. .~:-.'.- A:

 

4

. pr

;'-;\:tinu:
;

 

’m.’
“w;
.1“ (‘3‘4‘0
(0:55;. \ I; )0 ;
9

I
Iu'A'Hq A .2.
9' flit“:-
f)‘
:7 an

hi";

r‘JA’
.2 $39.;
0’ V

{5.1: ‘
"“535

1.x?“
$3.13
I

.
,. .. ,1 . .:‘1.7§\;r.«
l}“‘ . ’ .A;. m$e ‘2‘ 4f.) ,U’/.

 

 

 

 

 

 

 

 

 

 

1,

‘ 2

’ “psafgz’ine‘

" l u a"
,"u'u/

, .
u -
(4:. £121.; 4

u

l: 0’. .
A P)‘.’u’f{.’,’..g . .
4 . UAAA
‘ 1.2.4.4 ‘
l
C

”I
-( ta
Il!‘fi‘v”a>u,a -
f an Au:

" ding“: I
(Hi, {£571
,3 . . 4

Q?“
{2?}; n.1,:-

134. u...

a

9
v
S

"(1)

r u l q . _ 1': 3

we; f m "2.” . m, - 5:3,; 32%.;
“‘7: ' _nlh4 , {IL-9' ,fl A .

10m. A a . ?

32!}: SIS: If?! )7???
I”

M: .m 1]

"Vi.
" . 37.1.1 ‘
a/:%;A1. ,J'qrt'
ff?"

”[3; 5r , .
[4‘9 >- ‘l’f
9‘3!” 1".“

a. ‘JTAHUJIH 7': 1
{~- (.3
J

I
9.3:",n7,
”£9“, “9 ”Lg/{555‘ '

fag}; 3.1.. rags.

-\ J«;rft:4}:gv Jab:
Ira. wrung;

{f‘w

n It 3&3}; Mfii- “.49: ’ 3:521. 451;.
‘ . 4 .- d -- J J 3 ' . r9,
mgr; ” fw :2er 5‘5"}.
:2531- 35.55.71“. 1}}? ' a

‘94’15' ~59"? A?“ {fin/44")“

agffin’ LA.‘ 3 4.. a "A if; A}

‘ u... I"
4" (h ':-f “a" %:€{:§::;¢
”:33; 4 ""3“" _ .. ' P 34":th 5:41
»....¢;=:m.:~£*"~ r -. .r (51» mm

.4
( . 4 J
'L‘mr'a‘v’"\-‘>\‘~"- HI

w‘m

 

 

I"!

 

x: W) 2*“ ‘9 “l
llHllHllHHllHllHHHIllIllHlHHHIIUHIHHHHMHI
3 1293 00550 3408 r
LIBRARY

Michigan State
University

THEE“:

 

This is to certify that the

dissertation entitled

FURAN TERMINTED N-ACYLIMINIUM ION CYCLIZATIONS DI THE

SYNTHESIS OF FUSED-, SPIROCYCLIC-, AND BRIDGED—RING
CONTAINING ALKALOIDS

presented by

Lisa Ann Dixon

has been accepted towards fulfillment
of the requirements for

Ph. D. Chemistry

degree in

537;» (me)

 

Major professor

M. W775}?

mum-"Afr ,. . . ,1 m“ . . - - 0.12771

 

 

MSU

LlBRARlES

  

 

  

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

FURAN TERMINATED N—ACYLIMINIUM ION CYCLIZATIONS IN THE
SYNTHESIS OF FUSED-, SPIROCYCLIC-, AND BRIDGED- RING
CONTAINING ALKALOIDS

BY

LISA ANN DIXON

A DISSERTATION

Submitted to

Michigan State University

in partial fulfillment of the requirements

for the degree of

DOCTOR OF PHILOSOPHY

Department of Chemistry

1988

 

ABSTRACT

FURAN TERMINATED N-ACYLIMINIUM ION CYCLIZATIONS IN THE
SYNTHESIS OF FUSED-, SPIROCYCLIC-, AND BRIDGED- RING
CONTAINING ALKALOIDS

BY

LISA ANN DIXON

In an ongoing investigation of furans as terminators in cationic cyclization, we
have examined the use of M—acylimlnium ions as the initiators. This approach
has proven to be successful in the preparation of the linearly fused, bridged,
and spirocyclic ring systems present in many alkaloids.

We had hoped to utilize furans in both the cyclization sequence, and as
resident functionality to facilitate completion of chosen natural product
syntheses. Several questions had to be addressed during this study: 1) could
the required furan containing cyclization precursors be prepared, 2) would the
furans withstand the cyclization conditions, 3) would both the 2- to 3-furyl mode
of closure as well as the relatively more reactive 3— to 2 furyl closure be viable
processes, 4) could the furans be manipulated to provide useful synthetic
intermediates, and 5) could the intermediates be transformed into the targeted
alkaloids.

Studies directed towards answering these questions will be presented. To
summarize we will describe the successful preparation of 5,6 and 6,6 linearly
fused ring systems through the 2- to 3- furyl closure, and 5,6; 6,6; 5,7; and 6,7

linearly fused ring systems through the 3- to 2- furyl closure. The 6,6 fused

system (2- to 3- furyl closure) has been employed in a total synthesis of (+/-)-
epi-lupinine. The spiropiperidine ring system present in histrionicotoxin and
perhydrohistrionicotoxin has been prepared, and applied to a formal total
synthesis of the later. Finally, two bridged ring containing systems, the
aza[3.2.1]bicyclooctane, present in cocaine and the aza[4.2.1]bicyclononane,

present in anatoxin-a have been prepared.

 

 

DEDICATION

To Mom and Dad

ACKNOWLEDGEMENTS

The author wishes to express her sincere appreciation to Dr. Steven P. Tanis for
his guidance, patience, and advice, to Dr. William Reusch for his adept handling
of internal affairs and to them both for their excellent party throwing. Thanks
also go to Dr. Mike Fiathke for serving as my second reader and for his many
jokes, several of which were funny.

Financial support from Michigan State University and the NIH is gratefully
acknowledged.

The author would like to thank everyone past and present for their helpful
discussions and assistance, in particular: Jeffery Raggon and Mark McMills for
their contributions to this study, as well as for their many entertaining hours; and
Ben and Zepp for their companionship.

Many thanks go to Nancy for keeping me in touch with the outside world.

A special acknowledgment goes to my parents and family whose love and
encouragement kept me motivated (remember Mike- It's Dr. Dixon to you).
Thanks to Bubber for making me look forward to college, and always seeming
proud to have me as his little sister

Finally, a very special thanks goes to Paul for his advice, understanding,
support, and assistance.

TABLE OF CONTENTS

 

 

 

 

 

 

Page

LIST OF FIGURES vii
LIST OF EQUATIONS viii
LIST OF SCHEMES x
INTRODUCTION 1
TARGET STRUCTURES 19
LINEARLY FUSED SYSTEMS: LUPININE AND ELAEOKANINE-A ............... 20
SPIROCYCLIC ALKALOIDS: A FORMAL TOTAL SYNTHESIS OF
(+/-)-PERHYDROHISTRIONICOTOXIN a7
BRIDGED ALKALOIDS: AN APPROACH TO (+/-)-COCAINE AND
(+/-)-ANATOXlN-A 45

 

EXPERIMENTAI
LIST OF REFERENCES

 

vi

HQ

101

Figure 1

Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8

Figure 9
Figure 10

Figure 11
Figure 12
Figure 13
Figure 14
Figure 15
Figure 16
Figure 17

Figure 18

LIST OF FIGURES

 

 

 

 

 

 

 

 

 

 

Rage
Some Representative Examples of Furan—
Terminated CationicCyclizations 2
Furan Equivalencies ’i
lminium Ion and N—Acyliminium Ion 3
13C-NMR Data for lminium Ions and N—Acyliminium Ions .......... 4
Cyclization of Acyliminium Ion vs N-Acyliminium Ion .................. 4
Precursors of N—Acyliminium Ions 6
Cyclization of N-substituted Cyclic Carbinol-Amides .................. 1 0
Some Preliminary Studies from the Laboratories of
Speckamp 1 1
Cyclization vs Enamide Dimerization 13
Basic Stmctural Possibilities for Furan Terminated N-
Acyliminium Ion Initiated Cyclizations 19
Potential Alkaloid Targets 22
A Possible Approach to (+/-)-Perhydrohistrionicotoxin ............... 40
Possible Furan Elaboration Protocol 41
A Retrosynthetic Analysis for a Cocaine Synthesis ..................... 47
A Possible Route to (+/-)-Cocaine 50
Second Generation Approach to (+/-)-Cocaine ........................... 55
First Retrosynthetic Plan Leading to Anatoxin-a .......................... 61
Third Retrosynthetic Plan Leading to Anatoxin-a ........................ 64

(15)
(16)

(17)

LIST OF EQUATIONS

 

Amide Addition to a Carbonyl

 

N-Acylimidate Reduction

 

lmide to On-alkoxylactam

 

Cyclizations

 

Chair-like Transition State

 

 

Precursors

 

 

 

 

Intermediate 76

 

Intermediate 88

 

 

Page

The Tscherniac-Einhorn Reaction 5

7

7

8
Vinylsilanes as Terminators in N-Acyliminium Ion

12

12
Proposed Furan Oxidation of 40b to Ouinolizidine

98
Clausen-Kaas Oxidation of 40b 29
Preparation of Cyclized Substrate 64 ”-18
"Kinetic Ketalization" of Dione 63b 14
"Thermodynamic" Ketalization of Dione 63b ............................... 34
Attempted Thioketalization of 63b 64
Baeyer-Villager Oxidation of Pyrrolizidine Alkaloid

"<17
Clauson-Kaas Oxidation of Spiropiperldine

42
Attempted Preparation of 5—Silyl Furan 43

53

Li/NH3 Reduction of 118

 

Proposed Formal Total Synthesis of (+/-)-Cocaine

viii

(18)
(19)
(21)
(22)
(23)

from 143 57

Second Formal Total Synthesis of (+/-)-Cocaine from 143 ...... 57

 

Proposed Deprotection/Methylation of 143 ................................. 57
Preparation of Furanacetaldehyde Dimethylacetal 156 ............ 63
Preparation of Furanacetaldehyde 157 63
Preparation of Nitro-Alcohol 158 63

 

ix

Scheme

1

Scheme 2

Scheme 3

Scheme 4

Scheme 5

Scheme 6

Scheme 7

Scheme 8

Scheme 9

Scheme

Scheme
Scheme
Scheme
Scheme
Scheme
Scheme
Scheme

Scheme

10

11
12
13
14
15
16

LIST OF SCHEMES

Evans'25 Approach to Spiropiperldine Preparation ...............

Page

..... 14

 

Speckamp's26 Synthesis of Anatoxin-a

Rapoport'327 Synthesis of Anatoxin-a

 

Chamberlin's14 Construction of the Pyrrolizidine,

 

Indolizidine, and Related Ring Systems

An Approach to the Linearly Fused Alkaloids .........................

Linearly Fused Furan 3- to 2- Closure

Chamberlin's14 Synthesis of (+/-)-epi-Lupinine .....................

Tufariello’529b Synthesis of (+/-)-Lupinine

The Examination of a Silyl Furan as a Terminator .................

Second Generation Approach to Lupinine and
lndolizidine Alkaloids

Synthesis of (+/-)-Epi-Lupinine

Kishi's33Cyd Synthesis of (+/-)-Perhydrohistrionicotoxin .......
First Generation Spiropiperldine Construction .......................

A Formal Total Synthesis of (+/-)-Perhydrohistrionicotoxin

..... 24

9‘3

..... 26

27

..... 31

33

36

..... 38

..... 40

44

47

 

Tuffariello's31a Synthesis of (+/-)-Cocaine

The Construction of the Cocaine Ring System .......................

Cocaine Side Chain and Carbinol Manipulation ...................

Lewin's31b Synthesis of (+)-Cocaine

..... 48

..... 51

..53

 

Scheme 19

Scheme 20

Scheme 21

Scheme 22

Scheme 23

Scheme 24

Second Generation Preparation of the TrOpane
Ring System 56

 

 

Speckamp'525 Synthesis of Anatoxin-a 60

Second Generation Construction of the Anatoxin-a
Ring System 63

 

Third Generation Construction of the Anatoxin-a
Ring System as

 

Fourth Generation Construction of the Anatoxin-a
Ring System 66

 

Proposed Conclusion to the (+/-)-Anatoxln-a Synthesis ............ 68

xi

INTRODUCTION

 

.. .
-. J

INTRODUCTION

As part of a general program in furan chemistry , we have been examining
furan terminated cyclizations as a method for the construction of linearly-fused,
spirocyclic, and bridged-carbocyclic ring systems.1 We have found that by
using enones‘e, epoxidesld, and allylic alcohols1d as cationic initiators, the
above mentioned ring systems may be readily prepared (Figure 1). Our interest
in developing substituted furans as bis-nucleophllic synthons in annulative
processes stemmed from the variety of useful functional groups which might be
realized from the relatively unreactive furyl nucleus (Figure 2). We have utilized
this concept to accomplish a total synthesis of nakafuran-91d and a formal total
synthesis of (+/-)- and (+)-aphidicolin10. Should a nitrogen containing initiator
be employed, similar methodology might provide an entry into alkaloid natural
products containing the ring systems mentioned above. Possible initiators
include the lminium ion and the N—acyliminium ion (Figure 3).2

In choosing between these two initiators we need to consider several factors,
including the relative reactivity and stability of the two moieties. There is very
little quantitative data available concerning N-acyliminium vs lminium ions;
nonetheless, it is possible to make some qualitative distinctions.

As would be expected, substitution of an acyl group on an lminium ion results
in a more electron poor carbon. This was demonstrated by Wurthwein and
coworkers using 1SC spectroscopy (Figure 4).3 They observed a significant

downfield shift when the nitrogen was substituted with an electron withdrawing

1

carbonyl group. The implication here is that the N-acyliminium ion would be
more electrophilic and more reactive than the simple iminium ion. This
prediction is borne out experimentally as Figure 5 demonstrates.4 In this
example an olefinic terminator, which is a relatively unreactive nucleophile,
which is also quite hindered, succeeds in participating in the cyclization only

when an N-acyliminium ion is used as the initiator. In general, lminium ions

HOD—2H-
allylic alcohol

E26
//

61 -68°/o

HCD 2H
3° carbocation

\\o

2

\
E?)

56%

HCD H
\ / enorzte

O

\
go

O

72%

enone

79%

F
«El

Figure 1: Some Representative Examples of Furan-Terminated Cationic Cyclizations

3

require either very reactive terminators or very strongly acidic conditionsSto
participate in cyclization sequences. There are exceptions, such as when a
stable carbenium ion intermediate is formed upon termination.6 N-acyliminium
ions, on the other hand, rarely require extremely harsh conditions and need
much less reactive terminators for successful cyclizations.2 ~

Another rationale for the synthetic utility of N-acyliminium ions may lie in the

types of products formed. After generation of either type of ion they are usually

oi? O Q

 

 

\
/

HO/_\ OH

 

Figure 2: Furan Equivalencies

O
I
Rlyfi: R4 R1>=Gr§l¢ R5

R2 R3 R2

R1 , R2, R3 , R4: aryl, alkyl, H
R5 = C or heteroatom

Figure 3', lminium Ion and N—Acyl lminium Ion

l1
C=N C:
I I
Ph H Ph

170 7 152.0
184.6 x 189.7 0 190.6 0
P“ @M9 A 6) FR (9*
\ I \
Mk H

H
$130169 sncuo SbCle

Figure 4; 13C-NMR Data for lminium Ions and N-Acyliminium Ions

trapped by a nucleophile. In the case of an lminium ion the product is an amine;
in the case of an N-acyliminium ion the product is an amide. It is known that an
amine is more reactive under a variety of reaction conditions.7 Therefore,
another advantage of N-acyliminium ions over lminium ions may be the
production of a more stable product from the former over the latter. The efficacy
of forming a more stable amide instead of an amine can be exploited while
performing steps subsequent to the cyclization. An amine might be the ultimate
target; however most workers choose to keep it masked as an amide, only

exposing the free amine in the later stages of the synthesis.

R1
R1 or R2=O N 2
R1
N (9

Figure 5: Cyclization of Acyliminium Ion lam-Acyliminium ion

R‘ = R2 = H
No Reaction

Upon evaluation of the points just discussed, the increased reactivity of an N-
acyliminium ion, the milder conditions required for successful cyclization
reactions, and procurement of a more stable product, it is not surprising that the
N—acyliminium ion has become the initiator of choice in many syntheses. Since
our terminator is relatively unreactive and is sensitive to Bronsted acidic
conditions, we also chose to examine N-acyliminium ion in our furan terminated
cyclizations.

N-acyliminium ions have been known since the early 1900's when the

Tscherniac-Einhorn reaction (Eq. 1) was discovered.8 They may be generated

0 o O o O 0
H80
RkNHCHZOH 2 “ RANHCHZ R/‘kN=CH2 Ph” RANHCHzPh (I)
H

 

 

 

in a number of ways as outlined in Figure 6.2 The protonation of N-acylimines
(Figure 6, path b) has seen limited use as the starting materials are not very
stable.9 N-acylation of imines 10 (Figure 6, path a) has some synthetic utility;
the imine required can not often be isolated, but it can give rise to N-acyliminiun
ion intermediates.11 Enamides can in some instances be protonated to furnish
N-acyliminiun ions (Figure 6, path 0), however this reactions has limitations.12
Amides will undergo electrochemical oxidation (Figure 6 path d), usually in the
presence of methanol to give or—alkoxyamides.13 As will be seen shortly, on-
alkoxy amides are very efficient sources of N-acyliminiun ions.

Figure 6, path e describes the process most often used in synthetic
sequences for generating N-acyliminium ions; the heterolysis of a-substltuted

amides. Usually the heteroatom to be solvolyzed is oxygen, but halogen,

6

nitrogen, sulfur or phosphorus have been used. When X=OH or )R, either
Bronsted or Lewis acids can be employed to generate the N-acyliminium ion;2 if
R=OMs no acid catalysis is necessary.14

oc-Alkoxy or oc-hydroxy amides are, in turn, readily prepared from stable, often
commercially available starting materials. The first method, the nucleophilic
addition of amines to aldehydes or ketones (Eq. 2) requires use of very reactive
carbonyl compounds or intramolecular addition.2v15 The product hydroxy
amide is often not isolated, but carried to an N-acyliminium ion intermediate
directly. Equation 3 describes the reduction of N-acylimidates with NaBH4 to
provide oc-alkoxy amides, which might suffer C-O bond heterolysis to generate

the desired N—acyliminium ion.16

WQO RayRi/K

RJOKN/KR.

szRa/erolysis C)\XO

d) oxidation

RtiNz/iRCRS
o H
WAN/NR3
R2 R4

Figure 5; Precursors of N-Acyliminium Ions

7
Addition of some Grignard reagents to cyclic imides provides oc-hydroxy amides
in good yield (eq. 4, c,d).17 if R'=H two equivalents of Grignard reagent are

needed.

 

O o O ' OH
Pal/"\NHZ + #th3 R1 F312 (2)
R
——’ (3)
O N/ OE! EtOH O N OE!
H

Possibly the most widely used route to oc-alkoxy and oc-hydroxy substituted
amides is the partial reduction of cyclic imides. The pioneering work of
Speckamp in the early seventies demonstrated that this was a viable method,
and paved the way for the intense synthetic effort into N-acyliminium ion
research over the last 15 years. Speckamp accomplished this feat by reducing
either glutarimide or succinimide with NaBH4 in ethanol in the presence of acid
to provide oc-ethoxylactams in good to excellent yields (Eq. 4b).18 More recently
Chamberlin has demonstrated that cyclic imides can successfully be reduced to
oc-hydroxy amides using NaBH4 in methanol at low temperature (Eq. 4a).14 He
also reported that a-hydroxy lactams will form N-acyliminlum ions, by solvolysis,
in the presence of mesyl chloride and triethylamine, thus avoiding the use of
acid catalysis. These last methods discussed, the heterolysis of a-substituted
amides, account for an overwhelming majority of the approaches in which N-

acyliminiun ions are in synthesis.

- . R
R1 R1
Entry gengltlene _F11_ —Rz— R3_
a NaBH4 , MeOH alkyl H H
b NaBH4 , ROH. H+ alkyl H ' alkyl
6 R2ng alkyl alkyl H
d RzMgX (2 eq.) H alkyl H

We have described a number of routes for preparation of N-acyliminium ions
but have not touched on some other facets of their use. Still to be discussed is
the fate of an N-acyliminium ion: the types of terminators that have been
employed in cyclizations and the types of ring systems available from this
methodology. Specific issues to be considered are: 1) carbonolamides
opening to ketoamides; 2) enamide formation/enamide dimerization; 3)
regiochemical ambiguities in the termination step; 4) stereochemistry of the
cyclization. A few examples highlighting these topics follow (vida supra).

Prior to Speckamp's discovery of the reduction of cyclic imides only a few
stable terminators had been used in N-acyliminium ion cyclizations.
Speckamp's methodology opened the door for experimentation with a variety of
terminators, including olefins. Figure 7 outlines the proposed mechanism for an
olefin terminated N-acylimnium ion initiated cyclization, proceeding from a
carbinolamide ether, such as A. Treatment of oc-alkoxy amide with acid leads to
an N-acyliminium ion B which is in equilibrium with TI complex D or C (D has
also been depicted as a bridged carbenium ion D'). Complex C has a boat-like
conformation and is consequently thought to be less favorable than the chair
like conformation D. Attack of the terminating function on D will lead to either G

or H depending of the ring size of the lactam and the nature of the R groups.

9

Should the R groups be a strongly cation-stabilizing group; B may provide
discrete carbenium ions E or F which may then trap a nucleophile or lose a
proton to form the products. If the terminator is aromatic, loss of a proton and
rearomatization typically occurs.

The first problem Speckamp solved was inhibiting ring opening of the
carbinol amide to a ketoamide which was susceptible to further reduction with
NaBH4 to yield an amide alcohol. He accomplished this by the careful addition
of H2804 or HCl to the reaction medium during reduction to give the related oc-
alkoxyamides.18 With a source of oz-alkoxy-lactam assured, he then examined
the crucial cyclization. Figure 8 outlines a few terminators and the ring systems
first constructed.19 Treatment of 1a or b with formic acid for 72 hours provided,
after capture with formate and hydrolysis, keto-amide 2a in 97% (n=1) or 2b in
88% (n=2). Compound 16 under the same treatment for 18 hr provided formate
2c (R=CHO) in nearly quantitative yield. This can be contrasted with the
substituted olefin 1d which cyclizes in formic acid (-5° C - RT)) immediately or
under milder conditions (acetic acid) in 24 hr. The former conditions provide
formate 2d (R=COH); the latter condition acetate 2d (R=COCH3). It is
interesting to note that with the exception of 2d, extended periods of time in
stronger acid (HCOzH vs HOAc) were required to induce cyclization.

We see that six membered rings are generally favored, however there are
examples in which either a mixture or mainly the 5-membered ring are
obtainedfi‘in20 An elegant control method to avoid regiochemical ambiguities is
to utilize the unique ability of either a silicon21 or a sulfur”:22 atom substituted
terminator to direct the course of cyclization. Overman has used this
methodology in the preparation of linearly fused systemsr?1a his results are
outlined in equation 5. Reduction of imides 3a-c with NaBH4 and treatment of

the carbinolamide with trifluoroacetic acid afforded cyclized products 4a-c. By

10

using 1-bromo vinylsilane he is able to control the regiochemistry of the
cyclization and also produce an olefin that is differentially substituted for further

transformations.

O

products ‘_N_u_

 

products -N—u

 

II

Q

Figure 7: Cyclization of N-Substituted Cyclic Carbinol-Amides

N-acyliminlum ion cyclizations can proceed not only with excellent

regioselectivity, but, often with excellent stereoselectivity providing high yields

11

of single isomersZC. As shown in equation 6, the face of the lactam to be

attacked can be determined by the size of the substituents on the lactam. Attack

)n )n
acid 0 ,
O OEt N

 

R
R.
1. 2
L 1 i M
a 1 (CH2)20-CH =0 97%
b 2 (CHzlzC-CH =0 88%
2 1 (CH2)20H=CH2 H. OR "100%"
g 1 (CH2)2C(CH3)=CH2 CH3, OR "100%"

Figure 8: Some Preliminary Studies from the Laboratories of Speckamp

of formate from the equatorial position and cyclization gives a diasteroemeric
mixture of 6 and 7 (2:1, ca. 100%) when R’=H (entry a), but only one isomer, 6,
(ca. 100%) when R'=Et (entry b).23

The final point to be discussed is enamide formation/dimerization. Enamides
can be protonated to give N-acyliminium ions, which might cyclize; but if there
are other problems with the cyclization such as unfavorable ring size or a
terminator that is not sufficiently nucleophilic, enamides or enamide dimers may
be the only products isolated. For example, in the cyclization of 8 if R=Me a 7-
membered ring is formed in good yield however, if R: H then enamide
formation and dimerization becomes competitive (Figure 9).24 One might

assume that the competitive dimerization process is sterically inhibited when

 

R=Me. Next, three synthesis which demonstrate the preparation of three 3 types
of target ring systems will be presented. Only the critical N-acyliminium ion
formation/termination steps will be outlined. These will be discussed
individually with emphasis on the problems described above and on the relative

merits of each terminator employed.

 

1) NaBH4
N o 2) W 7 N R (5)

R3) trifluoroacetic acid

 

\
TMS
3. fl.
_4_a ”=1,R=H,92°/o
b n=2,R=H,91%
C n = 1, R: Br, 63°/o
O O
2 O
Reg/R4 Qzfi/OMe + WOMe
\ R R HOCO
R1 R3 OCHO R1 H Rh H
5. 5. 1

am a‘ 52 51 m
a H H 2:1 ca.100
b Et H 100:1 ca.100

 

U“U
/.
t5\.

: *6

Figure 9: Cyclization vs. Enamide Dimerization

t C:
2

:0

The first example is taken from the work of Evans who has used N-
acyliminium ions to prepare spirocyclic alkaloid325. As described in Scheme 1
Evans utilized the addition of a Grignard reagent to a cyclic imide to generate
his cyclization precursor. Thus, treatment of glutarimide in ether with MeMgl
followed by addition of the Grignard reagent derived from 1-bromo-4-nonene
provided both ketoamide 13 and carbinolamide 14 as a 1:1 mixture (61%).
Evans later found that he could avoid ketoamide formation if the Grignard
addition was performed in dichloromethane instead of ether. Having thus
obtained mainly the carbinolamide 14, it was subjected to HCOzH (48 hr) to
provide the desired Spiro-cyclic compound 18 (33% yield from glutarimide)
along with enamide dimer 15 (10%), enamide 16 (10%), 6,5-spirocycle 17

(20%) and epi-17 (10%). As can be seen, this route is troubled by relatively low

 

14

yields, and formation of unwanted side products; however it does concisely

construct a ring system that is amenable to transformation to

Senemeji Evans'25Approach to Spiropiperldine Preparation

c H; 0”
om B'M9(°H2>a/\/ 4 M + New
C C / m
0 Hz '2 NH2 0 C4Hg O

 

 

MgX
12 1.3 1.4.
HOOOH enamide
4 ——" dimer +
1_§ (10%) ]§_ (10%) 11 (30%, 2 : 1 B/a) 1g (33%, R = CHO)

perhydrohistrionicotoxin. Another potential drawback is that the terminator is
not very versatile, affording only a protected alcohol as residual functionality.
Speckamp's synthesis of anatoxin-a, published in 1986, will serve to
demonstrate the preparation of a bridged ring system (Scheme 2).26 Reduction
of 19 (NaBH4) in the presence of H2804 and EtOH provided 20 in 77% yield.
Elaboration of the side chain to the required terminator afforded cyclization
precursor 21 (83%). Treatment of 21 with HCI in MeOH at -50°C and gradual
warming to room temperature over 18 hr resulted in a mixture of 22 and 23 in
47% and 11% yield respectively. The mixture was exposed to DBN to provide
the enone 22 in 60% yield. This product was carried on to anatoxin—a. A
unique feature of this synthesis is the directed closure to the a-carbon of an
a,l3—unsaturated carbonyl compound as the nucleophilic terminator.

It may be interesting to compare the synthesis anatoxin-a using acyliminium

1 5
Scheme 2: Speckamp's26 Synthesis of Anatoxin-a

NaBH4
Q’QW EIOH. H’ 30/vi ROM
77°42

ooze: c025: 0025: o
1.2 22 2.1

EtOZCN O EtOZCN EtOz

0 CN 0 HN O
ll/eOH/HCI DBN __
\ + 60% \ \
Cl
22

anatoxin a

 

22 (47%) 2.3 (11%)

ion technology to the Rapoport sequence described in Scheme 3.27 Treatment
of 24 with POCI3 followed by HCI and MeOH did not produce the desired
enone containing bridged product. However, treatment of ketone, derived from
reduction of 24, under the same conditions did result in closure, generating
compound 25 in 50% yield. To facilitate completion of the synthesis the benzyl
group was removed by hydrogenolysis and the amine was protected with di-t-
butyl dicarbonate (45 %, 2 steps). In comparing the Rapoport and the
Speckamp syntheses a few points deserve comment. Speckamp was able to
use a terminator which incorporated the required double bond; and the N-
acyliminium ion initiated cyclization resulted in a slightly better cyclization, and
overall yield. Finally; this serves as an example of the utility of the N-acyl
portion of the molecule serving as a protecting group both before and after a
cyclization.

The final ring system to be considered is a linearly fused ring system drawn
from the work of Chamberlin (Scheme 4).14 Treatment of cyclic imides 27a-e
with NaBH4 in MeOH at 0°C provided carbinolamides 28 a-e which were not

isolated but immediately treated with MsCl/TEA in CH2Cl2 to afford cyclized

 

1 6
Seheme 3; Rapoport'sz78ynthesis of Anatoxin—a

1) H A
/ 2) p683 Ph N 1) H2. WC t—BOCN HN
HOZC"“(|\>"'I W 2%] W "T "’ Q ]
k NY 50% O 45% O O ‘
Ph
21 25 25 anatoxin-a

products, 29a-e, in excellent yields. These compounds were then converted to
a variety of linearly fused natural products. The advantages of using
ketenedithioacetals as the terminating functions are: 1) a single regiochemistry
was observed, and 2) the functionality remaining was useful for completing the
desired synthesis.

It is known that a large variety of terminators, including olefins, alkynes,
allenes, dienes, vinyl silanes, propargyl silanes, and allyl silanes, as well as
aromatic, terminators such as phenyl, substituted phenyl, thiophene,
lndole,naphthalene, and imidazole can also be employed.2 From these (vide
supra) and other examples we can see that the requirements for a useful
terminator are: 1) regiochemical predictability in the termination step; 2) that
the terminator be sufficiently nucleophilic for the cyclization to take place under
mild conditions, and 3) that the terminator residue provide useful functionality
for completing the synthesis at hand.

Furans could complement the list of known terminators very nicely; provided
they prove to be stable to the reaction conditions and are sufficiently
nucleophilic. As a result of our experience with furans as terminators in cationic
cyclizations we expect to observe the desired cyclization product without any
regiochemical ambiguities. Secondly, the product furan, after certain well

precedented manipulations (Figure 2), should provide a more highly

 

W

functionalized intermediate than many of the previously utilized terminators.

Questions to be addressed in this study are: 1) can the required cyclization

Sehemfl: Chamberlin's14 Construction of the Pyrrolizidine, Indolizidine, and Related Ring
Systems

3
0 Gt 3 S
( m S/j NaBH4 j MsCl /
N \fs MeOH N\—(//Ks TEA ' ( m )n
)n )n N
o 0 o
21 2.3
m °/_°

m
H
N

r
QNANA p0
(D
m

@QOU‘N
ANN—LA

precursors be readily prepared, 2) will these furan containing compounds
withstand the conditions needed to generate N-acyliminium ions, 3) can we
observe both the electronically preferred furan 3-2 closure as well as the 2-3
mode of closure, and 4) can the product furans be manipulated to form natural
product targets.

Three modes of closure could conceivably be performed by using furan
terminated N—acyliminium ion cyclizations. The proposed system consists of a
cyclic initiator, a furan terminator, and a tethering unit to connect the two.
Attachment of the tethering unit at different points on the initiating function will
result in the three skeletal types described in Figure 10. Thus, attachment at the
nitrogen of an imide derived N-acyliminium ion will provide a linearly fused

system, such as in A. Likewise attachment at the carbon adjacent to the

 

18

nitrogen in a similar system, B will provide a spirocyclic system; and as in C,
attachment at the a'-site of an N—acyl-lactam derived acyliminium ion will afford
bridged systems as described in Figure 10, C. The length of the tethering unit
will determine the size of the forming ring. Finally the furan may be appended
to the tethering unit at either the indicated 2- or the 3-position.

Our previous work in the carbocyclic series can lend some insight as to how
this endeavor may proceed. Six-membered rings have proven to be the easiest
to prepare; five-membered rings have not been prepared under any conditions.
We have constructed 7-membered rings via a furan terminated cyclization, but
often in low yield. In terms of the skeletal types formed, linearly fused systems
have generally been the easiest to form, followed by Spiro-cycles and finally the
bridged ring systems. A final critical observation is that the electronically more
favorable 3 to 2 closure has been observed to be a more general mode of
closure than the alternative furyl 2- to 3- cyclization.

The design of our substrates will be unique for each mode of closure (Figure
10). The preparation of linearly fused and spirocyclic ring type precursors,
without the requisite furyl teminator, are precendented in the literature. A well
established route to the systems needed for the linearly fused closure is found
in addition of an electrophile containing the teminator to an imide, which later
serves as the initiator (see Scheme 4). For the linearly fused systems we can
envision employing this strategy by coupling a furyl containing tethering unit
with the nucleophilic nitrogen of an imide. There is also a clear precendent for
the preparation of a spirocyclic type precursor in the work of Evans (Scheme 1).
Unlike the first two modes of closure there are several possible approaches to
the design of the bridged ring forming precursors, these will be discused in

more detail in later sections of this thesis.

 

 

+
N

“101:0

B

 

/

C

 

)n
N
.l / \
O

)n )n
0 K/ )m N )m
H — H _
\ 0 \ 0
)n )n
+/
N
R
m( / m( \
O l 0

Figure 19: Basic Structural Possibilities for Furan Terminated N—Acyliminium Ion Initiated
Cyclizations

Tar e ructure

The alkaloids present a large array of skeletal and structural types28. Among
these are the linearly fused systems of lupinine (31)14.29, and elaeokanine A30
(30), the bridged alkaloids cocaine (34)31 and anatoxin-a (35)26132, and the
spirocyclic histrionicotoxin (32) and perhydrohistrionicotoxin (33) (Figure 11).33
These alkaloids have been the target of intense synthetic interest because of
their challenging structural features and because many of them display intense

biological activities that has rendered them useful, among other things, as

 

20

biological probes. Many elegant approaches to these types of ring systems
have been developed, however methodology which allows the concise
preparation of a variety of ring systems from common intermediates is always
desirable.

We have presented arguments which suggestthe utility of furan terminated
N-acyliminium ion cyclizations, for the construction of a variety of ring systems.
We should have access to the linearly fused ring systems present in lupinine,
the Spiro-cyclic system of the histrionicotoxins, and the bridged ring systems of
the type found in cocaine and anatoxin-a. Upon closer examination of these
structures, it becomes clear that in order to obtain appropriately substituted,
synthetically useful intermediates we will need to use the 2 to 3 closure mode.
Assuming that we are able to observe these cyclizations, the furan will provide
residues at positions in these alkaloids where functionality is needed. However
these residues will need to be modified. Crucial to our synthetic strategy is that
the furans be easily and selectively manipulated to form the desired natural

products.

Lin rl F m : L Inln n Ela kanine-A

Initially we directed our efforts the relatively simple linearly fused indolizidine
and quinolizidine alkaloid skeletal types. It was hoped that we could routinely
access fused 5-6, 5-7, 6-6, and 6—7 membered ring systems using either the 2 to
3 cyclization mode or the relatively more reactive 3 to 2 closure mode. We
anticipated preparing the required cyclization precursors by Mitsuobu
coupling34 of appropriately substituted furyl alcohols to either succinimide or
glutarimide.35 Subsequent NaBH4 reduction should provide the desired N-
acyliminium ion precursors.” The first sequence examined is outlined in

Scheme 5. This study was designed to provide some insight on how readily the

 

21

furan precursors could be made, what cyclization conditions would be most
ideal and how well the furyl moiety would survive the reaction conditions.
Mitsunobu coupling of the readily available 3-(2-furyl)-1-propanol36 to
glutarimide provided 38d in 53% yield. Reduction of the imide according to the
procedure of Chamberlin (NaBH4, MeOH, -4°C) afforded the corresponding
carbinol amide in good yield (95%). Carbinol amide 39d was then subjected to
the cyclization conditions we had successfully employed in our sequences with
allylic alcohols and enone initiatorsla-lc. Exposure of 39d to a two phase
mixture of HCOgH and cC6H12 resulted only in the destruction of the starting
material. This led to an extensive examination of alternative cyclization
conditions to no avail. Submission of the carbinolamide to MsCl, TEA
according to the procedure of Chamberlin14 or to PPTs led to the corresponding
enamide; THF/HOAc,THF or rexyn 300 led to destruction of starting material;
with alumina as the acid catalyst some starting material was recovered but in
overall poor mass balance. Derivitization of the carbinol amide as its
acetate35,and activation with an238 failed as well. We also attempted to cyclize
the derived enamide, since some workers have reported success with this
approach.36e We treated the enamide with HCOgH-CC5H12, with
Hg(C020F3)2, and with NBS, and in each case observed destruction of the
starting material with no trace of product detectible by NMR. Hart has
accomplished cyclizations using N-acylimino radicals, derived from on-
phenylthiolactams, as initiators.35f We were able to prepare the phenyl thio
derivative corresponding to 39d, but upon treatment with nBU3SnH, and AIBN,
we obtained only recovered starting material with overall poor mass balance.
We had previously encountered such seven-membered, ring forming
problems in the carbocylic series, and there have been reports of similar

problems in the literature.‘2 As discussed earlier, Speckamp has succeeded in

 

22

 

/ _
%
elaeokanine-A lupinine histrionicotoxin
1Q 3.1 3.2

/7 COgMe O

HN
NH HO MeN
I OCOPh \

perhydrohistrionicotoxin cocaine anatoxin-a
3.3 $3.4 ~15

"-0

Figure 11: Potential Alkaloid Targets

preparing seven membered rings only when the beta position of the carbinol
amide is blocked with two alkyl substituents (Figure 9).24 Based upon our
previous experience, we did not expect to encounter such a bias in the
electronically more favorable 3- to 2-furyl closure, or in the six membered ring
forming reactions.‘ Investigation of this surmise was elegantly carried out by
co-worker Jeffery Raggon.

To examine the cyclization in the six membered ring forming sequences, we

 

23

prepared the requisite imides as described above (Scheme 5). Treatment of 2-
(2—furyl)ethanol 37 (m=1)<38 with either succinimide or glutarimide in the
presence of DEAD and Ph3P provided N-substituted imides 38a and 38b in
31% and 51% yields, respectively, after chromatography. Reduction of 38a and
38b, according to the procedure of Chamberlin (NaBH‘4, MeOH, -4°C)14,
provided the corresponding carbinolamides 39a and 39b in 88% and 95%
yields, respectively. Exposure of 39a to a two-phase mixture of anhydrous
HCOzH and c-CeH12 for 2-3 minutes gave the desired indolizidine alkaloid
precursor 40a in 70% yield. The reaction time (2-3 min.) was found to be
crucial as lengthening of this period (5-10min.) caused a substantial reduction
in yield and a poor mass balance. The isolation of good yield of 40a is
noteworthy in that it is but our third exampleld.e of the previously unknown and
relatively disfavored 2-substituted -to-3-furyl cyclization. Similarly,
carbinolamide 39b afforded quinolizidine precursor 40b in 71% yield after
purification by chromatography. As might be expected, the remaining seven
membered ring forming possibility represented by 400 also failed to cyclize

We had yet to demonstrate that seven membered rings could be prepared.
Therefore, we prepared the 3—substituted furan containing imides 42a (100%),
42b (100%), 42c (41%), and 42d (56%), from glutarimide or succinimide, and
2-(3-furyl)ethanol 41 (m=1)39 or 3-(3—furyl)propanol 41 (m=2) and subjected
these materials to the standard reduction and cyclization conditions (Scheme
6). The yields of product carbinolamides were uniformly high, and, to our
delight, all of these substrates provided good yields (66%, 71%, 50%, 67%) of
cyclized products 44a, b, c,and d after brief treatment with HCOgH/CC6H12.

At this point we had answered the first two questions posed above (vide
supra). We were able to prepare the required starting materials and show that

the product furans were stable to the reaction conditions. Next we needed to

 

24
Scheme 51 An Approach to Linearly Fused Alkaloids

O Q \
( w PhaP,DEAD \ o
n —— nI
NH + H0 m o THF N
O o . m
3.6 n 31 m 31 n m 16 °
1,2 1,2 a 1 1 31
b 2 1 69
c 1 2 51
d 2 2 53
HO

m

 

H \
NaBH4, MeOH ( HCOOH
-400 n N \ O CCSH12
O

3.2 n m l “V 4_Q n. m I %
a 1 1 88 a 1 1 70
b 2 1 95 b 2 1 71
c 1 2 94 c 1 2 0
d 2 2 95 d 2 2 0

demonstrate that the linearly fused ring systems prepared by this method
provides intermediates that are useful in natural product synthesis. We chose
the alkaloid lupinine as our first target.

Lupinine (31) is a quinolizidine alkaloid originally isolated from yellow lupine
seeds (Lupinus luteus.) by Cassola in 1835.41 It was obtained in pure form by
Baumert in 188142, and posseses little biological activity.28 Many routes to
lupinine and and its more stable epimer epilupinine have been published.29 It
has been reported that if care is not taken to avoid epimerization epilupinine is
the isomer that is most often isolated. One synthesis of epilupinine that has

been mentioned previously is that of Chamberlin.14 In this sequence (Scheme

 

25

7) the cyclized material 29 is deprotected with HgClg in MeOH and aqueous
HCIO4 to give the ester 46. Treatment of 46 with LAH completes his synthesis

Senemej; Linearly Fused Furan 3- to 2- Closure

O
O
0 MG Ph3P, DEAD ( \ I
"I + H0 m / THF " N
o O
3.6 n 5.1. m 9.2 a :11 ed %
1 2 1 2 a 1 1 100
' ' b 2 1 100
c 1 2 56
d 2 2 84
HO O \
H O \
NaBH4, MeOH n( \ l HCOOH n(
—4°c N €06th N im
0 m o
4.2 n m u n .m. ield %
a 1 1 a 1 1 66
b 2 1 b 2 1 71
c 1 2 c 1 2 so
d 2 2 d 2 2 67

of epilupinine (epi-31, 75%, 2 steps).

A synthesis that employs an approach other than N-acyliminium ions is that
of Tufariello described in Scheme 8.29b Tufariello utilizes a nitrone
cycloaddition to prepare the 6-6 membered linearly fused ring system. When
methanesulfonate 48 (prepared in 5 steps from 3-buten-1-ol) was exposed to

2,3,4,5-tetrahydropyridine-1-oxide in toluene at 0-5° C for 60 hrs., the salt 50

 

26

was isolated in 74% yield. After reduction of 50 with Zn and acetic acid (80%)
the product alcohol was dehydrated with POCI3 in pyridine to give the
unsaturated ester 52 in 75% yield. 52 has been carried on to lupinine (31) by
reduction of the double bond with H2/P12040, then reduction of the lactam with
LAH.41 Alternatively ester 53 can be epimerized with NaOMe in MeOH and
reduced with LAH to provide epilupinine (epi-31).

Sentinel; Chamberlin'sl“ Synthesis of (+/-)-epi-Lupinine

w qJW ch

 

CF 0&3:

KN

S S
l HCOZMe HCHZOH
ch12, MeOH : = = =

leCI aq HCIO4

IE; A, 2hrs THLAFH A

79% 75%

O
23 3_1
epi-Iupinine

Having the methodology to form the 6-6-membered fused ring system, our
next step toward a total synthesis is to transform the furan into the functional
groups present in lupinine. Referring back to the structure of lupinine (31), w
see that the furan provides an alkyl residue at an appropriate location on the 6-

6 membered ring system and an oxygen residue that we need to remove; we

 

27

must also consider a C-C bond cleavage as furan 40b possesses one more

carbon than does our target.

5% Tufariello's29b Synthesis of Lupinine

 

C302Me 002Me
l + —. _.

@MOO ”\o

OMs OMs
$1 $3 5.2

C02M6
Zn/HOAc POC'a
80% 75% '
§.Q 5.1.
002Me H002Me {

Towards this end the crucial oxidative cleavage of the 2,3-disubstituted furyl
moiety was examined. Numerous methods for transforming a wide variety of
variously substituted furans into their corresponding keto-enals (54) or
butenolides (55) have been reported (eq. 7). Of these methods, we initially
investigated oxidation with MCPBA in CHQCIQ buffered with NaHCOg46 or
unbuffered42v43, the chromium VI-based reagents (PCC44 and variants, such as

2-CNPCC45; and the more classical CIauson-Kaas oxidation, Bra in buffered

 

28

CH30H45), followed by hydrolysis of the intermediate a,a'-dimethoxy-dihydro

furan derivative.42y450v47

[0] or

 

In the event, the readily available quinolizidine precursor 40b was subjected
to the oxidation methods mentioned above. Thus, treatment of 40b with
MCPBA under a variety of reaction conditions (2.2 equiv., CH2C|2, 0° C to
reflux42,43; 2.2 equiv., NaHCOa, 0° C to reflux42; 2.2 equiv,. NaOAc, HOAc)
followed by reductive (NaBH4) workup48; and finally, 2.2 equiv.MCPBA,
CH2C|2, 0° C to 25° C followed by trifluoroacetic acid (TFA) quench42, led only
to recovery of the starting material, or a number of unidentified products with
overall poor mass-balance. Similarly ineffective in oxidizing the furyl residue of
40b was PCC, CHgClg, 25° C to reflux and the more reactive 2-CNPCC,
CH2CI2, 25° C to reflux.“ Clauson-Kaas oxidation (Brg, NagCO3, MeOH, -30°
C) of the indolizidine precursor 40b did provide the corresponding a,a"-
dimethoxy-dihydro derivative in 77% yield45 (eq. 8); however, we were unable
to isolate the presumably formed keto-enal upon acid hydrolysis of the crude
reaction mixture using a variety of known methods (i. 1% aqueous HOAc, A473;

or ii. 1N HCl, H20, 45° C473; or iii. 2% aqueous H2804, 25° C).

 

29

 

OMe
O / O
Bra, MeOH OM 8
N NaZCO3 N e < )
-30° C
o 77% o
4_0_l2 5.5

Other methods that have been successfully employed in oxidizing related
substituted furan systems, but which failed to oxidatively open the furyl residue
in 40b, were NBS, NaOAc, dioxane-HQO, followed by NaBH4 reduction“9 and
Ce'V(NH4)2(N03)5, H20-CH3CN, 25° C50. We can safely conclude after this
extensive examination of chemical oxidants that furan cleavages are non—
standard operations which are extremely substrate dependent. Since all
standard and even esoteric methods for oxidizing the furyl moiety of substrate
40b proved fruitless, efforts were directed towards a photochemical means of
achieving this necessary transformation. The use of photochemically generated
singlet oxygen to oxidize variously mono- and di-substituted furans has
received considerable attention over the years.51 Treatment of 40b with 102,
generated by bubbling oxygen through solution of substrate in CH30H or
C H 20l2 and either rose bengal52, hematoporphrin51 or
tetrahydroporphrin51y53, as sensitizers at 25° C using either a medium-pressure
Hanovia lamp or a 500W Tungsten filament source, failed to provide any of the
desired products and, in general, resulted in poor mass-recovery.
Consequently, the temperature at which the photolysis mixtures were quenched
with reducing agents, including NaBH4 in MeOH or i-PrOH54 and Ph3P52f153b
was decreased. In these low-temperature photolyses, only recovered starting

material was observed with no traces of oxidatively cleaved photoproducts,

 

30

such as butenolides or keto-enals, detected (eq.7). The failure of the standard
chemical and 102 oxidations, thus far examined, caused us to consider the
alternatives outlined below.

Based upon our previous experience in oxidizing furans55 and the studies of
others53cv56, we elected to increase the nucleophilicity of the furyl moiety in the
cyclized substrate 40b by introducing a TMS group at the unsubstituted-od-
position. Following a procedure developed by German workers, who
successfully silylated analogous pyrrole systems using Et3N and TMSOTf at 5°
C to 25° C57, we exposed furan 40b to TMSOTf and Eth. After a number of
attempts, we failed to obtain any of the desired C-silylated furan. In fact, it
appeared from a cursory examination of the EI-MS and 1H-NMFt (250MHz)
spectra that the lactam moiety had been silylated; a surmise which was
substantiated by treating the crude silylated mixtures with K2CO3 in MeOH
leading to recovery of 40b. Alternatively, the silyl group could be introduced
intact on the furyl piece prior to the Mitsunobu coupling reaction as is outlined in
Scheme 9.

The coupling of 2-(5-trimethylsilyl-2-furyl) ethanol 5758 with succinimide
using the Mitsunobu procedure (DEAD, Ph3P)34 provided imide 58 in 87%
yield after chromatography. Reduction (NaBH4, MeOH, -4° C)14 yielded
cyclization precursor 59 in quantitative crude yield. Attempted cyclization of
carbinolamide 59 using a variety of conditions (e.g., i. HCOQH, CC6H12; ii. 1N
HCI, CH2CI2; MsCl, Et3N, -23°C to 25°C) gave only desilylated cyclized
material 40b.

We had encountered similar difficulties in furan elaboration during our
formal total synthesis of (+/-)-aphidicolin. That problem of furan oxidation was
overcome by transforming the substrate 2,3-disubstituted furan to a 2,3,5-

trisubstituted furan; oxidative cleavage of the substance was then smoothly

 

31

Scheme 9: The Examination of a Silyl Furan as a Terminator

0
DEAD o o TMS
87%. _
5.1 5.8
’0
CCthz
5.6 5.0.9

accomplished”. Because of our concern for the survival of this system under
the strongly basic conditions required to introduce a CH3- onto the cyclized
40b, we elected to employ the strategy depicted in Scheme 10 and utilize a
furyl piece with the requisite group already in place.

Therefore, the required imides, 62a (n=1, 83%) and 62b (n=2, 55%), were
prepared from succinimide (n=1) or glutarimide (n=2) and 2-(5-methyl-2-furyl)
ethanol 60.59 Reduction provided crude carbinolamides 62a and 62b in
quantitative yields. Subjecting these two substrates to the standard cyclization
condition (HCOgH, cCeH12, 2 to 3 min.) afforded diones 63a (n=1) and 63b
(n=2) in 35% and 64% yields, respectively, as mixtures of epimers. This
observation is unique in that the spontaneous hydrolysis of the putative furan
intermediate is essentially unprecedented. Our experience in the parent
systems and in a related perhydrohistrionicotoxin spirocyclization1dve, which
will be discussed later, suggests that the serendipitious hydrolysis might result

from the presence of an sp2-hybridized nitrogen in the forming cycle and

 

32

perhaps might be related to torsional60 and or allylic strain61. We can
demonstrate (eq. 9) that there is nothing intrisically wrong with the furan
intermediate by submitting 62a to Chamberlins' solvolytic conditions (MsCl,
TEA) to give 64 in 81% yield. When 64 is dissolved in cCeH12 and HCOgH
containing 1 eq. of water, we obtain the previously observed dione 63a in 62%
yield.

With quinolizidine 63b in hand, we next examined its conversion to the
relatively simple naturally occurring alkaloid lupinine 31 (or epi-lupinine). To
complete a synthesis of lupinine we needed to reductively remove the lactam
and the "ring" carbonyls, and transform the "side chain" into a methanol
residue. We chose to leave the lactam carbonyl intact until the latest possible
step to avoid exposing the amine to harsh reaction conditions. One problem
that is immediately obvious is distinguishing between the two ketone functions.
The ability to selectively protect one of the ketones and perform manipulations
on the other unprotected ketone is crucial to the successful completion of the
synthesis.

We began our investigation by examining various ketalization conditions. It
was found that submission of 63b to a modification of Noyori's kinetic
ketalization conditions (excess TMSOCH20H20TMS, TMSOTf, CH20I2, -78° to
RT)‘52 provided a mixture of the ring ketal 65 and the bis-ketal 66 (10:1 to 1:1,
78-88%). The predominant formation of the ring ketalized material 65 was
demonstrated when the reaction mixture was submitted to NaBH4 reduction.
Two products, that were easily separable by column chromatography, were
obtained. The 1H—NMR (250MHz) spectrum of the major product ketal-alcohol
(lower Rf by tlc) revealed that the methyl singlet of the starting material, now
appeared as a pair of doublets, indicating that the major product under these

conditions was hydroxy ketal 67, prepared as a mixture of stereochemistries at

 

33

the carbinol center. Examination of the 1H-NMR (250MHz) spectrum of the
minor product indicated that the singlet corresponding to the methyl group had
migrated upfield relative to the starting dione, 63b. This observation would be
consistent with bis-ketal structure 66. Our presumption was substantiated by
El-MS. We found that by carefully limiting the amount of TMSOCHgCHgOTMS
employed in the reaction (one equivalent) that we could prepare 65 almost

exclusively (>98:<2, 93%)

Scheme 19: Second Generation Approach to Lupinine and lndolizidine Alkaloids

O O \
N-l + HO 0 THF N
o o

11:12 5.0. 5.1 n Yield—L221
a 1 83
b 2 55
O
HO
H \ 0
w. .1 N \ O mom .1
4°C cCsH12
O O
5.2 11 fl nibiLli/d
a 1 a 1 75
b 2 b 2 64

HO

H \ \ O
\ O MsC|,Et3N (9)
N T
O

123 5.4

34

TMSO(CH2)20TMS
TMSOTf, CH2Cl2
-24° c to RT

 
  
 

OO

HO(CHZ)2OH. stOH

4
benzene, A L+§i+§§

Om

attempted
micketalizau'on

S] S
0
$5 6§W§¢§¢§
+ 8+ 3+ + 3
O O O O O
54 1a 5.2 7.9 11

o
O O] OH
0‘3 NaBH‘I 0’»
0+ O W 0+
on on 011

(10)

(11)

(12)

Exposure of 63b to the "standard" ketalization conditions (HOCH20H20H,

TsOH, PhH, A)53 afforded a mixture of 64, 65, 66 (eq. 11). Knowing that the

ring oxo-ketal was readily available we surmised that if we could prepare the

ring thioketal (68) in an analogous manner, and if that thioketal could be

reductively removed with Raney nickele4 we would be in a position to complete

the synthesis by Baeyer-Villager oxidationfi5 followed by treatment of the

derived acetate with LAH14.

Treatment of dione 63b under thermodynamic thioketalization conditions (i.

HSCHgCHgCHgSH, stOH, refluxing benzene ii.

HSCH2CH28H, stOH,

35

refluxing benzene)65 led to furan 64. Noyori's kinetic ketalization procedure
(TMSSCHgCHgSTMS, TMSOTf, CH20|2,)52-67 provided an inseparable
mixture of products. Elucidation of the structures was not possible by
examination of the NMR. An examination of the 1H-NMR (250MHz) spectra of
both crude and chromatographed material indicated the presence of a methyl
group adjacent to a carbonyl group and a methyl group adjacent to a ketal.
Possible products include furan 64, ring ketal 68, bisketal 69, side chain ketal
70 or bicyclic ketal 71. We attempted to modify the reaction conditions to
provide the desired ring thioketal 68 to no avail. Thus exposure of dione 63b
to various thioketalization conditions resulted in either the formation of furan 64
(i. HSCchstH, BF3-OEt, CH20|229C; ii. TMSSCHgCHgSTMS, ang,
CH2CI257; iii. HSCHgCHZSH, Nafion, benzene58), a mixture of products (i.
TMSSCHgCHZSTMS, TMSOTf, CH20I262; ii. TMSSCHgCHgCHgSTMS,
TMSOTf, CH2C|266; iii. HSCHgCHgSH, polystyryl catalyst, CH3CN59; iv.
HSCH20H20H28H, polystyryl catalyst, CH30N59; v. TMSSCHgCHgSTMS,
ZnOTf, CH2CI270; vi. TMSSCHZCHZSTMSZnClz, CH2CI2), or recovered
starting material (i. t-BuSH, TMSCI, CH20l271; ii. TMSSCHgCHgSTMS, Mngz,
CH2Cl2; iii. TMSSCHgCHgSTMS, Mg-OEtg, CHQCIQ, iv. 2-phenyI-1,3,2-
dithiaborolane, SHCHgCHzSH, CHCI372).

In view of these problems, and our success with preparing oxo-ketal 65, we
elected to examine the route outlined in Scheme 11. Baeyer-Villager
oxidation65 of the ring oxoketal should provide acetate 73, which might be
transketalized to thioketal 74. Raney nickel reduction, and treatment with LAH
should provide either (+/-)—Iupinine or (+/-)- epi-lupinine

In the event, Baeyer-Villager oxidation of 72, using freshly prepared
trifluoroperacetic acid in CH2Cl2 buffered with anhydrous NagHPO455b,

provided the acetate 73 in 54%

 

36

55mm Synthesis of (+/-)-epi-Lupinine

0 OAc
0 WCHzizoms 0’) - TFPAA ~01}
Na2HPO4
-24°CtORT2 CH20I2
0
5.3.12

 

93% 54%

OAc

S/j
HSCHZCHZCHZSI THE A
BFSUE'Q- CREE: F“
60%;

87%

 

H epi—§_1_

yield. In an attempt to improve the yield of this step, other oxidative conditions
were examined. More mild conditions provided only recovered starting material
(i. TMSOOTM855d; ii. permaleic acid55k; iii. MCPBA), while more harsh
conditions using Lewis acid catalysts (BF30Et, H202559) resulted in
deprotection of the ring ketal and a variety of products. This seems to be a
problem inherent in the system, since conditions that will promote the Baeyer-
Villager oxidation will also promote deprotection of the oxoketal. However, a
recent report by Hart suggests that, even with a methylene unit in place of the
oxoketal, the yield may not be significantly improved73. Equation 13 illustrates
the Hart example.

Continuing the synthesis as outlined in Scheme 11, transketalization of the
oxoketal 73 to the thioketal 74 (HSCH2CHgCH28H, BF30Et, CH2C|2)

proceeded in 87% yield.74 Reductive removal of the thioketal (Raney nickel,

37

EtOH)70 provided the acetate 75 in 60% yield. Treatment 0175 with LAH led to
a compound (nearly quantitative crude) whose spectral data was similar to the
published values for (+/-)-epilupinine.14. Believing that we have successfully
accomplished the preparation of a linearly fused alkaloid 1i_a_ furan terminated
cyclization, we have demonstrated that the furyl residue left after cyclization

provides a convenient "handle" for completing alkaloid syntheses.

AA ——> isoretronecanol eq. 13
CHZCIZ
NazHPO4
56% 15

   

Soiro-Cvclic Alkaloids: A Formal Total Svnthesis of (+/-)-
P r ' ' i

The tropical "arrow poison frogs" of the genera Dendrobates have yielded a
host of structurally unique alkaloids.75v75 The C-19 alkaloid histrionicotoxin 32
and its non-natural hydrogenation product perhydrohistrionicotoxin 33, which
possess a novel azaspiro[5.5]undecane ring system have been the target of
intense synthetic interest for over ten years.77r33 The attention directed toward
32 and 33 is due to their unusual structures, their potent biological activities,
and their low natural abundance (ca. 200 pg HTX per frog); both alkaloids
block post-synaptic membrane depolarization.78 The realization that the
structurally more simple perhydrohistrionicotoxin possesses nearly the same
biological activity as histrionicotoxin has led to the former being a common

target for synthesis.

38

Recently Kishi culminated nearly a decade of effort with the first total

synthesis of (+/-)-histrionicotoxin itse|f79; however 34 remains a target of

interest. Kishi's total synthesis of 33 has established that lactam 83 is a useful

precursor to perhydrohistrionicotoxin (Scheme 12)33°1d.

Kishi treated 2-butyl-1,3—cyclohexandione (79, available in 2 steps from

methyl 4-(chloro-formyl)butyrate) with acidic ethanol followed by vinyl

magnesium bromide to provide 80. Compound 80 was subjected to i. NaOMe,

NHgCOCH2002Et; ii. aqueous NaOH; iii. aqueous HCI; iv. 100° C, dioxane

to afford 81 in 45% overall yield. lntramolecular Michael addition provided the

two epimers (82:83) in a 1:2 ratio (100%).

W Kishi's Synthesisaacvd of (+/-)-Perhydrohistrionicotoxin

 

 

o COzMe
I Na—<
1) EtOH,H’ 1) NaOMe. COZNHZ
2) aq.NaOH — _ _’
O ”MQB' C4H9 a) aq.HCl
O 4) 100°C
45%
12 5.9
HC(OEt)3, EtOH NaOMe g2; 53 Li/NHa
WW MHZ + NH O ‘CRQCIT’ 4 :1 50%
100%
O 1 :2 O
12 £3
1) DHP
1) Aozo_pyr 2) Meenivein reagent 9 AIH3
——_. -—-—fi—u— —.
2) P255 NH OH 3) 3 HL' NH OTHP 63.70%
3) OH‘ 4) H
90% S
5.5 §§

NH OTHP

39

Epimerization of the mixture with NaOMe in MeOH afforded a 4:1 ratio of Q
and 83 which were reduced with Li/NH3 to provide the desired orientation of
the hydroxyl group (84, 50%). Formation of the thiolactam (1) ACQO, pyr; 2)
P285; 3) 'OH, 90%) 85 followed by preparation of the imine (1. Meerwein's
reagent; 2. CH3(CH2)4Li) afforded 86. Reduction of the imine with AIH3 (70%)
completed the synthesis of perhydrohistrionicoton.

Evan's approach to perhydrohistrionicotoxin using an N-acyliminium ion
cyclization was discussed previously (Scheme 1).33e We can see that the with
the preparation of cyclized compound 18 all that remains to complete a
synthesis of perhydrohistrionicotoxin is attachment of the 5-carbon side chain.
Evans accomplished this in a manner analogous to Kishi's approach. While
Evans cyclization product is functionalized very well for a synthesis of
perhydrohistrionicotoxin, his approach would not be amenable to a synthesis of
histrionicotoxin itself.

As is outlined in Figure 12, furan terminated N-acyliminium ion cyclization of
89 should provide the spiropiperidine 88. Furan elaboration could conceivably
furnish an intermediate which might be utilized in the synthesis of either
histrionicotoxin or perhydrohistrionicotoxin. The aldehyde could provide a
"handle" for attachment of the eneyne side chain of histrionicotoxin, while
addition of a two carbon fragment at the position of the aldehyde should afford
Kishi's perhydrohistrionicotoxin intermediate 83.

Our first attempt to construct spirolactam 88 is described in Scheme 14.
Reaction of glutarimide with CH3Mgl followed by the Grignard reagent prepared
from 3-(2-furyI)-1-bromopropane80, by the procedure of Evans33e, afforded the
carbinolamide 89 in quantitative crude yield. Without purification 89 was
immediately exposed to the two-phase mixture HCOgH-CC5H12 (3 minutes)

which provided spiropiperidine 88 in 55% chromatographed yield.

40

R' R.
OH
\
NH OH: NH OH —_—> NH\ O 3 NH —
1:: ‘ o o o \ O
8.1 u 89

perhydrohistrionicotoxin

or
histrionicotoxin

Figure 12', A Possible Approach to (+/-)-Perhydrohistrionicotoxin

§eheme 13: First Generation Spiropiperldine Construction
0
/

/

$0 1)CH_3_______.MgI HCOzH. cCeH12
NH 0., NH

2)Bng\/\/O 55/
O O

Q

With the desired cyclization substrate in hand we initially turned our attention
to the preparation of perhydrohistrionicotoxin. As outlined in Figure 13 we
envisioned oxidative cleavage of the furan, followed by reduction of the
unwanted double bond. Alternatively hydrolytic cleavage of 88 should provide
keto-aldehyde 91 directly.

We subjected 88 to a variety of oxidative and hydrolytic conditions to no
avail. We observed either no reaction or we obtained unacceptably low yields
and mass balances. For example, MCPBA under a variety of conditions (0.9 — 2
equivalents in CHgClg at 0°C to reflux42'43; 0.9 - 2 equivalents in a two phase
mixture of CH20I2 and saturated aqueous NaHCO342; 2.7 equivalents, NaOAc

and HOAc48b; 1 equivalent, followed by

41

 

3.8
\ hydrolysis

Figure 13: Possible Furan Elaboration Protocol

addition of NaBH443; 1 equivalent followed by addition of CF2CO2H) led only to
recovery of starting material or a number of unidentifiable products with overall
poor mass balance.

Similarly ineffective in oxidizing the furyl residue in 88 were a variety of
oxidation conditions such as PCC44, CNPCC“5 and CAN50; Br2, pyridine,
acetone, H2O; NaOAc, AcOH, MCPBA, Me2S48b; NBS in dioxane and H20
provided a complex reaction mixture in which an aldehyde proton was
observed by 1H-NMR (250MHz), however it could not be recovered in good
yield.49 Clauson-Kaas oxidation of 88 (eq. 15) did provide the can'-
dimethoxy-dihydro derivative in 86% yield; hoWever, we were unable to cleanly
hydrolyze this intermediate to the desired keto-enal under a variety of known
methods (i. H2O at reflux“; ii) dowex resin82; iii) 0.7N HCI in THF; iv)
HOCH2CH2OH, benzene, stOH83; v) 0.005M H2SO484; vi) THF, H2O,
CF3CO2H; vii) 1% HOAcin H2O47a; viii) 2% H2SO447C185; iv) ppts,THF,
H2O). Attempted hydrolysis of 88 itself (HOCH2CH2OH, TsOH83; HCl,

MeOH86) led only to recovered starting material. As had been described

42

during our adventures leading to lupinine, a furan unravelling sequence
assisted by Si or a halogen might be considered. These substrates might
obviate our oxidation problems; therefore, the 5-bromo derivative of 88 was
prepared (NBS, DMF, 68%)37, but it could not be oxidized with PCC44d. We
attempted the preparation of the 5-silyl derivative of 88 by silylating the furan

—2—*fi' 29‘3"” 04>
O N 32 3 O
H H
\ \ / OMe
O O
OMe
8.8 9.2

according to a published procedure which successfully silylated pyrroles with
Et3N and TMSOTf57. Unfortunately, as is the lupinine synthesis submission of
88 to these conditions did not result in silylation of the furan, but apparently
silylation of the lactam.

The silyl furan could conceivably be brought in intact, by using 1-bromo-3-(2-
furyl)-5-si|ylpropane80 in the Grignard coupling, as outlined in equation 15.
This proved to be the case as coupling of 93 to glutarimide according to the
procedure of Evans provided 94 in nearly quantitative crude yield. Submission
of 94 to the usual cyclization conditions led to loss of silicon and a variety of
products.

We have already encountered numerous and described problems in
oxidizing furans; alternatively we have also had difficulties in keeping the furan

intact (vide supra). To summarize, we have observed that trisubstituted furans

43

13.. DWMB.— OH H + We
0 0mg 0 9 o o TMS ~

H H o TMS
as 9.4 9.:

are much more susceptible to either oxidative or hydrolytic cleavage. With
these precedents in mind, we elected to incorporate the needed two carbon unit
into the starting furyl organometallic compound; hopefully obviating the furan
manipulation difficulties encountered previously. A successful formal total
synthesis of (+/-)-33 employing this modification is outlined in Scheme 14. In
the event, treatment of the N-Mgl salt of glutarimide with the Grignard reagent
prepared form 1-bromo-3-(5-ethyl-2-furyl)propane30 gave the corresponding
carbinolamide 97 in excellent crude yield33e. Without purification 97 was
cyclized as described above (HCO2H, cC5H12) affording spiro-piperidine 98
(72%) with the furan intact. The furan contained in 98 did indeed suffer smooth
oxidative cleavage (MCPBA, CH2Cl2, saturated aqueous NaHCOg) yielding 99
(70%) after reduction (H2-Pd/C; EtOAc-aqueous HOAc) of the rather unstable
ene-dione. The side chain and ring ketone carbonyl moieties must now be
differentiated, and the unwanted side chain oxygen removed. -

Toward that end we exposed dione 99 to thermodynamic ethylene ketal
forming conditions (HOCH2CH2OH, stOH, PhH, reflux) and to the kinetic
ketalization conditions of Noyori (TMSOCH2CH2OTMS, TMSOTf, CH2Cl2, -
78°C to RT)53-. Under "thermodynamic" conditions 99 was cleanly converted to
the starting furan 98 (90%); however the "kinetic conditions" of Noyori led
almost exclusively to "side chain" ketal. (>95:<5, 55%). Having realized

selectivity in carbonyl protection we investigated removing the "side chain"

44

ketone via kinetic dithioketalization and reductive removal. Treatment of 99
with TMSSCH2CH2STMSG7 and TMSOTf in CH2CI2 selectively provided 100
(>98:<2) in 67% chromatographed yield. Attempts to improve this yield were
unsuccessful, some conditions examined included (i. thionyl chloride, silica gel,

HSCH2CH2SH, toluene88; ii. HSCH2CH2SH, stOH, refluxing benzene66; iii.

$912M. A Formal Total Synthesis of (+/-)-Perhydrohistrionicotoxin

Et
\ O
OH
O
1) CHSWI HCOZH. CC6H12
g4 \Afl NH 72%

2 M
)Br 9 Et 0

9_7_

O
9.6

1)MCPBA, CHZCIZ
aq. NaHCOa
° H \ W
3\— 10% aq. HOAc, EtOAc
0 70%

TMSS(CH2)28TNS Raney Nickel
TMSOTf O EtOH, A
S
67 % H 78%
SJ

1.0.9.

HSCH2CH2CH23H, BF3-OE12, CH2C|264C; IV. TMSSCHQCHQCHQSTMS, ang,

CH30N68) the majority of which resulted in formation of furan 98.

45

Reductive cleavage (78%) was accomplished with Raney Nickel in refluxing
ethanol to give a product which was identical in all respects to published data.
A low ratio of catalyst to substrate was used to avoid the problem of over
reduction of the ketone to a mixture of diastereomeric alcohols (the alcohols
could be oxidized to 83 under either Swern89 of Jone390 oxidation conditions).

In summary we have described a concise (6 steps, 26%) preparation of
lactam 83, thus constituting a formal total synthesis of (+/-)-
perhydrohistrionicotoxin 33. The advantages of this annulative approach are
brevity and the regiochemical integrity of the crucial furan terminated N-

acyliminium ion initiated cyclization.

 

Cocaine (34) will be the first of two bridged ring containing alkaloids to be
discussed. Cocaine is an aza[3.2.1]bicyclic ring containing alkaloid that is
isolated from Erythroxy/an coca. Historically it has figured prominently in the
development of local anesthesia, but it is more notorious for its use as a
psychoactive drug. More recently because of its toxic, highly addictive nature its
medical use has been limited to topical application primarily in opthamology91.
Numerous analogs of cocaine have been prepared and found to possess a
wide variety of biological activities. Because of its challenging structural
features, and the need for analogues, preparation of the tropane ring system
still attracts considerable attention92.

Willstatter initiated work in the synthesis of tropane alkaloids by
accomplishing a synthesis of tropinone starting from cycloheptanone310. This
was soon followed by Robinson's efficient preparation of tropinone involving the
condensation of succindialdehyde, methylamine, and the calcium salt of 1,3-

acetonedicarboxylic acid93. Attempts to adjust this approach to a synthesis of

46

cocaine encountered stereochemical difficulties. More recently there have
been several synthetic entries into the tropane ring skeleton94. However, like
Robinson's synthesis, most fail to address the stereochemical aspects of
cocaine itself. Examination of the structure of cocaine reveals two stereocenters
that need to be controlled in any successful synthesis. The'oxygen functionality
at C-3 has been controlled by reduction of the corresponding ketone. The axial
C-2 carbomethoxy residue has proven to be more difficult; it is readily
epimerized in basic medium to the more stable equatorial position. Tufariello
has published a total synthesis of cocaine that successfully addresses these
problems.31a

Scheme 15 outlines Tufariello's31a approach to cocaine. 1-Pyrroline-1-oxide
101 reacts with methyl-3-butenoate to provided adduct 102 in 96% yield.
Oxidation (MCPBA) to 102 (89%), followed by further oxidation with MCPBA
and subsequent addition of methyl acrylate affords 104 in 77% yield.
Formation of 105 (MsCl, pyridine,95%), followed by treatment with DBN, to
effect dehydration, provided 106 (86%). Refluxing in xylene results in
spontaneous nitrone cycloadditon of 107 to provide 108 (66%) directly.
Addition of Mel (109, 58%), and treatment of 109 with Zn and 50% aqueous
HOAc afforded ecgonine methyl ester in 47% yield. Benzoylation as described
by de Jong95 completed the synthesis of (d,l)-cocaine (37%).

Our retrosynthetic approach to (+/-)-cocaine is outlined in Figure 14.
Cocaine may be available from furan 111, which could be derived from N-
acyliminium ion 112. Acetal 113, which might serve as the cyclization
precursor, could be available by reduction of nitro-olefin 114 and protection of
the product amine. Compound 114 could in turn be available by a
condensation between furfural and 4-nitrobutanaldimethylacetal 115. One

major drawback of the illustrated approach is that we might be unable to

47

mm Tufariello'531a Synthesis of (+/-)-Cocaine

Z a; vC02M_9. m MCPBA e,
toluene, A N‘O CHZE '
00 96% 002m 89% o 0
1.0.1 102 mcone
MCPBA MsCl. pyr ' I DBN, PhH
0112012 0° c to RT ‘ 86%
95%
”C02“; ”0 COzMe M50 0 002Me
0° C to RT COzMe 002Me
77% 1.0.4 19.:
xylene. A 6) N’0 CO2MG Mel
’ 66% ether/CHZCIZ
' l 0 58°49
\ 0 \ O H
002MB
COzMQ 002MB
1.0.6 1.01 1.01
9 o
\N’ COZMS CH3 COZMQ benzoyl chloride CH N C02M°
50% aq. HOAc OH anh. N32003 3 OZCCBHS
H ' 9 47% 37%
1.0_9 1.1.Q fl.
COzMe
MeN RN \ \ @/
OCOPh .
2 O :> \ C02Et :>
\
11.1 1.11
Cocaine

/ O
H
CH(0Me)2:> Women/Ia» :>
”1 N0
. 2
COR

\

0 14.3

/

QCHO + NOZ/VYOMG

OMe

1.1.4 m

Eigure 14', A Retrosynthetic Analysis for a Cocaine Synthesis

pinpoint our exact problem should the acyclic furan containing amido-aldehyde

fail to provide the desired bridged furan. Finally, if the oxidation/hydrolysis step

fails we could adjust our retrosynthetic plan to include a disubstituted furan

synthon from the outset.

In the forward sense, Michael addition of nitromethane to acrolein catalyzed

by BU3P followed by in situ dimethyl acetal formation (HC(0Me3)3, MeOH,

stOH96) provided 4-nitrobutanaldimethylacetal in 35% yield (Scheme 16).

The nitroaldol reaction between 115 and furfural, catalyzed by nBuNH297, gave

nitroolefin 114 (69%), which was reduced with LAH to afford an amine (86%),

which was not purified but immediately treated with ClCO2Et and TEA in THF98

to provide cyclization precursor 117 in 65% yield.

SEEM The Construction of the Cocaine Ring System

/\CH0

1) CH3N02, 1311313, PhH

 

2) HC(OMe)3, MeOH, stOH

OMe
WOMQ
No2

1.1.4

35%

LAH Et20 A
86%

NHCOzEt
MOMS
O

1J1

OMe

011013 ‘

 

O N CH0
2 MOMS nBSNH
2
69%:
1.1.5
ClCOZEt TEA THF
65%

Hi

CFacOZH CH0

EIOZCNR/
0

1J_3

62%

49

With the successful preparation of the cyclization precursor we next
examined several cyclization conditions. Treatment of 117 under a variety of
acidic conditions provided a plethora of products99; however we found that
trifluoroacetic acid/H2O (1 :1) added to a solution of 117 in CHCI3993 provided a
62% yield of cyclized product 118; a compound-in which the furan was not
intact. Based upon our previous results we were surprised to observe the furan
hydrolysis under the reaction conditions. Recall that we had, in several cases,
failed to oxidize or hydrolyze a disubstituted furan. One might guess that the
added strain which would accompany rearomatization, in this bridged system,
drives the intermediate toward trifluoroacetate capture, and ultimately to keto-
aldehyde 118 after hydrolysis upon workup

To complete a synthesis of cocaine we need to consider several factors. First
the stereocenters at C-2 and C-3, need to be controlled, and we need to remove
a one carbon fragment from the C-2 side chain. The correct orientation of C-2
carbomethoxy residue in cocaine (Figure 15) could conceivably result from
exocyclic protonation of the dianion 120. For this to occur we would need
access to hydroxy ester 120 with the C-3 OH oriented in an equatorial position.
Having the OH equatorially oriented, and in the form of an anion should reduce
the possibility of p-elimination during the enolization step. This surmise is
further supported by the driving force for the hydrolysis of the initial cyclization
product. We could envision obtaining 121 from alkene 122 by some oxidative
process such as ozonolytic cleavage. Alkene 122 could in turn be prepared by
selective dehydration of a primary alcohol which should be available by
reduction of the cyclization product 123. Since this retrosynthetic plan hinges
on being able to selectively reduce the C-3 ketone to an equatorial alcohol, we

began by examining this reaction.

50

Our first attempt at selective reduction utilized NaBH4 as the reducing agent.
Treatment of 118 with NaBH4 in MeOH provided a 4:1 ratio of two diols 124a

and 124b in excellent yield (84%). Since we had two carbinol centers present

' MeO

CO2Me / 0' M“

' RN RN _ +
m 1_2.9

COzMe RN CHO
j W
1_2.3

1

RN%'/ OH 3 RN

121 122

Flggre 15: A Possible Route to (+/-)-Cocaine

in the molecule, it was difficult to determine the stereochemistry at the C-3
hydroxyl. In order to prepare a compound whose stereochemistry could be
unequivocally assigned and investigate the rest of the sequence we chose to
carry this diol through the dehydration step.

It is well known that primary and secondary alcohols may be conveniently
eliminated by first transforming them into a selenide, and then treating the
selenide with H202 to induce elimination100. A survey of the literature
suggested that the selenide formation was selective for primary alcoholsim.
Upon treatment of the mixture of 124a and 1 24b with o -
nitrophenylselenylcyanate10031102 and BU3P in THF, two primary selenides

were isolated, 125a 64.8% and 125b in 24% yield. In order to protect the

51

secondary alcohol during the oxidation steps the corresponding acetates were
formed. Selenide 124a was stirred with pyridine, DMAP, and acetic anhydride
for 2 days to furnish acetate 126a (97.5%) Selenide 125b was stirred with

 

pyridine, and acetic anhydride overnight to furnish acetate 126b (100%). Both
126a and 126b, after being dissolved in THF and treated with 30% H202,

undenNent smooth oxidation and elimination to provide olefins 127a and 127b

Scheme 17; Cocaine Side Chain and Carbinol Manipulation

SeCN
No2
EtO CN
\0 NBOH OH Bu3P, THF, RT
1.1.9. a a—OH 67% m
b B—OH 17%
EtOZCN mga A—GZO- EtOZCN OSQQ Hzo2 1].":
PW Ac N02
1.2.5. L2§

 

EIOZCN / EIOZCN +E102CN EtOZCN CO:M9
‘OAc 2 3 0 Ac 23m OAc + Em

E

in 99% and 88% yields respectively. At this point it was clear, by comparison of
the 1H-nmr's of 127a and 127b to published spectra of the epimers of cocaine,

that our major product was the undesired axial alcohol.103

 

52

Anticipating that we would be able to adjust the hydroxyl orientation, perhaps
by a more judicious choice of reducing agents, we continued with the synthesis.
We were concerned that 127 might not withstand harsh C=C cleavage
conditions; therefore we sought the mildest cleavage conditions possible.
Toward this end, we submitted a solution of 127b in MeOH at -78° C to
ozone104, followed by oxidative work up with H202, We obtained a good yield
of a mixture of compounds that appeared to consist of aldehyde 128, acid 129,
and ester 130. In the final route we will need to cleanly obtain one of these
oxidation products, preferably the methyl ester since that is our final goal. It has
been reported that ozonolysis of alkenes in a solution of anhydrous HCI in
MeOH allows clean isolation of methyl esters.‘°4e Knowing that our compound
will withstand ozonolytic cleavage we returned to the problem of hydroxyl
stereochemistry.

Extensive studies on the reduction of tropinone have been reported. With
tropinone or 2—(carbomethoxy)-3-tropinone the most efficient reduction to date
has been accomplished using sodium metal or sodium amalgam in alcoholic
solvents31b1105. We attempted to apply this protocol to our keto-aldehyde
system. Treatment of the keto-aldehyde with Na in EtOH or Na in s-BuOH failed
to provide any trace of alcoholic products. A more modern method, that, like the
Na/EtOH reduction, results in formation of the more stable equatorial alcohol (in
the absence of severe steric hinderance) is reduction by Li/NH3.106 When we
treated a solution of keto-aldehyde 118 in THF with Li/NH3 we isolated instead
of the desired alcohols, two isomeric lactols 131 and 132 . We could imagine
the major product resulting from the reduction of the aldehyde to an alcohol and
attack of the alcohol on the ketone. Likewise, reduction of the ketone and attack

of the alcohol on the aldehyde would result in Iactol 132.

OH
Et02CN GHQ LVNH EtOZCN EtOZCN
_3- o + o (16)
O OH

1.1.0. 1.31. 1.12

Lewin and coworkers have recently examined the reduction of 2-
(carbomethoxy)-3-tropinone (134) extensively (Scheme 18).31b In an attempt
to improve the ratio obtained with the standard sodium amalgam reduction, they
examined NaBH4 in MeOH at -30° 0; H2_ Pd/C; K, HMPA; lithium tri-sec-
butylborohydride; and potassium graphite. They conclude that the sodium
amalgam reduction is still the most viable synthetic route from 2-

(carbomethoxy)-3-tropinone 134 to ecgonine methyl ester 110b.

Scheme 1§z Lewin's31b Synthesis of (+)-Cocaine

O CH3N _
1) NaH, (MeO)2CO sodium amalgam
2) resolution 2% aq. H2804
80%

 

CHSOZC O 50%-70%
L33 1_$_4
CH N CH N 00 M
3 3 ¥ 2 e benzoyl chloride, pyr .
+ o (+)-cocaine
OH OH 89 /o
M9020
1J.0.a 1.1.0.12

1:2 to 2:3

54

There is no reason to believe that a 2-substituted system such as ours should
behave much differently than tropinone. A possible source for our problems is
the fact that the 2-substituted group is an aldehyde. Considering the difficulties
inherent in this system we decided to attempt a formal total synthesis by
preparing the Lewin intermediate 134. Having 2-(carbomethoxy)-tropinone as
our ultimate goal, another factor we now had to consider is how to prepare the
N-methyl group present in this molecule. The carboethoxy group that we were
presently using as a protecting group can be transformed into an N-methyl
group by reduction with LAH107, however LAH is a very strong reducing agent
that will attack many other functional groups. Removal of such a group requires
either strong acid, strong base, or TMSl, all of which are harsh conditions and
probably not compatible with the present system98. We decided to switch to a
protecting group that should be stable to oxidative conditions, but will be easier
to remove. The carbobenzyloxy group most closely meets these requirements.

Another potential problem with the synthesis is the low yields in preparing
the cyclization precursor. This is usually most troublesome when the starting
materials are difficult to obtain or when the sequence is not amenable to large
scale synthesis. Neither of these cases are true, since our starting materials,
nitromethane, acrolein, and furfural are very inexpensive and all of the reactions
can be easily performed on a large scale. Nevertheless it is aesthetically more
pleasing to see a reaction sequence where as many reactions as possible
proceed in high yield.

With this in mind we considered the retrosynthetic plan, incorporating the
Cbz protecting group, as outlined in Figure 16. The requisite cyclization
precursor should be available from protection of the amine, produced by the

reduction of a nitrocompound 136. The nitro compound could arise from

55

Michael addition of 2-(2-furyl)-ethanol with acrolein and subsequent acetal
formation.

In the fonNard direction (Scheme 19) treatment of 2-(2-furyl)-1-nitroethane108
and acrolein in MeOH with NaOMe then addition of HC(OMe)3 and HCI
provided 5-(2-furyl)-4-nitropentanaldimethylacetal -in 57% yield. Reduction of
the nitro-compound (NiBg, NaBH4, MeOH) provided amine 138 in 48%

NHCbz No2
OMe OMe
1.2.5 1.3.6

Jfl

1101111945.: Second Generation Approach to (+/-)-Cocaine

yield109, which was treated was treated with Cszl and TEA in THF to supply
the Cbz protected cyclization precursor (135) in 94% yield.

Submission of 135 to the usual reaction conditions (CHCI3, trifluoroacetic acid
H2O 1;1) resulted in formation of the cyclized product (61%) accompanied by
several unidentified compounds. After purification by column chromatography,
139 in MeOH was treated with NaBH4 to provide a mixture of diols (140,71%).
Treatment of the epimeric mixture of diols in THF with o-
nitrophenylselenylcyanate furnished two selenides (141) in 99% combined
yield100, which were oxidized and eliminated to afford the terminal alkenes 142
in 76% combined yield. Oxidation of the epimeric alcohols (Collins reagent)
provided the ketone 143 in 83.3% yield110. In order to obtained the desired
methyl ester directly the alkene was taken up in a solution of HCI in MeOH and

submitted to ozonolysis104e. Two products were obtained in good overall mass

56

balance, however 1H-NMR,El-MS, and ir spectral data indicated that neither
corresponded to the desired compound. It is possible, either before or after
ozonolytic cleavage, that there is elimination of the nitrogen functionality. This
would provide a compound that would be susceptible to further attack by ozone.
Alternatively, under the acidic conditions, 143 may be enolizing during the
ozonolysis. It is well known that ozone can attack enol derivatives.

In view of these problems a different approach seems necessary. Ozonolysis

W Second Generation Preparation of the Tropane Ring System

 

N02
m 1) NaOMe / \ .
MeOH OMe N123, NaBH
0 N02 /\CHO W O _ MeOH ‘
131 57% 1.3.6 OMe 48%
NH2 NHCbZ
/ \ cm, / \ CFchZH
O OMe ———.TEA' THF 0 0M9 CHCI,
OMe 94V° OMe 610/"
1.3.6 135
Csz CI III Csz
_~_-_.BMe°H 0H _°'_N°2P“S°°”—
O 52% OH Pngp, THF
99%
m 5 1.4.9
/ Cb N /
CbZN Se(o-N02)Ph H202 CbZN croa. pyr, (”"1205 Z W
OH 76% OH 83% 0
1.4.1 1.1.2 113.

of alkene 143 in CH2CI2 with an acidic work up may provide the corresponding
acid, but as a p-ketoacid it may be susceptible to decarboxylation. Ozonolysis

of alkene 143 in CH2Cl2, and workup with dimethyl sulfide could provide

 

57

aldehyde 144 which may be oxidized, according to Corey's (MnO2, CN',
MeOH)111 or Wuts' conditions (NaHSO3, DMSO, (Ac)2O, MeOH)112, to the
desired methyl ester. Another approach may be to protect the ketone as an
oxoketal and then perform the cleavage to a methyl ester or a carboxylic acid.
Also to be examined are other oxidizing agentssuch as 0504, RuO4, and
KMnO4. These oxidations could be performed on either the ketone or the ketal
(see eq. 17, 18, and 19).

After obtaining the desired keto-ester 145 all that remains to complete a

Csz / Csz Csz
W 03, reductive work up KCHO MnOz, CN'. MeOH, or fiCOZMQ (17)
O O NaH 30,, DMSO, A070 0

15.3. 13.4 141.5
(*2qu 1) Ketalization CbZNflCgZH deprotection CbZN CO2Me (18)
2) O ,oxidau've work-up
O 3 O\) 0
1.1.3. 13.5. 115
,2, L “4°” c0211.
2) CH20,H002H O (‘9)
1.3.1

formal total synthesis of cocaine is the deprotection of the Cbz protected amine
and methylation of the free amine. The deprotection could be accomplished by
treatment of 145 with H2 in the presence of Pd catalyst. Methylation of the
amine could be accomplished by a method such as the Eschweiler-Clarke
procedure (CH2O, HCO2H).113

A few comments about the desirability of the two approaches .to the

cyclization precursor should be made. The low yield on both of the Michael

 

58

additions (Scheme 16 and Scheme 19) might be improved by examining some
alternate conditions. The nitro group reduction in the second approach might
be more efficiently accomplished with LAH. Putting the yields of each reaction
sequence aside, the first approach stands out as being more efficient because
all of the starting materials are commercially available. In addition, this furan
terminated N-acyliminium ion cyclization constructs the bridged ring system of
the tropane alkaloids very quickly. We will continue to examine the synthesis of
(+/-)-cocaine by the procedures outlined in equations 17-19.

The final alkaloid to be discussed is anatoxin-a. Anatoxin-a was isolated by
Edwards, Gorham and coworkers from the fresh water blue-green alga
Anabaena flos-aquae.114 It was originally called "very fast death factor"
(VFDF) after it was deemed responsible for numerous incidence of livestock
and waterfowl poisoning115. As one of the most potent nicotinic acetylcholine
receptor agonists known, it has proven very useful in neuropharmacological
studies116. Abnormalities in acetylcholine mediated neurotransmission has
been linked to myasthenia gravis, Parkinson's disease and Alzheimer's
disease. Because of its low natural abundance and the need for convenient
sources for both anatoxin-a and anatoxin-a analogues, this bridged alkaloid
has generated considerable synthetic interest.

There have been relatively few approaches to anatoxin-a. Two of the most
efficient routes to this molecule are the Speckamp25 and Rapoport27 syntheses
presented earlier. The Speckamp route is described in detail in Scheme 20.

Formation of the Mg-salt of succinimide, addition of 2 eq of the Grignard
reagent derived from 4-bromo-1-butene, reduction with NaBH30N and
acidification afforded butenylpyrrolidone 19 (53%) Protection of the lactam
(LDA, THF,EtO2CCN, 69%) followed by reduction and in situ ethanolysis

provided 148 (77%). The olefin was then converted into the requisite

59

terminator by cleavage (i. 03, CH2Cl2, -78° C; ii. S(Me)2, RT) and Wittig
addition of dimethyl (2-oxopropyl)phosphonate(83°/o) under the Masamune-
Roush conditions. Submission of cyclization precursor 21 to HCI in MeOH (-
50° C to RT) afforded the desired cyclization product 22 (47%) and chloride 23
(11%). The mixture could be treated with DBN to provide 22 (60%).
Deprotection (TMSI, 55%) completed the synthesis of anatoxin-a.

Our retrosynthetic plan for anatoxin-a is analogous to the initial plan for
cocaine (Scheme 16). Figure 17 shows how the side chain in anatoxin-a could
be derived from the alkyl group residue of a furan while the double bond could
be obtained from the ketone residue. The bicyclic furan (50) could be available
from an acyclic precursor (152) which was obtained from a one carbon
analogue of the nitroolefin used in the cocaine synthesis. The extra carbon
atom could be obtained by using furanacetaldehyde (153) rather than furfural
in the initial Michael addition.

Unlike furfural, furanacetaldehyde (153) is not a commercially available
compound. This meant that our first task in the anatoxin-a synthesis was to
devise a convenient method for obtaining large quantities of furanacetaldehyde.
At the time, the only reported literature preparation was by the Darzens
condensation”? We were able to prepare small amounts of
furanacetaldehyde by this route, but it was always accompanied by side
products. During our search search for a more convenient route, we examined
the oxidation 2-(2-furyl)-ethano| to furanacetaldehyde under a variety of
conditions (Swern89; PDC118; PCC44C; PCC/pyridine; PCC/NaOAc; 8H3-
SMe2/PCC119; pyr-803120; CrO3-pyr109; DEAD121, Ph3P, NO2CH2CO2Et122;
BaMnO4123; K2Cr207, 9M H2804, CH2Cl2, BU4NHSO4124; DEAD; 5%NaOCl,
BU4NHSO4125). The best yield (33%) was obtained from the DEAD,'Ph3P,

NO2CH2CO2Et method, but the product was difficult to separate from the excess

 

60

Scheme 20: Speckamp's 26Synthesis of Anatoxin-a

1) MeMgCI,11-IF

 

 

 

Z S 2) 2 eq. MQBM / 1) LDA, THF, .78° C
H 4) SN HCI H
53% 12 69%
NaBH‘, H2804, EtOH 1) 03, CHZCIZ, .79° 0
0% —p77% EtO/Q\/\/ 2) S(Me)2, RT
1 1 ,,
002m 002m 751,
2.0. 13.6
9 o
H (EtO)2P\)K BOW HCI, MeOH
EtO [q 83°42 -50° C 10 RT
COzMe 0 C02Et ‘8 “l
1.4.2 2.1.
EtOZCN EtOZCN HN 0
1) DBN 50%
M M 2) TMSI 55% ETA/K
22 (47%) 2.6 (11%) anatoxin-a

reagents. With several of the other reagents mentioned, we were able to
observe small amounts of aldehyde formation, but the reaction never
proceeded in high enough mass balance to be considered synthetically useful.
We attempted to condense furfural with methoxytriphenylphosphorane‘25, then
hydrolyze the intermediate enol ether by refluxing in acetone, H20 and
PPTs.127 The only product observed was the product of aldol condensation
between two molecules of furanacetaldehyde.

During an attempted Michael addition employing 2-(2-fury|)-1-nitroethane as

the Michael donor we observed furanacetaldehyde dimethylacetal as a bi-

 

61

HN 0 “N (D/
90* 3 m 3 t2
0 /
1.5.0 1.51
OMe
/ \ / \ ' ‘
WOMe : [O-k/CHO + NOZ/WOMG
N02 OMe
1_5.2 1.5.6 m

Figure 17: First Retrosynthetic Plan Leading to Anatoxin-a

product.128 By treating 2-(2-furyl)-1-nitroethane with NaOMe in MeOH followed
by HCI we were able to obtain 156 exclusively in a 74% yield (eq. 20). This led
us to examine some modifications of the Nef reaction as routes to
furanacetaldehyde.129 Treatment of 2-(2-furyl)-1-nitroethane under a variety of
conditions failed to give satisfactory results (i. 1) THF, 2N NaOH130; 2) 10N
H2804; ii. 1) KOH/MeOH 2) KMnO4/MgSO4131; iii. 1) t-BuONa 2) KMnO4132; iv.
TiCl3133; v. H202; vi. 1) aq. NaOH 2) aq. H2SO4I134; vii. 1) 30% H202, K2003,
H20135; viii. 1) TEA, CTAP136). We finally found that if the lithium salt of 2-(2-
furyl)-1-nitroethane was oxidized, in an aqueous solution of Na2B4O7, with
KMnO4, furanacetaldehyde137 was produced (50% mass balance) cleanly (eq.
21).138 The aldehyde was then subjected without purification to nitroaldol
condensation with nitrobutanaldimethyl acetal (eq. 22). Under a variety of base
catalysis (i. LDA139; ii. NaOH, EtOH1391140; iii. KF141; iv. Amberlyst142), the

alcohol was obtained in only 20% yield (from furanacetaldehyde). As a result of

 

62

the low yield we investigated an alternate approach to the required cylization
precursor.

We surmized that the cyclization precursor could be prepared in a fashion
similar to the second sequence of the cocaine synthesis (Figue 19). Thus, the
required lactam could be derived from a nitro compund, which in turn is
prepared from the condensation between acrolein and an appropriate furan
compound

Condensation of 3-(2-fury|)-1-nitropropane (159)143 with acrolein in MeOH
with catalytic NaOMe, then addition of HCI provided acetal 160 in 47% yield31a.
Reduction to the corresponding amine (Ni2B, NaBH4)109 and immediate
protection provided 161 in 72% yield. Unfortunately submission of this acetal
to the usual cyclization conditions (CHCI3, trifluoroacetic acid H2O, 1:1) did not
provide cyclized material. Instead, a single, high molecular weight compound
was isolated. At this point we did not attempt further identification. We decided
that we should approach the cyclization in a more stepwise manner, since in the
current one pot cyclization protocol it is difficult to ascertain which step in the
multi-step process is problematic. _

With this in mind we designed an N—acyliminium ion precursor in which the
lactam ring is preformed. The requisite N-acyliminium ion should be available
from lactam 163 by one of the reductive methods described above, and the
lactam should be readily available from a nitro ester (164), which in turn could
be prepared by a Michael addition.

In the forward direction Triton B catalyzed condensation of 3—(2-furyl)-1-
nitroethane143 with freshly distilled ethyl acrylate in EtOH144 provided ethyl-7-
(2-furyl)-5-nitroheptanoate 164 in 94% yield (Scheme 22). M28 reduction and

in situ cyclization afforded lactam 165 (86%).109 Protection of the nitrogen by

 

63

OM
W 1) NaOMe,MeOH We
(20)
0 N02 2) HCI
1.31

 

o OMe
74%

1) ”om mcm (21)
N02 0

2) aq. NazB4O7, KMnO4

 

50%, 1 52
OH OMe
WC”? + N02/\/\(OMa E2. / \ (22)
O O OMe
0M9 ca. 20% N02
1.5.5. 1.1.5. 1.5.6

Seheme 21: Second Generation Construction of the Anatoxin-a Ring System

OMe

 

[M 1) NaOMe,acrolein / \ 1) M28, MeOH
N02 2) HC(OMe) ,lVeOH 2) ClCOzEt, TEA
O stOH 3 O OMB THF
57% N02 72%
1.5.9.
1.6.9
/ \ We 11* EtOzCN
o OMe 7’ m
NHCOzEt 0

m1 :3

64

HN 0 /\
W :> o :0
3.5.

1.5.3

o COZEI :> o NO2+ /\c02Et

N02
L63: L52

FiguLe 1§'. Third Retrosynthetic Plan Leading to Anatoxin-a

treatment of 165 in THF with LDA (-78° C) followed by ethyl cyanoformate gave
166 (84%)25. Reduction of the lactam according to Chamberlin's procedure
provided carbinolamide 167 in 90% crude yield.14 Alternatively, reduction of
166 according to Speckamp's procedure provided the corresponding ethoxy-
carbamate (nearly quantitaive crude yield).18 With both N-acyliminium ion
precursors in hand we submitted them both to various cyclization conditions. In
each case, the only product obtained was the same high molecular weight
compound observed previously. Closer examination of the 1H-NMR spectra
provided some insight into the structure of this unknown compound. There
were no resonances in either the aldehydic or the a-furyl regions. There were
several signals in the B—furyl and olefinic regions. One very characteristic
signal, a doublet of triplets, was observed at 4.9ppm A similar signal was
observed at 5.1ppm in the spectra of the enamide formed from 39d. All this
data pointed towards a compound such as 168. This could result from attack of
starting material at the 5-position of the furan onto the aldehyde of a cyclized,

hydrolyzed furan. If this is the case it suggests that there in nothing inherently

 

65

wrong with formation of a seven membered bicyclic ring system. A potential
way around this "dimer" problem is to block the 5-position of the furan with some
group. We decided that the easiest, most reliable group to utilize would be a
methyl group.

Preparation of the needed cyclization precursor proceeded exactly as before
with the unsubstituted furan (Scheme 23). Thus, Triton B catalyzed addition of
3-(5-methyl-(2-fury|))-nitropropane1413 to freshly distilled ethyl acrylate provided
ethyl-7-(5-methyl-(2-furyI))-5-nitroheptanoate 169 in 88% yield. Reduction
(NigB, NaBH4) and in situ cyclization provided lactam 170 in 98% yield.108
Protection as before (LDA, CNCOgEt) afforded lactam 171 (58%).26

Submission of 171 to Speckamp's reduction conditions provided 172 in 74%

mm; Third Generation Construction of the Anatoxin-a Ring System

QWNOZ :mtlfialate' MCOZ EtNi'8—>ZBMH MO

COZEt
152 1.5.4
/ \ / \
LDA. 80sz O O NaBH‘ 0 OH
. 99% .
34% 002E: 002Et
116 L61
HCOOH, cCsH12 EtOZCN
Of
151mm CFSCOZH.HZO,CHCl3 ’ / \ \
O O
m COzEt

 

 

66

yield.18 As predicted blocking the 5-position of the furan allowed smooth
cyclization. Submission of 172 to cCsH12/HCOZH afforded a mixture of furan
173 in 58% yield and dione 174 in 17% yield. Furan 173 was readily oxidized
to ene-dione 175 (83%) by treatment with MCPBA in a two phase mixture of
CH2C|2 and saturated aqueous NaHCOa.

With both dione and enedione in hand we had to consider their
transformation to anatoxin-a. The first question to answer is how to select
between the two carbonyl groups of 174 or 175. We had previously
encountered problems with furan formation during ketalization attempts. If the

ene-dione could be selectively protected we could avoid the furan formation

W Fourth Generation Construction of the Anatoxin-a Ring System

ethyl acrylate CQZEt NizB NaBH., MeOH
02 Triton B 98%

88%

MO LDA EtOZCCN THF MO NaBH4. EIOH H250.
58% 74%

coza

 

EtOch EtOZCN o
HCOZH ccsHiz m + W
O O
COZEtE 113 (58%) mum)
EIOZCN COzEtN
in M. / TMSOCHZCHZOTMS
81% o TMSOTf, CHZClz. 43° c —* ->
O 52%

Anatoxin- -a
115

 

67

problem since 175 is not in the same oxidation state as a furan. Toward this
end we submitted dione 174 to both thermodynamic (HOCHZCHZOH, p-TsOH,
refluxing benzene)‘53 and modified kinetic ketalization conditions (CHQCI2,
TMSOCH20H20TMS, TMSOTf, -78° C to RT).62 In both cases we observed
significant furan formation. Submission of enedione 176 to Noyori's kinetic
ketalization conditions (78° C, TMSOCHgCHgOTMS, TMSOTf)62 provided the
side chain ketal selectively in 52% yield. The low yield may be a result of the
low stability of ene-dione 175. If this is the case, modification of the ketalization
conditions may result in higher yields. Work toward this end will continue.

To complete a synthesis of anatoxin-a we need to modify the side chain and
insert a double bond into the ring. One way we envisioned accomplishing this
is outlined in Scheme 24. Conjugate addition of a cuprate reagent should
provide 178 with the required methyl group in place. Reduction of the ring
ketone and protection with an appropriate group should afford 179.
Deprotection of the side chain ketal and Baeyer—Villager oxidation should yield
acetate 181. Hydrolysis of the acetate and oxidation of the derived alcohol
would complete the synthesis of the side chain. Deprotection of the alcohol and
elimination to an alkene would provide 183, with the anatoxin-a ring skeleton in
place. All that would need to be done to complete the synthesis of anatoxin-a is
deprotection of the amine with TMSl.

In summary, we have demostrated the 2- and 3-furyl moieties are sufficiently
nucleophilic and stable to serve as terminator functions for N-acyliminium ion-
initiated cyclizations. These processes provide access to 5,6; 5,7; 6,6; and 6,7-
membered, fused-ring systems with both the electronically favored 3-to2-furyl
closure (all) and the regioisomeric 2- to-3-furyl closure (5,6- and 6,6-membered

rings only) being realized. In addition a 6-6 spiro-cyclic

 

68
scheme 24: Proposed Conclusion to the Anatoxin-a Synthesis

EtOZC EtOZC EtOZC
'MeZCuLi' 1) NaBH4 O Q
2) OH protection \ I

OR
11!
BO C EtO C O
2 .
deprotect Baeyer-Villager 2 0k 1) - ON:
0 2) oxidation
OR OR
1.8.9 1.8.1
E1020 EtOZC HN
O 1) OH deprotection O TMSI O
2) elimination \ \
OR
1&2 m anatoxin-a

ring, an 8-aza-bicyclo[3.2.1] and an aza-bicyclo[4.2.1] ring have been prepared.
By accomplishing this we have shown that furans meet two of the requirements
for terminators to be deemed useful: they cyclize under mild conditions, and
they proceed with regiochemical predictability. We have completed a total
synthesis of lupinine, a formal total synthesis of perhydrohistrionicotoxin, and
have prepared the functionalized ring systems present in cocaine and anatoxin-
a. in summary, we have shown that a furan is a highly functional and versatile
teminator for the N-acyliminium ion cyclization.

During the course of these studies we have repeatedly encountered an
extreme substrate dependent furan hydrolysis/oxidation relationship.The
pattern that seems to be emerging is that the more strained the system the more

susceptible to cleavage it becomes.

 

EXPERIMENTAL

 

EXPERIMENTAL SECTION

General. Tetrahydrofuran (THF), benzene, and diethyl ether were dried by
distillation under argon from sodium benzophenone ketyl; methylene chloride,
triethylamine (TEA), pyridine, n-butylamine, acetic anhydride, and
diisopropylamine were dried by distillation under argon from calcium hydride.
Acrolein was distilled from hydroquinone and copper sulfate and used
immediately. Ethyl acrylate was distilled and used immediately. Trimethylsilyl
trifluoromethanesulfonate (TMSOTf), 1,2-bis(trimehtylsilyloxy)ethane,
tributylphosphine, and furfural were distilled at reduced pressures prior to use.
Cyclohexane was dried over molecular sieves, and nitromethane was filtered
through basic alumina prior to use. MeMgI was prepared from dried Mg and
Mel that was used as received. Formic acid (98%) was purchased from Fluka
and was used as received. Diethyl azodicarboxylate and Pth were purchased
from Aldrich Chemical Comnpany, Miwaukee, Wisconsin and were 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 a Perkin-
EImer Model 167 spectrometer. Proton magnetic resonance spectra (1H-NMR)
were recorded on a Varian T-60 at 60MHz or a Bruker WM-250 spectrometer at

250MHz as mentioned in deuteriochloroform unless otherwise indicated.

69

 

70

Chemical shifts are reported in parts per million (5 scale) from internal standard
tetramethylsilane. Data are reported as follows: chemical shifts (muliplicity: s =
singlet, bs = broad singlet, d = doublet, t= triplet, m = multiplet, coupling constant
(Hz), integration). Electon impact (El/MS, 709V) mass spectra were recorded
on a Finnigan 4000 with an INCOS 4021 data system.

Flash chromatography was performed according to the procedure of Still et
al.145 by using the Whatman silica gel mentioned and eluted with the solvents

mentioned. The column outer diameter (0. d.) is listed in millimeters.

Kinetic ketalization of 63b.

To a solution of 63b (188.8mg, 0.84mmol) in CH2C|2 (3ml) cooled to -24° C
in a dry ice-CCI4 bath, was added TMSOCH20H20TMS (174.4mg, 0.845mmol)
followed by TMSOTf (31 drops). The cooling bath was maintained at -24° C for
3 hours, then stirring was continued overnight while allowing the yellowish
solution to warm to room temperature. Pyridine (15 drops) was added and the
solution was cast into saturated aqueous NaHCO;; (5ml) and the aqueous layer
was extracted with CH2CI2 (4 x 5ml). The combined organic layers were
washed with 1N HCI (10ml), saturated aqueous NaHCO3 (10ml), brine (10ml),
dried (NaQSO4), and concentrated in vacuo to provide 85.3mg of a yellow oil.
The crude product was purified on a column of silica (230-400 mesh, 209 silica,
20mm 0. d., ethyl acetate-methylene chloride-methanol, 8:1 :1, 4.5ml fractions)
using the flash technique. Fractions 13-23 provided 210.8mg, 93 %, of 72 as a
water white amorphous solid.
1H-NMR (00013): 5:4.77 (ddd, J=14.4, 6.4, 3.4Hz, 1), 3.98 (m, 4), 3.3 (m, 1),
2.7-2.16 (m, 5), 2.15 (s, 3), 1.91-1.75 (m, 4), 1.6-1.4 (m, 3); 13C—NMR: 207, 169,
108, 64.6, 65.0, 58.7, 44.2, 40.3, 39.7, 33.7, 32.9, 29.6, 27.5, 18.9; IR (CCI4):
2945, 2780, 1715, 1640, 1440, 1350, 1250, 1150, 1050, 940, 900, 650.; EI-MS

 

71

(70 eV): 268 (M++1, 3.43), 224 (10.24 ), 209 (53.16), 179 (19.37), 150 (15.39),
137 (31.02), 99 (base), 87 (64.12), 69 (23.19), 55(79.63), 43 (68.21).

B r-Villa r xi i n f72.

To a solution of 72 (349mg, 0.013mmol) in dry CH2C‘I2 (2ml) was added
anhydrous NagHPO4 (720mg, 5mmol), then freshly prepared triflouroperacetic
acid (0.6ml). The mixture was warmed to reflux for 4 hours. After cooling to
room temperature, saturated aqueous Na28204 was added until the reaction
mixture tested negative to peroxides by the starch iodide test. The aqueous
layer was concentrated in vacuo, taken up in CH2CI2 (10ml), and the combined
organic layers were dried (N32804) and concentrated in vacuo to give 38.8mg
of an amorphous solid which was purified on a column of silica gel (230-400
mesh, 49, 20mm o.d., ethyl acetate-methanol-methylene chloride 821:1, 1.5ml
fractions) using the flash technique. Fractions 5-8 provided 18.8mg, 54 %, of
73 as water white oil.
1H-NMR (250MHz): 5 = 4.7 (ddd, J=14.4, 6.4, 3.4Hz, 1), 4.25-4.09 (m, 2), 4.0
(m,4) 3.58 (m, 1), 2.7 (dt J=13.1, 3.4Hz, 1), 2.45 (m, 2), 2.06 (s, 3), 2.0 - 1.65 (m,
6), 1.56 (dt, J=14.4, 4.7Hz, 1); IR (CHCI3): 2980, 1735, 1625, 1470, 1450,
1420, 1350, 1250, 1205, 1165, 1050, 950; EI-MS (70eV): 284 (M++1, 6.83 ),
293 (M+, 2.79), 224 (10.70 ), 210 (20.66), 192 (2.93 ), 178 (2.72), 166 (3.01),
150 (180), 138 (9.74), 125 (10.51), 110 (7.10), 99 (base), 69 (28.41), 55 (78.82),
43 (77.06).

Exehange ketelizetien ef 73.

To a solution of 73 (26.4mg, 0.093mmole) in dry CH2CI2 (2ml) was added
1,3-propanedithiol (20.2mg, 0.186mmol) followed by BF3-OEt2 (13.2mg,
0.093mmol). After stirring overnight, 10% aqueous NaOH was added (8ml) and

 

72

the aqeous layer was extracted with CH2C|2 (4 x 8ml). The combined organic
layers were washed with brine (20ml), dried over Na2804 and concenetrated in
vacuo to give 30 mg of an oil which was purified on a column of silica gel (230-
400 mesh, 39, 20mm 0. d., EtOAC'MeOH'CHQCIQ, 40:1 :1, 2ml fractions) using
the flash technique. Fraction 8—13 provided 26.7mg, 87%, 'of 74 as a colorless
oil.

1H-NMR (250 MHz): 8 = 4.68 (m, 2), 4.2 (dd, J=11.9, J=5.5Hz, 1), 3.62 (m, 1),
319-28 (m, 4), 2.7-2.2 (m, 6), 2.03 (s, 3), 1.9-1.5 (m, 6); IR (CHCI3): 3010,
2960, 1730, 1620, 1440, 1420, 1230, 1210, 730; El-MS (709V): 330 (M++1,
10.34), 329 (M+. 81.03), 222 (60.34), 55 (65.52), 43 (base).

Reney Niekel Reggetign of 74.
To a solution of 74 in absolute EtOH (1ml) was added Raney nickel (0.25mi)

as a slurry in EtOH (commercial Raney Nickel was rinsed with 8x5 ml EtOH).
After 1h at room temperature followed by 6 hours at reflux, the catalyst was
quenched by adding 1N HCI (50ml) then a few drops of 6N HCI. The green
solution was extracted with CH2C12 (4 x 50ml), and the combined organic
phases were washed with saturated aqueous NaHCO3, brine, dried (Na2804)
and concentrated in vacuo to give 9.1 mg of a brownish green oil which was
purified on a column of silica gel (230-400 mesh, 0.59, pipet, EtOAC-CH2C|2-
MeOH 402121, 0.5 ml fractions) using the flash technique. Fractions 7—15
provided 5.4m9, 60%, of 75 as a water white oil.

1H-NMR (250Mz): 5:4.81 (m, 1), 4.06 (dd, J=11.01, 4.2Hz), 4.0 (dd, J=11.01,
4.2Hz),3.15(m,1),2.5-2.1 (m, 4), 2.04 (s, 3), 1.9-1.3 (m, 8); IR (CHCI3): 3040,
3005, 2960, 1730, 1620, 1440, 1200, 1145, 700, EI-MS (70eV): 227 (M++2,
4.41), 226 (M++1, 34.96), 251 (M+, 43.9), 197 (17.48), 182 (25.54), 166 (64.83),

 

73

151 (56.32), 138 (62.09), 124 (20.78), 112 (84.08), 96 (46.85), 84 (38.13), 69
(89.73), 55 (base).

Reggetien 9f 75.

To a suspension of lithium aluminum hydride (3.5mg, 0.091mmole) in THF
(1ml) was added a solution of 75 (4.1mg, 0.0182mmole) in THF (2ml). The
mixture was heated to reflux overnight, then cooled to room temperature, and
quenched with 15% aq. NaOH (5 drops) and H20 (5 drops). The mixture was
dried over Na2SO4 and concentrated in vacuo to give 19mg of 31 as an
orange-yellow amorphous solid.
1H-NMR (250MHz): 8 = 3.8 (m,2), 3.0 (m, 2), 2.2-1.6 (m, 10), 1.5-1.0 (m, 5); IR
(CHCI3): 3360, 2940, 1400, 1240, 1060, 1000, 750; EI-MS (70eV): 169 (M+,
40.99), 168 (52.50), 152 (59.24). 138 (60.78); 111 (48.76), 97 (34.98).

 

Methyl magnesium iodide (20ml, 1.2M in Et20) was added dropwise to a
solution of glutarimide (3.449, 26.2mmol) in methylene chloride (500ml). The
mixture was heated to reflux for 30 minutes, then cooled to room temperature
and 3-(2—furyl)-propyl magnesium bromide (35ml, 0.98mmole) was added over
20 min. The mixture was refluxed overnight (18 hours), cooled to 0° and
quenched with saturated aqueous ammonium chloride (170ml). The precipitate
was removed by filtering through a pad of celite and the layers separated. The
aqueous layer was extracted with methylene chloride (4x80 ml). The combined
organic layers were dried (M9804) and concentrated in vacuo to afford 5.39,
90.5%, of the crude carbinolamide 89 as a slightly yellow amorphous solid,

which was used without further purification.

 

74

1H—NMR (250MHz): 5 = 8.16 (bm, 1), 7.22 (m, 1), 6.93 (bs,1), 6.21 (m, 1), 5.93
(m, 1), 4.77 (5,2) 2.7-1.6(m, 10); IR (CHCI3): 3685, 3620, 3380, 3200, 2980,
2400, 1710, 1660, 1220, 760, 670 cm-1; El-MS(70eV): 205(M+-18,5.78),124
(22.82), 111(8.86), 94(base), 82 (1872), 55 (30.24), 41 (23.71).

Pregeretien ef 88.

To a vigoursly stirred solution of 89 (5.39, 23.7mmol) in cyclohexane (290ml)
was added HCOgH (29ml). The two phase mixture was stirred for 3 min then
immediately cast into CH20I2 (350ml) and H20 (500ml). The aqueous layer
was separated and extracted with CH20l2 (3x300ml). The combined organic
layers were washed with saturated aqueous NaH003 (500ml), brine (200ml),
dried (M9804), and concentrated in vacuo to give 4 9 of an off white solid. The
crude product was purified on a column of silica gel (230-400 mesh, 3009.
60mm o.d., ethyl acetate, 100ml fractions) using the flash technique. Fractions
11-21 yielded 2.96 9, 55%, of 88 as a white solid. mp 193-194
1H-NMR (250MHz): 5 = 7.19 (d, J=2.1 Hz, 1), 6.28 (d, J=2.1Hz, 1), 6.74 (bs, 1),
2.52 (m, 2), 2.36 (m. 2), 1.76 (m, 8); IR (CCI4): 3065, 2920, 1662, 1540, 1450,
1395, 1125, 752, 732cm-1; El-MS (70 eV): 205 (M+, 60.6), 190, (10.65), 177
(b), 162 (28.29), 149 (37.11), 134 (34.60), 118 (11.01), 107 (29.16), 91 (21.66),
77 (20.11), 55 (40.45)

Preearatien and cyclization of 6-(3-(5-etth-(2-furvlil-broovl1-5-hvdroxv-2-
gigeridinone 97.

Methyl magnesium iodide (4.9ml, 2.4M in EtQO) was added over 15 min. to a
solution of glutarimide (1.459, 12.8mmol) in CHQCIQ (210ml). The reaction
mixture was heated to reflux for 30min, cooled to room temperature and 3-(5-

ethyl-2-furyI)-propyl magnesium bromide (14ml, 1.1M) was added over 15min.

 

75

The mixture was refluxed overnight (18hr), cooled to 0° and quenched with
saturated aqueous NH4CI (700ml). The precipitate was removed by filtering
through a pad of celite and the layers separated. The aqueous layer was
extracted with Cchlg (4 x 300ml), the combined organic layers dried over
M9804, and concentrated in vacuo to afford 5.079, (ca.'100%) of the crude
carbinol amide as a slightly yellow amorphous solid, which was used without
further purification.

To a vigorously stirred solution of 96 (5.079, 22.7mmol) in cyclohexane
(275ml) was added HCOzH (27ml). The two phase mixture was stirred for 3
min, then cast into CH20I2 (500ml) and H20 (500ml). The aqueous layer was
separated, saturated with NaCl and extracted with CH20|2 (4 x 500ml). The
combined organic layers were washed with saturated aqueous NaHCO;;
(750ml), brine (75 ml), dried (M9804) and concentrated in vacuo to give 5.29 of
an amorphous solid which was purified on a column of silica gel (230-400
mesh, 2009, 50mm 0. d., ethyl acetate-methylene chloride-methanol, 40:1:1,
75ml fractions) using the flash technique. Fractions 13-21 provided 2.179,
72.7%, of 98 as a white solid mp. 128-130° C
1H-NMR (250 MHz): 5 = 5.91 (s,1), 5.83 (bs, 1), 2.59 (m, 4), 2.41 (bt, J=5.9Hz,
2), 1.88 (m, 8), 1.2 (t, J=7.6Hz, 3); IR (CHCI3): 3000, 2960, 1700, 1580, 1450,
1385, 1205, 980, 760. EI-MS: 233 (M+, 3.37), 205 (5.49), 195 (4.81), 176
(3.876), 149 (8.98), 124(32.43), 96 (20.77), 84 (b), 51 (15.74), 49 (53.78), 41
(54.28);

MCPBA Oxidation and Reductien of 98.

To a solution of 98 (81.3m9, 0.35mmol) in CHQCIZ (3ml) cooled in an ice-
water bath was added saturated aqueous NaHC03 (3ml) followed by MCPBA
(80-85%, 81.7mg, 0.40mmol) in one portion. Stirring was continued at 0° C for

 

76

2h then at room temperature for 3 h. After separating the two phase mixture, the
organic layer was dried (NagsO4) and the aqueous layer was concentrate in
vacuo. The residue was taken up in CH2CI2 and dried over NaZSO4. The two
oganic phases were combined and concentrated in vacuo to give 82.6mg of a
clear oil which was immediately dissolved in 20ml of 20 % aqueous EtOH to
which had been added 4 drops of HOAc. The solution was hydrogenated
(1atm) over 10% Pd/C (20mg) for 45 min. The mixture was immediately filtered
through a pad of celite, and the celite was rinsed with CH2CI2. To the combined
filtrates was added saturated aqueous NaH003 (10 drops) , and the mixture
was concentrated in vacuo to near dryness. The residue was taken up in
CHzCla, dried over NaQSO4 and concentrated in vacuo to give 78mg of a
yellow oil, which was purified on a column of silica gel (230-400 mesh, 109,
20mm o.d., ethyl acetate-methylene chloride-methanol-triethylamine 40:1:0.5:
0.5, 2ml fractions) using the flash technique. Fractions 9-16 provided 61.1mg,
70%, of 99 as a water-white oil.

1H-NMR (250MHz): 5: 6.01 (bs, 1), 3.12 (m), 3.09 (dd, J=8.9, 2.5Hz, 3), 2.6-
2.15 (m, 8), 185-15 (m, 6), 1.07 (t, J=7.6Hz, 3); IR (CHCI3): 3000,2950, 1705,
1655, 1450, 1375, 1215, 715; El-MS (70eV): 251 (M+, 7.98), 222 (4.3), 194
(41.2), 176 (293), 166 (6.83), 152 (5.73), 134 (6.55 ), 124 (b), 112 (33.67), 96
(31.86), 84 (45.07), 55 (84.91), 41 (36.72).

Pregaretien ef thioketel 109.

A solution of 99 (13.9mg, 0.0554mmol) and TMSSCHgCHgSTMS (14.5mg,
0.0609mmol) in CH2CI2 (0.5ml) was cooled to -24° in a dry ice-CCI4 bath, and 3
drops of TMSOTf was added. The solution was stirred at -24° C for 1.5h, then
allowed to warm to room temperature and stirred overnight. To the solution was

added 10 drops of pyridine and the solution was cast into saturated aqueous

77

NaHC03 (5ml). The aqueous layer was extracted with CHQCIZ (4 x 4ml), and
the combined organic layers washed with 1N HCI (7ml), saturated aqueous
NaHCOa (7ml), and brine (7ml), dried over NaZSO4, and concentrated in vacuo
to provide 14mg of an amorphous solid. The crude product was purified on a
column of silica gel (230-400 mesh, 159, 20mm 0. d., ethyl acetate-methylene
chloride-methanol-triethylamine 25:1:0.5:0.5, 3ml fractions) using the flash
technique. Fractions 20-29 yielded 12mg, 67% of 100 as a white solid.
mp=168.5-170.0° C

1H-NMR (250 MHz): 8: 6.05 (bs, 1), 3.21 (m, 4), 2.73 (m,2), 2.4(m, 3), 2.2 (m,
2), 1.9-1.6 (m, 10), 1.09 (t, J=7.6Hz, 3);130-NMR: 207, 172, 73, 62, 59, 41, 39,
39.2, 38, 37.5, 32, 31, 24, 22, 17, 11; IR: (CHCI3): 3350, 3000, 2960, 1715,
1650, 1450, 1380, 1210, 720, 660; EI-MS: 327 (M , 3.07), 298 (7.15), 280
(5.88), 268 (b ), 250 (17.36), 234 (28.24), 194 (40.74), 176 (19.56), 133 (92.09),
124 (62.86), 112 (25.49), 96 (15.42), 55 (25.63).

Raney nickel reduetion of 100.

To a solution of 100 in absolute EtOH (5ml) was added 1/8 teaspoon Raney
Nickel. The mixture was refluxed for 2 hours, cooled to room temperature and
quenched with 1 N HCI. The green solution was extracted with CH2CI2 (4 x
25ml). The combined organic extracts were washed with saturated aqeous
NaHCO3 (25ml), brine (25ml), dried over NaQSO4, and concentrated in vacuo
to give a yellow oil which was purified on a column of silica gel (230-400, 79, 20
mm o.d., ethyl acetate-methylene chloride-methanol-triethylamine 30:1 :.5:.5, 1.5
ml fraction) using the flash technique. Fraction 16-21 provided 45.6 mg, 78.5%,
of 83 as a water white oil.
1H-NMR (250MHz): 8 = 6.56 (bs, 1), 2.32 (m, 5), 2.2-1.0 (m, 14), 0.86 (bt, 3); IR
(CHCI3): 3020,2950, 1710, 1650, 1440, 1390, 1210, 720; EI-MS (70 eV): 237

 

78

(10.33), 194 (7.42), 175 (3.67), 138 (9.94), 124 (base). 112 (34.08), 96 (50.88),
82 (18.99), 55 (71.26).

4-Ni1roegtenelgimethylaeetel 115.

To acrolein (11.29, 0.2mol) and nitromethane (97.69, -1.6mol) in benzene
(200 ml) in a water bath at room temperature, was added a solution of BU3P
(0.089, 0.0004mol) in benzene (2ml) over 5 min. After 45 min MeOH (20.2ml,
0.5mol), HC(OMe)3 (24.1ml, .22mol) and stOH (0.49m, 0.002mol) were
added and the solution was heated to 35° for 45 min. The organic phase was
washed with 5% aqueous NaHCO3 (75ml), brine (75ml), dried over Na2804,
concentrated in vacuo, and the orange viscous liquid was purified by distillation
under reduced pressure bp (3.5mm) 110-113° to give 11.49 of 115, 35%, as a
water white liquid.
1H-NMR (60 MHz): 8:4.42-422 (m, 3H), 3.25 (s, 6H), 2.3-1.4 (m, 4H); IR
(CHCI3): 300, 2940, 2840, 1555, 1440, 1380, 1200, 1130. 1060, 950, 715, 670;
El-MS (70 eV): 163(M+,0.17),162(1.81),161 (8.55), 132 (58.83), 100 (11.25),
85 (67.73), 75 (base), 71 (39.11).

5-(2-furyl)-4-nitro-4-gentenaldimethylacetal 114.

To a solution of 4—nitrobutanaldimethylacetal 115 (2.459, 15mmol) in
absolute EtOH (1.5ml) was added furfural (1.449, 15mmol) and n-BuNH2
(52mg, 0.71mmol). The solution was heated to reflux for 8 hr, cooled to room
temperature, and the dark reaction mixture was then cast into brine (200ml).
The aqueous phase was extracted with EtQO (2 x 200ml), and the combined
organic phases were dried (N82804) and concentrated in vacuo to give 4.19 of

a dark viscous liquid which was purified on a column of silica gel (230-400

 

79

mesh, 2009, 50mm 0. d., ether-hexane 1:1, 100ml fractions) using the flash
technique. Fractions 12-19 provided 2.59, 69%, of 114 as a yellow liquid.
1H-NMR (250MHz, CDCI3): 8 = 7.7 (s, 1), 7.6 (m, 1), 6.8 (m, 1), 6.5 (m, 1), 4.4 (t,
J = 5.65Hz, 1), 3.22 (s, 6), 3-1.6 (m, 4); IR (CHCI3): 3010, 2950, 1650, 1550,
1510, 1440, 1310, 1130, 1070, 730; El-MS (70 eV): 210 (M+-31, 11.65), 179
(9.03), 163 (12.95), 152 (16.42), 137 (18.37), 119 (21.03), 106 (33.75), 91
(28.50), 75 (base), 71 (78.50).

4-amino-5- 2-fu I- nt nal ime h I tal 116.

To a suspension of LAH (2.679, 70.1 mmol) in Eth (300ml) cooled in an ice-
water bath was added 5-(2-furyl)-4-nitro-4-pentenedimethylacetal, 114, (6.7g,
28mmol) in Et20 (300ml) over 1 hr. After the addition was complete, the mixture
was heated to reflux for 2 hours, cooled to 0° and quenched with 20% NaOH
(35ml). The precipitate was filtered through a pad of celite and the celite rinsed
well with Et20. The combined organic filtrates were dried over Na2804,and
concentrated in vacuo to give 5.119, 86%, of 116 as dark yellow liquid, which
was used without further purification.
1H-NMR (250MHz): 5 = 7.3 (m, 1), 6.26 (m, 1), 6.02 (m, 1), 4.33 (t, J: 5.61 Hz, 1),
3.28 (s, 6), 3.0 (m, 1), 2.75 (dd, J=4.46, 14.75Hz, 1), 2.5 (dd, J=8.16, 14.75, 1),
1.8-1.2 (m, 6); IR (CCI4): 3400, 2940, 2825, 1600, 1510, 1450, 1390, 1365,
1190, 1130, 1070, 1115, 965, 730; El-MS (70eV): 214 (M++1, 10.07), 182
(3.30), 150 (12.72), 132 (20.55), 100(base), 81 (90.22), 75 (44.38), 68 (48.69).

Pregeration of 117.
To a solution of 116 (5.119, 24.0mmol) in THF (80ml), cooled in an ice-water

bath, was added Et3N (3.159, 31.2mmol), and ethylchloroformate-(2.6g,

24.0mmol). After stirring at room temperature overnight, the mixture was filtered

 

80

through celite and the filtrate washed with brine (50ml). The organic phase was
dried over M9804 and concentrated in vacuo to give 0.669 of a dark yellow oil
which was purified on a column of silica gel (230-400 mesh, 609, 40mm 0. d.,
ethyl acetate-hexane 1:2, 30ml fractions) using the flash technique. Fractions
10-18 provided 4.4g, 65%. of 117 as a yellow solid. mp 53-55° C

1H-NMR (250MHz): 5:7.3 (m, 1), 6.25 (m, 1), 6.05 (m, 1), 4.7 (bm, 1), 4.3 (1,
J=5.8Hz, 1), 4.07 (q, 7.1, 2), 3.88 (bm, 1), 3.27 (s, 3), 3.28 (s, 3), 2.8 (m, 2), 1.7-
1.3 (m, 4), 1.15 (t, J=7.1Hz, 3); IR (CHCI3): 3440, 3005, 2950, 1710, 1510,
1450, 1220, 1130, 1080, 1055, 750; El-MS (70eV): 222 (M+-63, 7.57), 204
(5.54), 172 (69.95), 132 (15.93), 81 (41.16), 75 (34.30), 68 (base), 53 (15.84),
43 (15.23).

chlizetien ef 117.
The acetal 117 (0.2g, 0.70mmol) was dissolved in CHCI3 (22ml), cooled in

an ice-water bath and trifluoroacetic acid-H20 (1 :1, 11.6ml) was added. The
mixture was stirred for 1.5 h at 0°, then 1.5 h at room temperature. The mixture
was cautiously cast into CH2C|2 (50ml), and saturated aqueous NaHCO3
(50ml). The organic layers were extracted with CH20|2 (3 x 25ml) and the
combined layers washed with brine (50ml), dried over NaZSO4 and
concentrated in vacuo to provide a dark yellow oil, which was purified on a
column of silica gel (230-400 mesh, 89, 20mm 0. d., ethyl acetate-hexane 1:1,
2ml fractions) using the flash technique. Fractions 9-17 provided 101.3mg,
60%, of 118 a light yellow oil.

1H—NMR (250MHz): 5 = 9.78 (bs, 1), 4.5 (bm, 1), 4.3 (bm, 1), 4.2 (q, J=7.1 Hz, 2),
3.26 (m, 1), 2.9 (m, 1), 2.7 (m, 1), 2.33 (dd, J=1.95, 1.5.1Hz, 1), 2.15 (dd, J=4.68,
17.7Hz, 1), 1.55 (d, J=7.45Hz, 2), 1.28 (1, J=7.1Hz, 3); IR (CHCI3): 3020, 3005,
2980, 1690, 1420, 1380, 1335, 1320, 1205, 1110, 750; EI-MS (709V): 239

 

81

(M++1, 3.79), 210 (8.43), 194 (2.49), 168 (2.59), 149 (5.40), 140 (base), 96
(14.63), 82 (10.21), 68 (93.18), 55 (31.43).

Redgetien ef 119.

To a solution of the dione (59.2mg, 0.247mmol) in MeOH (3ml), cooled to
0°C in an ice water bath, was added NaBH4 (28mg, 0.741 mmol) in one portion.
The solution was allowed to warm to room temperature over 4.5hr, then the
excess NaBH4 was quenched with H20 (0.15mi). The solution was
concentrated in vacuo, the residue taken up in Cchlg, dried over N82804, and
concentrated in vacuo, to provide a yellow oil which was purified on a column of
silica gel (230-400 mesh, 139, 20mm 0. d., ethyl acetate (30ml), then ethyl
acetate-methylene chloride-methanol 8:1 :1 (50ml), 2ml fractions) using the flash
technique. Fractions 10-16 provided 40.5mg, 67% of 124a, and fractions 21-
35, 10.2mg, 17% of 124b.

Data for 124a:

1H-NMR (250MHz): 5 = 4.2 (bm, 1), 4.1 (q, J=7.1Hz, 2), 4.06 (bt, 1), 3.9 (m, 1),
3.82 (m, 1), 3.7 (m, 1), 3.0 (bs, 2), 2.2 (m, 1), 2.1-1.5 (m, 7), 1.2 (t, J=7.1Hz, 3); IR
(CHCI3): 3620, 3430, 2950, 1670, 1435, 1325, 1230, 1110, 1045, 740; EI-MS
(70eV): 243 (M+, 2.11), 212 (1.82), 198 (2.91), 1961 (238), 180 (1.68), 170
(9.34), 158 (52.94), 139 (34.79), 126 (4.38), 82 (31.71), 68 (base), 55 (36.87).
Data for 124b:

1H-NMR (250MHz): 5 = 4.3 (bm, 1), 4.11 (q, J=7.1Hz, 2), 4.03 (bm, 1), 3.8 (bm,
1), 3.66 (bm, 2), 3.41 (bs, 2), 198-1.5 (m, 9), 1.22 (t, J=7.1Hz, 3); IR (CHCI3):
3380, 2980, 1675, 1430, 1385, 1330, 1215, 1115, 750; EI-MS (70eV): 244
(M++1, 3.36), 243, (M+, 2.62), 212 (1.44), 196 (2.50), 180 (1.34), 170 (10.25),
158 (52.28), 139 (35.84), 126 (5.50), 108 (6.71), 96 (13.29), 84 (72.62), 68
(base), 55 (30.07).

 

82

Pr rinf lni 12.

To 124 (153.0mg, 0.63mmol) as a mixture of diasteriomers, and o-
nitrophenylselenylcyanate (172.3mg, 0.76mmol) in THF (2ml) was added BU3P
(153mg, 0.76mmol) dropwise. After stirring at room temperature overnight, the
solvent was removed and the crude product purified on a-column of silica gel
(230-400 mesh, 309, 20mm 0. d., ethyl acetate-hexane 4:3 for 28 fractions then
ethyl acetate for 45 fractions, 9ml fractions) using the flash technique. Fraction
15-26 provided 174.6mg, 64.8%, of an amorphous solid 125a, and 29-45
provided 63.9mg, 24%, of a yellow oil 125b.

Data for 125a:

1H-NMR (250MHz): 8 = 8.22 (m, 1), 7.5 (m, 2), 7.26 (m, 1),4.2(bm,1),4.1 (q,
J=7.1Hz, 2), 4.05 (m, 2), 2.95 (m, 2), 2.2-1.6 (m, 10), 1.2 (t, J=7.1Hz, 3); IR
(CHCI3): 33610, 2990, 1665, 1575, 1480, 1410, 1305, 1275, 1170, 1080, 670;
EI-MS (70eV): 428 (M++1, 1.79), 355 (9.80), 254 (3.54), 295 (2.08), 242
(61.07), 224 (50.00), 196 (48.27), 158 (base), 152 (25.67), 140 (57.14), 106
(16.64), 82 (17.23), 68 (41.29).

Data for 125b:

1H-NMR (250 MHz): 5 = 8.82 (m, 1), 7.5 (m, 2), 7.3 (m, 1), 4.3 (bm, 1), 4.2 (m, 1),
4.1 (q, J-7.1 Hz, 2), 3.6 (dt, J=11.5, 5.77Hz, 1), 3.02 (m, 2), 2.2 (m, 1),1.95-1.5
(m, 9), 1.22 (t, J=7.1Hz, 3); IR (CHCI3): 3420, 3000, 2915, 1675, 1590, 1425,
1335. 1310, 1210, 1115, 700; EI-MS (70 eV): 428 (M++1,4.25), 426 (M+, 1.24),
355 (4.78), 254 (3.77), 242 (3.47), 224 (base), 196 (1.52), 158(24.20), 152
(21.17), 140 (33.38), 106 (11.91), 82 (11.85), 68 (21.72).

Preparation ef Aeetate 126b.
To a solution of equatorial alcohol 125b (62.4mg, 0.146mmol) in pyridine
(0.23mi) was added acetic anhydride (149.1mg, 1.46mmol) and the solution

83

was stirred overnight. The solvent was removed in vacuo, and the dark oil was
taken up in CH20I2 (5ml) and washed with saturated aqeous NH4CI (5ml). The
organic phase was dried over Na2SO4 and concentrated in vacuo to provide a
dark oil that was purified on a column of silica gel (230-400 mesh, 9.09, 20mm
0. d., ethyl acetate-hexane 1:1, 2ml fractions) .using the flash technique.
Fractions 7-18 provided 66.8mg, 97.5% of a yellow oil.

1H-NMR (250MHz): 8 = 8.18 (m, 1), 7.4 (m, 2), 7.2 (m, 1), 4.7 (dt, J=11.5,
5.77Hz, 1), 4.2 (bm,1) 4.1 (bm, 1), 4.09 (q, J=7.1Hz, 2), 2.84 (m, 2), 1.9 (s,3), 2-
1.4 (m, 12), 1.13 (t, J=7.1Hz, 3); IR (CHCI3): 3020. 2980, 1730, 1680, 1510,
1430, 1330, 1250, 1200, 1110, 1030, 700; EI-MS (70eV): 470 (M++1, 3.95),
411 (7.72), 3.39 (7.58), 284 (12.36), 268 (35.80), 224 (79.87), 208 (25.05), 186
(28.17), 152 (26.24), 140 (88.51), 106 (36.81), 82 (26.06), 68 (87.59), 43 (base).

Pregeration ef Aeetate 126a.

To a solution of axial alcohol 125a (72.5mg, 0.17mmol) in pyridine (0.32mi)
was added acetic anhydride (173.1mg, 0.17mmol), then a catalytic amount of
DMAP, and the solution was stirred at room temperature overnight. The solvent
was removed and the residue purified on a column of silica gel (230—400 mesh,
109, 20mm 0. d., ethyl acetate-hexane 1:1, 2ml fractions) using the flash
technique. Fractions 9-18 provided 79.6 mg, 100%, of a yellow oil.
1H—NMR (250MHz): 5 = 8.22 (m, 1),7.46 (m, 2), 7.28 (m, 1). 5.16 (bt, J=3.21 Hz,
1), 4.21 (bm,1),4.1 (bm,1),4.1 (q,J=7.1Hz, 2),2.86 (t, J=8.97, 2), 2.1-1.6 (m, 9),
2.03 (s, 3), 1.21 (t, J=7.05, 3); IR (CHCI3): 3020,2980, 1730, 1680, 1510, 1430,
1330, 1250, 1200, 1170, 1110, 1030, 700; El-MS (70eV): 470 (M+, 2.88), 411
(2.66), 284 (14.85), 268 (8.65), 238 (28.46), 224 (26.38), 208 (4.96), 198 (base),
186 (6.8), 178 (7.41), 152(6.45),14o (17.67), 106 (4.35), 82 (1.09). 68 (1.23).

 

84

Elimin inf I ni 12 .

To 126b (53.9mg, 0.115mmol) in THF (0.5ml) was added 30% H202 (0.1 ml).
After stirring overnight, the solution was cast into CHZCIZ (5ml) and 10%
Na2S203 (5ml). The aqueous solution was extracted with CH20|2 (2x5ml), and
the combined organic layers were washed with saturated aqueous NaHCO3
(10ml), brine (10ml), dried over Na2804, and concentrated in vacuo to give a
yellow oil that was purified on a column of silica gel (230-400 mesh, 4.59,
20mm 0. d., ethyl acetate-hexane 1:2, 2ml fractions), using the flash technique.
Fractions 4-8 provided 30.8mg, 100%, of 127b as a water-white oil.
1H-NMR (250MHz): 8 = 5.6 (m, 1), 5.12 (m, 1), 5.0 (m, 2), 4.3 (bm, 1), 4.12 (bm,
1), 4.1 (q, J=7.1Hz, 2), 2.43 (bm,1), 2.2-1.5 (m, 6), 1.95 (s, 3), 1.22 (t, J=7.1Hz,
3); IR (CHCI3): 3020, 1725, 1685, 1520, 1425, 1210, 1110, 1030, 720; EI-MS
(15eV): 267 (M+, 1.91), 208 (10.94), 180 (2.66), 7.14 (base), 118 (4.80), 68
(60.06), 43 (34.31).

Elimination of selenide 126a.

To 126a (79.6mg, 0.17mmol) in THF (0.7ml) was added 30% H202 (0.15mi).
After stirring overnight, the solution was cast into CH2CI2 (5ml), and 10%
Na282O3 (5ml). The aqueous solution was extracted with CH20l2 (2x5ml), and
the combined organic layers were washed with saturated aqueous NaHC03
(10ml), brine (10ml), dried over N32804, and concentrated to give a yellow oil
that was purified on a column of silica gel (230-400 mesh, 59, 20mm 0. d., ethyl
acetate—hexane 1:2, 2ml fractions) using the flash technique. Fractions 5-8
provided 40.0mg, 88%, of 127a as a water white oil.
1H-NMR (250MHz): 5 = 1.7 (m, 1), 5.12 (m,2), 5.09 (m, 1), 4.22 (bm, 1), 4.14
(bm,1).4.1 (q, J=7.1Hz, 2), 2.68 (bm, 1), 2.2 (m, 1), 2.0 (s, 3), 1.9 (m, 4), 1,7 (m,
1), 1.2 (t, J=7.1Hz, 3); IR (CHCI3): 3005,2960, 1770, 1720, 1680, 1770, 1520,

 

85

1420, 1380, 1320, 1225, 1170, 1120, 1040, 1000, 730; El-MS: 267 (M+, 0.27),
208 (13.35), 179 (1.49), 152 (3.93), 140 (base), 68 (46.96)-

- 2-f l-4-nir n n l imeth I l 6.

To a solution of freshly prepared NaOMe (7.58mmol) in MeOH (125ml),
cooled to -30° C (CH30N-dry ice) was added a solution of freshly distilled
acrolein (3.699, 65.9mmol) and 2-(2-furyI)-1-nitroethane (137) (6.249,
44.2mmol) in MeOH (500ml) drowise over 2.5 h. After stirring at -40° to -30° C
overnight, concentrated HCI (7.6ml) was added dropwise and the solution was
allowed to warm to room temperature. Saturated NaHCOa was added to
neutralize the reaction mixture, the solvent was removed. Water (350ml) was
added to the residue, and the aqueous layer was extrated with CH2Cl2 (4 x
300ml). The combined organic layers were dried over NaZSO4 and
concentrated in vacuo to give a yellow liqiud which was purified on a column of
silica gel (230-400 mesh, 5009, 60mm o.d., ethyl acetate-hexane 1:8 for 8
fractions, 1:7 for 16 fractions, 1:2 for 50 fractions, 50ml fractions. Fractions 12-
43 provided 6.149, 57.1%, of 136 as a yellow liquid.
1H-NMR (250MHz): 5 = 7.3 (m, 1), 6.25 (m, 1), 6.08 (m, 1), 4.78 (m, 1), 4.3 (t,
J=5.61Hz, 1), 3.29 (m, 1), 3.28 (s, 3), 3.27 (s, 3), 3.06 (, 1), 2.0 (m, 1), 1.88 (m, 1),
1.65 (m, 2); IR (CCI4): 3120, 2950, 2830, 1600, 1510, 1450, 1370, 1200, 1130,
1175, 1120, 925, 860, 730; El-MS (70eV): 212 (M+-31, 4.23), 196 (1.36), 181
(11.72), 164 (11.78), 133 (22.22), 107 (41.96), 101 (28.66), 81 (57.15), 75
(base), 71 (53.49)

4-amino-5-(2-fuml)-pentenalgimethylacetal 138.
To a solution of NiCl2-6H20 (0.49, 2.05) in MeOH (40ml) was added NaBH4

portionwise (0.239, 6.15mmol) and the black suspension was sonicated for 1h.

86

5-(2-furyl)-4-nitropentanaldimethylacetal, 136, (1g, 4.1mmol) in MeOH (5ml),
then NaBH4 (0.549, 14.4mmol) was added and the mixture was stirred for 6 h.
The mixture was filtered through a pad of celite, and rinsed with MeOH. The
solvent was removed in vacuo to provide a green liquid and white solid (1.69),
which was purified on a column of silica gel (230-400 mesh, 809, 40mm 0. d.,
ethyl acetate-hexane 1:2 for 10 fractions, ethyl acetate-methylene chloride-
methanoI-triethylamine 202221 :1 for 21, 50ml fractions) using the flash
technique. Fractions 16-21 provided 0.629, 71%, 138 as a yellow viscous
liqiud.

1H-NMR (250MHz): 5 = 7.3 (m, 1), 6.26 (m, 1), 6.02 (m, 1), 4.33 (t, J=5.61Hz, 1),
3.28 (s, 6), 3.0 (m, 1), 2.75 (dd, J=14.75, 4.46Hz, 1), 2.5 (dd, J=14.75, 8.16Hz,
1), 1.8-1.2 (m, 6); IR (CCI4): 3400,2940, 2825, 1600, 1510, 1450, 1390, 1365,
1190, 11309, 1070, 1115, 965, 730; EI-MS (709V): 214 (M++1, 1.07), 182
(3.30), 150 (12.72), 132 (20.55), 100 (base), 81 (90.22), 75 (44.38), 68 (48.69).

Pregaration 91 135.

T0138 (320mg, 1.5mmol) and TEA (198.1mg, 1.95mmol) in CH20I2 (15ml),
cooled to 0°C in an ice-water bath, was added benzylchloroformate (281 .5mg,
1.65mmol). A precipitate formed immediately. The mixture was stirred at room
temperature overnight, the precipitate removed by filtration through a pad of
celite, rinsed with CH20I2, and the filtrates concentrated in vacuo to provide
0.529 of a yellow solid, which was purified on a column of silica gel (230-400
mesh, 409, 30mm 0. d., ethyl acetate-hexane 1:2, 20ml fractions). Fraction 22-
34 provided 0.499, 94%, of 135 as a yellow solid. m. p. 68-69° C
1H-NMR (250MHz): 5 = 7.3 (m, 6), 6.25 (m, 1), 6.08 (m, 1), 5.1 (s, 2), 4.82 (bm,
1), 4.31 (t, J=5.7Hz, 1), 3.9 (bm, 1), 3.3 (s, 3), 3.29 (s, 3), 2.82 (bt, 2), 172-1.3 (m,
4); IR (CHCI3): 3440,3005, 2960, 1715, 1550, 1220, 1125, 1020, 775, 700; El-

 

87

MS (70 eV): 284 (M+-63, 0.1), 266 (0.24), 234 (4.89), 190 (4.87), 158 (1.72),
312 (1.64), 108 (4.05), 98 (2.31), 91 (base), 75 (20.63).

delizetign 9f 13;.

To a solution of 135 (255.2mg. 0.734mmol) in CHCI3 (23ml), cooled to 0°C,
in an ice-H20 bath, was added trifluoroacetic acid-H20 (1 :1 ,12ml). The mixture
was allowed to warm to room temperature and stirred overnight, then cautiously
cast into CH20l2 (50ml) and saturated aqueous NaH003 (50ml). The aqueous
layer was extracted with CH2CI2 (3 x 25ml), the combined organic layers were
washed with brine (50ml), dried over Na2804 and concentrated in vacuo to
give a dark oil, which was purified on a column of silica gel (230-400 mesh,
259, 30mm 0. d., ethyl acetate-hexane 1:2 for 20 fractions, 1:1 for 51, 7ml
fractions) using the flash technique. Fractions 39-51 provided 135.6mg,
61%,139 as a yellow oil.
1H-NMR (250MHz): 5 = 9.33 (bs, 1), 7.38 (m, 5), 5.22 (s, 2), 4.6 (bm, 1), 4.39
(bm, 1), 3.29 (m, 1), 2.95 (m, 1), 2.72 (m, 1), 2.33 (dd, J=15.1, 1.95Hz), 2.15 (dd,
J=17.7, 4.68Hz, 1), 1.57 (d, J=7.45Hz, 2). 2.0-1.7 (m, 2); IR (CHCI3): 3020,
1700, 1410, 1340, 1210, 1105, 700; El-MS (259V): 305 (M+, 6.19), 220 (5.20),
176 (5.12), 170 (5.74), 158 (3.10), 152 (11.44), 68 (23.72).

Reguction of 139.

To a solution of 139 (310mg, 1.03mmol) in MeOH (10ml) cooled to 0°C in an
ice-water bath, was added NaBH4 (117mg, 3.09mmol) in one portion. After
stirring overnight at room temperature, saturated aqueous NH4C| was added
(1ml) and the solution was stirred for 30min. The solvent was removed, the
residue taken up in CHQCI2, dried over NaQSO4 and concentrated in vacuo to

provide a yellow oil, which was purified on a column of silica gel (230-400

 

88

mesh, 109, 30mm 0. d., ethyl acetate-hexane 1:2, 12 fractions, ethyl acetate, 38
fractions, 10ml fractions) using the flash technique. Fractions 22-38 provided
220mg, 71%, 140 of 2 alcohols.

1H-NMR (250MHz): 8 = 7.38, (m, 5), 5.12 (s, 2), 4.3 (bm, 1), 4.03 (bt, 1), 3.95
(bm, 1), 3.8 (m, 1), 3.65 (bm, 2), 2,9 (bm, 2), 229-1 (m, 8);. IR (CHCI3): 3610,
3400, 3005, 2975, 1680, 1420, 1325, 1210, 1100, 1045, 720; EI-MS (25 eV):
305 (M+, 0.19), 220 (5.20), 176 (5.12), 170 (5.74), 158 (3.10), 152 (11.44), 91
(base), 68 (23.72).

Preparation ef selenide 141.
To 140 (220mg) and o-nitrophhenylselenylcyanate (200mg) in THF (3ml)

was added BU3P (175mg, 0.86mmol) dropwise. After stirring at room
temperature overnight the solvent was removed in vacuo and the crude product
purified on a column of silica gel (230-400 mesh, 309, 30mm 0. d.,ethyl acetate-
hexane 1:4 for 4 fractions, ethyl acetate-hexane 1:1 for 16, ethyl acetate for 28,
7 ml fractions) using the flash technique. Fractions 12-28 provided 350mg,
99%, of 141 as a orange oil.

1H-NMR (250MHz): 5 = 8.2 (m, 1), 7.4-7.15 (m, 8), 5.02 (m, 1), 5.01 (s, 2), 4.2
(bm, 1), 3.99 (bm, 1), 2.85 (bm, 2), 2.2-1.6 (m, 10); IR (CHCI3): 3610,3400,
3000, 2940, 1680, 1585, 1520, 1460, 1230, 1115, 1030, 720; EI-MS (25eV):
243 (M+-246, 10.44), 224 (1.60), 210 (4.31), 197 (22.75), 184 (base), 149
(6.67), 130 (13.11), 93 (16.58), 71 (91.10).

Elimination of 141.
To 141 (350mg, 0.715mmol) in THF (3ml) was added 30% H202 (0.62mi).
After stirring overnight, the solution was cast into CH2CI2 (10ml) and 10%

Na2S203 (10ml). The aqueous layer was extracted with CHQC|2 (3 x 5ml), and

 

89

the combined organic layers were washed with saturated aqueous NaH003
(10ml), brine (10ml), dried over Na2SO4, and concentrated in vacuo to give a
yellow oil that was purified on a column of silica gel (230-400 mesh, 199, 30mm
0. d., ethyl acetate-hexane 1:8 for 4 fractions, 1:1 for 25 fractions, 20 ml
fractions) using the flash technique. Fractions 6-8 provided~115.0 mg, 56%, 19-
25 45.4mg, 22.1%, of 142.

1H-NMR (250MHz): 5 = 7.3 (m, 5), 5.9 (m, 1), 5.28 (m, 2), 5.12 (s, 2), 4.3 (bm, 1),
4.22 (bm, 1), 4.05 (t, J=5.77Hz, 1), 2.6 (bm, 1), 231.7 (m, 7); IR (CHCI3): 3450,
3010, 1690, 1425, 1325, 1220, 1105, 1040, 930; El-MS (25 eV): 288 (M++1,
0.12), 287 (M+, 0.11), 220 (9.63), 176 (7.74), 152 (6.63), 91 (base), 68 (32.92).

W-

To a solution of pyridine (149.0mg, 1.88) in CH2C|2 (2.5ml) was added Cr03
(93.9, 0.94mmol). After stirring 15 min, a solution of 142 (45.0mg, 0.157mmol)
in CHgClg (4ml) was added and the dark mixture was stirred for 20 min. The
solvent was decanted and the residue rinsed with CHQCIQ (8 x 2ml). The
combined organic phases were washed with 5% NaOH (5ml), 1N HCI (5ml),
saturated aqeous NaHCO;; (5ml), brine (5ml) dried over Na2804 and
concentrated to give a dark yellow oil which was purified on a column of silica
gel (230-400 mesh, 3.59, 10mm 0. d., ethyl acetate-hexane 1:3, 1.5ml fractions)
using the flash technique. Fractions 2-8 provided 28.5mg, 64%, of 143 as a
yellow oil.
1H-NMR (250 MHz): 5 = 7.3 (m, 5), 5.9 (m, 1), 5.28 (m, 2), 5.18 (s, 2), 5.12 (m,
1), 4.58 (bm, 1), 4.5 (bm, 1), 2.7 (bm, 1). 2.35 (m, 1), 2.0-1.5 (m, 4); IR (CHCI3):
3005, 2960, 1680, 1410, 1315, 1260, 1210, 1095, 1015; EI-MS (70eV): 285
(M+, 3.46), 257 (1.73), 229 (0.29), 202 (1.35), 166 (2.33)., 158 (8.60), 150-(5.42),
91 (base), 65 (12.14)

 

90

6-(2-fuml)-4-nitrohexanaldimethylacetal 160.

To a freshly prepared solution of NaOMe (0.58mmol) cooled to -30°C in a
xylenes-liquid N2 bath was added slowly a solution of acrolein (0.2839,
5.05mmol) and 3-(2-furyI)-1-nitropropane 159 (0.539, 3.4mmol) in MeOH
(48ml) over 45 min. After stirring for 45 min at -30° the solution was allowed to
warm to room temperature overnight, then cooled to -30° and conc. HCI
(0.58mi) was added. The solution was allowed to warm to room temperature
over 6h, solid NaHCO3 was added and the solvent was removed. Water (75ml)
was added to the residue and the aqueous layer was extracted with CH2CI2 (4
x 50ml), brine (100ml), dried over N32804 and concentrated in vacuo to
provide 0.779 of a dark red oil which was purified on a column of silica gel (230-
400 mesh, 77g, 40mm 0. d., ethyl acetate-hexane 1:4, 25ml fractions) using the
flash technique. Fractions 3562 provided 411mg, 47%, of 160 as a yellow oil.
1H-NMR (250MHz): 5 = 7.3 (m, 1), 6.28 (m, 1), 6.01 (m, 1), 4.48 (m, 1), 4.29 (t,
J=5.61Hz, 1), 3.28 (s, 6), 2.65 (m, 2), 2.29 (m, 1), 2.03 (m, 2), 1.8 (m, 1), 1.6 (m,
2); IR (CHCI3): 3010,2940, 1550, 1440, 1370, 1200, 1130, 1070, 740; El-MS
(25 eV): 228 (M+-15, 4.94), 210 (1.62), 201 (2.60), 182 (1.21), 166 (8.30), 158
(5.52), 138 (2.17), 91 (base), 68 (7.39)

4-emine-6-(2-fgml)-hexenelgimethylaeetal.

To a solution of NiCl2-6H20 (68.9mg, 0.29mmol), in MeOH (5.6ml) was
added NaBH4 (32.9mg, 0.87mmol) portionwise. The black suspension was
sonicated for 30min, then a solution of 160 (150mg, 0.58mmol) in MeOH (1ml)
was added, followed by more NaBH4 (77.1mg, 2.0mmol). After stirring
overnight the mixture was filtered through a pad of celite, the celite rinsed with
MeOH, and the organic filtrate concentrated in vacuo to give 120mg, 91%, of a

yellow-green oil which was used without further purification.

91

1H-NMR (250MHz): 5: 7.3 (m, 1), 6.28 (m,1), 6.12 (m, 1), 5.9 (bm, 2), 4.33 (1.
J=5.6, 1), 3.32 (s, 6), 3.1 (bt, 1), 2.81 (t, J=7.7, 2), 2.0 (m, 2), 1.73 (m, 4).; El-MS
(70eV): 228 (M+1, 2.27), 227(M+, 0.84), 164 (13.22), 124 (20.92), 100 (27.40),
94 (7.52), 81 (base), 75 (38.78), 68 (56.97), 53 (25.63), 43 (26.93)

Pregeretien gt 1 §1.

To a solution of the ande amine (120mg, 0.53mmol) in THF (4ml) was added
TEA (70mg, 0.689mmol) and ClCOgEt (63mg, 0.583mmol). After stirring
overnight the mixture was filtered through a pad of celite and the celite rinsed
with CH20l2. The organic filtrates were concentrated in vacuo to provide
280mg of a yellow oil which was purified on a column of silica gel (230-400
mesh, 17g, 30mm 0. d., ethyl acetate-hexane 3:4, 10ml fractions) using the flash
technique. Fractions 13-21 provided 125.1mg, 78.8%, of 161 as yellow solid.
1H-NMR (250MHZ): 5:7.3 (m,1), 6.22 (m, 1), 5.94 (m,1 )4.31 (t, J=6Hz, 1), 4.1
(bm, 1), 4.08 (q, J=7.1Hz, 2), 3.29 (s, 6), 2.6 (m, 3), 1.8, 1.4 (m, 6), 1.1 (t,
J=7.1Hz, 3); IR (CHCI3): 3440, 3000, 2950, 1700, 1510, 1450, 1415, 1385,
1225, 1120, 1070, 740; El-MS (70eV): 268 (M+-31, 0.25), 235 (7.96), 196
(12.56), 146 (8.74), 107 (21.58), 97 (11.93), 81 (base), 75 (39.61), 71 (12.91),
68 (19.68), 53 (10.88)

(gyelizetion ef 161.

To 161 (52.7mg, 0.176mmol) in CHCI3 (6ml), cooled to 0°C in an ice-H20
bath, was added trifluoroacetic acid-H20 (1:1, 2.9ml). After stirring overnight the
two phase mixture was cast into H20 (5ml) and the aqueous layer was
extracted with CH2CI2 (3 x5ml). The combined organic layers were washed
with saturated aqueous NaHCOs (10ml), brine (10ml), dried over NaZSO4,

concentrated in vacuo to provide 59.8mg of a dark oil which was purified on a

92

column of silica gel (230-400 mesh, 59, 20mm 0. d., ethyl acetate hexane 3:4,
2m| fractions) using the flash technique. Fraction 7-9 provided 14.8 mg, 35%, of
168 as an amorphous solid.

1H-NMR (250MHz): 5 = 6.14-5.74 (m, 4), 4.84 (bdt, 2), 4.04 (m, 8),2.74-1.64 (m,
14), 1.19 (m, 6); IR (CHCI3): 3005,2980, 1780, 1560, 1420, 1380, 1345, 1120,
1010; El-MS (70eV): 470 (M+, 0.93), 397 (1.06), 381 (5.18), 292 (2.02), 266
(3.04), 266 (3.04), 234 (3.62), 222 (4.80), 196 (3.46), 172 (6.64), 159 (8.44), 120
(57.84), 107 (base), 81 (44.88), 68 (35.56).

Ethvl-6-12-furvI)-4-nitrohexenoete 164.

To a solution of 3-(2-furyl)-1-nitropropane 159 (29, 12.89mmol) and dioxane
(0.6ml) was added Triton-B (0.05mi) and the solution was warmed to 70°C
(internal temperature). Ethyl acrylate (0.659, 6.45mol) was added dropwise,
maintaining the temperature at 90° or below. After stirring at 70-75°C for 8h,
and cooling to room temperature, the solution was acidified with 1N HCI, cast
into CHZCI2 (35ml). The aqueous phase was separated, extracted with CH20I2
(2x5ml), and the combined organic layers were washed with saturated aqueous
NaH003 (10ml), H20 (10ml), dried over Na2804 and concentrated in vacuo to
provide a yellow liquid which was purified on a column of silica gel (230-400
mesh, 2209, 50mm 0. d., ethyl acetate-hexane 1:6 for 20 fractions, ethyl acetate-
hexane 1:4 for 48 fractions, 40ml fractions) using the flash technique. Fraction
27-48 provided 1.559, 94%, of 164 as a yellow liquid.
1H-NMR (250MHz): 5 = 7.28 (m, 1), 6.24 (m, 1), 6.0 (m, 1), 4.55 (m, 1), 4.1 (q,
J=7.2Hz, 2), 2.65 (m, 2), 2.35—1.95 (m, 6), 1.2 (t, J=7.2Hz, 3); IR (neat): 2980,
2940, 1720, 1550, 1440, 1380, 1200, 1100, 1020, 800, 740; EI-MS (25. eV):
225 (M+-30, 1.22), 207 (15.13), 163 (12.34), 135 (14.20), 133 (31.70), 126

93

(12.15), 121 (17.06), 120 (52.61), 119 (16.27), 107 (11.09), 97 (43.02), 94
(7.91), 81 (base).

-2-2-f l- h l-2- rr Ii in n1.

To a solution of NiCl2-6H20 (74.6mg, 0.314mmol) in MeOH (6ml) was added
NaBH4 (35.6mg, 0.94mmol) and the black suspension was sonicated for 1.5h.
Ethyl-5-(2-furyl)hexanoate 160 in MeOH (1ml) was added followed by NaBH4
(83mg, 2.19mmol). After stirring for 2 days the catalyst was removed by filtration
through a pad of celite, the celite rinsed with MeOH, concentrated in vacuo to
provide 165 as a greenish oil which was purified on a column of silica gel (230-
400 mesh, 109, 20mm 0. d., ethyl acetate, 5ml fractions) using the flash
technique. Fractions 6-30 provided 96.2mg, 85.6% of 165 as a white solid.
mp=72.0-73.5 ° C
1H—NMR (250MHz): 5: 7.32 (m,1), 6.29 (m,1), 6.02 (m, 1), 5.0 (b, 1), 3.7 (m, 1),
2.7 (t, J=7.5Hz, 1), 2.3 (m, 3), 1.85 (m, 3); IR (CHCI3): 3440,3020, 1695, 1220,
1130, 930, 1775, 710, 670; El-MS (70 eV): 179 (M+, 22.77), 162 (6.11), 161
(2.75), 120 (10.16), 98 (11.25), 97 (32.31), 94 (9.72), 84 (base), 81 (36.43), 69

(10.33), 56 (16.11), 55 (11.25), 53 (18.96), 41 (28.62).

 

To a solution of diisopropylamine (0.849, 8.26mmol) in THF (17ml), cooled to
-28°C in a dry ice-CCI4 bath was added nBuLi (3.4ml, 2.4M, 8.26mmol) after
30min, the solution was cooled to -78°C (i-PrOH-dry ice) and 165 (1.48,
8.26mmol) in THF (10ml) was added dropwise over 10min. Stirring was
continued for 25min, then ethylcyanoformate (0.829, 8.26mmmol) was added,
and the resulting mixture was allowed to stir at -78°C for 15min, then was

warmed to room temperature, and cast into CH2CI2 (50ml), and H20(50ml).

 

94

The aqueous layer was extracted CHchQ (3 x 30ml) and the combined organic
extracts were dried over Na2SO4, concentrated in vacuo to provide 2.219 of a
yellow liqiud which was purified on a column of silica gel (230-400 mesh, 2109
silica, 50mm 0. d. ethyl acetate-hexane 9:16 for 8 fractions, ethyl acetate
hexane 1:1 for 16 fractions) using the flash technique. Fractions 11—16 provided
1.769, 84%, of 166 as a yellow oil.

1H-NMR (250MHz): 5 = 7.28 (m, 1), 6.25 (m, 1), 6.03 (m, 1), 4.26 (q, J=7.2Hz);
IR (neat): 3120, 1785, 1750, 1715, 1600, 1510, 1450, 1375, 1280, 1150, 1050,
1010, 930, 750; El-MS (70eV): 251 (M+, 32.83), 206(5.76), 162 (36.78), 157
(36.85), 120 (61.31), 95 (21.83), 94 (18.77), 85 (28.61), 84 (base), 81 (59.67),
56 (19.89), 55 (18.29), 53 (32.46), 41 (43.39).

Redgetien of 166.

To the lactam 166 (99.4mg, 0.40mmol) in MeOH (4.0ml), cooled to -5° C in
an ice-water-NaCl bath, was added NaBH4 in one portion (153.0mg, 4.0mmol).
After stirring for 1h, the mixture was cast into saturated NaH003 (4ml), CH2CI2
(4ml) cooled to 0° and stirred rapidly for 15min. The aqueous layer was
extracted with CH2CI2 (4 x 4ml), the combined organic layers were washed with
brine, dried over Na2SO4 and concentrated in vacuo to provide 100mg, 98.8%,
of 167 as a water-white oil which was used without further purification.
1H-NMR (250MHz): 5: 7.26 (m, 1), 6.22 (m, 1), 6.01 (m, 1), 5.48 (m, 1), 4.14 (m,
2), 3.8 (bm, 1), 2.6 (m, 2), 2.2-1.5 (m, 7) 1.22 (t, J=7.2Hz, 3); IR (0014): 3600,
3440, 1685, 1420, 1380, 1340, 1265, 1190, 1115, 1110, 730; El-MS (70eV):
253 (M+, 1.52), 235 (37.97), 158 (17.05), 146 (19.51), 141 (57.65), 140 (20.98),
120 (23.70), 107 (12.68), 96 (12.68), 94 (15.44), 86 (17.79), 82 (12.25), 81
(57.45), 80 (15.66), 69 (48.74), 68 (base), 56 (10.84), 53 (25.23), 41 (34.97)

 

 

To a solution of 166 (50.3m9, 0.2mmol) in EtOH (1 .5ml) cooled to -20°C in a
dry ice-CCI4 bath was added NaBH4 (53.2mg, 1.4mmol). Every 15 min 2N
H2SO4 (2-3drops) was added. After stirring for 4h at -20°C , the solution was
acidified to pH2 with 2N H2SO4, warmed to room temperature, cast into CH20I2
(5ml), brine (5ml). The aqueous layer was extracted with CH2C|2 (4 x 5ml),
dried over NaZSO4, and concentrated to give 61 .6mg of a water white oil.
1H-NMR (250MHz): 5: 7.26 (m, 1), 6.22 (m, 1), 6.0 (m, 1), 5.3 (bs, 1), 4.12 (bm,
2), 3.83 (bm, 1), 3.52 (bm, 2), 2.61 (m, 2), 2.24 (b, 1), 2.04 (m, 1), 1.9-1.6 (m, 4),
1.22 (1. J=7.2Hz, 3), 1.13 (t, J=7.2Hz, 3); IR (CHCI3): 2990, 1685, 1415, 1380,
1350, 1320, 1200, 1115, 1010, 930, 730; El-MS (70eV): 235 (M+-46, 12.46),
196 (1.46), 152 (3.60), 141 (29.83), 140 (25.59), 120 (1.74), 107 (5.46), 94
(9.26), 91 (2.53), 81 (40.82), 68 (base), 53 (21.91).

EhI-- -m hI-2-f I-4-nirhxn 19.

To a solution of 3-(2-furyl)-1-nitropropane (7.6g, 4.9mmol) and dioxane
(2.2ml) was added Triton-B (0.45mi) and the solution was warmed to 70°C
(internal temperature). Ethyl acrylate (2.259, 22.46mmol) was added dropwise,
maintaining the temperature at 90° or below. After stirring at 70-75°C for 6h,
and cooling to room temperature, the solution was acidified with 1N HCI, cast
into CH2Cl2 (50ml). The organic layer was washed with saturated aqueous
NaHC03 (20ml), H20 (20ml), dried over N32802:, and concentrated in vacuo to
provide 10.79 of a yellow liquid which was purified on a column of silica gel
(230-400 mesh, 4009, 60mm 0. d., ethyl acetate-hexane 1:10 for 10 fractions,
ethyl acetate-hexane 1:6 for 18 fractions, 60ml fractions) using the flash

technique. Fraction 11-18 provided 5.39, 87.6%, of 169 as a yellow liquid.

 

96

1H-NMR (250MHz): 5: 5.9 (m, 1), 5.82 (m, 1), 4.58 (m, 1), 4.12 (q, J=7.2Hz, 2),
2.6 (m, 2),2.3-2.0(,6),2.21 (s, 3), 1.22 (t, J=7.2Hz, 3); IR (CCI4): 2980,2940,
1735, 1555, 1450, 1380, 1180, 1025, 740; El-MS (709V): 270 (M++1, 1.20),
269 (0.44), 252 (0.48), 239 (0.52), 221 (2.77), 193 (1.11), 177 (2.13), 147 (2.72),
108 (21.62), 95 (base).

5- 2- -me h l- 2-f l -e h I-2- rrolidin ne 170.

To a solution of NiCl2-6H20 (5.1g, 18.9mmol) in MeOH (18ml) was added
NaBH4 (1.079, 28.4mmol) and the black suspension was sonicated for 40min.
Ethyl-5-(2-furyl)hexanoate 169 in MeOH (12ml) was added followed by NaBH4
(2.59, 66.2mmol). After stirring for 2 days the catalyst was removed by filtration
through a pad of celite, the celite rinsed with MeOH, concentrated in vacuo to
provide 170 as a greenish oil which was purified on a column of silica gel
(230-400 mesh, 2209, 50mm 0. d., ethyl acetate-methylene chloride-methanol
20:1 :1, 40ml fractions) using the flash technique. Fractions 2-6 provided 3.579,
97.8% of 170 as a white solid. mp=69.5-70.0° C
1H-NMR (250MHz): 5 =5.85 (m, 2), 3.62 (m, 1), 2.6 (t, J=7.0Hz, 2), 2.3 (m, 2),
2.21 (s, 3), 1.8-1.5 (m, 5); IR (CHCI3): 3320, 3060, 2990, 1700, 1420, 1260,
1130, 900, 680; EI-MS (70eV): 193 (1.39), 134 (1.04), 105 (2.77), 95 (13.83),
84 (20.88), 69 (12.74), 56 (53.38), 43 (base).

1-carboethoxv-5-(2-(5-mghvl-(2-furvlllethvll-Z-bvrrolidinone 171.

 

To a solution of diisopropylamine (1.869, 18.4mmol) in THF (38ml), cooled to
-24°C in a dry ice-CCI4 bath was added nBuLi (7.7ml, 2.4M, 18.4mmol) after
30min, the solution was cooled to -78°C (i-PrOH-dry ice) and 170 (3.569,
18.4mmol) in THF (18ml) was added dropswise over 10 min. Stirring was

continued for 25 min and ethylcyanoformate (1.89, 18.4mmmol) was added

 

 

 

97

dropwise over 10 min. The solution was stirred at -78°C for 15 min, allowed to
warm to room temperature, and cast into CH20I2 (75ml), and H20(75ml). The
aqueous layer was extracted with CH20I2 (3 x 40ml) and the combined organic
extracts were dried over Na2SO4, concentrated in vacuo to provide 5.269 of a
yellow liqiud which was purified on a column of silica gel (230-400 mesh, 2209
silica, 50mm 0. d. ethyl acetate-hexane 9:16 for 8 fractions, ethyl acetate
hexane 1:1 for 16 fractions 50ml fractions) using the flash technique. Fractions
6-12 provided 2.859, 58%, of 171 as a yellow oil.

1H-NMR (250MHz): 5:5.86 (m, 1), 5.8 (m, 1), 4.26 (q, J=7.2Hz, 2), 4.18 (m, 1),
2.6-2.3 (m, 4), 2.19 (s, 3), 2.1 (m, 2), 1.78 (m, 2 (1.28 (t, J=7.2Hz, 3); IR (CCI4):
2980, 2940, 1735, 1555, 1450, 1380, 1180, 1025, 740; El-MS (70eV): 265
(M+, 21.80), 249 (1.72), 192 (3.69), 176 (34.24), 157 (15.55), 148 (12.07), 134
(35.41 ), 121 (12.68), 109 (29.80), 95 (98.77), 84 (base), 55 (27.83), 43 (77.09).

Pregaratien ef 172.
To a solution of 171 (777mg, 2.93mmol) in EtOH (22ml) cooled to -20°C in a

dry ice-CCI4 bath was added NaBH4 (779mg, 20.5mmol). Every 15 min 2N
H2804 (2-3 drops) were added. After stirring for 4h at -20°C , the solution was
acidified to pH2 with 2N H2804, warmed to room temperature, cast into CHZClg
(40ml), brine (30ml). The aqueous layer was extracted with CH2CI2 (4 x 30ml),
dried over NaQSO4, and concentrated in vacuo to give 0.849 of a yellow oil
which was purified on a column of silica gel (230-400mesh, 409 silica, 30mm 0.
d., ethyl acetate-hexane 1:8, 20ml fractions) using the flash technique.
Fractions 9-18 provided 640mg, 74%, of 172 as a yellow oil.

1H-NMR (250 MHz): 5:5.86 (m, 1), 5.8 (m, 1), 5.3 (bm, 1), 4.12 (bq, 2),,3.82
(bm, 1), 3.55 (bm, 2), 2.56 (m, 2), 2.2 (s, 3), 2.1-1.6 (m, 6), 1.25 (t, J=7.2-Hz, 3),
1.16 (t, J=7.2Hz, 3); IR (CHCI3): 3010, 2980, 1700, 1420, 1385, 1200, 1125,

 

98

700; El-MS (70eV): 295 (M+, 1.02), 266 (2.35), 249 (24.76), 191 (13.49), 160
(10.62), 141 (33.62), 95 (base), 68 (65.84).

nglizeiign Qf 172.

To a vigorously stirred solution of 172 ( 0.619, 2.07mmol) in cyclohexane
(33ml), was added rapidly HCOzH (7.5ml). After stirring for 3 min. the two
phase mixture was immediately cast into H20 (50ml), and CH2Cl2 (50ml). The
aqueous layer was separated, extracted with CH20|2 (3 x 50ml) and the
combined organic layers were washed with saturated aqueous NaHCO;;
(100ml), brine (100ml), dried over NaZSO4, and concentrated in vacuo to
provide a yellow oil which was purified on a column of silica gel (230—400 mesh,
459, 30mm 0. d., ethyl acetate-hexane 1:6 for 16 fractions ethyl acetate-hexane
1:1 for 12 fractions) using the flash technique. Fractions 7-12 provided 0.269,
51%, of 173 as a yellow oil, and fractions 19-25 provided 0.129, 22%, of 174
as a yellow oil.

Data for 172:

1H-NMR (250MHz): 5:5.79 (m,1), 4.8 (m, 1), 4.45 (bm, 1), 4.08 (q, 2), 2.8 (m,
2), 2.2 (s, 3), 2.2-2.0 (m, 3), 1.23 (m, 30, 1.2 (t, J=7.2Hz, 3); IR (CHCI3): 3005,
2980, 1685, 1470, 1430, 1425, 1335, 1305, 1210, 1115, 1030; EI-MS (709V):
249 (M+, 33.46), 221 (4.65), 193 (4.49), 176 (12.78), 160 (45.64), 148 (27.80),
134 (69.01), 117 (12.00), 105 (12.83), 91 (26.95), 77 (18.59), 68 (13.72), 55
(16.57), 43 (base).

Data for 174:

1H-NMR (250MHz): 5 =4.42 (bm, 1), 4.4 (bm, 1), 4.1 (q, J=7.2Hz, 2), 3.2-2.35
(m, 5), 2.18 (s, 3), 2.1-1.4 (m, 4), 1.22 (t, J=7.2Hz, 3); IR (CHCI3): 3000,2980,
1680, 1695, 1425, 1385, 1425, 1385, 1330, 11170, 1120, 1020; El-MS (709V):
267 (M+, 2.94), 224 (25.89), 206 (1.04), 191 (21.9), 178 (10.99), 168 (3.55), 153

99

(20.78), 135 (10.88), 110 (6.58), 96 (9.00), 82 (31.84), 68 (25.59), 55 (42.76), 43

(base)

To 173 (105.7m9), 0.424mmol) in CH20I2 (3ml) and-saturated aqueous
NaHCO3 (3.5ml) cooled to 0°C in an ice-water bath was added MCPBA (84mg)
in one portion. After stirring for 2h at 0°C the layers were separated, the organic
layer washed with 10% Na28203 (5ml), dried over Na2804, and concentrated
in vacuo to provide a yellow oil which was purified on a column of silica gel
(230-400 mesh, 79, 20mm o.d., ethyl acetate hexane 1:1, 2m| fractions) using
the flash technique. Fractions 9-15 provided 89.4mg, 79.5% of 175 as a yellow
oil.
1H-NMR (250MHz, benzene): 5:5.4 (bs, 0.5), 5.38 (bs, 0.5), 4.6 (m, 1), 4.3 (m,
1), 4.0 (bm, 2), 2.35 (m, 4), 1.8 (s, 30, 1.6-1.3 (m, 4), 1.03 (1, J=7.2Hz, 3); IR
(CHCI3): 3020, 2980, 1685, 1420, 1380, 1330, 1220, 1200, 1125, 925, 720; El-
MS (70eV): 265 (M+. 18.16), 237 (3.44), 22 (6.28), 208 (5.47) 176 (12.95), 150
(30.32), 122 (16.93), 94 (20.28), 68, (22.23), 55 (24.60), 43 (base).

Kinetie ketalizatien ef 175.

To a solution of 175 (195.6mg, 0.74mmol) in CH2CI2 (2ml) cooled to -78° C
in a i-PrOH-dry ice bath was added bis-trimethylsilylethyleneglycol (149.6mg,
0.74mmol), and TMSOTf as a solution in CH2Cl2 (0.0987M, 0.15mi) After
stirring at -78°C for 8h, pyridine (5 drops) was added and the solution was cast
into CH2C|2 (5ml) and saturated aqueous NaH003 (5ml). The aqueous layer
was extracted with CH2Cl2 (3 x 5ml), the combined organic layers dried over
Na2304, and concentrated in vacuo to give a yellow oil which was purified on a

column of silica gel (230-400 mesh, 209, 30mm 0. d., ethyl acetate-hexane 1:2,

 

100

7ml fractions) using the flash technique. Fractions 12-20 provided 107.1mg,
52% of 176 as an amorphous solid.

1H-NMR (250MHz): 5= 5.3-5.14 (bs, 1), 4.9-4.7 (m, 1), 4.1 (m, 7), 2.2-1.7 (m, 6),
1.56 (s, 3), 1.3 (m, 2), 1.03 (1, J=7.2Hz, 3); IR (CHCI3): 3005, 2980, 1780, 1430,
1385, 1335, 1205, 1120, 950, 720; El-MS (709V): 309 (M+, 3.18), 291 (4.67),
266 (9.51 ), 249 (8.38), 220 (5.00), 160 (41.77) 134, (39.78), 43 (base).

 

 

LIST OF REFERENCES

LIST OF REFERENCES

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

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

c) Tanis., S.P.; Chuang, 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.

CT

For reviews of N-acyliminium ion cyclizations, see:

) Speckamp, W.N. Rec/.Trav. Chim. Pays-Bas 1981, 100, 345.

) Zaug, H. E. Synthesis 1984, 85, 181.

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

) For a review of N-acyliminium ions in alkaloid synthesis, see:
Hiemstral, H.; Speckamp, W. N. The Alkaloids Brossi, A., ed.,
Academic Press, Inc. 1988, vol. 2, pp271-344.

0'91

c
d

a) Wurthwein, E. U.; Kupfer, R.; Kaliba Angew. Chem. Supp. 1983,
264.
b) Kupfer, R.; Drolomarbeit, University of Erlangen-Nurnberg, 1983.

Boekelheide, M.; Muller, M.; Jack, J.; Grossnickle, T. T.; Chang, M. J.
Am. Chem. Soc. 1959, 81, 3955.

Tramontini, M. Synthesis 1973, 703.

a) Rischke, H.; Wilcock, J. D.; Winterfeldt, E. Chem. Ber. 1973, 106,
3106.

b) Grob, C. A.; Wohl, R. A. He/v. Chim. Acta 1966, 49, 2175.

c) Bohlmann, F.; Winterfeldt, E. Chem. Ber. 1960, 93, 1956.

Grob, C. A. Angew Chem. 1969, 81, 543.
a) Tscherniac, J. Ger. Pat. 134979, 1904; Chem. Zentr. 1902, II,
1084, 1902.
b) Einhorn, A. Liebigs. Ann. Chem. 1905, 343, 207.
Weinreb, S. M.; Levin, J. I. Heterocyc/es 1979, 12, 949.

a)
b) Hall, H. K.; Miniutti, D. C. Tetrahedron Lett. 1984, 29, 943.‘
c) Jendrzejewski, S.; Steglich, W. Chem Ber. 1982, 114, 1337.

101

102

Lasne, M. C.; Ripol, J. L.; Thuilier, A. J. Chem. Res. (8) 1982,
214.

Krow, G. R.; Pyun, C.; Leitz, C.; Markowski, J.; Ramey, K. J. Org.
Chem. 1974, 39, 2449

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

March, J. Advanced Organic Chemistry 3 ed. John Wiley and
Sons, New York 1985.

James, T. C.;Judd, C. W. J. Chem. Soc. 1914, 105, 1427.
Bohme, H.; Hartke, K. Chem Ber. 1963, 96, 600.

Cohen, T.; Lipoitz, J. J. Am. Chem. Soc. 1964, 86, 2514.

Bose, A. K.; Spiegelman, G.; Manhas, M. S. Tetrahedron Lett.
1971, 3167.

Govindachari, T. R.; Chinnasamy, P.; Rajeswari, S.;
Chandrasekaran,Premila, M.S.; Natarajan, S.; Nagarajan, K.; Pal,
B. R. Heterocyc/es 1984,22, 585.

Lenz, G. R. Synthesis 1978, 489.

Boar, R. B.; McGhie, J. F.; Robinson, M.; Barton, D. H. R.; Horwell,
D. C.; Stick, R. V. J. Chem. Soc. Perkin Trans. l 1975, 1237.
Nyberg, K. Synthesis 1976, 545.

Shono, T.; Matsumura, Y.; Tsubata, K.; Sugihara, Y.; Yamane, S.;
Kanazawa, T.; Aoki, T. J. Am. Chem. Soc. 1982, 104, 6697.
Csizmadia, V. M.; Koshy, K. M.; Lav, K. C. M.; McClelland, R. A.;
Nowlan, V. J.; Tidwell, T. T. J. Am. Chem. Soc. 1979, 101, 974.

For a thorough reniew, see: Shono, T. Tetrahedron 1984, 40,
811

Ross, S. D.; Finkelstein, M.; Petersen, R. C. J. Org. Chem. 1966,
31, 128.

Nyberg, K.; Servin, R. Acta Chem. Scand. Ser.B 1976, 30, 640.
Shono, T.; Hamaguchi, H.; Matsumura, Y. J. Am. Chem. Soc
1975,97, 4264.

Mitzlaff, M.; Warning, K.; Jensen, H. Justus Liebigs. Ann/n. Chem.
1978,171 3.

Palasz, P. D.; Utley, J. H. P.; Hardstone, J. D. J. Chem. Soc.Perkin
Trans. il, 1984, 807.

Chamberlin, A. R.; Nguyen, H. D.; Chung, J. Y. C. J. Org. Chem. 1984,
49, 1682.

999m

8

Gilbert, Synthesis 1972, 30.

Zoller, U.; Ben-lshai, D. Tetrahedron 1975, 31, 836.

Wuonola, M. A.; Woodward, R. B. Tetrahedron 1976, 32, 1085.
Confalone, D. N.; Lo, 8. S. Tetrahedron Lett. 1984, 37, 185.

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

 

20.

21.

22.

23.
24.

103

b) Matsuda, F.; Yanagiya, M.; Matsumoto,T. Tetrahedron Lett
1982,23, 4043.

a) Flitsch, W. Justus Liebigs Annln Chem. 1965, 684, 141.

b) Gouriou, Y.; Fayat, C.; Foucaud, A.; Bull. Soc. Chim. Fr. 1970,
2293.

c) Chiron, R.; Graft, Y. Ibid., 1971, 2145.

d) Wrobel, J. T.; Cybulski, J.; Dabrowski, 2. Synthesis 1977, 686.

e) Dijkink, J.; Speckamp, W. N. Heterocycles 1979, 12, 1147.

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

b) Hubert, J.C.; Steege, W.; Speckamp, W. N. Tetrahedron 1975,
31, 1437.

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

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

f) Shoemaker, H. E.; Dijkink, J.; Speckamp, W. N. lbid. 1978, 34,
163.
a) Dijkink, J.; Shoemaker, H. E.; Speckamp, W. N. Tetrahedron Lett.

1 975, 4043.
) Dijkink, J.; Speckamp, W. N.; Tetrahedron Lett. 1975, 4047.

O"

) Rapopart, H. Lect. Heterocycl. Chem. 1978, 4, 47.

) Falling, S. N.; Rapoport, H. J. Org. Chem. 1980, 45, 1260, and
references cited therein.

) Cook, A. G., Ed. "Enamines"; M. Dekker: New York 1968.

0'9)

0

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

) Hiemstra, H.; Speckamp, W. N. Tetrahedron Lett. 1983,24,
1407.

) Hiemstra, H.; Sno, M. H. A. M.; Vijn, R. J.; Speckamp, W. N. J. Org.
Chem. 1985, 50, 4014.

0'

O

a) Nossin, P. M. M.; Speckamp, W. N. Tetrahedron Lett. 1979,4411.
b) Chamberlin, A. R.; Chung, J. Y. L. lbid. 1982, 23, 2619.

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

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

b) Schoemaker, H. E.; Dijkink, J.; Speckamp, W. N. Tetrahedron,
1978,34, 163.

c) Hubert, J. C.; Wijnberg, J. B. P. A.; Speckamp, W. N. lbid. 1975,
31, 1437. ’

 

25.

26.

27.

28.

29.

30.

31.

32.

104

Evans, D. A.; Thomas, E. W.; Cherpeck, R. E. J. Am. Chem. Soc. 1982,
104, 3695.

Melching, K. H.; Hiemstra, H.; Klaver, W. J.; Speckamp, W. N.
Tetrahedron Lett. 1986, 27, 4799.

a) Koskinen, A. M. P.; Rapoport, H. J. Med. Chem., 28, 1301.
b) Petersen, J. S.; Fels, G.; Rapoport, H. J. Am. Chem. Soc. 1984,
106, 4539. -

For reviews of alkaloids, see:

a) Brossi, A., Ed., The Alkaloids, Academic Press, San Diego, 1988.

b) Pelletier, S. W., Ed., Alkaloids: Chemical and Biological
Perspectives John Wiley and Sons, New York,.(1987)

c) Grundon, M. F.Ed. The Alkaloids, The Royal Chemical Society
Great Britain (1983). Continued as Natural Products Reports.

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

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

Bremmer, M. L.; Khatri, N. A.; Weinreb, S. M. J. Org. Chem. 1983,

48, 3661.

d) Nagao, Y.; Dal, W. -M.; Ochiai, M.; Tsukagoshi, S.; Fujita, E. J. Am.
Chem. Soc. 1988, 110, 289.

e) Haddad, M.; Celerier, J. -P.; Lhommer, G. Heterocycles 1987,
26, 2335.

f) Takahata, H.; Yamabe, K.; Suzuki, T.; Yamazaki, T. lbid.

1986,24, 37.

9.38

For a review of Elaeocarpus alkaloids, see:

a) Hart, N. K.; Johns, S. R.; Lamberton, J. A. Aust, J. Chem,
1972,25, 817.

b) Gribble, G. W.; Switzer, F. L.; Soll, R. M. J. Org. Chem. 1988, 53,
3164.

c) Johns, S. R.; Lamberton, J. A. The Alkaloids; Manske, R. H.
F.,Ed.; Academic, New York, 1 973; 14, pp325-346.

For approaches to the synthesis of cocaine, see:
a) Tufariello, J. J.; Muller, G. B.; Tegeler, J. J.; Trybulski, E. J.; Wong,
8. C.;Ali, Sk. A. J. Am. Chem. Soc. 1979, 101, 2435.
b) Lewin, A. H.; Naseree, T.; Carroll, F. I. J. Heterocyc/ic Chem.
1987,24, 19.
) Willstatter, R.; Wolfes, O.; Mader, H. Ann. Chem. 1923, 434, 111.
) Bazilezshaya, G. l.; Bainova, M. A.; Gura, D. V.; Dyumaev, K. M.;
Preobrazhenskii, N. A.; lzv. Vyssh. Uchebn. Zaved. Khim. Khim.
Technol. 1958, 2, 75; Chem. Abstr. 1959, 53, 423h.
e) Bainova, M. S.; Bazilevskaya, G. I.; Preobrazhenskii, N. A.
Zh.Obshch. Khim. 1960,30, 3258.

c
d

For approaches to the synthesis of anatoxin-a, see.

 

33.

34.
35.

36.

37.

38.

105

a) Danheiser, R. L.; Morin Jr., J. M.; Salaski, E. J. J. Am. Chem. Soc.
1985, 107, 8066.

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

c) Petersen, J. S.; Fels, G.; Rapoport, H. J. Am. Chem. Soc. 1984,
106, 4539.

d) Melching, K. H.; Hiemstra, H. Klaver, W. J.; Speckamp,W. N.
Tetrahedron Lett. 1986,27, 7.99

9) Campbell, H. F; Edwards, O. E.; Edler, J. W; Kolt, R. J. PoIJ.
Chem. 1979,53, 27.

For approaches to the synthesis of perhydrohistionicotoxin, see:

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

b) Corey, E. J.; Arnett, J. F.; Widiger, G. N. J. Am. Chem. Soc.
1975,97, 430.

c) Aratani, M.; Dunkerton, C. V.; Fukuyama, T.; Kishi, Y.; Kakoi, H.;
Sugiura, S.; Inoue, S. J. Org. Chem. 1975, 40, 2009.

d) Fukuyama, T.; Dunkerton, C. V.; Aratani, M.; Kishi, Y. Ibid. 1975,
40, 2011.

e) Evans, D. A; Thomas, E. W.; Cherpeck, R. E. J. Am. Chem. Soc.
1982, 104, 3695.

f) Tufariello, J. J.; Trybulski, E. J. J. Org. Chem. 1974, 39, 3378.

g) Shoemaker, H. E.; Speckman, W,. N. Tetrahedron, 1980, 36,
951.

h) Corey, E. J.; Petrzilka, M.; Ueda, Y. Helv. Chim. Acta 1977, 60,
2294, and references cited therein.

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

For examples of the Mitsunobu reaction, see:

a) Hart, D. J.; Kanai, K. J. Org. Chem. 1982, 47, 1555.

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

c) Hart, D. J. J. Org. Chem. 1982, 46,. 367.

) Schoemaker, H. E.; Boer-Terpstra, J. J.; Dijkink, J.; Speckamp, W.

N. Tetrahedron, 1980, 36, 143.

e) Wijnberg, B. P.; Speckamp, W. N. Tetrahedron Lett. 1981, 22,
5097.

f) Burnett, P. A.; Choi, J.-K.; Hart, D. J.; Tsai, Y. -M. J. Am. Chem.
Soc. 1984,106, 8201.

Hofle, G.; Steglich, W. Synthesis 1972, 619.

Reetz , M. T. ; Huttenhain, S.; Fubner, F. Synth. Commun. 1981 11,
217.

See: Sargent, M. V, Crisp, T. M. in Comprehensive Organic Chemistry,
Sammes, P. (3.; Ed., Permagon Press, Oxford, 1979. Compound 37
was prepared by reacting ethylene oxide with 2- lithiofuran:
Ramanathan, V.; Levine, R. JOrg. Chem. 1962, 27, 1216, orby

 

39.

40.
41.
42.

43

44.

45.
46.

47.

48.

49.

50.

106

hydroboration of 2-vinylfuran: Brown, H. C.; Prasad, J. V. N. V.; Zee, S.
-H. J. Org. Chem. 1986,51, 439.; Brown, H. C.; Prasad, J. V. N. V.; Zee,
S. -H. J. Org. Chem. 1985, 50, 1582.

2-(3-furyl)-1-ethanol was prepared by reacting ethylene oxide with 3-
lithiofuran.

Boekelheide, V.; Lodge, J. P., J. Am. Chem. Soc. 1951, 73, 3681.
Goldberg, S. F.; Lipkin, A. H. J. Org. Chem. 1970, 35, 242.
Williams, P. 0., Ph. D. Dissertation, Michigan State University (1982).

a) Williams, P. D.; LeGoff, E. J. Org. Chem. 1981, 46, 4143.
b) Gingerich, S. B.; Campbell, W. H.; Bricca, C. E.; Jennings, P. W. J.
Org. Chem. 1981, 46, 2589.

For a recent review of Cr Vl-mediated oxidations, see:

a) Cainelli, G.; Cardillo, G., Eds., Chromium Oxidations in Organic
Chemistry, Springer Verlag, 1984.

b) Piancatelli, G.; Scettri, A.; D'Auria, M. Tetrahedron, 1980, 36,
661; and references cited therein.

For a convenient synthesis of PCC, see:

0) Corey, E. J.; Suggs, J. W. Tetrahedron Lett. 1975, 16, 2647.

d) Piancatelli, G.; Scettri, A.; D'Auria, M. Tetrahedron Lett. 1979,
1507.

Corey, E. J.; Mehrotra, M. M. Tetrahedron Lett. 1985, 26, 2411.

a) Levisalles. J. Bull. Soc. Chim. France 1957, 997.
b) Elming, N. Adv. Org. Chem. 1960, 67, 115.
c) Lee, T. -J. Tetrahedron Lett. 1979,20, 2297/

Hirsch, J. H.; Szui, A. J. J. Heferocyc/ic Chem. 1972, 9, 523.
Floyd, M.B. J. Org. Chem. 1978,43,1641.

MacLeod, J. K.; Bott, G.; Cable, J. Aust, J. Chem. 1977, 30,
2561.

99.8;

a) Wiesner, K.; Tsai, T. Y. R.; Kumai, R.; Sivaramakrichnan, H. Helv.
Chim.Acta 1984, 67, 1128.

b) Wiesner, K.; Tsai, T. Y. R.; Jin, H. Ibid. 1985, 68, 300.

a) Wiesner, K.; Tsai, T. Y. R.; Sen, A.; Kumar, R.; Tsubuki, M. Helv.
Chim. Acta 1983,66, 2632.

b) Ferland, J. M. Can. J. Chem. 1974, 52, 1652.

c) Ferland, J. M.; Lefebvre, Y.; Deghenghi, R. Tetrahedron Lett.
1966, 3617.

LePage, L. Lepage, Y. Synthesis 1983, 1018.

 

51.

52.

53.

54.
55.
56.

57.
58.

107

a) Schaap, A. P., Ed., Single Molecular Oxygen Dowden,
Hutchinson and Ross, Stroudesburg, Pennsylvania, 1976.

b) Wasserman, H. H.; Murray, R. W., Eds, Singlet Oxygen Academic
Press, New York 1979.

For reviews on the application of singlet oxygen in synthesis, see:

c) Ohloff, G. Pure Appl. Chem 1975,43, 481.

d) Matsumoto, M.; Londo, K. J. Syn. Org. Chem. Jap. 1977, 35,
188. , .

e) Wasserman, H. H.; Ives, J. L. Tetrahedron 1980, 37, 1825.

a) Wasserman, H. H.; Liberles, A. J. Am. Chem. Soc. 1960, 82,
2086.

b) Foote, C. S.; Wexler, S. J. Am. Chem. Soc. 1964, 86, 3879.

c) Foote, C. S.; Wexler, S. Ibid. 1964, 86, 3880.

d) McKeown, E.; Waters, W. A. J. Chem. Soc. 1966,.8, 1040.

e) Foote, C. S.; Wexler, S.; Ando, W.; Higgins, R. J. Am. Chem. Soc.
1968, 90, 975.

f) Foote, C. S.; Wuestfogg, M. T.; Wexler, S.; Burstain, I. G.; Denny,R.

Tetrahedron 1967,23, 2583.
g) Naya, K.; Matsuura, T.; Makiyama, M.; Tsumura, M. Heterocycles,

1978, 10, 177.
h) Fukuda, H.; Takeda, M.; Sato, Y.; Mitsunobu, 0. Synthesis, 1978,
368.

a) Foote, C. 8.; Peterson, E. R.; Lee, K. -W. J. Am. Chem. Soc.
1972, 94, 1032.

b) Gollnick, K.p; Griesbeck, A. Angew. Chem. Int. Ed. Engl. 1983,
22, 726.

c) Katsumura, S.; Hori, K.; Fujiwara, S.; lsoe, S. Tetrahedron Left.
1985,26, 4625.

d) Gollnick, K.; Schnatterer, A. Tetrahedron, Lett. 1985,26, 5029.

Corey, E. J.; Mehrotam, M. M. J. Am. Chem. Soc. 1984, 106, 3384.
Tanis, S.P.; Head, D. B. Tetrahedron Left. 1984, 25, 4451.

Kuwajima, l.; Urabe, H. Tetrahedron Lett. 1981, 22, 5191.
Adam, W.; Rodriguez, A. Ibid. 1981,22, 3503.

deGroot, A.; Jansen, B. J. M. J. Org. Chem. 1984, 49, 2034.
Takano, Y.; Yasuda, A.; Urabe, H.; Kuwajima, l. Tetrahedron Lett.
1985,26, 6225.

9994

Frick, W.; Simchen, G. Synthesis, 1984, 929.

Compound 57 was prepared by treating the corresponding 2-(2-furyl)-
ethanol with n-BuLi (2 equiv.), TMSCI (2 equiv.), followed by
NaOMe/MeOH cleavage of the siloxy group.

 

59.

60.

61.
62.

63.

64.

65.

108

Compound 60 was prepared by reacting ethylene oxide with 5-methyl-

2-lithio-furan

a) Masumune, T.; Ono, M.; Matsue, H. Bull. Chem. Soc. Jpn. 1975,
4 1.

b) Buchi, G.; Wuest, H. J. Org. Chem 1966,31, 977.

For a discussion, see: Eliel, E. L.; Allinger, N. L.; Angyal, S. J.;
Morrison, "Conformational Analysis", Wiler-Interscience, New York,
1965, pp 5-12. For applications of the concept of tortional angle to
conformational descriptions, see: Bucourt, R. Top. Stereochem. \
1974, 8, 159.

Johnson, F. Chem. Rev. 1968,68, 375.

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

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

For a preparation of bistrimethylsilyI-ethylene glycol, see:

c) Langer, S. H.; Connell, S.; Wender, l. J. Org. Chem. 1957, 23, 50

Cole, J. E.; Johnson, W. S.; Robins, P. A.; Walker, J. J. Chem. Soc.
1967, 244.

) Roberts, J. D.; Moreland Jr., W. T.; Frazer, W.

) Hauptmann, H.; Walter, W. F. Chem. Rev. 1962, 342.

) Rao, P, N. J. Org. Chem. 1971, 36, 2427.

) Hatch, R. P.; Shringarpure, J.; Weinreb, S. M. Ibid. 1978, 43,
4172.

QOUDJ

Netter, H. Helv. Chim. Acta 1981, 64, 761.

Emmons, W. D.; Lucas, G. B. J. Am. Chem. Soc. 1955, 77, 2287.

Fung, 8.; Siccall, J. B.

Suzuki, M.; Takada, H.; Noyori, R. J. Org. Chem. 1982, 47, 902.

McClure, J. D.; Williams, P. H. J. Org. Chem. 1962, 27, 24.

Matsubara, S.; Takai, K.; Nozaki, H. Bull. Chem. Soc. Jpn. 1983,

56, 2029.

g) Grieco, P. A.; Yoloyama, Y.; Gilman, S.; Ohfune, Y. J. Chem. Soc.
Chem. Comm. 1977, 870.

h) Jacobson, S.; Mares, F.; Zambri, P. M. J. Am. Chem. Soc. 1979,

35339.38

101, 6938.

i) Rastetter, W. H.; Richard, T. J.; Lewis, M. D. J. Org. Chem. 1978,
43, 3163.

)) Tsunokawa, Y.; Iwasaki, S.; Okudo, S. Chem. Pharm. Bull. 1983,
31 , 4578

k) White, R. W.; Emmons, W. D. Tetrahedron 1962, 17, 31.

I) Corey, E. J.; Arnold, 2.; Hutton, J. Tetrahedron Lett. 1970, 307.
m) Chambers, R. P.; Clark, M. Tetrahedron Left. 1970, 2741.

11) Gibson, D. H.; Joseph, J. T. Tetrahedron Left. 1972, 3483; ‘

o) For reviews on the Baeyer-Villiger oxidation, see:

 

66.

67.

68.

69.

70.

71.
72.

73.

74.

75.

76.

109

Hassall, C. H. Org. React. 1957, 9, 73.
p) Lewis, S. N. In "oxidation", Augustine, R. L., Ed.; Marcel Dekker:
New York, 1969; vol. 1, pp213.

For a review of thioketalizations, see:
Patai, S. The Chemistry of the Thiol Group Wiley, New York,
1974, vol. 2, pp521-532.

For a preparation of trimethylsilylthiols, see:

a) Evans, D. A.; Truesdale, L. K.; Grimm, K. G. Nesbitt, S L.; J. Am.
Chem. Soc. 1977’, 99, 5009

b) Glass, R. S. J. Organmet. Chem. 1973, 61, 83.

Olah, G. A.; Narang, S. C.; Meidar, D.; Salem, G. F. Synthesis 1981,
282.

a) Caputo, R.; Ferreri, C.; Noviello, S.; Palumbo, G. Synthesis,
1986, 499.
b) Caputo, R.; Ferreri, C.; Palumbo, G. Synthesis 1987, 386.

Corey, E. J.; Shimoji, K. Tetrahedron Lett. 1983,24, 169.
Ong, B. 8.; Chan, T. H. Synthetic Commun. 1977, 7, 283.

a) Morton, D. R.; Hobbs, S. J. J. Org. Chem. 1979, 44, 656.

b) Abel, E. W.; Armitage, D. A.; Bush, R. P. J. Chem Soc. 1965,
3045.

c) Abel, E. W.;Armitage, D. A.; Bush, R. P. Ibid. ,7098.

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

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

a) Witkop, B. Experentia 1971, 27, 1121

b) Daley, J. W.; Karle, l. L.; Meyers, W.; Tokuyama, T.; Walters, J. A.;
Witkop, J. A.; Witkop, B. Proc. Natl. Acad. Sci. U. S. A. 1971, 68,
1870.

c) Tokuyama, T.; Uenoyama, K; Brown, G. Daly, J. W.; Witkop, B.
Helv. Chim. Acta 1974, 57, 2597.

d) Karle, I. L. J. Am. Chem. Soc. 1973, 95, 4036.

e) Daly, J. W.; Tokutama, T.; Fujiwara, T.; Higher, R. J.; Karle, I. KL.
Ibid. 1980, 102, 830.

For recent reviews, see:

a) Inubushi, Y.; lbuka, T. Heterocycles, 1977, 8, 633.

b) Daly,J. W. Brwon, A. B, Mensah- Dwumah, M.; Meyers, C. W.
Toxicon.1978, 16,163.

 

77.

78.

79.
80.

81.
82.
83.

84.
85.

86.
87.

88.

89.

110

For pertinent model studies directed toward the synthesis of the 1-

azaspiro[5.5]undecane ring system, see:

a) Corey, E. J.; Balanson, R. D. Heterocycles 1976, 5, 445.

b) Bond, F. T.; Stemke, J. E.; Powell, D. W. Synth. Commun. 1975,
5, 427.

c) Gossinger, E.; Imbof, R.; Wenrli, H. Helv. Chim. Acta 1975,58,

d) Tufariello, J. J.; Trybulski, E. J. J. Org. Chem. 1974, 39, 3378.

3) Elliot, 8.; Raftery, M. A. Biochemistry 1979, 18, 1968.

b) Albuquerque, E. Y.; Kuba, K.; Daly, J. W. J. Pharmacol Exp.
Therap. 1974, 189, 513.

c) Albuquerque, E. Y.; Barnard, E. P.; Chui, T. H.; Lapa, A. J.; Dolly,
J.O.; Jansson, S.; Daly, HJ.; Witkop, B. Proc. Natl. Acad. Sci USA
1973, 70, 949.

d) Takahashi, K.; Witkop, B.; Brossi, A.; Malequie, M.; Albuquerque,
E. Y. Helv. Chim. Acta 1982,65, 252.

Carey, S. C.; Aratani, M.; Kishi, Y. Tetrahedron Lett. 1985, 26, 5887.

a) Keck, G. E.; Webb, R. Tetrahedron Lett. 1979, 1185.
b) Keck, G. E.; Webb, R.; Yars, J. B. Tetrahedron 1981, 37, 4007.

Rice, K. L.; Dyer, J. R. J. Heterocycl. Chem. 1975, 1325.
Lee, T. -J. Tetrahedron Lett. 1979, 2297.

Johnson, W. S.; Li, T, -T.; Harbert, C. A.; Bartlett, W. R.; Herrin, T. R.;
Staskum, 8.; Rich, P. H. J. Am. Chem. Soc. 1970, 92, 4461.

MacLeod, J. K.; Bott, G.; Cable, J. Aust. J. Chem. 1977, 30, 2561.

Achmatowicz Jr., 0.; Bukowski, P.; Szechner , B.; Zwierzchowska,
2.; Zamojski, A. Tetrahedron 1971,27, 1973.

Kametani,T.;Nemoto, H.;Fukumoto, K. Heterocycles, 1974, 2, 639.

3) Mitchell, R. H.; Lai, Y. H.; Williams, R. V.; J. Org. Chem. 1979, 44,
4733.

b) Aiello, E.; Dattolo, G.; Cirrincione, G.; Almerico, A. M.; D'Asdia, l. J.
Heterocycl. Chem. 1982, 19, 977.

c) Antonioletti, R.; D'Auria, M.; DeMico, A.; Piancatelli, G.; Scettr,
A.Synfhesis 1984, 280.

d) Scettri, A. Tetrahedron 1985,41, 5141

Kamitori, Y.; Hojo, M.; Masuda, R.; Kimura, T.; Yoshida, T. J. Org.Chem.
1986,51, 1427.

For a review of activated DMSO reagents and their use in oxidations,

 

90.

91.

92.

93.

94.

95.
96.

97.
98.

99.

111

see:
a) Mancuso, A. 8.; Swern, D. Synthesis, 1981, 165.
b) Omura, K.; Swern, D. Tetrahedron 1978, 34, 1651.

a) Bowden, K.; Heilbron, I. M.; Jones, E. R. H.; Weedon, B. C. L. J.
Chem. Soc. 1946, 39.

b) Bowers, A.; Halsell, T. G.;Jones, E. R. H.; Lemin, A. J. Ibid. 1953,
2548.

a) DiPalma, J. R. Drill's Pharmacology in Medicine 4 ed.; McGraw-
Hill, New York; pp206 and 208.

b) Grabowski, J., Ed., Cocaine Pharmacology, Effects, andTreafmenf
of Abuse, National Institute of Drug Abuse Research Monograph
50, Superintnedent of Documents, U. S. Government Printing
Office, Washington, D. C. 1984.

For reviews on tropane alkaloids, see:

a) Holmes, H. L. The Alkaloids, vol. 1; Manske, R. H. F. Ed.;
Academic Press: New York, 1950; chapter 6.

b) Fodor, G. Ibid., Vol VI; 1960; chapter 5.

c) Ibid.; Vol. IX; 1967; chapter 7.

d) Ibid.; Vol. XIII; 1971; chapter 8.

e) Clarke, R. L. Ibid., Vol. VXI; 1977; chapter 2.

f) Alkaloids (London) 1971, 1, 55. 1972, 2, 54. 1973, 3,, 67.
1974, 4, 78. 1975, 5, 65. 1977, 7, 47

Robinson, R. J. J. Chem. Soc. 1917, 111, 762. For a mechanistic
treatmetn of related reactions, see Paquette, L. A.; Hermaster, J. W. J.
Am. Chem. Soc. 1966, 88, 763.

a) Macdonald, T. L.; Dolan, R.
b) see 34a, and references cited therein.

de Jong, A. W. K. Reel. Trav. Chim. Pays-Bas 1940, 59, 27.

a) White, D. A.; Baizer, M. M. Tetrahedron Left. 1973, 3597.
b) Miyakoshi, T.; Saito, S. Yakugaku, 1982, 31, 35.

Hass, H. B.; Susie, A. G.; Heider, R. L. J. Org. Chem. 1950, 15, 8.

Greene, T. W. Protective Groups in Organic Synthesis John Wiley
and Sons, New York, 1981.

a) 50% CFCO2H/CHCI3-H20: Ellison, R. A.; Lukenbach, E. R.; Chiu,
C. -W. Tetrahedron Lett. 1975, 499.

b) H2O or 10% (C02H)2 or 15% H2804 absorbed on silica: Huet,
F.; Lechevallier, A.; Pellet, M.; Conia, J. M. Synthesis, 1978, 63.

c) THF, 0.7N HCI: Wenkert, E.; Goodwin, T. E. Synthesis
Commun. 1977, 7, 409.

100.

101.

102.

103.

104.

105

106.

107.
108.

109.

38

112
HCOzH, CC5H12: ref. 1a, 1C.
Sharpless, K. 8.; Young, M. W. J. Org. Chem. 1975, 40, 947.

Grieco, P. A.; Gilman, 8.; Nishizawa, M. J. Org. Chem. 1976, 41,
1485.

Almansa, C.; Carceller, E.; Moyano, A.; Serratosa, F. Tetrahedron
1986,42, 3637.

vv

a
b

Bauer, H. Chem. Ber. 1913,46, 92.
Waitkins, C. R.; Shutt, R. lnorg. Synth. 1946, 2, 186.

Carroll, F. I.; Coleman, M. L.; Lewin, A. H. J. Org. Chem. 1982,
47, 13.
Chazin, W. J.; Colebrook, L. D. J. Org. Chem. 1986, 51, 1243.

Baily, P. S., Ozonafion in Organic Chemisty, Academic Press,
New York, 1978.

Scheiber, S. L.; Claus, R. E.; Reagan, J. Tetrahedron Lett. 1982,
23, 3862.

Fremery, M. |.; Fields, E. R. J. Org. Chem. 1963, 28, 2537.
Henne, A. L.; Hill, P. J. Am. Chem. Soc. 1943, 65. 752.
Neumeister, J.; Keul, H.; Mahendra, P. 8.; Griesbaum, K. Angew.
Chem. Int. Ed. Engl. 1978, 17, 939.

Beckett, A. H.; Harper, N. J.; Balon, A. D. J.; Watts, T. H. E.
Tetrahedron 1954, 6, 319.

Nickon, A.; Fieser, L. F. J. Am. Chem. Soc. 1952, 74, 5566.
Fodor, G. Tetrahedron 1957, 1, 86.

Hayakawa, Y.; Noyori, R. Bull. Chem. Soc. Jpn. 1974, 47, 2617.
Mirza, R. Nature, 1952, 170, 630.

For a discussion, seezHouse, H. 0. Modern Synthetic Reactions
2 ed., W. A. Benjamin |nc,.Menlo Park, CA. 1972.

lhara, M.; Katogi, M.; Fukumoto, K.; Kametani, T. J. Chem. Soc.
Chem. Commun. 1987, 721.

Charles, J. T.; Huffman, J. W. J. Am. Chem. Soc. 1968,90, 6486.
For a review, see:

Kaiser, E. M. Synthesis, 1972, 391.

Fung, V. A.; Degraw, J. I. Synthesis 1976, 311.

a)

b)

For the preparation of nitro—olefins, see:

Gairand, C. B.; Cappin, G. R. J. Org. Chem. 1953, 18, 1.

For the reduction of nitro-olefins, see:

Sinhababu, A. K.; Borehardt, R. Tetrahedron Lett. 1983, 24, 227.

Osby, J. D.; Ganem, B. Tetrahedron Left. 1985, 26, 6413.

110.
111.

112.
113.
114.

115.

116.

117.

118.
119.
120.
121.
122.

123.

124.

113
Ratcliffe, R.; Rodehors, R. J. Org. Chem. 1970, 35, 4000.

Corey, E. J.; Gilman, N. W.; Ganem, B. E. J. Am. Chem. Soc. 1968, 90,
5616.

Wuts, P. G. M.; Bergh, C. L. Tetrahedron Lett. 1986,27, 3995
Moore, M. L. Org. React. 1949, 5, 301.

a) Carmichael, W. W.; Biggs, D. F.; Gorham, P. R. Science
Washington DC 1975, 542.

b) Devlin, J. P.; Edwards, 0, E.; Gorham, P. R.; Hunter, N. R.; Pike, R.
K.; Stavric, B. Can. J. Chem. 1977, 55, 1367.

Moore, R. E. Bioscience 1977,27, 797.

Carmichael, W. W.; Gorham, P. R. In "Algae Biomass"; Shelef, G.;
Seder, C. J.; Eds.; Elsevier, Amsterdam, 1980, pp437-448 and
references cited therein.

38

a) Carmichael, W. W.; Biggs, D. F.; Peterson, M. A. Toxicon 1979,
17, 229.

b) Spivak, C. E.; Waters, J.; Witkop, 8.; Albuquerque, E. X. Mol.
Pharmacol 1983,23, 337.

a) Seyden-Penne, J.; RouxpSchmitt, M. C.; Roux, A. Tetrahedron

1970,26, 2649.

) Borsch, R. F.Tetrahedron Lett. 1972, 3761.

c) Oka, Y.; Motuhashi, M.; Sugihara, H.; Miyashita, 0.; Itoh, K.;
Nishikawa, M.; Yurugi, S.Chem. Pharm. Bull. 1977., 25, 632.

d) Rathke

e) Schreibler, H.; Tutundzitsch, R. 8. Chem. Ber. 1931, 2916.

f) Ballester, M. Chem. Rev. 1954, 283.

Corey, E. J.; Schmidt, G. Tetrahedron Lett. 1979, 399.

Brown, H. C.; Kulkarni, 8. V.; Rao, C. G. Synthesis, 1980, 151.

Schiehser, G. A; White, J. D. J. Org. Chem. 1980, 45, 1864.

Yoneda, F.; Suzuki, K.; Nltta, Y. J. Org. Chem. 1967, 32, 727.

a) For a review of the use of nitroacetic acid esters in synthesis, see:
Shipchandler, M. T. Synthesis 1979, 666.

b) Mitsunobu, 0.; Yoshida, N. Tetrahedron Lett. 1981, 22, 2295.

Firouzabadi, H.; Ghaderi, E. Tetrahedron Lett. 1978, 839.

Pletcher, D.; Tait, S. J. D. Tetrahedron Lett. 1978, 1601.

 

125.
126.
127.

128.
129.

130.
131.
132.

133.
134.
135.

136.

137.

138.

139.

140.
141.
142.

143.

114
Lee. G. A.; Freedman, H. H. Tetrahedron Lett. 1976, 1641.

Wittig, G.; Knauss, E. Angew. Chem. 1959, 71, 127.

Colvin, G. W.; Raphael, R. A.; Roberts, J. 8. J. Chem. Soc. Chem.
Commun. 1971, 858.

Jacobson, R. M. Tetrahedron Lett. 1974, 3215.

For a review of the Net reaction, see:
Noland, W. E. Chem. Rev. 1955,55, 137.

Dubs, P.; Stussi, R. Helv. Chim. Acta 1978, 61, 990.
Steliou, K.; Doupart, M. -A. J. Org. Chem. 1985, 50, 4971.

Kornblum, N.; Erickson, A. 8.; Kelly, WI JI; Henggeler, B. J. Org. Chem.
1982,47, 4534.

McMuray, J. E.; Melton, J. J. Org. Chem. 1973, 38, 4367.
Johnson, K.; Degering ED. F. J. Org. Chem. 1943, 8, 10.

Olah, G. A.; Arvanaghi, M.; Vankar, Y. D.; Prakash, G. K. 8. Synthesis,
1980, 662.

Vankar, D. 8.; Rathore, R.; Cahndrasekaran, 8. Synth. Commun.
1987, 17, 195.

a) Kienzle, F.; Holland, G. W.; Jernow, J. L.; Kwon, 8.; Rosen, P. J.
Org. Chem. 1973, 38, 3440.

b) Ballini, R.; Petrini, M. Synth. Commun. 1987, 17. 543.

For a furanacetaldehyde synthesis that appeared after this study, see:

Chikasita,

Synth. Commun. 1987, 17, 677.

Seebach, D.; Beck, A. K.; Mukhopadhyah, T.; Thomas, E. Helv. Chim.
Acta 1982, 65, 1101.

Sprang, Ca.; Degering, ED. F. J. Am. Chem. Soc. 1942, 64, 1063.
Wollenberg, R. H.; Miller, S. J. Tetrahedron Lett. 1978, 35, 3219.
Rosini, G.; Ballini, R.; Petrinin, M. Synthesis 1985, 269.

a) Kornblum, N; Larson, H. 0.; Blackwood, R. K.; Mooberry, D. D.;
Oliveto, E. P.; Graham, G. E. J. Am. Chem. Soc. 1956, 78, 1497.

 

115

b) for a related preparation, see: Crumble, R. L.; Nimitz, J. 8.;
Mosher, H. 8. J. Org. Chem. 1982, 47, 4040.

144. Moffett, R. B. Org. Synth. IV 652.
145. 8till,WC.; Mitra,A.;Khan, M. J. Org. Chem. 1978,41, 2923.

 

 

 

 

"IlllllllllIlIlllIIIIlI