Synthesis and biological evaluation of nitric oxide-releasing derivatives of oleanolic acid as inhibitors of HepG2 cell apoptosis

Article abstract of DOI:10.1016/j.bmcl.2007.03.068

A total of 106 nitric oxide-releasing derivatives of oleanolic acid were synthesized and their effects on the inhibition of anti-Fas-mediated HepG2 cell apoptosis were evaluated in vitro. Several compounds inhibited anti-Fas-mediated HepG2 cell apoptosis

Full text of DOI:10.1016/j.bmcl.2007.03.068

Bioorganic & Medicinal Chemistry Letters 17 (2007) 2979–2982
Synthesis and biological evaluation of nitric oxide-releasing
derivatives of oleanolic acid as inhibitors of HepG2 cell apoptosis
Li Chen,a,b Yihua Zhang,a,* Xiangwen Kong,a Sixun Penga and Jide Tianc,*
aCenter of Drug Discovery, China Pharmaceutical University, 24 Tongjiaxiang, Nanjing 210009, PR China
bDepartment of Phytochemistry, China Pharmaceutical University, 1 Shennonglu, Nanjing 210038, PR China
cDepartment of Molecular & Medical Pharmacology, University of California Los Angeles, Los Angeles, CA 90095, USA
Received 20 December 2006; revised 19 March 2007; accepted 22 March 2007
Available online 25 March 2007
Abstract—A total of 106 nitric oxide-releasing derivatives of oleanolic acid were synthesized and their effects on the inhibition of
anti-Fas-mediated HepG2 cell apoptosis were evaluated in vitro. Several compounds inhibited anti-Fas-mediated HepG2 cell apop-
tosis in a dose-dependent manner. Within this series of compounds, 8b is the most potent inhibitor. The development of new NO-
releasing derivatives of oleanolic acid may aid in the design of NO-based medicines for the intervention of human liver inflammatory
diseases.
ꢀ 2007 Elsevier Ltd. All rights reserved.
Nitric oxide (NO), synthesized from L-arginine by a
family of constitutive and inducible NO synthases
(cNOS and iNOS), is a small, diffusible, highly reactive
molecule with dichotomous biological functions, such
as blood vessel dilatation, neurotransmission, modula-
tion of the hair cycle, and penile erections.1 NO can also
be generated from synthetic NO-releasing compounds,
such as nitrate, furoxan, hydroxyguanidine, S-nitroso-
thiol, diazeniumdiolate, and others.2 Higher concentra-
tions of NO can promote the apoptosis of many types
of cells, while lower concentrations of NO usually pro-
tect cells, such as hepatocytes, from inflammation-med-
iated apoptosis.3 Several mechanisms have been
proposed to explain the anti-apoptosis effects of NO,
including increased cGMP production, inhibition of cas-
pase activation as well as prevention of mitochondrial
permeability transition that leads to the release of cyto-
chrome C into the cytosol.4
of NO replacement therapy in clinic. However, the di-
verse bioactivities of NO limit its clinical application.
New types of NO-releasing compounds/drug hybrids
need to be developed for obtaining tissue-specific NO-
related function.
Oleanolic acid (OA) is a triterpenoid compound, widely
found in natural plants.5 Systemic treatment with the
purified OA leads predominately to the distribution
and metabolism of OA in the liver. OA has been shown
to protect rodent liver from CCl4 and many other toxi-
cant-induced hepatotoxicity and chronic cirrhosis.6
Importantly, OA has been clinically used as a safe non-
prescription drug for treatment of hepatitis in China for
more than 20 years,7 although its therapeutic efficacy is
limited. The therapeutic effect, the property of liver-spe-
cific metabolism, and wide availability make it to be an
ideal base for the design of new NO-releasing com-
pounds for the production of NO specifically in the liver.
The combination of NO and OA probably provides syn-
ergic protection of hepatocytes from inflammation- and
toxicant-mediated liver damage.
Increased scientific evidence indicates that NO defi-
ciency is implicated in many physiological and patholog-
ical processes within the mammalian body. This fact
provides a plausible biologic basis for the application
In the present studies, 106 novel NO-releasing deriva-
tives of OA were synthesized by connecting NO-donat-
ing moiety to the OA-28-COOH/OA-3-OH through
varying lengths of linkers. The various linkers containing
anti-oxygen functionalities, such as ferulic acid,
p-hydroxyl cinnamic acid, and vanillic acid,8 were
Keywords: Oleanolic acid; Nitrate; NO-donors; Biological evaluation;
Apoptosis.
*
Corresponding authors. Tel./fax: +86 25 86635503 (Y.Z.); tel.: +1 310
206 3350; fax: +1 310 825 6267 (J.T.); e-mail addresses:
zyhtgd@hotmail.com; jtian@mednet.ucla.edu
0960-894X/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.03.068

2980
L. Chen et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2979–2982
designed to increase the ability of these objective com-
pounds to scavenge free radicals. The bioactivities of
all derivatives of OA were evaluated. Several compounds
were found to inhibit anti-Fas-mediated hepatocyte
apoptosis and their anti-apoptotic effects were
dose-dependent. One of the compounds, 8b, inhibited
hepatocyte apoptosis at a lower nanomolar level.
The development of new NO-releasing derivatives of
OA may aid in the design of NO-based new medicine
for the intervention of human inflammatory liver
diseases.
The synthetic routes of OA-nitrate conjugates (8a–8g,
9a–9g, 11a, 11b, 12a, and 12b) are summarized in
Schemes 2 and 3. Several strategies were attempted to
esterify OA-28-COOH directly with hydroxyl com-
pounds, but failed, perhaps because of the large steric
hindrance and weak acidity of OA-28-COOH. Alterna-
tively, OA was first treated with trifluoroacetic anhy-
dride to form a mixed anhydride in a quantitative
yield after stirring at room temperature for 10 min.
The resulting mixed anhydride was then reacted with hy-
droxyl compounds (2a–2g or 6a and 6b) in toluene to af-
ford 3-O-trifluoroacetyl OA esters (7a–7g or 10a and
10b) in good yields (70–78%), respectively.9,10 Com-
pounds 7a–7g or 10a and 10b were further converted
to the corresponding nitrates 8a–8g or 11a and 11b,
respectively, with AgNO3 in THF/CH3CN. Compounds
9a–9g or 12a and 12b were prepared by the hydrolysis of
8a–8g or 11a and 11b, respectively, with diluted KHCO3
to remove trifluoroacetyl group at C-3–OH without
affecting other ester bonds. The resulting products were
purified by column chromatography and their structures
were characterized by IR, 1H NMR, MS, and elemental
analysis.11
The synthetic routes of very important intermediates
(2a–2g, 6a, and 6b) are outlined in Scheme 1. Ferulic
acid (1a) or p-hydroxyl cinnamic acid (1b) was first trea-
ted with dibromoalkanes bearing two to six carbons in
the presence of Et3N and acetone at 50 ꢁC to generate
compounds 2a–2g in 60–73% yields. In a similar way,
compounds 6a and 6b were obtained in 61–65% yields
by treatment of vanillic acid (5) with 1, 3-dibromopro-
pane and 1, 4-dibromobutane, respectively.
Compounds 8a–8g, 9a–9g, 11a, 11b, 12a, 12b, and con-
trols, OA and NCX-1000, were in vitro evaluated for
their protective effects on anti-Fas-mediated HepG2 cell
apoptosis determined by LDH assay.12 As shown in
Figure 1, treatment with different concentrations of
OA failed to protect the HepG2 cells from anti-Fas-in-
duced apoptosis as there was no significant difference
in the percentage of survived HepG2 cells between the
presence and absence of different concentrations of
OA. Treatment with NCX-1000, a NO-releasing deriva-
tive of urodeoxycholic acid (UDCA) generated by add-
ing an ONO2 moiety through an antioxidant spacer,
showed moderate protective effects and NCX-1000 at
10À6 M protected 33.4% of HepG2 cells from anti-Fas-
induced apoptosis, consistent with previous report.3
Furthermore, five of the tested compounds displayed
great protective effects, which were 10- to 1000-fold
OH
OH
R
R
i
CH=CHCOOH
CH=CHCOO(CH2)nBr
2a R=OCH3, n=2 2b R=OCH3, n=3
2c R=OCH3, n=4 2d R=OCH3, n=6
2e R=H, n=2
1a R=OCH3
1b R=H
2f R=H, n=3
2g R=H, n=4
OH
OH
OH
OH
OCH3
OCH3
OCH
3 iv
OCH3
ii
iii
CHO
COONa
COOH
COO(CH2)nBr
6a n=3
6b n=4
5
4
3
Scheme 1. Reagents and conditions: (i) Br(CH2)nBr, Et3N, 50 ꢁC, 4 h
(60–73%); (ii) AgOH, NaOH (65%); (iii) HCl; (iv) Br(CH2)nBr (n = 3
or 4), Et3N, 50 ꢁC, 4 h (61–65%).
30
19
29
21
22
20
18
H
R
H
13
14
12
11
25
26
17
1
COOH
9
i
CH=CHCOO(CH2)nBr
COO
28
2
16
8
10
15
H
H
6
HO3
7a R=OCH3, n=2
7c R=OCH3, n=4
7e R=H, n=2
7b R=OCH3, n=3
7d R=OCH3, n=6
7f R=H, n=3
27
4
7
F3CCO
O
5
24
23
OA
H
7g R=H, n=4
R
ii
COO
R
CH=CHCOO(CH2)nONO2
H
8a R=OCH3, n=2 8b R=OCH3, n=3
8c R=OCH3, n=4 8d R=OCH3, n=6
F3CCO
O
iii
8e R=H, n=2
8g R=H, n=4
8f R=H, n=3
H
COO
CH=CHCOO(CH2)nONO2
H
9a R=OCH3, n=2 9b R=OCH3, n=3
9c R=OCH3, n=4 9d R=OCH3, n=6
HO
9e R=H, n=2
9g R=H, n=4
9f R=H, n=3
Scheme 2. Reagents and condition: (i) (CF3CO)2O, 2a–2g, <90 ꢁC, 6 h (70–78%); (ii) THF/CH3CN, AgNO3, reflux (67–75%); (iii) KHCO3, rt
(90–95%).

L. Chen et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2979–2982
2981
H3CO
COO
H3CO
COO
H
H
COO(CH ) Br
COO(CH2)nONO2
i
ii
2 n
OA
H
H
11a n=3
11b n=4
10a n=3
F3CCO
O
F3CCO
O
10b n=4
H3CO
COO
H
COO(CH2)nONO2
iii
H
12a n=3
12b n=4
HO
Scheme 3. Reagents and conditions: (i) (CF3CO)2O, 6a and 6b, <90 ꢁC, 6 h (71–73%); (ii) THF/CH3CN, AgNO3, reflux (67–73%); (iii) KHCO3, rt
(90–95%).
suggests that these compounds may be potential for
the treatment of inflammatory diseases in clinic.
OA
60
50
40
30
20
10
0
NCX-1000
8b
8c
As shown in Figure 2, these compounds produced low
levels of NO (about 12 ng/ml for 8b, 12 ng/ml for 8c,
21 ng/ml for 8d, 18 ng/ml for 8f, 20 ng/ml for 8g, and
28 ng/ml for 11b, respectively) determined by in vitro
Griess reaction. This is consistent with the notion that
low dose of NO protects cells from inflammation-medi-
ated apoptosis. Notably, some compounds that pro-
duced higher levels of NO in vitro showed potent
cytotoxicity on human liver cancer cells (unpublished
observations). In addition, our preliminary studies
showed that oral treatment with 8b at 128 mg kgÀ1
significantly inhibited CCl4-induced hepatic fibrosis
and liver damage in rats (data not shown). Collectively,
NO-releasing derivatives of OA with low level of NO
productivity protected human hepatocytes from
anti-Fas-mediated apoptosis.
8d
11b
8f
8g
1.E-06 1.E-07 1.E-08 1.E-09 1.E-10 1.E-11
Concentrations (M)
Figure 1. NO-releasing derivatives of OA-protected hepatocytes from
anti-Fas-mediated apoptosis. The anti-apoptotic effects of NO-releas-
ing derivatives of OA were evaluated by LDH assays. Briefly, HepG2
cells at 104/well were plated in triplicate in 96-well plates in 4% FCS
and 2 lg/ml cycloheximide in phenol-red free DMEM. The cells were
treated with anti-Fas (50 ng/ml for CH-11 or 1.5 lg/ml for DX2) in the
presence or absence of indicated concentrations of OA, NCX-1000 or
NO-releasing derivatives of OA at 37 ꢁC 5% CO2 for 18 h. Unmanip-
ulated cells or the cells treated with 1% Triton X-100 were used as
negative or positive controls for cytotoxicity. Their supernatants were
harvested and their contained LDH activities were measured by LDH
assays using the LDH cytotoxicity detection kit (Roche) and following
the manufacturer’s instructions at O.D. (490/650). The protective effect
(%) was determined by the formula: the protective effect
(%) = (1 À (O.D. experiments À O.D. negative controls)/(O.D. positive
controls À O.D. negative controls)) · 100. Data presented as means of
three independent experiments and the O.D. values of intra-group
variations were less than 10%.
Analysis of structure–activity relationships suggests that
the A-ring of OA may be a good target for further mod-
ification as the compounds 8b–8d, 8f, and 8g, whose C-
3-OH of A-ring was connected with trifluoroacetyl,
showed potent anti-apoptotic activities, while the com-
pounds 9b–9d, 9f, and 9g with corresponding free
30
NCX-1000
8b
25
20
8c
8d
8f
15
10
5
8g
11b
stronger than that of NCX-1000. The lowest concentra-
tion that significantly inhibited anti-Fas-induced HepG2
cell apoptosis was 10À10 M for 8b, 10À9 M for 8c,
10À8 M for 8f or 10À7 M for 8d and 8g, respectively. A
similar pattern of protective effects was observed when
using the p53À/À Hep3b cells, indicating that the protec-
tive effects of these NO-releasing derivatives of OA were
independent of p53. Given that Fas/FasL-mediated
hepatocyte apoptosis was involved in many inflamma-
tory- and toxicant-mediated liver diseases the potent
anti-apoptotic effect of NO-releasing derivatives of OA
0
0
100
200
300
400
Time (min)
Figure 2. NO-releasing derivatives of OA produced low levels of NO
À
in vitro. NO2 concentrations which represent the quantity of NO
wereÀdetermined by Griess assay. Griess reagent could combine with
NO2 and form the chromophore after 10 min at 30 ꢁC, the
absorbance then was measured at 540 nm.

2982
L. Chen et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2979–2982
Iwamoto, J.; Ayabe, T.; Kasai, S. J. Surg. Res. 2004, 118,
183; (c) Jenkins, D. C.; Charles, L. L.; Thomsen, D. W.;
Moss, D. W.; Holmes, L. S.; Baylis, S. A.; Rhode, P.;
Westmore, K.; Emson, P. C.; Moncada, S. Proc. Natl.
Acad. Sci. U.S.A. 1995, 92, 4392.
C-3-OH were inactive. Second, the length of the linkers
connecting to ferulic acid or p-hydroxyl cinnamic acid
appears crucial for their bioactivities. The 8b with a
3-carbon long linker had 10- to 100-fold more potent
anti-apoptotic activity than 8c or 8d, respectively, while
8a was inactive. Similarly, the compounds 8f and 8g, but
not 8e, had anti-apoptotic activity. Apparently, a 3-car-
bon long linker was optimal, while compounds with
longer linkers were accompanied by decreased bioactiv-
ities. Furthermore, the potent anti-apoptotic activities
were governed by appropriate anti-oxygen reagent, like
ferulic acid, at the C-28 of OA esters as compounds 8b
and 8c showed higher anti-apoptotic activities than the
corresponding compounds 8f, 8g and 11a, 11b, even if
they had the same length of carbon chain. Finally, these
OA-nitrate conjugates at less than 10À6 M, but not OA,
inhibited anti-Fas-mediated HepG2 cell apoptosis in our
experimental system, suggesting that the NO-releasing
moieties may be responsible for the inhibitory activities
of these derivatives of OA.
4. Moncada, S.; Erusalimsky, J. D. Nat. Rev. Mol. Cell. Biol.
2002, 3, 214.
5. Liu, J. J. Ethnopharmacol. 1995, 49, 57.
6. (a) Liu, Y.; Hartley, D. P.; Liu, J. Toxicol. Lett. 1998, 95,
77; (b) Han, D. W.; Ma, X. H.; Zhao, Y. C.; Yin, L.; Ji, C.
X. J. Tradit. Chin. Med. 1982, 2, 83; (c) Liu, J.; Liu, Y.;
Madhu, C.; Klaassen, C. D. J. Pharmacol. Exp. Ther.
1993, 266, 1607; (d) Liu, Y.; Kreppel, H.; Liu, J.;
Choudhuri, S.; Klaassen, C. D. J. Pharmacol. Exp. Ther.
1993, 266, 400.
7. (a) Liu, J.; Liu, Y.; Parkinson, A. J. Pharmacol. Exp. Ther.
1995, 275, 768; (b) Wang, L. X.; Han, G. X.; Liu, W. Y.
J. Pharm. Practice 2001, 19, 104.
8. (a) Ozaki, Y. Chem. Pharm. Bull. 1992, 40, 954; (b) Hu,
H. J.; Hang, B. Q. J. China Pharm. Univ. 1990, 21, 279;
(c) Lin, Y. H.; Chen, W. W. Acta Pharm. Sin. 1994, 28,
267.
9. (a) Parish, R. C.; Stock, L. M. Tetrahedron Lett. 1964, 20,
1285; (b) Bourne, J. E.; Stacey, M.; Tatlow, J. C.; Tedder,
J. M. J. Chem. Soc. 1949, 2976.
10. Chen L. Ph.D. dissertation. China Pharmaceutical Uni-
versity at Nanjing in China, 2004.
A series of OA-nitrate conjugates were synthesized and
their biological functions were evaluated. Within the ser-
ies of compounds, five of them showed to inhibit anti-
Fas-mediated hepatocyte apoptosis in vitro and their
anti-apoptotic effects were dose-dependent. Notably,
the compound 8b was the most potent inhibitor and pro-
tected hepatocytes from anti-Fas-mediated apoptosis at
a lower nanomolar level. The preliminary SAR analysis
of these compounds revealed that connection of the
A-ring of OA with an optimal length of linker and the
C-28 of OA esters with an anti-oxygen reagent was cru-
cial for their anti-apoptotic activities. Together, these
findings provide a new framework for the rational
design of NO-based medicine for the intervention of hu-
man inflammatory liver diseases in clinic.
11. Analytical data for 8b: mp 112–114 ꢁC; IR (KBr, cmÀ1) :
1779, 1750, 1715, 1635; 1H NMR (CDCl3, 300 MHz): d
0.86 (s, 3H, CH3), 0.92 (s, 6H, 2· CH3), 0.94 (s, 3H, CH3),
0.97 (s, 6H, 2· CH3), 1.19 (s, 3H, CH3), 2.10–2.18 (m, 2H,
CH2), 2.94–3.00 (m, 1H, C18–H), 3.83 (s, 3H, OCH3), 4.32
(t, 2H, OCH2, J = 6.0 Hz), 4.61 (t, 2H, OCH2, J = 6.0 Hz),
4.68–4.73 (m, 1H, H-3a), 5.33 (br s, 1H, H-12), 6.36 (d,
1H, @CH, J = 16.0 Hz), 6.95–6.97 (m, 1H, ArH), 7.07–
7.12 (m, 2H, ArH), 7.65 (d, 1H, @CH, J = 16.0 Hz); ESI-
MS: 854 [M+Na]+; Anal. C45H60F3NO10. Found (%): N
1.56, C 65.34, H 7.31; Calcd (%): N 1.68, C 64.79, H
7.27.Analytical data for 9b: mp 98–100 ꢁC; IR (KBr,
cmÀ1) : 3523, 1748, 1715, 1634; 1H NMR (CDCl3,
300 MHz): d (ppm): 0.82 (s, 3H, CH3), 0.88 (s, 3H,
CH3), 0.95 (s, 3H, CH3), 0.97 (s, 3H, CH3), 1.00 (s, 3H,
CH3), 1.03 (s, 3H, CH3), 1.23 (s, 3H, CH3), 2.96–3.02 (m,
1H, H-18), 3.25 (br s, 1H, H-3a), 3.85 (s, 3H, OCH3), 4.35
(t, 2H, OCH2, J = 6 Hz), 4.64 (t, 2H, OCH2, J = 6.0 Hz),
5.36 (br s, 1H, C12–H), 6.38 (d, 1H, @CH, J = 16.0 Hz),
6.99 (d, 1H, ArH, J = 7.8Hz), 7.11–7.16 (m, 2H, ArH),
7.68 (d, 1H, @CH, J = 16.0Hz); ESI-MS: 758 [M+Na]+;
Anal. C43H61NO9Æ0.5H2O. Found (%): N 1.64, C 69.34, H
8.52; Calcd (%): N 1.88, C 69.31, H 8.32.
Acknowledgments
We thank Ms. Xiaochun Feng and Ms. Xiuying Chen for
their technical supports in some biological assays. This
study was partially supported by grant from the National
Institute of Health (NIH) (CA116846, Jide Tian).
Analytical data for 11b: mp 64–66 ꢁC; IR (KBr, cmÀ1) :
1779, 1756, 1720, 1632; 1H NMR (CDCl3, 300 MHz): d
0.85 (s, 3H, CH3), 0.92 (s, 6H, 2· CH3), 0.94 (s, 3H, CH3),
0.97 (s, 6H, 2· CH3), 1.19 (s, 3H, CH3), 2.94–3.00 (m, 1H,
H-18), 3.85 (s, 3H, OCH3), 4.36 (br s, 2H, OCH2), 4.53 (br
s, 2H, OCH2), 4.68–4.73 (m, 1H, 3a-H), 5.33 (br s, 1H,
C12–H), 7.00 (d, 1H, ArH, J = 7.9 Hz), 7.61–7.64 (m, 2H,
ArH); ESI-MS : 823 [M+NH4]+, 828 [M+Na]+; Anal.
C44H60F3NO10. Found (%): N 1.62, C 64.16, H 7.45;
Calcd (%): N 1.71, C64.45, H 7.38.
References and notes
1. Grisham, M. B.; Jourdheuil, D.; Wink, D. A. Am. J.
Physiol. 1999, 276, G315.
2. (a) Hou, Y. C.; Janczuk, A.; Wang, P. G. Curr. Pharm.
Des. 1999, 5, 417; (b) Wang, P. G.; Xian, M.; Tang, X.;
Wu, X.; Wen, Z.; Cai, T.; Janczuk, A. J. Chem. Rev. 2002,
102, 1091.
3. (a) Fiorucci, S.; Mencarelli, A.; Palazzetti, B.; Del Soldato,
P.; Morelli, A.; Ignaro, L. J. Proc. Natl. Acad. Sci. U.S.A.
2001, 98, 2652; (b) Ebisawa, Y.; Kono, T.; Yoneda, M.;
Asama, T.; Chisato, N.; Sugawara, M.; Ishikawa, K.;
12. Fernandez-Cruz, E.; Escartin, P.; Bootello, A.; Kreisler,
M.; Segovia de Arana, J. M. Clin. Exp. Immunol. 1978, 31,
436.