Diastereoselective synthesis of 3,5-trans-(+)-(3R,5R)-3-carbomethoxycarbapenam from 3-hydroxypyridine: Questioning the stereochemical assignment of the natural product

Article abstract of DOI:10.1016/S0040-4039(01)02073-1

An enantio- and diastereoselective synthesis of the methyl ester of 3,5-trans-(3R,5R)-carbapenam-3-carboxylic acid from 3-hydroxypyridine via (2R,5S)-trans-5-acetoxy-2-allylpiperidine has been achieved by employing the piperidine-pyrrolidine ring-contract

Full text of DOI:10.1016/S0040-4039(01)02073-1

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 93–96
Diastereoselective synthesis of
3,5-trans-(+)-(3R,5R)-3-carbomethoxycarbapenam from
3-hydroxypyridine: questioning the stereochemical assignment of
the natural product
Hideyuki Tanaka, Hideki Sakagami and Kunio Ogasawara*
Pharmaceutical Institute, Tohoku University, Aobayama, Sendai 980-8578, Japan
Received 11 October 2001; revised 26 October 2001; accepted 1 November 2001
Abstract—An enantio- and diastereoselective synthesis of the methyl ester of 3,5-trans-(3R,5R)-carbapenam-3-carboxylic acid
from 3-hydroxypyridine via (2R,5S)-trans-5-acetoxy-2-allylpiperidine has been achieved by employing the piperidine–pyrrolidine
ring-contraction reaction as the key step. The synthesis indicates that the configuration of the naturally occurring 3,5-trans-car-
bapenam-3-carboxylic acid is not the revised (3S,5S), but rather the originally assigned (3R,5R) configuration. © 2001 Elsevier
Science Ltd. All rights reserved.
The isolation of 3,5-trans-carbapenam-3-carboxylic
acid 1 from strains of Serratia and Erwinia spp. was
reported by Bycroft and co-workers.1 Although its
absolute configuration was assigned as (3R,5R) based
on spectroscopic analysis, this assignment was later
revised2 to be the opposite (3S,5S) configuration, since
the synthetic carbapenam methyl ester (3R,5R)-2 syn-
thesized from (R)-glutamic acid was found to be the
antipode of the one obtained from the natural com-
pound (Fig. 1).
to explore the enantio- and diastereocontrolled synthe-
sis of the methyl ester 2 of the carbapenam-3-carboxylic
acid 1 to extend the synthetic utility of our chiral
product. We report here the stereoselective synthesis of
the methyl ester 2 of the carbapenam 1 having the
(3R,5R) configuration. This configuration argues
strongly that the natural 3,5-trans-carbapenam-3-car-
boxylic acid 1 has the originally assigned (3R,5R)
configuration, and not the revised (3S,5S) configuration
(Scheme 1).
Because carbapenems and carbapenams with a 5S
configuration had not been known before, this finding
was of particular interest to the biosynthetic
researchers. Since we had discovered3,4 the unprece-
dented N-acylative reduction of 3-hydroxypyridine 3 to
produce the 3-hydroxytetrahydropyridine enamide 4
and its diastereoselective transformation into the enan-
tiopure trans-2,5-disubstituted piperidine 7, we decided
The starting enantiopure (2R,5S)-trans-5-acetoxy-2-
allylpiperidine was prepared from 3-hydroxypyridine 3
following the five-step sequence of our established
route3,4 using N-acylative reduction, lipase-mediated
resolution, methanol addition, acetylation, and
diastereoselective allylation. Oxidative cleavage of the
olefin functionality followed by esterification5 trans-
formed the allyl piperidine (+)-7 into the tert-butyl ester
(−)-8, [h]2D5 −9.1 (c 1.1, CHCl3). The N-carbobenzoxy
(N-Cbz) functionality of the ester (−)-8 was then
replaced by the N-benzyl functionality by sequential
H
H
5
5
3
3
N
N
H
H
H
O
O
present
study
OH
ref. 3
CO2H
CO2H
OH
N
(3R,5R)-1
(3S,5S)-1
H
N
O
N
CO2Me
Cbz
(+)-4
Figure 1.
* Corresponding author. E-mail: konol@mail.cc.tohoku.ac.jp
3
(3R,5R)-2
Scheme 1.
0040-4039/02/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)02073-1

94
H. Tanaka et al. / Tetrahedron Letters 43 (2002) 93–96
hydrogenolysis and N-benzylation to give the tertiary
amine (+)-9, [h]2D6 +15.8 (c 1.3, CHCl3), the acetyl
functionality of which was removed by alkaline
methanolysis to yield the (2R,5S)-5-piperidinol (−)-10,
[h]2D7 −4.4 (c 0.8, CHCl3) (Scheme 2).
acetate (+)-9 and was subsequently recycled. Both prod-
ucts presumably came about via an intervention of the
common aziridinium intermediate 12, generated ini-
tially from the mesylate 10 by an intramolecular SN2
reaction. The two products were formed diastereoselec-
tively by two competitive intermolecular SN2 pathways,
namely, the pyrrolidine (+)-13 via route a and the
piperidine (+)-9 via route b (Scheme 3).
To carry out the key piperidine–pyrrolidine ring-con-
traction reaction, the (2R,5S)-piperidinol 10 was trans-
formed into the mesylate 11 which was exposed to
cesium acetate6a,c in warm (ꢀ70°C) DMF for 5 h to
furnish the (2R,5R)-trans-2,5-disubstituted pyrrolidine
(+)-13, [h]2D3 +64.1 (c 0.9, CHCl3), and the trans-2,5-di-
substituted piperidinol acetate (+)-9, [h]2D7 +15.4 (c 1.1,
CHCl3), in yields of 75 and 18%, respectively. The
resulting acetate was identical to the above intermediate
To adjust the oxidation stage of the C5 functionality to
that of the target molecule 2, the 5-substituted prolinol
acetate (+)-13 thus obtained was first transformed into
the 5-substituted prolinol (+)-14, [h]2D8 +52.7 (c 1.4,
CHCl3), by alkaline methanolysis. Since its direct oxi-
dation was found to be difficult, the benzylamine (+)-14
OAc
OH
NaBH4
Cbz-Cl
OH
OH
Lipase PS
vinyl acetate
CH2Cl2
+
N
N
NaHCO3
EtOH
(70%)
N
(+)-5
(+)-4
N
Cbz
Cbz
Cbz
47%:>99%ee
48%:>99%ee
4
3
1) p-TsOH(cat.)
OAc
OAc
allyl-TMS
MeOH
(+)-4
2) Ac2O, Et3N
ZnCl2
MeO
N
N
CH2Cl2
–78°C - rt
(80%)
DMAP(cat.)
CH2Cl2
Cbz
Cbz
(+)-7
6
(78%)
1) OsO4(cat.), NMO
then NaIO4
1) H2, Pd(OH)2
OR
OAc
O
O
EtOH
2) BnBr
Hünig base
2) NaClO2, NaH2PO4
t-BuO
N
t-BuO
N
2-methyl-2-butene
aq.t-BuOH
Bn
Cbz
MeCN
(97%)
(+)-9: R=Ac
(–)-8
K2CO3
MeOH
3) Boc2O, DMAP(cat.)
(–)-10: R=OH
t-BuOH
(100%)
(60%)
Scheme 2.
b
–OAc
OMs
O
O
MsCl, Et3N
CsOAc
+
N
a
(–)-10
t-BuO
N
t-BuO
DMF
CH2Cl2
Bn
11
MsO–
Bn
12
70°C, 5h
OAc
O
O
OAc
+
t-BuO
N
t-BuO
N
Bn
Bn
(+)-13
(75%; 2 steps)
(+)-9
(18%; 2 steps)
Scheme 3.

H. Tanaka et al. / Tetrahedron Letters 43 (2002) 93–96
95
O
O
H2, Pd(OH)2
EtOH
K2CO3
OR
OTBDPS
t-BuO
t-BuO
(+)-13
N
N
R
MeOH
(98%)
Bn
16: R=H
(+)-17: R=Boc
Boc2O
1N NaOH
dioxane
(+)-14: R=H
(+)-15: R= TBDPS
TBDPS-Cl
imidazole
DMF
O
O
1) Swern oxidation
2) NaClO2, NaH2PO4
Bu4NF
THF
(98%: 4 steps)
OMe
OH
t-BuO
t-BuO
N
N
O
2-methyl-2-butene
then CH2N2
Boc
Boc
(+)-18
(+)-19
(98%)
Scheme 4.
was transformed into the tert-butyl carbamate (Boc)
(+)-18, [h]2D3 +44.5 (c 1.1, CHCl3), by a four-step
sequence of reactions involving O-protection,
hydrogenolysis, carbamoylation, and O-deprotection
via the tert-butyldiphenylsilyl (TBDPS) ether 15, the
secondary amine 16, and the N-tert-butylcarbamate
(N-Boc) 17. Oxidation of the carbamate (+)-18 pro-
ceeded without difficulty to give the methyl ester, (+)-
19, [h]2D6+56.6 (c 1.1, CHCl3), by sequential Swern
oxidation,7 chlorite-mediated oxidation,4 and esterifica-
tion (Scheme 4).
After the trans-2,5-disubstituted pyrrolidine (+)-19 with
the two distinguishable ester functionalities was
obtained, it was exposed to trifluoroacetic acid in
dichloromethane at room temperature to give the half
ester 23 by selective removal of the tert-butyl ester
functionality. When the methyl ester 23, after conver-
sion into the hydrochloride, was dissolved in water,
facile hydrolysis occurred to give the diacid (+)-22, mp
130–131°C, [h]3D0 +113.0 (c 1.0, H2O), after purification
using an ion-exchange resin (Dowex 50W-X8). In order
to obtain the target carbapenam ester 2, the half ester
23 was refluxed with di(2-pyridyl) disulfide,
triphenylphosphine and triethylamine in acetonitrile8
for 8 h. Although the reaction did not proceed as
efficiently as we had hoped, the expected carbapenam
ester 2, [h]3D0 +199.1 (c 0.2, CHCl3), having the (3R,5R)
configuration, was nevertheless obtained in 17% yield.
Its spectroscopic data were identical in all respects with
those reported for the (3R,5R)-ester 2,2,9 [h]D30 −197 (c
0.5, CHCl3), obtained from (R)-glutamic acid. How-
ever, the present product showed the opposite sign in
its optical rotation, indicating that it was enantiomeric
with the one reported.2 Although we cannot account
for this discrepancy, we still believe that our product
has the correct structure since the starting material was
carefully checked and the absolute structure has been
defined by its conversion into several structurally-
defined compounds3a,10 including tert-butyl (+)-2,5-
trans-(2R,5R)-1-tert-butoxycarbonyl-5-(2-hydroxyethyl)
Scheme 5.
prolinate11,12 from the same intermediate (+)-17 used in
the present synthesis (Scheme 5).
In conclusion, using the enantiopure piperidine precur-
sor, we have synthesized the (3R,5R)-3-carbomethoxy-
carbapenam that is identical with the methyl ester of
the naturally occurring carbapenam-3-carboxylic acid.
The key step of this synthesis is utilizing the ring
contraction. The present synthesis indicates unambigu-
ously that the absolute configuration of the naturally
occurring 3,5-trans-carbapenam-3-carboxylic acid is
(3R,5R) and not the opposite (3S,5S), as had previ-
ously been proposed.
Acknowledgements
We thank the Ministry of Education, Culture, Sports,
Science and Technology, Japan, for financial support.
We also thank the Japan Society for the Promotion of
Science for a fellowship (to H.S.).

96
H. Tanaka et al. / Tetrahedron Letters 43 (2002) 93–96
Boelens, M.; Contreras, J.; DeKimpe, N.; Knight, D. W.
References
Tetrahedron Lett. 2000, 41, 2507; (c) Sakagami, H.;
Ogasawara, K. Synlett 2001, 45.
7. Mancuso, A. J.; Swern, D. Synthesis 1981, 165.
8. Kobayashi, S.; Iimori, T.; Izawa, T.; Ohno, M. J. Am.
Chem. Soc. 1981, 103, 2405.
9. For a racemic synthesis of the 3,5-trans-ester ( )-2, see:
Gilchrist, T. L.; Lemos, A.; Ottaway, C. J. J. Chem. Soc.,
Perkin Trans. 1 1997, 3005.
1. (a) Bycroft, B. W.; Maslen, C.; Box, S. J.; Brown,
A. G.; Tyler, J. W. J. Chem. Soc., Chem. Commun.
1987, 1623; (b) Bycroft, B. W.; Marslen, C.; Box,
S. J.; Brown, A. G.; Tyler, J. W. J. Antibiot. 1988, 41,
1231.
2. Bycroft, B. W.; Chhabra, S. R. J. Chem. Soc., Chem.
Commun. 1989, 423.
3. (a) Sakagami, H.; Kamikubo, T.; Ogasawara, K. Chem.
Commun. 1996, 1433; (b) Sakagami, H.; Ogasawara, K.
Synthesis 2000, 521.
10. (a) Sakagami, H.; Kamikubo, T.; Ogasawara, K. Synlett
1997, 221; (b) Sakagami, H.; Ogasawara, K. Heterocycles
2001, 54, 43.
4. Lindgren, B. O.; Nolsson, T. Acta Chem. Scand. 1973, 27,
888.
11. Mulzer, J.; Schu¨lzchen, F.; Bats, J.-W. Tetrahedron 2000,
56, 4289.
5. Takeda, K.; Akiyama, A.; Nakamura, H.; Takizawa, S.;
Mizuno, H.; Takayanagi, H.; Harigaya, Y. Synthesis
1994, 1063.
12. Compound (+)-17 afforded tert-butyl (+)-2,5-trans-
(2R,5R)-1-tert-butoxycarbonyl-5-(2-hydroxyethyl)pro-
linate, [h]2D7 +32.3 (c 0.2, CHCl3) {lit.11 for the (2S,5S)-
enantiomer: [h]2D0 −31.2 (c 1.12, CHCl3)}, on a sequence
of six steps of reactions.
6. (a) Poitout, L.; LeMerrer, Y.; Depezay, J.-C. Tetrahedron
Lett. 1996, 37, 1609; (b) Tehrani, K. A.; Syngel, K. V.;
.