Total synthesis of polyoximic acid

Article abstract of DOI:10.1139/v01-171

The structure and stereochemistry of polyoximic acid, a degradation product of polyoxins, was originally designated as trans-3-ethylidene-L-azetidine-2-carboxylic acid. However, total synthesis revealed that the correct structure was in fact cis-3-ethylidene-L-azetidine-2-carboxylic acid, which was confirmed by X-ray crystallography. The synthesis of the trans-isomer was also done and its identity was confirmed by X-ray analysis as well. The key step for constructing the four-membered ring was a rhodium catalyzed carbenoid insertion into the N - H bond of a β-amino acid derivative. The stereoselectivity of the exo-double bond was controlled by conducting a Horner-Emmons-Wadsworth or a Wittig reaction to generate the trans- and cis-isomers, respectively. Weinreb's amide was used as a latent methyl group for the separation of trans and cis mixtures. The double bond stereochemistry of polyoximic acid in the parent polyoxin was also confirmed to be cis by extensive 2D NMR studies.

Full text of DOI:10.1139/v01-171

1812
Total synthesis of polyoximic acid
Stephen Hanessian and Jian-min Fu
Abstract: The structure and stereochemistry of polyoximic acid, a degradation product of polyoxins, was originally
designated as trans-3-ethylidene-L-azetidine-2-carboxylic acid. However, total synthesis revealed that the correct struc-
ture was in fact cis-3-ethylidene-L-azetidine-2-carboxylic acid, which was confirmed by X-ray crystallography. The syn-
thesis of the trans-isomer was also done and its identity was confirmed by X-ray analysis as well. The key step for
constructing the four-membered ring was a rhodium catalyzed carbenoid insertion into the N—H bond of a b-amino
acid derivative. The stereoselectivity of the exo-double bond was controlled by conducting a Horner-Emmons-
Wadsworth or a Wittig reaction to generate the trans- and cis-isomers, respectively. Weinreb’s amide was used as a la-
tent methyl group for the separation of trans and cis mixtures. The double bond stereochemistry of polyoximic acid in
the parent polyoxin was also confirmed to be cis by extensive 2D NMR studies.
Key words: diazoinsertion, azetidine, olefination.
Résumé : La structure et la stéréochimie originalement attribuées à l’acide polyoximique, un produit de dégradation
des polyoxines, étaient celles de l’acide trans-3-éthylidène-L-azétidine-2-carboxylique. Une synthèse totale a toutefois
permis de démontrer qu’il s’agit plutôt de son isomère, l’acide cis-3-éthylidène-L-azétidine-2-carboxylique, ce qui a été
confirmé par diffraction des rayons X. On a aussi réalisé la synthèse de l’isomère trans et son identité a aussi été
confirmée par diffraction des rayons X. L’étape clé dans la construction du cycle à quatre chaînons est une insertion de
carbénoïde dans la liaison N—H d’un dérivé d’acide b-aminé, catalysée par le rhodium. La stéréosélectivité de la
double liaison exo est contrôlée en procédant à une réaction d’Horner-Emmons-Wadsworth ou d’une réaction de Wittig
qui conduisent respectivement aux isomères trans et cis. L’amide de Weinreb a été utilisé comme groupe méthyle latent
pour la séparation des mélanges trans et cis. En se basant sur des études approfondies de RMN 2D, on a aussi
confirmé que la stéréochimie de la double liaison de l’acide polyoximique de la polyoxine de base est aussi cis.
Mots clés : diazoinsertion, azétidine, oléfination.
[Traduit par la Rédaction] 1826
Introduction
Hanessian and Fuhydroxymethyl uracil (3), polyoxin C (4), polyoxamic acid
(5), and polyoximic acid (1). The latter is a unique amino
acid that is a component of polyoxins A, F, H, and K. Total
syntheses of polyoxin C (9), polyoxins B and D (10),
polyoxins J and L (11), and polyoxamic acid (12) have been
reported. Only two reports in the literature address the total
synthesis of polyoximic acid (13, 14). Baumann and Duthaler
(13) described an approach using a Claisen rearrangement or
cycloaddtion reaction for the synthesis of key intermediates.
Emmer (14) reported a total synthesis of racemic trans-
polyoximic acid using a rhodium catalyzed intramolecular
carbenoid insertion reaction.
Initially, racemic polyoximic acid (rac-1) was obtained
from polyoxin A by either basic or acidic hydrolysis
(Scheme 2) (5). Upon ozonolysis, the characterizable product
6 was produced, thus establishing the atomic composition of
the molecule. The stereochemistry of the exocyclic double
bond was established as trans based on an NOE study using
a 60 MHz NMR instrument. Additionally, hydrogenolysis of
the parent polyoxin A produced the saturated intermediate
In the 1960’s, Isono and co-workers (1a, c–j) first isolated
and characterized a series of nucleoside antibiotics desig-
nated as the polyoxins from culture broths of Streptomyces
cacacoi var asoensis. Since then, new members were added
to this family that constitute the fifteen known polyoxins
A–O (2). These compounds are potent inhibitors of chitin
synthetase and could be used as fungicides for treating plant
diseases (3) and other applications (for other biological uses
of polyoxins, see ref. 4). Consequently a great deal of effort
has been devoted to the structural elucidation (5),
biosynthesis (6), structural–activity relationships (7), and to-
tal synthesis (8) of this small family of antibiotics. The
structures and configurations of the polyoxins, represented
by polyoxin A and its degradation product polyoximic acid
(1), were proposed as depicted in Fig. 1.
The structural elucidation was based on the analysis of the
degradation products (5). For example, treatment of polyoxin
A (Scheme 1) with acid or base led to the generation of 5-
Received May 10, 2001. Published on the NRC Research Press Web site at http://canjchem.nrc.ca on November 21, 2001.
Dedicated to Professor Victor Snieckus: a friend, scholar, and chemist extraordinaire.
S. Hanessian.1 Department of Chemistry, Université de Montréal, Montréal, QC H3C 3J7 Canada.
J.-m. Fu. Pharmacia Corp. 7255-209-706, Structural, Analytical and Medicinal Chemistry, 301 Henrietta St., Kalamazoo, MI 49007, U.S.A.
1Corresponding author (telephone: (514) 343-6738; fax: (514) 343-5728; e-mail: stephen.hanessian@umontreal.ca).
Can. J. Chem. 79: 1812–1826 (2001)
© 2001 NRC Canada
DOI: 10.1139/cjc-79-11-1812

Hanessian and Fu
1813
Fig 1.
5''
O
COOH
O
6''
2''
7
Me
CH2OH
4
HN 3
4''
5
6
N
OH
O
2
O
5
COOH
6
5'''
O
N
1
5'
4'
2
Me
O
4'''
2'''
3'''
1'''
H2N
O
N
H
1'
4
NH
2'
OH
NH2
3'
OH OH
1. trans-Polyoximic acid
Polyoxin A (original structure)
6''
Me
O
COOH
5''
2''
7
CH2OH
4
HN 3
4''
6
Me
5
N
O
OH
O
2
O
6
5
COOH
5'''
O
N
1
5'
2
O
4'''
2'''
3'''
1'''
H2N
O
N
H
4'
1'
4
NH
2'
OH
NH2
3'
OH OH
2. cis-Polyoximic acid
Polyoxin A (revised structure)
Scheme 1.
Polyoxin A
or 3 N HCl
reflux, 1 h
0.5 N NaOH
65oC, 4 h
O
N
CH2OH
COOH
HN
O
COOH
H2N
COOH
O
CH2OH
Me
O
HN
OH
H2N
NH
HO
O
N
H
OH OH
CH2OH
1
3
4
5
dihydropolyoxin A 7, and subsequent basic hydrolysis gen-
erated the dihydro polyoximic acid 8, which was converted
to its N-dithiocarbethoxy derivative whose positive Cotton
effect was compared to that of the known analogous deriva-
tive prepared from L-azetidine-2-carboxylic acid. Thus, the
absolute configuration of the amino acid was established as
(S) (5).
In the course of our studies (15a) directed at the total syn-
thesis of polyoximic acid, we discovered that the structure
was originally assigned incorrectly as the trans-isomer (15b).
Here we wish to report in detail the total synthesis of this
four-membered amino acid as the cis-isomer, as well as a
synthesis of the trans-isomer.
Results and discussion
We envisioned that the total synthesis of polyoximic acid
could be accomplished from D-serine, as illustrated by the
disconnections shown in Scheme 3. The carboxylic acid could
be derived from a latent functional group like an alcohol 9,
and the exo-double bond would originate from a Wittig reac-
tion on the azetidinone 10. The four-membered ring of 10
© 2001 NRC Canada

1814
Can. J. Chem. Vol. 79, 2001
Scheme 2.
O
COOH
COOH
O
Me
Base
or
O3
Me
Polyoxin A
NR'
NH
rac-1
Acid
6
H2/Pd-C
COOH
COOH
Me
Me
Base
NH
O
N
8
Dihydro polyoxin A, 7
Scheme 3.
O
X
COOH
OR
OR
Me
NR'
10
NR'
NH
9
1
O
O
OH
OR
HO
NH2
NHR'
11
N2
12
Scheme 4.
O
O
O
CH2R
Method Aa
or Method Bb
Rh2(OAc)4
CH2Cl2
*
N2
CH2R
CH2R
*
*
HO
NPG
temperature
NHPG
NHPG
13
14
15
would be constructed by an intramolecular rhodium catalyzed
carbene insertion into the N—H bond of the diazoketone,
which in turn could be generated from the amino acid D-serine
(12). This approach to the synthesis does not require us to
build any new stereogenic centers on an sp3 carbon, and relies
on the general design principles embodied in the “Chiron”
approach (16).
Thus, the formation of azetidinone 10, from the rhodium
catalyzed intramolecular carbenoid insertion reaction, would
be the pivotal step of the total synthesis. Although rhodium
catalyzed carbene insertions to C—H and O—H bonds have
been extensively studied (17), to our knowledge, there was
no precedence in the literature of using rhodium catalysis to
promote an intramolecular carbene insertion into an N—H
bond to form a four-membered ring from enantiopure a-amino
acids (18). Hence, a methodological study was carried out
first on simple amino acid templates such as phenylalanine
and alanine (Scheme 4, Table 1) to explore the appropriate
conditions. The diazoketones 14a–f were prepared from dif-
ferent amino acids by reacting with diazomethane via either
a mixed anhydride (method A) or an acyl chloride
(method B). Subjection of the phenylalanine diazoketone
14a to catalysis by rhodium(III) at different temperatures (rt,
–20, –40, –78°C) revealed that –40°C was the optimal tem-
perature giving a 70% yield of the expected 3-azetidinone.
This temperature (–40°C) appeared to work well for the
alanine system (Table 1, entries 6 and 7) with Cbz and Ts as
the amino protecting groups. In contrast, when the amino
protecting group was changed to Boc, the reaction led to the
formation of unidentified materials (Table 1, entries 5 and 9)
and no product could be isolated. Normally, carbene inser-
tion reactions are carried out at rt or higher (17).
These low temperature conditions were then applied to the
synthesis of polyoximic acid, starting with D-serine (12). In
sharp contrast to the phenylalanine and alanine systems, the
Boc group emerged as a better amino protecting group than
Cbz (Scheme 4, Table 1, entries 10 and 11). In addition, the
TBDPS (19) hydroxyl protecting group was found better
than TBDMS (20). From the rhodium catalyzed reaction of
diazoketone 14h, two side products 16 and 17 were isolated
in addition to the desired product 15h. Compound 16 was
formed by insertion of the carbene into the oxygen—silicon
© 2001 NRC Canada

Hanessian and Fu
Table 1.
1815
Product 14
(yield, %)
Temp.
(°C)
Product 15
(yield, %)
Entry
R
PG
Configuration
Methoda
1
2
3
4
5
6
7
8
9
10
11
12
13
Ph
Ph
Ph
Ph
Ph
H
H
H
Cbz
Cbz
Cbz
Cbz
Boc
Cbz
Cbz
Ts
Boc
Cbz
Boc
Boc
Boc
R
R
R
R
R
R
S
R
rac
R
A
A
A
A
A
A
A
B
A
A
A
A
A
14a (74)
rt
15a (0)
–20 J rt
–40 J rt
–78 J rt
–40 J rt
–40 J rt
–40 J rt
–40 J rt
–78 J rt
–40 J rt
–40 J rt
–78 J rt
–95 J rt
15a (61)
15a (70)
15a (46)
15b (0)
15c (71)
15d (68)
15e (75)
15f (0)
15g (19)
15h (50)
15h (57)
15h (67)
14b (71)
14c (69)
14d (75)
14e (77)
14f (95)
14g (35)
14h (97)
H
OTBDMS
OTBDPS
OTBDPS
OTBDPS
R
R
R
aMethod A: t-BuOCOCl, N-methylmorpholine, CH2N2·Et2O, CH2Cl2, –10 to 0°C ; Method B: (i) PCl5, Et2O, (ii)
CH2N2·Et2O, 0°C.
O
Table 2.
OTBDPS
NHBoc
O
Entry
Solvent
Yield (%)
E:Z
1
2
3
4
5
THF
54
50
19
55
24
33:67
37:63
37:63
37:63
61:39
O
NH
Toluene
HMPA
THFa
THFb
O
O
16
17
aSalt free conditions: NaHMDS, 0°C.
Scheme 5.
bWittig–Schlosser conditions (ref. 21).
OTBDPS
O
OTBDPS
Ph3P+CH2CH3Br-
Me
n-BuLi/Solvent
0oC
RT
To better control the stereochemistry of the double bond,
we changed the Wittig ylide from an unstabilized one to a
stabilized one with either an aldehyde or ester (Scheme 6),
since stabilized ylides normally favor E-isomer formation
(22). Moreover, the aldehyde or ester could be considered as
a latent methyl group as required in polyoximic acid. The re-
sults (Table 3, entries 1–6) showed that the stereoselectivity
was not improved, and the E:Z mixture of olefins was still
inseparable for both the aldehyde and the ester cases. Alter-
natively, an Horner-Emmons-Wadsworth reaction was con-
sidered. Indeed, E-isomer formation was favored, and the
ratio was improved as expected (Table 3, entries 7–10). We
then turned our attention to the cyano group (Table 3, entries
11–17), and found that the E:Z mixture of cyano olefins
could be separated by iterative preparative TLC. It appeared
that the phosphine oxide was the best choice with respect to
obtaining a higher E:Z ratio (Table 3, entries 16 and 17).
With both isomers in hand, we could determine the double
bond geometry by an NOE study (Fig. 2). Although the en-
hancement was not significant, the E- and Z-isomers were
differentiated by comparing the corresponding enhancements
of the two isomers. Retrospectively, this evidence was used
to confirm the stereochemical outcome of the Wittig and Hor-
ner-Emmons-Wadsworth reactions for the inseparable E- and
Z-product mixtures (Schemes 5 and 6) by comparing their
1H NMR spectra.
NPG
18
NPG
15h
bond, and 17 was derived from the ylide, which was formed
by reaction of the carbene with the Boc carbonyl. Since the
four-membered ring product 15h was the kinetic product,
lower temperature was presumed to be beneficial. When the
reaction was carried out at –78°C, the yield increased to
57% (Table 1, entry 12). A further decrease of the reaction
temperature to –95°C furnished 15h in 67% yield. A temper-
ature lower than –95°C was not pursued since the reaction
mixture starts to freeze at –95°C in CH2Cl2. The highest
yield (75%) was observed in the case of the N-Ts analog
15e. However, the synthesis was continued with the N-Boc
derivative 15h for ease of deprotection.
Having secured a method to construct the four-membered
ring skeleton, the next stage of the synthesis was to intro-
duce the ethylidene group by conducting the Wittig reaction.
Under normal Wittig conditions (Scheme 5 and Table 2, en-
try 1), a 1:2 mixture of E:Z olefins 18 was produced in favor
of the Z isomer, which were not separable by normal chro-
matography methods. The stereoselectivity of the double
bond formation was not improved by solvent variations (to-
luene, HMPA, Table 2, entries 2 and 3), or by using salt free
conditions (entry 4). In addition, Schlosser’s conditions (21)
(entry 5) reversed the stereoselectivity in favor of the E iso-
mer (E:Z, 2:1).
After separation of the cyano olefins 19 and 20, the E-
isomer was subjected to reduction to obtain the correspond-
ing aldehyde. Unfortunately, using DIBAL-H (1–5 equiv.) in
© 2001 NRC Canada

1816
Can. J. Chem. Vol. 79, 2001
Scheme 6.
Y
O
5
O
OTBDPS
Y
OTBDPS
Reagent
3
2
OTBDPS
5
+
3
2
Base/Solvent.
Temp.
NBoc
15h
O
NBoc
1
4
NBoc
20 1
19
4
Scheme 7.
Me
MeO N
O
P
O
O
Me
EtO
EtO
OMe
Me
O
OTBDPS
N
MeO N
OTBDPS
+
21
OTBDPS
Base/Solvent.
o
NBoc
O
NBoc
0 C
RT
NBoc
22
23
15h
Scheme 8.
Me
MeO N
OTBDPS
HO
OTBDPS
OTBDPS
H
LiAlH4
+
THF
o
O
O
NBoc
-78 C
NBoc
(21%)
NBoc
26
25
(68%)
22
NaBH /CeCl .H O/EtOH/RT/30min
4
3
2
(76%)
Ph P
4
3
CBr
(75%)
CH Cl
RT
2
2
Br
OTBDPS
OH
OTBDPS
TBAF
THF
LiBH
DMSO
4
Me
Me
RT/14h
(66%)
o
NBoc
NBoc
NBoc
0 C
3h
RT
29
28
(94%)
27
Jone's
Reagent
acetone
o
0 C
3h
RT
COOH
COOH
Me
Me
HCOOH
RT
NBoc
NH
trans-(E)-Polyoximic acid
1
30
Et2O or toluene at 0, –40, or –78°C gave at best 30% yield
of the expected aldehyde. Either products of decomposition or
starting material was recovered from the reduction. There-
fore this route using the cyano group as the latent methyl
group in the natural product was not pursued further.
Nevertheless, we discovered that the phosphonate amide
by Nahm and Weinreb (23) provided a solution for this prob-
lem since it could be reduced more easily than the cyano
group. Different conditions for the Horner-Emmons-
Wadsworth reaction (Scheme 7) of compound 15h with re-
agent 21 were explored. A mixture of E- (22) and Z- (23)
isomers was obtained and separated by column chromatogra-
phy. Sodium hydride in THF (Table 4, entry 5) was found to
give the best result with a 94:6 ratio in favor of the E-iso-
© 2001 NRC Canada

Hanessian and Fu
1817
Table 3.
Temp.
(°C)
Yield
(%)
Entry
Reagent
Base
Solvent
Y
E:Z
1
2
3
4
5
6
7
8
9
Ph3P=CHCO2Et
Ph3P=CHCO2Et
Ph3P=CHCO2Et
Ph3P=CHCO2Et
Ph3P=CHCO2Et
Ph3P=CHCHO
(i-PrO)2P(O)CH2CO2Et
(EtO)2P(O)CH2CO2-t-Bu
Me
—
—
—
—
—
—
n-BuLi
n-BuLi
NaH
CH3CN
MeOH
MeOH
DMF
CF3CH2OH
CH3CN
DME
Reflux
0 J rt
–78 J rt
0 J rt
0 J rt
Reflux
rt
EtO
EtO
EtO
EtO
EtO
H
EtO
t-BuO
MeO
77
78
84
93
84
35
82
52
71
26:74
17:83
14:86
25:75
61:39
50:50
80:20
87:13
80:20
DME
THF
rt
0 J rt
N
P(O)CH CO Me
2
2
N
Me
10
NaH
THF
0 J rt
MeO
32
91:9
Bn
N
P(O)CH CO Me
2
2
N
Bn
11
12
13
14
15
16
17
(EtO)2P(O)CH2CN
(EtO)2P(O)CH2CN
(EtO)2P(O)CH2CN
Ph3P=CHCN
Ph3P=CHCN
Ph2P(O)CH2CN
Ph2P(O)CH2CN
n-BuLi
NaH
KH
KH
KH
DME
DME
DME
CH3CN
PhH
rt
rt
rt
CN
CN
CN
CN
CN
CN
CN
78
61
77
78
78
53
51
73:27
83:17
79:21
55:45
64:36
89:11
77:23
Reflux
Reflux
rt
KH
KH
DME
THF
rt
Fig 2.
Table 4.
3.4%
Entry
Base
Solvent
Yield (%)
22:23
1
2
3
4
5
6a
a
n-BuLi
NaH
KH
n-BuLi
NaH
NaH
DME
DME
DME
THF
THF
THF
65
61
66
56
7
88:12
87:13
85:15
86:14
94:6
H
H
OTBDPS
NC
1.1%
NBoc
H
83
88:12
H
Ph
19 trans, (E-)
O
P
O
N
N
OMe
Me
N
0.5%
24
Ph
CN
H
OTBDPS
H
Luche conditions (25) to the alcohol 26 in good yield. The
latter was converted to the bromide 27, which was further
reduced to the methyl compound 28. Removal of the
TBDPS group of 28 and attempts to oxidize the alcohol
group in 29 to the carboxylic acid 30 proved problematic.
The corresponding aldehyde, obtained from Swern or Dess-
Martin oxidations was unstable. Eventually, the classic Jones
oxidation emerged as the best method to yield 30 in 40%
yield. Finally, the N-Boc protecting group was removed by
stirring compound 30 with formic acid to produce the trans-
(E)-polyoximic acid 1, as a crystalline solid, whose structure
was definitively confirmed by X-ray analysis. After com-
pleting the synthesis of trans-(E) 1, and being unaware of
the incorrect stereochemistry of the natural product, we com-
NBoc
2.7%
H
H
20 cis, (Z-)
mer. The chiral non-racemic phosphonamide reagent 24, de-
veloped in our laboratories (24), was also attempted and
gave an E:Z ratio of 88:12.
Our efforts toward the synthesis of the trans-(E)-isomer of
polyoximic acid is shown in Scheme 8. Reduction of the amide
22 with LAH gave a mixture of the a,b-unsaturated aldehyde
25 and the alcohol 26. The aldehyde was reduced under
1
pared the H NMR spectrum of the synthetic sample with an
© 2001 NRC Canada

1818
Can. J. Chem. Vol. 79, 2001
Scheme 9.
Me
Me
MeO N
O
MeO N
OTBDPS
O
OTBDPS
OTBDPS
Ph3P=CHCON(OMe)Me
O
NBoc
+
NBoc
MeOH/-78oC
(73%)
22
NBoc
(90:10)
15h
23
H
O
HO
OTBDPS
LiAlH4
OTBDPS
23
+
THF
-78oC
NBoc
NBoc
(16%)
(52%)
32
31
NaBH4/CeCl3.H2O/EtOH/RT/30min
(74%)
Ph3P
CBr4
CH2Cl2
RT
(80%)
Br
Me
TBAF
THF
LiBH4
THF
Me
OH
OTBDPS
OTBDPS
RT/14h
(79%)
0oC
3h
RT
NBoc
NBoc
NBoc
(97%)
reagent,
acetone
o
35
(40%)
34
33
0 C
3h
RT
Me
Me
COOH
COOH
HCOOH
RT
NH
NBoc
cis(Z)-Polyoximic acid
36
2
authentic one kindly provided by Professor Isono. The spec-
tra were in fact different, and we had synthesized the incor-
rect isomer.
logical choice as a (Z)-olefinating reagent in methanol
(Scheme 9). Under these conditions, the E:Z-mixture of
olefins was obtained in 73% yield and in a ratio of 10:90. In
acetonitrile or THF, the ratios were 64:36 and 61:39, respec-
Faced with this dilemma, we were fortunate that the sample
of racemic polyoximic acid received from Professor Isuno
was in excellent physical state and amenable to X-ray analy-
sis. The result shown in Fig. 3 confirms unambiguously that
the geometry of the exocyclic double bond of the natural prod-
uct is indeed cis-(Z), contrary to the original assignment (5).
With this important stereochemical revision, we turned our
attention to the synthesis of the natural cis-(Z)-polyoximic
acid 2. The ready availability of the 3-azetidinone 15h
(Scheme 6), urged us to develop methodology directed at en-
riching the desired Z-isomer. Our studies on olefin formation
had shown that the Wittig reaction favors the Z-product par-
ticularly in a protic solvent (Table 3, compare with Table 4).
Therefore the Weinreb ester of the Wittig reagent became a
1
tively, as determined by H NMR spectroscopy. With a sat-
isfactory method to obtain the (Z)-isomer 23, the synthesis
was carried out in a manner similar to that of the (E)-isomer,
affording crystalline cis-(Z)-polyoximic acid 2 (Scheme 9).
Comparison of the spectroscopic and physical data of the
synthetic material 2 with that of the natural product, con-
firmed the proposed structure, and the geometry of the dou-
ble bond. The spectra in CD3OD for the cis-(Z)- and trans-
(E)-synthetic products, as well as that of the racemic natural
product are shown in Fig. 4. It is interesting that the vinylic
proton and the methyl protons in the trans-isomer 1 are
shifted downfield and upfield, respectively, compared to the
signals of the cis-isomer 2. This pattern was also found in
© 2001 NRC Canada

Hanessian and Fu
1819
Scheme 10.
Me
HR
Me
H
HS
HS
HR
COOH
COOH
COOH
COOH
Me
?
Me
H
NH2
NH
Me NH2
Ref. 26a
Me NH2
HO
B:
B:
ref. 26b
1
With the establishment of the cis stereochemistry for
polyoximic acid, the proposal of an elimination of the pro-R
hydrogen at C-4 of L-isoleucine (26a), and the possible
intermediacy of a hypothetical precursor, L-2-amino-3-
hydroxymethyl-3-pentenoic acid (26b), must be reassessed
(Scheme 10). If the double bond is indeed introduced by an
enzymatic process, it most likely involves a synperiplanar
elimination.
Fig. 3. Ortep drawing and structure of racemic cis-polyoximic acid.
Finally, it is of interest to point out that the structures of
polyoximic acids 1 and 2, and related structures resulting
from the methodology developed in this work represent con-
strained analogs of amino acids (27) such as L-isoleucine, L-
vinyl glycine (28), and their congeners. As such, they are in-
teresting structures for exploitation as b-turn mimetics (for se-
lected reviews, see ref. 29) in a variety of peptide-like motifs.
In summary, after more than 30 years since the isolation
and structure determination of polyoxin (5), the diminutive
yet intriguing amino acid polyoximic acid was synthesized
in enantiopure form, and the geometry of the double bond
was reassigned as cis rather than trans through NMR spec-
troscopy and X-ray crystallography.
D2O and DMSO-d6. The amino acid 2 was partially racemized
in methanol containing a drop of Et3N (~50% within 20 h at rt).
The geometry of the exocyclic double in the 3-azetidinone
15h, could be controlled by conducting either the Horner-
Emmons-Wadworth reaction or the Wittig reaction to gener-
ate a (E)- or (Z)-olefin preponderantly. Apparently, the Hor-
ner-Emmons-Wadsworth reaction gives the thermodynamic
product as a major isomer at 0°C, and the Wittig reaction
gives the kinetic product. To what extent, the nature of the
Weinreb amide influences the respective transition states com-
pared to normal esters is a matter of speculation (Tables 3
and 4). Dipolar and stereoelectronic effects could be playing
important roles in both reactions. In addition, the nature of the
solvent and cation is important in the phosphonate case, while
the protic medium may promote a faster collapse of the more
congested (Z)-betaine via the corresponding oxaphosphetane.
Since polyoximic acid was the degradation product of
polyoxin, we had to be assured that no isomerization of the
double bond had occurrred during the hydrolysis process or
subsequent isolation. The 1D 1H NMR spectrum of polyoxin
A provided by Professor Isono is shown in Fig. 5, where the
6>>-methyl, 5>>-vinylic, and 2>>-a-carboxyamide protons are
clearly distinguishable. These correspond in pattern and in
chemical shift differences to the cis- rather than the trans-
polyoximic acid moiety.
Proton connectivities by the J-correlated spectroscopy
(COSY) technique located the “intact” polyoximic acid moi-
ety. Next, we carried out a 2D NOE correlated spectroscopy
experiment (NOESY) to secure the cis orientation of the
ethylenic group vis-à-vis the carboxamide group. Positive
enhancements were observed between the ethylene methyl
group and the a-carboxamide hydrogen, which is consistent
with a cis-type geometry. These were independently corrob-
orated by NOE studies on the cis- and trans-polyoximic ac-
ids in D2O and DMSO-d6.
Experimental
General procedure
Flash chromatography was carried out using 230–400
mesh silica gel. Mixtures of ethyl acetate–hexanes were used
as the eluents unless otherwise specified. Analytical thin
layer chromatography (TLC) was performed on glass plates
coated with 0.02 mm layer of silica gel 60 F-254. Melting
points were uncorrected. IR spectra were recorded as films.
1H NMR and 13C NMR spectra were recorded at 300 and
75 MHz, respectively, and the chemical shifts are reported in
parts per million on the d scale with CDCl3 as reference un-
less otherwise specified. MS and HRMS spectra were re-
corded using electron ionization (EI) or by the FAB
technique. Elemental analyses were performed by Guelph
Laboratories, Guelph, Ontario. Optical rotations were mea-
sured in CHCl3 at 23°C. All reactions were carried out under
nitrogen or argon atmosphere unless otherwise specified.
The –78°C temperature is approximate as achieved by a dry
ice – acetone bath. The –95°C temperature is approximate as
achieved by a liquid nitrogen – methanol bath. The term
“normal workup” means the addition of saturated aqueous
NH4Cl solution to the reaction mixture, followed extraction
with ethyl acetate, drying over Na2SO4, filtration, and evapo-
ration of the filtrate in vacuo to afford the crude product.
© 2001 NRC Canada

1820
Can. J. Chem. Vol. 79, 2001
1
Fig. 4. 300 MHz H NMR (CD3OD) spectra of synthetic trans 1, synthetic cis-2 obtained by acid hydrolysis (ref. 5).
(m, 10H). 13C NMR d: 54.6, 58.7, 67.0, 127.0, 128.0, 128.1,
128.4, 128.5, 128.6, 129.1, 129.2, 135.8, 136.0, 155.6, 192.6.
General procedure for the preparation of diazoketones
(method A)
To a solution of the protected amino acid (1.00 mmol) in
dichloromethane at –10°C was added simultaneously
isobutyl chloroformate (1.07 mmol) and 4-methylmor-
pholine (1.07 mmol). After stirring for 45 min at –10°C, a
solution of freshly prepared diazomethane in ether (18.75 mL,
6.68 mmol) was added, and stirring continued for 7 h at rt,
followed by concentration in vacuo to dryness. The residue
was subjected to column chromatography to afford the de-
sired product.
(3R)-1-Diazo-3-(N-tert-Butyloxycarbonylamino)-4-phenyl-2-
butanone (14b)
According to method A, the title compound was prepared in
71% yield: mp 87–89°C (hexanes–CH2Cl2); [a]D –8.1 (c 1.02,
CHCl3). FAB-MS (NBA) m/z (rel intensity): 290 ([M + 1]+,
18), 206 (100), 164 (31), 154 (70), 136 (52), 120 (84). HRMS
calcd. for C15H20N3O3: 290.1506; found: 290.1472. IR (KBr)
ꢀmax (cm–1): 3340, 2102, 1690, 1640. 1H NMR d: 1.39 (s, 9H),
2.98–3.09 (m, 2H), 4.35–4.48 (br s, 1H), 5.14 (d, J = 7.7 Hz,
1H), 5.23 (s, 1H). 13C NMR (CDCl3) d: 28.1, 38.4, 54.2, 58.3,
79.9, 126.8, 128.5, 129.2, 136.2, 155.0, 193.2.
(3R)-1-Diazo-3-(N-benzyloxycarbonylamino)-4-phenyl-2-
butanone (14a)
According to method A, the title compound was prepared
in 74% yield: mp 82–83°C (hexanes–CH2Cl2); [a]D –13
(c 1.2, CHCl3). FAB-MS (NBA) m/z (rel intensity): 324
([M + 1]+, 3), 307 (13), 289 (16), 154 (100), 136 (91), 106
(58). HRMS calcd. for C18H18N3O3 + H: 324.1349; found:
324.1329. IR (KBr) ꢀmax (cm–1): 3335, 3082, 2118, 1700,
(3R)-1-Diazo-3-(N-benzyloxycarbonylamino)-2-butanone (14c)
According to method A, the title compound was prepared
in 69% yield: mp 92–94°C (hexanes–CH2Cl2); [a]D +4.1
(c 1.0, CHCl3). FAB-MS (NBA) m/z (rel intensity): 268 ([M +
1]+, 16), 240 (38), 197 (62), 154 (100), 139 (20). HRMS
calcd. for C12H14N3O3 + H: 248.1036; found: 248.1037. IR
1
1626. H NMR d: 3.04 (d, J = 6.6 Hz, 2H), 4.46–4.55 (m,
1H), 5.08 (s, 2H), 5.20 (s, 1H), 5.38–5.40 (m, 1H), 7.15–7.38
1
(CHCl3) ꢀmax (cm–1): 3326, 2124, 1730, 1650. H NMR d:
© 2001 NRC Canada

Hanessian and Fu
1821
CHCl3). FAB-MS (NEA) m/z (rel intensity): 468 ([M + H]+,
49), 410 (16), 199 (64), 135 (100). HRMS calcd. for
C25H34N3O4Si (M + H): 468.2320; found: 468.2266. IR
Fig. 5. 2D COSY NMR spectrum of natural polyoxin A in
DMSO-d6 at 400 MHz (top); 2D NOESY NMR spectrum in
DMSO-d6 at 400 MHz (bottom).
1
(CHCl3) ꢀmax (cm–1): 2103, 1711, 1641. H NMR d: 1.06 (s,
9H), 1.46 (s, 9H), 3.79–3.84 (dd, J = 10.2, 4.8 Hz, 1H),
3.96–4.01 (dd, J = 10.2, 3.9 Hz, 1H), 4.20–4.38 (br s, 1H),
5.35–5.38 (d, J = 7.8 Hz, 1H), 5.49 (s, 1H), 7.40–7.45 (m,
6H), 7.61–7.64 (m, 4H). 13C NMR d: 19.1, 26.4, 28.2, 54.1,
59.0, 64.0, 80.0, 127.7, 129.8, 132.6, 135.4, 155.2, 192.8.
(3R)-(+)-1-Diazo-3-(N-4-toluenesulphonylamino)-2-
butanone (14e) (method B)
To a solution of N-4-toluenesulphonyl-D-alanine (5.73 g,
23.6 mmol) in ether (55.0 mL) was added pentachloro-
phosphorus (10.1 g, 48.3 mmol). After vigorous stirring at rt
for 1 h, the reaction mixture was filtered. To the filtrate was
added hexane (200 mL). The generated solid was filtered
and collected to give 5.00 g (82%) of the chloride com-
pound. The chloride was dissolved in ether (50.0 mL). To
this solution was added diazomethane-etherate solution
(140 mL) at 0°C. After 1 h, the reaction mixture was filtered
through a pad of silica gel. The filtrate was concentrated in
vacuo to dryness. The residue was subjected to column chro-
matography (EtOAc–hexane, 1:2) to afford 3.94 g (77%) of
the desired product: oil. [a]D +116.2 (c 2.36, CHCl3). FAB-
MS m/z (rel intensity): 268 ([M + 1]+, 16), 240 (35), 197
(61), 154 (100). HRMS calcd. for C11H14N3O3S: 268.0757;
found: 268.0788. IR (neat) ꢀmax (cm–1): 3260, 2105, 1735,
1
1641, 1366. H NMR d: 1.24 (d, J = 7.1 Hz, 3H), 2.40 (s,
3H), 3.76–3.89 (m, 1H), 5.45 (s, 1H), 5.51 (d, J = 7.5 Hz,
1H), 7.28 (d, J = 8.0 Hz, 2H), 7.69–7.73 (m, 2H). 13C NMR
(CDCl3) d: 19.4, 21.4, 53.9, 55.2, 127.0, 129.7, 136.6, 143.7,
192.7.
General procedure for the preparation of azetidinone
15 (method C)
To a solution of the diazo compound 14 (1.00 mmol)
in dichloromethane (10.0 mL) at the reaction temperature
indicated was added rhodium acetate dimer (0.009 g,
0.020 mmol). The resulting reaction mixture was stirred that
reaction temperature for 1 h and allowed to warm to rt for
14 h. The solvent was removed and the residue was sub-
jected to column chromatography.
1.35 (d, J = 7.1 Hz, 3H), 4.20–4.37 (m, 1H), 5.11 (s, 2H),
5.40 (s, 1H), 5.47 (br, 1H), 7.31–7.37 (m, 5H). 13C NMR
(CDCl3) d: 18.5, 53.4, 53.5, 66.9, 128.0, 128.1, 128.4, 136.0,
155.5, 193.6.
(2R)-1-Benzyloxycarbonyl-2-benzyl-3-azetidinone (15a)
The reaction was carried out at –40°C to give 70% of
product 15a: mp 85–87°C (hexanes–CH2Cl2). [a]D –124.6
(c 1.0, CHCl3). EI-MS m/z (rel intensity): 295 (M+, 3), 267
(18), 204 (38), 132 (30), 88 (100). HRMS calcd. for
C18H17NO3: 295.1209; found: 295.1238. IR (neat) ꢀmax
rac-1-Diazo-3-(N-tert-butyloxycarbonylamino)-2-butanone (14f)
According to method A, the title compound was prepared
in 95% yield: mp 87–89°C (hexanes–CH2Cl2). FAB-MS
(NBA) m/z (rel intensity): 214 ([M + 1]+, 35), 158 (50), 154
1
(cm–1): 1819, 1691. H NMR d: 3.14 (dd, J = 14.3, 4.0 Hz,
1H), 3.24 (dd, J = 14.3, 6.3 Hz, 1H), 4.08 (dd, J = 16.5,
4.4 Hz, 1H), 4.59 (d, J = 16.5 Hz, 1H), 5.20 (s, 2H), 5.21–
5.25 (m, 1H), 7.12–7.15 (m, 2H), 7.23–7.29 (m, 2H), 7.35–
7.41 (m, 6H). 13C NMR (CDCl3) d: 35.4, 67.3, 69.2, 83.5,
126.9, 128.1, 128.2, 128.4, 128.5, 129.7, 134.9, 136.0,
155.8, 198.7.
1
(80), 144 (26), 136 (71), 130 (100). H NMR d: 1.31 (d, J =
7.1, 3H), 1.43 (s, 9H), 4.14–4.29 (br s, 1H), 5.09–5.21 (br s,
1H), 5.44 (s, 1H). 13C NMR (CDCl3) d: 18.3, 28.2, 53.0,
53.2, 79.9, 155.0, 194.3.
(3R)-1-Diazo-3-(N-tert-butyloxycarbonyl)amino-4-tert-butyl-
diphenyl-silyloxy-2-butanone (14h)
According to method A, the title compound was prepared
in 97% yield: mp 89–90°C (hexanes). [a]D –0.67 (c 1.05,
(2R)-1-Benzyloxycarbonyl-2-methyl-3-azetidinone (15c)
The reaction was carried out at –40°C to give 71% of
product 15c: mp 80–81°C (hexane–CH2Cl2). [a]D –39.7
© 2001 NRC Canada

1822
Can. J. Chem. Vol. 79, 2001
(c 0.99, CHCl3). EI-MS m/z (rel intensity): 219 (M+, 1), 191
1.16 mmol) in THF (10.0 mL) was added. After stirring at rt
for 14 h, water (30.0 mL) was added, THF was removed in
vacuo and the aqueous solution was extracted with EtOAc
(3 × 30 mL). The extracts were dried (Na2SO4) and filtered.
Column chromatography (EtOAc–hexanes, 1:8) afforded
0.281 g (54%) of product 18 as a colorless oil. IR (film)
(4), 91 (100). HRMS calcd. for C12H13NO3: 219.0896;
1
found: 219.0935. IR (CHCl3) ꢀmax (cm–1): 1816, 1694. H
NMR d: 1.49 (d, J = 7.0 Hz, 3H), 4.67 (dd, J = 16.5, 4.2 Hz,
1H), 4.97–5.07 (m, 1H), 5.17 (s, 2H), 7.27–7.39 (m, 2H).
1
ꢀmax (cm–1): 1705. H NMR d: 1.05 (s, 9H), 1.40 (s, 9H),
(2R)-1-p-Toluenesulphonyl-2-methyl-3-azetidinone (15e)
According to method C, the reaction was carried out at
–40°C to give 75% of product 15e: mp 76–78°C (hexane–
CH2Cl2). [a]D –65.1 (c 1.09, CHCl3). EI-MS m/z (rel inten-
sity): 240 (M+, 8), 211 (16), 155 (20), 105 (17), 91 (100).
HRMS calcd. for C11H14NO3S: 239.0617; found: 239.0605.
1.58–1.60 (m, 3H), 3.79–3.86 (m, 1H), 3.90–4.10 (br s, 1H),
4.29–4.41 (m, 2H), 4.67 (br s, 1H), 4.80 (br s, 1H), 7.34–
7.46 (m, 6H), 7.65–7.74 (m, 4H).
(2S,E)-1-tert-Butyloxycarbonyl-2-tert-butyldiphenylsilyl-
oxymethyl-3-cyanocarbonyl methylidene-azetidine (19)
and (2S,Z)-1-tert-butoxycarbonyl-2-tert-butyldiphenysilyl-
oxymethyl-3-cyanocarbonyl-methylidene-azetidine (20)
To a solution of diethyl cyanomethylphosphonate (62.0 mg,
0.250 mmol) in DME (1.50 mL) at 0°C was added n-
butyllithium (0.14 mL, 2.30 M in hexanes, 0.32 mmol). The
resulting light yellow solution was stirred at 0°C for 20 min
and rt for 20 min. The solution of compound 15h (0.128 g.
0.29 mmol) in DMF (1.50 mL) was added. After stirring at
rt for 2 h, the reaction was quenched with saturated NH4Cl
solution (1.00 mL). The aqueous solution was extracted with
EtOAc (3 × 5.00 mL) and the combined EtOAc solutions
were dried (Na2SO4) and filtered. The filtrate was concentrated
in vacuo to dryness. Column chromatography (EtOAc–hexanes,
1:3) afforded 77.0 mg (57%) of product 19 and 22.0 mg (16%)
of product 20, both as colorless oils. For 19: [a]D –24.0
(c 1.22, CHCl3). FAB-MS m/z (rel intensity): 463 ([M + H]+,
47), 407 (66), 349 (46), 329 (52), 198 (52), 154 (70), 137
(71). 106 (59), 57 (100). HRMS calcd. for C27H34N2O3Si
(M+ H): 463.2419; found: 463.2435. IR (film) ꢀmax (cm–1):
1
IR (KBr) ꢀmax (cm–1): 1831, 1599, 1345. H NMR d: 1.46
(d, J = 7.0 Hz, 3H), 2.47 (s, 3H), 4.43–4.60 (m, 2H), 4.76–
4.80 (m, 1H), 7.38–7.41 (d, J = 8.3 Hz, 2H), 7.78–7.82 (m,
2H). 13C NMR (CDCl3) d: 15.6, 21.5, 69.5, 80.9, 128.3,
130.0, 131.8, 144.9, 196.7.
(2R)-1-tert-Butyloxycarbonyl-2-tert-butyldiphenylsilyloxy-
methylene-3-azetidinone (15h)
The reaction was carried out at –95°C to give 67% of
product 15h as a colorless oil. [a]D –41.9 (c 1.50, CHCl3).
FAB-MS (NBA) m/z (rel intensity): 440 ([M + H]+, 4), 384
(10), 326 (57), 306 (63), 198 (78), 197 (77), 135 (100).
HRMS calcd. for C25H34NO4Si (M + H): 440.2258; found:
1
440.2294. IR (film) ꢀmax (cm–1): 1831, 1705. H NMR d:
1.03 (s, 9H), 1.46 (s, 9H), 3.90–3.94 (dd, J = 11.4, 1.7 Hz,
1H), 4.00–4.20 (br s, 1H,), 4.68–4.70 (m, 2H), 4.88 (br s,
1H), 7.37–7.47 (m, 6H), 7.62–7.73 (m, 4H). 13C NMR d:
19.1, 26.5, 28.2, 60.3, 70.0, 80.6, 84.0, 127.7, 129.8, 132.6,
135.4, 155.1, 198.9.
1
(2R)-2-(tert-Butyloxycarbonylamino)-4-oxacyclopentanone
(16)
2221, 1710. H NMR d: 1.07 (s, 9H), 1.42 (s, 9H), 3.85–
3.90 (dd, J = 10.6, 3.3 Hz, 1H), 3.97–4.08 (br s, 1H), 4.63–
4.65 (dd, J = 2.9, 2.9 Hz, 2H), 4.84 (m, 1H), 5.38 (m, 1H).
7.37–7.48 (m, 6H), 7.62–7.67 (m, 4H). 13C NMR d: 19.1,
26.6, 28.1, 56.7, 62.8, 70.5, 80.5, 93.1, 114.3, 127.7, 129.8,
132.5, 135.4, 155.4. 160.7.
For 20: [a]D +61.3 (c 1–22, CHCl3). FAB-MS m/z (rel in-
tensity): 463 ([M + H]+, 66), 407 (71), 349 (49), 329 (52),
307 (42), 137 (37), 106 (68), 57 (100). HRMS calcd for
C27H34N2O3Si (M + H): 463.2419; found: 463.2397. IR (film)
ꢀmax (cm–1): 2219, 1706. 1H NMR d: 1.06 (s, 9H), 1.42 (s, 9H),
4.08–4.25 (br s, 2H), 4.51–4.58 (ddd, J = 15.4, 2.0, 1.9 Hz,
1H), 4.61–4.68 (ddd, J = 15.4, 4.2, 2.3 Hz, 1H), 4.95–4.97 (m,
1H), 5.45–5.47 (m, 1H). 7.36–7.47 (m, 6H), 7.63–7.73 (m,
4H). 13C NMR d: 19.3. 26.7. 28.3. 57.2, 61.2. 72.3, 80.4, 92.9,
114.2, 127.7. 129.8, 132.9, 135.5, 155.2, 160.4.
Compound 16 was isolated as a side product when prepar-
ing compound 15h (12%): oil. IR (neat) ꢀmax (cm–1): 3431,
1
1770, 1711. H NMR d: 1.45 (s, 9H), 3.78 (dd, J = 10.4,
9.0 Hz, 1H), 3.93 (d, J = 17.2 Hz, 1H), 4.09–4.21 (br s, 1H),
4.20 (d, J = 17.2 Hz, 1H), 4.64–4.72 (m, 1H), 4.91–5.01
(br s, 1H).
rac-2-(tert-Butyldiphenylsilyloxymethyl)-3-aza-5-oxa-1,4-
cyclohexanedione (17)
Compound 17 was isolated as a side product when prepar-
ing compound 15h (8%): mp 137–139°C. EI-MS m/z (rel in-
tensity): 327 ([M + 1]+, 38), 326 (M+, 57), 296 (62), 282
(46), 248 (48), 199 (100). HRMS calcd. for C17H16NO4Si:
326.0849; found: 326.0872. IR (KBr) ꢀmax (cm–1): 3270,
1
1736, 1720. H NMR d: 1.04 (s, 9H), 3.80 (dd, J = 10.3,
2.8 Hz, 1H), 3.91–3.94 (br s, 1H), 4.56 (d, J = 17.0 Hz, 1H),
4.65 (d, J = 17.0 Hz, 1H), 6.49 (s, 1H), 7.37–7.46 (m, 6H),
7.58–7.67 (m, 4H). 13C NMR (CDCl3) d: 19.0, 26.5, 61.1,
64.6, 71.8, 127.9, 130.1, 132.0, 135.5, 154.3, 200.9.
(2S,E)-1-tert-Butyloxycarbonyl-2-tert-butyldiphenylsilyl-
oxymethyl-3-(N-methoxy-N-methylcarbamoyl)methylidene-
azetidine (22)
To a suspension of NaH (1.76 g, 60% in oil, 44.0 mmol,
washed three times with ether) in THF (500 mL) at 0°C was
added the solution of diethyl (N-methoxy-N-methylcarba-
moylmethyl)phosphonate 21 (11.0 g, 46.0 mmol) in THF
(50.0 mL). The resulting reaction mixture was stirred at 0°C
for 10 min and at rt for 20 min, a solution of compound 15h
(17.9 g, 42.0 mmol) in THF (50.0 mL) was added. After
stirring at rt for 14 h, water (30.0 mL) was added, THF was
evaporated, and the aqueous solution was extracted with
(2S,E/Z)-1-tert-Butyloxycarbonyl-2-tert-butyldiphenyl-
silyloxymethyl-3-ethylidene azetidine (18)
To a solution of (ethyl)triphenylphosphonium bromide
(1.94 g, 5.23 mmol) in THF (30.0 mL) at rt was added n-butyl-
lithium (2.12 mL, 2.19 M in hexanes, 4.65 mmol). The re-
sulting orange red solution was stirred for 30 min and cooled
down to 0°C. A solution of compound 15h (0.511 g,
© 2001 NRC Canada

Hanessian and Fu
1823
EtOAc (3 × 30.0 mL). The combined EtOAc solution was
dried (Na2SO4) and filtered. Column chromatography
(CH2Cl2–C6H6–EtOAc, 4:2:1) afforded 16.0 g (73%) of
product 22 as a colorless oil and 1.0 g (4.6%) of product 23
as a colorless solid. For 22: [a]D +3.2 (c 1.7 1, CHCl3).
FAB-MS m/z (rel intensity): 525 ([M + H]+, 52), 469 (73),
451 (18), 411 (47), 213 (21), 197 (72), 154 (76), 136 (100).
HRMS calcd. for C29H40N2O5Si (M + H): 525.2786; found:
product 26 as a colorless oil. For 25: [a]D –16.0 (c 2.01,
CHCl3). FAB-MS (NEA) m/z (rel intensity): 466 ([M + H]+,
5), 427 (9), 352 (12), 197 (74), 154 (64), 135 (100). HRMS
calcd. for C27H36N3O4Si (M + H): 466.2415; found:
1
466.2390. IR (CHCl3) ꢀmax (cm–1): 2739, 1703, 1696. H
NMR d: 1.05 (s, 9H), 1.43 (s, 9H), 3.87–3.91 (dd, J = 10.6,
2.7 Hz, 1H), 4.00–4.17 (br s, 1H), 4.86–4.88 (m, 2H), 4.93
(br s, 1H), 6.07–6.10 (m, 1H), 7.35–7.47 (m, 6H), 7.61–7.67
(m, 4H), 9.62–9.65 (d, J = 6.8 Hz, 1H). 13C NMR d: 19.4,
26.9, 28.4, 60.0, 62.2, 63.4, 71.6, 79.6, 122.4, 128.6, 130.1,
133.9, 135.9, 155.4, 159.2, 187.8.
1
525.2723. IR (film) ꢀmax (cm–1): 1718, 1644. H NMR d:
1.06 (s, 9H), 1.40 (s, 9H), 3.23 (s, 3H), 3.65 (s, 3H), 3.93–
4.01 (br s, 2H), 4.83–4.91 (m, 3H), 6.52–6.55 (br s, 1H),
1
7.34–7.46 (m, 6H), 7.65–7.69 (m, 4H). H NMR (C6D6) d:
1.13 (s, 9H), 1.39 (s, 9H), 2.87 (s, 3H), 3.10 (s, 3H), 3.86–
4.30 (br s, 2H), 4.72–4.79 (br s, 1H), 5.12–5.27 (m, 2H),
6.60–6.70 (br s, 1H), 7.21–7.26 (m, 6H), 7.73–7.80 (m, 4H).
13C NMR d: 19.1, 26.6, 28.2, 31.8, 59.3, 61.6, 63.3, 70.2,
79.6, 110.3, 127.6, 129.6, 129.7, 132.9, 135.4, 154.4, 165.9.
13C NMR (C6D6) d: 19.5, 26.9, 28.4, 31.8, 60.2, 61.1, 63.7,
70.7, 79.2, 110.9, 128.1, 128.6, 130.0. 133.6, 136.0, 154.8,
166.0.
(2S,E)-1-tert-Butyloxycarbonyl-2-tert-butyldiphenyisilyl-
oxymethyl-3-(hydroxymethyl)methylidene-azetidine (26)
To a solution of 25 (5.11 g, 11.0 mmol) in ethanol (300 mL)
at rt was added sodium borohydride (0.415 g, 11.0 mmol)
and cerium trichloride heptahydrate (4.10 g, 11.0 mmol) si-
multaneously. The resulting reaction mixture was stirred at rt
for 30 min and quenched with H2O (300 mL). The aqueous
solution was extracted with ethyl acetate (3 × 500 mL) and
the combined ethyl acetate solution was dried (Na2SO4) and
filtered. The filtrate was concentrated in vacuo to dryness.
Column chromatography (CH2Cl2–hexanes–EtOAc, 2:1:1) af-
forded 4.54 g (89%) of product 26 as a colorless oil: [a]D
–13.3 (c 1.56, CHCl3). FAB-MS (NBA) m/z (rel intensity):
468 ([M + H]+, 20), 429 (18), 354 (44), 334 (43), 307 (40),
198 (63), 154 (90), 136 (100). HRMS calcd. for C27H38NO4Si
(M + H): 468.2571; found: 468.2618. IR (CHCl3) ꢀmax (cm–1):
(2S,Z)-1-tert-Butyloxycarbonyl-2-tert-butyldiphenylsilyl-
oxymethyl-3-(N-methoxy-N-methylcarbamoyl)methylidene-
azetidine (23)
A mixture of compound 15h (0.647g, 1.47 mmol) and N-
methoxy-N-methyl-2-(triphenylphosphoranylidene)acetamide
(0.642 g, 1.77 mmol) in MeOH (15.0 mL) was stirred at –78°C
for 1 h and at rt for 24 h. After concentration to dryness, col-
umn chromatography (CH2Cl2–C6H6–EtOAc, 4:2:1) afforded
0.515 g (67%) of product 23 and 0.059 g (7.7%) of product
22. For 23: mp 119–121°C (CH2Cl2–hexanes). [a]D +15.9
(c 0.79, CHCl3). FAB-MS m/z (rel intensity): 525 ([M + H]+,
44), 469 (27), 451 (16), 411 (40), 212 (54), 197 (66), 154
(82), 136 (100). HRMS calcd. for C29H40N2O5Si (M + H):
525.2786; found: 525.2836. IR (film) ꢀmax (cm–1): 1711,
1
3420, 1700, 1676. H NMR d: 1.06 (s, 9H), 1.41 (s, 9H),
3.83–3.87 (dd, J = 10.4, 3.0 Hz, 1H), 3.95–4.09 (br s, 1H),
4.10–4.12 (d, J = 5.7 Hz, 2H), 4.51 (br s, 2H), 4.73 (br s,
1H), 5.61 (m, 1H), 7.35–7.43 (m, 6H), 7.65–7.70 (m, 4H).
13C NMR d: 19.2, 26.7, 28.3, 55.8, 60.0, 63.5, 70.1, 79.5,
120.8, 127.6, 129.6, 133.3, 133.4, 135.5, 155.8.
1
1645. H NMR d: 1.03 (s, 9H), 1.40 (s, 9H), 3.18 (s, 3H),
(2S,E)-1-tert-Butyloxycarbonyl-2-tert-butyldiphenylsilyloxy-
methyl-3-(bromomethyl)methylidene-azetidine (27)
3.70 (s, 3H), 4.05–4.15 (br s, 1H), 4.34–4.39 (m, 1H), 4.44–
4.53 (m, 1H), 4.64–4.73 (m, 1H), 5.18–5.20 (br s, 1H),
6.34–6.41 (br s, 1H), 7.29–7.41 (m, 6H), 7.59–7.68 (m, 4H).
1H NMR (C6D6) d: 1.16 (s, 9H), 1.40 (s, 9H), 2.86 (s, 3H),
3.06 (s, 3H), 4.22–4.40 (br s, 2H), 4.66–4.76 (m, 2H), 5.37–
5.38 (br s, 1H), 6.17 (br s, 1H), 7.21–7.29 (m, 6H), 7.76–
7.84 (m, 4H). 13C NMR d: 18.9, 26.4, 28.0, 31.6, 56.9, 61.0,
61.3, 73.4, 79.2, 110.7, 126.8, 129.0, 133.1, 133.4, 135.1,
153.2, 165.1. 13C NMR (C6D6) d: 19.6, 27.1, 28.5, 31.8,
57.4, 61.1, 61.8, 74.1, 79.1, 111.6, 128.6, 129.8, 134.0,
134.3, 136.1, 153.8, 165.9. Anal. calcd for C29H40N2O5Si:
C 66.38, H 7.68, N 5.34; found: C 66.38, H 7.62, N 5.25.
To a solution of 26 (1.98 g, 4.23 mmol) in dichloromethane
(85.0 mL) at rt was added triphenylphosphine (1.67 g,
6.35 mmol) and carbon tetrabromide (2.11 g, 6.35 mmol).
The resulting reaction mixture was stirred at rt for 14 h and
concentrated in vacuo to dryness. Column chromatography
(EtOAc–hexanes, 1:6) afforded 1.67 g (75%) of product 27
as a colorless oil: [a]D –10.9 (c 2.27, CHCl3). FAB-MS
(NBA) m/z (rel intensity): 532 ([M + H]+, 3), 530 ([M + H]+, 1),
418 (14), 416 (12), 199 (55), 197 (52), 135 (100). HRMS
calcd. for C27H37BrNO3Si (M + H): 530.1727; found:
1
530.1783. IR (CHCl3) ꢀmax (cm–1): 1702. H NMR d: 1.06
(s, 9H), 1.41 (s, 9H), 3.84–3.88 (m, 3H), 3.90–4.10 (br s,
1H), 4.51 (br s, 2H), 4.76 (br s, 1H), 5.77 (m, 1H), 7.36–
7.46 (m, 6H), 7.66–7.73 (m, 4H). 13C NMR d: 19.1, 26.6,
27.3, 28.2, 54.3, 63.1, 69.9, 79.7, 118.4, 127.6, 129.6, 133.1,
135.5, 138.3, 155.6.
(2S,E)-1-tert-Butyloxycarbonyl-2-tert-butyldiphenylsilyl-
oxymethyl-3-(formyl)methylidene-azetidine (25)
To a solution of 22 (8.70 g, 16.6 mmol) in THF (170 mL) at
–78°C was added lithium aluminum hydride (3.15 g,
82.9 mmol). The resulting reaction mixture was stirred at –78°C
for 1 h and quenched with H2O (3.15 mL) at –78°C. After
warming to rt, a 15% NaOH solution (3.15 mL) was added
and the slurry was stirred for 15 min at rt followed by addi-
tion of H2O (6.30 mL). The reaction mixture was filtered
and the filtrate was concentrated in vacuo to dryness. Col-
umn chromatography (EtOAc–hexanes, 1:3) afforded 5.23 g
(68%) of product 25 as a colorless oil and 1.65 g (21%) of
(2S,E)-1-tert-Butyloxycarbonyl-2-tert-butyldiphenylsilyl-
oxymethyl-3-ethylidene-azetidine (28)
To a solution of 27 (2.85 g, 5.36 mmol) in dimethyl
sulphoxide (110 mL) at rt was added lithium borohydride
(5.36 mL, 2.0 M solution in THF, 10.7 mmol). The resulting
reaction mixture was stirred at rt for 14 h. Water (110 mL)
was added followed by extraction with ethyl acetate (3 ×
© 2001 NRC Canada

1824
Can. J. Chem. Vol. 79, 2001
200 mL). The combined ethyl acetate solution was dried
(Na2SO4) and filtered. The filtrate was concentrated in
vacuo to dryness. Column chromatography (EtOAc–hex-
anes, 1:8) afforded 1.60 g (66%) of product 28 as a colorless
oil: [a]D –10.7 (c 1.34, CHCl3). FAB-MS (NBA) m/z (rel in-
tensity): 452 ([M + H]+, 9), 396 (7), 338 (100), 318 (84),
199 (83), 154 (42), 135 (100). HRMS calcd. for
C27H37NO3Si (M + H): 452.2685; found: 452.2667. IR
(CHCl3) ꢀmax (cm–1): 1701. 1H NMR d: 1.05 (s, 9H), 1.40 (s,
9H), 1.58–1.60 (m, 3H), 3.79–3.83 (dd, J = 10.3, 2.9 Hz,
1H), 3.90–4.10 (br s, 1H), 4.41 (br s, 2H), 4.67 (br s, 1H),
5.39–5.50 (m, 1H), 7.34–7.46 (m, 6H), 7.65–7.72 (m, 4H).
13C NMR d: 13.5, 19.5, 26.6, 28.3, 55.1, 63.6, 69.9, 79.2,
116.8, 127.5, 129.5, 131.5, 133.4, 135.5, 155.9.
m/z (rel intensity): 129 ([M + H]+, 98), 84 (100). HRMS
calcd. for C6H10NO2 (M + H): 128.0794; found: 128.0722.
IR (neat) ꢀmax (cm–1): 3700–2000, 1614, 1379, 1291, 759,
1
690. H NMR (CD3OD) d: 1.56–1.58 (m, 3H), 4.50–4.70
(m, 2H), 5.09 (br s, 1H), 5.70–5.83 (m, 1 H). 1H NMR
(D2O) d: 1.53–1.56 (m, 3H), 4.60–4.73 (m, 2H), 5.24 (br s,
1H), 5.66–5.77 (m, 1H). 13C NMR (CD3OD) d: 13.1, 53.3,
70.7, 121.3, 128.4, 170.6.
(2S,Z)-1-tert-Butyloxycarbonyl-2-tert-butyldiphenylsilyl-
oxymethyl-3-(formyl)methylidene-azetidine (31)
Compound 23 (0.505 g, 0.962 mmol) was reduced with
lithium aluminum hydride (0.192 g, 95%, 4.81 mmol) at
–78°C for 2 h to afford 0.274 g (61%) of 31 and 0.109 g
(24%) of 32 both as colorless oils. For 31 : [a]D +35.9
(c 1.79, CHCl3). FAB-MS (NBA) m/z (rel intensity): 466
([M + H]+, 31), 427 (17), 352 (12), 198 (72), 154 (76), 135
(100). HRMS calcd. for C27H36NO4Si (M + H): 466.2415;
found: 466.2371. IR (CHCl3) ꢀmax (cm–1): 2739, 1701, 1685.
1H NMR d: 1.06 (s, 9H), 1.44 (s, 9H), 3.87–3.91 (d, J =
9.3 Hz, 1H), 4.55–4.61 (m, 1H), 4.62–4.69 (m, 2H), 5.20
(br s, 1H), 6.09–6.13 (m, 1H), 7.35–7.47 (m, 6H), 7.61–7.76
(m, 4H), 9.68–9.71 (d, J = 7.3 Hz, 1H). 13C NMR d: 19.1,
26.6, 28.2, 56.9, 62.7, 63.1, 71.7, 80.2, 123.8, 127.6, 129.7,
129.8, 132.6, 155.1, 159.3, 189.0.
(2S,E)-1-tert-Butyloxycarbonyl-2-hydroxymethyl-3-
ethylidene-azetidine (29)
To a solution of compound 28 (0.589 g, 1.30 mmol) in
THF (40.0 mL) at 0°C was added tetrabutylammonium fluo-
ride (1.96 mL, 1.0 M solution in THF, 1.96 mmol). The re-
sulted reaction mixture was stirred towards rt for 3 h. The
reaction mixture was concentrated in vacuo to dryness. Column
chromatography (EtOAc–hexanes, 1:3) afforded 0.261 g (94%)
of product 29 as a colorless oil: [a]D –8.9 (c 1.56, CHCl3).
FAB-MS (NBA) m/z (rel intensity): 214 ([M + H]+, 78), 182
(54), 158 (100), 136 (98), 126 (81), 107 (96), 97 (74). HRMS
calcd. for C11H20NO3 (M + H): 214.1444; found: 214.1422.
(2S,Z)-1-tert-Butyloxycarbonyl-2-tert-butyldiphenylsilyl-
oxymethyl-3-(hydroxymethyl)methylidene-azetidine (32)
Compound 31 (1.40 g, 3.01 mmol) was reduced with so-
dium borohydride (0.125 g, 3.31 mmol) in the presence of
cerium trichloride heptahydrate (1.24 g, 3.31 mmol) to af-
ford 1.16 g (82%) of 32 as a colorless oil: [a]D –0.2 (c 1.53,
CHC13). FAB-MS (NBA) m/z (rel intensity): 468 ([M + H]+,
21), 412 (14), 354 (52), 334 (46), 198 (71), 154 (85), 136
(100). HRMS calcd. for C27H38NO4Si (M + H): 468.2571,
found 468.2598. IR (neat) ꢀmax (cm–1): 3420, 1700, 1676. 1H
NMR d: 1.08 (s, 9H), 1.34 (s, 9H), 3.89–3.92 (dd, J = 10.3,
2.8 Hz, 1H), 3.95–4.21 (br s, 3H), 4.37–4.49 (m, 2H), 4.96
(br s, 1H), 5.63 (m, 1H), 7.36–7.45 (m, 6H), 7.66–7.70 (m,
4H). 13C NMR d: 19.1, 26.7, 28.2, 56.0. 59.7, 64.0, 70.0,
79.5, 122.3, 127.7, 129.7, 132.8, 133.3, 135.4, 155.6.
1
IR (CHCl3) ꢀmax (cm–1): 3425, 1701, 1658. H NMR d: 1.43
(s, 9H), 1.50–1.53 (m, 3H), 3.66–3.78 (m, 2H), 4.31–4.42 (m,
2H), 4.81 (br s, 1H), 5.30–5.40 (m, 1H). 13C NMR d: 13.3,
28.1, 55.1, 65.6, 71.3, 80.3, 117.7, 128.7, 156.9.
(2S,E)-1-tert-Butyloxycarbonyl-carboxylic acid 3-
ethylidene-2-azetidine (30)
To a solution of 29 (64.6 mg, 0.30 mmol) in acetone
(3.00 mL) at 0°C was added chromic acid (1.16 mL, 2.6 M
solution, 3.03 mmol). The resulting reaction mixture was
stirred towards rt for 3 h. Isopropanol was added to destroy
the residual chromic acid followed by the addition of water
(5.00 mL). The mixture was extracted with ether (3 × 10.0 mL)
and the combined ether solution was dried (Na2SO4), fil-
tered, and the filtrate was concentrated in vacuo to dryness.
Column chromatography (EtOAc–hexanes, 1:2, with 1% of
AcOH) afforded 26.5 mg (39%) of product 30: mp 138–
140°C (CH2Cl2–hexane). [a]D –63.5 (c 1.28, CHCl3). FAB-
MS (NBA) m/z (rel intensity): 228 ([M + H]+, 8), 220 (45),
172 (25), 154 (100), 136 (98), 107 (89). HRMS calcd. for
C11H18NO4 (M + H): 228.1236; found: 228.1244. IR (CHCl3)
(2S,Z)-1-tert-Butyloxycarbonyl-2-tert-butyldiphenylsilyl-
oxymethyl-3-(bromomethyl)methylidene-azetidine (33)
Compound 32 (0.315 g, 0.672 mmol) was reacted with
triphenylphosphine (0.265 g, 1.01 mmol) and carbon tetra-
bromide (0.335 g, 1.01 mmol) to afford 0.285 g (80%) of 33
as a colorless oil: [a]D +60.3 (c 1.19, CHCl3). FAB-MS
(NBA) m/z (rel intensity): 532 ([M + H]+, 3), 530 ([M + H]+,
2), 417 (11), 415 (10), 198 (37), 195 (34), 135 (100). HRMS
calcd. for C27H37BrNO3Si (M + H): 530.1727; found:
1
ꢀmax (cm–1): 3700–2500, 1729, 1707, 1687. H NMR d: 1.45
(s, 9H), 1.57–1.61 (m, 3H), 4.39–4.57 (m, 2H), 5.19 (br s,
1H), 5.78 (br s, 1H). 13C NMR d: 13.5, 28.2, 55.5, 67.9,
81.7, 120.7, 124.9, 156.6, 171.7. Anal. calcd for C11H17NO4:
C 58.14, H 7.54, N 6.16; found: C 57.99, H 7.57, N 6.12.
1
530.1754. IR (neat) ꢀmax (cm–1): 1701. H NMR d: 1.09 (s,
9H), 1.39 (s, 9H), 3.81–3.96 (m, 2H), 3.96–4.15 (br s, 2H),
4.39–4.53 (m, 2H), 4.93 (br s, 1H), 5.73 (m, 1H), 7.37–7.47
(m, 6H), 7.69–7.72 (m, 4H). 13C NMR d: 19.1, 26.7, 27.5,
28.2, 55.7, 63.2, 69.9, 79.6, 119.4, 127.6, 129.6, 132.9,
135.4, 137.4, 155.4.
(2S,E)-3-Ethylidene-azetidine-2-carboxylic acid (1)
A solution of compound 30 (0.104 g, 0.459 mmol) in for-
mic acid (5.00 mL) was stirred at rt for 5 h. After filtration
through a fine sintered glass funnel, the filtrate was concen-
trated in vacuo on an oil pump to dryness to give 0.058 g
(100%) of product 1: mp 147–149°C (triturated with metha-
nol and isopropanol). [a]D +17.3 (c 2.10, MeOH). CI-MS
(2S,E)-1-tert-Butyloxycarbonyl-2-tert-butyldiphenylsilyl-
oxymethyl-3-ethylidene-azetidine (34)
Compound 33 (0.285 g, 0.537 mmol) was reacted with
© 2001 NRC Canada

Hanessian and Fu
1825
lithium borohydride (0.54 mL, 2.0 M solution in THF,
1.08 mmol) to afford 0.181 g (79%) of 34 as a colorless oil:
[a]D +4.4 (c 1.38, CHCl3). FAB-MS (NBA) m/z (rel inten-
sity): 452 ([M + H]+, 20), 396 (19), 352 (18), 338 (54), 318
(51), 225 (52), 199 (61), 154 (82), 136 (100). HRMS calcd.
for C27H38NO3Si (M + H): 452.2685; found: 452.2598. IR
studies in carbenoid insertions, and Dr. Y. Tu for NMR
experiments. We are grateful to Professor K. Isono for an
authentic samples of polyoximic acid and polyoxin A and to
Dr. Michel Simard for X-ray analyses.
References
1
(neat) ꢀmax (cm–1): 1700. H NMR d: 1.08 (s, 9H), 1.43 (s,
1. (a) K. Isono. Biotechnol. Ser. 28, 597 (1995); (b) K.Y.I. El-
Shahed. Egypt. J. Microbio. 29, 315 (1995); (c) M. Uramoto,
M. Matsuoka, J.G. Liehr, J.A. McCloskey, and K. Isono. Agr.
Biol. Chem. (Tokyo) 45, 1901 (1981); (d) S. Suzuki, K. Isono,
M. Uramoto, H. Okamoto, and M. Matsuoka. Japan patent,
No. 5 2020 555 (1977); (e) K. Isono and S. Suzuki. Tetrahedron,
24, 1133 (1968); (f) K. Isono, K. Kobinata, and S. Suzuki.
Agr. Biol. Chem. (Tokyo) 32, 1193 (1968); (g) K. Isono and S.
Suzuki. Agr. Biol. Chem. (Tokyo) 32, 792 (1968); (h) S. Suzuki,
K. Isono, J. Nagatsu, Y. Kawashima, K. Yamagata, K. Sasaki,
and K. Hashimoto. Agr. Biol. Chem. (Tokyo) 30, 817 (1966);
(i) S. Suzuki, K. Isono, J. Nagatsu, T. Mizutani, K. Kawashina,
and T. Mizuno. J. Antibiotic A18, 131 (1965); (j) K. Isono, J.
Nagatsu, Y. Kawashima, and S. Suzuki. Agr. Biol. Chem. (To-
kyo) 29, 848 (1965).
9H), 1.59–1.61 (m, 3H), 3.85–3.89 (dd, J = 10.7, 2.1 Hz,
1H), 4.00–4.25 (br s, 1H), 4.30–4.52 (m, 2H), 4.82 (br s,
1H), 5.39–5.50 (m, 1H), 7.35–7.47 (m, 6H), 7.67–7.75 (m,
4H). 13C NMR d: 13.6, 19.2, 26.8, 28.3, 56.1, 62.5, 70.2,
79.1, 117.3, 127.5, 129.5, 130.8, 133.3, 135.5, 155.6.
(2S,Z)-1-tert-Butyloxycarbonyl-2-hydroxymethyl-3-
ethylidene-azetidine (35)
Compound 34 (0.912 g, 2.02 mmol) was reacted with
tetrabutylammonium fluoride (3.03 mL, 1.0 M solution in
THF, 3.03 mmol) to afford 0.418 g (97%) of 35 as a color-
less oil: [a]D –12.3 (c 1.98, CHCl3). FAB-MS (NBA) m/z
(rel intensity): 214 ([M + H]+, 20), 158 (55), 136 (100), 107
(92). HRMS calcd. for C11H20NO3 (M + H): 214.1444;
found: 214.1460. IR (neat) ꢀmax (cm–1): 3423, 1701, 1658.
1H NMR d: 1.45 (s, 9H), 1.57–1.61 (m, 3H), 3.78–3.85 (dd,
J = 11.6, 7.6 Hz, 1H), 3.90–3.95 (dd, J = 11.6, 7.6 Hz, 1H)
4.32–4.34 (m, 2H), 4.96 (br s, 1H), 5.31–5.42 (m, 1H). 13C
NMR d: 13.6, 28.2, 55.9, 64.5, 71.3, 80.4, 119.0, 128.1, 154.6.
2. S. Suzuki. Hakko To Taisha, 31, 16 (1975); Chem. Abstr. 83,
193 595 (1975).
3. (a) K. Isono. In Shokubutsu byogai no kagaku. Edited by A.
Ichihara and T. Ueno. Gakkai Shuppan Senta, Tokyo. 1997. p.
263; Chem. Abstr. 130, 263 433 or 85 598 (1999); (b) I.K.O.
Cann, Y. Kobayashi, A. Onoda, M. Wakita, and S. Hoshino.
J.Appl. Bacteriol. 74, 127 (1993).
4. (a) M. Reuveni, H. Cohen, T. Zahavi, and A. Venezian. Crop
Prot. 19, 393 (2000); (b) Y. Jiang. Int. J. Food Sci. Technol.
32, 535 (1997); (c) Y. Elad, B. Kirshner, and Y. Gotlib. Crop
Prot. 12, 69 (1993); (d) E. Krainer, J.M. Becker, and F.J.
Naider. Med. Chem. 34, 174 (1991).
(2S,Z)-1-tert-Butyloxycarbonyl-3-ethylidene-azetidine-2-
carboxylic acid (36)
Compound 35 (0.178 g, 0.840 mmol) was oxidized with
chromic acid (3.22 mL, 2.6 M solution, 8.36 mmol) to
afford 76.0 mg (40%) of 36: mp 83–84°C (CH2Cl2–hex-
ane). [a]D –15.3 (c 2.10, CHCl3). EI-MS m/z (rel intensity):
([M + H]+, 21), 182 (52), 172 (58), 154 (33), 142 (58), 126
(74), 76 (100). HRMS calcd. for C11H18NO4 (M + H):
228.1236; found: 228.1214. IR (neat) ꢀmax (cm–1): 3700–
5. K. Isono, K. Asabi, and S. Suzuki. J. Am. Chem. Soc. 91,
7490 (1969) and refs. cited therein.
6. (a) T. Sato, K. Hirasawa, J. Uzawa, T. Inaba, and K. Isono.
Tetrahedron Lett. 36, 3441 (1979); (b) K. Isono, T. Sato, K.
Hirasawa, S. Funayama, and S. Suzuki. J. Am. Chem. Soc.
100, 3937 (1978); (c) K. Isono and R. Suhadolnik. J. Arch.
Biochem. Biophys. 173, 141 (1976); (d) K. Isono, K.
Funayama, and R. Suhadolnik. Biochemistry, 14, 2992 (1975).
7. H. Sakurai, T. Akashiba, and S. Suzuki. Noyaku Seisan Gijutsu,
21, 13 (1970); Chem. Abstr. 74, 52 425 (1971).
1
2500, 1731, 1702, 1690. H NMR d: 1.45 (s, 9H), 1.69–1.72
(m, 3H), 4.36–4.42 (m, 1H), 4.50–4.58 (m, 1H), 5.21 (m,
1H), 5.48–5.59 (m, 1H). 13C NMR d: 13.5, 28.2, 56.8, 68.0,
81.0, 121.3, 125.1, 155.8, 172.4. Anal. calcd for C11H17NO4:
C 58.14, H 7.54, N 6.16; found: C 57.81, H 7.42, N 6.06.
8. D. Zhang and M.J. Miller. Curr. Pharm. Des. 5, 73 (1999) and
refs. cited there in.
(2S,Z)-3-Ethylidene-azetidine-2-carboxylic acid (2)
A solution of 36 (0.107 g, 0.472 mmol) in formic acid
(5.00 mL) was stirred at rt for 5 h. After filtration through a
fine sintered glass funnel, the filtrate was concentrated in vacuo
on an oil pump to dryness to afford 0.060 g (100%) of 2: mp
153–155°C (triturated with methanol and EtOAc). [a]D +41.3
(c 0.55, MeOH). CI-MS m/z (rel intensity): 128 ([M + H]+,
20), 84 (43), 82 (100), 68 (48). HRMS calcd. for C6H10NO2
(M + H): 128.0794; found: 128.0722. IR (neat) ꢀmax (cm–1):
3700–2000, 1640, 1383, 1290, 851, 690. 1H NMR (CD3OD) d:
1.76–1.80 (m, 3H), 4.45–4.63 (m, 2H), 5.14–5.16 (br s, 1H),
9. (a) O. Tamura, K. Gotanda, J. Yoshino, Y. Morita, R. Terashima,
M. Kikuchi, T. Miyawaki, N. Mita, M. Yamashita, H. Ishibashi,
and M. Sakamoto. J. Org. Chem. 65, 8544 (2000); (b) P. Me-
rino, S. Franco, F. Merchan, and T. Tejero. J. Org. Chem. 65,
5575 (2000); (c) A. Chen, E. J. Thomas, and P.D. Wilson. J.
Chem. Soc., Perkin Trans. 1, 3305 (1999); (d) K. Kato, C.Y.
Chen, and H. Akita. Synthesis, 1527 (1998); (e) A.K. Ghosh
and Y. Wang. J. Org. Chem. 63, 6735 (1998); (f) P. Merino, S.
Franco, F.L. Merchan, and T. Tejero. Tetrahedron Lett. 39,
6411 (1998); (g) J.-Q. Wang, C. Archer, J. Li, S.G. Bott, and
P.G. Wang. J. Org. Chem. 63, 4850 (1998); (h) H. Akita, K.
Uchida, C.Y. Chen, and K. Kato. Chem. Pharm. Bull. 46, 1034
(1998); (i) H. Akita, K. Uchida, and C.Y. Chen. Heterocycles,
46, 87 (1997); (j) C. Dehoux, E. Fontaine, J.-M. Escudier, M.
Baltas, and L. Gorrichon. Adv. Chitin Sci. 2, 279 (1997); (k)
D. Zhang and M.J. Miller. J. Org. Chem. 63, 755 (1998); (l)
D.M. Gethin and N.S. Simpkins. Tetrahedron, 53, 14417 (1997);
(m) V.K. Aggarwal and N.J. Monteiro. Chem. Soc. Perkin
1
5.46–5.56 (m, 1H). H NMR (D2O) d: 1.63–1.68 (m, 3H),
4.49–4.68 (m, 2H), 5.29 (m, 1H), 5.50–5.60 (m, 1H). 13C
NMR (CD3OD) d: 13.6, 54.7, 71.2, 122.8, 127.7, 170.1.
Acknowledgments
We are grateful to NSERC for financial support of this
work. We thank Drs. J.-L. Chiara and R. DiFabio for initial
© 2001 NRC Canada

1826
Trans. 1, 2531 (1997); (n) C.M. Evina and G. Guillerm. Tetra-
Can. J. Chem. Vol. 79, 2001
Brunner. Angew. Chem. Int. Ed. Engl. 31, 1183 (1992); (e) A.
Padwa and K.E. Krumpe. Tetrahedron, 48, 5385 (1992); (f)
D.F. Taber. In Comprehensive organic synthesis. Vol. 3.
Edited by B.M. Trost, I. Fleming, and G. Pattenden. Pergamon
Press, New York. 1991. p. 1045; (g) J. Adams and D.M.
Spero. Tetrahedron, 47, 1765 (1991); (h) A. Padwa and S.F.
Hombuckle. Chem. Rev. 91, 263 (1991) and refs. cited therein.
18. (a) G. Lawton, C.J. Moody, and C.J. Pearson. J. Chem. Soc.
Perkin Trans. 1, 899 (1987); (b) M.P. Moyer, P.L. Feldman,
and H. Rapoport. J. Org. Chem. 50, 5223 (1985).
19. S. Hanessian and P. Lavallee. Can. J. Chem. 53, 2975 (1975).
20. E.J. Corey and A. Venkateswarlu. J. Am. Chem. Soc. 94, 6190
(1972).
21. M. Schlosser, K.-F. Christmann, and A. Piskala. Chem. Ber.
103, 2814 (1970).
22. B.E. Maryanoff and A.B. Reitz. Chem. Rev. 89, 863 (1989).
23. S. Nahm and S. Weinreb. Tetrahedron Lett. 22, 3815 (1981).
24. S. Hanessian and Y.L. Bennani. Chem. Rev. 97, 3161 (1997);
S. Hanessian, Y. Bennani, and Y. Leblanc. Heterocycles, 35,
1411 (1993).
25. J.-L. Luche. J. Am. Chem. Soc. 103, 5454 (1981).
26. (a) R.K. Hill, S.W. Rhee, K. Isono, D.H.G. Crout, and R.J.
Suhadolnik. Biochemistry, 20, 7040 (1981); (b) R. Doyle and
B. Levenberg. Biochemistry, 7, 2457 (1968).
hedron Lett. 37, 163 (1996); (o) A. Dondoni, F. Junquera, F.
Merchan, P. Merino, and T. Tejero. Tetrahedron Lett. 35, 9439
(1994); (p) A. Chen, I. Savage, E.J. Thomas, and P.D. Wilson.
Tetrahedron Lett. 34, 6769 (1993); (q) Y. Auberson and P.
Vogel. Tetrahedron, 46, 7019 (1990); (r) A.G.M. Barrett and
S.A. Lebold. J. Org. Chem. 55, 3853 (1990); (s) P. Garner and
J.M. Park. J. Org. Chem. 55, 3772 (1990); (t) G. Emmer, N.S.
Ryder, and M.A. Grassberger. J. Med. Chem. 28, 278 (1985).
10. K. Uchida, K. Kato, K. Yamaguchi, and H. Akita. Heterocycles
53, 2253 (2000).
11. (a) K. Uchida, K. Kato, and H. Akita. Synthesis, 1678 (1999);
(b) K.M.K. Kutterer and G. Just. Heterocycles, 51, 1409
(1999); (c) A.K. Ghosh and Y.J. Wang. Org. Chem. 64, 2789
(1999); (d) H. Akita, K. Uchida, and K. Kato. Heterocycles,
47, 157 (1998); (e) A. Dondoni, S. Franco, F. Junquera, F.L.
Merchan, P. Merino, and T. Tejero. J. Org. Chem. 62, 5497
(1997); (f) A. Dondoni, F. Junquera, F.L. Merchan, P. Merino,
and T.J. Tejero. Chem. Soc. Chem. Commun. 2127 (1995); (g)
N. Chida, K. Koizumi, Y. Kitada, C. Yokoyama, and S. Ogawa.
J. Chem. Soc. Chem. Commun. 111 (1994).
12. (a) I. Savage and E.J. Thomas. J. Chem. Soc. Chem. Commun.
717 (1989); (b) P. Garner and J.M. Park. J. Org. Chem. 53,
2979 (1988); (c) H. Kuzuhara, M. Kimura, and S. Emoto.
Carbohydr. Res. 45, 245 (1975).
13. H. Baumann and R.O. Duthaler. Helv. Chim. Acta, 71, 1025,
1035 (1988).
27. S. Hanessian, N. Bernstein, R.-Y. Yang, and R. Maguire. Bioorg.
Med. Chem. Lett. 9, 1437 (1999).
28. (a) W.J. Krol, S. Mao, D.L. Steele, and C.A. Townsend. J.
Org. Chem. 56, 728 (1991); (b) S. Mzengeza and R.A. Whit-
ney. J. Org. Chem. 53, 4024 (1989); (c) D.H.R. Barton, D.
Crich, Y. Herve, P. Potier, and J. Thierry. Tetrahedron, 41,
4347 (1985); (d) S. Hanessian and S.P. Sahoo. Tetrahedron
Lett. 25, 1425 (1984); (e) A. Afzali-Ardakin and H.J. Rapoport.
J. Org. Chem. 45, 4817 (1980).
29. (a) S. Hanessian, C. McNaugton-Smith, H.-G. Lombart, and
W.D. Lubell. Tetrahedron, 53, 12789 (1997); (b) J. Rizo and
L.M. Giersach. Ann. Rev. Biochem. 61, 387 (1992); (c) G.
Holzmann. Kontakte, 3 (1991); (d) J.B. Ball and P.F. Alewood.
J. Mol. Recogn. 3, 55 (1990); (e) M. Kahn, S. Wilke, B. Chen,
K. Fujita, Y.-H. Lee, and M.E. Johnson. J. Mol. Recogn. 1, 75
(1988).
14. G. Emmer. Tetrahedron, 48, 7165 (1992).
15. (a) S. Hanessian, J.-M. Fu, J.L. Chiara, and R. DiFabio. Tetra-
hedron Lett. 34, 4157 (1993); (b) S. Hanessian, J.-M. Fu, Y.
Tu, and K. Isono. Tetrahedron Lett. 34, 4153 (1993).
16. S. Hanessian. In Total synthesis of natural compounds: The
“Chiron” approach. Vol. 3. Edited by J.E. Baldwin. Organic
Chemistry Series, Pergamon Press, Oxford. 1983.
17. (a) M.P. Doyle, M.A. McKervey, and T. Ye. In Modern cata-
lytic methods for organic synthesis with diazo compounds:
From cyclopropanes to ylides. Wiley, New York. 1998; (b) M.
Doyle, M.A. McKervey, and T. Ye. In Reaction synthesis with
a-diazocarbonyl compounds. Wiley, New York. 1997; (c) T.
Ye and M.A. McKervey. Chem. Rev. 94, 1091 (1994); (d) H.
© 2001 NRC Canada