Synthetic study of polyoxypeptin: Stereoselective synthesis of (2S,3R)-3-hydroxy-3-methylproline

Article abstract of DOI:10.1016/S0040-4039(01)00867-X

The first synthesis of (2S,3R)-3-hydroxy-3-methylproline, which is a novel amino acid of polyoxypeptin, a potent inducer of apoptosis in human pancreatic carcinoma AsPC-1, was accomplished. The key feature of the synthesis is the palladium-catalyzed intra

Full text of DOI:10.1016/S0040-4039(01)00867-X

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 5253–5256
Synthetic study of polyoxypeptin: stereoselective synthesis of
(2S,3R)-3-hydroxy-3-methylproline
Yasuo Noguchi,† Hiromi Uchiro, Tatsuhiro Yamada and Susumu Kobayashi*
Faculty of Pharmaceutical Sciences, Science University of Tokyo, Ichigaya-Funagawara-machi, Shinjuku-ku,
Tokyo 162-0826, Japan
Received 23 April 2001; accepted 18 May 2001
Abstract—The first synthesis of (2S,3R)-3-hydroxy-3-methylproline, which is a novel amino acid of polyoxypeptin, a potent
inducer of apoptosis in human pancreatic carcinoma AsPC-1, was accomplished. The key feature of the synthesis is the
palladium-catalyzed intramolecular N-allylation of alkenyloxirane to the pyrrolidine ring. © 2001 Elsevier Science Ltd. All rights
reserved.
Polyoxypeptin A (1),1 isolated by Umezawa et al. has
attracted a great deal of attention due to its significant
apoptosis-inducing activity in human pancreatic car-
cinoma AsPC-1. Structurally, polyoxypeptin is a unique
hexadepsipeptide containing novel subunits such as the
acyl side-chain moiety and (2S,3R)-3-hydroxy-3-
methylproline (2) (Scheme 1). Therefore, polyoxypeptin
is considered to be an interesting target molecule from
a synthetic point of view as well as from medicinal
interest.
synthesis of 2 based on palladium-catalyzed cyclization
as the key-step.
Our strategy for the synthesis of the (2S,3R)-3-
hydroxy-3-methylproline (2) is shown in Scheme 2. We
were interested in the possibility of the palladium-cata-
lyzed cyclization of alkenyloxylane 4. There is no prece-
dent for such pyrrolidine formation.3 Alkenyloxirane 4,
in turn, might be prepared in an enantioselective man-
ner from geraniol using asymmetric epoxidation.4
During the course of our investigation towards the total
synthesis of polyoxypeptin, we recently reported the
first stereoselective synthesis of the acyl side-chain seg-
ment.2 Here we wish to report the first stereoselective
Alkenyloxirane 4, a substrate for Pd-catalyzed cycliza-
tion, was synthesized from known aldehyde 6, readily
prepared from geraniol in three steps.5 The aldehyde 6
was oxidized to carboxylic acid 7, and the latter was
Me
OH
NH
N
N
O
HO
O
N
NH
Me
OH
O
O
N
O
HO
Me
Me
N
N
H
O
O
Et
O
CO2H
HO
N
H
i-Pr
O
Me OH
(2S, 3R)-3-Hydroxy-3-methylproline (2)
OH
i-Pr
Polyoxypeptin A (1)
Scheme 1. The structure of polyoxypeptin A and (2S,3R)-3-hydroxy-3-methylproline.
* Corresponding author. Tel.: +81-3-3260-8848; fax: +81-3-3260-8848; e-mail: kobayash@ps.kagu.sut.ac.jp
† Visiting scientist from Sankyo Co., Ltd.
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)00867-X

5254
Y. Noguchi et al. / Tetrahedron Letters 42 (2001) 5253–5256
Me
Asymmetric
Epoxidation
Me
Pd-catalyzed
Cyclization
Me
OH
OH
O
CO2Et
NH
Z
CO2Et
N
Z
CO2H
N
H
4
3
2
Me
Me
Me
OH
Me
OH
NH
geraniol
5
Z
Scheme 2. Retrosynthetic analysis of (2S,3R)-3-hydroxy-3-methylproline (2).
subjected to a modified Curtius rearrangement using
(PhO)2P(O)N3 and benzyl alcohol, followed by
When 4 was treated with a catalytic amount (5 mol%)
of Pd(PPh3)4 in THF at room temperature in the pres-
ence of NaH, no reaction occurred, and 4 was recov-
ered in 60% yield. When the above reaction was carried
out under reflux, a complex mixture of products was
formed. On the other hand, cyclization proceeded
smoothly in THF under reflux without a base affording
a mixture of pyrrolidine derivative 3 and 3% in total 74%
yield (90:10).
6
deacetylation to afford an allylic alcohol 5 in 67% yield
(two steps). The allylic alcohol 5 was next subjected to
Sharpless asymmetric epoxidation4 using
D
-(−)-tartrate
to give epoxyalcohol 8 in 55% yield with 97% ee.7
Epoxyalcohol 8 was oxidized by Swern oxidation, and
the resulting aldehyde was reacted with triethyl phos-
phonoacetate to obtain (E)-alkenyloxirane 4 in 58%
yield (two steps from 8) in stereochemically pure form
(Scheme 3). The key reaction, Pd-catalyzed cyclization
of 4, was next examined (Table 1).
The stereochemistry of the major isomer 3 was tenta-
tively assigned as shown based on the mechanism (Fig.
1) (PhO)2P(O)N3, Et3N
benzene reflux 2h, then
BnOH, benzene, reflux 22h
Me
Me
NaClO4, NaH2PO4
2-methyl-2-butene
ref.5
geraniol
OAc
OAc
2) K2CO3, MeOH, r.t. 1h
67% (two steps)
t-BuOH - H2O, r.t., 2h
CHO
HO2C
96%
6
7
Me
Me
Me
O
1) (COCl)2, DMSO, CH2Cl2
TBHP, Ti(OiPr)4
D-(-)-DET, MS4A
O
-78 , 1h then Et3N, r.t. 1h
OH
OH
NH
Z
NH
2) (EtO)2P(O)CH2CO2Et,NaH
THF, 0 , 1h
CO2Et
NH
Z
CH2Cl2, -23 , 1h
55%
Z
5
4
(97%ee)
58% (two steps)
8
Scheme 3. Synthesis of the alkenyloxirane 4.
Table 1. Pd-catalyzed cyclization
Me
Me
Me
OH
3'
OH
Pd (0)
O
+
Conditions
CO2Et
CO2Et
NH
Z
N
Z
CO2Et
N
Z
4
3
Entry
Pd Cat.
Base
Conditions
THF, rt, 24 h
THF, reflux, 5 h
DMF, rt, 24 h
Yield
N.R.a
Ratio 3:3%
1
2
3
4
5
5 mol% Pd(PPh3)4
5 mol% Pd(PPh3)4
5 mol% Pd(PPh3)4
5 mol% Pd(PPh3)4
None
NaH
NaH
NaH
None
None
–
–
–
Complex mixture
Complex mixture
74%
THF, reflux, 5 h
THF, reflux, 24 h
90:10
–
N.R.b
a 4 was recovered in 60% yield.
b 4 was recovered in quantitative yield.

Y. Noguchi et al. / Tetrahedron Letters 42 (2001) 5253–5256
5255
Me
Me
O-
Me
O
OH
Pd (0)
Pd+
CO2Et
CO2Et
CO2Et
NH
Z
N
Z
NH
Z
3
4
A
Pd(0)
Me
OH
Me
O-
Pd+
CO2Et
N
Z
CO2Et
NH
Z
3'
B
Figure 1. Mechanism of the Pd-catalyzed cyclization.
Me
Me
Me
OH
OH
OH
CHO
O3, MeOH, -78
+
then Me2S, 0
CO2Et
CHO
9 (82%)
N
Z
N
Z
N
Z
3 (90 : 10 mixture)
9' (9%)
Me
OH
Me
NaClO4, NaH2PO4
2-methyl-2-butene
tBuOH - H2O
OH
H2(1atm), Pd-C
9
MeOH, r.t., 24h
quant
26
CO2H
N
α
[
]
D = -40.2 (c 0.42, H2O)
D = -41 (c 0.4, H2O)
CO2H
N
H
r.t., 1.5h
88%
18
Z
[α
]
10
lit:
2
Scheme 4. Synthesis of (2S,3R)-3-hydroxy-3-methylproline (2).
ture of the present synthesis is that the stereochemistry
at C-2 and C-3 was controlled by Pd-catalyzed cycliza-
tion of alkenyloxirane without a base. Preparation of
other segments and the coupling toward the total syn-
thesis of polyoxypeptin are now in progress.
1), and was unambiguously confirmed by successful
correlation to 2. The isomeric 3% might be derived from
p-allyl palladium B formed by an intermolecular SN2-
type attack of Pd(0) to p-allyl palladium A. The forma-
tion of 3% is also possible by the direct epoxide opening
by 5-endo mode. However, this seems unlikely because
no reaction occurred without Pd(PPh3)4 in refluxing
THF.
References
Transformation of 3 into (2S,3R)-3-hydroxy-3-methyl-
proline (2) is summarized in Scheme 4. Cleavage of the
double bond using O3 oxidation gave the desired alde-
hyde 9 in 82% yield. The C2-epimer 9%, which was
derived from 3%, was separated from 9 by silica gel
column chromatography (9% yield). The aldehyde 9
was then oxidized with NaClO2, and finally deprotec-
tion of the benzyloxycarbonyl group with H2–Pd/C
completed the first synthesis of 2.8 Spectral data as well
as the optical rotation value of the synthetic 2 were in
good accordance with those of the natural (2S,3R)-3-
hydroxy-3-methylproline reported by Umezawa’s
group.1b
1. (a) Umezawa, K.; Nakazawa, K.; Uemura, T.; Ikeda, Y.;
Kondo, S.; Naganawa, H.; Kinoshita, N.; Hashizume, H.;
Hamada, M.; Takeuchi, T.; Ohba, S. Tetrahedron Lett.
1998, 39, 1389–1392; (b) Umezawa, K.; Nakazawa, K.;
Ikeda, Y.; Naganawa, H.; Kondo, S. J. Org. Chem. 1999,
64, 3034–3038.
2. (a) Noguchi, Y.; Yamada, T.; Uchiro, H.; Kobayashi, S.
Tetrahedron Lett. 2000, 41, 7493–7497; (b) Noguchi, Y.;
Yamada, T.; Uchiro, H.; Kobayashi, S. Tetrahedron Lett.
2000, 41, 7499–7502; (c) Recently, synthesis of the acyl
side-chain segment of polyoxypeptin was reported: Lorca,
M.; Kurosu, M. Tetrahedron Lett. 2001, 42, 2431–2434.
3. (a) Similar Pd-catalyzed intramolecular N-allylation of
carbamate, see: Takao, K.; Nigawara, Y.; Nishino, E.;
Takagi, I.; Maeda, K.; Tadano, K.; Ogawa, S. Tetrahedron
In conclusion we were able to achieve the first synthesis
of (2S,3R)-3-hydroxy-3-methylproline 2. The key fea-

5256
Y. Noguchi et al. / Tetrahedron Letters 42 (2001) 5253–5256
1994, 50, 5681–5704 and references cited therein; (b) Pd-
sion to (+)-MTPA ester by chiral HPLC analysis [column,
Chiralcel AD (4.6f×250 mm); eluent, 20:1 n-hexane:EtOH
mixture; flow rate, 0.8 ml/min; temperature 35°C; wave-
length, 254 nm].
catalyzed O-allylation of alkenyloxirane, see: Suzuki, T.;
Sato, O.; Hirama, M. Tetrahedron Lett. 1990, 31, 4747–
4750.
1
4. Katsuki, T.; Sharpless, K. B. J. Am. Chem. Soc. 1980, 102,
5974–5976.
5. Tago, K.; Arai, M.; Kogen, H. J. Chem. Soc., Perkin
Trans. 1 2000, 2073–2078.
6. Ninomiya, K.; Shioiri, T.; Yamada, S. Tetrahedron 1974,
30, 2151–2157.
7. The enantiomeric excess of 8 was determined after conver-
8. Data for 2: mp 198–201°C. [h]2D6 −40.2 (c 0.42, H2O). H
NMR (D2O, 500 MHz): l 1.60 (s, 3H), 2.13–2.16 (m, 2H),
3.45 (ddd, J=11.6, 7.0 and 4.6 Hz, 1H), 3.55 (ddd, J=
11.6, 10.7 and 7.9 Hz, 1H), 3.86 (s, 1H). 13C NMR (D2O,
125 MHz): l 171.21, 78.83, 70.09, 43.70, 39.84, 24.26.
FAB-MS (pos.); m/z 146 (M+H)+. FAB-HRMS (pos.)
calcd for C6H12NO3 (M+H)+ 146.0817, found 146.0815.
.