Preparation of 4- and 6-desphenyl analogues of (-)-clausenamide and alternative synthesis of (+)-epi-clausenamide

Article abstract of DOI:10.1016/j.cclet.2011.07.002

4- and 6-desphenyl analogues of (-)-clausenamide, 6 and 7, were prepared in optical active form from commercially available d-pyroglutamic acid and the known racemic pyrrolidinone 13, respectively. In order to confirm the absolute stereochemistry of (+)-

Full text of DOI:10.1016/j.cclet.2011.07.002

Available online at www.sciencedirect.com
Chinese Chemical Letters 22 (2011) 1261–1264
www.elsevier.com/locate/cclet
Preparation of 4- and 6-desphenyl analogues of (ꢀ)-clausenamide
and alternative synthesis of (+)-epi-clausenamide
*
Jian Jun Xue, Yu Mei Zhou, Xiao Ming Yu
Institute of Materia Medica, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing 100050, China
Received 31 March 2011
Available online 6 September 2011
Abstract
4- and 6-desphenyl analogues of (ꢀ)-clausenamide, 6 and 7, were prepared in optical active form from commercially available
D-pyroglutamic acid and the known racemic pyrrolidinone 13, respectively. In order to confirm the absolute stereochemistry of (+)-
and (ꢀ)-7, intermediate 19b was transferred into the (+)-epi-clausenamide 8.
# 2011 Xiao Ming Yu. Published by Elsevier B.V. on behalf of Chinese Chemical Society. All rights reserved.
Keywords: (ꢀ)-Clausenamide; Desphenyl analogue; (+)-epi-Clausenamide; Synthesis
(ꢁ)-Clausenamide ((ꢁ)-1, Fig. 1) is a racemic natural product found in the leaves of Clausena lansium (Lour)
skeels [1]. The levo rotatory enantiomer, ((ꢀ)-clausenamide 1, Fig. 1), was a potent nootropics [2,3]. A phase II
clinical trial of this chiral compound as therapeutics for Alzheimer’s disease is to be launched soon.
Along with the identification of several more potent stereoisomers of (ꢀ)-1 [4], we are interested in the substituent
related structure activity relationship (SAR). Six one-substituent modified analogues of (ꢀ)-1 (2–7, Fig. 1) were thus
designed, and compound 2–5 were synthesized [5]. In this letter, we report the preparation of 6 and 7, the 4- and 6-
desphenyl analogues of (ꢀ)-1, in enantiomerically pure form. In addition, partly as a stereochemistry confirmation of
7, we also present an alternative synthesis of (+)-epi-clausenamide 8, a stereoisomer of (ꢀ)-1 that exhibited much
stronger long-term potentiation (LTP) enhancing effect in animal model [4].
As summarized in Scheme 1, easily available D-pyroglutamic acid 9 was selected to be starting material of 6
because of its identical C5 configuration to that of the target molecule. Compound 10 was prepared from 9 in 48%
yield by repeating a known three-step procedure [6]. The following addition of PhMgBr to 10 in THF, however,
afforded an inseparable mixture of diastereomers (11a/11b ꢂ 1/2) [7]. Optimization of this reaction by changing
solvent and lowering the reaction temperature, unfortunately, did not result in significant improvement. In that this
mixture was oxidized with KMnO4 to give ketone 12 in 92.6% yield, with the hope that it could be reduced in a
stereospecific manner. After screening of several commonly used reducing agents including iPrOH/Al(iPrO)3 [8],
NaBH4, LiBHEt3 [9] and CF3COOH/PhMe2SiH [10], we found palladium catalyzed hydrogenation [11] of 12 in THF
gave 11b as the major product (11a/11b, 1/10) in 71.2% isolated yield after recrystallization. Although C6
configuration of this intermediate is opposite to the desired one, it was easily inverted by application of Mitsunobu
* Corresponding author.
E-mail address: mingxyu@imm.ac.cn (X.M. Yu).
1001-8417/$ – see front matter # 2011 Xiao Ming Yu. Published by Elsevier B.V. on behalf of Chinese Chemical Society. All rights reserved.
doi:10.1016/j.cclet.2011.07.002

1262
J.J. Xue et al. / Chinese Chemical Letters 22 (2011) 1261–1264
Fig. 1. (ꢀ)-Clausenamide, its designed derivatives and (+)-epi-clausenamide.
H
H
H
O
Ph
Ph
a
b
Ref. 6
O
O
O
HOOC
O
N
Me
N
N
N
H
9
HO
O
Me
Me
10
12
11a(6S)/11b(6R) c.a. 1/2
OH
H
H
H
d, e
Ph
Ph
Ph
f
c
O
O
O
N
Me
11b
N
Me
11a
N
Me
6
HO
HO
HO
Scheme 1. Synthesis of compound 6. Reagent and conditions: (a) PhMgBr, 83.1%; (b) KMnO4, 92.6%; (c) Pd/C, H2, 71.2%; (d) DEAD/Ph3P/
ClCH2COOH; (e) NaOH, 80.7%, two steps; (f) LDA, Davis oxidant, 52% based on recovered 11a.
reaction [12] and the following hydrolysis to provide 11a in 80.7% yield over two steps. Installation of the 3-a-OH on
the basis of 11a was an uncertain step even when the synthetic plan was drawn. Nevertheless, steric induction of the
C5-b substituent was expected to be harnessed by using bulky oxidants. Compound 11a was thus treated with 3
equivalences of LDA followed by 4 equivalents of Davis oxidant, and 6 was isolated in 52% yield [13,14] based on
54% recovered 11a.
Synthetic route to 7 (Schemes 2 and 3) was, to a large extent, designed on the basis of Hartwig’s total synthesis of
(ꢁ)-clausenamide [2]. The key intermediate, (ꢁ)-16, was prepared from ethyl cinnamate and diethyl
acetamidomalonate in four steps, by following a slightly modified procedure [2] (Scheme 2). Transformation of
14 into 15a was first described by Hartwig and Born [2], in which the starting material was hydrolyzed into the
corresponding monoester and then subjected to pyrolysis at 170 8C to give a 2/1 mixture of 15a and 15b. We found that
this two-step procedure could be replaced by simple treatment of 14 with catalytic LiCl in DMSO under heating,
perfect cis selectivity and without losing yield. Similar modification was also made to the reduction of 15a that less
expensive LiAlH4 was used instead of LiBHEt3.
Scheme 2. Synthesis of compound (ꢁ)-16. Reagent and conditions: (a) EtONa, 76.6%; (b) NaH/MeI, 92.0%; (c) LiCl/DMSO, 170 8C, 65.3%; (d)
LAH, 72.4%.

J.J. Xue et al. / Chinese Chemical Letters 22 (2011) 1261–1264
1263
Ph
Ph
Ph
OH
a
BnO
b
HO
BnO
O
O
N
Me
O
N
Me
N
Me
)-18
17
( )-
16
)-
(
(
OAc
Ph
OH
Ph
OH
Ph
O
Ph
e
d
BnO
HO
BnO
O
O
O
O
N
N
N
Me
( )-18
Me
Me
20
[α]D20 = + 92.2o(c 0.7, chloroform)
−
[α]D = −264.3o(c 0.7, methanol)
( )-7
−
19a
c
+
OAc
Ph
OH
Ph
OH
Ph
O
Ph
e
d
BnO
HO
BnO
O
O
O
O
N
N
N
Me
(+)-18
Me
Me
19b [α]D20 = + 202.1o(c 1.2, chloroform)
(+)-7 [α]D20 = + 267.2o(c 0.4, methanol)
Scheme 3. Optical resolution based synthesis of (+)- and (ꢀ)-7. Reagents and conditions: (a) NaH, BnBr; 75.1%; (b) LDA, O2, P(OEt)3; 64.2%; (c)
EDCI; (ꢀ)-O-acetyl mandelic acid, TEA; 14a, 38.5%; 14b, 40.6%; (d) NaOH; (S)-15, 93.1%; (+)-15, 89.4%; (e) Pd/C; (S)-7, 88%; (+)-7, 94.1%.
Optical resolution of (ꢁ)-16 was intend to form esters with chiral acids including (ꢀ)-menthyloxylacetic acid, (+)-
(S)-O-acetyl mandelic acid and (S)-2-(1, 3-dioxoisoindolin-2-yl) propanoic acid, but this was unsuccessful since all
diastereomeric mixtures showed little separation on chromatography. Therefore, this racemic intermediate was
subjected to hydroxyl protection to give benzyl ether (ꢁ)-17 (Scheme 3), and subsequently oxidized with gaseous
oxygen in the presence of 2 equivalents of LDA to afford (ꢁ)-18 [2]. After an EDCI effected condensation of racemic
18 with (+)-(S)-O-acetyl mandelic acid, separation of the resulting diastereomers appeared to be easy. 19a and 19b
were obtained in pure form in 38.5% and 40.6% yield, respectively. Stepwise hydroxyl deprotection of 19a by basic
hydrolysis and palladium catalyzed hydrogenation afforded (ꢀ)-7 in 81.9% overall yield; likewise treatment of 19b
gave (+)-7 in 84.1% yield [14].
According to Hartwig’s report [2], compound 19a and 19b are conceivably good precursors of (+)- and (ꢀ)-epi-
clausenamide ((+)- and (ꢀ)-8). Since structural data of both enantiomers of 8 is readily available [4], transformation of
19a or 19b into (+)- or (ꢀ)-8 would provide unambiguous evidence for the configuration of the two optical resolution
products and thereby the resulting (+)- and (ꢀ)-7. Moreover, LTP enhancing effect exhibited by (+)-epi-clausenamide
8 is nearly two fold as high as (ꢀ)-clausenamide 1, indicating that this compound is likely a more potent nootropics
[4], but the published synthesis of 8 suffered from the use of harsh chromic acid oxidant and the extremely low yield of
the last two steps. Therefore, the synthetic route starting from 19a or 19b might constitute a better alternative access to
this biologically important isomer of (ꢀ)-clausenamide. Hence, 19b was randomly selected and subjected to the
transformation depicted in Scheme 4. Hydrogenation of 19b successfully delivered 20, of which the 2-O-Ac subunit of
Ph
Ph
H
OH
Ph
O
b, c
a
O
Ph
HO
HO
19b
O
O
N
Me
8
N
Me
20
[ ] 20= + 204.1o(c 0.445, methanol)
α
D
lit. [α]D20= + 201o(c 0.345, methanol) [4]
Scheme 4. Synthesis of (+)-epi-clausenamide 8. Reagents and conditions: (a) Pd/C; 93.1%; (b) Swern oxide; (c) PhMgBr; 59.4% in two steps.

1264
J.J. Xue et al. / Chinese Chemical Letters 22 (2011) 1261–1264
the mandelic acid moiety was removed spontaneously. Swern oxidation of 20 followed by Grignard reaction with
excessive PhMgBr afforded a product that was eventually characterized to be (+)-epi-clausenamide 8, based on a
comparison of NMR spectra and optical rotation with known data. This result clearly indicated on one hand that (ꢀ)-7
was the desired 6-desphenyl analogue of (ꢀ)-clausenamide that has 3S, 4R and 5R configuration, and on the other
hand 19a is a more valuable intermediate than previously reported neo-clausenamidone for the synthesis of 8.
In summary, 4- and 6-desphenyl analogues of (ꢀ)-clausenamide, 6 and 7, were prepared for the first time in
enantiomerically pure form from commercially available D-pyroglutamic acid and the known compound 13,
respectively. In addition, the transformation of 16b into (+)-epi-clausenamide 8 constitutes an unambiguous
configuration confirmation of (+)- and (ꢀ)-7 and an improved synthesis of this biologically interesting compound as
well.
Acknowledgment
This project was financially supported by Natural Science Foundation of China (No. 30672530).
References
[1] M.H. Yang, Y.H. Cao, W.X. Li, et al. Acta Pharm. Sin. 22 (1987) 33.
[2] W. Hartwig, L. Born, J. Org. Chem. 52 (1987) 4352.
[3] J.T. Zhang, X.Y. Jiang, Collection of the Academic Papers of Scientific Annual Meeting of Peking Union Medical College, Beijing Medical
University & Peking Union Medical College Publishing House, Beijing, 1994, p. 356.
[4] Z. Feng, X. Li, G. Zheng, et al. Bioorg. Med. Chem. Lett. 19 (2009) 2112.
[5] J.J. Xue, X.M. Yu, Chin. Chem. Lett. 22 (2011) 761.
[6] P.D. Williams, D.S. Perlow, L.S. Payne, et al. J. Med. Chem. 34 (1991) 887.
[7] The ratio of the two diastereomers was determined by 1H NMR.
[8] J. Yin, M.A. Huffman, K.M. Conrad, et al. J. Org. Chem. 71 (2006) 840.
[9] C. Wiles, P. Watts, S.J. Haswell, Tetrahedron Lett. 47 (2006) 5261.
[10] M. Fujita, T. Hiyama, Org. Synth. 69 (1990) 44.
[11] W. Yang, Y. Wang, J.Y. Roberge, et al. Bioorg. Med. Chem. Lett. 15 (2005) 1225.
[12] J.A. Dodge, J.S. Nissen, et al. Org. Synth. 73 (1996) 110.
[13] The relative stereochemistry of 6 was determined by NOESY. Key NOE correlations were observed between C3-H (d 4.18) and C4-b-H (d
2.28), and between C4-a-H(d 1.55) and C5-H (d 3.75), suggesting a trans geometry between C3 and C5 substituents.
β OH
H
H α
H
O
H
3
4
Ph
HO
5
N
Me
6
20
[14] Compund 6: mp: 218.9–220.1 8C; ½aꢃD þ 52:1 (c 0.2, MeOH); 1H NMR (600 MHz, CD3COCD3): d 1.55 (dt, 1 H, J = 12.6, 8.4), 2.28 (dd, 1 H,
J = 12.6, 8.4), 2.89 (s, 3 H), 3.75 (dd, 1 H, J = 8.4, 1.2), 4.18 (t, 1 H, J = 8.4), 5.10 (d, 1 H, J = 1.2), 7.25–7.46 (m, 5 H); 13C NMR (150 MHz,
CD3COCD3): d 175.6, 142.8, 128.9, 127.9, 127.0, 71.6, 69.4, 63.8, 28.4; FT-IR (cmꢀ1): 3423.6, 3328.8, 1648.6; FAB-MS: 222.1(M + 1);
20
HRMS m/z (M + 1) calcd. for C12H14NO3, 221.1046; found 221.1006. Compound 7: mp: 204.8–205.7 8C; ½aꢃD ꢀ 264:3 (c 0.7, MeOH); 1H
NMR (300 MHz, DMSO-d6 + D2O): d 2.78 (s, 3 H), 2.94 (dd, 1 H, J = 2.7, 11.7), 3.39 (dd, J = 2.4, 11.7), 3.41 (dd, 1 H, J = 8.7, 11.1), 3.64
(ddd, 1 H, J = 2.7, 2.4, 8.7), 4.66 (d, 1 H, J = 11.1), 7.21–7.38 (m, 5 H); 13C NMR (125 MHz, DMSO-d6 + D2O): d 174.3, 137.4, 128.6, 128.3,
126.7, 70.7, 61.2, 57.0, 50.7, 28.2; HRMS m/z (M + 1) calcd. for C12H16NO3, 222.1125; found 222.1125.