~3‘0m

 

~—

4'.-4L.,...

""'--n....,,

 

'(
:5 v'qla'K-i - .21“
.._-

vs'u

hf. '

i

I". 4 J

.1)

.
. .
1-. aw.»

. ' “4‘5, .1

 

("4.1; ' :j-‘sw
. , '3.

mun". ~""""."."""
-. .
....

r '13,!"

$5
.,§
1.0.} v‘ 74).:
“J. \3‘. :Tx;

.-‘
’44.;-
4.
‘

. 4:;
3:” ' {kt-”5‘ 4““:4‘.’ 33'

~41
)}hl‘$‘

4 " .y'g'ztk
L

s' . - ‘
I '1‘ l'v.‘ :":‘ .":' ‘3‘»" (Fr. ’Ki‘.-:
:4! f) 5-4 b'v

34"}:

'v' V‘K'f‘istv (

"a
‘ \ I i
'33:?

a»:
'1
5h

“In.“ '
4 v. o. ".47." 1
\

.

2 .

. s .
‘;.:‘4‘ ‘X

14
"win...

u
an

'47:»
'f .4
73

’4 Sci'
4'9.

k. t

In.‘ '
.
“5,53 ‘
l - C
II 5-,
fié‘fi’ .f
‘ ' 5
‘V
v
15"“:-

5" .‘u .
..».5-'7;.,.={‘ 13.; ‘
$61.33“! grgj4CW~

" ' H3 3.30
"1 ' 1"
4).

4
~ ~. .
. ‘A
1
i:-
,,
‘47
:- .a'2w»
.’ J 7". .
3’ "5"?“
.\ ’ g
. .
.4
‘ ~

_.
.l A . l
" - I
"Di" . ‘4. .
35:25‘391‘
’3

499
«f. M3? is”. ‘
a [3'

A” :4
(1'
¥1< :1. ,, .. .. . 7: .
"‘4' ‘ '1‘ " ' V I“! '
’1‘ 1| “‘ h‘f’" ' 4:}. V." 217': f\:u;:4' :;:4.>""
. 135%,”; 1-!“ ‘1“? _, \f-
NW- ‘3. .w - ‘

'5 "‘vxw

€I4J" :“

3.37.4
g».
,9

$54.41;! ‘42.
. .L 5

$19.? V 1 ӣ1
’1' 1,? Cfil,‘

' t‘ " ' 4'4
41.l"-' '“Z. '5‘ r'
;.w . l.‘ 5“?" 39'.
‘ "V‘Jjé ”I" 13.4.! L
7 ’ 211-1.“-
fi' "1 “QM-'2'
I,‘

r,
4" 4 .i
.'.’ , H A
(I. .35.”: "4' 5'.- , £311. . . 4
’ * ,. 1‘ . 5 firm Ti»? “5:“ 61$
" I?" 4" 4 V :4’ ‘1' ’ ' A. '3‘; j’? »
r7 - 2:43»: 4: .31sz .J
$5.15- 4“ J 4':- 5;; .4- .41" A
'1"

.4 1

3.?
“any”; '
' «.44
.r ‘f"
F L
1 #:‘VI‘.
1/; ’4'.

1.

‘1
1 .
4

4

4

4 a
5‘2; 2
§

1 u " ‘1 ,
' :Ejfih 0732432 3'2: 'l‘fi

‘1r‘.4:l”f'_ «My 2:..." . «£43972,
”yr/‘4' .fifitj;
‘3! ”E”.’}':t:- "w

 

5.‘“z.'.‘42‘

4;}:11'1- VI“
' "w
‘ r123”

I .
w fl.h,4_ . ‘nfi
C22" Wifav.

Ir-
a

. ., .4. u». ' ’ ' fi
¢E_&§~J£X“’§¢.gf锑

113000117

MICHIGAN STATE UNIVERSIT IBRARIES

I N Ill/llfll’lll [Hill/fill[Milli/(ll

3 1293 00563 3221

 

 

 

 

 

 

 

 

 

 

 

 

 

 

LIERARY
Michigan State

J___llnlmmlty .

 

 

This is to certify that the

dissertation entitled

FURAN TERMINATED CATIONIC CYCLIZATIONS AND THEIR USE IN
SYNTHESIS. APPROACHES TO THE SYNTHESES OF HELENANOLIDE
PSEUDO GUAIANOLIDES, DAPHNANE DITERPENES, AND INDOLIZIDINE
ALKALOID NATURAL PRODUCTS.

presented by

MARK CHAD MCMILLS

has been accepted towards fulfillment
of the requirements for

Ph . D . degree in Chemi stry

 

 

KNEW

Major professor

Date January 17, 1989

 

MS U is an Affirmative Action/Equal Opportunity Institution 0-12771

 

 

MSU

LIBRARIES
.—3—.

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. flfl§§_will
be charged if book is
returned after the date
stamped below.

 

 

 

 

FURAN TERMINATED CATIONIC CYCLIZATIONS AND THEIR USE IN
SYNTHESIS. APPROACHES TO THE SYNTHESES OF HELEN ANOLIDE
PSEUDOGUAIANOLIDES, DAPHNAN E DITERPENES, AND INDOLIZIDINE
ALKALOID NATURAL PRODUCTS.

By

Mark Chad McMills

A DISSERTATION

Submitted To
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Chemistry

1989

ABSTRACT

FURAN TERMINATED CATIONIC CYCLIZATIONS AND THEIR USE IN
SYNTHESIS. APPROACHES TO THE SYNTHESIS OF HELEN ANOLIDE
PSEUDOGUAIANOLIDE, DAPHNANE DITERPENES, AND INDOLIZIDINE
ALKALOID NATURAL PRODUCTS.

BY

MARK CHAD MCMILLS

In an ongoing investigation of furans as terminators in cationic cyclization,
we have examined the use of various thioesters, enones, and a-hydroxy-
amides as initiator functions. This approach has been used previously for the
successful generation of linearly fused, bridged, and spirocyclic six and seven
membered rings present in carbocyclic and azacyclic natural products.

We had hoped to utilize furans in both the cyclization sequence, and as
resident functionality to facilitate completion of chosen natural product
syntheses. During this study we addressed several questions: 1) could the
required cyclization precursors be prepared easily, 2) would furan withstand
the conditions needed for cyclization, 3) could the furans be manipulated to
provide useful synthetic intermediates, and 4) could these intermediates be
transformed into the targeted natural products.

Studies directed toward answering these questions will be presented. We
will describe the successful preparation of two bicyclo[5.3.0]decane systems
utilizing a 3-92 mode of furan closure, and one linearly fused indolizidine
system using a similar type of closure. Currently work is continuing to
complete formal total syntheses of compounds such as fastigilin-C, resin-
iferonol, and elaeokanine A.

FOR MOM, DAD, AND MIKE
THANK YOU

iii

ACKNOWLEDGMENTS

The author would like to express his deep appreciation to Steven P. Tanis
for four great years of chemistry and life. He provided more than a mere
education, but helped me learn how to think about problems for novel
solutions. A special thanks to Dr. William (The Waz) Reusch for his adept
handling of all the internal affairs we needed protection from and for serving
as my second reader and friend.

Financial support from Michigan State University and the National
Institutes of Health (GM 33947 1985-1988) are gratefully acknowledged and
appreciated.

The author would like to thank all those past and present who have
helped so much with both chemistry and life, in particular: Lisa, Paul, and
Bryon for friendship, chemical suggestions, and fun times away from the
chemistry building. Thanks to Steve Steffke for the synthesis of the tri-
substituted furan. A special thanks to Jeff who stood up for me when it was
most important. To Michele and Greg, Megan and David thanks for the
listening ears, support, and love over the highs and lows of the last four
years. Finally to Lauren, we did it together, its been six months and gets better
with each passing day.

A special note of thanks to the members of the tenure and promotions
committee for opening my eyes to the fact that being a good, conscientious
scientist is not always important in a tenure decision.

iv

TABLE OF CONTENTS

Egg
LIST OF TABLES - INTRODUCTION, CHAPTERS I, 11, AND 111 .................. vi
LIST OF FIGURES - INTRODUCTION, CHAPTERS I, II, AND 11 ................. vii
LIST OF EQUATIONS - INTRODUCTION, CHAPTERS I, 11, AND 111 .......... ix
LIST OF SCHEMES - INTRODUCTION, CHAPTERS I, 11, AND III .............. xii
LIST OF SPECTRA - INTRODUCTION CHAPTERS I, 11, AND III .............. xiv
INTRODUCTION ....................................................................................................... 1
CHAPTER I - APPROACHES TO FASTIGILIN-C ................................................ 6
EXPERIMENTAL ...................................................................................................... 37
LIST OF REFERENCES ............................................................................................ 43
CHAPTER II - APROACHES TO RESINIFERONOL ......................................... 48
EXPERIMENTAL ...................................................................................................... 62
LIST OF REFERENCES ............................................................................................ 66
CHAPTER III - APPROACHES TO ELAEOKANINE A .................................... 69
EXPERIMENTAL ...................................................................................................... 86
LIST OF REFERENCES ............................................................................................ 91
CONCLUSIONS ........................................................................................................ 85

LIST OF TABLES

 

Page
CHAPTER I
Table 1 Effect of Silyl Substituents ................................................... 23

vi

LIST OF FIGURES

 

 

 

 

Page

INTRODUCTION
Figure 1 Furan Oxidation States .......................................................... 2
Figure 2 Generalized Cyclization Modes ........................................... 6
CHAPTER I
Figure 3 Generalized Pseudoguaianolides ........................................ 6
Figure 4 Possible Retrosyntheses for Fastigilin-C ............................ 8
Figure 5 Stereochemistry of Ketone Reduction ............................. 22
CHAPTER 11
Figure 1 Tigliane, Daphnane, and Ingenane Natural

Products ................................................................................... 48
Figure 2 PKC Promoters ...................................................................... 50
Figure 3 Synthetic Pharmacophore ................................................... 50
Figure 4 First Generation Retrosynthetic Analysis ....................... 54
Figure 5 Second Generation Retrosynthetic Analysis .................. 54
CHAPTER III
Figure 1 Alkaloid Natural Products .................................................. 69
Figure 2 Iminium and N-Acyliminium Ion Structures .............. 70

vii

Figure 3

Figure 4

Figure 5

Figure 6

Eneamide Dimerization ...................................................... 71

Regiochemical Results of N-Acyliminium

Ion Cyclizations ..................................................................... 71
Stereochemical Outcome of Cyclization .......................... 72
Overman Exo, Endo Cyclizations ...................................... 72

viii

LIST OF EQUATIONS

 

 

Page

INTRODUCTION
(1) Allylic Alcohol Cyclization ................................................... 1
(2) Tertiary Carbocation Cyclization ......................................... 1
(3) Enone Spirocyclization .......................................................... 1
(4) Bridged Cyclization ................................................................. 1
(5) Spirocyclic Cyclization ........................................................... 6
CHAPTER 1
(6) Possible Bioactivity Mechanism .......................................... 7
(7) Unsuccessful Vinyl Anion Addition ............................... 22
(8) Ireland Ester Enolate Claisen Precursor 129 .................... 23
(9) Model Enolate Claisen ......................................................... 23
(10) T hioenol Ether Addition to Cyclohexenone .................. 25
(11) Thioenol Ether Addition to Cyclopentenone ................. 25
(12) Thioenol Ether Addition to

2-Methylcyclopentenone ..................................................... 25
(13) Michael Addition/Aldol Sequence to 140 ....................... 26

ix

(14)

(15)

(16)

(17)

(18)

(19)

(20)

(21)

(22)

(23)

(24)

CHAPTER II
(1)

(2)

(3)

(4)

Attempted Protection of 140 ............................................... 27
Successful Acylation of 140 ................................................. 27
Unsuccessful Cyclization of 141 ......................................... 27
Lactonization of 141 .............................................................. 28

Successful Cyclization to Fastigilin-C

Precursor 144 .......................................................................... 28
Unsuccessful Ketalization of 144 ....................................... 32
Sodium Borohydride Reduction of 144 ........................... 32
Barton Deoxygenation of 166 .............................................. 33
Second Generation Cyclization Precursor ....................... 34
Unsuccessful Cyclization of 170 ......................................... 35

Unsuccessful Michael Addition/Alkylation

Sequence ................................................................................. 35
Synthesis of Disubstituted Furan 47 ................................. 57
Ojima a,[3 Enone Reduction .............................................. 59
Keinan 0i,[3 Enone Reduction ............................................ 61
Attempted a,B Enoate Reduction ..................................... 61

LIST OF SCHEMES

 

 

Page

INTRODUCTION
Scheme 1 Total Synthesis of Nakafuran 9 ........................................... 3
Scheme 2 Formal Total Synthesis of Aphidicolin ............................. 4
CHAPTER I
Scheme 3 Biosynthetic Pathway for the Formation

of Pseudoguaianolides ........................................................... 7
Scheme 4 Heathcock's Synthesis of Confertin .................................. 10
Scheme 5 Wender's Synthesis of Confertin ...................................... 11
Scheme 6 Semmelhack's Synthesis of Confertin ............................. 12
Scheme 7 Ziegler's Synthesis of Aromatin ........................................ 14
Scheme 8 Lansbury's Synthesis of Aromatin .................................... 15
Scheme 9 Grieco's Synthesis of Helenalin ......................................... 17
Scheme 10 Schlessinger's Synthesis of Helenalin .............................. 18
Scheme 11 Lansbury's Synthesis of 2,3-Dihydrofastigilin ................ 20
Scheme 12 Lansbury's Unsuccessful Lactone

Transposition ......................................................................... 21

xi

Scheme 13

Scheme 14

Scheme 15

Scheme 1 6

CHAPTER II
Scheme 1

Scheme 2

Scheme 3

Scheme 4

Scheme 5

Scheme 6

Scheme 7

Scheme 8

Scheme 9

CHAPTER III

Scheme 1

Scheme 2

Scheme 3

Schultz’s Synthesis of Confertin ....................................... 3O
Schultz's Synthesis of Aromatin ....................................... 31
Future Plans ........................................................................... 34
Completion of Fastigilin-C ................................................. 36
Wender's First Generation Phorbol Route ..................... 52

Wender's Second Generation Phorbol

Synthesis ................................................................................. 53
First Generation Phorbol Precursor .................................. 55
First Generation Desmethyl Phorbol ................................ 56
Synthesis of Phorbol Adduct .............................................. 58
Furan Manipulation ............................................................ 59
Attempted Ene—Dione Hydrogenation ............................. 60
Current Monoreduced Furan Sequence .......................... 60
Future Synthetic Strategy .................................................... 61
Chamberlin's Eleaokanine Synthesis ............................... 74
Speckamp Eleaokanine Synthesis ..................................... 74

Weinreb Hetero Diel-Alder Eleaokanine
Synthesis ................................................................................. 75

xii

Scheme 4 Dixon Synthesis of Epilupinine ......................................... 77

Scheme 5 Dixon Synthesis of Perhydrohistionicotoxin .................. 78
Scheme 6 Dixon Synthetic Route Toward Cocaine .......................... 80
Scheme 7 First Generation Elaeokanine Precursors ........................ 80
Scheme 8 Synthesis of Trisubstituted Furan ..................................... 82
Scheme 9 Second Generation Eleaokanine Precursors ................... 82
Scheme 10 Synthesis of Reduced Trisubstituted Furan .................... 83
Scheme 11 Synthesis of an Advanced Eleaokanine

Precursor ................................................................................. 84
Scheme 12 Future Synthetic Scheme .................................................... 84

xiii

LIST OF SPECTRA

 

Page
CHAPTER I
(1) Cyclized Fastigilin Precursor 144 ....................................... 29

xiv

INTRO DUCT ION

INTRODUCTION

Our research group has been interested in the development of the furan
terminated cationic cyclization as a synthetic method. These studies have
culminated in the synthesis of spirocyclic-, linearly-fused, and bicydic alkaloid
and terpenoid natural productsl, utilizing both 2-, and 3- substituted furans as
terminators2 in the cyclization process. Epoxides,1d allylic alcohols,1d
enones,1e and carbinol amidesle have served as initiator functions in these
reactions, allowing us to prepare various 5, 6, and 7 membered ring
containing systems; and the furan moiety has helped establish both regio- and
stereochemical control in a number of systems that would otherwise be
difficult to handle. Generic (3—-)2) furan terminated cyclizations are illustrated
in Eqs. 1-4.

HO \
\ HCOOH

61-68% (1)

(2)

 

Enone (3)

 

BFa'OElz

Enone o (4)
79 °/o

2

The introduction of a furan into the newly constructed molecule serves
several functions. Initially, furan serves as a regiochemical control element
for the introduction of one or two oxygen atoms and functionalized carbon
molecules. A fused furan ring can further serve to restrict the freedom of a
conformationally mobile system such as a cycloheptyl ring by adding the
enforced rigidity of a double bond to the framework. Finally, furan can
function as the operational equivalent of a number of 1,4-dioxygenated chains
as well as five and six membered rings after additional manipulation (Fig. 1).

Recently, we have successfully concluded the syntheses of several targets
that serve as examples of our ongoing research directions. In the case of
bridged carbocyclic systems, Herrinton and Tanis3 have completed the syn-
thesis of N akafuran 9, a natural fish antifeedant. Furyl Grignard reagent 1
(CuCN) was added in SN2' fashion to vinyl epoxide 2, to give allylic alcohol 3
(62%). Oxidation, conjugate addition (MeCu, BF3-(OEt)2)4, and introduction of
an enone double bond (selenation, oxidation) gave compound 5 that serves as
the precyclization substrate. Acid catalyzed cationic cyclization then gave an
excellent 79% yield of the bicyclo[4.3.1]decane system 6. The synthesis was
completed by Wittig olefination and acid catalyzed double bond migration to
give nakafuran 9 (8).

 

 

00

Figure 1: Furan Oxidation States

3

As an example of the linearly fused terpenes which might be prepared via
a furan terminated cyclization, Scheme 2 outlines our formal total synthesis
of aphidicolin5, a diterpene tetraol that has shown considerable activity
against both Herpes and leukemia.. Especially interesting in this case, is the
fact that the molecule has been synthesized in both racemic and chiral forms,
demonstrating that a complete transfer of asymmetry is possible in the furan
terminated cationic cyclization. In the event geraniol was converted to chiral
epoxide 11 through a Sharpless6 asymmetric epoxidation, further reactions
added a furan ring to give compound 13. Triethylamine—moderated BF3~OEt2
cyclization gave a excellent (72%) yield of 14, with no trace of the alternative
enantiomer detected. The conversion of 14 to the known aphidicolin
precursor 20 is presented in the latter portion of Scheme 2.

HO \
/ \ CuCN \ 1) mo
0 62% 2) MeCu,
3

 

BF3'OEi2, 62%

‘- ‘<\O
ESe Br CCEOHiz i '
72% 79%

P: H O 0
be; JED: ——~"“°’°“ d“ .
° 80%
95 :5

7 8
Scheme1: Total Synthesis of Nakafuran 9

 

/

I

 

 

1) DIPT

 

 

OBz 1" Ci 08211—0 1NaOMe

\ 0H 2) 3900? TBHP, 83% 2) nBuLi

TBHp H 2) BnBr, 88% stCl

LiCI, 85%
9

0'1) 3-Furyl MgCI: BF3 OEtg Swern

LigcuCL;B EtaN, -78°C:H 97%

84% 72%

 

 

 

  

1) MgBrz
L-Selectride:

2) LAH Hoe“
89% 8.521 Ho_s'

 

16

 

 

1) N88. DMF; MQEBE
2) nBuLi, Mel ‘ 97%
65% +0;

 

Scheme 2: Formal Total Synthesis of Aphidicolin

Finally, in the case of the spirocyclic carbocycles, an interesting bicyclo-
[5.4.0]decane has been prepared by McMills and Tanis7(Equation 5). This
system evolved, somewhat unexpectedly, from a separate study directed
toward the synthesis of simple guaianolides such as estafiatin. Toward that
end, thioketal 21 was metallated (nBuLi) and was then added to 3-(3-
furyl)ethanal to give alcohol 22 (90%). When this substrate was treated with
formic acid, did not cyclize to form the linearly fused compound as
anticipated, but instead opened and reclosed the thioketal function, with the
furan attacking the 2-position of the double bond to give spirocycle 23 (84%).

These simple examples clearly indicate that furan terminated cationic
cyclizations can be employed in the synthesis of a variety of alkaloids and
terpenoids. These data when considered together with many examples from
our laboratories of the preparation of 7-membered rings suggested the
application of this methodology to the synthesis of the functionally and

5

stereochemically complex bicyclo[5.3.0]decane containing natural products
such as the pseudoguaianolides or tigliane/daphnane diterpenes. We
reasoned that the inclusion of the furyl 2,3-double bond into the 7-membered
ring would provide well defined conformations of the normally flexible and
troublesome carbocycle, thus enabling us to resort to standard cyclic
methodology for the establishment of additional asymmetric centers in a
predictable fashion. To this end, we have chosen to incorporate three
different cyclization modes, designated as A, B, and C. With these three
simple disconnections and furan placements, a variety of diterpenes, pseudo-
guaianolide, and tigliane/daphnane diterpenes have become available as
synthetic targets. These closures will be discussed in more detail in
subsequent sections of this dissertation.

 

 

r-\ l_\
S S S S \
1) nBuLi :
2) 3-(3-Furyl)— ( 5)
Br pmpanal
90% HO
2 1 2 2 2 3

R: H: Estaflatin
R: H: Compressanolide
R: CH3 : Damsin

 

   

o F RAN R .. [3 CH3 .. Confertin
R = (1 CH3 = Fastigillin

O \
\
CLOSURE .0 WSW Tigliane Diterpenes
R R'

Figure 2: Generalized Cyclization Modes

APPROACHES TO FASTIGILIN—C

APPROACHES TO FASTIGILIN-C

I“
O..“

     

O 00 O :
HO

0 .‘
O
(i) Contertin 24 (i) Aromatin 25 (i) Helenalin 26 O=<=<

(i) Fastigilin-C 27

Flgure 3: Pseudoguaianolide Natural Products

Pseudoguaianolides8 are a class of natural products that contain a bicyclo-
[5.3.0]decane ring system. These natural products are thought to arise from a
cascade of intermediates starting with trans, trans farnesyl pyrophosphate.9
Thus, bond formation initiated by loss of pyrophosphate gives a
germacradiene (or triene), and enzymatic oxidation of carbon 6 or 8 with
subsequent lactonization generates two possible germacranolides 32, and its
C6 regioisomer. Cation induced cyclization and trapping by water leads to the
guaianolides that are the biological precursor of the pseudoguaianolides.
Migration of a methyl group from C4 to C5 generates two regiochemically
different types of pseudoguaianolides, as seen in Scheme 3.

Much of the biological activity of the sesquiterpene lactones has been traced
to the ability of the a-methylene lactone to act as a Michael acceptor for thiol
containing enzymes10 (see Equation 6). This inhibits amino acid
incorporation into proteins causing a disruption of metabolism at the cellular
level. Cytotoxic activity has also been linked to the presence of the a-
methylene lactone, but tends to be enhanced by the presence of a conjugated
cyclopentenone and hydroxyl groups about the periphery. These compounds
are suspected to cause an inhibition of DNA transcription or synthesis.4

7

Contact dermatitis is another problem associated with these compounds". It
is thought that a Michael reaction occurs with a skin protein forming an
antigen that sensitizes some lymphocytes.6 This causes an allergic response
that can range from very mild to extremely severe. Finally, a number of these
sesquiterpene lactones have exhibited both phytotoxic and antimicrobial
properties. Hager has postulated that attack of sulfur at an auxin receptor
occurs, then, once bound to the protein receptor, the thiol (RSH) is set free.12
The formation of the sulfur bond then causes irreversible inhibition of plant
growth. In many cases, these lactones are stress metabolites (they are formed
during periods of stress to a plant example pest attack, drought or
overexposure) and act as a chemical defense for the plant.

\
/ \ 1) - H t
2) l 0]
3 0 0H
1) 08 or 06 O
022) Lactonization
326

. Caged;

 

3 3
SCHEME 3: Biosynthetic Pathway Toward the PseudoguaianolidesO

S-Enz
(fie . Ema. __..... 43L;

In recent years the pseudoguaianolides were discovered to have potent
anti-inflammatory and antitumor activity. In arthritis screens, compounds
containing a-methylene lactones, a,B-unsaturated cyclopentenones and
epoxy-cyclopentanones showed significant activity at levels as low as 2.5

8

mg/kg/day.14 Induced pleurisy, anaphylaxis and hypersensitivity were
suppressed as well. Studies of the inhibition of lysosomal enzyme activity by
derivatives of helenalin were conducted to determine what structural
features were necessary for activity.” It was found that masking the a-
methylene lactone, loss of the hydroxyl at C6, or changing the lactone from cis
to trans all resulted in reduced activity. These pseudoguaianolides, applied at
5 x 10'4 M, compare favorably with standard antiinflammatory agents such as
salicylates (1 x 10'3 M) and phenylbutazone (1 x 10'4 M).16

Helenalin is an active cytotoxic agent for human KB or H. Ep-2 carcinoma
cell lines.17 At a therapeutic level (8 mg/kg), helenalin has shown little
toxicity in mice, but at a level of 25 mg/ kg some cardiac toxicity has been
noted. In the case of P-388 lymphocytic leukemia, the bis ester of helenalin
with succinic anhydride has been shown to act by suppressing DNA synthesis
as well as inhibition of protein and RNA synthesis.18 Hall et al. have cited
the lack of sufficient quantities of the natural products and limited
information about their mode of action as reasons why the
pseudoguaianolides have not been used in clinical trials.19 One goal of this
study is to develop an efficient route to these compounds and to provide
adequate quantities of representative compounds to make possible human
studies of the various biological activities.

The pseudoguaianolides are grouped in two categories 1) ambrosanolides,
with a B-oriented methyl group at C10 and 2) helenanolides, which are
biologically more complex with an a-methyl group at C10. The ambros-
anolides are illustrated by the compounds mexicanin and confertin. These
relatively simple ambrosanolides are contrasted by the more highly
functionalized and biologically potent helenanolides such as helenalin and
fastigilin-C. Fastigilin-C is currently the only member of this class of
compounds to have eluded total synthesis.

5
o

H H ‘-
0'“ CO

Amhmsaneum W

Flgure 4: Generalized Pseudoguaianolides

9

Many groups have examined the preparation of, or approaches toward the
synthesis of representative pseudoguaianolides. Thus, syntheses of confertin,
range in overall yields from a high of 10-15% by Wender and Schlessinger20
to low overall yields of 0.8 - 0.9% by Marshall21 and Heathcock have been
reported. In most cases, ring A has come from substituted methyl
cyclopentenone with some notable exceptions. Schultz's ring A was derived
from methyl 1,3-cyclopentanedione and Heathcock's ring A from a
substituted cyclohexenone. We will describe the preparation of repre-
sentative pseudoguaianolides in Schemes 4-12.

Heathcock22 (Scheme 4) begins his approach to confertin 24 from the
known ketal acetate 38. Osmylation from the less bulky B-side (blocked by the
dioxolane), tosylation of the secondary alcohol (stCl, pyridine) and
deprotection of the acetoxy-diol (K2CO3, MeOH) gave compound 40. The
resulting diol tosylate was exposed to solvolytic conditions (LiOH, tBuOH,
65°C) to give 41 and 42 epimeric at C1 (85%), in an equilibrium ratio of ~ 4:1.
Protection of the free hydroxyl as an acetate and subsequent reaction with
Me3Al, MeLi and quenching with NH4Cl/H20 gave enone 44 as the sole
product in 70% yield. At this point, Heathcock can diverge with this enone 44
to produce either the helenanolides loot-CH3 orientation .(Li/NH3) or the
ambrosanolides IOB-CH3 configuration (Hz/Rh-A1203). In the event,
reduction of enone 44 with Hz/Rh-Aleg gave 45 (83%), which was enolized
(LDA) and the enolate then alkylated with methyl bromoacetate to afford 46
(56%) selectively with C-ring elements in place. Hydrolysis of the ester and
subsequent closure (excess HClO4) gave crude butenolide 47 (34%) which was
hydrogenated (H2, Rh/A1203) to provide butyrolactone 48 (34%). Lactone 49
was then smoothly converted to confertin 24 (31%) as shown.

TsO

>1) 0804 OH O
6. o NMO,74% .. o
2) stCl
82%

A00
3 8 A°° 3 9

   
 
 
  
 
 
 

LiOH
M C
85%

 

1) MeaAl
2) MeLi

3) NH4C|
H20, 70°/o

1 LDA

acetate, 56%

1) xs HCIO4_
2) Ac20 7
002H AcO

 

 
     
  

 

 

   

1) KOH, MeOH. 80% 1) CH3I
2) LDA, 2) NaHCOa
|\’|82'\|==C"'2I A00 0 3) Jones, 31% O 0

4 9 NMe2 2 4
SCHEME 4: Heathcock's Synthesis of Confertin

Two substantially different synthetic routes toward confertin were taken by
Wender and Semmelhack. Wender23 (Scheme 5) chose a divinyl cyclo-
propane rearrangement to form the hydroazulene system 54. The
preparation of the substrate suitable for rearrangement began with the
addition of lithio cyclopropane 52 to 3-ethoxy-2-methylcyclopentenone
followed by hydrolysis the enol ether to give 53 (72%). Irradiation of the
divinyl cyclopropane at 98°C resulted in cyclization to form the
bicyclo[5.3.0]decane 54 (80%). Dieneone 54 was then protected (ethylene glycol,
benzene, reflux) and oxidized (PCC) to give a 9:1 ratio of dieneone 56 (70%)
regioisomers. Epoxidation (H202, NaOH, 70%) and Horner-Emmons
olefination (Na(EtO)2POCHCO2Et, 80%) of the ketone gave the epoxy triene
57. Finally, compound 57 was subjected to ester hydrolysis (aq. H+), epoxide

1 1
opening (10% H2504) to form a lactone, followed by dehydration (30% N aOH)
to furnish triene-lactone 59. Hydrogenation (H2, Pd) established the ring
fusion, the C10-CH3, and butyrolactone stereocenter in the desired sense

affording lactone 60, which was then converted to (1)-confertin by the method
of Heathcock.

O

 

 

 

 

 

 

CH CH3 CH3
fi/I'. 3 &/ CH3 ~H 72% “flifi‘; m
OEt L' Li
5 1
ethylene
m) gtyool PCC
98°C .0 benzene 0.0 70% O.
90% O reflux K10 0
5 4 K15
OH
1) H202 80°/o
O
2) Na(EtO)2POCHCOZEt \
95 :5 2: E O
80% NaOH; H2, Pd ' 1) LDA :
2) M92N=CH2 l
(IL/o O 3) CHal

60 4) NaHCOa

5)10% H2304
O
O O

2 4
SCHEME 5: Wenders Synthesis of Confertin

Semmelhack24 employed yet another interesting variation, a metal
promoted cyclization-lactonization, in his successful synthesis of (i)-confertin
24 (Scheme 6). Using a tandem Michael addition/alkylation sequence,
cyclopentanone 62 was prepared in 85% yield, as illustrated in Scheme 6.
Reduction (LAH) and oxidation (CrO3) gave aldehyde 63 in 85% yield ; and
this was added in Horner-Emmons fashion to the sodium or lithium salt

12

formed by reaction of sodium or lithium isopropylmercaptide with trimethyl
phosphonoacrylate to give a-thiomethyl enoate 64 (70-85%) as a Z/E mixture
(4:1(Na) to 1:8(Li)). Hydrolysis (pTSA, MeOH) of the methoxymethyl
protecting group, followed by careful Moffatt oxidation (dicyclohexyl-
carbodiimide, TFA) of the resulting hydroxyl group generated an aldehyde in
a 9:1 isomer ratio, contaminated with some of the opposite olefin
stereoisomer. Treatment of the mixture with "Magic Methyl" (CH3S03F)
resulted in a sulfur ylide 65, which was treated with either Zn(0) or Ni(0) to
promote cyclization. Zinc (zinc-copper couple) gave a product that was
determined to be the a-cis fused lactone (30%), whereas nickel (nickel cyclo-
octadiene) gave a 2:1 ratio of the B-cis fused lactone (66), (12)-confertin, to the
a-cis fused lactone in 43% yield. Semmelhack has not yet explained the role
of either the double bond geometry or the nature of the metal employed in
the stereochemical outcome of the cyclization.

 

 

 

    

 

 

1) tBuLi \o ‘o
2) 3,3-dimethyl H 0" H J
B butynyl copper“): ' 1) LAH ' 0
GAO 3) 2-methyl 0020113 2) CrOa CHO
| cyclopentenone O 85% O
6 1 4) methyl 85% 5 2 5 3
bromoacetate
1)>-S\_<.P0(0CH3)2 '3 OH L" CH0
C02Cl13"> 1) Moffatt _
70-85% 0 O G)
6 4 S-< 6 5 5-<
I
:1
NigCOD); ' 1) nPrSH :
43% 2:1 2) H2, pd
O 3) CH30802F 0
4 N
6 6 ) 82003 2 4

SCHEME 6: Semmelhack's Synthesis of Confertin

The simple helenanolides, represented by aromatin 25, have also served as
targets for total synthesis. Ziegler's25 approach to the synthesis of aromatin
used dithianylidene anion chemistry, as described in Scheme 7. Thus,
dithianyl anion 67 was added to 2-methyl-cyclopentenone and the resulting

13

copper enolate (CuI, P(OMe)3) was trapped by reaction with allyl bromide to
give compound 68 in 50% yield. Ozonolysis then provided ketones 69 (67%)
and its epimer in a 18:1 ratio, and aldol cyclization (2% KOH-MeOH) afforded
a single aldol product 70 (85%), in which the C10-CH3 has epimerized to the
favored (it-(equatorial) orientation. Dehydration (MeOH—P2Os) of 70, followed
by reduction (LAH), gave diol 71 (83%), which was ideally suited for the C-
ring annulation via Claisen technology. The Eschenmoser variant of the
Claisen rearrangement (Me2NCH3CH(OMe)2) afforded eneamide 72 (72%)
after saponification. Iodolactonization (I2, aq. THF) followed by reductive
elimination ((Bu)3SnH) of iodine generates 73, thus establishing the
bicyclo[5.3.0]decane ring system. Ziegler completed the synthesis of (i)
aromatin with an interesting elaboration of the C-ring to form the on-
methylene butyrolactone. Reaction of lactone 73 with bis(dimethyl-
amino)methoxymethane (Bredereck's Reagent) gave the related vinylogous
carbamate which provided (i)-aromatin after reduction (DIBAL), acidification
(NH4C1), and oxidation (PCC) to the 4-one product. Ziegler has also
completed a synthesis of (i) confertin from compound 72 via iodine
elimination (DBN) and hydrogenation (H2, Pd) of the resulting olefin. The
latter reduction gave an excellent yield of the B-methyl product.

Lansbury26 has also completed a synthesis of (:t)-aromatin as described in
Scheme 8. Propargylation of 2-methyl-1,3-cyclopentanedione gave 77 (85%),
which added propenyl lithium to afford compound 78. Cyclization of 78 by
treatment with formic acid (90% HCOOH, 80°C) yielded dione 79. Compound
79 was reduced (H2, Pd), providing the undesired B—methyl stereochemistry, a
situation that was remedied, but at the cost of a number of steps, as illustrated
in (Scheme 8).27 With the desired C10 a-methyl ketone 84 in hand, Lansbury
then added the elements of the C-ring butyrolactone by first forming the c.0-
unsaturated ester 85 (100%) via Petersen olefination, followed by double bond
migration to the 7,8-position by kinetic-controlled enolate protonation.
Reduction of both the double bond and ester functions by diborane gave diol
87 (95%) after alkaline peroxide treatment. The C-ring was then formed by a
Pt/O2 oxidation giving lactone 88 (50%), which led to (i)-aromatin, as
described in Scheme 8.

   
   
  

1) CH3303H

2)LAH

 

 

N,N Dimethylacetamide ! '- :13.
Dirnethyl Acetal : . 1) 390- '2 39% . O Bredereck's:
Reflux, 72% 2) nBuaan 94% Reagent
O 1 00%
H0 CONMe2 HO
7 2 7 3
i -. PCC
97% 1 00%
HO O
7 5

   

SCHEME 7: Ziegler's Synthesis of Aromatin

15

O 0 HO
N H - -
+ __\ 21—3322. 2 Llproraege:
Br 85%
0 o
76 78

 

 

90% HCOOH .‘ H2. Pd 1) NaH :
80°C 2) MeSOzPh
50% O O

7 9
H

 

 
 
 
 
 
 
 
 

2 7° 2*

A020
.. S\0 CH3303H .‘ SE
0 O

96%
8 1

 

O
8 2
H ’1
, 9
EN... . 1) LDA ;
96°/o 2) TMSCH2CO2E1
O 0 1 00%
8 4

 

SCHEME 8: Lansbury’s Synthesis of Aromatin

16

The final examples of pseudoguaianolide syntheses discussed here are
directed toward the more highly oxygenated helenanolides, such as helenalin,
and fastigilin-C. As might be expected for these more functionalized ring
systems, the number of syntheses reported decreases as the complexity of
functionality and stereochemistry increase. Helenalin has been synthesized
by Grieco and Schlessinger. Fastigilin-C has not yet yielded to total synthesis,
however, a synthesis of 2,3-dihydrofastigilin was recently reported by
Lansbury.

Grieco28 began his synthesis of (:t) helenalin 26 (Scheme 9) in a fashion
very similar to his prior pseudoguaianolide preparations from rigid bicyclic
precursors. The known bicyclo[2.2.1]-heptanone 90 was alkylated (LDA, MeI)
to give exclusively the endo methylated ketone 91 (94%). A base catalyzed
Baeyer-Villiger oxidation (H202,OH') and subsequent lactone opening gave 92
after esterification (CH2N 2, 86%). These steps unmasked the A-ring as a
cyclopentanol, having the desired ring fusion stereochemistry; however the
C10-CH3 has the ambrosanolide configuration. This latter concern was
readily corrected by an equilibrium-driven epimerization favoring the desired
C10-0i-CH3 (helenanolide) orientation. Toward that end, 92 was converted to
the desired lactone 93 in a series of five steps as drawn. Reduction of 93
(DIBAL) and Wittig olefination (@3P=CHOCH3) gave an enol ether, which led
to keto-aldehyde 95 after hydrolysis, MeLi addition, and oxidation (PCC, 45%
over 5 steps). Aldol condensation (KOH, MeOH) and dehydration (stOH)
then gave enone 96 (70%), which was epoxidized (tBuOOH, Triton B, 87%)
and reduced (NaBH4, 99%) to give alcohol 98. The final stages of this
helenalin synthesis involved dilithioacetate addition to the epoxide, removal
of the C4 oxygen benzyl protecting group (Li, NH3) and acidic quenching to
give lactone 99 (86%). This was smoothly converted to the target compound
as shown in Scheme 9.

OTHP
LDA, Mel
94%

 

l 7
OTHP
H
(E 1) H202, NaOl-L '
o 2) CH2N2 T

 

 

 

 

C02M6
3) NaH, BnBr
52% BnO 9 2 OTHP
H I H g
-.- O -
1)stOH. M6011: 1) DIBAL : ' ’CHOM" 1) H91
2) KOH, EtOH O 2) 03P:CHOM9 2) MeLi
3) stCl. 030 BnO 87% 3) P00
0 BnO OH
67 /o 9 3 9 4 76%
o ..
H H'i H ‘1
KOH MeOH Bnc tBuOOlJ O
CH O 70% Triton B
870/0 '0, ‘.
BnO Bn 0 'O
9 5
l: . 11-:
NaBH 0' O 1) Dilithioacetatet HO'E‘ 1) DHP, H+ _
93% 2) Li, NH3 2) LDA CHZO ’
'o. ’ 86% O 3) MsCl, pyridine
3"0 O 4) DBU 61% overall
H9 9

 

1) ACOH, H20 787:
2) Mn02 65%

   

SCHEME 9: Grieco Synthesis of Helenalin

In the Schlessinger synthesis of helenalin29 an epoxy alcohol intermediate
(110) very similar to that described by Grieco, was prepared by a completely
different procedure. The known enone 101 was converted to lactam 102 using
the Barton rearrangement (MeNHOH-HCl, 40°C, stCl) . Addition of lithio
dimethyl methylphosphonate to lactam 102 occurred with subsequent ring
opening to a metallated cyclopentyl enamine that gave aldehyde 103 after
hydrolytic workup. Retroaldol fragmentation of aldehyde 103 on treatment
with KOtBu generated a Horner-Emmons intermediate which closed to
enone 104, after intramolecular capture of the aldehyde. Cuprate addition
(MeMgBr, CuI) to enone 104 then introduced the C10-CH3 from the desired

and less sterically encumbered or face leading to 105 (67%). To complete the

18

synthesis of known intermediate 109, Schlessinger was then faced with the
task of introducing two double bonds at the 2,3 and 6,7 positions as well as
manipulating of the oxidation states at C4 and C8. Toward that end, ketone
105 was deprotected (H+), the C8 ketone was blocked as the corresponding
dioxolane (ethylene glycol, stOH), and the C4-one was introduced by
oxidation (PCC). As described in Scheme 10, ketone 106 was converted to the
A-ring enone; this was reduced, the resulting alcohol protected as a THP
ether, and the A6,7-double bond was introduced via Saegusa oxidation of the
silyl enolether after hydrolysis of the dioxolane. The remainder of
Schlessinger's synthesis followed that of Grieco to afford (:t)-helenalin 26.

 

 

 

(CH3O)20P
NCH, OHC OH
.6 O1) MeNHOH. HCI: OLiCH2PO(OCH3)2
2) stCl
tBu.
1 0 1
Hi.
KOtBu 1) MeMgBr 1) HCI, MeOH _
50% overall 2) Cul 87% 2) Ethylene glycol
benzene, heat
tBu O tBu O 3) FCC 97%

   

 

 

. Hi
1)NaH.0282 4 051) DIBAL
.. :3 2) MCPBA ' C. OJ-?)HC| MeOH Page: 0
o

 

 

 

 

 

O 3) 110°, P(OMe)3 3) DHP H+ TH
1 06 92% 1 o 7 84% 1 08
s, a,
E1 t' 1) Dilithioacetate
1) LiHMDS a». 01) H202. NaOH . OH HMPA L
2) TMS-Cl 2) TRIBAL 2) TMS-Cl
3) Pd£OAC)2 THPO THPO 55% from 109
73 / 09
t!"- H‘=
a. o 1)MMC 76% 2
2)CH2O, ElgNH'
TMSO 2'. o 3) MnOz
TMSO
111 26

SCHEME 10:80hlessinger's Synthesis of Helenalin

19

Fastigilin-C 27, one of the most complex of the pseudoguaianolides is
functionalized at all seven carbons about the B—ring. As a result of this factor,
and two peculiarities noted during Lansbury's30 attempted synthesis of 27,
fastigilin-C has resisted total synthesis to date. The efforts of Lansbury, which
comprised routes A and B are described in Schemes 11 and 12 respectively.

Route A begins with the bromination (LiHMDS, NBS) of the enol silyl
ether derived from ketone 112 to give 113, followed by reduction (N aBH4) of
the C9 ketone to give alcohol 114, with the incorrect stereochemistry at C9 for
fastigilin-C. Therefore the hydroxyl group was eliminated (zinc) to form
olefin 115 (90%), and a "Wet" Prevost reaction (HOAc, H20, Ag(OAc)2, 12) was
employed to give a 5:1 ratio of the 86,913 diacetate (Ac20, DMAP) 116 in 90%
yield to the related 8a,9a diacetate. Saponification (aq. N aOH) of 116 followed
by acid treatment and heat afforded a 9:1 mixture of the 9B-hydroxy-7,8 cis
lactone 117 (89%). With all stereochemical aspects of fastigilin correctly
established in diol lactone 117, Lansbury simply needed to introduce the A 2,3
ene-4-one double bond and functionalize both the C6 alcohol and the lactone
methylene carbon. Toward that end, Lansbury examined route B the
introduction of the A 2,3 double bond at an earlier stage prior to the B-ring
elaboration. As seen in Scheme 12, keto diacetate 123 was selenylated by the
method of Grieco (HCl, PhSeCl), the selenide oxidized, eliminated, and the C4
ketone converted to the 4B—SEM ether 126 ( a. NaBH4, CeC13, b. SEM-Cl). At
this stage, Lansbury was unable to transpose the lactone, therefore, he chose to
return to the nearly completed fastigilin skeleton and introduce the enone as
the final step for the completion the synthesis.

1) Lil-lMDS
__IM&QL.
2) NBS

tBu 70% tBu o

       

   
   

1) HOAc, Ag(OAc)2
l2

 

1 NaOH EtOH
A %
0 ° 2) 10% HCI

89%

2) A020, DMAP
90% 5 Z 1

87%
tBu

1 00%

Selenylate
CL ;

Brominate

No Elimination

 

1 2 2
SCHEME 11: Lansbury's 2,3-Dihydroiastigilin-C

Instead Lansbury chose to further examine route A to attempt to
functionalize the completed fastigilin ring precursor (see Scheme 11). 0t-
methylation of lactone 117 (LDA, CO2, 87%), then alkylation with
Eschenmoser salt (CH2=NMe2I) of the carboxylactone gave a-methylene
butyrolactone 119 (86%). Methylene lactone 119 was selectively converted to
the diester (C9-OH, TFAA, pyridine, 89%) and when heated to reflux with [3,8
dimethylacryloyl chloride gave the senecoiate ester 121 (100%). Hydrolysis of
the C4 ether (stA) and oxidation (PCC) gave 2,3-dihydrofastigilin-9-

21

trifluoroacetate 122 which resisted all attempts at introduction of the A2,3
double bond. The mild conditions necessary to ensure survival of the
resident functionality (CLCS trans orientation) and the steric hindrance
provided by the axial 5B-CH3 and 6a-ester conspire to prevent C-3
functionalization. Thus the synthesis of (1)-27 remains unfinished.

H2: OAC
0 Ac 1) 03901, HCI:

  
  
 

 

NaBH4, Ceca;

 

2) Na|04
_. OAc
. SEM-Cl W No Lactone
C. OAC OAC Transposition
SEMO

 

SCHEME 12: Lansbury's Unsuccessful Lactone Transposition

Our interest centered upon the synthesis of fastigilin-C 27 via our
previously established furan terminated cationic cyclization methodology.
We reasoned that with proper masking of the resident functionality in a
more robust form, and incorporation of the butyrolactone in latent form as a
furan, that we would be able to overcome the the functionalization and
conformational problems encountered by Lansbury. One potential retro-
synthetic approach to fastigilin-C is presented in Figure 5 (route A). The
crucial steps in this proposed sequence are the furan terminated acylium ion
initiated cyclization which establishes the 7-membered B-ring and the three
component coupling which establishes the cyclization substrate. Our concern
with the latter were twofold; i) what equivalent of a 2-metallated acrylate
would add to 2-methylcyclopentenone, and ii) could we expect any control in
the development of the C6 oxygen bond?

Marino31 has demonstrated the utility of the anion prepared from 2-
bromoacrolein in conjugate addition sequences, however, we were unable to
detect even trace amounts of products which correspond to 1,4 addition after
quenching with H+, TMS—Cl, or 3-furaldehyde (eq. 7), therefore, the route was
abandoned.

 

Figure 5 : Possible Retrosynthesis for Fastigilin-C

O
O OCH3 g \ \
O " H

Another alternative which was considered was postulated from the retro-
synthetic analysis presented in Figure 5 (route B). We examined a route
which would establish the desired cyclization substrates via an Ireland32 type
enolate Claisen rearrangement. Some precedent for this type of construction
was found in the work of Burke.33 Important questions to be answered in
this approach were i) will the depicted rearrangement occur through the
cyclopentyl unit or the furyl moiety, ii) what will be the selectivity at the 2
methyl bearing centers, and iii) can the C5-CH3 and the C6-OH be added with
the desired stereochemistry?

23

Toward those ends, we constructed a model system as outlined in
Equations 8 and 9. Metallation of 2-bromocyclopentenone—ethylene ketal
(nBuLi) and quenching with (CH20)n by the procedure of Smith“, afforded on-
hydroxymethyl-ketal 128 (70%), which was acylated with propionyl chloride
to give propionate 129 (96%). Enolization of 129 under standard Ireland
conditions (LDA, THF, - 8°C) followed by silylation with TBDMS-Cl provided
the E-enolsilyl ether35 which was warmed to room temperature. We
obtained a 90% yield of esters in a ca. 10:1 ratio, with the stereochemistry
anticipated to be as predicted by literature precedent.

 

 

 

‘ ‘ I ‘ o (\0
o o O O 0
Br - i- 0 ON 0
W 0H .. : (a)
2) (CHO)n Pyridine
70% 96% O
1.21 133 12.2
l/\o I—\
o o O O
o 1) LDA-78° H30:
: -——-.- (9)
l/l§02) TBDMS Cl 0 74% O
”fag/RT : OTBDMS = OH
° H 5 H 5
123 129 1.3.1
Table 1. Effect of Silyl Substituent
\
1’\ \
o
o o
0 gum-78° A
2) RasiCI 7
(KO 3) H30" H3 \ 0
CH3
COzH
1.3.2 1.3.5
flailQL
t—BuMeg
(Mela

 

24

With that information in hand, we next examined a system where furan
had been incorporated. Our hydroxy methyl component required an
additional furan to be placed at the hydroxymethyl carbon. Therefore we
substituted 3-furaldehyde for formaldehyde in the cuprate reaction (127,
nBuLi, CuI) of vinyl anion formed to give an alcohol (89%). The alcohol was
esterified (propionyl chloride, pyridine) to give 132 in a 95% chromato-
graphed yield. We then subjected propionate 132 to the enolate Claisen
conditions (Table 1) used in our model system (LDA, -78°C, TBDMS-Cl, RT)
and we were pleased to find, upon hydrolysis, (5% aq HCl) a 69% yield of a
two component mixture in a 95:5 ratio. The major product was found to be
the expected mono-substituted furan acid 133, the minor product was
determined to be a disubstituted furan 134, a product that arose from an ester
enolate Claisen reaction with one of the double bonds of the furan instead of
the cyclopentenyl double bond.36 We discovered that the nature of the
substituents on the silylating reagent had a profound effect (Table 1) upon the
course of the reaction.

Our next task was to modify the enone and in the process introduce the C5-
CH3 and the C6-OH groups. We examined a number of sequences including
enone reduction and alkylation, epoxidation and epoxide opening, and were
frustrated by our inability to introduce the CH3 correctly with either the
absence or the presence of the OH. These difficulties, and an interesting
report by Mukaiyama, caused us again to modify our approach.

While we were struggling with our attempts to introduce the C5-CH3 and
the C6 oxygen to the enolate Claisen product, we took note of a series of
reports by Mukaiyama37 which described a trityl cation catalyzed addition of
ketene acetals to enones, including cycloalkenones. Mukaiyama reported
good yields of either m or 131211 (eq 10,11) products as a function of substrates
and reaction conditions. Of greater importance was the report of the addition
of enolsilyl ethers of t-butylthiopropionate to 2-methylcyclopentenone to give
predominantly the trans arrangement (eq 12) about cyclopentane ring, and
also afforded the correct relative stereochemistry at C10, C1, C5. The utility of
the thioester moiety for cationic processes was also appreciated for our
projected furan terminated ring closure.

25

 

 

 

 

 

o
OTBDMS TrClO j °/o
d * JV 2 -4(3m° )= (10)
RS -pyndyl methanol
1 3 5
o
+ OTBDMS TerCls (3 mol %)_ (1 1)
RSJV 2-pyridyl methanol 7
o
+ mews TerCl6 (3 mol VOL (1 2)
IRS/S 2-pyridyl methanol?
1 3 8

Given Mukaiyama's success in securing a number of centers needed for the
preparation of fastigilin-C, we decided to examine this approach as an
alternative to the troublesome Claisen functionalization. Toward that end, t-
butyl thiopropionate (available from Columbia Organics ) was deprotonated
according to Gennari (LDA, -78°C) and trapped (TMS—Cl) to form the
relatively stable thio silyloxy enol ether 136.38 Alternatively, we also prepared
136 by treating the thioester with TMSOTF (RT).39 In each of these cases we
obtained the desired ketene acetal in ca. 80% yield as a 95:5 ratio of Z to E
olefin isomers, as determined by gas chromatography and comparison of our
data with that of Gennari. The other components for the three component
Michael addition-Aldol condensation, 2-methylcyclopentenone and 3-
furaldehyde are commercially available in research quantities. With the
reagents in hand, we studied the experimental sequence utilizing the
following protocol: to 2—methylcyclopentenone in dry CHzClz cooled to -95°C
was added the distilled ketene acetal followed by 3-5 mole% of Ph3CSbC16
(TerClg), and then finally 3-furaldehyde in CH2C12 was added dropwise
(Equation 13). During our study of this sequence, we examined the variation
of time intervals between additions upon the outcome of the reaction and
discovered that timing of the aldehyde addition is most crucial. The addition
of 3-furaldehyde 20-30 minutes after the Michael addition components have
been added led to a 72% yield of a mixture of compounds which exhibited
>95:5 NMR ratio of the trans to cis C10-C1 stereochemistry, and an aldol ratio
(C5-C6) of ~4-6:1 (CH3, 0.82 ppm), these products were also accompanied by ca.

26

1% of what appeared to be a lactone. The initial Michael stereochemistry was
also examined by addition of the thio silyloxy enol ether to 2-
methylcyclopentenone and then quenching of the resulting enolate (H20) to
give two products in a 95:5 ratio as determined by 1H NMR and GC. The
Michael addition and aldol ratios were nearly identical to those reported by
Mukaiyama, however the relatively poor aldol (ca. 6:1) selectivity observed
was the cause of some concern as it appeared that we would have to carry a
complicated mixture some distance into the sequence before we might
consider rectifying the GO stereochemistry. This mixture would add
complexity to the 1H NMR spectrum and complicate our analysis of
subsequent reactions. Regardless, we opted to continue with the mixture as
we were unable to improve the aldol stereochemistry beyond 6:1, and next
examined the crucial cyclization.

 

     

o O a. 5+
oms H ‘ o
1% T
+ j s’\ + \ 3"” 'Sbc's: 72% (13)
O '95°C 0
CH3 0 .
H0
138 140
95 :5 C1/C10
6:1 ce/cs

To prepare for cyclization, the alcohol produced from the tandem Michael
addition 3-furaldehyde aldol condensation was protected to eliminate the
possibility of a lactone being formed, or a retroaldol decomposition. We had
decided to try three protecting groups, with a possibility of fine tuning the
reaction with either size or type of protective reagent. Attempted protection
with both t-butyl diphenylsilyl chloride“0 and pivaloyl chloride41 failed, with
no trace of silylated or acylated products detected (Equation 14). Acetylation
(AcCl or AczO, Et3N)42 did not appear (TLC) to be productive , however, when
this reaction was monitored by GC, we discovered that the desired acetate had
indeed been formed (80%, eq 15).

27

tBqusi-Clk

 

or No Reaction (14)
tBuCOCl

 

      

MO
140 141

With our cyclization precursor 141 in hand, conditions were needed to
cyclize the furyl thioester. It was thought that three different reagents for
cyclization might be effective based on literature precedent, they were
Cu(II)OTf'PhH, Me3+OBF4-, and Hg(CF3COz)2. Our first attempts were done
using Cu(II)OTf-PhH as the acylation catalyst using the work of Kozikowski35
as a model. Kozikowski"3 found that seleno esters could be induced to acylate
furan, pyrrole or thiophene intermolecularly in the presence of Cu(II)-
OTf-PhH ( in 64-100% isolated yields). Upon treatment of compound 141 with
1.2 eq. of Cu(II)OTf-PhH no change in TLC was observed, and after workup,
the starting material was recovered unchanged. Similar results were obtained
with heating or reaction times of up to 24 hours.

CUMOWPhH : No Reaction (15)

 

 

141

Our second effort involved the methylation of sulfur utilizing Meerwein's
reagent (Me30+BF4-). Upon addition of Me3O+BF4- to a dichloromethane
solution of 141, there was an immediate change in the TLC and two spots of

28

lower Rf were detected. After work up, it was found that we had not cyclized
through furan acylation, but instead had formed the diastereomeric lactones
142 and 143 in a 4-6:1 ratio (aldol center) in good chemical yield (BO-90%).

MgO‘BH‘
tau-90%

   

141 142 143

With our options becoming few, we chose to use a potent thiophile to
initiate cyclization. Based on the work of Masamune44, we decided to look at
various thiophilic metal complexes such as Ag(CF3COz)2, AgBF4, and
Hg(CF3COz)2. Our first attempt involved mercuric trifluoroacetate (2 eq.) in
dry acetonitrile (Equation 18). Upon addition of the mercury salt there was a
change in color to a greenish hue, then a white precipitate formed. Thin layer
chromatography showed a new spot had formed that was _n_o_t the same Rf as
either the starting material or the lactones (142,143) observed previously.
After workup, we isolated a 61% yield of a white - yellow solid, mp 158.5-160
°C (uncorrected). The NMR of the product produced a typical proton doublet
pattern (8:6.87, 6.58, both a d, I=1.71Hz) for the two furan protons of a 2,3
disubstituted compound. The 1H NMR also revealed a doublet pattern for
the C10 methyl (1.15ppm, I=6.85 Hz), a singlet for the C5 angular methyl (0.49
ppm) and a singlet for the methine at C6 (6.19 ppm). The 1H NMR and GC
seemed to indicate that we may have obtained a single isomer in which the
minor aldol product seems to have disappeared. One explanation would be
the selective destruction of one isomer possibly by retroaldol.

   

2 eq. Hg(OTFA)2‘
anhyd. CHacN i

66%
/ 0 1lsomer (‘3)

 

 

29

c6

ldidi‘4i

o.—

quJl‘l—lfld

\\_

 

 

9N

14—dl‘dd_‘d4‘l‘4

od 0....

udj—l4q<d—4d44 111
n

d 1

o.m 0.0 o8

dlflillel‘l—diliJ-4141—4JJ4dJlfldJi_‘l111‘Jd11

 

 

 

 

 

 

 

mommDUmmm mMHA ZHQHUHBmdm

ad ad

‘djl‘J—dd‘ld—44lddddill—

"H <m90mmm

30

High resolution mass spectrometry, 13C NMR, and FT-IR all confirm the
presence of the cyclized bicyclo[5.3.0]decane product. Spectroscopic studies of
the products are continuing (COSY45, NOESY45, HOHAHA47 1H NMR) to
determine the exact stereochemistry of the compound formed. Crystals have
been obtained, but as yet the structure has not been elucidated by x-ray

crystallography.

 

 

Me02C OH MeOzc Br 2-Methyl-1 ,3-
1) PBra g cyclopentanedionL
/o\ 2) nPrCOsz I \ KOtBu V
O Ki
1 45

 

Pl'lMegN+ Bra'
1 00%

  

  

   

 

 

   

1 NaBH Quinoline, Cu(O) "BULL TMS-Cl;
2) NaOH, MeOH ° 86% 100%
93%
HO
1 53 COgH
1) 4O °/o CH3003H
NaOAc ;
2) A020. PYridine
SiMe 0 O O
3 5% A00 156 A00 157

SCHEME 13: Schultz's Synthesis of Confertin

With compound 144 in hand, we were in a position to complete a total

synthesis of fastigilin C and / or aromatin, confertin and helenalin. Aromatin
and helenalin appeared as particularly desirable targets since Schultz43 and
his group had completed formal total syntheses of both confertin and

31

aromatin from a common furan intermediate 150 (Scheme 13). More
importantly, Schultz completed his syntheses of aromatin/confertin with a
furan oxidation to obtain the butyrolactone needed in these systems, similar
to a process we had envisioned to complete our synthetic effort.

 

 

 

 

 

 

1)NaOH,MeOH‘
2) quinoline, Cu°
60%
5% H? H?
C. o nBuLi, TMS-Cl: .‘. o 40% 011300311; .’ o
l/ 90% // NaOAc /
0 O SiM63 43% 0 0
162 163 164
if:
H2,Rh '
— C o
57%
o o
165

SCHEME 14: Schultz's Synthesis of Aromatin

Intending to complete a synthesis of fastigilin-C, our first goal was the
protection of C4 carbonyl as the ethylene ketal. Unfortunately,
thermodynamic (ethylene glycol/ benzene/ reflux) as well as kinetic
((TMSOCH2)2, TMSOTf, -78 °C) conditions for ketalization“9 gave no reaction
(Equation 19). Needing to differentiate the C4 and C9 carbonyls, we attempted
a sodium borohydride reduction. Using 1.5 eq. of NaBH4 gave what appears
to be a single isomer of alcohol 166 (97%) in which the "conjugated" C=O has
been reduced to the exclusion of the more hindered C4 C=O (eq. 20).

32

ethylene glycol, pTSA _

 

or No Reaction (1 9)
(T MSOCH2)2. TMSOTt

 

1 .5eq. NaBH4 ~

 

   

OAc
144 155

A comparison of alcohol 166 to a related compound (160) prepared by
Schultz, we found the coupling constant of the doublet for C9 methine proton
(9.69 Hz), was in accord with that found by Schultz (]=9.4 Hz.). Schultz
attributed the coupling constant to a 180° dihedral angle between the two
protons. An inspection of Dreiding models clearly showed two distinct
conformations in which hydride attack could occur. In conformation A, the
seven membered ring was nearly chair-like making the B-face accessible to
hydride attack giving a product epimeric to Schultz's. If, on the other hand,
the ground state conformation looks more boat like as in B, then only the 0:-
face is accessible to hydride attack giving the [3 alcohol. Since our coupling
constant of the C9 hydrogen was so similar to that of Schultz, we tentatively
concluded that our assignment for the position of the alcohol is [5. In
addition, we have performed MM250 calculations on structure 166 and have
arrived at a minimized structure with a C9-C10 dihedral angle of ca. 170°, in
relatively good agreement with that derived from NMR data.

33

   

0A0

Flgure 6: Stereochemistry oi Ketone Reduction

With mono-protected diol 166 in hand, we turned our attention to the
synthesis of aromatin, and helenalin. In order to prepare these compounds,
we examined the possible methods of deoxygenating the C6 and C9 positions
for aromatin, and remove the C9 oxygen to arrive at helenalin. We have
investigated various Barton51 deoxygenation methods (NaH, C52, nBugan)
to remove these oxygens and have met with little success, these include
attempts to reductively remove each alcohol separately (including various
Barton methods and halogenation of the secondary alcohols) and a bis
reduction (2 eq. NaH, 15 eq. C52) from the keto diol derived from 166
(Equation 21).

a) NaH, C82, Mel

b) Buaan, AIBN No Product Recovered (21)

 

(Mo
1 44

Two possible approaches toward aromatin and/ or helenalin are currently
under study. The first requires the selective formation of either the
dithioketal 167 as indicated in Scheme 15, or formation of thiophenyl
compound 168. The thioketal and/ or C9-SPh compounds could be reduced
(Ra-Ni, and Bu3SnH respectively) to afford the deoxy furan 169 (Scheme 15).
This mono-acetoxy furan might then be converted to helenalin in a manner
similar to that employed by Schultz during his synthesis of aromatin.

 

   

/—\ ”1
TMSS STMS __ nBuaan O.
TMSOTf AIBN
—78°
OAc
1 6 9

 

OAc
1 6 8
Scheme 15: Future Plans

 

The second approach requires the synthesis of a cyclization precursor with
a different group protecting the C6 oxygen. Manipulation at this stage is a
requirement as we have found that altering the C6-OH protecting function is
very difficult post cyclization, as the C6 oxygen becomes too hindered for
replacement by any other protecting group. We have completed a pilot study
in which alcohol 140 was protected as the SEM-ether52 (SEM-Cl,
diisopropylamine) in 97% yield (eq 21). When subjected to 2 eq. of
Hg(CF3COz)2, no cyclized product was seen and after 24 hour exposure to
Hg(II), the TLC showed predominantly starting material and/ or deprotected
alcohol. It seems that SEM is not stable to the reaction conditions. We are in
the process of employing either MEM (methyloxyethoxymethyl)53 or MOM
(methoxy-methyl)54 functions as protection for the C6-O during the
cyclization. Should these techniques succeed, then we might have a
compound that can withstand reductive (Barton) or oxidative conditions
anticipated during assaults on helenalin and/ or aromatin.

86% (22)

 

Zeq. Hg(OTFA)2_
anhyd CHacN

 

   

Another possibility was to enter the product manifold without the C6
oxygen, thus avoiding the question of oxygen removal for the preparation of
aromatin. This might be accomplished if we could replace 3-furaldehyde in
the initial Michael-Aldol reaction with 3-halomethyl furan, thus converting
the process to a Michael addition-enolate alkylation. As was previously
mentioned, the initial Michael addition of the thiopropionate enol ether to 2-
methyl cyclopentenone was smoothly accomplished and the cyclopentyl enol
ether was isolated and subjected to Mukaiyama55 reaction conditions for silyl
enol ether alkylation (TiCl4).

Our initial electrophile for reaction with thiosilyloxy enol ether (138) was
3-chloromethyl furan (3-furylmethanol, MsCl, collidine, LiCl) (179).
Unfortunately we observed only the product of a Michael reaction without
any further alkylation. Our assumption was that the chloro compound may
not be sufficiently electrophilic to react with silylenol ether (138). Currently,
we are synthesizing the bromomethylfuran compound to form the protected
bicyclo[5.3.0]decane ring system of aromatin or confertin.

O

OTMS
{r [W m... ,, _
+ + II ’

tBUS \ O 3 ITIOI °/o

 

(24)

 

172

A major portion of our remaining work will focus on the total synthesis of
fastigilin-C, which includes, manipulation of the A ring to generate an
eliminatable group for formation of the enone once the furan has been
oxidized and reduced to the butyrolactone. This question of enone formation
was a key problem in Lansbury's synthetic scheme, however we anticipate
that our intermediates should not suffer these problems as the rigidity

3 6
introduced into the B-ring by the furyl moiety should lessen the interactions
of the axial groups at C5 and C6 which Lansbury felt were the source of his
difficulties.

 

 

    

RSH or ROH;
ethylene glycol

 

SCHEME 16: Completion of Fastigilin-C

EXPERIMENTAL

37

EXPERIMENTAL SECTION

General. Tetrahydrofuran (THF) and benzene were dried by distillation
under argon from sodium benzophenone ketyl; methylene chloride (CH2C12),
triethylamine, methanesulfonyl chloride (mesyl chloride), pyridine, boron
trifluoride etherate (BF3-OEt2), hexamethylphosphorus triamide (HMPA),
chlorotrimethylsilane (TMS-Cl), and diisopropylamine were dried under
argon by distillation from calcium hydride. Formic acid (98%) was purchased
from Fluka and used as received. All lithium reagents were purchased from
Aldrich Chemical used as a known molarity. Petroleum ether refers to 35-
60°C boiling point fraction of petroleum benzin. Diethyl ether was purchased
from Columbia Chemical and used as received. All other reagents were used
as received unless otherwise stated; all reactions were performed under argon
with the rigid exclusion of moisture from all reagents and glassware unless
otherwise mentioned.

Melting points were determined on a Thomas-Hoover capillary melting
point apparatus and are uncorrected. Infrared spectra were recorded on
Perkin-Elmer Model 167 and N icolet IR-44 Fourier Transform spectrometers
with polystyrene as standard.

Proton magnetic resonance spectra (1H or 13C NMR) were recorded on a
Varian T-60 at 60 MHz, a Varian FT-80 at 80 MHz, a Bruker WM-250
spectrometer at 250 MHz, a Bruker WM-300 at 300 MHz, or a Bruker WM-SOO
at 500 MHz as mentioned in dueteriochloroform or deut‘eriobenzene.
Chemical shifts are reported in parts per million (8 scale) from residual
proton resonance. Data are reported as follows: chemical shifts (multiplicity:
s = singlet, bs = broad singlet, d = doublet, t = triplet, q = quartet, h = heptet, m
= multiplet, a = apparent), coupling constant (Hz), integration.

Electron impact (El-MS, 70 or 25ev) mass spectra were recorded on a
Finnigan 4000 with an INCOS 4021 data system. High Resolution Mass
Spectra (HR-MS) were recorded at the Michigan State University Regional
Mass Spectroscopy Facility (NIH), Department of Biochemistry.

38

Gas Chromatograms were obtained on a Hewlett Packard 5770A using a 35
meter capillary column containing methyl silicone. Parameters were as
follows; Injector =250°C, Detector =350°C, Column =100-220°C @ 10°/ min

Flash column chromatography was performed according to the method of
Still55 et. al. using Merck silica gel and eluted with solvents mentioned. The
column outer diameter (OD) is listed in millimeters.

Mukaiyama Michael AdditionZAldol Adduct 140
To a solution of thiosilylenol ether 137 (1.67g, 7.66 mmol) in CH2C12

(10 m1) at -95°C was added 2-methylcyclopentenone (0.797 ml, 8.12 mmol) and
tn'tyl hexachloroantimonate (0.221 g, 0.383 mmol, 5 mol%). After stirring for
20-30 min. at 95°C (time is critical in the addition of the aldehyde, longer
time periods cause destruction of the Michael product and substantially
lowered yields), 3-furaldehyde (0.636 ml, 0.706g, 7.35 mmol) in CH2C12 (2 ml)
is added dropwise over 10 min. The solution was stirred at -95°C for 12 hrs.,
quenched with sat. NaHCO3, extracted with CHzClz (3 x 25 ml), the organic
phases combined, washed with brine (2 x 50 m1), dried over MgSO4, and
evaporated in vacuo. The residual oil obtained was chromatographed (flash
technique, SiOZ, 230-400 mesh, 270g, 50 mm O.D., ether/ hexane, 1:1) and gave
two fractions, 140 (1.88g, 72.6%) and an unwanted lactone product, 142 (0.19g).

1H NMR (250 MHz, C6D5) 5 = 7.79 (m, 1H), 7.06 (m, 1H), 6.43 (m, 1H), 4.73 (d,
I=10.8 Hz, 1H), 4.11 (d, J=10.8 Hz, 1H), 2.71 (m, 1H), 2.66 (m, 1H), 2.35 (m, 1H),
1.38 (s, 3H), 1.30 (s, 9H), 0.82 (d, I=6.4 Hz, 3H).

13C NMR (62.95 MHz, C6D6) 6 = 225.38, 203.80, 142.93, 141.75, 126.58, 109.86,
71.83, 55.45, 49.77, 48.21, 43.57, 42.11, 38.11, 29.51, 23.71, 16.40.

IR (neat) 3496, 2967, 2927, 2877, 1739, 1682, 1476, 1374, 1277, 1159, 1078, 965, 800,
738 cm'l.

EI-MS (70eV) m/e = 338 (M+), 264 (0.27), 248 (1.45), 214 (1.48), 185 (6.05), 146
(11.42), 125 (19.66), 97 (72.20), 96 (32.54), 95 (38.03), 90 (14.98), 83 (32.77), 69
(16.24), 57 (100), 55 (68.93), 41 (64.02).

HR-MS m/ e = 338.1552 calculated 338.1549 found

QC = 6.72 minutes, 6.96 minutes ~4:1 ratio

39

Acetylated Cyclopentanone 141
To a solution of aldol alcohol 140 (1.02g, 3.02 mmol) in CH2C12 (40 ml)

was added triethylamine (2.11 ml, 15.05 mmol), DMAP (0.03g, catalytic) and
acetic anhydride (0.712 ml, 7.54 mmol) dropwise. After stirring at RT for 12
hrs., the solution was quenched carefully with sat. NaHC03 (15 ml dropwise),
extracted with CH2C12 (4 x 15 ml), the organic phases combined, dried over
MgSO4, and evaporated in vacuo. Chromatography (flash technique, SiOz,
140g, 40 mm O.D., ether/ hexane, 1:1) of the resulting orange viscous oil gave
acetylated product 141 (0.923g, 80.3%) and a product that arose from retroaldol
(11.0%).

1H NMR (250 MHz, C6D6) 6 = 7.81 (m, 1H), 7.01 (t, #173 Hz, 1H), 6.55 (m, 1H),
6.09 (s, 1H), 2.76-2.34 (m, 3H), 1.95-1.31 (m, 3H), 1.68 (s, 3H), 1.49 (s, 9H), 1.07 (s,
3H), 0.85 (d, #986 Hz, 3H).

13C NMR (62.95 MHz, C6D6) 5 = 217.55, 203.98, 169.25, 143.04, 142.61, 122.52,
110.50, 70.83, 54.40, 49.43, 48.42, 43.13, 38.57, 31.84, 29.72, 23.25, 22.92, 20.51, 17.27,
15.59.

11; (neat) 2968, 2927, 2903, 1746, 1702, 1676, 1456, 1366, 1234, 1162, 1080, 1023,
967, 957, 752 crn'l.

CI-MS (CH4) m/ e = 380 (M+, 18.5), 338 (16.8), 321 (29.5), 293 (6.6), 265 (100), 249
(17.5), 231 (5), 203 (3.3), 166 (2.4), 139 (2.5), 97 (2.2), 57 (3.1).

HR-MS = 380.1657 calculated, 380.1655 found

g = 13.40 minutes, 13.46 minutes, ~4:1 ratio

Formation of Lactones 142 and 143
To a solution of the acetylated thioester 141 (0.025g, 0.065 mmol) in

CHzClz (4 ml) was added trimethyl oxonium tetrafluoroborate (0.011g, 0.0723
mmol). After stirring at RT (4 hrs.), the solution was quenched with sat.
N aHCO3 (2 ml), extracted with CH2C12 (3 x 5 ml), dried over Na2304, and
concentrated in vacuo.. Chromatography of the residue (flash technique,
SiOz, 230-400 mesh, 3g, 10 mm O.D., ether/ hexane, 1:1) gave two fractions of
142 (0.014g, 93%) and 143 (0.001 g, 6%) which were presumed to be lactones.

4 0

142: 1H NMR (250 MHz, c6136) 6 = 7.39 (m, 1H), 7.05 (t, J=1.66 Hz, 1H), 6.46 (m,
1H), 4.40 (s, 1H), 2.03 (dq, J=8.6 Hz, 5.63 Hz, 1H), 1.86-1.52 (m, 2H), 1331.11 (m,
2H), 0.97 (d, I=8.6 Hz, 3H), 0.70 (m, 1H), 1.60 (s, 3H).

143: 1H NMR (250 MHz, C6D6) 6 = 7.26 (m, 1H), 6.93 (t, I=1.77 Hz, 1H), 6.22 (m,
1H), 4.10 (s, 1H), 2.01 (m, 1H), 1.73 (m, 1H), 1.4-0.96 (m, 4H), 0.96 (d, I=7.5 Hz,
3H), 0.83 (s, 3H).

Hexahydro-6g-acetoxy-5310g,-Dimethyl-4,8-Dioxoazuleno-[7,8]-fura_n_144

To a solution of acetylated thioester 141 (8.66g, 22.79 mmol) in
anhydrous acetonitrite (90 ml) was added mercuric trifluoroacetate (19.44g,
45.58 mmol) in several portions. The solution changed from a water white to
green, then finally forming a white precipitate. The solution was quenched
with sat. NaHCO3, extracted with CH2C12, the organic phases combined,
washed with brine (3 x 20 ml), dried over MgSO4, and concentrated in vacuo.
The solid recovered was chromatographed (flash technique, SiOz, 230-400
mesh, 400g, 60 mm O.D., ether/hexane, 3.5:1) and gave 4.37g (66.1%) of
cyclized furan 144 as a white crystalline solid. mp = 158.5-160°C uncorrected.

1H NMR (500 MHz, C6D6) 8 = 6.87 (d, ]=1.71 Hz, 1H), 6.58 (d, I=1.71 Hz, 1H),
6.19 (s, 1H), 2.47 (dt, 1:11.58, 6.33 Hz, 1H), 2.07 (dq, #924, 6.85 Hz, 1H), 1.94 (dd,
1:16.43, 8.01 Hz, 1H), 1.69 (ddd, 1:14.21, 12.34, 8.13 Hz, 1H), 1.60-1.40 (m, 1H),
1.50 (s, 3H), 1.15 ((1, ]=6.85 Hz, 3H), 0.75 (ddd, 1:16.45, 9.24, 8.13 Hz, 1H), 0.49 (s,
3H).

13C NMR (62.95 MHz, C6D6) 8 = 217.06, 189.25, 168.96, 149.15, 146.70, 127.32,
115.29, 70.34, 53.84, 45.65, 41.01, 36.91, 24.80, 21.19, 14.80, 14.46.

B (neat) 2990, 2943, 2877, 1750, 1662, 1488, 1415, 1370, 1225, 1058, 1015, 976, 892
cm'l.

EL-Mfi (70eV) m/ e = 248 (6.63), 247 (19.78), 215 (4.33), 165 (4.05), 124 (5.89), 97
(10.58), 95 (5.03), 91 (8.04), 77 (8.97), 55 (19.66), 43 (100).

HR-MS = 290.1154 calculated, 290.1132 found

G_C = 11.63 minutes, >97:3 one compound

41

Hexahydro-6g-acetoxy—Sfi-hydroxy-Sfl,1OQ-Dimethyl-4-oxoazuleno-[7,8]-furan
1%

To a solution of cyclized furan 144 (0.248g, 0.855 mmol) in methanol (10
ml) was added N aBH4 (0.048g, 1.28 mmol) in one portion. The reaction (12
hrs.) was quenched with sat. NaHCO3, extracted with CHzClz (3 x 15ml),
washed with brine (2 x 10 ml), the organic fractions combined, dried over
MgSO4, and evaporated in vacuo. The resulting solid was chromatographed
(flash technique, SiOz, 230-400 mesh, 25g, 2.0 mm O.D., ether/ hexane, 5:1) and
gave 0.241 g (96.7%) of a single alcohol product 166.

1H NMR (250 MHz, C6D6) 8 = 6.87 (m, 1H), 6.65 (d, J=1.79 Hz, 1H), 6.09 (s, 1H),
4.19 (d, I=9.69 Hz, 1H), 2.23 (dq, I=8.75, 6.14 Hz, 1H), 1.98 (dd, 1:18.27, 9.00 Hz,
1H), 1.80-1.43 (m, 5H), 1.51 (s, 3H), 1.06 (d, I=6.48, 3H), 0.57 (s, 3H).

13C NMR (62.95 MHz, C6D6) 8 = 217.84, 169.15, 153.17, 140.21, 115.91, 115.36,
73.98, 70.45, 53.63, 42.33, 38.17, 36.16, 24.01, 20.36, 16.10, 15.58.

_I_R_ (neat) 3224, 2972, 2924, 1743, 1701, 1464, 1458. 1373, 1236, 1124, 1080, 1016,
1003, 983, 937, 893, 748, 611 cm'l.

ISL—MS (25eV) m/ e = 232 (35.63), 175 (10.92), 161 (19.20), 159 (13.05), 147 (23.34),
126 (17.94), 124 (14.05), 97 (16.69), 69 (18.19), 44 (17.57), 43 (100).

g; = 11.08 minutes, 11.47 minutes, 96:4 ratio

HexaLhydro-6Q-hydroxy-5310g-Dimetlgl-4,8-Dioxoazuleno-[7.8]-furan

To a solution of cyclized furan 144 (0.200g, 0.689 mmol) in
methanol/ water (15 ml/l ml) was added K2CO3 (0.100g, 0.724 mmol). After
stirring for 5 hrs., the reaction was quenched (15 ml H20), extracted with
CH2C12 (4 x 10 ml), the organic phases combined, dried over MgSO4, and
concentrated in vacuo. The solid obtained was chromatographed (flash
technique, SiOZ, 230-400 mesh, 20g, 20 mm O.D., ether/ hexane, 5:1) and gave
0.170g (99.4%) of a white solid 166A.

1H NMR (250 MHz, C6D6) 8 = 7.11 (d, I=1.70 Hz, 1H), 6.31 (d, I=1.7O Hz, 1H),
4.84 (bs, 1H), 3.80 (m, 1H), 2.76 (dt, 1:11.86, 6.32 Hz, 1H), 2.26-1.84 (m, 2H), 1.57
(dd, I=4.05, 2.13 Hz, 1H), 1.19 (d, I=6.91 Hz, 3H), 1.89 (adq, I=9.19, 3.30 Hz, 1H),
0.59 (8, 3H).

42

13C NMR (62.95 NIHz, C6D6) 8 = 220.42, 190.70, 157.28, 146.96, 131.99, 114.71,
69.39, 55.29, 45.97, 39.57, 38.10, 25.10, 15.17, 14.90.

_IR_ (neat) 3445, 2974, 2936, 2880, 1741, 1701, 1653, 1487, 1417, 1278, 1263, 1122,
1060, 974, 794 crn'l.

EI-MS (25eV) m/ e = 163 (0.22), 135 (1.36), 127 (1.39), 124 (1.80), 115 (2.02), 100
(5.46), 97 (12.94), 96 (17.61), 86 (19.06), 84 (33.36), 67 (25.00), 60 (17.96), 58 (70.16),
51 (30.61), 49 (100).

LIST OF REFERENCES

2)
3)
4)

5)
6)

7)

LIST OF REFERENCES

a) Tanis, S. P.; Herrinton, P. M. I. Org. Chem. 1983, 48, 4572.

b) Tanis, S. P.; Herrinton, P. M. Ibid. 1985, 50, 3988.

c) Tanis, S. P.; Chaung, Y. -H.; Head, D. B. Tetrahedron Lett. 1985, 26,
6147.

d) Tanis, S. P.; Herrinton, P. M.; Dixon, L. A. Tetrahedron Lett. 1985,
26, 5347.

e) Tanis, S. P.; Dixon, L. A. Tetrahedron Lett. 1987, 28, 2495.

See Reference 1.
See Reference 1 b.

a) Yamamoto, Y.; Maruyama, K. I. Am. Chem. Soc. 1977, 99, 8086.

b) Yamamoto, Y. Angew. Chem. Int. Ed. Engl. 1986, 25, 947.

Tanis, S. P.; Chaung, Y. -H.; Head, D. B. J. Org. Chem. 1988, 53, 4929.

Gao, Y.; Hanson, R. M.; Klunder, I. M.;Ko, S. Y.; Masamune, H.;

Sharpless, K. B. I. Am. Chem. Soc. 1987,109, 5765.

a) McMills, M. C. M.S. Michigan State University 1987.

b) Tanis, S. P.; Johnson, G. M.; McMills, M. C. Tetrahedron Lett.
1988, 29, 4521.

a) Romo, 1.; Romo de Vivas, A. "Forschritte der Chemie Organisher
N aturstoffe", Springer Verlag, New York, 1963.

b) Yoshioka, H.; Mabry, T.].; Timmerman, B.N. "Sesquiterpene
Lactones", University of Tokyo Press, Tokyo, 1973.

c) Heathcock, C.H.; Graham, S.L.; Pirrung, M.C.; Plavac, F.; White, CT.
"The Total Synthesis of Natural Products", Apsimon, ].W., Ed.,
Wiley:New York, 1982, Vol. 5.

a) Fischer, N .H.; Oliver, E.].; Fischer, H.D. "Forschritte der Chemie
Organosher N aturstoffe", Springer Verlag, New York, 1979.

b) Heywood, H.; Harbone, I.B.; Turner, B.L. "The Biology and
Chemistry of the Compositae, Vols. 1 and 2", Academic Press,
London, 1977.

43

10.

11.
12.

13.
14.

15.

16.

17.

18.

19.

20.

20.
21.

44

c) Rucker, G. Angew. Chem. Ant. Ed. Engl. 1973,12, 793.

a) Rodriguez, E.; Towers, G.H.N.; Mitchell, ].C. Phytochemistry 1976,
15, 1573.

b) Iiono, Y.; Tanako, A.; Yamashita, K. Agric. Biol. Chem. 1972, 36,
2505.

c) Gross, D. Phytochemistry 1975,14, 2105.

d) Cassady, I.M.; B yrn, S.R.; Stamos, I.K.; Evans, S.M.; McKenzie, A. I.
Med. Chem. 1978, 21, 815.

e) Fujita, E.; Nagus, Y. Bioorg. Chem. 1977, 6, 287.

f) Schlewer, G.; Stampf, ].L.; Beneyra, C. I. Med. Chem. 1980, 24, 1031.

Spring, 0.; Kupka, 1.; Maier, B.; Hager, A. Z. Naturforsch 1982, C37, 1087.

a) Barbier, P.; Benezra, C. Naturwisseuschaften 1982, 69, 296.

b) Benezra, C.; Dupuis, G. Recherche 1983, 14, 1062.

c) Dupuis, G.; Benezra, C. "Allergic Contact Dermatitis to Simple
Chemicals: A Molecular Approach", Marcel Dekker, New York,
1982.

d) Stampf, I.L.; Benezra, C.; Klecak, G.; Geleick, H.; Schulz, W.; Hausen,
B. Contact Dermatitis 1982, 8, 16.

e) Papageorgiou, C.; Benezra, C. Tetrahedron Lett. 1984,25, 1303.

Dupuis, G.; Mitchell, ].C.; Towers, G.H.N. Can. I. Biochem. 1974, 52, 575.

a) Spring, 0.; Hager, A. Planta 1982, 156, 433.

b) Spring, 0.; Albert, K.; Hager, A. Phytochemistry 1982, 21, 2551.

Hall, I.H.; Starnes, Ir., C.O.; Lee, K.H.; Waddell, T.G. I. Pharm. Sci. 1980,

69, 537.

Stephenson, R.P. Br. I. Pharmacol. 1956,11, 379.

a) Whitehouse, M.W.; Dean, P.D.G. Biochem. Pharmacol. 1965,14,
557.

b) Whitehouse, M.W.; Haslam, ].M. Nature 1962,196, 1323.

Hall, I.H.; Lee, K.H.; Stames, C.O.; ElGebaly, S.A.; Ibuka, T.; Liou, Y.F.;

Haruna, M. I. Pharm. Sci. 1978, 67, 1235.

Hall, I.H.; Williams, Ir., W.L.; Chaney, S.G.; Gilbert, C.].; Holbrook, D.].;

Muraoka, O.; Kiyokowa, H.; Lee, K.H. I. Pharm. Sci. 1985, 74, 250.

Hall, I.H.; Grippo, A.A.; Lee, K.-H.; Chaney, S.G.; Holbrook, DJ. Pharm.

Res. 1987, 4, 509.

Quallich, G.I.; Schlessinger, R.H. I. Am. Chem. Soc. 1979,101, 7627.

Marshall, ].A.; Ellison, R.H. I. Am. Chem. Soc. 1976, 98, 4312.

24.

26.

27.

28.

29.
30.

31.

32.

33.

45

Heathcock, C.H.; DelMar, E.G.; Graham, S.L. I. Am. Chem. Soc. 1982,

104, 1907.

Wender, P.A.; Eissenstat, M.A.; Filosa, M.P. I. Am. Chem. Soc. 1979, 101,

2196.

Semmelhack, M.F.; Yamashita, A.; Tomesch, 1.C.; Hirotsu, K. I. Am.

Chem. Soc. 1978, 100, 5565.

Ziegler, F.E.; Fong, 1.-M.; Tam, CC. I. Am. Chem. Soc. 1982,104, 7174.

a) Lansbury, P.T.; Hangauer, 1r., D.G.; Vacca, 1.P. I. Am. Chem. Soc.
1980, 102, 3964.

b) Lansbury, P.T.; Vacca, 1.P. Tetrahedron 1982, 38, 2797.

Lansbury, P.T.; Serelis, A.K.; Hengeveld, 1.E.; Hangauer, 1r., D.G.

Tetrahedron 1979, 36, 2701.

a) Grieco, P.A.; Ohfune, Y.; Majetich, G.F.; Wang, C.-L.1. I. Am. Chem.
Soc. 1982,104, 4233.

b) Ohfune, Y.; Grieco, P.A.; Wang, C.-L.1.; Majetich, G. I. Am. Chem.
Soc. 1978, 100, 5946.

Quallich, G.1.; Schlessinger, R.H. I. Am. Chem. Soc. 1979, 101, 7626.

a) Lansbury, P.T.; Nickson, T.E.; Vacca, 1.P.; Sindelar, R.D.; Messinger
II, 1.M. Tetrahedron 1987,43, 5583.

b) Lansbury, P.T.; Vacca, 1.P. Tetrahedron Lett. 1982, 263, 2623.

c) Lansbury, P.T.; Serelis, A.K.; Hengeveld,1.E.; Hangauer, 1r., D.G.
Tetrahedron 1980, 36, 2701.

a) Marino, 1.P.; Farina, 1.8. Tetrahedron Lett. 1975, 3901.

b) Marino,1.P.; Farina, 1.5. I. Org. Chem. 1976, 41, 3213.

c) Marino, 1.P.; Floyd, D.M. I. Am. Chem. Soc. 1974, 96, 7138.

d) Grieco, P.A.; Wang, C.-L.1.; Majetich, G. I. Org. Chem. 1976, 41, 726.

a) Ireland, R.E.; Mueller, R.H.; Willard, A.K. I. Am. Chem. Soc. 1978,
98, 2808.

b) Ireland, R.E.; Wilcox, C.S. Tetrahedron Lett. 1977, 2839.

c) Ireland, R.E.; Willard, A.K. Tetrahedron Lett. 1975, 3975.

d) Ireland, RE; Mueller, R.H. I. Am. Chem. Soc. 1972, 94, 5897.

a) Burke, S.D.; Fobare, W.F.; Pacofsky, G1 I. Org. Chem. 1983, 48,
5221.

b) Burke, S.D.; Pacofsky, G.1. Tetrahedron Lett. 1986, 27, 1986.

Smith, A. B.; Branca, S. 1.; Guaciaro, M. A.; Wovkulich, P. M.; Korn, A

Org. Syn. 1983, 61, 65.

35.

37.

38.

39.

41.

42.
43.

a)

b)
a)

b)
c)

d)

a)

b)

c)

d)

a)
b)

a)
b)
a)

b)
a)

b)

46

Raucher, S. 1.; Lui. A. S-T.; MacDonald, I. E. I. Org. Chem. 1979, 44,
1885.

Nemoto, H.; Shitara, E.; Fukumoto, K. Heterocycles 1985, 23, 549.
Mukaiyama, T.; Tamura, M.; Kobayashi, S. Chemistry Lett. 1986,
1017.

Kobayashi, S.; Mukaiyama, T. Chemistry Lett. 1986, 1805.
Mukaiyama, T.; Tamura, M.; Kobayashi, S. Chemistry Lett. 1986,
1817.

Mukaiyama, T.; Tamura, M.; Kobayashi, S. Chemistry Lett. 1987,
743.

Gennari, C.; Baretta, M.G.; Bernardi, A.; Moro, G; Scolastico, C.;
Todeschini, R Tetrahedron 1986, 42, 893.

Gennari, C.; Bernardi, A.; Cardani, S.; Scolastico, C. Tetrahedron
1984, 40, 4059.

Gennari, C.; Bernardi, A.; Cardani, S.; Scolastico, C. Tetrahedron
Lett. 1985, 26, 797.

Evans, D.A.; McGee, L.R. Tetrahedron Lett. 1980, 21, 3975.
Simchen, G.; West W. Synthesis 1977, 247.

Emde, H.; Domsch, D.; Feger, H.; Frick, U.; Gotz, A.; Hergott, H.H.;
Gofmann, K.; Kober, W.; Krageloh, K.; Oesterle, T.; Steppan, W.;
West, W.; Simchen, G. Synthesis 1982, 1.

Hanessian, S.; Lavallee, P. Can. 1. Chem. 1975, 53, 2975.
Hanessian, S.; Lavallee, P. Can. I. Chem. 1977, 55, 562.

Robins, M.1.; Hawrelak, S.O.; Kanai, T.; Siefert,1.-M.; Mengel, R I.
Org. Chem. 1979, 44, 1317.

van Boeckel, C.A.A.; van Boom, 1.H. Tetrahedron Lett. 1979, 3561.
Hofle, G.; Steglich, W.; Vorbruggen, H. Angew. Chem. Int. Ed. Engl.
1978, 17, 569.

Kisfabidy, L.; Mohacsi, T.; Low, M.; Drepler, F. I. Org. Chem. 1979,
44, 654.

Kozikowski, A.P.; Ames, A. I. Am. Chem. Soc. 1980, 102, 860.

a)

b)

Masamune, S.; Hayase, Y.; Schilling, W.; Chan, W.K.; Bates, G5. I.
Am. Chem. Soc. 1977, 99, 6756.

Masamune, 5.; Kanata, S.; Schilling, W. I. Am. Chem. Soc. 1975,
97, 3515.

45.

47.

48.

49.

50.

51.
52.
53.

55.

4 7

c) Masamune, S.; Yamamoto, H.; Kamata, S.; Fukuyawa, A. I. Am.
Chem. Soc. 1975, 97, 3513.

d) Meinwald,1.; Huang, 1. I. Am. Chem. Soc. 1981, 103, 861.

1enner, 1.; Meier, B.H.; Backman, P.; Ernst, RR. I. Am. Chem. Soc. 1979,

101, 6441.

Haasnoot, C.A.G.; van de Ven, F.1.M.; Hilbers, C.W. I. Mag. Resonance

1984, 56, 347.

Inept Like Spectra, T. Skayhill, The Upjohn Co., Personal

Communication

Schultz, A.G.; Motyka, L.A.; Plummer, M. I. Am. Chem. Soc. 1986, 108,

1056.

a) ku, 1.R; Werzel 1.M. I. Org. Chem. 1985,50, 3948.

b) Tsunoda, T.; Suzuki, M.; Noyori, R. Tetrahedron Lett. 1980, 21,
1357.

PCMODEL MacIntosh Version 1.1 supplied by Sereena Software,

Bloomington, Indiana, 47041.

a) Barton, D. H. R; McCombie, S. W. I. Chem. Soc., Perkin Trans. 1
1975, 1574.

b) Barton, D. H. R.; Motherwell, W. B. Pure Appl. Chem. 1981, 53, 15.

Lipschutz, B. H.; Pegram, 1. 1. Tetrahedron Lett. 1980, 21, 3343.

Corey, E. 1.; Gras, 1. L.; Ulrich, P. Tetrahedron Lett. 1976, 809.

a) Kluge, A. F.; Untch, K. G.; Fried, 1. H. I. Am. Chem. Soc. 1972, 94,
7827.

b) Stork, G.; Takahashi, T. I. Am. Chem. Soc. 1977, 99, 1275.

a) Mukaiyama, T. Angew. Chem. Int. Ed. Engl. 1977, 16, 817.

b) Mukaiyama, T. Pure Appl. Chem. 1986, 58, 505.

Still, W. C.; Kahn, M.; Mitra, A. I. Org. Chem. 1978, 43, 2923.

APPROACHES TO RESINIFERONOL

APPROACHES TO RESINIFERONOL

 

Phorbol Myristate Acetate 1 Resiniteronol 2 lngenol 3

Figure 1: Tigliane, Daphnane, and Ingenane Natural Products

Tigliane,1 daphnane,2 and ingenane3 diterpenes such as phorbol 1, resin-
iferonol 2, and ingenol 3 are examples of compounds known as tumor
promoters.4 Administration of a subthreshold dosage of a potent carcinogen
such as 7,12-dimethylbenz[a]anthracene followed by application of the tumor
promoter results in numerous papillomas and squamous carcinomas when
applied to the skin of mice.5 Phorbol itself is not a carcinogen, but seems to
amplify the effect of a carcinogen, thus increasing the risk of cancer
development. Recently, a similar two step carcinogenic mechanism has been
delineated to account for human carcinogenesis.6

Phorbol has been shown to exhibit activity at concentrations comparable to
that of mammalian hormones, and therefore has been suggested to mimic
one of these hormones in enzyme binding and activation. Protein kinase C
has been implicated as a major receptor for phorbol, and is thought to be
responsible for hormonal signal transduction in cells.7

The conformations of some tumor promoters such as phorbol,
resiniferonol, ingenol, and teleocidin have been correlated by Wender8 in
terms of heterocyclic electron density to attempt to explain the similar
responses from such dissimilar compounds. Oxygen functionality at C-4, C-9,

48

49

and C-20 of phorbol have been related to the indole nitrogen N-l,
benzylamine nitrogen N -13, and hydroxyl at C-24 of teleocidin. The lipophilic
side chain of phorbol myristate acetate can also be correlated to the cyclohexyl
carbocycle portion of teleocidin. A similar argument has been made for the
activity of ingenol. The C-4 hydroxyl, C-20 hydroxyl, and C-9 ketone compare
favorably to the areas of high electron density in both PMA and teleocidin. A
possible note of support comes from the fact that 4a phorbol, which exhibits a
vastly modified conformation in the region occupied by the A-ring, is
completely inactive as a tumor promoter. A comparison of the structures of
both phorbol epimers reveals very little difference in structure except for the
region about the hydroxyl group. These compounds have further been
modeled with respect to diacyl glycerols, the endogenous substrates for
protein kinase C, which these promoters presumably mimic.9 A natural
extension for these hypotheses would be the synthesis of analogs that
compare favorably in conformation and relative heteroatom placement and
electron density to PMA and other promoters. Wender10 has designed
several new compounds that appear to have structural features necessary for
binding and activation. The basic requirements are as follows: 1) a rigid
structure to hold the pharmacophore in the correct spatial arrangement, and
2) place heteroatoms at positions similar to those of PMA, DAG and
teleocidin. The rigid template was available from an aromatic nucleus, and
the use of nitrogen or oxygen with the appropriate spacer atoms should form
a compound that might show similar activity to the natural substrates.
Toward that end, Wender has synthesized a trisubstituted benzene (Figure 3)
with heteroatoms in good agreement with their model. The decyl group on
nitrogen serves as the mimic of the the lipophilic ester subunit contained in
PMA and teleocidin. The substrates were tested using a protein kinase C
binding assay and were found to bind to PKC receptors. Although binding
affinity was less than that of PMA, it was shown to be within an order of
magnitude of the endogenous diacyl glycerol activators.

50

HO

 

AcO OAc
Teleocidin 4 Diacylglycerol 5 Phorbol 6

 

Flgure 2: PKC Promoters

OH
0 °
H1909\ NACHa
H

I
CH3
Figure 3: Synthetic Pharmacophore

Wender's11 group has embarked on a program directed toward the total
synthesis of phorbol, resiniferonol, and ingenol from a bicyclic BC ring
structure. Wender assumed that tiglianes and daphnanes arose directly from
a biological cascade of intermediates, whereas the ingenanes might be realized
from a rearrangement of the C ring which might provide a biosynthetic
approach to the ingenanes. Reducing the problem to that of the synthesis of a
bicyclic B,C-ring is a worthwhile synthetic problem, however it is an
oversimplification because of the flexibility of the bicyclo[5.4.0]undecane ring.
A relative lack of control of the of the conformations available to the
bicyclo[5.3.0]undecane nucleus serve to magnify the difficulty of this problem
as it renders the usually reliable cyclic control of peripheral stereochemistry
useless. Wender chose to relieve this difficulty by adding an oxygen bridge
that would substantially reduce the degrees of freedom available to 1.
Wender has now nearly completed two different syntheses of phorbol shown
in Scheme 112 and 2. Starting with 2-furylmethanol, the alcohol was
protected as its TBS ether, and the side chain added at the five position for
formation of the C ring after metallation (n-BuLi) and addition of the furyl
lithium to lithium propanoate. An aldol reaction was then used to increase
the chain length, and engender the correct relative stereochemistry at the
secondary methyl bearing center of the future C-ring. The aldol was protected

51

(AcCl), the ketone reduced (N aBH4), and the furan was oxidized to afford the
desired pyrone 10. The stage is now set for the crucial [5+2] cyclization of the
olefinic terminus to the derived pyrilium ion. In the event, 10 was acetylated
and base induced pyrilium formation resulted in the formation of the highly
functionalized rigid tricycle 11 (89%). The A-ring was then appended as
follows; the enone was reduced (H2 / Pd), the product ketone treated with
Ph3P=CH2, allylic oxidation (SeOz) and oxidation of the C-4 alcohol then
furnished the desired exomethylene ketone 12. Two carbons, comprising C3
and C4, were then added via higher order vinyl cuprate addition (13). The A-
ring carbonyl and angular OH are deftly introduced as a TMS-cyanohydrin
from the addition of TMSCN to the ketone at C5. Oxime formation and
oxidation formed the nitrile oxide resulting in a [3+2] cycloaddition with the
available double bond to give 15. Reduction (Ra-Ni) of the N-O bond,
hydrolysis of the imine, and base induced elimination (DBU) of the resulting
hydroxy compound completed the A,B,C-ring containing tricyclic system (16).
The synthesis was completed by functionalization of the C ring and addition
of the dimethylcyclopropane. The Wender group concluded this approach
with an acetonide formation, deprotection, oxidation, phenyl sulfide
addition, acetoxylation, and elimination to form the a-acetoxyenone 17.
Sulfur ylide addition then yielded the completed tetracyclic phorbol model
(18). Rings A,B, and C were formed in 14% overall yield, averaging 90% for
each individual step.

 

O O
16 OR X 17 OR X

a)TBSCl, noozu 5) unmask. ncno c) AcCl d) NaBH, e) MCPBA 1) A020 9) oeu h) H2, Pd h) PhaPCHZ
i) 8002 )) Mno2 k) (vinyl)20uCNLi2 I) TMSCN. an2 m) DIBAL n) HONH2.HCI o) Bleach p) RaNi, H2 q) 8220
r) DBU s) DIBAL 1) R4NF u) 03H50Me v) DIBAL w) PCC x) LDA, PhSSOzPh y) mom), 2) MCPBA, heat
1)ArZSCMez

18 on

Scheme 1: Wender’s First Generation Phorbol Route

Wender13 has published a second route to phorbol using a formal [4+2]
Diels-Alder approach. This synthesis began with a hetero Diels-Alder
reaction giving a dihydropyran that formed the basis of ring B. Treatment of
the Diels-Alder adduct with LDA and 1-bromo-2,4-pentadiene added the
sidechain that formed the bulk of the C ring. Selective oxidation of the
methoxy enolether (MCPBA, MeOH), followed by Swern oxidation of the
resulting epimeric alcohols, gave a monoprotected 1,2-diketone. Exo-
methylenation via aldol condensation with acetaldehyde proved difficult, but
with the addition of lithium bromide, the reaction proceeded smoothly to
give the requisite ketoalcohol. The alcohols were then dehydrated (MsCl,

53

DBU) by elimination to give trieneone 20 required for Diels-Alder cyclization.
Compound 20 was heated in xylene at 145°C to give compound 21 exclusively,
in 52% yield for the four steps. The exo selectivity in this cyclization is
thought to result from steric congestion of the C-4 methoxy group in the endo
transition state. Introduction of the A- and D-rings began with the
transposition of the ketone from C-10 to G4 in a fashion similar to the
previous synthesis (Scheme 1). In this case, instead of a cuprate addition, a
second hetero Diels—Alder was used to form a single ortho lactone which
upon protonation provided ketoester 22. At this juncture, the C ring olefin
was cyclopropanated (PhHgCBrz), and then a reduction] oxidation (DIBAL,
Swern), aldol sequence was used to obtain the A-ring. To finish the sequence
required more A ring manipulation, and finally oxidation (SeOz),
elimination (SOClz), addition to complete the approach as outlined in
Scheme 2.

   
   
  

OCH3
1 9

EtOzC

glhlili
91%

a) MCPBA b)Swem c) UNUMS». CHQCHO d) MsCl. Eth e)DBU 1) 145°C. xylene g)Ph3P-:CH2 h) (c0211),

i)CH2:C(OTBS)OEt an2 DHF k) PhHgC&3,80°C I) TMSCN, 20'2 m) DlBAL n)Swem o) 3621114201300,

p) DIBAL mMeZCuCNle, Mel r) WM, H202 s) 8220 t) 2012. TMSCN u)T120 v)Bu4Nl w)t-Bulj
x)TMS-lm ”$902 z)SOClzi)KOAc.AgOAC 2)TBAF

Scheme 2: Wender's Second Generation Phorbol Synthesis

Our strategy for the synthesis of tigliane and daphnane natural products
differs from that of Wender in that our initial goal was the synthesis of an A,

54

B-ring system with a functional equivalent of a cyclohexenone present to
form the C-ring. As has been previously described, a furan can function as
the operational equivalent of a wide variety of a number of useful organic
residues, including a six-membered ring. This equivalence, and the
proximity of this residue to the A-B ring fusion suggested the application of a
furan terminated cationic cyclization sequence to generate a furan containing
bicyclo[5.3.0]decane nucleus 30 (Figure 4). This analysis allowed us to dissect
the 7-membered B ring at the cyclopentanone 31. We might further simplify
32 by disconnecting the side chain in a "normal" polarity sense affording the
hypothetical cyclopentanone-2-nuc1eophile-3-electrophi1e and a residue com-
prised of an allylic electrophile and furyl nucleophile.

   

\
+ 0
= U- ..
PO(OCH3)2
O + -
CH COZCHa
33 3 4

Figure 5: Second Generation Retrosynthetic Analysis

55

With this analysis accomplished, we examined our "intermediates" in
order to insure the bond formations occur in the forward direction for the
desired sequence. This was readily achieved via the reaction of a
cyclopentanoid carbanion, possessing an incipient electrophilic B-carbon (2-
1ithio-2-cyclopentenone ethylene ketal) with a furan containing allylic halide.
The latter component (1-ethyl-2-bromomethyl-(E)-3-(5-methyl)-3-furyl
acrylate) would then possess the furyl carbanion in latent form as the neutral
furan. The furyl acrylate might then be realized as outlined in the lower
portion of Scheme 3. Our initial effort began with a two component
alkylative sequence that would provide the desired furylacrylate 38. Furan 37
was easily prepared by the reaction of 3-furaldehyde with methyl
phosphonopropionate to give 37 in 98%. Bromination14 with NBS in carbon
tetrachloride (hv) provided 38 in 95% yield. We were in a position to study
the coupling of 38 with 2-lithiocyclopentenone ethylene ketal. In the event,
41 (Li, Scheme 4) prepared by the procedure of Smith15, was treated with
bromide 38 under a wide variety of reaction conditions, to no avail. We were
unable to observe even trace amounts of the desired 31, but instead we
obtained a lone furan containing product 39, the results of an undesired
(likely electrocyclic) cyclization. We also discovered that bromide 38 gives 39
on prolonged storage.

002R
H CH3 _ CH
A _uatL.. 3
/ \ '1' R020 PO(OR)2 98% / \
3% O
3 7
_C02R

NhBS / O\ 'electrocyclic' C.—002R
95%

Unwanted Product
SCHEME 3: 1St Generation Phorbol Precursor

We were then forced to look at possible alternative routes to secure the
requisite cyclization precursor. One attractive alternative involved a one pot,
three component coupling of an anion derived from Smith's protected

56

bromoenone with trimethyl phosphonoacrylate16 to give a Horner-Emmons
like intermediate that could be quenched with 3-furaldehyde to give adduct
43. This type of system would follow from our second generation
retrosynthetic disconnections (Figure 5).

This possibility was realized when 41, prepared by lithium-halogen
exchange of bromoenone 40 was treated with CuI to form a cuprate. The
cuprate was exposed to trimethylphosphonoacrylate at -78°C and the mixture
was quenched with 3-furaldehyde to give 43 (32%). After workup, we
obtained a 6:1 ratio of the cis to trans compounds 43. Hydrolysis of the
resulting ketal yielded an enone in nearly quantitative yield. With the enone
in hand, we then examined the crucial cyclization sequence. The desired
closure was effected by BF3°OEt217 (Scheme 4) in dichloromethane to give the
tricycle in a 64% unoptimized yield.

 

 

 

 

O/_\O Ol—\O P0(0CH3)2 OHC
‘3; 1) n-BuLi, -78°C_ ‘fiuu /‘002CH3__ “POmCHg-z
Q 261,...- O '0 -
_ . C3020113
4 0 41

 

  

3-Furtural 1) H”, H20, 94%
fl :
31.7% 2) BFa'OEtg, 64%

 

k/0 4 3 002m“

SCHEME 4: First Generation Desmethyl Phorbol

Having demonstrated the viability of the reaction, we elected to undertake
the preparation of the cyclization intermediate possessing the furyl-CS-CH3
needed to build the C-ring. We began this endeavor with the preparation of
2-methyl-3-furaldehyde. Synthesis of the disubstituted furan was found to be
more difficult than anticipated despite (scant) literature precedent.18 The
ketone of ethyl levulinate was protected as its dioxolane, which was then
coupled with ethyl formate via the corresponding potassium enolate. The
crude or-formyl ketone was then treated with acid (conc. H2804) to give the
furan in a reproducible but disappointing 29% overall yield.

57

o 0
M08 1) Ethylene Glycol W051” LAH. age/L fiH
0 2) KH, EtOCHO, 61% o 2) MnOz, 79070 0 (1)
45 47

3 00110. H O ,52%
45 ) 2S 4

 

The synthetic route utilized in the preparation of 44 was then applied to
furaldehyde 47. To the anion derived from the protected cyclopentenone 40
(Scheme 5) was added a stoichiometric amount of Cu119 to form the lower
order cuprate (-78°C->-45°C—>-78°C), the use of catalytic amounts of copper led
to drastically lower yields. Trimethylphosphonoacrylate20 (TI-IF, -78°C) was
added to the preformed cuprate via a chilled cannula and the resulting
solution was warmed to -20°C, then to 0°C, followed by quenching with 5-
methyl-3-furaldehyde. This procedure resulted in an increase of our yields of
45 from 35% to a reproducible 50-60%. The only difficulty with this procedure
was that the olefin was obtained with poorer stereoselectivity (3-4:1 vs 6-8:1)
in the Wittig olefination than previously observed. The addition product
was deprotected (5% aq. HCl, 86%) and enone 45 was cyclized with BF3-OEt2
(1.0 eq.) to afford the desired bicyclo[5.3.0]decane product 46 (60-70%).

An analysis of the 1H N MR allowed us to determine that 46 was in fact a
mixture of 2 olefin regioisomers 46 and 47 which were nearly inseparable by
TLC. However, careful flash chromatography provided both compounds, and
we determined that along with the expected product 46, there was a second
compound with the double bond in the C6-C7 position instead of the desired
C7-C8 position. The NMR clearly showed the vinyl proton as a doublet (8:
7.62 ppm,1=1.86 Hz) instead of a singlet (7.58 ppm) for 46. Olefin 47 also
exhibited a doublet of doublets (3.89 ppm, 1=21.3, 1.56 Hz) that was not present
in compound 46. This second product was obtained as a 1:3-8 mixture (47/48)
with 46, as determined by gas chromatographic analysis. Although this
outcome was initially disconcerting we could in principle utilize this
byproduct. We might isomerize the double bond to the C7-C8 position by
rhodium21 or acid22 catalysis, or secondly the double bond could be
hydrogenated and reintroduced in a manner similar to that employed by
Wender.23

   

l—\
o o
120(00ng
+

COzCHa
4 2

BFa'O E12
67% -

 

SCHEME 5: Synthesis of Methyl Phorbol Adduct

With the isolated tricycle 46 in hand, we needed to oxidatively open the
furan to an ene—dione and selectively manipulate a compound containing
three ketone carbonyls. We felt that we could indeed selectively protect a side
chain ketone in preference to the 7-membered ring carbonyl, however the A-
ring ketone added an undesired degree of complexity to the scheme.
Therefore, we elected to eliminate this problem by reducing the various
carbonyls (LAH, 92%). We protected the primary allylic alcohol as a SEM
ether24 (SEM-Cl, 67%, Scheme 6) 49, and the more hindered A-ring secondary
alcohol was then masked as the corresponding TBDMS ether25 (TBS-Cl, 76%)
50. With the potentially problematic ketones masked and the ester reduced,
the furan might be oxidized26 to furnish a dione after ene-dione reduction. In
the event, the furyl compound was treated with 1.2 equivalents of MCPBA
(aq. N aHCO3), to give, after workup, the desired ene-dione 51 in 51% yield.
The relatively unstable enedione, produced above, should afford the more
stable dione upon selective reduction. Initially, we attempted hydrogenation
with Lindlars catalyst (Pd/Pb/CaCO3)27, but no reduction was observed. We
then attempted reduction over Adams catalyst (PtzO)28, but recovered only a
fully reduced tetrahydrofuran as the sole product. I

 

  

TBS on - /
513” MCPBA
76% 51%
TBSO H OSEM TBSO H OSEM
5 0 5 1

SCHEME 6: Furan Manipulation

Ojima29 (Equation 2) has found that hydrosilylation of a,B-unsaturated
carbonyls occured using dimethylphenylsilane and Wilkinsons catalyst
((Ph3P)3RhCl) leading to 1,4 reduction of the olefin with little over reduction.
He has also found that the same conditions will reduce the 0t,B-unsaturated
double bond of an enone in preference to the 8,7—double bond. It has been
proposed that silane attacks rhodium, then forms an 112 complex to the
ketone, silicon migration to oxygen and addition of rhodium to carbon then
occured. Once the a-(siloxyallyl)rhodium hydride was formed, there was an
isomerization to the Y-(siloxyallyl)rhodium hydride accompanied by a hydride
shift to form the 1,4 adduct. On an extremely small scale (Scheme 7), this was
in fact what was seen. Mixing the enedione, Wilkinsons catalyst, and
dimethylphenyl-silane in benzene resulted in conjugate reduction.

Ojima Example

\ o (03P)3RhC1 o +
-"“ PhMezsiHr e“

 

 

91
94% Overall Yield

6 0
Pd/CaCOale
or

   

No Reaction

 

(@3P)3RhCI:
PhMBzSiH

 

 

SCHEME 7: Attempted Ene-Dione Hydrogenation

Concurrent with the ene-dione reduction mentioned above, we examined
the elaboration of a furan-containing bicyclo[5.3.0]decane in which the ene-
oate had not been reduced to an allylic alcohol. We studied the borohydride
reduction of 46 to give 93% (Scheme 8) of a single alcohol 53, which was not
protected. Oxidation of the furan was accomplished using 1.2 eq. of MCPBA
in a two phase mixture with sodium bicarbonate as a buffer. Ene dione 54 was
isolated in 40% yield with a recovery of 58% starting material. Prolonged
exposure to the peracid caused a destruction of both starting material and
product probably due to multiple sites of unsaturation available for oxidation.

ZRCIZ
——>
(PhaP)4Pd

HO H 002CH3
5 5

SCHEME 8: Current Monoreduced Furan Sequence

Keinan30 has reported a variation of Ojima's hydrosilylation that will
reduce an 01,13-unsaturated ketone or aldehyde without reducing an 01,0

61

unsaturated ester, using a soluble palladium catalyst with zinc chloride and
diphenylsilane. Hydride addition at the B-carbon occured from the less
hindered face forming a palladium enolate then quenching with water. Zinc
chloride is reported to act not merely as an acid, but as some sort of "Zn-H" or
"Zn-H-Si" species that transfered hydride to palladium. We are currently
exploring the possible uses of this reation to maintain oxygen functionality in
various oxidation states throughout the molecule.

Keinan Example

 

 

o anSin NOR
phM 2'02 : Ph 100% Yield (3)
Pd(PPh3)4
0 131125in
pn/VKQCHS Z”C'2 : No Reaction (4)
Pd(PPh3)4

The C-ring construction can then be completed after selective protection
and addition of a one carbon aldehyde equivalent and aldol-cyclodehydration.
Compound 58 once obtained and oxygenated at C-5 and C-10 might enable us
to search for any competitive binding with protein kinase C. We expect our
intermediate to be flexible enough for analog synthesis as well as serving as
an advanced intermediate for the preparation of the less complex

resiniferonol.
O i

\
.‘°

1) Ketalization
2) PhaPCHzOCHa

HO 5 6 COzCHg O

O

+ -

1)H .1120 : —_"’ : _ Resiniteronol

2) Base Catalyzed
AldOI

 

 

 

SCHEME 9: Future Synthetic Strategy

EXPERIMENTAL

EXPERIMENTAL

Formation of Diol 46.

To a solution of 46 (0.0683g, 0.262 mmol) in anhydrous ether (5 ml) was
added LAH (0.119g, 0.263 ml, 0.315 mmol, 1.0 w over 5 min. The mixture
was quenched with sat. NaHCO3 (5 ml) and NaOH (1 ml, 2N), extracted with
CH2C12 (3 x 15 ml), EtOAc (1 x 15 ml), the organic phases combined, dried over
MgSO4, and concentrated in vacuo. The residue was chromatographed (silica
gel, 230-400 mesh, 6g, 10 mm O.D., EtzO/ Acetone, 5:1) using the flash
technique, and gave 0.056g (92%) of diol 48.

1H N MR (250 MHz, (CD3)2CO) 5 = 6.13 (m, 1H), 5.81 (m, 1H), 4.38 (dt, ]=7.12,
3.55 Hz, 1H), 4.05 (b5, 2H), 3.96 (bs, 1H), 2.65 (dd, 1:10.65, 2.1 Hz, 1H), 2.52 (dd,
=10.3, 1.02 Hz, 1H), 2.16 (bs, 3H), 2.28-1.41 (m, 7H).

3 (neat) 3445, 2948, 2920, 2862, 1592, 1441, 1238, 1218, 1140, 1133, 1018, 984, 772,
685 cm'l.

EI-MS (70eV) m/ e = 235 (M++1, 10.71), 234 (M"', 70.89), 233 (M+-1, 19.61), 203
(53.32), 175 (27.60), 159 (15.46), 145 (14.03), 131 (15.01), 123 (21.34), 122 (100), 121
(71.34), 109 (28.36), 108 (21.57), 43 (40.35).

Monoprotected SEM Ether 49.

To a solution of diol 48 (0.056g, 0.239 mmol) and diisopropyl-
ethylamine (0.0708 m1, 0.406 mmol) in CHzClz (10 ml) was added SEM-Cl
(silyl ethoxy methyl chloride) (0.0508 ml, 0.287 mmol)) dropwise over 5 min.
The solution was stirred overnight, quenched with sat. NH4C1 (8 ml),
extracted with CH2C12 (3 x 15 ml), the organic phase combined, dried over
MgSO4, and concentrated in vacuo. Chromatography on silica gel (230-400
mesh, 6g, 10 mm O.D., EtzO/Hexane, 1:1) gave SEM protected 49 as a light
yellow oil (0.0581 g, 66.8%).

62

6 3
1H NMR (250 MHZ, C6D6) 5 = 6.39 (m, 1H), 5.82 (m, 1H), 4.68 (m, 2H), 4.11 (m,
3H), 3.62 (m, 3H), 2.70 (m, 2H), 2.25-1.75 (m, 6H), 2.03 (b8, 3H), 0.98 (m, 2H), -0.2
(m, 9H)
IR (neat) 3442, 2993, 2924, 2870, 1595, 1327, 1152, 1048, 1020, 946, 855, 740, 688
cm'l.

Diprotected SEMI TBDMS Ether 50.
To a solution of the SEM-protected compound 49 (0.058g, 0.159 mmol)

in dimethylformamide (5 ml) was added imidazole (0.0238g, 0.350 mmol) and
TBDMS-Cl (t-Butyldimethylsilyl chloride) (0.033g, 0.223 mmol) in one
portion. After stirring overnight, the solution was quenched with sat. NH4C1
(10 ml), extracted with CH2C12 (3 x 20 ml), the organic phases combined, dried
over MgSO4, and concentrated in vacuo. Chromatography (silica gel, 230-400
mesh, 6g, 10 mm O.D., Hexane/ Ether, 15:1) of the resulting oil gave 0.058g
(76.3%) of diprotected compound 50 using the flash technique.

1H NMR (250 MHZ, C6D6) 5 = 6.51 (bs, 1H), 5.73 (bs, 1H), 4.71 (S, 2H), 4.19 (8,
2H), 3.72 (dd, I=11.9, 10.3 Hz, 2H), 3.05 (m, 1H), 2.70 (d, 1=19.5 Hz, 1H), 2.25-1.33
(m, 3H), 2.01 (5, 3H), 0.99 (dd, 1=11.9, 10.3 HZ, 2H), 0.92 (8, 9H), 0.03 (8, 3H), -0.02
(s, 12H).

B (neat) 2957, 2900, 2893, 1590, 1435, 1330, 1267, 1242, 1148, 1052, 991, 820, 698
cm'1.

EI-MS (256V) m/ e = 479 (M++1, 2.04), 478 (M"', 6.49), 420 (3.12), 347 (3.08), 332
(4.69), 331 (14.16), 330 (21.35), 273 (9.33), 199 (21.16), 198 (29.59), 197 (27.27), 183
(11.51), 172 (7.20), 75 (100), 73 (34.71).

Formation of Ene-Dione 51.

To a solution of diprotected furan 50 (0.044g, 0.092 mmol) in CHzClz (5
ml) at 0°C was added MCPBA (0.0186g, 0.092 mmol, 85%) in one portion. The
reaction was warmed to RT and stirred for 2 hrs. The reaction was quenched
by the addition of sat. NaHCO3 (5 ml), extracted with CH2C12 (3 x 10 ml), the
organic layers combined, dried over MgSO4, and concentrated in vacuo.
Chromatography (flash technique, SiOz, 230-400 mesh, 5g, 10 mm O.D.,
ether/ hexane, 1:1) of the resulting oil gave 0.023g (51.1%) of ene-dione 51.

64

1H NMR (250 MHZ, C6D6) 5 = 6.23 (bs, 1H), 5.54 (s, 1H), 4.61 (s, 2H), 3.97 (bs,
2H), 3.67 (m, 2H), 3.41 (m, 1H), 2.34 ((1, 1:20.52 Hz, 1H), 2.81-1.42 (m, 7H), 1.73
(s, 3H), 1.01 (m, 2H), 0.91 (s, 9H), 0.02 (s, 3H), -0.02 (s, 12H).

13C NMR (62.95 MHz, C6D6) 5 = 204.13, 196.40, 155.25, 150.36, 123.35, 121.03,
94.59, 80.06, 75.59, 71.65, 65.54, 51.77, 45.45, 43.63, 32.93, 29.72, 25.99, 19.81, 18.25,
-1.28, -4.62.

B (neat) 2940, 2884, 1700, 1695, 1630, 1445, 1365, 1050, 943, 733 crn‘l.

EI-MS (25eV) m/ e = 476 (M+-18, 3.96), 437 (14.60), 379 (19.06), 328 (11.47), 295
(11.39), 289 (42.33), 261 (9.90), 215 (8.83), 197 (21.20), 196 (23.10), 171 (16.09), 147
(12.79), 103 (15.43), 75 (60.73), 73 (100).

Monoreduced Ester 53.
To a solution of 46 (0.244g, 0.938 mmol) in MeOH (15 ml) was added

N aBH4 (0.053g, 1.408 mmol) in one portion and stirred (RT) for 4 hrs. The
reaction was quenched with sat. aq. NaHCOg (10 ml), extracted with CH2C12 (3
x 25 ml), the organic phases combined, dried over MgSO4, and concentrated in
vacuo. Chromatography (silica gel, 230-400 mesh, 25g, 20 mm O.D.,
EtOAc/ hexane, 1:1) of the resulting oil gave 0.228g (92.6%) of compound 53
using the flash technique.

1H NMR (250 MHz, C6D6) 8 = 7.67 (s, 1H), 5.65 (q, I=1.15 HZ, 1H), 4.04 (dt,
I=8.71, 5.06 HZ, 1H), 3.78-3.5 (m, 3H), 3.51 (8, 3H), 2.90 (In, 1H), 2.12-1.33 (m, 5H),
1.92 (s, 3H).

13C NMR (62.95 MHZ, C6D6) 5 = 168.62, 158.77, 150.84, 132.49, 129.99, 117.58,
107.87, 75.17, 51.63, 44.41, 40.68, 29.98, 29.46, 22.54, 13.08.

Ll: (neat) 3420, 2951, 2905, 2862, 1704, 1699, 1694, 1682, 1435, 1261, 1249, 1132,
1103, 948, 786 Girl.

_E_I_-_M§ (70eV) m/e = 263 (M++1, 2.38), 262 (M+, 12.35), 229 (8.37), 216 (8.37), 203
(12.77), 185 (13.18), 145 (16.51), 115 (20.31), 91 (16.57), 77 (10.33), 59 (12.65), 57
(23.69), 55 (23.28), 43 (100).

minor 1H - trans compound 8 = 7.71 (s, 1H), 5.62 (m, 1H), 3.49 (s, 3H).

6 5

Formflon of Ene-Dione 54.

To a solution of 53 (0.1043g, 0.398 mmol) in CHzClz (8 ml) and sat. aq.
N aHCOg (2 ml) was added MCPBA (0.080g, 0.469 mmol, 85%) in one portion.
The mixture was stirred 2 hr., cast into CH2C12 / Aq. N aHCO3 (20 ml each),
extracted (CHzClz, 3 x 20 ml), the organic phases combined, tested with starch
iodine paper, dried over MgSO4, and concentrated in vacuo. Column
chromatography (silica gel, 230-400 mesh, 10g, 10 mm O.D., EtZO/ hexane, 5:1)
of the resulting oil gave two fractions 53 (0.056g) and 54 (0.044g, 40%).
Fraction 53 corresponded to starting material.

1H NMR (250 MHZ, C6D6) 8 = 7.09 (s, 1H), 5.44 (s, 1H), 3.71 (q, I=3.55 Hz, 1H),
3.63-3.15 (m, 3H), 3.42 (s, 3H), 2.68 (m, 1H), 2.32-1.96 (m, 2H), 1.69 (s, 3H), 1.65-
1.28 (m, 3H).

13C NMR (62.95 MHz, C6D6) 5 = 204.79, 196.89, 167.25, 151.73, 139.59, 136.87,
135.06, 131.48, 75.12, 52.27, 52.01, 46.51, 33.99, 29.60, 24.63, 21.58.

113 (neat) 2920, 2900, 2875, 1705, 1695, 1690, 1620, 1455, 1390, 1194, 1093, 1015,
906, 855, 732 cm‘l.

M (25eV) m/ e = 260 (M+-18, 2.66), 228 (5.79), 219 (5.94), 203 (8.81), 200 (7.15),
149 (10.79), 129 (8.27), 111 (9.62), 97 (13.01), 85 (18.97), 83 (17.52), 71 (26.48), 57
(46.18), 44 (31.88), 43 (100).

LIST OF REFERENCES

LIST OF REFERENCES

a) Evans, F. 1.; Soper, C. I. Lloydia 1978, 41, 193.

b) Blumberg, P. M. Critical Review of Toxicology 1980, 153.

c) Ibid., Idem. 1981, 199.

a) Evans, F. 1.; Schmidt, R I. Phytochemistry 1976, 15, 333.

b) Ibid.; Idem. 1976,15, 1778.

c) Hergenhahn, M.; Adolf, W.; Hecker, E. Tetrahedron Lett. 1975,
1595.

a) Hecker, E. Pure Appl. Chem. 1977, 49, 1423.

b) Paquette, L. A.; Nitz, T. 1.; Ross, R. 1.; Springer, I. P. 1. Am. Chem.
Soc. 1984, 106, 1446.

c) Funk, R. L.; Bolton, G. L. I. Am. Chem. Soc. 1986, 108, 4655.

d) Rigby, I. H.; Moore, T. L.; Rege, S. I. Org. Chem. 1986, 51, 2400.

Hecker, E. "Carcinogenesis - A Comprehensive Survey", Slaga, T.I.;

Sivak, A.; Boutwell, R.I<., Eds., Vol. 2, Raven Press, New York, 1978, pp.

11—48.

a) Van Duuren, B.L. Progr. Exp. Tumor Res. 1968, 11, 31.

b) Boutwell, R.I<. CRC Crit. Rev. Toxicology 1974, 2, 419.

c) Hecker, E. "Methods in Cancer Research", Busch, H., Ed., Vol. 6,
Academic Press, New York, pp. 439-484 (1971).

d) Scribner, ].D.; Suss, R. Int. Rev. Exp. Path. 1978, 18, 137.

a) Hecker, E.; Adolf, W.; Hergenhahn, M.; Schmidt, R.; Sorg, B.
"Cellular Interactions by Environmental Tumor Promoters", Fujiki,
H., et.a1., (Eds), Japan Sci. Soc. Press, Utrecht, pp. 3-36, 1984.

b) Berenblum, I. "Risk Factors and Multiple Cancer", Stoll, 8., Ed.,
Wiley & Sons, New York, 1984.

Nishizuka, Y. Nature 1984, 308, 693-698.

Wender, P.A.; Koehler, K.F.; Sharkey, N .A.; Dell 'Aquila, M.L.;

Blumberg, P.M. Proc. Natl. Acad. Sci. USA 1986, 83, 4214.

66

10.

11.

12.

13.

14.

15.

16.
17.

18.

19.

20.

67

a) Boutwell, R.K.; Robrschneider, L.R. Nature New Biology 1973, 243,
121.

b) Wilson, S.R.; Huffman, I.C. Experientia 1976, 32, 1489.

c) Smythies, ].R.; Benington, F.; Morin, RD. Psychoneuroendocrin-
ology 1975, 1, 123.

Wender, P.A.; Koehler, K.F.; Wilhelm, RS.; Williams, P.D.; Keenan,

RM.; Lee, H.Y. "New Synthetic Methodology and Functionally

Interesting Compounds", Yoshida, Z.I., Ed., Elsevieerew York, 1986, pp.

163-182.

Wender, P.A.; Keenan, R.M.; Lee, H.Y. I. Am. Chem. Soc. 1987, 109, 4390.

See Reference 10.

Zvak, V.; Kovak, 1.; Kriz, M. Collection Czech. Chem. Comm. 1980, 45,

906.

Smith, A.B.; Branca, S.I.; Guaciaro, M.A.; Wovkulich, P.M.; Korn, A.

Org. Syn. 1983, 61, 65.

a) Semmelhack, M.F.; Tomesch, I.C.; Czarny, M.; Boettger, S. I. Org.
Chem. 1978, 42, 1259.

b) Kleschick, W.A.; Heathcock, C.H. I. Org. Chem. 1978, 42, 1256.

Tanis, S. P.; Herrinton, P.M. I. Org. Chem. 1985, 50, 3988.

a) Kotsuki, H.; Mondeu, M.; Ochi, M. Chemistry Lett. 1983, 1007.

b) Jones, RG. 1. Am. Chem. Soc. 1955, 77, 4069.

c) Scott, L.T.; Naples, 1.0. Synthesis 1973, 209.

a) Posner, G.H. Org. React. 1972,19, 1.

b) Posner, G.H. "An Introduction to Synthesis Using Organocopper
Reagents", Wiley, New York, 1980.

a) McIntosh, 1.M.; Sieler, RA. Can. I. Chem. 1978, 56, 226.

b) Available from Fluka Chemicals, #79499.

Trost, B.M.; Kony, RA. I. Org. Chem. 1974, 39, 2475.

Tanis, S.P.; Herrinton, P.M. I. Org. Chem. 1985, 50, 3988.

Wender, P.A.; Keenan, R.M.; Lee, H.Y. I. Am. Chem. Soc. 1987, 109, 4390.

Lipshutz, B.; Pegem, I. I. Tetrahedron Lett. 1980, 21, 3343.

Corey, E.I.; Venkateswarlu, A. I. Am. Chem. Soc. 1972, 94, 6190.

a) Williams, P.D.; LeGoff, E. I. Org. Chem. 1981, 46, 4143.

b) Bingerich, S.B.; Campbell, W.H.; Bricca, C.E.; Jennings, P.W. I. Org.
Chem. 1981,46, 2589.

c) Kuwajima, I.; Urabe, H. Tetrahedron Lett. 1981,22, 5191.

26.
27.

28.

29.

68

(1) Adam, W.; Rodriguez, A. Tetrahedron Lett. 1981, 22, 3503.

e) deGroot, A.; Jansen, B.I.M. I. Org. Chem. 1984, 49, 2034.

f) Takano, Y.; Yasuda, A.; Urabe, H.; Kuwajima, I. Tetrahedron Lett.
1985, 26, 6225.

Lindlar, H.; Dubuis, R. Org. Syn. 1973, V, 880.

Rylander, P. N. "Catalytic Hydrogenation in Organic Synthesis."

Academic Press, New York, 1979.

a) Ojima, 1.; Kogure, T. Organometallics 1982,1, 1390.

b) Ojima, 1.; Kogure, T.; Nagai, Y. Tetrahedron Lett. 1972, 5085.

Keinan, E.; Greenspoon, N. Tetrahedron Lett. 1985, 26, 1353.

APPROACHES TO ELEAOKANINE A

APPROACHES TO ELAEOKANINE A

O COQCHa

CH3 ocopn

\

Eleaokanine A 1 Cocaine 2 \
Histrionicotoxin 3

Flgure 1: Alkaloid Natural Products

As an extension of our work with carbocyclic furan terminated cationic
cyclization, we surmised that the use of a nitrogen containing cyclization
initiator might allow access to a large array of fused-, bridged-, and spirocyclic
alkaloids. After considering either an iminium or N -acyliminium ion as the
initiating function, we elected to examine the utility of the more reactive N-
acyliminium ion1 for the task alluded to above. Some important reasons for
selecting an N-acyliminium ions as the initiator of choice include: 1) ease of
formation from readily available starting materials, 2) formation of a
relatively unreactive amide after cyclization, and 3) generally require less
reactive terminators for successful reactions. An additional consideration
was the extensive literature precedent for the use of this function, notably in
the pioneering work of Speckampz, and then observed in the efforts of
Evans3, Hart4, and Chamberlin5 among others. These workers have applied
N-acyliminium ion cyclizations to the construction of a wide variety of
structurally diverse alkaloids including perhydrohistrionicotoxin6,
gephyrotoxin7, and eleaokanine A.8

69

70

O
F"\C'D ./u\G
/N:=CH2 R IINI=CH2
R R
Iminium Ion A N-Acyliminium Ion B

Flgure 2: Iminium and N-acyliminium Ion Structures

Another attractive feature of the N-acyliminium ion as a cyclization
initiator is the diversity of precursors which will give rise to the same
reactive intermediates. For example, in the early 1900's, Tscherniac9a and
Einhorn9b exposed a—OH—amides to H2504 in benzene to obtain an a-Ph-
amide. This compound resulted from the attack of neutral benzene, as a
latent nucleophile, onto the reactive N-acyliminium ion which resulted from
dehydration (H2804) of the a-OH-amide. Although potentially useful, this
technique suffered from relatively low yields in intermolecular processes.
Since that time there have been many new methods to generate an N-
acyliminium ion“). The current method of choice is facile monoreduction of
a cyclic imide, then protonation and elimination (H20) of the on-
hydroxyamide or the generation of related a-alkoxy-amide and elimination of
the elements of ROH to form the desired N-acyliminium ion.11 This reaction
intermediate was then captured by an internally held nucleophile in a facile
intramolecular process. Speckamp12 has extensively studied the mono-
reduction of cyclic imides (NaBH4) in ethanol to give the corresponding 0:-
alkoxy-lactam. This protocol was adopted by Speckamp to circumvent a
nagging problem, overreduction to the (o-OH-amide. Capture as the a-OR-
lactam ether prevents carbinol-amide opening and subsequent amido-
aldehyde reduction. However the a-OR—lactam thus formed, although very
useful, cannot be employed universally. Chamberlin13 solved the problem of
overreduction with a simple modification, treatment of the imide in
methanol with an excess of N aBH4 at -4°C, which cleanly reduced the imide.
This discovery has opened the door for studies of the cyclization under
solvolytic as well as acidic conditions. Chamberlin14 has reported successful
cyclizations of the oc-OH-lactam after treatment with MsCl and triethylamine
followed by capture of the N-acyliminium ion produced after expulsion of the
excellent leaving group MsO'.

71

A multitude of possible outcomes await the derived N-acyliminium ion
upon electrophilic attack, with the majority being either destructive or
unproductive. In order to maximize the production of the desired cyclization
products we must consider 1) enamide formation, 2) enamide dimerization 3)
regiochemical ambiguities, and 4) questionable stereochemical control.
Enamide formation can be a major pathway in any acid initiated N—
acyliminium ion reaction. The selection of reaction conditions and the
relative rate of deprotonation versus cyclization govern the outcome. One
might consider this path reasonable as the enamide should be the operational
equivalent of the N-acyliminium ion fie; the addition of a proton. Many
workers have discovered that the conditions needed to add H+ to the
enamide are indeed much harsher than utilized with the a-OH or a-OR-
lactam and can often lead to destruction of the starting material or the
product. Should the enamide pathway be the reaction of choice, then we
must concern ourselves with the possibility of irreversible enamide
dimerization (see Figure 3).15

ADO—7U [U

Flgure 3: Eneamide Dimerization

 

Flgure 4 : Regiochemical Outcome of Acyliminium Ion Closure

Regiochemical problems are numerous, for example, in a study by
Speckamp16 of unsubstituted olefin and alkyne terminators which can afford
either 6-endo, or 5-exo products he obtains both 5-exo and 6-endo products
(Figure 4). In most unsubstituted cases, the formation of the six membered

7 2
ring is prevalent, however terminator design is crucial to obtain the product
desired.

Overman17 has shown that substitution of the olefin with sulfur or silicon
directed the cyclization as in Figure 5. Finally stereochemistry must be
addressed. In a case studied by Speckamp13, cyclization of the disubstituted
versus monosubstituted olefin showed a marked preference for isomer 7
when the olefin is substituted (Figure 6).

R1
R1=SKCHala y K /

R2 R1 7
Pk /
GU ‘
3 - R2=~‘-‘J(CH$)3 : KN \

F12
5

FIGURE 5: Overman Exo, Endo Cyclizations

0 o
2
Q0549 OMe R— O OCHa
\ a? ___.HC°°H R2 + Hco
H
Hcoo H R, H

H1 Hi
6 7 a

a‘ .32 m ma

H H 2:1 ~1oo

Et H 100:1 ~100

Flgure 6: Stereochemical Outcome of Acyliminium Cyclization

The Eleaocarpus19 and quinolizidine20 alkaloids are groups of compounds
that exhibit little bioactivity, but have been used as a showcase for specific
methodology under study. As was the case for systems analyzed above , the
relative placement of the sidechain and pendant functionality to the
piperidine ring suggested employing a furan terminated cyclization for the
preparation of the basic ring system bearing in mind that for each target
alkaloid we must cleave a C-C bond to correctly place a C=O or C-R function.

73

Our major concerns for the study of these molecules were: 1) are both the 3-2
and 2-3 furan closures useful in alkaloid syntheses, 2) is furan amenable to
the functionality required in the eleaocarpus and quinolizidine alkaloids, 3)
will it be neccessary to use acidic methodology for closure, and 4) can we
complete the synthesis of one or more of the natural products?

Chamberlin and Speckamp have both used succinimide as the A-ring
precursor in the synthesis of Eleaokanine A or B. Each has chosen a unique
terminator for the cationic process. Chamberlin21 elected (Scheme 1) to
cyclize using a ketene dithioacetal as his terminator. The ketene dithioacetal
was selected because of the ability of sulfur to stabilize the carbocation that is
formed from attack of the N-acyliminium ion. Also important was the ability
of the ketene dithioacetal to function both as an acyl anion equivalent and as
a latent carbonyl. Using a modification of Corey's procedure for protecting
lactones as their cyclic dithioorthoester, Chamberlin formed the ketene
dithioacetal alcohol (MezAlS(CH2)3SAlMe2), heat) from valerolactone in
reasonable yields. Mitsunobu condensation (Ph3P, DEAD) of the dithioacetal
alcohol with succinimide gave an overall 62% yield of 11. Reduction
(N aBH4) and cyclization (MsCl, Et3N, -20°C to RT) of the crude hydroxylactam
gave compound 12 in 71% yield. Reduction of the amide (LAH) and
subsequent alkylation of the thioketal anion (LDA, nPrI) resulted in the
formation of advanced intermediate 13 Deprotection of the dithioketal
(HgClz, H20) gave Eleaokanine A in 48% for the three step process.

Speckamp has used a similar system except for the cyclization terminator.
In his case, Speckamp22 (Scheme 2) employed a protected enone to trap the N-
acyliminium ion formed. Cyclization (HCl, MeOH) across the olefin occured
with the regiochemistry indicated, to give the chloro-ketone 18. Deprotection
occured concurrent with cyclization, but was probably not competitive since
in a separate experiment the enone was not reactive enough to form the
product. Halide elimination (DBN) and reduction (LAH) gave an unstated
yield of eleaokanine B.

74

o s s
S/j °'
Phap 1) NaBH4
N” + H0 / S DEAD' 2) MsCl
62 °/. EbN
9 o 1 o O 1 1 7o %

 

 

s s 0
l
1)LAH H9012, Hzo_ \
gfi 2)LDA 755/ so % ,

0 ‘2 95%,43% . 14
SCHEME 1: Chamberlin's Elaeokanine Synthesis

 

 

 

Mill-3.3341.» 1Lsuocinimide 0&0
I/OY 2)C'°3’8°% / 2) Phap5 51°/ U/Kk
pyridine HO DEAD °
0 3) HCI,MeOH
14 1 5
l \ o

1) (TMSOCH2)2 0 0 .6”

mson _ 0” HCI MeOH DBN
2) NaBH4 ' | 100% ' 100%

100%

1) NaBH;
0

2) DIBAL

 

1 9 2 0
SCHEME 2: Speckamp Elaeokanine Synthesis

A final example was a non N-acyliminium ion formation of eleaokanine
B. Weinreb23 looked at eleaokanine as the product of a hetero Diels-Alder
reaction of acyl imine 28 (Scheme 3). A standard Horner-Emmons reaction
(piperidine, HOAc, benzene) reaction forms the disubstituted
dihydrothiophene 22, which, after oxidation (MCPBA) of both the sulfur and

75

olefin, gave epoxy-sulfone 23 in 92% yield. Epoxide 23 was treated with
periodic acid/chromic acid to give carboxylic acid 24 (84%). The addition of
ammonia to the acid chloride prepared from the corresponding carboxylic
acid, gave amide 25 (58%) which led to thioamide 26 in 62% yield. The amide
acetate 26 was then prepared from 27 with mercuric acetate (1 eq.) in glacial
acetic acid followed by silylation. Pyrolytic cyclization of 27 (370°C) gave a 68%
yield of eleaokanine B, which led to eleaokanine A (62%) after Swern
oxidation.

 

  

 

 

 

 

/ o
0 o

/ HSCHZCHQ MCPBA

E13N ’ 92 % | so,

CH3 K‘OCHW 56% CH3 CH3
0
2 1
COZH CONH2
o
1) EEtOCOCI
A°°‘°"° , I 8'02 2) NHa 58%H 02 62%
34% CH
3
02>CH3$ NaBH OCH 35 >1) H9(0AC)2 82%
090's 2) TMS-Cl 96%
90%
o
N
T1180 0
1) 370 °c 68 °/.
A00 2) P00. 66% A I

I 502 3) DIBAL 91 %'
CH3 4) TFA/ DMSO

2 a 62% 2 9

SCHEME 3: Weinreb Non Acyliminium Cyclization to Elaeokanine

The Tanis research group24 has been interested in the use of furans as
butenolide, cyclopentenone, cyclohexane, and 1,4 dione equivalents in

76

cationic cyclizations. The majority of our efforts have employed standard
"carbenium ion" initiator functions en route to heteroatom deficient
terpenoids. More recently we have initiated a series of studies designed to
construct a variety of alkaloid skeletal types via N-acyliminium ion initiated

furan terminated cyclizations. Tanis, Dixon, and Raggonz5 have examined
the use of this protocol in preparing a variety of linearly fused-, spirocyclic-,
and bridged ring containing systems.

The basic concepts employed are

presented in Figure 7 for the electronically favored 3—92 mode of cyclization.

O
N

R0

°o
N

 

T

 

 

O
N
—- . Spirocyclic
0 /
O
—-> / Linear
o /
—-> O Bridged
/

Flgure7: Generalized 3—)2 Cyclization Modes

Dixon and Tanis:26 have constructed the linearly fused system 39 (Scheme
4) that represents a total synthesis of the quinolizidine alkaloid epi-lupinine.
Using a system similar to Chamberlin and Speckamp, disubstituted furan 31
(2-methylfuran, nBuLi, ethylene oxide) was coupled with glutarimide
utilizing Mitsunobu27 chemistry (Ph3P, DEAD) to give a 55% yield of 32.
Reduction (N aBH4, MeOH, -4°C) gave a-hydroxyamide 33(~100%).
Cyclization (HCOOH, cC6H12) produced, directly, 1,4 dione 34 in 75% yield.

 

77

Differentiation of the two ketones formed upon cyclization was of paramount
importance. Attempted thioketalization under thermodynamic conditions
((HSCH2)2, benzene, reflux) led primarily to reclosed furan. Kinetic
thioketalization28 ((TMSSCH2)2, TMSOTf, -40°C—>-20°C) gave a number of
products as an inseparable mixture. Kinetic ketalization29 ((TMSOCH2)2,
TMSOTf, -78°C ->RT) gave a predominance of one product (ring ketal) 35 as
shown by 1H N MR Baeyer-Villiger oxidation (MCPBA) provided the acetate
36 in 55%. The previously reported work of Hart30 suggested that this might
be an optimized yield. Transketalization?’1 (ethanedithiol, BF3-OEt2) rapidly
gave the thioketal 37, which, after reduction (RaNi) of the thioketal, and
reduction (LAH) of both the acetate and amide gave epilupinine 39 in 74%.

O \ O
OHM—4L7 Ph P NaBHg
DEAD MeOH
83% .400

O
/k 32
0

$0
OHCOOH = SOCH2)2
75% TMSOTf
78%

 

 

03 4
AcO\_ c0\-
' O ' S’>
CF3C03H ‘ " o HSCH,CH,SH_ ' s RaNi
NazHPO4 ' BF3' oazfi EtOH / A
55% 83% 60%
o 3 6 o 3 7
ACO\- (OH
74%
038 39

SCHEME 4: Dixon Synthesis of Epilupinine

78

In the case of the spirocyclic alkaloids, Dixon and Tanis32 have completed a
formal total synthesis of perhydrohistrionicotoxin (Scheme 5), a compound
which was reported to block postsynaptic membrane depolarization. Initially,
the synthesis was attempted using a monosubstituted furan, however, after
cyclization, oxidation with various reagents such as MCPBA, Brz/MeOH, hv
etc. proved useless.33 We had previously observed, during our formal total
synthesis of aphidicolin“, that increasing substitution on furan made it more
susceptible to oxidation (MCPBA) affording an ene-dione, which gave the
dione 42 after hydrogenation (H2/ Pd). This was indeed the case as described
in Scheme 5. Cyclization of 40 (HCOOH, cC6H12) gave 41 (72%) which was
smoothly oxidized (MCPBA) and reduced (H2, Pd) to furnish 42 in 70% yield.
Differentiation of the ene—dione was accomplished using kinetic ketalization
methodology developed by Noyori. In the event, kinetic thioketalization of
42 selectively (11:1) gave the product of side chain thioketalization 43 in 67%
yield. Reduction (RaNi) of the thioketal gave the parent hydrocarbon,
completing a formal total synthesis of perhydrohistrionicotoxin in 26% yield.

Et

/ O
. / 0
WWW 0H ma.

2) Bng 0 E1 72%

N

 

 

 

3 o O 4 O
1) MCPBA V’
aq. NaHCOa _ o (T MSCH2)2 _
2) H2/ 10% Pd ' 0 TMSOTF ’
10% aq. HOAc 67 %
7o % o
4 2
RaNi *
_.__... o
78 % o
4 4

SCHEME 5: Dixon Synthesis of Perhydrhistrionicotoxin

79

The bridged alkaloids, represented by the aza-bicyclo[3.2.1]octane and aza-
bicyclo[4.2.1]nonane ring system of cocaine35 and anatoxin-A36 respectively
were the final systems to be investigated. To date, a formal total synthesis has
not been completed, however, the results of Dixon are summarized in
Scheme 6. The synthesis of the cocaine ring system (Scheme 6) proceeded
from a mono-cyclic (furan ring) precursor which was prepared as follows:
nitromethane was added to acrolein under Michael conditions and the
product 4-nitro-butanal was protected as the related dimethyl acetal 45.
Nitroaldol addition to furfural (nBuNHz) and dehydration gave the vinyl
nitro adduct 47 (69%) which was then reduced (LAH), and acetylated (EtCOCl,
56%, 49). Carbamate 49 was reacted with TFA in CHC13 to afford the
azabicyclo[3.2.1]octane-one-al 50 in 62% yield. In this step we have observed
not only a cyclization to an a-alkoxy carbamate, followed by an
intramolecular furan [2-—>3] terminated N-acyliminium cyclization, but, in
addition, the disubstituted furyl moiety has suffered hydrolysis to the related
keto-aldehyde. We surmise that strain prevented rearomatization and that
CF3COz- capture of the intermediate oxonium ion affords a stable enol-ketol
which was hydrolyzed upon workup.

This study, directed toward the synthesis of the elaeokanine alkaloids,
began with the disubstituted furan, 2-methyl-5-furylethanol. This compound
was prepared by the addition of ethylene oxide to the lithium salt of 2-methyl
furan to give 31 in 45% yield. Mitsunobu coupling (Ph3P, DEAD) of 2-methyl-
5-furylethanol with succinimide gave imide 53 (83%), which was reduced
with sodium borohydride (MeOH, -4°C) to give 54 in quantitative crude yield.
Cyclization of the hydroxy lactam in a two phase mixture of formic acid in
cyclohexane gave a 75% yield of dione 55, a result which stands in contrast to
that found by Dixon and Tanis37 in the non-methylated case in which the
cyclized-l unopened furan compound 56 was obtained in 71% yield. This
hydroxy lactam was also cyclized according to Chamberlin's non acidic
conditions (MsCl, Et3N) to afford desired furan 56 (81%). Exposure of 56 to the
same two phase cyclization conditions (HCOOH, cC6H12) containing water (2
eq.) gave, as expected, the dione 55 (62%) demonstrating the feasibility of this
process in a well defined ring system.

N02
4W; QY: —2»":;:,H I; 0....
47

 

OCH3
NHz NHCOzEt
_LA_§_. / \ / OCH Clco Et / \ / OCH3
86 /o O 65 % O
4 8 OCH3 OCHa
OH
CF3C02H ElOzCN NaBH 5102C
———>
62 °/o 84 °/o
4: 1 01:8 OH
5 1
oSeCN
N02
1) E1020 /
BuaP 88 %
2) ACQO,100°/:> 0A6
3) H202. 88 % 5 2

SCHEME 6: Dixon Synthetic Route Toward Cocaine

O ’0
CE +211 ° ‘ A...
O HO O DEAD MeOH

9 31

83% -4°0
~Quant.

 

 

O
HCOOH O
.-- ———> 6
HO O CC5H12 N 63 /°
\
O 5 5
O 5 4 \ 0
M80 81 °/
EM °
- 20° to RT

0 5 6
SCHEME 7: First Elaeokanine Generation

81

After demonstrating the viability of the cyclization process, our task was to
manipulate the resulting dione so as to eliminate the excess carbon chain.
We considered two possible strategies, first we thought that the addition of a
propyl sidechain onto the existing 1,4 dione, and second that we could
incorporate this three carbon residue from the outset. The former approach
would require either a selective alkylation of the sidechain C=O or a
protection protocol similar to those described earlier. After careful
consideration, we decided that a more efficient synthesis might be realized if
the addition of a sidechain came initially from a trisubstituted furan. We
hoped that closure of the this furan would proceed as expected resulting in a
dione with a propyl sidechain placed in the correct position. This strategy
required us to develop the synthesis of a rather complex trisubstituted
furylethanol (Scheme 8 or 10). We were pleased to find an inexpensive,
commercially available starting material 2-methyl-3-methylfurancarboxylate
(Aldrich). Our first route toward the trisubstituted furan 59 is outlined in
Scheme 8. A two carbon Wittig homologation (Ph3P=CHCH3) to 2-methyl-3-
furfural gave furan (58) with a propenyl side chain. Previous work in our
laboratories”, had suggested that reduction of the propenyl side chain,
without concurrent reduction of the furan to a disubstituted tetrahydrofuran
was possible. Reduction catalysts such as Lindlar's (Pd/Pb/CaC03) and NizB
gave either no reduction or a poor mixture of products; platinum or
palladium generally afforded products of overreduction (60). At this point,
we considered conducting the reaction sequence with the olefin intact.
Toward that end, we reacted the propenyl-methyl furan with 2-3 equivalents
of butyllithium, then quenched with either gaseous ethylene oxide or a 1.0M
solution of ethylene oxide in ether to give 59 in 40-70% yields.

With 59 in hand, we proceeded as described in Scheme 9. Furan 59 was
coupled with succinimide to furnish 61 (79%), which was reduced (NaBI-I4,
MeOH, -4°C) to yield carbinol-amide 62 Cyclization (HCOOH, cC6H12) led to
64 in 71% yield. All attempts at further functionalization have been
thwarted.

82

 

O
OEt OH
1 MnO ,86% -
n we 2>—— (£9
0 82% o 2) @3PCHZCH3 Br
Aldrich 57
$0.25/g
NizB N R of
or Lindlar 0 ea '0"
/ \ \
HO 0
59 Pd/C
orPt/C H

SCHEME 8: Synthesis of Trisubstituted Furan
O \
PhaP
DEAD
79%

O \
NaBH HCOOH
Me H CC6H12
-400
Quant. o 62 715%
° 63
SCHEME 9: Second Generation Elaeokanine Synthesis

 

83

In the hope of circumventing these problems we chose modify the furan
and incorporate the desired propyl sidechain instead of the propenyl
compound. Methyl-Z-methyl-S-furancarboxylate was reduced (LAH) to give
2-methyl-3-furanmethanol in 89% yield, which was brominated, according to
Padwa“0 (Scheme 10) to give unstable methyl bromomethylfuran 65 as a
crude product. After numerous attempts, we were unable to purify 65,
therefore, 65 was stored as a solution shielded from light at - 20°C. Alkylation
of 65 with a large excess of ethyl magnesium bromide gave a reasonable 57%
overall yield of 3-propyl-2-methylfuran. Furan 66 was then lithiated (nBuLi)
and ethylene oxide was added to the resulting anion to give 67 (39%)
contaminated with n-hexanol.

[£01133 (“QB' EtMgBr Cf/ 1)neuu
~100% 57%
o cmde 2) <10 39%
57
W66
H O

6 7
Scheme 10: Synthesis 01 Reduced Trisubstituted Furan

Purification of 65 followed by Mitsunobu coupling (Ph3P, DEAD) of 67 with
succinimide gave 68 (85%), which was reduced (NABH4, MeOH, - 4°C), then
exposed to HCOOH, cC6H12 to yield a mixture (49:52, 99%) of furan 71 and
dione 70. As of this writing, we are examining the cyclization conditions in
an effort to obtain a single product. We are also attempting to open the furan
obtained with acid catalyzed hydrolysis. We have found that the dione is
very susceptible to closure, and as a result we are attempting to find
conditions that will trap the dione in the open form. Should we succeed in
this endeavor we anticipate completion of eleaokanine 2 as outlined in
scheme 12. The results will be reported in due course.

84

O
Q° .\
DEAD

85%
O \
NaBH‘ C)HCOOH
M428” OC6H12
' 99%
O
Quant. 6 9

 

Scheme 11: Synthesis of Elaeokanine Precursor

O
Theorrrno.
0
Kinetic O CFacgaH
N Ketalization:

 

O 72 74
OTS 1)}(2003
Deprotect OHTSE 2)[0 l .
NaBH, 3) -—D_—>BU Eleaokamne
o 75 o 76 4’ LAH

Scheme 12: Future Synthetic Scheme

Starting with the hypothesis that furan terminated cationic cyclizations are
useful for the synthesis of various carbocyclic and azacyclic systems, we have
shown that indeed, furan cyclizations (2—)3 and 3—)2) are useful for the
preparation of alkaloid, pseudoguaianolide, and tigliane/daphnane systems.
These processes have provided both 5,5 and 5,6 membered fused rings
required for the alkaloid systems under study, and also the 5,7,5 membered
ring systems required for the pseudoguaianolides, and as a template for the
tetracyclic daphnane products. We have also shown that furan terminated

8 5

cationic cyclizations meet the criteria we require to be useful: 1)they cyclize
under mild conditions (both acidic and neutral), and 2) they proceed with
regiochemical predictability. During the course of these studies, we have
found that furan oxidation and hydrolysis are the limiting factors in the
effectiveness of the cationic process. These factors tend to be substrate
dependent therefore making conclusions about the generality of furans use
difficult. We hope that through the many examples we have shown that
furan oxidation and hydrolysis can become predictable. Currently, we are in
the process of completing formal total syntheses of several molecules
including phorbol/resiniferonol, fastigilin-C, aromatin, and elaeokanine A.
We should also be able to control the stereochemistry of various functional
groups about the periphery of all three rings contained in the bicyclo[5.3.0]-
decane skeleton.

EXPERIMENTAL

EXPERIMENTAL

Cyclization to Indolizidine Furan 55

To a stirring solution of hydroxy amide 54 (0.130g, 0.622 mmol) in
CH2C12 (-20°C) was added triethylamine (0.203ml, 1.86 mmol) and mesyl
chloride (0.089g, 0.933 mmol) sequentially. The reaction was warmed to RT
over a 3 hr. period, stirred overnight, and quenched with sat. aq. N aHCO3 (20
ml). The aqueous phase was extracted (3 x 20 ml CHzClz), the organic phases
combined, dried over MgSO4, concentrated in vacuo to give clear oil 56.
Chromatography (230-400 mesh, 10g, 20 mm O.D., EtOAc/Hexane, 2:1) on
silica gel using the flash technique gave 0.103g (86.6%) of 56, the closed furan
product

1H NMR (250 MHZ, CDCl3) 5 = 5.84 (S, 1H), 4.59 (m, 1H), 4.43 (dd, ]=9.36, 6.21
HZ, 1H), 3.55 (m, 1H), 3.05-2.33 (m, 4H), 2.26 (S, 3H), 1.73 (m, 2H).

B (neat): 2950, 2922, 2880, 1702, 1565, 1431, 1384, 1240, 1210, 1180, 992, 825, 685
cm'1.

ELL/Q (70eV) m/ e = 192 (M++1, 13.78), 191 (M+, 100), 190 (80.41), 176 (18.29), 148
(24.86), 135 (27.61), 134 (44.59), 133 (13.54), 431 (17.98), 91 (17.71), 77 (16.58), 55
(12.10), 43 (16.02).

Mitsunobu Product 61.

To a solution of 59 (1.66g, 10 mmol) in THF (25 ml) was added
succinimide (0.990g, 10 mmol), triphenylphosphine (3.01g, 11.5 mmol), and
DEAD (diethylazo-dicarboxylate) (1.81 ml, 2.00g, 11.5 mmol). The resulting
solution was stirred overnight, then concentrated in vacuo. Ether (60 ml) was
added to the residue and, after filtration, the solution was concentrated to
give fluffy white solid 61. Solid 61 was chromatographed (230-400 mesh, 200g,
50 mm O.D., EtOAc/Hexane, 70:30 on silica) using the flash technique to give
purified 61 (1.96g, 79.3%).

86

87

1H NMR (250 MHz, C6D6) 5 = 6.24 (S, 1H), 6.11 (m, 1H), 6.51 (dq, ]=9.25, 7.07
HZ, 1H), 3.65 (t, I=7.18 Hz, 2H), 2.84 (t, I=7.22 Hz, 2H), 2.00 (s, 3H), 1.75 (dd,
I=7.07, 1.73 Hz, 3H), 1.60 (S, 4H).

13C NMR (62.95 MHZ, C6D6) 5 = 175.88, 149.90, 131.9, 124.05, 120.91, 118.6,
108.20, 56.5, 37.22, 27.81, 26.34, 14.66, 11.75, -0.71.

IR (neat) 2955, 2918, 2856, 1701, 1592, 1422, 1399, 1233, 1218, 1175, 952, 830 cm'l.
EI-MS (70eV) m/e= 249 (MM-2, 0.48), 248 (M++1, 4.68), 247 (M+, 29.56), 149
(12.92), 148 (100), 147 (33.60), 135 (36.15), 133 (23.83), 105 (6.91), 91 (9.13), 55
(10.62), 43 (28.47).

Reduction to Q-Hydroxy Amide 62.

To a solution of 61 (1.30g, 5.28 mmol) in MeOH (15 m1, -4°C) was added
in one portion. After stirring for 2 hr. at -4°C, the solution was warmed to
RT, cast into aq. N aHCOg, extracted with CH2C12 (3 x 41 ml) the organic phases
combined, dried over MgSO4, and concentrated in vacuo to give 62 (~100%).
The product was used without further purification.

1H NMR (250 MHZ, C6D6) 5 = 6.25 (s, 1H), 6.10 (m, 1H), 5.51 (dq, J=9.36, 6.55
HZ, 1H), 4.82 (m, 1H), 4,53 (m, 1H), 3.94-3.29 (m, 3H), 2.87 (dt, I=7.64, 3.40 HZ,
1H), 2.30 (m, 1H), 2.00 (s, 3H), 2.00-1.54 (m, 3H), 1.77 (dd, I=7.10, 1.66 HZ, 3H).

Formation of Indolizidine Dione 64 and Furan 63.

To a vigorously stirring solution of 62 (1.30g, 5.22 mmol) in cyclo-
hexane (70 ml) was added 90% formic acid (0.25 ml) dropwise. After 5 min.
the reaction mixture was cast into a mixture (1:1) of CH2C12 and aq. NaHCO3
(100 ml). the phases were separated, the aqueous was washed with CH2C12 (4
x 20 ml), the organic fractions combined, dried over MgSO4, and concentrated
in vacuo to give a faint yellow oil. Chromatography (230-400 mesh, 100g 30
mm O.D., EtOAc/ MeOH/ TEA, 20:1:0.5) using the flash technique gave 0.93g
(71.5%) of 64 and a second fraction of 0.26g of 63.

8 8
64: 1H NMR (250 MHz, C6D6) 5 = 5.92 (m, 0.5H), 5.86 (m, 0.5H), 5.54 (dq, I=9.41,
6.62 HZ, 1H), 4.42 (m, 1H), 4.01 (m, 1H), 2.42 (m, 1H), 2.11 (m, 4H), 1.98 (S, 3H),
1.88-1.64 (m, 2H), 1.44 (dd,] = 7.22, 1.51 HZ, 3H), 1.26 (m, 2H).
13C NMR (62.95 MHZ, C6D6) 5 = 173.2, 147.4, 132.6, 128.9, 120.4, 115.5, 53.8, 37.7,
32.1, 26.3, 24.8, 16.2, 13.6.
$(neat) 2920, 2848, 1730, 1690, 1589, 1440, 1420, 1271, 1180, 1121, 947, 922, 723,
696 cm'l.
M (70eV) m/ e = 232 (M++1, 19.26), 231 (M+, 100), 230 (67.51), 216 (42.80), 188
(44.94), 174 (42.02), 160 (16.46), 145 (16.49), 131 (10.24), 121 (15.54), 91 (17.90), 77
(17.97), 55 (21.16), 43 (78.31).

63: 1H NMR (250 MHz, C6D6) 8 = 5.51 (m, 2H), 4.11 (ddd, 1:12.58, 6.41, 3.39 Hz,
1H), 3.04 (m, 1H), 2.76 (m, 1H), 2.34 (m, 2H), 2.15-1.63 (M, 6H), 2.04 (s, 3H), 1.22
(dd, 1:731, 1.39 Hz, 3H).

B (neat) 2930, 2910, 2851, 1718, 1705, 1445, 1430, 1280, 1230, 1180, 1130, 951, 843,
696 cm‘l.

Eli/13(70eV) m/e = 249 (M+, 1.61), 149 (12.23), 148 (100), 135 (23.98), 133 (11.59),
98 (7.11), 91 (11.22), 68 (23.31), 55 (8.84), 43 (58.23).

Mitsunobu Adduct 68.
To a solution of 67 (0.650g, 4.77 mmol) in THF (22 ml) was added

succinimide (0.568g, 5.73 mmol), triphenylphosphine (1.75g, 6.69 mmol) and
DEAD (diethylazo-dicarboxylate) (1.18g, 6.69 mmol). The resulting solution
was stirred overnight, then concentrated in vacuo. Ether (40 ml) was added to
the residue, and after filtration, the solution was concentrated to give a clear
oil. The oil was chromatographed (230-400 mesh, 100g, 40 mm O.D.,
EtOAc/Hexane, 70:30) using the flash technique to give 68 (0.818g, 84.9%)

1H NMR (250 MHZ, C6D5) 5 = 5.78 (S, 1H), 3.64 (t, I=7.3 HZ, 2H), 2.81 (t, I=7.4
HZ, 2H), 2.12 (t, I=7.3 Hz, 2H), 1.99 (S, 3H), 1.78 (S, 4H), 1.42 (S, I=7.4 HZ, 2H), 0.82
(t, J=7.3 Hz, 3H).

13C NMR (62.95 MHZ, C6D6) 5 = 176.14, 149.41, 146.26, 119.66, 188.67, 37.46,
27.89, 27.06, 26.42, 23.81, 13.80, 11.30.

B (neat) 2959, 2932 2872, 1776, 1705, 1441, 1435, 1402, 1242, 1159, 1145, 821, 663
cm'l.

89

Reduction to d-Hydroxy Amide 69.

To a solution of 68 (0.284g, 1.14 mmol) in MeOH (5 ml) cooled to 0°C,
was added NaBH4 (0.064g, 1.71 mmol) in one portion. After stirring at RT for
2 hrs., the excess N aBH4 was quenched with H20 (2 ml), concentrated, the
residue taken up in CHzClz (10 ml), dried over NazSO4, and concentratged in
vacuo to provide 0.278g (97%) of 69 as a yellow oil. The product was used
without further purification.

1H NMR (250 MHZ, C6D6) 5 = 5.85 (s, 1H), 3.80 (m, 1H), 3.62 (m, 1H), 2.89 (dt,
J=7.33, 3.51 HZ, 1H), 2.44 (m, 1H), 2.15 (m, 4H), 2.01 (s, 3H), 2.00-1.69 (m, 5H),
1.44 (d, ]=7.43 HZ, 2H), 0.84 (t, I=7.29 HZ, 3H).

13C NMR (62.95 MHZ, C6D6) 5 = 175.14, 150.46, 145.87, 119.76, 108.44, 83.56,
39.28, 29.16, 26.40, 27.13, 26.98, 23.83, 13.84, 11.31.

IR (neat) 3317, 2957, 2930, 2872, 1695, 1670, 1576, 1464, 1458, 1421, 1284, 1211,
1068, 989, 800.

Cyclization to Indolizidine Furan 21 and Dione 20.

To a vigorously stirred solution of 69 (0.140g, 0.557 mmol) in
cyclohexane (15 ml) was added HCOOH (0.084 ml) rapidly. After stirring 5
min., the two-phase mixture was immeidately cast into H20 (10 ml) and
CHzClz (10 ml). The aqueous layer was separated, extracted with CHzClz (3 x
15 ml), the combined organic layers were washed with sat. N aHC03 (50 m1),
brine (50 ml), dried (Na2504), and concentrated in vacuo. The clear oil was
purified on a column of silica gel (230-400 mesh, 14g, 20 mm O.D., EtOAc,
CH2C12, MeOH, 8:1:1) to provide 0.064 g (49.3%) of 70 and 0.073g (52.1%) of 71
using the flash technique.

71: 1H NMR (250 MHz, C6D6) 8: 4.41 (dd, 1:12.61, 5.04 Hz, 1H), 3.95 (m, 1H),
2.36 (m, 2H), 2.18-2.00 (m, 4H), 1.97 (s, 3H), 1.70 (m, 1H), 141.1 (m, 4H), 0.81 (t,
1:7.3 Hz, 3H).

13C NMR (62.95 MHz, C6D6) 8 = 172.33, 1465.61, 145.11, 119.79, 116.48, 53.89,
36.41, 31.62, 26.46, 26.14, 23.51, 23.45, 13.96, 11.36.

m (neat) 2959, 2932, 2870, 1695, 1419, 1269, 1149, 945, 884, 705 cm-1

9 0
EI-MS (70eV) m/ e = 233 (M++2, 12.55), 232 (M++1, 10.08), 204 (11.59), 176 (4.62),
150 (4.26), 137 (3.02), 121 (5.40), 96 (7.87), 91 (6.43), 77 (9.02), 68 (14.23), 55 (28.56),
43 (100).

70: 1H NMR (250 MHz, C6D6) 8 : 4.07 (ddd, 1:13.08, 6.54, 3.89 Hz, 1H), 3.42 (bq,
I=7.01 HZ, 1H), 3.25 (m, 1H), 2.43 (m, 1H), 2.13-1.73 (m, 5H), 1.94 (S, 3H), 1.64-
1.00 (m, 6H), 0.85 (t, 1:7.27, 3H).

13C N MR (62.95 MHZ, C6D6) 5 = 208.83, 206.21, 173.20, 62.07, 58.42, 57.19, 50.06,
39.52, 38.54, 33.68, 29.91, 29.93, 21.49, 14.21.

B (neat) 2996, 2910, 1720, 1692, 1495, 1435, 1376, 1320, 1287, 1175, 1003, 865, 690
cm‘l.

MS (708V) 252 (M++1, 0.30), 208 (2.90), 153 (5.47), 152 (65.65), 125 (6.15), 115
(6.99), 114 (7.15), 110 (3.97), 96 (24.91), 84 (22.74), 71 (29.09), 68 (34.61), 55 (48.26),
43 (100).

LIST OF REFERENCES

10.

11.

LIST OF REFERENCES

a) Speckamp, W.N. Recl. Trav. Pays-Bus 1981, 100, 345.

b) Zaug, H.E. Synthesis 1984,85, 1981.

c) Speckamp, W.N.; Heimstra, H. Tetrahedron 1985, 41, 4367.

d) Heimstra, H.; Speckamp, W.N. "The Alkaloids", Brossi, A., ed.,
Academic Press, Inc., 1988, Vol. 2, pp. 271-344.

See Reference 1 c.

Evans, D.A.; Thomas, E.W.; Cherpeck, RE. I. Am. Chem. Soc. 1982,104,

3695.

Hart, D]. I. Org. Chem. 1981,46, 3576.

Chamberlin, A. R; Nguyen, H. D.; Chung, J. Y. L. J. Org. Chem. 1984, 49,

1682.

a) Keck, G. R; Yates, J. B. I. Org. Chem. 1982,47, 3590.

b) Aratani, M.; Dunkerton, L. V.; Kishi, Y.; Kakoi, H.; Sugiura, S.;
Inoue, S. J. Org. Chem. 1975, 40, 2009.

c) Shoemaker, H. E.; Speckamp, W. N. Tetrahedron 1980, 36, 951.

Hart, D. J.; Kanai, K. I. I. Am. Chem. Soc. 1983,105, 1255.

See References 21-23..

a) Tscherniac, J. Ger. Pat. 134979, 1904.

b) Einhom, A. Leibigs. Ann. Chem. 1905, 343, 207.

a) Weinreb, S.M.; Levin, J. D. Heterocycles 1979, 12, 949.

b) Hall, H.K.; Minutti, D.C. Tetrahedron Lett. 1984, 29, 943.

c) Campbell, K.N.; Sommers, A.H.; Campbell, B.K. I. Am. Chem. Soc.
1944, 66, 82.

d) Shono, T.; Aoki, T. I. Am. Chem. Soc. 1982, 104, 6697.

e) Shono, T. Tetrahedron 1984, 40, 811.

a) Matobu, K.; Yamazaki, T. Chem. Pharm. Bull. 1974, 22, 2999.

b) Matsuda, F.; Yanagiya, M.; Matsumoto, T. Tetrahedron Lett. 1983,
23, 4043.

91

 

12.

13.
14.

15.

16.
17.

18.
19.

20.

21.

a)

b)

c)

92

Hubert, J.C.; Wijnberg, J.B.P.A.; Speckamp, W.N. Tetrahedron 1975,
31 , 1437.

Hubert, J.C.; Speckamp, W.N.; Huisman, H.O. Tetrahedron Lett.
1972, 4493.

Shoemaker, H.E.; Dijkink, J.; Speckamp, W.N. Tetrahedron 1978,
34, 163.

Chamberlin, A. R.; Chung, J.Y.L. I. Am. Chem. Soc. 1982, 104, 3653.

a)
b)

a)
b)

C)

Chamberlin, A.R; Chung, J.Y.L. Tetrahedron Lett. 1982, 23, 2619.
Chamberlin, A.R.; Nguyen, H.D.; Chung, J.Y.L. J. Org. Chem.
1984, 49, 1682.

Hamersma, J.A.M.; Speckamp, W.N. Tetrahedron 1982, 38, 3255.

Shoemaker, H.E.; Dijkink, J.; Speckamp, W.N. Tetrahedron 1978,

34, 163.

Hubert, J.C.; Wijnberg, J.B.P.A.; Speckamp, W.N. Tetrahedron

1975, 31, 1437.

Dikjink, J.; Speckamp, W. N. Heterocycles 1979,12, 1147.

Overman, L.E.; Malone, T.C.; Meier, G.P. I. Am. Chem. Soc. 1983,105,
1407.

Wijnberg, B.P.; Speckamp, W.N. Tetrahedron Lett. 1980, 21, 1987.

a)
b)
c)

a)
b)
C)
d)

e)

0

Hart, N .K.; Johns, S.R.; Lamberton, J.A. Aust. I. Chem. 1972, 25, 817.
Gribble, G.W.; Switzer, EL; 5011, RM. I. Org. Chem. 1988, 53, 3164.
Johns, S.R.; Lamberton, J.A. "The Alkaloids", Manske, R.H.F., ed.,
Academic Press, New York, 1973, Vol. 14, pp.325.

Grieco, P.A.; Parker, D.T. I. Org. Chem. 1988, 53, 3325.

Tufariello, I.J.; Tegeler, J.J. Tetrahedron Lett. 1976, 4037.

Bremmer, M.L.; Khatri, N.A.; Weinreb, S. M. I. Org. Chem. 1983,
48, 3661.

N agas, Y.; Dai, W.-M.; Ochdai, M.; Tsukagoshi, S.; Tujita, E. I. Am.
Chem. Soc. 1988, 110, 289.

Haddad, M.; Celerier, J.-P.; Lhommer, G. Heterocycles 1987, 26, 2335.
Takahata, H.; Yamabe, K.; Suzuki, T.; Yamayaki, T. Heterocycles
1986, 24, 37.

Chemberlin, A.R; Nguyen, H.D.; Chung, J.Y.L. I. Org. Chem. 1984, 49,
1682.
Wijnberg, B.P.; Speckamp, W.N. Tetrahedron Lett. 1981, 22, 5079.

24.

26.
27.
28.

29.
30.
31.
32.
33.

35.

36.

37.
38.
39.

40.

93

Khatri, N .A.; Schmitthenner, H.F.; Shringarpure, J.; Weinreb, S.M. I.

Am. Chem. Soc. 1981,103, 6387.

Raggon, J.W., Ph.D. Thesis, Michigan State University, 1986.

Tanis, S. P.; Dixon, L. A.; McMills, M. C.; Raggon, J. R Manuscript in

Preparation

Dixon, L., Ph.D. Thesis, Michigan State University, 1988.

Mitsunobu, O.E. Synthesis, 1981, 1.

a) Tsunoda, T.; Suzuki, M.; Noyori, R. Tetrahedron Lett. 1980, 21,
1357.

b) ku, J.R.; Werzel, J.M. I. Org. Chem. 1985, 50, 3948.

See Ref. 23.

Hart, D.J.; Tsai, Y.M. I. Am. Chem. Soc. 1984,106, 8209.

Yang, Y.-S.; Manna, S.; Falck, J.R. I. Am. Chem. Soc. 1984,106, 3811.

Tanis, S.P.; Dixon, L.A. Tetrahedron Lett. 1987, 28, 2495.

Tanis, Raggon and Dixon, unpublished observations.

Tanis, S.P.; Chuang, Y.-H.; Head, D.B. I. Org. Chem. 1988, 53, 4929.

a) Tufariello, J.J.; Muller, G.B.; Tegeler, J.J.; Trybulski, E.J.; Wong, S.C.;
Ali, Sk.A. I. Am. Chem. Soc. 1979, 101, 2435.

b) Lewin, A.H.; Naseree, T.; Carroll, F.I. I. Heterocyclic Chem. 1987, 24,

19.

a) Danheiser, RL.; Morin, Jr., J.M.; Salaski, E]. I. Am. Chem. Soc.
1985, 107, 8066.

b) Tufariello, J.J.; Meckler, A.; Senaratne, K.P.A. I. Am. Chem. Soc.
1984, 106, 7979.

c) Melching, K.H.; Heimstra, H.; Klaver, W.I.; Speckamp, W.N.
Tetrahedron Lett. 1986, 27, 4799.

See Reference 26.

See Reference 33.

Tanis, S. P. Unpublished Data for the Preparation of Various Substituted

Furans.

Padwa, A.; Gasdaska, J. R. Tetrahedron 1988, 44, 4147.

HIGRN STRTE UNIV. IB

HIC L

312930056332

RRRIES
IININHI
21