Synthesis and anti-HIV activity of oleanolic acid derivatives

Article abstract of DOI:10.1016/S0960-894X(01)00647-3

Thirteen oleanolic acid derivatives were prepared and evaluated for anti-HIV activity in H9 lymphocytes. Saturating the C₁₂-C₁₃ double bond and converting the C₁₇-carboxyl group to an aminomethyl group led to compounds 13-15 and 19-20, respectively, which showed improved anti-HIV activity. Compound 15 was the most potent derivative with EC₅₀=0.0039 μg/mL and TI=3570.

Full text of DOI:10.1016/S0960-894X(01)00647-3

Bioorganic & Medicinal Chemistry Letters 11 (2001) 3115–3118
Synthesis and Anti-HIV Activity of Oleanolic Acid Derivatives
Yong-Ming Zhu,a Jing-Kang Shen,a Hui-Kang Wang,b L. Mark Cosentinoc
and Kuo-Hsiung Leeb,*
aShanghai Institute of Materia Medica, Chinese Academy of Sciences, 294 Taiyuan Road, Shanghai 200031, China
bNatural Products Laboratory, School of Pharmacy, University of North Carolina at Chapel Hill, Chapel Hill,
NC 27599-7360, USA
cBiotech Research Laboratories, 217 Perry Parkway, Gaithersburg, MD 20877, USA
Received 13 July 2001; accepted 14 September 2001
Abstract—Thirteen oleanolic acid derivatives were prepared and evaluated for anti-HIV activity in H9 lymphocytes. Saturating the
C12–C13 double bond and converting the C17-carboxyl group to an aminomethyl group led to compounds 13–15 and 19–20,
respectively, which showed improved anti-HIV activity. Compound 15 was the most potent derivative with EC50=0.0039 mg/mL
and TI=3570. # 2001Elsevier Science Ltd. All rights reserved.
In order to combat the debilitating disease acquired
immune deficiency syndrome (AIDS), the USA FDA
has approved new drugs for the treatment of human
immunodeficiency virus (HIV) infection. These drugs
include nucleoside analogue HIV reverse transcriptase
(RT) inhibitors,1,2 non-nucleoside RT inhibitors,3 and
protease inhibitors.4 However, the efficacy of these HIV
enzyme inhibitors is limited by the development of drug
resistance and their toxic side effects.5,6 Thus, current
anti-HIV drug discovery is largely focused on new
agents with novel mechanisms of action.
Table 1. Esterification of 1 at C3-hydroxy gave 2–5,
which were as active or less active than the lead com-
pound. Saturation of the C12–C13 double bond in 1 gave
10 and led to a 3-fold increase in activity and in TI
value. Esterification of 10 with appropriate anhydrides
yielded compounds 11–15. While 11 was less active than
the parent acid (10), 12–14 were 5-fold more active than
10, and 15 showed remarkable activity. Converting the
C17-carboxyl of 1 to aminomethyl gave 18, which could
not be tested due to its insolubility in DMSO. Intro-
ducing 30,30-dimethylglutaryl or 30,30-tetramethylglutaryl
groups at both the C3 and C28 positions of 18 produced
19–20, which displayed more than a tenfold increase in
activity compared with 1.
In our continuing search for new anti-HIV drug candi-
dates, we previously reported that modification of
betulinic acid yielded extremely potent anti-HIV
compounds.7ꢀ9 Recently, we also discovered that olea-
nolic acid exhibited anti-HIV activity with an EC50 of
1.7 mg/mL and therapeutic index (TI) of 12.8.10 Based
on this finding, oleanolic acid was also considered to be
a potential anti-HIV lead triterpenoid. To explore the
roles of the acyl substituent and other functional
groups, we developed a method to saturate the C12–C13
double bond and, in addition, converted the C17-carboxyl
to an aminomethyl group (Fig. 1).
In conclusion, oleanolic acid derivatives 12–15, 19, and
20 showed high anti-HIV activity. Compound 15 was
the most active with EC50 and TI values of 0.0039 mg/
mL and 3570, respectively. Saturation of the C12–C13
double bond could be a major cause of anti-HIV activ-
ity enhancement, whereas a C3 acryl side chain was
essential for optimal activity. In addition, changing
the C17-carboxyl to aminomethyl could significantly
enhance anti-HIV activity. Additional structure–activity
relationship studies are in process.
Compounds 2–5, 10–15, 18–20 and AZT were examined
for anti-HIV activity in H9 lymphocytes as shown in
A general synthesis for compounds 2–5 and 11–15 was
performed according to the following procedure. Oleanolic
acid or C12, C13-dihydrooleanolic acid was heated to reflux
overnight with 2 equiv of the appropriate anhydride in
anhydrous pyridine in the presence of 4-(dimethylamino)
*Corresponding author. Tel.:+1-919-962-0066; fax:+1-919-966-3893;
e-mail: khlee@email.unc.edu
0960-894X/01/$ - see front matter # 2001Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00647-3

3116
Y.-M. Zhu et al. / Bioorg. Med. Chem. Lett. 11 (2001) 3115–3118
Figure 1.
(10) was produced by Wolff–Kishner reduction of 9 in
tripolyethylene glycol solution.
Table 1. Anti-HIV activity of oleanolic acid and derivatives
Compd
Anti-HIV activitya
EC50 (mg/mL)
Cytotoxicitya
IC50 (mg/mL)
Therapeutic index
Tl=IC50/EC50
The preparation of compound 18 is illustrated in
Scheme 2. 3-Acetoxyoleanolicacid (6) was treated with
oxalyl chloride to yield 3-acetoxyoleanolic chloride (16).
A solution of 16 in freshly distilled THF was added
dropwise to concentrated ammonia solution cooled with
an ice-salt bath to afford 3-acetoxyoleanolic amide (17).
Compound 17 was refluxed with 2.0 equiv of lithium
aluminum hydride in anhydrous ether to give 18. Com-
pounds 19 and 20 were prepared by heating 18 with 3
equiv of 3,3-dimthylglutaryl and 3,3-tetramethylglutaryl
anhydride, respectively, in anhydrous pyridine in the
presence of 4-(dimethylamino) pyridine (catalytic) over-
night at 100 ꢁC. The reaction mixture was worked up as
for compounds 2–5 and 11–15.
1
1.7
7.11 2.1
8.3
1.5
1.2
0.5
2.6
0.1
0.18.3
0.1
0.0039
NT
0.1
0.1
0.01500
21.8
12.8
2
4.6
3
4
5
27.5
14.2
4.7
17.4
12.9
12.3
83
10.1
13.7
NT
21.2
19.5
50,000
3.3
9.3
3.9
34.8
4.9
123
10
11
12
13
14
15
18
19
20
AZT
101
3570
NT
212
195
aNT, not tested due to solubility. All data are an average of at least two
experiments.
The biological evaluation of HIV-1inhibition was car-
ried out according to established protocols. The T cell
line, H9, was maintained in continuous culture with
complete medium (RPMI 1640 with 10% fetal calf
serum supplemented with l-glutamine at 5% CO2 and
37 ꢁC). Aliquots of this cell line were used in experi-
ments only when in log-phase growth. Test samples
were first dissolved in dimethylsulfoxide. The following
final drug concentrations were routinely used for
screening: 100, 20, 4, and 0.8 mg/mL. For active agents,
additional dilutions were prepared for subsequent test-
ing so that an accurate EC50 value (defined below) could
be achieved. As the test sample was being prepared, an
aliquot of the H9 cell line was infected with HIV-1(IIIB
isolate) while another aliquot was mock-infected with
complete medium. The stock virus used for these studies
typically had a TCID 50 value of 104 Infectious Units/
mL. The appropriate amount of virus for a multiplicity
of infection (moi) between 0.1and 0.01Infectious
Units/mL was added to the first aliquot of H9 cells. The
other aliquot received only culture medium, and these
pyridine (catalytic). When TLC indicated complete
consumption of starting material, the reaction solution
was diluted with CHCl3 and washed with 3 N HCl
solution. The organic layer was then dried over MgSO4
and subjected to column chromatography. Compound
15 was obtained from preparative HPLC.11
The synthesis of compound 10 is outlined in Scheme 1.
Oleanolic acid was acetylated with 10 equiv of acetic
anhydride in anhydrous pyridine to give 3-acetoxy-
oleanolic acid (6) and then treated with ethereal diazo-
methane to yield methyl 3-acetyloleanolic acid (7). On
oxidation with 1.2 equiv of 3-chloroperbenzoic acid in
dry CHC13 at 0 ꢁC overnight, compound 7 afforded
methyl 3-acetoxy-12,13-epoxyoleanolate (8).12 Com-
pound 8 was refluxed with ethereal boron trifluoride in
fresh benzene for 30 minutes to give methyl 3-acetoxy-
12-oxo-13-oleanolate (9).13,14 Dihydrooleanolic acid

Y.-M. Zhu et al. / Bioorg. Med. Chem. Lett. 11 (2001) 3115–3118
3117
Scheme 1.
Scheme 2.
mock-infected cells were used for toxicity determina-
tions (IC50 defined below). After a 4 h incubation at
37 ꢁC and 5% CO2, both cell populations were washed
three times with fresh medium and then added to the
appropriate wells of a 24-plate containing the various
concentrations of the test drug or culture medium
(positive infected control negative drug control). In
addition AZT was also assayed during each experiment
as a positive drug control. The plates were incubated at
37 ꢁC and 5% CO2 for 4 days. Cell-free supernatants
were collected on day 4 for use in our in-house p24
antigen ELISA assay. P24 antigen is a core protein of
HIV and therefore is an indirect measure of virus pre-
sent in the supernatants. Toxicity was determined by
performing cell counts by a Coulter Counter on the
mock-infected H9 cells, which had either received cul-
ture medium (no toxicity), test sample, or AZT. If a test
sample had suppressive capability and was not toxic, its
effects were reported in the following terms: IC50, the
concentration of the test sample which was toxic to 50%
of the mock-infected H9; EC50, the concentration of test
sample which was able to suppress HIV replication by
50%; and Therapeutic Index (TI), the ratio of IC50 to
EC50.
Acknowledgements
This investigation was supported in part by grant AI-
33066 from the National Institute of Allergies and
Infectious Diseases awarded to K. H. Lee.

3118
Y.-M. Zhu et al. / Bioorg. Med. Chem. Lett. 11 (2001) 3115–3118
10. Kashiwada, Y.; Wang, H. K.; Nagao, T.; Kitanaka, S.;
References and Notes
Yasuda, I.; Fujioka, T.; Yamagishi, T.; Cosentino, L. M.;
Kozuka, M.; Okabe, H.; Ikeshiro, Y.; Hu, C. Q.; Yeh, E.; Lee,
K. H. J. Nat. Prod. 1998, 61, 1090.
1. Debyser, Z.; Pauwels, R.; Andries, K.; Declereq, E. J.
Enzyme Inhib. 1992, 6, 47.
2. Barre-Sinooussi, F.; Chermann, J. C.; Rey, F. Science 1983,
226, 868.
3. Crob, P. M.; Cao, Y.; Muchmore, E.; Ho, D. D.; Norris, S.;
Pav, J. W.; Shih, C. K.; Adams, J. Nature Medicine 1997, 3, 665.
4. Meek, T. D. J. Enzyme Inhib. 1992, 6, 65.
11. Compound 15: yield 46% [after HPLC separation with n-
hexane/ethyl acetate (10:1)], a white amorphous powder, mp
289–290 ꢁC (MeOH–H2O), 1H NMR (CDCl3, 400 MHz)
d, 0.80 (3H, s, H-25), 0.83 (9H, s, H-23, H-24, H-29), 0.90 (3H,
s, H-26), 0.93 (3H, s, H-30), 0.99 (3H, s, H-27), 1.27, 1.32 [3H
each, both s, 30-(CH3)2], 2.15 (1H, d, J=14.2 Hz, H-18), 2.40,
2.90 (1H, each, both d, J=15.1 Hz, H-20), 4.52 (1H, dd,
J=11.3, 3.4 Hz), IR (KBr): 1729.9, 1697.1, 1269.0, 993.2
cmꢀ1. Anal. calcd for C36H58O6, C 73.68, H 9.96, found C
73.81, H 9.95.
12. Boar, R. B.; Joukhadar, L.; Deluque, M.; McGhie, J. F.;
Barton, D. H. R.; Arigoni, D.; Brunner, H. G.; Giger, R. J.
Chem. Soc., Perkin Trans. 1 1977, 2104.
13. Farina, C.; Pinza, M. Gazz. Chim. Ital. 1987, 117, 561.
14. Boar, R. B.; Knight, D. C.; McGhie, J. F.; Barton,
D. H. R. J. Chem. Soc. (C) 1970, 678.
5. Schinazi, R. F.; Mead, J. R.; Feorino, P. M. Human
Retrovir. 1992, 8, 963.
6. Shirahata, T.; Yarchoan, R.; O’Brien, M. C.; Husson,
R. N.; Anderson, B. D.; Kojima, E.; Shimoda, T.; Broder, S.;
Mitsuya, H. Proc. Natl. Acad. Sci, U.S.A. 1993, 91, 562.
7. Kashiwada, Y.; Hashimoto, F.; Cosentino, L. M.; Chen,
C. H.; Garrett, P. E.; Lee, K. H. J. Med. Chem. 1996, 39, 1016.
8. Sun, I. C.; Shen, J. K.; Wang, H. K.; Cosentino, L. M.;
Lee, K. H. Bioorg. Med. Chem. Lett. 1998, 8, 1267.
9. Sun, I. C.; Wang, H. K.; Kashiwada, Y.; Shen, J. K.;
Cosentino, L. M.; Chen, C. H.; Yang, L. M.; Lee, K. H. J.
Med. Chem. 1998, 41, 4648.