdC12(PPh3)2

(702 mg, 1.0 mmol) and dissolved in 30 mL freshly distilled pyrrolidene under argon.
The flask was wrapped with aluminium foil, and iododiene 4a (6.49 g, 20 mmol) as the
pure E,Z-isomer was added neat via cannula. The solution darkened slightly, and was
briefly stirred before 3-butyn-2-ol (1.402 g, 1.57 mL, 20 mmol) was added in one portion
via syringe. The reaction was stirred at room temperature and followed by TLC until the

starting material had disappeared.

The reaction was quenched by adding excess saturated NH4C1 solution at 0 °C,

and then the mixture was further diluted with 150 mL ether. The mixture was poured into

a separatory funnel and the layers were separated. The aqueous layer was extracted with

230

ether (2 x 60 mL). The combined organic layers were washed with saturated NH4C1 (1 x
150 mL), saturated NazSZO3 (1 x 100 mL), water (2 x 100 mL), and brine (1 x 80 mL),
dried over anhydrous MgSO4, and concentrated to a thick brown oil. The oil was taken up
in approximately 30 mL of ether and stored at —40 °C overnight, giving an orange

solution containing a precipitated orange solid. The solid was filtered off through Celite,
and the orange solution was concentrated to an orange oil. This oil was purified by
chromatography on silica gel (4:1 pentane/ether — UV visualization) to give the product

153 in 87% yield (4.66 g, 17.5 mmol) as an orange oil.

1H NMR (300 MHz, CDCI3): 6 0.09 (s, 6H), 0.88 (s, 9H), 1.50 (d, 3H, J = 6.5
Hz), 4.29 (d, 2H, J = 4.3 Hz), 4.68 (dq, 1H, J = 1.7, 6.6 Hz), 5.42 (d, 1H, J = 10.4 Hz),
5.95 (dt, 1H, J = 15.3, 4.6 Hz), 6.40 (t, 1H, J = 10.9 Hz), 6.78 (m, 1H); 13C NMR (75
MHz, CDCI3): 6 -4.84, 18.79, 24.83, 26.05, 59.36, 63.51, 81.44, 97.35, 108.20, 126.70,
137.04, 140.20; IR (neat film on NaCl): 3360 (m), 2980-2850 (m), 1463 (s), 1362 (m),

1256 (m), 1073 (s), 837 (m), 777 (m) cm"; Rf = 0.38 (4 :1 pentane/ether).

 

HO
anCu-Ag T380
TBSO . — _
_ // 2:1 MeOH/1120 _ OH
153 — 80% 154

“Activated Metal Reduction to Give Triene 154. A 100 mL flask was charged

with zinc dust (10 g, 0.154 mol, 99.9%, 150-325 mesh, Alfa/Aesar), suspended in 50 mL

HPLC grade water and sparged with argon for 15 min. Anhydrous copper (II) acetate (1.0

231

 

g, 0.006 mol) was added, the flask was capped with a rubber septum, and the slurry was
stirred for 15 minutes. Silver nitrate (1.0 g, 0.006 mol) was then added and the flask
warmed noticeably while stirring was continued for 30 minutes. The black suspension of
activated metal was isolated by filtration on a Buchner funnel followed by sequential
washings with HPLC grade water, methanol, acetone, and ether.

The black solid was immediately added to a solution of dienyne 153 (133 mg, 0.5
mmol) in 15 mL of 2:1 methanol/water. The contents of the flask was placed under an
argon atmosphere and stirred for 20 h. The reaction mixture was filtered through Celite
and the black metal filter cake was rinsed with 50 mL ether. The liquid was poured into a
separatory funnel and the aqueous layer was extracted with ether (2 x 30 mL). The

combined organic layers were washed with brine (1 x 50 mL), dried over anhydrous
MgSO4, and concentrated to a yellow oil. The oil was purified by chromatography on
silica gel (5:1 hexane/EtOAc, UV/KMnO4 visualization) to give the product triene 154

(100 mg, 0.37 mmol) in 74.6% yield as a pale yellow oil. TLC showed only one spot at

Rf = 0.28 (5:1 hexanes/EtOAc). No over reduced products were isolated or observed.

1H NMR (400 MHz, CDC13): o 0.09 (s, 6H), 0.93 (s, 9H), 1.29 (d, 3H, J =
6.3 Hz), 4.27 (d, 2H, J = 1.35 Hz), 4.82 (m, 1H), 5.52 (t, 1H, J = 10.1 Hz), 5.81 (dt, 1H, J
= 15.0.4.9 Hz), 6.07 (t, 1H, J = 10.9 Hz), 6.18 (t, 1H, J = 11.5 Hz), 6.40 (t, 1H, J = 11.4
Hz), 6.67 (t, 1H, J = 11.2 Hz); 13C NMR (100 MHz, CDC13): o -537, 14.66, 23.35, 25.84,

63.40, 63.83, 122.92, 123.99, 124.27, 130.63, 135.23, 135.56; IR (neat film on NaCl):

3370 (w), 2959 (m), 2855 (m), 1426 (w), 1253 (m), 1121 (m), 1056 (m), 835 (s), 775 (m)

232

cm'l; E1 mass spectrum m/z (% rel intensity) (M+-29) 239 (2), 226 (2), 211 (3), 197 (6),
183 (5), 169 (7), 145 (7), 117 (43), 89 (46), 75 (100), 59 (23); RI = 0.28 (5 : 1

hexane/EtOAc); Anal calcd for CISHZSOZSi: C 67.11, H 10.51. Found: C 67.13, H 10.46.

HO T880

T830 // TBSOTf,NE13 _ T880

100% 2 — //

153 155

 

Dienyne 229. The proceedure was the same as that for compound 202 and was
run on a 0.188 mmol scale. Dienyne 229 (71.4 mg, 0.188 mmol) was obtained as a

colorless oil in 100% yield.

‘H NMR (300 MHz, c1303): 0 0.06 (s, 6H), 0.11 (d, 6H, J = 3.0 Hz), 0.89 (s,

9H), 0.90 (s, 9H), 1.42 (d, 3H, J = 6.6 Hz), 4.59 (d, 2H, J = 3.3 Hz), 4.67 (dq, 1H, J =
1.9, 6.6 Hz), 5.4 (d, 1H, J: 10.7 Hz), 5.88 (dt, 1H, J: 15.1, 5.2 Hz), 6.33 (t, 1H, J = 11.0

Hz), 6.68-6.80 (m, 1H); 13(3 NMR (75 MHz, CDCI3): o -531, 4.99, 4.60, 18.17, 18.31,
25.43, 25.76, 25.88, 59.48, 63.29, 80.20, 97.83, 108.43, 126.72, 136.02, 139.11, (1 sp3 c

not located). Rf = 0.2 (100:1 hexane/EtOAc).

233

TBSO // 2:1 MeOH/F120 — OTBS
60% 156

155

Activated Metal Reduction to Give Triene 156. The procedure was the same as

that for compound 154, and was run on a 0.0635 mmol scale. Triene 156 (14.5 mg, 0.038

mmol) was obtained as a colorless oil in 60% yield. TLC showed only one spot at Rf =

0.36 (99:1 pentane/EtOz). N 0 over reduced products were isolated or observed.

1H NMR (300 MHz, CDC13): o 0.01 (s, 3H), 0.03 (s, 3H), 0.05 (s, 3H), 0.06 (s,
3H), 0.85 (s, 9H), 0.90 (s, 9H), 1.19 (d, 3H, J = 6.3 Hz), 4.24 (d, 1H, J = 4.9 Hz), 4.76
(quintet, 1H, J = 7.4 Hz), 5.48 (t, 1H, J = 10.4 Hz), 5.81 (dt, 1H, J = 14.8, 5.0 Hz), 6.04
(t, 1H,J=11.3 Hz), 6.14 (t, 1H, J =11.3 Hz), 6.30 (t, 1H, J: 11.3 Hz), 6.68 (dd, 1H, J =
14.8, 11.2 Hz); 13C NMR (75 MHz, CDCl3): o 495,447,420, 18.45, 24.91, 24.99,

26.11, 63.74, 65.25, 121.62, 123.59, 124.76, 129.81, 134.81, 137.69, (1 sp3 C not

located); IR (neat film on NaCl): 2957 (m), 2928 (m), 2857 (m), 2363 (m), 2336 (m),
1653 (s), 1474(8), 1458 (s), 1256 (s), 1123 (m), 1078 (m), 1005 (m), 835 (m), 775 (m);
E1 mass spectrum m/z (% rel intensity) 382 M+ (18), 325 (18), 250 (38), 237 (45), 189

(18), 147 (100), 119 (34), 91 (25), 73 (98); Yield: 14.5 mg (60%).

234

 

 

o
| OTBS B, 30% PdC|2(DPPF)
l \ H \ \ 0TBDPS ”“3323 6"
0TBS 4:114:34)

(freeze-thaw degassed)

  

OTBS
0 18.18.4234

 

Alkyne (R,R,R)-234: To a solution of pure E,Z-bromide 4d (45 mg, 0.116 mmol)
in 1.5 mL of freshly distilled pyrrolidine, was added Pd(dppf)2C12 (7 mg, 0.0087 mmol)

and pure ketone 213 (15 mg, 0.029 mmol) in 1 mL of pyrrolidine. The reaction was

freeze-thaw degassed (3 cycles) and left under an argon atmosphere for 6 days at ambient
temperature. The reaction was quenched with saturated NH4C1 (2 mL), diluted with ether
(5 mL) and the water layer extracted three times with ether (5 mL each). The combined
organic layers were washed with water (5 mL) and brine (5 mL) and dried over MgSO4.

Column chromatography on silica gel with 20:1 pentane to ether gave 91 % of fostriecin

core 234 (22 mg, 0.026 mmol) as a yellow oil.

1H NMR (300 MHz, CDC13): 6 0.04 (s, 3H), 0.05 (s, 3H), 0.13 (s, 3H), 0.16 (s,
3H), 0.90 (s, 9H), 0.91 (s, 9H), 1.06 (s, 9H), 1.16 (d, 3H, J = 6.2 Hz), 1.18 (d, 3H, J = 6.2
Hz), 1.88-1.98 (m, 1H), 1.98-2. 15 (m, 3H), 3.97 (sept, 1H, J = 6.2 Hz), 4.29 (d, 2H, J =
4.6 Hz), 4.42 (dd, 1H, J = 8.2, 3.3 Hz), 4.544.66 (m, 1H), 4.70-4.88 (m, 1H), 5.12 (broad

s, 1H), 5.40 (d, 1H, J = 10.6 Hz), 5.74 (d, 1H, J = 10.2 Hz), 5.93 (dt, 1H, J = 15.2, 4.9

235

Hz), 5.96-6.04 (m, 1H), 6.36 (t, 1H, J = 10.8 Hz), 6.69 (dd, 1H, J = 15.8, 1.8 Hz), 6.72-
6.84 (m, 1H), 6.94 (dd, 1H, J = 17.2, 4.2 Hz), 7.33-7.47 (m, 6H), 7.64-7.69 (m, 4H); Rf =

0.69 (10:1 pentane/ether).

 

O OH
Methylating Reagent
Conditions 0.
218 219

219 from Methylation of 218. The procedure was the same as that for (R,R,R,R)-

220 and run on a 0.0075 mmol scale. Yield (99%, 0.0074 mmol).106

1H NMR (300 MHz, CDCI3): 6 1.56 (s, 3H), 1.76-1.97 (m, 4H), 2.70-2.88 (m,
2H), 7.04—7.09 (m, 1H), 7.13-7.24 (m, 2H), 7.58 (dd, 1H, J = 7.5, 1.5 Hz); 13C NMR (75

MHz, CDC13): 6 20.37, 29.89, 30.69, 39.71, 70.55, 126.29, 126.31, 127.03, 128.76,

136.20, 142.83; white solid.

236

 

Experimental data for Chapter 5.

 

TBS? OTBS NIS, AgNO3 TBS? OTBS
MeO " = MeO "
Q acetone, r.t Q
o (R,R)-205 TMS 99% O (R,R)-235 I

Iodoacetylene (R,R)-235: To a solution of TMS protected alkyne (R,R)-205 (30
mg, 0.066 mmol) in 4 mL of acetone was added AgNO3 (12.3 mg, 0.072 mmol) and NIS
(17.8 mg, 0.079 mmol). The solution was stirred for 3 hours then cooled to 0 °C and
diluted with 5 mL of EtOAc. The reaction was quenched with 5 mL of H20. The aqueous
layer was extracted with EtOAc. (3 x 5 mL) and the organic layers combined and dried
over Na2S04. The solution was concentrated on the rotary evaporator and

chromatographed on silica gel (pentane/EtOAc 10:1). Iodoacetylene (R,R)-235 was

obtained as a colorless oil in 99% yield (33 mg, 0.065 mmol).

1H NMR (300 MHz, CDC13): 6 0.03 (s, 3H), 0.06 (s, 3H), 0.10 (s, 3H), 0.13 (s,
3H), 0.87 (s, 9H), 0.88 (s, 9H), 1.90-2.16 (m, 2H), 3.70 (s, 3H), 4.34 (dd, 1H, J = 8.0, 4.1
Hz), 4.69 (dd, 1H, J = 8.4, 4.9 Hz); 13C NMR (75 MHz, CDC13): 6 -5.26, -4.80, -4.68, -

4.03, 18.19, 18.24, 25.75, 25.82, 44.52, 51.90, 60.05, 68.64, 95.49, 173.71 (lsp C not

located); IR (neat film on NaCl): 2957 (s), 2930 (s), 2859(8), 1759 (s), 1472 (s), 1385 (s),

1252 (s), 1094(8), 837 (s), 779 (s), 667 (s) cm]; EI mass spectrum m/z (% rel intensity)

497 M+ -15 (1), 455 (7), 369 (l), 323 (3), 295 (7), 291 (8), 229(4), 189(5), 147 (12), 115

237

(5), 89 (40), 73 (100), 57 (25). Anal calcd for C19H37IO4Si2: C 44.52, H 7.28. Found: C

44.41 , H 7.49. Rf = 0.60 (10:1 pentane/EtOAc) [01]D 44.40 (c 1.0, acetone).

 

Tesg ores N33,... N33 Tesg OTBS
M o i = M o i 1
° § THF, r.t e '—
o (Rm-235 l 93% o (R,R)-236

Vinyl Iodide (R,R)-236: To a solution of iodoacetylene (R,R)-23S (26 mg, 0.051

mmol) in 0.5 mL of THF and 0.5 mL of iPrOH was added Et3N (11 aL, 0.076 mmol) and
NBSH100 (22 mg, 0.102 mmol). The mixture was stirred for 14 hours then quenched

with 2 mL of saturated NaHCO3 and diluted with 4 mL of EtOAc. The aqueous layer was
extracted with EtOAc (3 x 5 mL) and the organic layers combined and dried over
NazSO4. The solution was concentrated on the rotary evaporator and chromatographed on

silica gel (Pentane/EtOAc 10:1). Vinyl iodide (R,R)-236 was obtained as a colorless oil in

93% yield (24.3 mg, 0.047 mmol).

1H NMR (300 MHz, CDC13): o 0.03 (s, 3H), 0.07 (s, 6H), 0.08 (s, 3H), 0.86 (s,
9H), 0.92 (s, 9H), 1.74 (ddd, 1H, J = 13.7, 8.2, 3.3 Hz), 1.93 (ddd, 1H, J = 13.8, 7.1, 3.6
Hz), 3.68 (s, 3H), 4.35 (dd, 1H, J = 8.2, 3.3 Hz), 4.57 (ddd, 1H, J = 8.8, 7.1, 3.6 Hz),
6.12-6.26 (m, 2H); 13C NMR (75 MHz, CDC13): o -510, 4.68, 4.51, -357, 17.98, 18.23,
25.83, 44.80, 51.71, 68.69, 72.32, 80.54, 143.69, 174.11, (1 sp3 C not located); IR (neat
film on NaCl): 2955 (s), 2930 (s), 2859 (s), 1757 (s), 1472 (s), 1362 (s), 1258 (s), 1134

(s), 1092 (s), 1005 (s), 837 (s) cm]; FAB mass spectrum m/z (% rel intensity) 515 M+ +1

238

(10), 499 (10), 457 (69), 383 (30), 325 (20), 297 (50), 283 (18), 251 (15), 229 (20), 203
(30), 154 (20), 147 (25), 136 (30), 115 (20), 89 (35), 73 (100), 59 (15). Anal calcd for

C19H39104Si2: C 44.35, H 7.64. Found: C 44.31, H 8.02. Rf = 0.56 (10:1 pentane/EtOAc)

[a]D 332° (c 1.0, ether).

 

Meoresq ores 1 Wm” Pd(CH30N)2Cl2 _
— + TBDPSO DMF ,
o (R,R)-236 4d 76%
T1339 ores
M60 — '_ — 0TBDPS
o (Rm—237

Triene (R,R)-237: To a solution of vinyl iodide (R,R)-236 (10 mg, 0.0195 mmol)

and stannane 4d (47 mg, 0.078 mmol) in dry DMF in a Schlenk flask was added
Pd(CH3CN)2C12 (0.5 mg, 0.002 mmol). The solution was freeze-thaw degassed (3 cycles)
and sealed under an argon atmosphere. The Schlenk flask was wrapped in foil and the

solution stirred at 0 °C for 24 h. The reaction mixture was diluted with 4 mL of EtZO and
quenched with 2 mL of saturated NaHCO3. The aqueous layer was extracted with EtZO (3

x 5 mL) and the organic layers combined and dried over Na2804. The solution was

concentrated on the rotary evaporator and chromatographed on silica gel (pentane/EtOAc
10:1). Triene (R,R)-237 was obtained as a yellow oil in 76% yield (10.5 mg, 0.0148

mmol).

239

1H NMR (500 MHz, CDC13): 6 0.017 (s, 3H), 0.021 (s, 3H), 0.05 (s, 3H), 0.06 (s,
3H), 0.86 (s, 9H), 0.91 (s, 9H), 1.05 (s, 9H), 1.75 (ddd, 1H, J = 13.3, 8.2, 3.3 Hz), 1.93
(ddd, 1H, J = 13.3, 7.1, 3.6 Hz), 3.66 (s, 3H), 4.26 (d, 2H, J = 4.7 Hz), 4.33 (dd, 1H, J =
3.9 Hz), 4.76—4.88 (m, 1H), 5.40 (t, 1H J = 9.9 Hz), 5.80 (dt, 1H, J = 15.1, 4.9 Hz), 6.03
(t, 1H, J = 10.8 Hz), 6.16 (t, 1H, J = 12.1 Hz), 6.33 (t, 1H, J = 11.0 Hz), 6.72 (dd, 1H, J =
12.6, 13.5 Hz), 7.30-7.44 (m, 6H), 7.65 (dd, 4H, J = 6.6, 0.6 Hz); 13C NMR (125 MHz,
CDC13): 6 -5.09, -4.62, -4.60, -3.51, 18.17, 18.28, 19.28, 25.82, 25.97, 26.86, 44.48,
51.70, 64.22, 65.08, 69.24, 123.11, 123.31, 124.52, 127.71, 129.71, 130.18, 133.60,

134.33, 135.04, 135.59, 174.29; IR (neat film on NaCl): 2957 (s), 2924 (s), 2853 (s),
1755 (s), 1620 (s), 1462 (s), 1260 (s), 1094 (s), 801 (s), 702 (s) cm'l; FAB mass spectrum
m/z (% rel intensity) 708 M” (1), 693 (1), 651(6), 577 (5), 519 (6), 491 (4), 452 (5), 327
(6), 321 (l l), 229 (15), 197 (60), 147 (40), 135 (99), 89 (40), 73 (100), 59 (22); HRMS
calcd for C40H64058i3 m/z 708.4065, meas 708.4062. Rf = 0.56 (10:1 pentane/EtOAc).

[(111) 3.6° (c 1.0, ether).

 

TBS? OTBS
M30 : _ _ __ CH3PO(OMe)2(4.Oeq)
0TBDPS . =
O (R,R)-237 n-BULI (3.759(1)
overnight, 88%
TESO: 0TBS
Meo‘p/\II/t\/'\:/=¥-\/OTBDPS
“”0 ('5 o (R,er

240

Phosphonate (R,R)-238. To a solution of dimethyl methyl phosphonate (24.5 11L,
0.226 mmol) in 2 mL of dry toluene at —78 °C was added n-BuLi (1.6 M, 132.4 11L, 0.212
mmol). After 1 h, a solution of ester (R,R)-237 (40 mg, 0.057 mmol) in 1 mL of dry

toluene was added and the reaction mixture stirred at —78 °C for 30 min. The reaction

mixture was quenched with 2 mL of saturated NaHCO3 and diluted with CHZCIZ (8 mL).
The aqueous solution was extracted with CH2C12 (3 x 5 mL) and the organic layers

combined and dried over NaZSO4. The solution was concentrated and the product was

purified by column chromatography (1:1 pentane/EtOAc) to give (R,R)-238 as a yellow

oil in 88% yield (40.1 mg, 0.05 mmol).

1H NMR (500 MHz, CDC13): 6 0.01 (s, 3H), 0.02 (s, 3H), 0.06 (s, 3H), 0.07 (s,
3H), 0.86 (s, 9H), 0.91 (s, 9H), 1.05 (s, 9H), 1.62-1.75 (m, 1H), 1.83-2.00 (m, 1H), 3.10
(dd, 1H, J = 22.4, 15.1 Hz), 3.22 (dd, 1H, J = 21.2, 15.4 Hz), 3.73 (d, 3H, J = 2.5 Hz),
3.76 (d, 3H, J = 2.5 Hz), 4.26 (d, 2H, J = 4.4 Hz), 4.264.30 (m, 1H), 4.74-4.85 (m, 1H),
5.37 (t, 1H J = 10.2 Hz), 5.81 (dt, 1H, J = 15.1, 4.9 Hz), 6.04 (t, 1H, J = 10.4 Hz), 6.16 (t,

1H, J = 10.7 Hz), 6.34 (t, 1H, J = 11.5 Hz), 6.74 (dd, 1H, J = 15.6, 11.3 Hz), 7.31-7.42
(m, 6H), 7.65 (d, 4H, J = 6.0 Hz); 13C NMR (125 MHz, CDCl3): 6 -4.85, -4.70, -4.69, -
3.66, 18.09, 19.22, 25.79, 25.89, 26.80, 35.43 ((1, J = 134.1 Hz), 43.23, 52.88, 64.14,
65.12, 75.82, 123.03, 123.48, 124.34, 127.67, 129.68, 130.54, 133.49, 134.57, 134.69,
135.52, 203.96, (1 sp3 C not located); FAB mass spectrum m/z (% rel intensity) 807 M+

+7 (4), 743 (2), 537 (2), 469 (2), 461 (2), 413 (6), 401 (5), 355 (6), 341 (6), 327 (8), 325

(8), 281 (16), 252 (100), 221 (20), 207 (24), 147 (55), 123 (75), 106 (20), 73 (99), 59

241

(22); Anal calcd for C42H6907PSi3 C 62.96, H 8.68. Found: C 62.60, H 8.26. R1. = 0.79

(1: 1 pentane/EtOAc).

Tesq ores

MO\ _ . ' .
e _ _ oreops Et3NL1C12a:

M80 '6 O (R,R)-238 THF, 98%

 

 

0 (R,R,R)—239

HWE Olefination-Ketone (R,R,R)-239. A solution of LiCl (3.0 mg, 0.070
mmol) in 3 mL of THF was added to a solution of phosphonate (R,R)-238 (40 mg, 0.050

mmol) in 1 mL of THF at room temperature and stirred for 5 minutes. The solution was
then cooled to 0 °C and Et3N (9.76 11L, 0.070 mmol) was added and the solution stirred

for 30 minutes at ambient temperature. At this point, the solution was re-cooled to 0 °C
and the purified major isomer of aldehyde (R)-Za (25.5 mg, 0.150 mmol) was added

dropwise. The flask was wrapped in foil and the solution was stirred overnight at ambient
temperature. The solution was then quenched with H20 (5 mL) and extracted with ether
(10 mL). The organic layer was washed with brine and dried over MgSO4. Column

chromatography (2:5 ether/pentane) gave the purified ketone (R,R,R)-239 in 98 % yield

(40.1 mg, 0.048 mmol) as a colorless oil.

242

1H NMR (500 MHz, CDC13): 6 0.02 (s, 3H), 0.03 (s, 3H), 0.06 (s, 3H), 0.07 (s,
3H), 0.86 (s, 9H), 0.91 (s, 9H), 1.05 (s, 9H), 1.15 (d, 3H, J = 6.4 Hz), 1.17 (d, 3H, J = 6.4
Hz) 1.67 (ddd, 1H, J = 4.4, 7.6, 13.7 Hz), 1.87 (ddd, 1H, J = 4.4, 8.2, 13.4 Hz), 2.01—2.09
(m, 2H), 3.97 (sept, 1H, J = 6.4 Hz), 4.26 (d, 2H, J = 4.4 Hz), 4.34 (dd, 1H, J = 8.3, 3.9
Hz), 4.57 (dd, 1H, J = 9.5, 5.8 Hz), 4.81 (td, 1H, J = 8.3, 4.4 Hz), 5.11 (s, 1H), 5.40 (t,
1H J = 9.8 Hz), 5.72 (d, 1H, J = 7.3 Hz), 5.82 (dt, 1H, J = 15.1, 5.4 Hz), 5.94-6.01 (m,
1H), 6.04 (t, 1H, J = 11.2 Hz), 6.19 (t, 1H, J =11.2 Hz), 6.32 (t, 1H, J = 11.2 Hz), 6.65
(ddd, 1H, J = 16.1, 4.8, 2.0 Hz), 6.64—6.75 (m, 1H), 6.91 (dd, 1H, J = 15.6, 4.4 Hz), 7.33-
7.43 (m, 6H), 7.65 (dd, 4H, J = 6.3, 1.4 Hz); 13C NMR (125 MHz, CDCl3): o -4.80, -
4.64, -4.38, -3.49, 18.16, 18.23, 19.28, 22.07, 23.86, 25.86, 25.98, 26.85, 29.88, 44.13,
64.19, 65.29, 69.84, 75.18, 77.25, 93.09, 123.18, 123.23, 123.65, 124.44, 126.31, 127.71,
128.02, 129.71, 130.31, 133.56, 134.46, 135.13, 135.57, 145.69, 201.12; IR (neat film on
NaCl): 2959 (s), 2922 (s), 2851 (s), 1653 (s), 1559 (s), 1462 (s), 1260 (s), 1094 (s), 1022
(s), 801 (s) cm"; FAB mass spectrum m/z (% rel intensity) 844 W (0.5), 785 (0.5), 713

(0.9), 653 (0.8), 517 (1.2), 505 (3.9), 491 (2.3), 457 (1.6), 373 (2.5), 340 (2.8), 301 (2.5),
239 (8), 223 (6), 209 (8), 197 (40), 171 (14), 147 (16), 135 (76), 73 (100); HRMS calcd

for C49H7606Si3 m/z 844.4955, meas 844.4950. Rf = 0.40 (10:1 pentane/EtOAc). [on]D

138° (c 1.0, ether).

243

MeLi-CeC137H20
0TBDPS -

THF, -78°C
98% d.r = 3:1

 

 

 

Me '0” (R,R,R,R)-240

Tertiary Alcohol (R,R,R,R)-240: CeCl3.7H20 (121.2 mg, 0.326 mmol) was
heated under vacuum (0.2 mmHg) at 70 °C. The temperature was raised from 70 °C to
100 °C slowly over three hours and then allowed to stay at 100 °C overnight. The
temperature was then raised to 140 °C and kept there for 12 h. At this point the reaction
flask was allowed to cool to room temperature under argon and 2 mL of THF was added
to a grayish-white solid and stirred for 10 h. The solution was then cooled to -78 °C and
MeLi (1.5 M, 212 14L, 0.32 mmol) was added. This reaction mixture was stirred for 10
min at —78 °C and 10 min at room temperature. The beige/orange solution was then re-

cooled to -78 °C and ketone (R,R,R)-239 (7 mg, 0.0087 mmol) was added in 0.5 mL of

dry THF. A saturated solution of NaHCO3 (1 mL) was used to quench the reaction, which
was extracted with CHZCI2 (3 X 5 mL) and chromatographed (4:1 pentane/ether) on silica

gel (pretreated with 5% Et3N in hexanes). Alcohol (R,R,R,R)-240 was obtained in 98%

(7.3 mg, 0.085 mmol) as a 3:1 ratio of diastereomers of a colorless oil. Pure major isomer

of 240 was obtained when Preparative TLC was carried out on a silica gel plate (20 x 20

244

 

cm, 250 um) pre-treated with 5% Et3N/hexanes and eluted with 1% Et3N, 5% EtOAc and
94% hexanes. Alcohol 240 was used immediately for the next step as a mixture of 3:1
diastereomers in hopes that an easier separation would be achieved when lactone 232 is
prepared. Boger achieved an easier separation on lactone 232 (Chapter 5, Figure V-l 1).24

However for the purpose of characterization, on one occasion alcohol 240 was separated

into its major and minor isomers.

Tertiary Alcohol (R,R,R,R)-240 (major isomer): 1H NMR (500 MHz, CDC13):
6 0.02 (s, 3H), 0.05 (s, 3H), 0.07 (s, 3H), 0.10 (s, 3H), 0.86 (s, 9H), 0.88 (s, 9H), 1.05 (s,
9H), 1.15 (d, 3H, J = 6.3 Hz), 1.20 (d, 3H, J = 6.4 Hz) 1.23 (s, 3H), 1.91 (ddd, 1H, J =
4.9, 9.8, 14.6 Hz), 2.01-2.14 (m, 3H), 2.72 (s, 1H), 3.64 (t, 1H, J = 4.6 Hz), 3.98 (sept,
1H, J = 6.0 Hz), 4.27 (d, 2H, J = 3.6 Hz), 4.41-4.46 (m, 1H), 4.72 (td, 1H, J = 9.1, 3.6
Hz), 5.10 (s, 1H), 5.38 (t, 1H, J = 11.3 Hz), 5.69 (d, 1H, J = 9.9 Hz), 5.80-5.92 (m, 3H),
5.96-6.02 (m, 1H), 6.04 (t, 1H, J = 10.9 Hz), 6.13 (t, 1H, J = 10.9 Hz), 6.34 (t, 1H, J =
11.3 Hz), 6.74 (dd, 1H, J = 14.4, 11.4 Hz), 7.34-7.44 (m, 6H), 7.65 (dd, 4H, J = 6.3, 1.4

1112); 13C NMR (125 MHz, CDCl3): 6 -4.45, -4.34, -3.72, -3.49, 18.15, 19.25, 22.09,

23.89, 25.93, 26.80, 29.69, 30.76, 42.87, 64.14, 66.07, 67.06, 69.55, 74.95, 75.90, 93.18,

122.76, 123.02, 124.36, 126.10, 127.68.128.67, 129.34, 129.69, 130.35, 133.50, 134.55,
134.94, 135.39, 135.54, (2 sp3 c not located); FAB mass spectrum m/z (% rel intensity)
860 M+ (0.2), 843 (0.2), 801 (0.3), 669(1), 589(4),491 (5), 461 (2), 401 (3), 327 (5),
325 (3), 239(6), 221 (6), 207 (10), 197 (25), 171 (15), 147 (25), 121 (10), 107 (18),91

(25), 73 (100), 55 (20); RI = 0.31 (10:1 pentane/EtOAc).

245

Tertiary Alcohol (R,R,R,R)-240 (minor isomer): 1H NMR (500 MHz, CDC13):
6 0.02 (s, 3H), 0.05 (s, 3H), 0.08 (s, 3H), 0.12 (s, 3H), 0.86 (s, 9H), 0.88 (s, 9H), 1.06 (s,
9H), 1.15 (d, 1H, J = 6.0 Hz), 1.20 (d, 1H, J = 6.1 Hz) 1.23 (s, 3H), 1.91-2.14 (m, 4H),
2.69 (s, 1H), 3.68 (t, 1H, J = 4.0 Hz), 3.98 (sept, 1H, J = 6.1 Hz), 4.27 (d, 2H, J = 3.8
Hz), 4.40-4.48 (m, 1H), 4.72 (td, 1H, J = 9.1, 3.6 Hz), 5.08 (s, 1H), 5.38 (t, 1H, J = 10.4
Hz), 5.69 (d, 1H, J = 8.8 Hz), 5.80-5.92 (m, 3H), 5.96—6.02 (m, 1H), 6.04 (t, 1H, J = 11.2
Hz), 6.13 (t, 1H, J = 10.9 Hz), 6.34 (t, 1H, J = 11.3 Hz), 6.74 (dd, 1H, J = 14.9, 12.1 Hz),
7.34-7.44 (m, 6H), 7.65 (d, 4H, J = 6.6 Hz); l3C NMR (125 MHz, CDC13): 6 -4.42, -3.50,
1.01, 18.09, 19.24, 19.72, 22.07, 23.88, 25.92, 26.80, 29.79, 31.91, 37.08, 64.12, 66.13,
66.74, 69.47, 74.71, 80.62, 93.11, 122.79, 123.03, 124.32, 126.09, 127.67, 128.28,

128.55, 129.68, 130.35, 133.48, 134.54, 135.15, 135.39, 135.52, (2 Sp3 C not located); R

= 0.31 (10:1 pentane/EtOAc).

246

MeOH, K2003, H20
0TBDPS - 240, 50%

 

-20°C, 211

 

      

Me "’ores (R. R, R, er42 Me "’ores (R, R, R,R)-243
25%, 50% based on SM recovered 20%, 40% based on SM recovered

Secondary Alcohol (R,R,R,R)-242: To a solution of alkyne (R,R,R,R)-240 a 3:1

mixture of C8 diastereomers (10 mg, 0.012 mmol) in MeOH (1 mL) at -20 °C, was added
[(ZCO3 (4.2 mg, 0.03 mmol) and H20 (0.125 mL). The mixture was stirred for 2 h after
which it was quenched with H20 (1 mL), extracted with EtOAc (3 X 5 mL) and dried
over MgSO4. The solution was concentrated on the rotary evaporator and

chromatograghed with Et3N/EtOAc/hexane (1:5: 144) on a preparatory thin layer
chromatography (PTLC) silica gel plate (20 x 20 cm, 250 pm) that was pre-treated with

Et3N/EtOAc/hexane (5: 12:83).
Compound (R,R,R,R)-243 (C IS-trans-isomer) was obtained in 20% yield (2.0 mg,

0.0024 mmol) as an inseparable 3:1 diastereomeric mixture as a colorless oil. The ratio

was determined by intergration of the hydroxyl proton (6 = 2.87 ppm major, 6 = 2.69

ppm minor). The stereochemistry of the major diastereomer was assumed to be the C8(R)

247

 

epimer based on the starting alcohol (R,R,R,R)-240 from which it was derived.
Compound (R,R,R,R)-242 was obtained in 25% yield (2.5 mg, 0.003 mmol) as a colorless
oil as a separable 3:1 diastereomeric mixture. Alcohol (R,R,R,R)-240 was recovered in
50% yield (5 mg, 0.006 mmol). The order of elution is as listed above with trans-
(R,R,R,R)-243 first followed by compound (R,R,R,R)-242 and the starting material
alcohol (R,R,R,R)-240 last.

Alcohol 242 was used immediately for the next step as a mixture of 3:1

diastereomers in hopes that an easier separation would be achieved when lactone 232 is
prepared. Boger achieved an easier separation on lactone 232 (Chapter 5, Figure V-l 1).24

However for the purpose of characterization on one occasion alcohol 242 was separated

into its major and minor isomers.

Secondary Alcohol (R ,R ,R,R)-242 (major isomer): 1H N MR (500 MHz,
CDC13): 6 0.01 (s, 3H), 0.05 (s, 6H), 0.06 (s, 3H), 0.86 (s, 9H), 0.87 (s, 9H), 1.06 (s, 9H),

1.17 (d, 3H, J = 5.9 Hz), 1.22 (d, 3H, J = 5.9 Hz) 1.24 (s, 3H), 1.94—2.16(m, 4H), 2.83 (d,
1H, J = 1.5 Hz), 3.66 (d, 1H, J = 10.2 Hz), 3.98 (sept, 1H, J = 6.3 Hz), 4.27 (d, 2H, J =
4.4 Hz), 4.41-4.48 (m, 1H), 4.88 (t, 1H, J = 8.1 Hz), 5.09 (s, 1H), 5.47 (t, 1H, J = 9.8 Hz),

5.62-5.86 (m, 4H), 5.94—6.08 (m, 2H), 6.17 (t, 1H, J = 10.9 Hz), 6.30 (t, 1H, J = 11.2 Hz),
6.74 (dd, 1H, J = 14.2, 12.4 Hz), 7.34-7.44 (m, 6H), 7.65 (dd, 4H, J = 6.6, 1.7 Hz); Rf =

0.48 (10:1 Pentane/EtOAc).

248

 

 

 

Secondary Alcohol (R,R,R,R)-242 (minor isomer): 1H N MR (500 MHz,
CDC13): 6 0.01 (s, 3H), 0.05 (s, 6H), 0.08 (s, 3H), 0.86 (s, 9H), 0.87 (s, 9H), 1.06 (s, 9H),
1. 15 (d, 3H, J = 6.4 Hz), 1.18 (d, 3H, J = 6.3 Hz) 1.24 (s, 3H), 1.95-2.16 (m, 4H), 2.71 (d,
1H, J = 2.9 Hz), 3.60 (d, 1H, J = 11.7 Hz), 3.97 (sept, 1H, J = 6.1 Hz), 4.27 (d, 2H, J =
4.9 Hz), 4.40-4.52 (m, 1H), 4.87 (t, 1H, J = 7.8 Hz), 5.08 (s, 1H), 5.48 (t, 1H, J = 10.0
Hz), 5.64 (dd, 1H, J = 16.0, 6.1 Hz), 5.70-5.76 (m, 1H), 5.75-5.86 (m, 2H), 5.94-6.05 (m, ‘1
2H), 6.17 (t, 1H, J = 11.2 Hz), 6.30 (t, 1H, J = 10.5 Hz), 6.74 (dd, 1H, J = 13.4, 13.4 Hz),
7.34—7.44 (m, 6H), 7.66 (d, 4H, J = 7.3 Hz); 13C NMR (125 MHz, CDCl3): 6 -5.10, -4.34,
-2.22, 18.12, 19.22, 19.71, 22.09, 23.85, 25.81, 25.83, 26.79, 29.69, 31.92, 37.07, 64.14,
66.27, 66.62, 69.56, 74.99, 77.52, 93.12, 121.95, 123.35, 124.49, 126.06, 127.66, 128.51,

129.66, 130.15, 133.51, 134.04, 135.13, 135.52, 136.06, (1 sp3 and 1 sp2 C not located);

Rf = 0.48 (10:1 pentane/EtOAc).

The following spectral data was taken on a 3:1 mixture of diastereomers.

Trans-Triene-(R,R,R,R)-243 (major isomer): 1H NMR (500 MHz, CDCl3): 6

0.03-0.10 (m, 12H), 0.83 (s, 9H), 0.84 (s, 9H), 1.05 (s, 9H), 1.15 (d, 3H, J = 6.3 Hz), 1.20
(d, 3H, J = 6.0 Hz) 1.28 (s, 3H), 1.90-2.10 (m, 4H), 2.87 (s, 1H), 3.66 (t, 1H, J = 10.2
Hz), 3.98 (sept, 1H, J = 6.0 Hz), 4.23 (d, 2H, J = 4.6 Hz), 4.30-4.50 (m, 1H), 4.88 (t, 1H,
J = 8.2 Hz), 5.08 (s, 1H), 5.39 (t, 1H, J = 8.5 Hz), 5.60-5.80 (m, 4H), 5.90 (t, 1H, J = 11.2
Hz), 5.94-6.01 (m, 1H), 6.17 (dd, 1H, J = 14.1, 10.9 Hz), 6.30 (dd, 1H, J = 15.0, 11.2

Hz), 6.40 (dd, 1H, J = 14.0, 12.1 Hz), 7.34-7.44 (m, 6H), 7.65 (d, 4H, J = 6.6 Hz);

 

249

 

13C NMR (125 MHz, CDC13): o -512, 4.29, -221, -214, 18.20, 19.25, 21.25,

22.25, 23.93, 25.86, 26.85, 30.66, 39.16, 64.24, 66.36, 66.86, 69.90, 74.79, 77.54, 93.39,

126.28, 127.09, 127.39, 127.66, 128.40, 129.64, 129.86, 130.50, 130.35, 133.20, 133.65,

134.99, 135.15, 135.56, (2 Sp3 C not located); FAB mass spectrum m/z (% rel intensity)
859 M+ -1 (0.2), 800 (0.3), 728 (1.3), 711 (0.3), 685 (0.3), 669 (1.5), 651 (0.6), 611 (1.0),

559 (1.2), 491 (4), 413 (4), 403 (2), 373 (3), 325 (l 1), 267 (10), 239 (15), 197 (60), 185

(50), 135 (99), 91 (15), 75 (70), 73 (100), 59 (18); HRMS calcd for C50H84N068i3 m/z (M

 

+ N H4)+ 878.5643, meas 878.5643. Rf = 0.7 (10:1 pentane/EtOAc).

Minor isomer: 6 = 2.69 (d, 1H, J = 2.5 Hz) only distinguishable proton.

PPTS (0.25 eq). EtOH
0TBDPS :

 

r.t, 3.5h, 92%

 

    

Ma 201-BS
(R, R, R,R)-244 (Boger‘s Intermediate)

Ethyl Acetal (Boger’s Intermediate) (R,R,R,R)-244 : To a solution of isopropyl

acetal (R,R,R,R)-242 (5.5 mg, 0.0067 mmol) in 1 mL of EtOH was added PPTS (0.4 mg,

0.0017 mmol) at room temperature. The reaction was stirred for 3.5 h then diluted with

250

 

2.5 mL of CHzCl2 and quenched with 1 mL of NaHCO3. The layers were separated and

the aqueous layer extracted twice with CHzClz. The organic layer was dried over NaZSO4
and concentrated. Preparative TLC was carried out on a silica gel plate (20 x 20 cm, 250
um) pre-treated with 5% Et3N/hexanes and chromatographed 28% EtOAc/hexanes.

Ethyl acetal (R,R,R,R)-244 was isolated as a colorless oil in 92% yield (5.2 mg, 0.0062
mmol) as an inseparable 3:1 mixture of diastereomers. The ratio was determined by

intergration of the hydroxyl proton (6 = 2.98 ppm major, 6 = 2.87 ppm minor). The

stereochemistry of the major diastereomer was assumed to be the C8(R) epimer based on

the spectral data provided by Boger.24 The 1H NMR spectrum and IR spectrum matched
those of an authentic sample. These spectra were kindly provided by professor Boger
(also a 3:1 mixture of epimers at C8). Copies of these spectra as well as those of 244 can

be found below.

The following spectral data was taken on a 3:1 mixture of diastereomers.

1H NMR (500 MHz, CD3CN) major isomer: 6 0.01 (s, 3H), 0.07-0.09 (m, 9H),

0.88 (s, 9H), 0.89 (s, 9H), 1.06 (s, 9H), 1.12-1.20 (m, 3H), 1.28-1.32 (m, 3H) 1.66-1.75
(m, 1H), 1.98-2.10 (m, 2H), 2.10-2.20 (m, 1H), 2.98 (d, 1H, J = 4.4 Hz), 3.46-3.56 (m,
1H), 3.61(dd, 1H, J = 10.0, 3.9 Hz), 3.71-3.78 (m, 1H), 4.27-4.38 (m, 3H), 4.88-4.93 (m,
1H), 4.98 (broad s, 1H), 5.48-5.54 (m, 1H), 5.65-5.82 (m, 3H), 5.88 (dt, 1H, J = 15.1, 4.9
Hz), 5.95-6.02 (m, 1H), 6.08 (t, 1H, J = 10.7 Hz), 6.23 (t, 1H, J = 11.7 Hz), 6.32 (t, 1H, J

= 11.2 Hz), 6.74 (dd, 1H, J = 14.4, 12.2 Hz), 7.39-7.48 (m, 6H), 7.67 (d, 4H, J = 6.8). IR

251

 

 

(neat film on NaCl): 3422 (w), 2957 (s), 2928 (s), 2857 (s), 1472 (s), 1462 (s), 1429 (s),
1385 (s), 1363 (s), 1260 (s), 1105 (s), 1022 (s), 837 (s), 802 (s), 777 (s), 702 (s) cm]; Rf

= 0.65 (10:1 pentane/EtOAc).

Minor isomer: 6 = 2.87 (d, 1H, J = 4.8 Hz) only distinguishable proton.

252

PCC (1.5 eq), NaOAc (4 eq)

 

0TBDPS
r.t, 2h, 40%

 

    

Me "0H (R,R,R,R)-241

Lactone (R,R,R,R)-241 To a flamed dried flask cooled under argon was added

PCC (1.6 mg, 0.007 mmol) and NaOAc (1.6 mg, 0.020 mmol) followed by 2 mL of
CHzClz. The suspension was stirred for 15 minutes before rapidly adding the acetal (4

mg, 0.005 mmol). After 2.5 hours the suspension was filtered through a small silica gel
pad and rinsed eight times with ether. The solution was concentrated in vacuo and
purification on a Preparative TLC plate (silica gel, 20 x 20 cm, 250 um) gave lactone

(R,R,R,R)-241 as a colorless oil in 40% yield (1.5 mg, 0.002 mmol).

(Spectra was obtained as a 1:1 mixture) 1H NMR (500 MHz, CDC13): 6 0.03—0.09
(m, 12H), 0.86-(s, 9H), 0.87 (s, 9H), 1.05 (s, 9H), 1.23 (s, 3H), 1.80-2.04 (m, 1H), 2.20-
2.49 (m, 3H), 2.87 (s, 1H), 3.50—3.75 (m, 1H), 4.26 (d, 2H, J = 4.4 Hz), 4.34 (d, 1H, J =
4.0 Hz), 4,644.78 (m, 1H), 4.874.98 (m, 1H), 5.65-6.22 (m, 4H), 6.22-6.50 (m, 2H),

6.64—6.90 (m, 2H), 7.32-7.43 (m, 6H), 7.65 (d, 4H, J = 6.0 Hz); Rf = 0.33 (CHzClz).

253

Amberlyst A-26 (Cl' form)

 

0 'C-r.t 1h

   

(R,R,R,R)-221 (R,R,R,R)-247

Alkyne (R,R,R,R)-247: To a solution of alkyne (R,R,R)-221 (20 mg, 0.036 mmol)
with a 7:1 mixture at C8 in MeOH (1 mL) was added amberlyst resin A-26 (Cl ion form,
16.0 mg) which was prewashed with MeOH. The reaction was run for 1 h and then
filtered and washed sequentially with MeOH (5 mL), EtZO (5 mL) and CHzCl2 (5 mL).

The combined rinses were concentrated in vacuo and the crude oil chromatographed on
silica gel with EtOAc/Pentane 1:20. Alkyne (R,R,R,R)-247 was isolated as a colorless oil
in 23% yield (4 mg, 0.076 mmol) and 44% (8.8 mg, 0.014 mmol) of starting alcohol

(R,R,R,R)-221 was recovered. Compound (R,R,R,R)-247 was assigned as the compound
having the C8 and C 11 hydroxyls protected with TBS based on incomplete HMQC and

HMBC data collected. Secondary alcohol (R,R,R,R)-247 was isolated as a 7:1 inseparable
mixture of diastereomers. The ratio was determined by intergration of the hydroxyl

proton (6 = 2.90 ppm major, 6 = 2.66 ppm minor). The stereochemistry of the major
diastereomer was assumed to be the C8(R) epimer based on the starting material 221.

Compound’s 221 preparation is outlined in chapter 4 experimental.

Major isomer: 1H N MR for secondary alkyne (R,R,R,R)-247: 6 0.04 (s, 3H), 0.06

(s, 3H), 0.09 (s, 3H), 0.12 (s, 3H), 0.13 (s, 9H), 0.85 (s, 9H), 0.86 (s, 9H), 1.14 (d, 3H, J =

254

 

6.0 Hz), 1.20 (d, 3H, J = 6.3 Hz), 1.31 (s, 3H), 1.75-1.86 (m, 1H), 1.98-2.10 (m, 3H),
2.90 (s, 1H), 3.71 (d, 1H, J = 13.9 Hz), 3.97 (sept, 1H, J = 6.0 Hz), 4.38—4.48 (m, 1H),
4.65 (dd, 1H, J = 7.9, 3.2 Hz), 5.08 (broad s, 1H), 5.65-5.84 (m, 3H), 5.94-6.04 (m, 1H).
FAB mass spectrum m/z (% rel intensity) (M*-1) 609 (0.2), 591 (0.2), 493 (1), 477 (0.8),

463 (0.8), 419(6), 361 (1.5), 325 (12), 267 (25), 241 (30), 185 (15), 157 (15), 147 (15),

135 (20), 73 (100); R; = 0.4 (4:1 pentane/ether).

Minor isomer: 6 = 2.66 (s, 1H) only distinguishable proton.

255

REFERENCE

1 Tunac, J. 8.; Graham, B. D.; Dobson, W. E. J. Antibiot 1983, 36, 1595.

2 The three compounds were reported as being 1.0g of pure (CI-920), and
unreported amount of pure PD 113,271, and 1.1g of PD 113,270 isolated
from a fermentation run in a 760 L tank.

3 Murray, A. w. Nature 1992, 359, 599.

4 De Jong, R.; De Vries, G. 15.; Mulder, N. H. Anti-Cancer Drugs 1997, 8,
413.

5 De Jong, R.; De Vries, G. 13.; Mulder, N. H. Br. J. Cancer 1999, 79, 882.

6 Chen, G. L.; Yang, L.; Rowe, T. C.; Halligan, B. D.; Tewey, K. M.; Liu, L.
F. J Biol. Chem. 1984, 259, 13560.

7 Yang L.; Rowe, T. C.; Halligan, B. D.; Tewey, K. M.; Liu, L. F. Science
1984, 266, 466.

8 Nelson, E. M.; Tewey, K. M.; Liu, L. F. Proc. Natl. Acad. Sci. USA 1984,
81, 3161.

9 Boger, D. L.; Soenen, D. R.; Gauss, C. —M; Lewy, D. S. Curr. Med.
Chem. 2002, 9, 2005.

'0 Frosina, G.; Rossi, 0. Carcenogenesis 1992, I3, 1371.
'1 Ingebristen, T. S.; Cohen, P. European J. Biochem. 1983, 132, 255.

12 Cheng, A.; Balczon, R.; Zuo, Z.; Koons, J. S.; Walsh, A. H.; Honkanen, R.
E. Cancer Res. 1998, 58, 3611.

'3 Roberge, M.; Tudan, C.; Hung, S. M. F.; Harder, K. W.; Jirik, F. R.;
Anderson, H. Cancer Res. 1994, 54, 6115.

‘4 Walsh, A. H.; Cheng, A.; Honkanen, R. E. FEBS Lett. 1997, 416, 230.
‘5 Hastie, c. 1.; Cohen, P. T. w. FBES Lett. 1998, 431, 357.

16 Ho, D. T.; Roberge, M. Carcinogenesis 1996, I 7, 967.

256

 

‘7 Guo, X. W.; Swank, R. A.; Anderson, H.; Tudan, C.; Bradbury, E. M.;
Roberge, M. EMBO J. 1995, I4, 976.

18O’Connor, P. M.; Ferris, D. K.; Pagano, M.; Draetta, G.; Pines, J.; Hunter,
T.; Longo, D. L.; Kohn, K. W.; J. Biol. Chem. 1993, 268, 8298.

b) Hartwell, L. H.; Weinert, T. A. Science, 1989, 246, 629. C) Brown, H.
C.; Brown, C. A. J. Am. Chem. Soc. 1963, 85, 1005.

‘9 Weinbrenner, C.; Baines, C. P.; Liu, G. 8.; Armstrong, S. C.; Ganote, C.
E.; Walsh, A. H.; Honkanen, R. E.; Cohen, M. V.; Downey, J. M.
Circulation 1998, 98, 899. D) Armstrong, S. C.; Gao, W.; Lane, J. R.;
Ganote, C. E. J. Mol. Cell. Cardiol. 1998, 30, 61.

20 Boger, D. L.; Honkanen, R. E.; Bonness, K. M.; Swingle, M. R.; Hwang,
1.; Gauss, C. -M.; Buck, S. B.; Hardouin, C.; Ichikawa, S.; Soenen, D. R.
J. Am. Chem. Soc. 2003, 125, 15694.

2‘ Hokason, G. C.; French, J. C. J. Org. Chem. 1985, 50, 462.

22 Boger, D. L.; Hikota, M.; Lewis, B. M. J. Org. Chem. 1997, 62, 1748.

23 Rychynovsky, S. D.; Rogers, B.; Yang, G. J, Org. Chem. 1993, 58, 3511.

2“ Boger, D. L.; Ichikawa, s.; Zhong, w. J. Am. Chem. Soc. 2001, 123, 4161.

25 Jacobsen, E. N.; Chavez, D. E. Angew. Chem. Int. Ed. 2001, 40, 3667.

26 Reddy, Y. K.; Falck, J. R. Org. Lett. 2002, 6, 969.

27 Miyashita, K.; Ikerjiri, M.; Kawasaki, H.; Maemura, S.; Imanishi, T.
Chem. Comm. 2002, 742.

28 Hatakeyama, S.; Tomoyuki, E.; Okamoto, N. Chem. Comm. 2002, 3042.

29 Fujii, K.; Maki, K.; Kanai, M.; Shibasaki, M. Org. Lett. 2003, 5, 733.

3" Wang, Y. -M.; Kobayashi, Y. Org. Lett. 2002, 4, 4615.

31Trost, B. M.; Frederiksen, M. U.; Papillon, J. P. N.; Harrington, P. E.;
Shin, S.; Shireman, B. T. J. Am. Chem. Soc. 2005, 127, 3666.

32 Just, G.; O’Connor, B. Tetrahedron Lett. 1988, 29, 753.

33 Cossy, 1.; Fabienne, P.; BouzBouz, s. Org. Lett. 2001.3, 2233.

257

34 Ramachandran, P. V.; Liu, H.; Reddy, M. V. R. R.; Brown, H. C. Org.
Lett. 2003, 5, 3755.

35 Brown, H. C.; Brown, C. A. J. Am. Chem. Soc. 1963,85, 1005.

36 For reviews on the AD reaction, see: a) Kolb, H. C.; VanNieuwenhze, M.
S.; Sharpless, K. B. Chem. Reviews 1994, 94, 2483. B) Johnson, R. A.;
Sharpless, K. B. Catalytic Asymmetric Synthesis; Ojima, 1., Ed.; (VCH:
New York), 1993, 227. c) Lohray, B. B. Tetrahedron Asymmetry 1992, 3,
1317.

37 Still, w. C.; Gennari, C. Tetrahedron Lett. 1983, 24, 4405.

38 Corey, E. J.; Fuchs, P. L. Tetrahedron Lett. 1972, 3769.

39 Uenishi, J .; Kawahama, R.; Yonemitsu, 0.; Tsuji, J. J. Org. Chem. 1998,
63, 8965. b) Uenishi, J .; Kawahama, R.; Yonemitsu, 0.; Tsuji, 1.; Shiga,
Y. Tetrahedron Lett. 1996, 37, 6759. c) Uenishi, J .; Kawahama, R.;
Yonemitsu, 0.; Tsuji, J. J. Org. Chem. 1996, 61, 5716.

4" Evans, D. A.; Black, w. C. J. Am. Chem. Soc. 1993, 115, 4497. b) Gibbs,
R. A.; Krishnan, U. Tetrahedron Lett. 1994, 35, 2509.

4‘ Jung, M. E.; Light, L. A. Tetrahedron Lett. 1982, 23, 3851.
42 Imamoto, T. Pure Appl. Chem. 1990, 62, 747.

43 Evans, D. A.; Gage, J. R.; Leighton, J. L. J. Am. Chem. Soc. 1992, 114,
9434.

44 Hafner, A.; Duthaler, R. 0.; Marti, R.; Rihs, G.; Rothe-Streit, P,;
Schwarzenbach, F. J. Am. Chem Soc. 1992, 114, 2321.

45 Grubbs, R. H.; Chang, S. Tetrahedron 1998, 54, 4413.

46 Tokunnaga, M.; Larrow, J. F.; Kakiuchi, F.; Jacobsen, E. N. Science 1997,
2 77, 936. b) Jacobsen, E. N.; Wu, M. H. Comprehensive Asymmetric
Catalysis (Eds.: Jacobsen, E. N .; Pfaltz, H.; Yamamoto, H.), Springer,
New York, 1999, Chapter 35.

4" Wipf, P.; Xu, w. Tetrahedron Lett. 1994.331, 5197. b) Wipf, P.; Ribe, s.
J. Org. Chem. 1998, 63, 6454.

258

 

48 Dossetter, A. G.; Jamison, T. F.; Jacobsen, E. N. Angew Chem. Intl. Ed.
1999, 38, 2398.

49 Noyori, R.; Matsumura, K.; Hashiguchi, S.; Ikariya, T. J. Am. Chem. Soc.
1997, 119, 8738.

5" Stille, J. K.; Groh, B. L. J. Am. Chem. Soc. 1987, 109, 813.

5‘ Brown, H. C.; Jadhav, P. K. J. Am. Chem. Soc. 1983, 105, 2092.

52 Miyuara, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457.

53 Noyori, R.; Tomino, 1.; Tanimoto, Y.; Nishizawa, M. J. Am. Chem. Soc.
1984, 106, 6709. b) Noyori, R.; Tomino, 1.; Tanimoto, Y.; Nishizawa, M.
J. Am. Chem. SOC. 1984, 106, 6717.

54 Kiyotsuka, Y.; Igarashi, J .; Kobayashi, Y. Tetrahedron Lett. 2002, 43,
2725.

55 Sonagishira, K.; Tohda, Y.; Hagihara, N. Tetrahedron Lett. 1975, 4467.
56 Parasi, M. Ph.D. thesis, University of Chicago, Illinois, CHICAGO, 1999.
57 Dr. Su Yu and Dr. W. D. Wulff unpublished results.

58 Dr. Ken Wilson’s and Dr. w. D. Wulff unpublished results.

59 Lei, M.; Oh, 13.; Shih, Y.; Liu, H. J. Org. Chem. 1992,57, 2741.

60 Yamada, K.; Miyura, N .; Itoh, M.; Suzuki, A. Synthesis 1977, 679.
b) Midland, M. M.; Tramontano, A.; Cable, J. R. J. Org. Chem. 1980, 45,
28.

6‘ Lindlar, H.; Dubuis, R. Org. Synth. v. 1973, 880.

62 Suemune, H.; Oda, K.l Saeki, S.; Sakai, K. CPB, 1988, 36, 172.

63 Crimmins, M. T.; King, B. w. J. Am. Chem. Soc. 1988, 120, 9084.
b) Stammers, T. A.; Lautens, M. Synthesis 2002, I4, 1993.

64 Kobayashi, M.; Wang, W.; Tsutsui, Y.; Sugimoto, M.; Mukarakami, N.
Tetrahedron Lett. 1998, 39, 8291.

259

65 Marshall, J. A.; Garofalo, A. w. J. Org. Chem. 1996,61, 8291.
66 Stork, G.; Zhao, K. Tetrahedron Lett. 1989, 30, 2173.

67 Marco, J. A.; Costero, M. A.; Ferrer, P.; Luis, s. V.; Gavina, F. J. Chem.
Soc. Chem. Commun. 1983, 296.

68 Kadota, 1.; Takamura, H.; Sato, K.; Ohno, A.; Kumiko, M.; Yamamoto, Y.
J. Am. Chem. Soc. 2003, 125, 46.

69 Powers, T. s.; Shi, Y.; Wilson, K. J.; Wulff, w. D.; Rheingold, A. L. J.
Org. Chem. 1994, 59, 6882. b) Powers, T. S.; Shi, Y.; Quinn, J.; Wulff, W.
D.; Rheingold, A. L.; Jiang, W.; Hsung, R.; Parasi, M.; RaHm, A.; Jiang,
X. W.; Glenn, P. A. J. Organomet. Chem. 2001, 617, 182.

70 2-Alkynals were prepared by modification of the procedure reported by
Brandsma, L. in Studies in Organic Chemistry 34: Preparative Acetylenic
Chemistry, 2““. Ed., (Elsevier: Amsterdam) 1988: p. 102. See experimental
section for detailed procedure illustrated for the preparation of
trimethylsilyl propynal l65c.

7' Shi, Y. Ph.D. thesis, University of Chicago, Illinois, CHICAGO, 1995.
72 Lefour J. M.; Loupy, A. Tetrahedron 1978, 34, 2597.
73 Wulff, W. D. Organometallics 1998, l 7, 3116.

74 Kelly, S. E. Comprehensive Organic Synthesis, Trost, B. M.; Fleming, 1.
Eds.; (Pergamon: Oxford), 1991, l, 761. b) Lawrence, N. J. Preparation of
Alkenes: A Practical Approah, Williams, J. M. J. Ed.; (Oxford University
Press: Oxford), 1996, 35. c) Maryanoff, B. E.; Reitz, A. B., Chem.
Reviews, 1989, 89, 863. d) Seyden-Penne, J. Bull. Chim. SOC. Fr. 1988,
238. e) Walker, B. J. Organophosphorus Compounds in Organic
Synthesis, Cadogan, J. I. G. Ed.; (Academic Press: New York), 1979, 155.
f) Wadsworth, W. S. Organic Reactions, 1977, 73, 25. G) Boutagy, J .;
Thomas, R. Chem. Rev. 1974, 74, 87.

75 Evans, D. A.; Chapman, K. T.; Carriera, E. M. J. Am. Chem. Soc. 1988,
110, 3560.

76 Prasad, J. 5.; Leibeskind, L. s. Tetrahedron Lett. 1987,29, 1857.

77 Rychnovsky, s. D.; Skalitshy, D. J. Tetrahedron Lett. 1990, 31, 945

260

 

b) Evans, D. A.; Rieger, D. L.; Gage, J. R. Tetrahedron Lett. 1990, 31,
7099. For related work using carbon-13 spectral data to determine
different sized acetonide rings, see: Buchanan, J. Carbohydrate Research,
1982, 100, 75.

73 Perrin, C. L.; Engler, R. E. J. Am. Chem. Soc. 1997, 119,585.
79 Holmes, A. 8.; Gilbert, I. H. Tetrahedron Lett. 1990, 31, 2633.

80 Iorga, B.; Eymery, F.; Carmichael, D.; Savignac, P. Eur. J. Org. Chem.
2000, 3103.

8‘ Kim, S.; Kee, 1. 3.; Park, J. H. Synlett 1991 183.
82 Bruzik, K. S.; Tsai, M. D. J. Am. Chem. Soc. 1992, 114, 6361.
83 Corey, E. J.; Erickson, B. w. J. Org. Chem. 1991,36, 3553.

84 Hazra, B. G.; Padmakar, L. J .; Bharat, B. B.; Narshinha, P. A.; Vandana,
S. P.; Chordia, M. D. Tetrahedron 1994, 50, 2523.

85 Corey, E. J.; Samuelson, B. J. Org. Chem. 1984, 49, 4735. b) Just, 6.;
O’Connor, B. Tetrahedron Lett. 1987, 28, 3235.

86 Postels, H. T.; Konig, w. A. Liebegs Ann. Chem. 1992, 1281.

87 Savage, 1. ; Thomas, E. J.; Wilson, P. D. J. Chem. Soc. Perkin Trans. 1,
1999, 3291. b) McCarthy, D. G.; Collins, C. C.; O’Driscoll, J. R;
Lawrence, S. E. J. Chem. Soc. Perkin Trans. 1, 1999, 3667.

88 Data obtained from unpublished results of Xuejun Lui and W. D. Wulff.

89 Evans, D. A.; Starr, J. T. J. Am. Chem. Soc. 2003, 125, 13531.

90 Furrow, M. E.; Schaus, s. E.; Jacobsen, E. N. J. Org. Chem. 1998,63,
6776.

9' Vollhardt, P.; Vollhardt, C.; Berris, B. C. Tetrahedron, 1982, 38, 2911.

92 Munakata, R.; Katakai, H.; Ueki, T.; Jurosaka, J.; Takao, K.; Tadano, K.
J. Am. Chem. Soc. 2003, 125, 14722.

93 Wadsworth, w. s.; Emmons, w. D. J. Am. Chem. Soc. 1961, 83, 1733.

261

 

9“ Sharpless, K. B.; Katsuki, T. J. Am. Chem. Soc. 1980, 102.5974.
95 Satoh, T.; Uwaya, S.; Yamakawa, J. Chem. Lett. 1987, 1259.

96 Arbuzov, A.; Russ, J. Phys. Chem. Soc. 1906, 38, 687.

97 Imamoto, T. Pure Appl. Chem. 1990, 62, 747.

98 Fargeas, V.; Ménez, P. L.; Berque, 1.; Aldisson, J .; Panctazi, A.
Tetrahedron, 1996, 52, 6613.

99 Jeffry, T. Tetrahedron Lett. 1985, 26, 2667.
'00 Meyers, A. G.; Zheng, B. J. Org. Chem. 1997,62, 7507.
'01 Fetizon, M.; Golfier, M.; Mourgues, P. Tetrahedron Lett. 1972, 4445.

'02 Information found on Alfa Aesar label for CeCl3.7HZO.

103 Mapp, K. A.; Heathcock, C. H. J. Org. Chem. 1999, 64, 23-

262

 

 

11111111111111111