Novel phenolic antioxidants as multifunctional inhibitors of inducible VCAM-1 expression for use in atherosclerosis

Article abstract of DOI:10.1016/S0960-894X(02)00516-4

A series of novel phenolic compounds has been discovered as potent inhibitors of TNF-α-inducible expression of vascular cell adhesion molecule-1 (VCAM-1) with concurrent antioxidant and lipid-modulating properties. Optimization of these multifunctional agents led to the identification of 3a (AGI-1067) as a clinical candidate with demonstrated efficacies in animal models of atherosclerosis and hyperlipidemia.

Full text of DOI:10.1016/S0960-894X(02)00516-4

Bioorganic & Medicinal Chemistry Letters 12 (2002) 2545–2548
Novel Phenolic Antioxidants as Multifunctional Inhibitors of
Inducible VCAM-1 Expression for Use in Atherosclerosis
Charles Q. Meng,* Patricia K. Somers, Carolyn L. Rachita, Lisa A. Holt, Lee K. Hoong,
X. Sharon Zheng, Jacob E. Simpson, Russell R. Hill, Lynda K. Olliff, Charles Kunsch,
y
Cynthia L. Sundell, Sampath Parthasarathy, Uday Saxena, James A. Sikorski
and Martin A. Wasserman
AtheroGenics, Inc., 8995 Westside Parkway, Alpharetta, GA 30004, USA
Received 19 February 2002; accepted 25 June 2002
Abstract—A series of novel phenolic compounds has been discovered as potent inhibitors of TNF-a-inducible expression of vas-
cular cell adhesion molecule-1 (VCAM-1) with concurrent antioxidant and lipid-modulating properties. Optimization of these
multifunctional agents led to the identification of 3a (AGI-1067) as a clinical candidate with demonstrated efficacies in animal
models of atherosclerosis and hyperlipidemia. # 2002 Elsevier Science Ltd. All rights reserved.
6
Coronary artery disease (CAD), primarily as a result of
atherosclerosis, remains the leading cause of death in
chemoattractant protein-1 (MCP-1) that help propa-
gate the formation of mature atherosclerotic plaques.
Cytokines, such as TNF-a and IL-1b, in turn further
induce the expression of inflammatory response genes
1
,2
industrialized countries. Atherosclerosis
is a multi-
faceted disease characterized by vascular inflammation,
deposition of lipids in the arterial vessel wall, and
smooth muscle cell proliferation resulting in a narrow-
ing of the vessel passages. While drugs that target single
mechanisms (e.g., the cholesterol-lowering statins) have
proven to be clinically beneficial for the treatment and
7
including VCAM-1 and MCP-1. A recent study
showed that VCAM-1 but not intercellular adhesion
molecule-1 (ICAM-1) plays a dominant role in the
initiation of atherosclerosis, although both VCAM-1
and ICAM-1 are upregulated in atherosclerotic lesions.
Inhibitors that selectively target VCAM-1 expression
are likely to have potential as therapeutic agents.
8
3
prevention of CAD, there is an ongoing need for
improved drugs that could more effectively address the
4
multiple underlying pathological mechanisms of CAD.
In this Letter, we report our successful efforts to com-
bine anti-inflammatory, lipid-modulating, and anti-
oxidant properties into single molecules as potential
treatments for CAD.
A variety of agents have been reported as inhibitors of
VCAM-1 expression or VCAM-1 activity/function.
9ꢀ17
Establishing pharmacological proof-of-concept for
VCAM-1 modulation as a therapeutic target, a cyclic
depsipeptide effectively inhibited VCAM-1 expression
and reduced inflammation in a dermal model of inflam-
As early atherosclerotic lesions form, localized endo-
thelial expression of vascular cell adhesion molecule-1
1
1
mation, a monoclonal antibody against VCAM-1
inhibited neointimal formation in a mouse model of
(
VCAM-1) leads to the recruitment of monocytes to
4
,5
12
adhere to the developing lesion. Subsequent conver-
sion of monocytes to foamy macrophages results in the
synthesis of a wide variety of inflammatory cytokines,
growth factors, and chemoattractants such as monocyte
arterial wall injury, and a disubstituted 1,4-diazepine
diminished the increase in paw thickness in a mouse
1
7
model of collagen-induced arthritis.
Oxidative stress is also an important contributor to
1
8
atherosclerosis. In particular, the facile uptake of oxi-
19
*Corresponding author. Tel.:+1-678-336-2540; fax:+1-678-336-2501;
e-mail: cmeng@atherogenics.com
Department of Gynecology and Obstretics, Emory University School
of Medicine, Atlanta, GA 30322, USA.
dized low-density lipoprotein (LDL) by macrophages
helps convert the recruited monocytes to lipid-engorged
macrophage foam cells that form the underlying fatty
y
0
960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00516-4

2546
C. Q. Meng et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2545–2548
streak of atherosclerosis. Thus, anti-inflammatory com-
pounds that selectively block inducible VCAM-1
expression and also exhibit antioxidant properties
should have high potential as anti-atherosclerotic agents.
Table 1. Inhibition of TNF-a-inducible VCAM-1 expression by car-
boxy-substituted esters
33
Probucol (1) is a modest lipid-lowering agent with
0,21
powerful antioxidant properties2
inhibits the oxidative modification of LDL indepen-
that effectively
Compd
R
VCAM-1 IC50 (mM)
2
2
dently of its lipid-lowering effect. The absorption of
probucol after oral administration is limited and vari-
3a
3b
3c
6
2
1
able, probably due to its high lipophilicity. While
probucol has demonstrated a reduction in the incidence
2
3
of restenosis in patients, its clinical use in CAD or any
other indications is severely limited by observations that
it progressively lowers the levels of high-density lipo-
>50
2
0
protein (HDL) and causes significant prolongation of
2
20
4,25
the QTc interval. In addition, several potentially
toxic metabolites of probucol have been identified that
2
1
arise from the highly reactive spiroquinone 2.
3
3
3
3
3
3
3
d
e
f
17
9
25
37
100
45
40
It is well known that the strong antioxidant properties
of probucol arise from the two phenol groups flanked
2
1
by tertiary-butyl moieties. We reasoned that suitably
modified monosubstituted compounds 3 might retain
the beneficial antioxidant and lipid-lowering properties
of probucol through the remaining phenol group but
could have an improved safety profile based on their
inability to form the spiroquinone metabolite 2. We also
discovered that, in contrast to 1, several of these mono-
substituted compounds 3 were unexpectedly potent and
selective inhibitors of TNF-a-inducible VCAM-1
expression on endothelial cells, with demonstrated effi-
cacy as anti-inflammatory agents in several preclinical
g
h
i
j
2
6,27
animal models.
The reaction of 1 with either a suitable acid chloride or
anhydride in the presence of base produces a mixture of
monosubstituted 3 and disubstituted 4, that are readily
separable by typical column chromatography procedures.
As exemplified in Scheme 1, treating 1 with succinic
anhydride in the presence of sodium hydride produced
acts as a chemical antioxidant and is as potent as pro-
bucol. Both 3a and 1 showed comparable inhibitory
effects (ꢁ85% inhibition at 70 mM) in preventing the
oxidation of linoleic acid by 15-lipoxygenase at the same
molar concentrations in an LMB (N-benzoylleucomethy-
lene blue) assay. Based on these highly encouraging
results, additional analogues were evaluated as potential
inhibitors of inducible VCAM-1 expression to further
explore the structure–activity relationships of this series
(Table 1).
2
8,29
the desired monosubstituted compound 3a
yield after chromatography.
in 28%
3
1
As shown in Table 1, 3a is a surprisingly potent inhi-
bitor of TNF-a-inducible VCAM-1 expression with an
IC50 of 6 mM. In contrast, neither probucol (1) (IC >
50
3
0
5
0 mM) as previously reported, nor the disubstituted
ester analogue 4a (IC >50 mM) significantly inhibited
The potency of 3a as an inhibitor of VCAM-1 expres-
sion drops slightly when the side chain is extended by
one methylene unit as in 3e and more dramatically when
either extended by 4 methylene units as in 3i or shor-
tened to the oxalate 3b. Such variations in homologues
have been observed in other series and can be related to
lipophilicities of the individual compounds.3 More
extensive modifications of the side chain in either 3a or
5
0
the expression of TNF-a-inducible VCAM-1. Although
a has only one instead of two phenol groups, it still
3
2
3e through the introduction of polar substituents (3c and
3f), unsaturation (3d), multiple fluorine atoms (3h), or
Scheme 1.

C. Q. Meng et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2545–2548
2547
replacement of one methylene unit by an oxygen as in 3g
all reduced their ability to inhibit VCAM-1 expression.
Therefore, compound 3a probably possesses the optimal
side chain length for the inhibition of VCAM-1 expres-
sion. Interestingly, 3j, the corresponding methyl ester of
it as a prototype for additional testing in order to better
define its selectivity and potential for in vivo efficacy. In
contrast to its potent inhibition of VCAM-1 expression,
3a showed no effect on inducible ICAM-1 expression at
concentrations up to 10 mM and exhibited reduced
potency for inhibiting TNF-a-inducible E-selectin
(IC =25 mM) expression. However, 3a was a potent
3
a, is significantly weaker as a VCAM-1 inhibitor,
thereby further corroborating the important contribution
of the free carboxy group in 3a to its VCAM-1 modulat-
ing activity. See below for further discussion. Compound
5
0
inhibitor of inducible MCP-1 expression (IC =10 mM).
5
0
MCP-1 is also highly expressed in human atherosclerotic
3
4
3
f was prepared by acylation of 1 with (R)-5-oxotetrahy-
lesions. Also, 3a inhibited the proliferation of human
aortic smooth muscle cells (ꢁ50% inhibition at 5 mM),
whereas probucol exhibited no significant effect at con-
centrations up to 100 mM. Smooth muscle cell prolifera-
tion contributes to late-stage formation of atherosclerosis.
drofuran-2-carbonyl chloride in the presence of base fol-
lowed by hydrolysis of the resulting intermediate lactone.
The carboxylic groups of the compounds in Table 1
could all be selectively reduced to a primary alcohol
with borane in THF. As shown in Table 2 , the resulting
monohydroxyesters 3k–3o are all generally less potent
than the corresponding carboxylic acids. However, the
potency of 3m (IC =40 mM) is increased 5-fold when a
Compound 3a is stable in solution (85:15 acetonitrile:
water) for weeks at room temperature, and 3a is also
stable to mild acid or base and can be stored in the solid
state without change for extended periods (years). This
extended stability is presumably attributable to the highly
hindered ester moiety in 3a. Preclinical animal studies and
human clinical trials also showed that 3a was not sig-
nificantly metabolized to probucol. For example, 3a was
the predominant component in plasma following oral
5
0
second hydroxyl group is introduced into its side chain
as in 3p (IC =8 mM). Compound 3p was prepared by
5
0
acylation of 1 with 2,3 - diacetoxy - 3 - chlorocarbonyl-
propionic acid methyl ester in the presence of sodium
hydride and subsequent DIBAL-H reduction.
1
4
administration of [ C]-3a in both the dog and rat. Fol-
1
4
Since there is no carboxylic group in 3p, the carboxylic
group in the compounds in Table 1 is clearly not abso-
lutely necessary for the inhibition of VCAM-1 expres-
sion. Rather, the carboxylic group in 3a or the dihydroxy
substituents in 3p likely bring these compounds into the
right lipophilicity range required for activity in the
VCAM-1 assay.
lowing administrating a single oral dose of [ C]-3a to
six healthy male volunteers, most of the radioactivity in
plasma (94.37–97.84%) was unchanged 3a. Thus, 3a is
not a prodrug of probucol, and the biological efficacy of
3a is completely attributable to the parent molecule.
In summary, 3a exhibits many of the in vitro properties
desirable in a molecule to treat atherosclerosis. It func-
tions as a potent antioxidant, selectively and potently
inhibits inducible VCAM-1 and MCP-1 expressions,
and inhibits human aortic smooth muscle cell prolifera-
The ability of 3a to inhibit TNF-a-inducible VCAM-1
expression at low micromolar levels prompted us to select
tion. Indeed, while details of its pharmacological prop-
26,27
Table 2. Inhibition of TNF-a-inducible VCAM-1 expression by
erties will be reported separately,
3a was well-
hydroxyl-substituted esters33
tolerated in animals and lowered LDL cholesterol in
mice, hamsters, rabbits and monkeys with neutral or
2
7
elevating effects on HDL levels after oral dosing.
Moreover, 3a also inhibited the progression of athero-
sclerosis in the aorta by 94% after oral dosing in cho-
lesterol-fed rabbits and 43–66% in LDL receptor-
Compd
R
VCAM-1 IC50 (mM)
2
6,27
knockout mice.
1
Based on these results, 3a (AGI-
067) was selected for clinical evaluation. Recently,
3
3
3
3
3
3
k
30
Phase II studies have demonstrated that AGI-1067 has
beneficial effects on the prevention of restenosis in
2
5
patients without prolongation of the QTc interval.
l
30
40
110
55
8
Acknowledgements
m
n
o
p
The authors thank Drs. Richard F. Arrendale and
Bruce M. Johnson for providing the chemical and
metabolic stability data for 3a and fruitful discussions.
References and Notes
1. McGill, H. C. In Atherosclerosis and Coronary Artery Dis-
ease; Fuster, V., Ross, R., Topol, E. J., Eds.; Lippincott-
Raven: Philadelphia, 1996; pp 25–41.

2548
C. Q. Meng et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2545–2548
2. Ross, R. N. Engl. J. Med. 1999, 340, 115.
3. Auer, J.; Eber, B. Curr. Opin. Invest. Drugs 2001, 2, 382.
4. Krause, B. R.; Auerbach, B. J.; Bocan, T. M. A. Curr.
22. Kita, T.; Nagano, Y.; Yokode, M.; Ishii, K.; Kumi, N.;
Ooshima, A.; Yoshida, H.; Kawai, C. Proc. Natl. Acad. Sci.
U.S.A. 1987, 84, 7725.
Med. Chem.: Immunol., Endocr. Metab. Agents 2001, 1, 39.
. Orlandi, Q.; Loscalzo, J. In Adhes. Mol. Health Dis.; Paul,
L. C., Issekutz, T. B., Eds. Dekker: New York, NY, 1997; pp
23. Tardif, J. C.; Cote, G.; Lesperance, J.; Bourassa, M.;
Lambert, J.; Doucet, S.; Bilodeau, L.; Nattel, S.; De Guise, P.
New Engl. J. Med. 1997, 337, 365.
5
4
6
7
01–425.
. Harrington, J. R. Stem Cells 2000, 18, 65.
. Davies, M. J.; Gordon, J. L.; Gearing, A. J.; Pigott, R.;
24. Klein, L. Arch. Intern. Med. 1981, 141, 1102.
25. Tardif, J. C.; Gregoire, J.; Schwartz, L.; Laramee, L.;
Title, L.; Reeves, F.; Lambert, J.; Guertin, M. C.; Lesperance,
J. Circulation; 104 (Suppl. II) 2001, 11, 544.
Woolf, N.; Katz, D.; Kyriakopoulos, A. J. Pathol. 1993, 171,
2
8
23.
. Cybulsky, M. I.; Iiyama, K.; Li, H.; Zhu, S.; Chen, M.;
26. Sundell, C. L.; Daugherty, A.; Stalvey, A. L.; Hammes, P.;
Landers, L. K.; Medford R. M.; Saxena, U. Circulation 1999,
100, pp 1–42.
27. Sundell, C. L.; Stalvey, A. L.; Hammes, P.; Landers, L.
K.; Somers, P. K.; Suen, K. L.; Meng, C. Q.; Hoong, L. K.;
Hill, R. R.; Zheng, X. S.; Simpson, J. E.; Rachita, C. L.;
Medford, R. M.; Saxena, U. Arterioscler. Thromb. Vasc. Biol.
manuscript submitted and personal communications (Sundell,
C. L.).
Iiyama, M.; Davis, V.; Gutierrez-Ramos, J. C.; Connelly,
P. W.; Milstone, D. S. J. Clin. Invest. 2001, 107, 1255.
9
. Bennett, C. F.; Condon, T. P.; Grimm, S.; Chan, H.;
Chiang, M. Y. J. Immunol. 1994, 152, 3530.
0. Hommel, U.; Weber, H. P.; Oberer, L.; Naegeli, H. U.;
Oberhauser, B.; Foster, C. A. FEBS Lett. 1996, 379, 69.
1. Foster, C. A.; Besemer, J.; Meingassner, J. G.; Naegeli,
H. U.; Schon, G.; Bevec, D.; Brend, T. Skin Pharmacol. 1996,
1
1
28. TLC (silica gel, 60:40 ether/hexane + 10 drops HOAc,
Rf=0.35); Flash Column Chromatography (silica gel, 70:30
9
1
, 149.
2. Oguchi, S.; Dimayuga, P.; Zhu, J.; Chyu, K.-Y.; Yano, J.;
1
hexane:ether to 100% ether); H NMR (CDCl
7.61 (s, 2H), 7.43 (s, 2H), 5.38 (s, 1H), 2.97 (t, J=6.8 Hz, 2H),
3
, 400 MHz) d
Shah, P. K.; Nilsson, J.; Cercek, B. Arterioscler. Thromb.
Vasc. Biol. 2000, 20, 1729.
1
G. C. N.; Mackman, N. FEBS Lett. 1996, 382, 323.
2.76 (t, J=6.8 Hz, 2H), 1.45 (s, 8H), 1.42 (s, 16H), 1.32 (s,
ꢂ
3. Cobb, R. R.; Felts, K. A.; McKenzie, T. C.; Parry,
18H); white to off-white solid, mp 132–134 C.
29. Somers, P. K. U.S. Patent 6 121 319, 2000.
30. Zapolska-Downar, D.; Zapolski-Downar, A.; Mar-
kiewski, M.; Ciechanowicz, A.; Kaczmarczyk, M.; Nar-
uszewicz, M. Atherosclerosis 2001, 155, 123.
31. Somers, P. K.; Medford, R. M.; Saxena, U. Free Radic.
Biol. Med. 2000, 28, 1532.
32. Wermuth, C. G. In The Practice of Medicinal Chem-
istry; Wermuth, C. G., Ed.; Academic: London, 1996; pp 181–
202.
14. Sakai, A. Inflammation Res. 1996, 45, 224.
15. Dodd, D. S.; Shen, Z.; Nishi, T.; Graber, N.; Bealls, D.;
Fong, M.; Ebert, T. Bioorg. Med. Chem. Lett. 1996, 6, 2693.
6. Meng, C. Q.; Zheng, X. S.; Holt, L. A.; Hoong, L. K.;
1
Somers, P. K.; Hill, R. R.; Saxena, U. Bioorg. Med. Chem.
Lett. 2001, 11, 1823.
17. Nakao, H.; Doi, T.; Suda, M.; Umetani, M.; Ohtaka, M.;
Shiratsuchi, M.; Kodama, T. J. Atheroscler. Thromb. 1998, 4, 149.
1
1
8. Kunsch, C.; Medford, R. M. Circ. Res. 1999, 85, 753.
9. Steinbrechner, U. P.; Parthasarathy, S.; Leake, D. S.;
33. Human aortic endothelial cells (90–95% confluent) are
incubated with compounds in DMSO and 100 U/mL TNF-a
overnight. Upon washing the cells are incubated with anti-
VCAM-1 monoclonal antibody and then an anti-mouse IgG/
HRP-conjugate. Color is developed using TMB substrate and
the reaction is stopped with 2 N sulfuric acid. Spectro-
photometric readings are then taken at wavelength 450 nm.
For a more detailed procedure, see ref 16.
Witztum, J. L.; Steinberg, D. Proc. Natl. Acad. Sci. U.S.A.
1
984, 81, 3883.
0. Meng, C. Q. Curr. Opin. Cardiovasc. Pulm. Renal Invest.
2
Drugs 2000, 2, 294.
1. Barnhart, J. W.; Wagner, E. R.; Jackson, R. L. In Anti-
2
lipidemic Drugs: Medicinal, Chemical, and Biochemical Aspects;
Witiak, D. T., Newman, H. A. I., Feller, D. R., Eds.; Elsevier
Science: Amsterdam, 1991; pp 277–299.
34. Reape, T. J.; Groot, P. H. Atherosclerosis 1999, 147,
213.