A convenient synthesis of 2β,3α-dihydroxyurs-12-en-28-oic acid as a natural diastereoisomer of corosolic acid

Article abstract of DOI:10.1016/j.cclet.2013.04.022

2β,3α-Dihydroxyurs-12-en-28-oic acid (6) is a naturally occurring diastereoisomer of corosolic acid with glycogen phosphorylase inhibitory activity. A new strategy for the semi-synthesis of 6 was developed. Using the commercially available ursolic acid (1) as the starting materials, 6 was synthesized through five facile reactions with a high stereoselectivity and an overall yield of 47.3%. The structure of 6 was confirmed by optical rotation, ESI-MS, ¹H NMR and ¹³C NMR data.

Full text of DOI:10.1016/j.cclet.2013.04.022

Chinese Chemical Letters 24 (2013) 617–618
Contents lists available at SciVerse ScienceDirect
Chinese Chemical Letters
journal homepage: www.elsevier.com/locate/cclet
Original article
A convenient synthesis of 2
b,3a-dihydroxyurs-12-en-28-oic acid as a natural
diastereoisomer of corosolic acid
*
Shi-Sheng Wang , Li-Juan Luo, Zhi-Gang Gao, Tong-Wei Guan, Wei-Jie Zhao
School of Pharmaceutical Science and Technology, Dalian University of Technology, Dalian 116023, China
A R T I C L E I N F O
A B S T R A C T
,3 -Dihydroxyurs-12-en-28-oic acid (6) is a naturally occurring diastereoisomer of corosolic acid
Article history:
2
b a
Received 10 January 2013
Received in revised form 25 March 2013
Accepted 28 March 2013
Available online 18 May 2013
with glycogen phosphorylase inhibitory activity. A new strategy for the semi-synthesis of 6 was
developed. Using the commercially available ursolic acid (1) as the starting materials, 6 was synthesized
through five facile reactions with a high stereoselectivity and an overall yield of 47.3%. The structure of 6
was confirmed by optical rotation, ESI-MS, 1H NMR and 13C NMR data.
ß 2013 Shi-Sheng Wang. Published by Elsevier B.V. on behalf of Chinese Chemical Society. All rights
reserved.
Keywords:
2b,3a-Dihydroxyurs-12-en-28-oic acid
Ursolic acid
Glycogen phosphorylase
Corosolic acid
1. Introduction
for the synthesis of 6 starting from ursolic acid in five steps and a
47.3% overall yield.
2
b a-Dihydroxyurs-12-en-28-oic acid (6), a naturally occur-
,3
ring pentacyclic triterpene, was first isolated from Lagerstroemia
floribunda (Lythraceae) [1]. Compound 6 is the 2 ,3 -isomer of
2. Experimental
b
a
corosolic acid that has recently attracted much attention due to its
various biological properties, such as glucose lowering effect [2],
anti-inflammatory [3], anti-tumor [4], antiviral [5] and cardiovas-
cular activities [6]. Corosolic acid is known as a ‘plant insulin’ for its
remarkable ability in reducing blood glucose level post glucose
challenge, which can be probably attributed to its glycogen
phosphorylase (GP) inhibitory activity [7–9]. Compound 6 was
Chemicals for the synthesis were analytical grade. Ursolic acid
(1) was purchased from Shanxi Yongjian Pharmaceutical Co., Ltd.
The reference compound, corosolic acid (2a,3b-dihydroxyurs-12-
en-28-oic acid), was purchased from Shanghai Yuanye Technology
Co., Ltd. The purity of the compounds was checked by thin layer
chromatography and high performance liquid chromatography
(HPLC) analyses. The known compounds were characterized by
comparing their spectral data with those reported in the literature.
As shown in Scheme 1, ursolic acid was reacted with benzyl
chloride to protect the 28-COOH as a benzyl ester. Then the
dehydration product 3 [15] was obtained by treating alcohol 2 with
methanesulfonyl chloride. Compound 3 was mixed with m-
chloroperoxybenzoic acid (mCPBA) in NaHCO3/CH2Cl2 to produce
reported to be a more potent GP inhibitor (IC50 1.1
corosolic acid (IC50 20 mol/L), which means that the configura-
tion of 2,3-dihydroxy at the ring A is a critical factor for GP
inhibitory activity and 2 ,3 -configuration is more favorable than
the 2 ,3 -configuration [10,11].
mmol/L) than
m
b
a
a
b
Unlike oleanolic acid or ursolic acid, 6 and corosolic acid have
very limited distribution and low abundance in nature. Thus, their
semi-syntheses from available pentacyclic triterpenes have been
investigated [12,13]. For the synthesis of 6, Sun et al. reported a
2a,3a-epoxyurs-12-en-28-oic acid benzyl ester (4) [16]. 2b,3a-
Dihydroxyurs-12-en-28-oic acid benzyl ester (5) [17] was
obtained by a ring-cleaving reaction of 4 with THF/H2O (3/1, (v/
v)) as the solvent and HClO4 as the catalyst. Thereafter, the final
product 6 was obtained as a white powder after deprotection in
five-step semi-synthetic method starting from the 2b-isomer of
corosolic acid [11], which was not commercially available and
synthesized from ursolic acid in six steps [10] in an overall yield of
8.8%. Herein, we present a more practical and convenient method
96.2% yield. ½a 2D2
ꢀ
+62.9 (c 0.062, pyridine). ESI-MS (m/z): 471.4
[MꢁH]ꢁ. 1H NMR (400 MHz, pyridine-d5):
d 0.94, 1.06, 1.17, 1.21 (s,
each 3H), 0.99 (d, 3H, J = 4.8 Hz), 1.28 (s, 6H), 2.63 (d, 1H,
J = 11.0 Hz), 3.98 (d, 1H, J = 6.7 Hz), 4.34 (dd, 1H, J = 6.4, 13.2 Hz),
5.49 (s, 1H). 13C NMR (100 MHz, pyridine-d5):
* Corresponding author.
E-mail address: wangss@dlut.edu.cn (S.-S. Wang).
d 18.6, 18.7, 20.9,
1001-8417/$ – see front matter ß 2013 Shi-Sheng Wang. Published by Elsevier B.V. on behalf of Chinese Chemical Society. All rights reserved.
http://dx.doi.org/10.1016/j.cclet.2013.04.022

[
S.-S. Wang et al. / Chinese Chemical Letters 24 (2013) 617–618
COOH
b
COOBn
COOBn
c
a
HO
HO
1
2
3
e
COOBn
COOBn
COOH
d
HO
HO
HO
HO
O
6
4
5
Scheme 1. Synthesis of 2b,3a-dihydroxyurs-12-en-28-oic acid. (a) BnCl, DMF, 50 8C, 93.7%; (b) methanesulfonyl chloride, DMAP, pyridine, reflux, 89.1%; (c) NaHCO3, mCPBA,
CH2Cl2, 71.5%; (d) HClO4, THF, H2O, r.t., 82.4%; (e) 10% Pd/C, H2, THF, 96.2%.
21.1, 22.6, 24.7, 25.0, 25.1, 26.2, 28.4, 29.8, 32.3, 34.5, 38.7, 38.8,
39.1, 40.6, 40.7, 41.4, 44.0, 47.1, 49.3, 49.8, 52.0, 54.9, 71.9, 79.5,
127.1, 140.4, 181.2.
Acknowledgment
Financial support from the Fundamental Research Funds for the
Central Universities (No. DUT12JB08) is greatly appreciated.
3. Results and discussion
References
Using the synthetic route illustrated in Scheme 1, the target
product 6 was successfully synthesized in a total yield of 47.3%. The
crucial steps of the whole route included the epoxidation of alkene
3 to produce epoxide 4 and the subsequent stereoselective
hydrolysis to yield diol 5. In our research, several oxidants were
screened to transform the 2,3-olefinic bonds in 3 to an epoxide,
such as hydrogenperoxide, tert-butyl hydrogenperoxide, performic
acid and mCPBA. The reaction with mCPBA gave the best results
with the fewest by-products, possibly derived from the epoxida-
tion of 11,12-olefinic bond in 3. The epoxide 4 was obtained using
[1] Y. Zhao, H. Dou, R.P. Zhang, et al., Constituents of three species of Lagerstroemia,
Biochem. Syst. Ecol. 33 (2005) 639–642.
[2] W. Zong, W.S. Xia, B.L. Cui, J.Z. Wan, Screening of hypoglycemic active constitu-
ents of Lagerstroemia speciosaleaves, Shipin Yu Shengwu Jishu Xuebao 25 (2006)
67–71.
[3] B. Norihiro, A. Toshihiro, T. Harukuni, et al., Triterpene acids from the leaves of
Perilla frutescens and their anti-inflammatory and antitumor-promoting effects,
Biosci. Biotechnol. Biochem. 68 (2004) 85–90.
[4] C. Murakami, K. Myoga, R. Kasai, et al., Screening of plant constituents for effect on
glucosetransport activity in Enrlich ascites tumour cells, Chem. Pharm. Bull. 41
(1993) 2129–2231.
[5] S.K. Gupta, M. Modi, Nutan, et al., Plant extract comprising corosolic acid with
anti-human immunodeficiency virus (HIV) activity and process of preparation
thereof, Indian Pat. Appl. 2010DE00242A.
[6] L.S. Chen, Y. Lu, Y. Xin, S.W. Xu, Preparation of natural high-purity corosolic acid
and maslinic acid from Eriobotrya japonica leaf, CN 2008 101240008A.
[7] C.P. Hu, L.S. Chen, Y. Xin, Q.X. Cai, Determination of corosolic acid in Eriobotrya
japonica leaves by reversed-phase high performance liquid chromatography,
Chin. J. Chromatogr. 24 (2006) 492–494.
[8] M. Fukushima, F. Matsuyama, N. Ueda, et al., Effect of corosolic acid on post-
challenge plasma glucose levels, Diabetes Res. Clin. Pract. 73 (2006) 174–177.
[9] H.B. Sun, X.A. Wen, J. Xia, et al., Pentacyclic triterpenes. Part 5: synthesis and SAR
study of corosolic acid derivatives as inhibitors of glycogen phosphorylases,
Bioorg. Med. Chem. Lett. 17 (2007) 5777–5782.
[10] G.O. Nikos, X.A. Wen, H.B. Sun, et al., Naturally occurring pentacyclic triterpenes
as inhibitors of glycogen phosphorylase: synthesis, structure activity relation-
ships, and X-ray crystallographic studies, J. Med. Chem. 51 (2008) 3540–3554.
[11] H.B. Sun, K.G. Cheng, P. Zhang, K.G. Cheng, J. Xie, Practical synthesis of brede-
molic acid, a natural inhibitor of glycogen phosphorylase, J. Nat. Prod. 71 (2008)
1877–1880.
mCPBA and its relative configuration was deduced to be 2a,3a
according to the configuration of the final product 6. The
stereoselectivity of the epoxidation can be explained by the fact
that top face was blocked by two axial methyl groups [14]. In the
step of the epoxide opening, reaction solvent with a proper
dielectric constant and solubilizing ability was thought to be a
pivotal factor for the success of the reaction. After screening several
solvents, THF was considered to be optimal and the highly
stereoselective ring-opening of epoxide to 2
achieved with HClO4 as an acidic catalyst. Due to the SN2 nature of
the reaction, the 2 ,3 -diol 5 was preferentially obtained.
b,3a-dihydroxyl was
b
a
The structure and stereochemistry of 6 were elucidated by
comparing its optical rotation, 1H NMR, 13C NMR, MS data with
those reported in the literature [11]. According to the spectral data
reported in the literatures [10,11], the configurations at 2,3-
positions can be established by the coupling constants (J values) of
[12] Z.Y. Wang, S.H. Wang, Synthesis of corosolic acid, CN 2008 101805389A.
[13] H.B. Sun, X.A. Wen, P.Z. Ni, Method for preparation of corosolic acid and maslinic
acid, CN 20051634971A.
[14] Y.M. Zhang, J. Guo, X.R. He, T. Xiong, Z.T. Li, Improved synthesis of 2a,3a-epoxy-
16a-bromo-5a-androstan-17-one, an intermediate of rocuronium bromide, Chin.
J. Med. Chem. 18 (2008) 61–63.
the H-3 signal in 1H NMR, which was ꢂ9.4 Hz for 2
a
,3
b
, ꢂ2.6 Hz
for 2
a,3
a
, ꢂ3.6 Hz for 2
b
,3
b
and ꢂ7.5 Hz for 2
b,3a, respectively.
The optical rotation, MS, 1H NMR and 13C NMR of 6 were consistent
[15] Spectra data of compound 3 (urs-2,12-dien-28-oic acid benzyl ester): ESI-MS: m/z
529.3 [M+H]+. 1H NMR (400 MHz, CDCl3): d 0.71, 1.00, 1.08, 1.10, 1.25 (s, each 3H),
0.86 (d, 3H, J = 5.6 Hz), 0.94 (d, 3H, J = 6.90 Hz), 4.99 and 5.09 (d, each 1H,
J = 12.5 Hz), 5.26 (s, 1H), 5.37 and 5.41 (d, each 1H, J = 10.2 Hz), 7.33 (m, 5H).
[16] Spectra data of compound 4 (2a,3a-epoxyurs-12-en-28-oic acid benzyl ester):
ESI-MS: m/z 545.4 [M+H]+. 1H NMR (400 MHz, CDCl3): d 0.61, 1.00, 1.08, 1.10, 1.25
(s, each 3H), 0.86 (d, 3H, J = 6.4 Hz), 0.92 (d, 3H, J = 5.6 Hz), 2.78 (d, 1H, J = 3.5 Hz),
3.18 (1H, m), 4.99 and 5.08 (d, each 1H, J = 12.5 Hz), 5.25 (s, 1H), 7.33 (m, 5H).
with those of 2b,3a-dihydroxyurs-12-en-28-oic acid.
4. Conclusion
We designed a practical route for the synthesis of 2b,3a-
dihydroxyurs-12-en-28-oic acid in five steps under mild condi-
tions in high yields and with low cost. The overall yield of the route
was much higher than those previously reported. The structure
and configuration of the final product were clarified by various
spectral techniques.
[17] Spectra data of compound
5 (2b,3a-dihydroxyurs-12-en-28-oic acid benzyl
ester): ESI-MS: m/z 563.4 [M+H]+. 1H NMR (400 MHz, CDCl3): d 0.62, 0.99,
1.06, 1.25, 1.27 (s, each 3H), 0.84 (d, 3H, J = 6.0 Hz), 0.92 (d, 3H, J = 5.6 Hz),
3.63 (d, 1H, J = 10.3 Hz), 3.71 (m, 1H), 4.99 and 5.08 (d, each 1H, J = 12.5 Hz), 5.25
(s, 1H), 7.33 (m, 5H).