Stereoselective synthesis of (3R,5R)-cis-3-hydroxy-5-phenylpyrrolidine

Article abstract of DOI:10.1055/s-2001-18106

Cis-3-hydroxy-5-phenyl pyrrolidine could be synthesized stereoselectively from inexpensive starting materials, (R)-(+)-ethyl-4-chloro-3-hydroxybutanoate and (R)-(+)-4-chloro-3-hydroxybutyronitrile. Key feature involves face- and chemo-selective hydrogenat

Full text of DOI:10.1055/s-2001-18106

1808
LETTER
Stereoselective Synthesis of (3R,5R)-cis-3-Hydroxy-5-phenylpyrrolidine
Stereoselective
Sy
e
nthesis of (3
n
R,5R)-cis-3
j
-Hydro
i
xy-5-phenylp
M
yrrolidine aeda,* Yuhei Yamamoto, Koji Tomimoto, Toshiaki Mase
Process Research, Process R&D, Laboratories for Technology Development, Banyu Pharmaceutical Co., Ltd, Kamimutsuna 3-chome-9-1,
Okazaki, Aichi 444-0858, Japan
Fax +81(564)517086; E-mail: maedakj@banyu.co.jp
Received 9 August 2001
synthesis. Herein we wish to report a general stereoselec-
Abstract: Cis-3-hydroxy-5-phenyl pyrrolidine could be synthe-
tive synthesis of 4.
sized stereoselectively from inexpensive starting materials, (R)-(+)-
ethyl-4-chloro-3-hydroxybutanoate and (R)-(+)-4-chloro-3-hy-
droxybutyronitrile. Key feature involves face- and chemo-selective
hydrogenation of cyclic imine 5 as the common key intermediate
using Pt/C catalyst. Transformations described here will allow a
practical synthesis of novel carbapenems, 1 and 2.
Our strategy for an efficient construction of 4 is outlined
in Scheme 1. Diastereo and chemoselective reduction of
the cyclic imine 5 would be one of most promising ap-
proaches to the desired pyrrolidine ring 4. However, it has
been known that 5 is quite unstable due to its liability to
lose the silyloxy group to form the corresponding pyr-
role.6 This nature makes the synthesis more challenging
although cyclic imines have been well recognized to be
valuable intermediates for alkaloid synthesis.6 In order to
remove our apprehension, efficient generation of the cy-
clic imine 5 followed by spontaneous reduction under
very mild condition would be key to the success. We en-
visioned that hydrogenation of azide 7 affords the desired
pyrrolidine 4 via the reductive cyclization of the resulting
aminoketone 6 in one operation.7 This approach can pro-
vide a domino reaction to obviate the isolation of 5. Azide
7 would be readily prepared from inexpensive (R)-(+)-
ethyl-4-chloro-3-hydroxybutanoate (8).
Key words: reductive amination, cyclic imine, hydrogenation, pyr-
rolidine, 1 -methylcarbapenem
1 -methylcarbapenem antibiotics having a (3S,5S)-cis-3-
thio-5-substituted pyrrolidine ring as the C-2 side chain
2
such as meropenem,1 S-4661, and BO-2727,3 which
have a broad gram-positive and -negative antibacterial
spectrum, are widely known, and their syntheses have
been well established. On the other hand, it has been re-
cently reported that 1 and 2 possessing an unique (3S,5R)-
trans-3-thio-5-substituted pyrrolidine ring C-2 side chain
showed ultra-broad antibacterial activity even against
MRSA (Figure).4
Azide 7 was prepared as shown in Scheme 2. The hydroxy
group of 8 was protected as TBS ether. Conversion of es-
ter 9 to Weinreb’s amide 10 utilizing William’s proce-
dure,8 followed by addition of phenyl Grignard reagent
gave ketone 11 in good overall yield. Direct displacement
of 11 with various metal azides was very sluggish even
under forcing conditions and when bromo or iodo deriva-
tives were used. These observations were probably due to
the huge steric hindrance at the reactive site. In fact, azide
group was introduced smoothly under the same conditions
after removal of the TBS group. Finally, re-protecton of
Although the (3S,5R)-trans-disubstituted pyrrolidine skel-
eton seems to be essential for appearance of these remark-
able activities, development of an efficient synthesis has
not been fully investigated.5 Therefore, in the course of
our studies toward a practical and large scale synthesis of
1 or 2, a stereoselective synthesis of the thiol prototype 3
was extensively studied. Since the thiol group is prefera-
bly introduced by SN2 fashion in a late stage of the synthe-
sis, stereoselective construction of (3R,5R)-cis-3-
hydroxy-5-phenyl pyrrolidine (4) is key for success of the
R
HO
HO
H
H
N
H
H
N
5
5
CONMe2
3
3
S
S
NH
NH
O
O
CO2H
Meropenem
CO2H
XHCl
·
NH2
CONH2
1
2
R=
R=
NHMe
Figure
Synlett 2001, No. 11, 26 10 2001. Article Identifier:
1437-2096,E;2001,0,11,1808,1810,ftx,en;Y16801ST.pdf.
© Georg Thieme Verlag Stuttgart · New York
ISSN 0936-5214

LETTER
Stereoselective Synthesis of (3R,5R)-cis-3-Hydroxy-5-phenylpyrrolidine
1809
TBSO
HS
TBSO
TBSO
O
H2N
Ph
Ph
Ph
Ph
N
N
Boc
N
Boc
3
4
5
6
TBSO
HO
O
N3
Cl
CO2Et
Ph
8
7
Scheme 1
OTBS
OTBS
CO2Et
O
i
ii
iii
8
Cl
Cl
NMe(OMe)
9
10
OR
OR
O
O
Cl
N3
v
11 R = TBS
12 R = H
13 R = H
7 R = TBS
vi
iv
Scheme 2 Reagents and conditions: (i) TBSCl, Im, DMF, 89%; (ii) Me(MeO)NH·HCl, i-PrMgBr, THF, –20 °C, 98%; (iii) PhMgCl, THF,
–20 °C to 0 °C, 99%; (iv) CSA, MeOH, 98%; (v) NaN3, DMF, 80 °C, 74%; (vi) TBSCl, Im, DMF, 90%.
the hydroxyl group as TBS ether afforded the desired tempts,12 finally we found that the desired cyclic imine 5
azide 7.
was smoothly generated from chloronitrile 15, readily ob-
tainable from TBS-protection of inexpensive (R)-(+)-4-
chloro-3-hydroxybutyronitrile.13 In this reaction, solvent
effect was critical for controlling both of the Grignard ad-
dition and the cyclization. Non-polar solvent such as
MTBE (metyl t-butyl ether)14 was necessary for the addi-
tion reaction while neither THF nor DME was successful.
Instead, either DME or THF was essential for the comple-
tion of the cyclization. Subsequent hydrogenation under
the above conditions followed by Boc protection in one
flask afforded the desired pyrrolidine 4 in 89% yield15
(Scheme 4).16
With the desired azide in hand, we then tried the reductive
domino cyclization of 7 under usual hydrogenation condi-
tions. Hydrogenation of 7 using Pd/C at atmospheric pres-
sure for 6 hours followed by direct addition of Boc2O
resulted in only the formation of the undesired open-chain
product 14 in 85% overall yield. 14 would be probably
formed via hydrogenolysis of the resulted pyrrolidine ring
at the benzylic position before Boc protection. In contrast,
platinum catalysts completely suppressed formation of
the over-reduction product 14. When Pt/C was used as a
catalyst, the desired cis-pyrrolidine 4 could be stereose-
lectively produced in 73%9 yield in ratio of 22:1
(cis:trans)10 (Scheme 3). The steric bulkiness of the TBS
ether would contribute to the stereoselectivity.11
H2 (balloon)
Pt/C
TBSO
TBSO
BocHN
7
+
EtOH, 16 h
then Boc2O
N
Boc
Domino-cyclization of the azide 7 under the hydrogena-
tion condition led to much success on diastereoselective
construction of the desired pyrrolidine ring. However,
transformation to 7 was tedious because it includes the
deprotection/protection sequence for introduction of the
azide function. Therefore, we re-investigated an alterna-
tive method to generate the cyclic imine. After several at-
4
14
0 %
73 %
cis:trans=22:1
Scheme 3
TBSO
TBSO
TBSO
PhMgBr
MTBE
Pt/C, H2
DME or THF
0 °C to rt
4
Cl
CN
Cl
0 °C
Ph
then Boc2O
89%
N
Ph
N
MgBr
16
15
5
Scheme 4
Synlett 2001, No. 11, 1808–1810 ISSN 0936-5214 © Thieme Stuttgart · New York

1810
K. Maeda et al.
LETTER
(6) (a) Fry, F. D.; Fowler, C. B.; Dieter, R. K. Synlett. 1994, 836;
and references cited therein. (b) Fry, F. D.; Brown, M.;
McDonald, J. C.; Dieter, R. K. Tetrahedron Lett. 1996, 37,
6227.
In this way, we succeeded in the stereoselective synthesis
of (3R,5R)-cis-3-hydroxy-5-phenyl pyrrolidine by highly
selective reduction of the cyclic imine. These methods en-
able us to provide not only practical and large scale syn-
thesis of new carbapenems, 1 and 2 but also a general
method for construction of the cis-2,4-disubstituted pyrro-
lidine skeleton. The whole synthetic work will be de-
scribed elsewhere as a full article.
(7) (a) Hanessian, S.; Frenette, R. Tetrahedron Lett. 1979,
3391. (b) von der Osten, C. H.; Sinskey, A. J.; Barbas, C. F.
III; Pederson, R. L.; Wang, Y.-F.; Wong, C.-H. J. Am. Chem.
Soc. 1989, 111, 3924. (c) Straub, A.; Effenberger, F.;
Fischer, P. J. Org. Chem. 1990, 55, 3926. (d) Liu, K. K.-C.;
Kajimoto, T.; Chen, L.; Zhong, Z.; Ichikawa, Y.; Wong, C.-
H. J. Org. Chem. 1991, 56, 6280. (e) Kajimoto, T.; Chen,
L.; Liu, K. K.-C.; Wong, C.-H. J. Am. Chem. Soc. 1991, 113,
6678. (f) Enders, D.; Jegelka, U. Synlett 1992, 999.
(g) Takaoka, Y.; Kajimoto, T.; Wong, C.-H. J. Org. Chem.
1993, 58, 4809. (h) Henderson, I.; Laslo, K.; Wong, C.-H.
Tetrahedron Lett. 1994, 35, 359.
Acknowledgement
We thank Mr. H. Imamura and Dr. Y. Sugimoto at Banyu Tsukuba
Research Institute for their technical assistance.
(8) Williams, J. M.; Jobson, R. B.; Yasuda, N.; Marchesini, G.;
Dolling, U.-H.; Grabowski, E. J. J. Tetrahedron Lett. 1995,
36, 5461.
(9) 13% and 29% yield of the desired product(4)were obtained
with PtO2 and with Pt/Al2O3, respectively.
(10) The ratio was determined by HPLC analysis.
(11) Hydrogenation of 13 under the same conditions resulted in
poor selectivity (cis:trans = 3.3:1, 70% yield).
(12) Aza-Wittig reaction of 4, aza-Peterson reaction of 17 and
acid promoted cyclic imination of N-Boc-amino ketone 18
were not successful. In most cases, the corresponding
pyrrole 19 caused from elimination of the siloxy group was
produced.
References and Notes
(1) Sunagawa, M.; Matsumura, H.; Inoue, T.; Fukasawa, M.;
Kato, M. J. Antibiotics 1990, 43, 519.
(2) Iso, Y.; Irie, T.; Nishino, Y.; Motokawa, K.; Nishitani, Y. J.
Antibiotics 1996, 49, 199.
(3) (a) Ohtake, N.; Okamoto, O.; Mitomo, R.; Kato, Y.;
Yamamoto, K.; Haga, Y.; Fukatsu, H.; Nakagawa, S. J.
Antibiotics 1997, 50, 598. (b) Nakagawa, S.; Hashizume, T.;
Matsuda, K.; Sanada, M.; Okamoto, O.; Fukatsu, H.;
Tanaka, N. Antimicrob. Agents Chemother. 1993, 37, 2756.
(4) (a) Imamura, H.; Ohtake, N.; Shimizu, A.; Jona, H.; Sato, H.;
Nagano, R.; Ushijima, R.; Yamada, K.; Hashizume, T.;
Morishima, H. Bioorg. Med. Chem. Lett. 2000, 10, 109.
(b) Imamura, H.; Ohtake, N.; Shimizu, A.; Sato, H.;
Sugimoto, Y.; Sakuraba, S.; Kiyonaga, H.; Suzuki-Sato, C.;
Nakano, M.; Nagano, R.; Yamada, K.; Hashizume, T.;
Morishima, H. Bioorg. Med. Chem. Lett. 2000, 10, 115.
(c) Nagano, R.; Shibata, K.; Adachi, Y.; Imamura, H.;
Hashizume, T.; Morishima, H. Antimicrob. Agents
Chemother. 2000, 44, 489. (d) Shibata, K.; Nagano, R.;
Hashizume, T.; Morishima, H. J. Antimicrob. Chemother.
2000, 45, 379.
(5) (a) Imamura, H.; Ohtake, N.; Sakuraba, S.; Shimizu, A.;
Yamada, K.; Morishima, H. Chem. Pharm. Bull. 2000, 48,
310. (b) Imamura, H.; Shimizu, A.; Sato, H.; Sugimoto, Y.;
Sakuraba, S.; Nakano, R.; Yamada, K.; Hashizume, T.;
Morishima, H. J. Antibiotics 2000, 53, 314. (c) Imamura,
H.; Shimizu, A.; Sato, H.; Sugimoto, Y.; Sakuraba, S.;
Nakajima, S.; Abe, S.; Miura, K.; Nishimura, I.; Yamada,
K.; Morishima, H. Tetrahedron 2000, 56, 7705.
TBSO
TBSO
O
BocHN
O
N
TMS
N
H
17
18
19
(13) This approach using bromonitrile has been reported by Fry
et al. See ref. 6.
(14) MTBE is the best solvent of toluene, benzene, hexane and
MTBE.
(15) The trans-isomer was not seen in 1H NMR analysis.
(16) Typical experimental procedure: To a solution of the
chloronitrile 15 (1.80 g, 7.69 mmol) in MTBE (55 mL) at
0 °C was added drop wise phenyl MgBr (3.85 mL of 3.0 M
solution in Et2O, 11.54 mmol) and the mixture was allowed
to reach room temperature. After 15 min, DME (10 mL) was
added drop wise over 5 min followed by addition of EtOH
(6.5 mL). To this reaction mixture was added 3.0 g of 5% Pt/
C in one portion and stirred under H2 atmosphere at room
temperature for 18 h and then Boc2O (3.58 mL, 15.6 mmol)
was added. After 5 h, filtration followed by SiO2 flash
chromatography afforded the desired pyrrolidine 4 (2.58 g,
6.83 mmol) in 89% yield.
(d) Sugimoto, Y.; Imamura, H.; Shimizu, A.; Nakano, M.;
Nakajima, S.; Abe, S.; Yamada, K.; Morishima, H.
Tetrahedron: Asymmetry 2000, 11, 3609.
Synlett 2001, No. 11, 1808–1810 ISSN 0936-5214 © Thieme Stuttgart · New York