Synthesis and activity of oleanolic acid derivatives, a novel class of inhibitors of osteoclast formation

Article abstract of DOI:10.1016/j.bmcl.2005.01.061

Two series of oleanolic acid derivatives were synthesized and their inhibitory activity on the formation of osteoclast-like multinucleated cells (OCLs) induced by 1α,25-dihydroxy vitamin D₃ was evaluated in a co-culture assay system. The structure-activity relationships, together with electronic structure based on the frontier molecular orbitals, for example, HOMO and LUMO, related to different amino acid substituents were studied. Derivatives with proline or phenylalanine showed a tendency to enhance the inhibitory activity.

Full text of DOI:10.1016/j.bmcl.2005.01.061

Bioorganic & Medicinal Chemistry Letters 15 (2005) 1629–1632
Synthesis and activity of oleanolic acid derivatives, a novel class
of inhibitors of osteoclast formation
a
b
c
Yuan Zhang,a Jian-xin Li,a,b, Jianwei Zhao, Shao-zhong Wang, Yi Pan,
*
Ken Tanakad and Shigetoshi Kadotae
aKey Lab of Analytical Chemistry for Life Science, School of Chemistry and Chemical Engineering,
Nanjing University 210093, PR China
bInstitute of Medicinal Chemistry, Nanjing University 210093, PR China
cDepartment of Chemistry, School of Chemistry and Chemical Engineering, Nanjing University 210093, PR China
dNational Police Agency, 2-1-2 Kasumigaseki, Chiyoda-Ku, Tokyo 100-8974, Japan
eInstitute of Natural Medicine, Toyama Medical and Pharmaceutical University, Toyama 930-0194, Japan
Received 1 December 2004; revised 24 January 2005; accepted 25 January 2005
Abstract—Two series of oleanolic acid derivatives were synthesized and their inhibitory activity on the formation of osteoclast-like
multinucleated cells (OCLs) induced by 1a,25-dihydroxy vitamin D3 was evaluated in a co-culture assay system. The structure–activ-
ity relationships, together with electronic structure based on the frontier molecular orbitals, for example, HOMO and LUMO,
related to different amino acid substituents were studied. Derivatives with proline or phenylalanine showed a tendency to enhance
the inhibitory activity.
Ó 2005 Elsevier Ltd. All rights reserved.
Osteoporosis is a disease characterized by low bone min-
eral density and structural deterioration of bone tissue,
which leads to bone fragility and increased susceptibility
to fractures.1 The disease is widely recognized as a major
public health problem, particularly among postmeno-
pausal women. The cause of the disease is thought to
stem from an imbalance of the bone remodeling process
with bone resorption exceeding bone formation. Since
potentiated bone resorption by osteoclasts, the primary
bone-resorption cells, is one of the major causes of osteo-
porosis, osteoclasts are one of the targets of chemother-
apy for osteoporosis. Inhibitors of osteoclasts may
therefore present useful agents to prevent the excessive
bone resorption associated with osteoporosis.2
those extracts, a methanol extract from the root of
Achyranthes bidentata Blume (Amaranthaceae) showed
the most potent activity. In vivo studies showed that
the butanol soluble fraction of the methanol extract pre-
vented the bone loss characteristic of osteoporosis in
OVX rats effectively.4 Further chemical isolation and
structure elucidation associated with in vitro assay indi-
cated that the active components are oleanolic acid type
of triterpenoids. We also discovered that oleanolic acid,
the genin of these triterpenoids, have inhibitory effect of
bone resorption stimulated by PTH.
Oleanolic acid (1), a natural product, has been reported
to have numerous pharmacological activities including
anti-inflammatory,5 anti-cancer,6 anti-HIV,7 and hep-
ato-protective effects.8 Many derivatives of oleanolic
acid have been synthesized for a variety of other biolog-
ical activities.9–12 It was also reported that amino acid
conjugates of natural product could produce effective
antitumor agents13 and modification of ring C of olean-
olic acid could increase the inhibitory potency of nitric
oxide production.11 These reports, along with our own
preliminary screening results of oleanolic acids as poten-
tial anti-osteoporosis lead compounds prompted us to
embark on an investigation to modify the structure to
improve the inhibitory activity on osteoclast formation.
In searching of biologically active substances from top-
ical traditional herbal drugs, we have screened a number
of crude drug extracts for their inhibitory activity on
bone resorption that stimulated by parathyroid hor-
mone (PTH) in a bone tissue culture system.3 Among
Keywords: Oleanolic acid derivatives; Inhibitor; Osteoclast formation;
Frontier molecular orbitals.
*
Corresponding author. Tel.: +86 25 83686419; fax: +86 25
83317761; e-mail: lijxnju@nju.edu.cn
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.01.061

1630
Y. Zhang et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1629–1632
To explore the effects of the amino acid substituents and
other functional groups on anti-osteoporosis agents, we
converted C11-methylene to a carbonyl group and C17-
carboxyl to an amide using several natural amino acids.
In this paper, we describe the synthesis, inhibitory activ-
ity of osteoclast formation and structure–activity rela-
tionships for these compounds.
(3) with oxalyl chloride in dry CH2Cl2. Treatment of
compound 5 with corresponding amino acid methyl es-
ters16 in the presence of triethylamine in CH2Cl2 affor-
ded amides 3a–g. Hydrolysis of the acetyl group and
the methyl esters of compounds 3a–g using aqueous so-
dium hydroxide yielded compounds 1a–g, respectively.
Compounds 2a–g were synthesized in the same protocol
as 1a–g from 3-acetyl-11-oxooleanolic acid (4) (Fig. 1).
The preparation of compound 2 was outlined in Scheme
1. 3-Acetyloleanolic acid (3) was synthesized by stirring
oleanolic acid with acetic anhydride in dry pyridine in
the presence of 4-dimethylaminopyridine (DMAP). On
oxidation of compound 3 with excess of tert-butyl chro-
mate in dry carbon tetrachloride at 65 °C overnight
afforded 3-acetyl-11-oxooleanolic acid (4) in 53%
yield.14 Deprotection of the hydroxyl group of 4 using
potassium carbonate in methanol at 45–60 °C for 19 h
gave 11-oxooleanolic acid (2).15
Two series of compounds 1, 1a–g and 2, 2a–g (Fig. 1)
were tested for their inhibitory activity on 1a,25-di-
hydroxy vitamin D3 [1a,25(OH)2D3] induced TRAP-
positive osteoclast-like multinucleated cells (OCLs)
formation in a co-culture assay system with mouse bone
marrow cells and osteoblast-like cells reported by
Takahashi et al.17 Elcatonin, a clinically available anti-
osteoporosis drug was used as a positive control.18 As
shown in Table 1, all of the tested compounds except
for 1b, 2a, 2b, 2d retained inhibitory effects on OCLs
formation. Introducing proline and phenylalanine at
C28-position of lead compound (1) displayed an increase
tendency in inhibitory effects compared with 1. How-
A general synthesis for compounds 1a–g were conducted
according to the following procedures (Scheme 2). Com-
pound 5 was prepared by stirring 3-acetyloleanolic acid
Scheme 1. Reagents and conditions: (a) Ac2O/DMAP/pyridine, rt, 99%; (b) (tert-BuO)2CrO2/CCl4, 53%; (c) K2CO3/MeOH, 95%.
Scheme 2. Reagents and conditions: (a) ClCOCOCl/CH2Cl2; (b) SOCl2/MeOH, 83–86%; (c) Et3N/CH2Cl2, 90–93%; (d) 4 M NaOH, MeOH/THF,
95–98%.

Y. Zhang et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1629–1632
1631
Figure 1.
Table 1. Effects of tested compounds on 1a,25(OH)2D3 induced
TRAP(+)-OCLs formation
of carcinogenic polycyclic aromatic hydrocarbons
(PAHs) and derived a simple rule to predict bioactivity
of PAHs.19 This methodology was based on the energy
separation values between frontier molecular orbitals
and in their relative contribution to the local density
of electronic states over specific molecular regions. In
order to further explore their structure–activity relation-
ships, a preliminary work on semi-empirical and ab ini-
tio calculations, which concerns the locations and the
relative energies of the frontier molecular orbitals,
namely HOMO (the highest occupied molecular orbital)
and LUMO (the lowest unoccupied molecular orbital)
on all of the synthesized molecules, was taken at the
semi-empirical parametric method 3 (PM3) and HF/3-
21* levels, respectively. The final RMS gradient less than
0.01 kcal/mol was achieved for all fully optimized mole-
cules. The results indicated that the geometrical struc-
tures of derivatives of 1 were similar to those of
corresponding ones of 2, however, surprisedly, a big dif-
ference on the locations and the relative energies of
HOMO and LUMO (EHOMO and ELUMO) was ob-
served. For the compound 1 and its derivatives, HOMO
mainly located on ring C and partially on ring D, and
LUMO on the double bond and tended to move to
the amino acid substituent if used. On the contrary,
for the compound 2 and derivatives HOMO mainly lo-
cated on the double bond and amino acid substituent,
and LUMO on ring C. A graphitic comparison of
HOMO and LUMO of representative molecules, 1f
and 2f, was given in Figure 2. For the EHOMO and
ELUMO, each derivative of 1 possessed higher level than
each corresponding one of 2 (Table 2). In addition, in 1
and its derivatives, 1, 1f, and 1g with more potent activ-
ity showed higher EHOMO, which accounts for the elec-
tron donor ability, the same tendency was observed in
2 and its derivatives except for 2c and 2d, which will
Compd
OCLs (%)
Compd
OCLs (%)
Control
Elcatonin
100.0 16.2#
29.0 9.1**
27.6 8.5**
61.5 12.0*
87.8 2.9
41.3 9.5*
53.5 7.9*
51.2 10.7*
16.3 7.9**
21.0 4.7**
Normal
2
20.0 10.0
38.5 3.8**
83.3 11.7
76.3 20.0
59.0 12.8*
62.0 15.1
69.5 8.3*
36.5 14.4**
40.3 4.9**
1
1a
1b
1c
1d
1e
1f
1g
2a
2b
2c
2d
2e
2f
2g
Control: cultured with 1a,25(OH)2D3 (10À8 M). Elcatonin: cultured
with 1a,25(OH)2D3 (10À8 M) and elcatonin (2 U/mL). Normal: cul-
tured without any additions. Samples: cultured with 1a,25(OH)2D3
(10À8 M) and each compound (20 lM). Each value was expressed as
mean SD, n = 4. The data of control group was pegged as 100%
(OLC number was 104), while other data were calculated relative to it.
Data were analyzed by studentÕs t test. Significant differences in
TRAP(+)-OCLs compared with control group, *p < 0.05, **p < 0.01,
#p < 0.05 versus normal group.
ever, other amino acid derivatives showed less activity.
The alanine derivatives of both 1 and 2 were inactive.
The skeleton modification (C11-methylene of to a car-
bonyl group) of the lead compound (1) led to a decrease
in activity compared with 1. Furthermore, all of the
derivatives of 2 were less active in comparison with
the parent compound and the corresponding derivatives
of 1. These results suggested that the C11-methylene
group of lead compound is important for inhibitory
potency.
Recently, a theoretical approach has been used to corre-
late electronic indices to the biological activity of a class

1632
Y. Zhang et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1629–1632
In summary, oleanolic acid derivatives 1f and 1g showed
a high inhibitory activity on formation of OCLs. Com-
pound 1f was the most potent with 16.3% inhibition of
OCL formation relative to the control (100%). Proline
and phenylalanine substituents of oleanolic acid showed
a tendency to enhance the activity, other amino acids
that used in the experiments had no benefit on improve-
ment of the activity. Removal of the hydrogen atoms on
C11 gave a big impact on the electronic structure and
could be a major cause of decrease of anti-OCLs forma-
tion activity. The locations and the relative energies of
HOMO and LUMO might closely relate to the activity.
Further syntheses of derivatives, studies on bioactivities
and analyses of the frontier molecular orbitals HOMO
and LUMO are in progress and will be reported in
due course.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.bmcl.
2005.01.061.
References and notes
1. Crandall, C. Postgrad. Med. 2003, 114, 21.
2. Farina, C.; Gagliardi, S. Curr. Pharm. Des. 2002, 8,
2033.
3. Li, J. X.; Kadota, S.; Li, H. Y.; Miyahara, T.; Namba, T.
J. Trad. Med. 1996, 13, 50.
4. Li, J. X.; Li, H. Y.; Miyahara, T.; Tezuka, Y.; Kadota, S.;
Tonami, S.; Seto, H.; Namba, T. J. Trad. Med. 1997, 14,
366.
Figure 2. Comparison of the frontier molecular orbitals HOMO and
LUMO of compounds 1f and 2f.
´ ˇ
5. Giner-Larza, E. M.; Manez, S.; Recio, M. C.; Giner, R.
Table 2. Energies (eV) of the frontier molecular orbitals for the OA
derivatives
´
M.; Prieto, J. M.; Cerda-Nicolas, M.; Rıos, J. L. Eur. J.
Pharmacol. 2001, 428, 137.
´
´
Compd EHOMO
ELUMO Compd EHOMO
ELUMO
6. Hsu, H. Y.; Yang, J. J.; Lin, C. C. Can. Lett. 1997, 111, 7.
7. Zhu, Y. M.; Shen, J. K.; Wang, H. K.; Cosentino, L. M.;
Lee, K. H. Bioorg. Med. Chem. Lett. 2001, 11, 3115.
8. Joeng, H. G. Toxicol. Lett. 1999, 105, 215.
9. Assefa, H.; Nimrod, A.; Walker, L.; Sindelar, R. Bioorg.
Med. Chem. Lett. 1999, 9, 1889.
1
À9.4197 1.0450
2
À10.1327 À0.0741
À10.1610 À0.1515
À10.1628 À0.1517
À10.0674 À0.1585
À10.0711 À0.1623
À10.1427 À0.1551
À9.6960 À0.1108
À9.6819 À0.0594
1a
1b
1c
1d
1e
1f
1g
À9.5186 0.5616
À9.4879 0.6895
À9.4979 0.7840
À9.5010 0.8179
À9.4940 0.7725
À9.4375 0.1894
À9.3953 0.8632
2a
2b
2c
2d
2e
2f
2g
10. Zaprutko, L.; Partyka, D.; Bednarczyk-Cwynar, B. Bio-
org. Med. Chem. Lett. 2004, 14, 4723.
11. Honda, T.; Rounds, B. V.; Bore, L.; Finlay, H. J.;
Favaloro, F. G., Jr.; Suh, N.; Wang, Y.; Sporn, M. B.;
Gribble, G. W. J. Med. Chem. 2000, 43, 4233.
12. Favaloro, J. F. G.; Honda, T.; Honda, Y.; Gribble, G. W.;
Suh, N.; Risingsong, R.; Sporn, M. B. J. Med. Chem.
2002, 45, 4801.
The calculations were carried out at semi-empirical PM3 level.
be investigated further. These results suggested that con-
version of the C11-methylene to a carbonyl group
strongly impacted the energies and locations of HOMO
and LUMO, and the compound with high EHOMO might
have the potent activity. It is worth to indicate that a lot
of efforts remained in order to reach a simple rule to pre-
dict the activity. Hopefully, further detailed analyses
could clarify the relationship between the electronic
structure and activity, create electronic indices including
the optimal locations and the energies of HOMO and
LUMO, environmental aspects, such as hydrophobicity,
and give a guide to synthesis of the more potent
inhibitors.
13. Jeong, H. J.; Chai, H. B.; Park, S. Y.; Kim, D. S. Bioorg.
Med. Chem. Lett. 1999, 9, 1201.
14. Barnes, R. A.; Pereira, A. L.; Scofield, T. C. V.; Filho, R.
B.; Pinto, A. C. Chem. Pharm. Bull. 1984, 32, 3674.
15. Linde, H. Arch. Pharm. (Weinheim, Germany) 1979, 312,
832.
16. Brenner, M.; Huber, W. Helv. Chim. Acta 1953, 36, 1109.
17. Takahashi, N.; Akatsu, T.; Udagawa, N.; Sasaki, T.;
Yamaguchi, A.; Moseley, J. M.; Martin, T. J.; Suda, T.
Endocrinology 1988, 123, 2600.
18. Okubo, Y.; Bessho, K.; Fujimura, K.; Kusumoto, K.;
Ogawa, Y.; Iizuka, T. Biochem. Biophys. Res. Commun.
2000, 269, 317.
19. Niu, J.; Yu, G. Environ. Toxicol. Pharmacol. 2004, 18, 39.