Enantioselective synthesis and complement inhibitory assay of A/B-ring partial analogues of oleanolic acid

Article abstract of DOI:10.1016/S0960-894X(01)00210-4

A series of oleanolic acid A/B-ring partial analogues was synthesized and tested for their complement inhibitory activity as well as cytotoxic properties. All target compounds and one intermediate exhibited moderate complement inhibitory potency. These co

Full text of DOI:10.1016/S0960-894X(01)00210-4

Bioorganic & Medicinal Chemistry Letters 11 (2001) 1619–1623
Enantioselective Synthesis and Complement Inhibitory Assay of
A/B-Ring Partial Analogues of Oleanolic Acid
Haregewein Assefa,a Alison Nimrod,b Larry Walkerb,c and Robert Sindelara,b,
*
aDepartment of Medicinal Chemistry, School of Pharmacy, University of Mississippi, University, MS 38677, USA
bNational Center for Natural Product Research, School of Pharmacy, University of Mississippi, University, MS 38677, USA
cDepartment of Pharmacology, School of Pharmacy, University of Mississippi, University, MS 38677, USA
Received 10 November 2000;accepted 16 March 2001
Abstract—A series of oleanolic acid A/B-ring partial analogues was synthesized and tested for their complement inhibitory activity
as well as cytotoxic properties. All target compounds and one intermediate exhibited moderate complement inhibitory potency.
These compounds also showed cytotoxicity on malignant melanoma cell line, SK-MEL. # 2001 Elsevier Science Ltd. All rights
reserved.
Complement is an essential component of the innate
immune system that provides a first line defense and
immune complex clearance in the blood stream.1,2 Fur-
thermore, complement plays a role in various functions
of the adaptive immune response.3,4 Yet overactivation
of complement is implicated in various inflammatory
diseases and xenotransplant rejection.5À8 Complement
inhibitors have been found to ameliorate these deleter-
ious conditions, and a continued search for complement
inhibitors has resulted in the identification of numerous
natural and synthetic compounds.9À11 However, no
complement inhibitor has been approved for clinical use
in the US due to the lack of potency and selectivity. The
triterpene natural product, oleanolic acid (3b-hydroxy-
olean-12-en-28-oic acid, 1) inhibits the C3-convertase of
the classical pathway in vitro.12 Oleanolic acid also
inhibits complement-mediated inflammation in animal
models.13 Earlier, we reported the synthesis and com-
plement inhibitory activity of various semisynthetic
analogues of oleanolic acid.14 To further explore the
regions of the molecule that confer complement inhibi-
tory activity, we have designed and synthesized A/B-
ring partial analogues of oleanolic acid. This dissection
of the molecule also provides an easy access to a wider
variety of analogues. The A/B-ring of oleanolic acid
with its stereochemistry has been retained in the partial
analogues while the carboxylic moiety has been tethered
to the A/B-ring using a styrene (2a–c) or a benzyl spacer
group (3a–c) (Fig. 1).
The optically pure compound 7 was synthesized using
the method developed by Hagiwara and Uda.15
Accordingly, condensation of 2-methyl-1,3-cyclohex-
anedione (4) with ethyl vinyl ketone in the presence of
Figure 1. Oleanolic acid (1) and its partial analogues (2a–c and 3a–c).
*Corresponding author. Tel.: +1-662-915-7101;fax: +1-662-915-5638;e-mail: sindelar@olemiss.edu
0960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00210-4

1620
H. Assefa et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1619–1623
Scheme 1. Reagents and conditions: (a) ethyl vinyl ketone, Et3N/THF, reflux;(b) l-Phe, d-CSA/DMF;(c) 2-ethyl-2-methyl-1,3-dioxolane, ethylene
glycol, d-CSA, 50 ꢀC;(d) (i) Li/liq NH 3;(ii) CH 3I;(e) NaBH 4/C2H5OH;(f) NaH, p-CH3OPhCH2Cl/DMF;(g) 1 N HCl/CH 3COOH/THF[1/2/3];(h)
NaH/DMSO, CH3OCH2PPh3Cl;(i) HCl/THF.
Scheme 2. Reagents and conditions: (a) NaH/DMSO, (p-carbomethoxybenzyl)triphenylphosphonium bromide;(b) DDQ/CH 2Cl2/H2O;(c) aq
NaOH/CH3OH/THF.
triethylamine under reflux followed by enantioselective
cyclization of the resulting triketone using l-phenylala-
nine (l-Phe) as a chiral auxiliary afforded the S-enan-
tiomer 5 as a major product. The pure S-enantiomer 5
obtained by crystallization was used as a starting mat-
erial for the subsequent steps. Selective protection of the
unconjugated keto group of 5 using 2-ethyl-2-methyl-
1,3-dioxolane in the presence of d-camphorsulfonic acid
(d-CSA) and treatment of the ketal with Li/liq NH3
followed by iodomethane yielded the trans-decalin 6.
Reduction of the keto group of 6 using NaBH4 gave the
b-epimer 7 as the major product, which was separated
from its a-epimer by flash chromatography to yield up
to 81% of stereochemically pure 7. Protection of the
hydroxyl group of compound 7 using p-methoxybenzyl
chloride and NaH and deprotection of the keto group
using 1 N HCl and acetic acid in tetrahydrofuran resul-
ted in the common intermediate 8.16 Wittig reaction of 8
with (methoxymethyl)triphenylphosphonium chloride
using dimsylsodium as a base followed by hydrolysis of
the resulting enol ether using hydrochloric acid in tet-
rahydrofuran (THF) afforded the b-aldehyde 9 as the
major product.17 Even though the a-epimer was formed
as a minor product, column chromatographic separation
gave a high yield of the pure epimer 9.
between the angular methyl and the aldehyde protons
clearly indicates the b-orientation of the formyl group.
This stereochemical assignment was supported by
molecular modeling studies.18 The distance between the
angular methyl and the formyl group hydrogens in the
b-epimer 9 was calculated to be as close as 2.04 A, while
the minimum distance between these two hydrogens in
the a-epimer was calculated to be 4.5 A.
The synthesis of the target compounds 2a–c began with
the Wittig reaction of 9 with the respective triphenyl-
phosphonium salts19 using dimsylsodium as a base.20
Thus, reaction of the aldehyde 9 with (p-carbomethoxy-
benzyl)triphenylphosphonium bromide gave the trans
compound 10 as the major product, which was easily
separable from its cis-isomer (minor product) by flash
chromatography. Deprotection of the benzyl protecting
group of 10 using 2,3-dichloro-5,6-dicyano-1,4-benzoqui-
none (DDQ)21 gave compound 11, which on hydrolysis
using NaOH yielded the compound 2a as the final product.
Reaction of 9 with (m-cyanobenzyl)triphenylphos-
phonium bromide yielded 12 as a cis–trans inseparable
1
mixture (1:3 ratio, H NMR). Therefore, the mixture
was subjected to deprotection of the hydroxyl group
with DDQ to give 13. Hydrolysis of 13 using KOH in
THF and methanol under reflux yielded the hydroxy
acid (85%) along with the corresponding primary amide
The stereochemistry of the aldehyde 9 was confirmed by
2-D NMR spectroscopic analysis. NOESY correlation

H. Assefa et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1619–1623
1621
Scheme 3. Reagents and conditions: (a) NaH/DMSO, (m-cyanobenzyl)triphenylphosphonium bromide;(b) DDQ/CH 2Cl2/H2O;(c) aq KOH/
CH3OH/THF, Á;(d) NaH/DMSO, ( o-cyanobenzyl)triphenylphosphonium bromide;(e) aq KOH/ethylene glycol, Á.
Scheme 4. Reagents and conditions: (a) LDA, methyl p-(bromomethyl)benzoate/THF;(b) NaBH 4/C2H5OH;(c) NaH, CS 2, CH3I/THF;(d) n-
Bu3SnH/toluene, Á;(e) aq NaOH/CH 3OH/THF;(f) H 2/Pd–C.
Scheme 5. Reagents and conditions: (a) LDA, m-cyanobenzyl bromide or o-cyanobenzyl bromide/THF;(b) NaBH 4/C2H5OH;(c) NaH, CS 2, CH3I/
THF;(d) n-Bu3SnH/toluene, Á;(e) aq KOH/ethylene glycol, Á;(f) H /Pd–C.
2
(15%). The pure trans product 2b was obtained by
crystallization from methanol. Wittig reaction of 9 with
(o-cyanobenzyl)triphenylphosphonium bromide gave
product. Cleavage of the p-methoxybenzyl group of 14
using DDQ afforded the alcohol 15. Recrystallization
afforded the pure trans-isomer, which was subjected to
hydrolysis to give the final compound 2c.
1
the trans-product 14 (>95%, H NMR) as the major

1622
H. Assefa et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1619–1623
The synthesis of 3a–c was initiated by alkylation of 8
using LDA and benzyl bromides in THF.22 Therefore,
alkylation of 8 with methyl p-(bromomethyl)benzoate
afforded the a-epimer 16 as the major product. Recrys-
tallization from hexane afforded the pure a-epimer 16.
Compound 16 was also obtained by stirring with
K2CO3 in methanol probably due to the epimerization
of the b-epimer (axial benzyl substituent) to the more
stable a-epimer (equatorial benzyl substituent). The ste-
reochemistry of 16 was assigned by 2-D NMR analysis
in which the NOESY correlation between the angular
methyl and the proton alpha to the carbonyl group was
considered as a clear indication of the a-orientation of
the benzyl substituent. The keto group was successfully
converted to the corresponding methylene group by
reducing the ketone to the corresponding alcohol using
NaBH4 and removing the resultant hydroxyl group
using Barton deoxygenation protocol.23 Hydrolysis of
the ester group of 17 by stirring with aqueous NaOH in
methanol and THF afforded 18. Then the final product
3a was obtained by hydrogenolysis of the p-methoxy-
benzyl group of 18 using Pd–C as a catalyst.
Table 1. Classical pathway complement inhibition and cytotoxicity
assays of oleanolic acid and its partial analogues
Compd
Complement
inhibition IC50 (mMa)
Cytotoxicity
IC50 (mM)
T.I.b
1
72.3 (Æ5.8)
na
112
na
1.55
—
0.6
—
0.5
0.6
—
0.6
0.4
—
11
2a
15
2b
2c
18
3a
21b
21c
3b
3c
623 (Æ34)
380
155
269
380
200
378
180
166
333
484
na
550 (Æ86)
633 (Æ83)
na
610 (Æ9)
488 (Æ11)
na
601 (Æ51)
0.6
0.8
616 (Æ66)
aValues are means of three experiments, standard deviation is given in
parentheses (na=not active).
bIn vitro therapeutic index (IC50 cytotoxicity/IC50 complement
inhibition).
compound 21b retains the activity may indicate that the
meta position of the carboxylic group is more favorable
for complement inhibitory activity.
Alkylation of 8 with m-cyanobenzyl bromide or o-cya-
nobenzyl bromide afforded the a-epimer of compounds
19b and 19c as major products. The pure epimer (a-
epimer) of each compound was obtained by stirring the
mixture with K2CO3 in methanol. The relative stereo-
chemistry was determined using the same method
employed for the stereochemical assignment of com-
pound 16. Molecular modeling studies using the same
program and methods described for 9 indicate that the
distance between the angular methyl hydrogen and the
hydrogen alpha to the carbonyl group is as close as 2.01
A. The carbonyl group of 19b and 19c was reduced to
the corresponding methylene group using the protocol
described for the conversion of compound 16 to 17. The
target compounds 3b and 3c were obtained by hydro-
lysis of the cyano group of 20b and 20c using aq KOH
in ethylene glycol under reflux, followed by hydro-
genolysis of the p-methoxybenzyl group.
All the target compounds 2a–c and 3a–c as well as the
intermediates 15, 18, 21b, and 21c showed cytotoxic
activity. Cytotoxicity and complement inhibitory activ-
ity appear to be dissociated as indicated by compounds
15, 18, and 21c. Although these partial analogues
showed moderate complement inhibitory potency with
low value of T.I., the flexible synthetic route developed
should allow easy access to various analogues for fur-
ther structure–activity relationship studies. It may also
be possible to enhance the cytotoxicity to useful levels
should the mechanism of cytotoxicity prove novel.
Acknowledgements
This work was supported in part by the United States
Department of Agriculture, Agriculture Research Ser-
vice Specific Cooperative Agreement No. 58-6408-7-012.
The compounds were bioassayed for the classical path-
way complement inhibitory activity in vitro following
the protocol described earlier.24 The cytotoxic property
of the compounds was assessed in a human malignant
melanoma cell line, SK-MEL. Solution of the com-
pounds were incubated with 25,000 SK-MEL cells/well
for 72 h. The number of remaining viable cells was
assessed using the supravital dye, neutral red.25 Briefly,
cells were washed with saline, incubated for 1.5 h with a
0.17% solution of neutral red in serum-free RPMI, and
washed again to remove extracellular dye. Following
solubilization with 0.04 N HCl in isopropanol, absor-
bance was read at 490 nM.
References and Notes
1. Complement, 2nd ed.;Law, S. K. A., Reid, K. B. M., Eds.;
IRL: New York, 1995.
2. Dodds, A. Biochemist 1998, 20, 18.
3. Taylor, P.;Botto, M.;Walport, M.
R259.
Curr. Biol. 1998, 8,
4. Carroll, M. C. Semin. Immunol. 1998, 10, 279.
5. Robbins, R. A.;Russ, W. D.;Rasmussen, J. K.;Clayton,
M. M. Am. Rev. Respir. Dis. 1987, 135, 651.
6. Kemp, P. A.;Spragg, J. H.;Brown, J. C.;Morgan, B. P.;
Gunn, C. A.;Taylor, P. W. J. Clin. Lab. Immunol. 1992, 37,
147.
All the target compounds 2a–c and 3a–c, and the inter-
mediate compound 21b have shown moderate comple-
ment inhibitory potency (Table 1). The lack of
complement inhibitory activity with compounds 11 and
15 shows the importance of the free carboxylic group
for complement inhibition. The lack of complement
inhibitory activity with compounds 18 and 21c while
7. Homeister, J. W.;Lucchest, B. R. Annu. Rev. Pharmacol.
Toxicol. 1994, 34, 17.
8. Rogers, J.;Griffin, W. S. T. In
Neuroinflammation:
Mechanisms and Management;Wood, P. L., Ed.;Humana:
Totowa, 1998;p 177.
9. Hagmann, W. K.;Sindelar, R. D. Annu. Rep. Med. Chem.
1992, 27, 199.

H. Assefa et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1619–1623
1623
10. Makrides, S. C. Pharmacol. Rev. 1998, 50, 59.
11. Knaus, U.;Wagner, H. Phytomedicine 1996, 3, 77.
12. Kapil, A.;Sharma, S. J. Pharm. Pharmacol. 1994, 46, 922.
13. Kapil, A.;Sharma, S. J. Pharm. Pharmacol. 1995, 47, 585.
14. Assefa, H.;Nimrod, A.;Walker, L.;Sindelar, R. Bioorg.
Med. Chem. Lett. 1999, 9, 1889.
15. Hagiwara, H.;Uda, H. J. Org. Chem. 1988, 53, 2308.
16. Snitman, D. L.;Tsai, M.-Y.;Watt, D. S. Synth. Commun.
1978, 8, 195.
19. The triphosphonium salt was synthesized by refluxing the
respective substituted benzyl bromide with triphenylphosphine
in benzene.
20. Greenwald, R.;Chaykovsky, M.;Corey, E. J.
Chem. 1963, 28, 1128.
J. Org.
21. Horita, K.;Yoshioka, T.;Tanaka, T.;Oikawa, Y.;Yone-
mitsu, O. Tetrahedron 1986, 42, 3021.
22. Grieco, P. A.;Ferrino, S.;Oguri, T. J. Org. Chem. 1979,
44, 2593.
17. Thompson, T. N.;Sierra, M. G.;Mcchesney, J. D. J. Org.
Chem. 1985, 50, 4447.
18. SYBYL1 6.4 Molecular Modeling Software Manual.
Tripos Inc., 1699 Hanley Road, St. Louis, MO 63144-2913,
USA.
23. Tice, C. M.;Heathcock, C. H. J. Org. Chem. 1981, 46, 9.
24. Srivastava, R. P.;Zhu, X.;Walker, L. A.;Sindelar, R. D.
Bioorg. Med. Chem. Lett. 1995, 5, 2429.
25. Borenfreund, E.;Babich, H.;Martin-Alguacil, N. Toxic.
In Vitro 1988, 2, 1.