Discovery of novel phenolic antioxidants as inhibitors of vascular cell adhesion molecule-1 expression for use in chronic inflammatory diseases

Article abstract of DOI:10.1021/jm049685u

Vascular cell adhesion molecule-1 (VCAM-1) mediates recruitment of leukocytes to endothelial cells and is implicated in many inflammatory conditions. Since part of the signal transduction pathway that regulates the activation of VCAM-1 expression is redox

Full text of DOI:10.1021/jm049685u

6420
J. Med. Chem. 2004, 47, 6420-6432
Discovery of Novel Phenolic Antioxidants as Inhibitors of Vascular Cell
Adhesion Molecule-1 Expression for Use in Chronic Inflammatory Diseases
Charles Q. Meng,* Patricia K. Somers, Lee K. Hoong, X. Sharon Zheng, Zhihong Ye, Kimberly J. Worsencroft,
Jacob E. Simpson, Martha R. Hotema, M. David Weingarten, Mathew L. MacDonald, Russell R. Hill,
Elaine M. Marino, Ki-Ling Suen, Jayraz Luchoomun, Charles Kunsch, Laura K. Landers,
Dimitria Stefanopoulos, Randy B. Howard, Cynthia L. Sundell, Uday Saxena, Martin A. Wasserman, and
James A. Sikorski
AtheroGenics, Inc., 8995 Westside Parkway, Alpharetta, Georgia 30004
Received April 29, 2004
Vascular cell adhesion molecule-1 (VCAM-1) mediates recruitment of leukocytes to endothelial
cells and is implicated in many inflammatory conditions. Since part of the signal transduction
pathway that regulates the activation of VCAM-1 expression is redox-sensitive, compounds
with antioxidant properties may have inhibitory effects on VCAM-1 expression. Novel phenolic
compounds have been designed and synthesized starting from probucol (1). Many of these
compounds demonstrated potent inhibitory effects on cytokine-induced VCAM-1 expression
and displayed potent antioxidant effects in vitro. Some of these derivatives (4o, 4p, 4w, and
4x) inhibited lipopolysaccharide (LPS)-induced secretion of pro-inflammatory cytokines such
as tumor necrosis factor-R (TNF-R), interleukin-1â (IL-1â), and IL-6 from human peripheral
blood mononuclear cells (hPBMCs) in a concentration-dependent manner in vitro and showed
antiinflammatory effects in an animal model. Compounds 4ad and 4ae are currently in clinical
trials for the treatment of rheumatoid arthritis (RA) and prevention of chronic organ transplant
rejection, respectively.
Introduction
probucol that arise from the highly reactive spiro-
quinone 2 have been identified.7
A key regulator of leukocyte trafficking to sites of
inflammation, VCAM-1 has been implicated in numer-
ous inflammatory diseases. It may be the predominant
adhesive molecule responsible for migration and adher-
ence of leukocytes to inflamed synovium in rheumatoid
arthritis (RA), and soluble VCAM-1 levels have been
correlated with clinical severity and progression of RA.1,2
Increased expression of endothelial adhesion molecules
including VCAM-1 has been observed during human
renal, cardiac, and liver transplant rejection, and mono-
clonal antibodies directed against VCAM-1 have been
shown to prolong allograft survival in animal models.3
It has been established that the activation of VCAM-1
expression on endothelial cells is partly regulated by a
redox signal transduction pathway that is sensitive to
inhibition by antioxidants.4,5 In light of the role of
antioxidants in the inhibition of VCAM-1 expression,
we set out to look into known antioxidants for a starting
point to discover novel inhibitors of VCAM-1 expression
for use in chronic inflammation.
Probucol (1) is a powerful antioxidant with modest
lipid-lowering properties.6,7 It effectively inhibits the
oxidative modification of low-density lipoprotein (LDL)
independently of its lipid-lowering effect.8 While probu-
col has demonstrated a reduction in incidence of rest-
enosis in patients,9 its clinical use is severely limited
by observations that it progressively lowers the levels
of HDL,6 causes significant prolongation of the QTc
interval in some patients,10 and has limited oral bio-
availability.7 Several potentially toxic metabolites of
It is well-known that the strong antioxidant effects
of probucol stem from the phenol groups flanked by tert-
butyl moieties.7 Since formation of spiroquinone 2 would
require oxidation of both phenol groups of probucol, we
reasoned that probucol derivatives suitably monosub-
stituted at one of the two phenol groups might retain
the beneficial antioxidant properties (which might be
required for inhibition of VCAM-1 expression) through
the remaining phenol group, but could have an im-
proved safety profile based on their inability to form the
spiroquinone metabolite 2. We have recently reported11
our findings with probucol monoesters. A lead compound
from that series, AGI-1067 (3), exhibited many proper-
ties desirable in a molecule to treat atherosclerosis.
Compound 3 potently inhibited cytokine-induced endo-
thelial expression of VCAM-1 and MCP-1 and smooth
muscle cell proliferation in vitro, as well as progression
of atherosclerosis in experimental animals. In clinical
trials, 3 did not cause QTc prolongation, while probucol
(1) did.12
Herein we disclose the discovery of novel probucol
monoether derivatives 4 as potent inhibitors of VCAM-1
expression with antiinflammatory effects, some of which
also have lipid-modulating properties. The ether linkage
of this series of compounds renders better chemical
* Correspondence: 678-336-2540 (Tel.); 678-336-2501 (Fax); cmeng@
atherogenics.com.
10.1021/jm049685u CCC: $27.50 © 2004 American Chemical Society
Published on Web 10/30/2004

Phenolic Antioxidants as Inhibitors of VCAM-1 Expression
Journal of Medicinal Chemistry, 2004, Vol. 47, No. 25 6421
Scheme 1
stability than the monoester series and, therefore,
enables the synthesis of compounds with a wider variety
of functional groups and probably also improves overall
in vivo efficacy of the compounds due to potential better
metabolic stability.
drochloride (EDCI) gave 4ah and subsequent hydrolysis
yielded 4ai. Compounds 4aj, 4ak, 4al, and 4am (Table
1) were obtained in a similar manner, with or without
a deprotection step. Compound 4an was prepared by
treating probucol (1) with iodoacetonirile in the presence
of potassium fluoride and compound 4ao derived from
treating 4an with hydroxylamine. Compound 5a (Scheme
4) was obtained as a side product from a large-scale
synthesis of 4ad.
Chemistry
The synthesis of all the compounds reported herein
started with alkylation of probucol (1) as shown in
Scheme 1. Dialkylated products (5) were observed in
almost all the alkylation reactions, but the desired
monoalkylated products (4) could be easily isolated
using routine column chromatography in moderate
isolated yields (up to 40%). As shown in Scheme 2, 1
was coupled with glycidol under Mitsunobu conditions13
to give epoxide 4a. Hydrolysis of 4a under basic condi-
tions in the presence of aqueous ethanol gave 4b and
4c. Alternatively, in the presence of acetic acid, acetates
4d and 4e were obtained from 4a. Aminolysis of 4a was
achieved with an amino acid to afford 4f or 4g (Scheme
2). Polyether derivative 4h was obtained in a similar
manner as 4a from 1. A Mitsunobu reaction of a
protected alcohol with probucol (1) gave 4i or 4j and
subsequent deprotection afforded compound 4k or 4l.
Coupling of probucol with a tartrate-derived diol14 gave
compound 4m. Compound 4n, the enantiomer of 4m,
was prepared in the same manner. Intriguingly, neither
dialkylated probucol nor diether of the diol was obtained
in significant amounts from this reaction. Upon hy-
drolysis of the protecting ortho ester, alditol 4o or 4p
(Table 1) was obtained in moderate yields. Similarly,
the coupling of probucol with D-ribonic γ-lactone ethyl
ortho ester15 (Scheme 2) gave compound 4q, deprotec-
tion of which produced diol 4r. Hydrolysis of 4r gave
carboxylic acid 4s. Reduction of 4s with lithium alumi-
num hydride (LAH) yielded lactol 4t and tetraol 4u.
Reduction of lactone 4r with LAH also led to 4t. In a
similar manner, probucol was coupled with 1,2,3,4-tetra-
O-acetyl-â-glucopyranose to give commpound 4v. Re-
moval of the acetyl groups of 4v produced lactol 4w.
Reduction of 4w with LAH afforded pentanol 4x.
Acrylic derivative 4y was obtained when probucol was
treated with ethyl propiolate (6) in the presence of
triethylamine (Scheme 3). Compound 4z (Table 1) was
generated as a side product from this reaction. Reduc-
tion of compound 4y with LAH yielded allylic alcohol
4aa. Alternatively, hydrolysis of 4y gave carboxylic acid
4ab.
Results and Discussion
As shown in Table 1, probucol (1) did not show any
inhibitory effect on cytokine-induced VCAM-1 expres-
sion in endothelial cells even at high concentration (100
µM). Simple ethers 4a and 4z and poly(ethylene glycol)
ether 4h did not show any inhibitory effects at 50 µM.
Compounds with one free hydroxy group on the side
chain (4b, 4aa, and 4k) showed weak potencies (IC50
>
50 µM). The only difference between 4b and 4d is the
extra carbonyl group on the side chain in the latter,
which probably helps increase the potency (from IC50
> 50 µM to IC50 ) 41 µM) by lowering the lipophilicity
of the compound. Compounds with two hydroxy groups
on the side chain (4c and 4l) in general showed better
potency (IC50 ) 13-30 µM) than compounds with only
one hydroxy group. A comparison of 4c with 4b reveals
that the removal of the ethyl group from 4b and
resulting decrease in lipophilicity enhances the potency
of the product (4c) as shown in Table 1. When a
methylene unit is inserted between the secondary
hydroxyl and the side chain of compound 4c, the
lipophilicity of the resulting 4l increases slightly (Figure
1), but the potency on inhibition of VCAM-1 expression
increases from IC50 ) 30 µM to 13 µM, rather than
decreasing as would be predicted from the above lipo-
philicity-potency relationship. Another difference be-
tween these two compounds is that 4l has two terminal
primary hydroxy groups while 4c has only one terminal
primary hydroxy group. The two regioisomers 4d and
4e showed slightly different inhibitory effects, the one
with a primary hydroxy group being more potent than
the one with a secondary hydroxy group. These results
indicate that other factors such as special configuration
may also affect the potency of compounds.
When a third hydroxy group is introduced to the side
chain of monosubstituted probucol derivatives, the
inhibitory potency on VCAM-1 expression is further
enhanced. Compound 4o, having one more hydroxy-
methylene unit than 4c, showed an IC50 value of 6 µM
as compared to 30 µM for 4c (Table 1). Interestingly,
there is a difference in potency between the two enan-
tiomers 4o and 4p. The R,R-isomer (4p) is 6-fold more
potent than the S,S-isomer (4o). This, together with the
regioisomeric selectivity discussed above, suggests a
well-defined molecular interaction of these inhibitors
with a yet undiscovered biological target.
As shown in Scheme 4, probucol (1) was alkylated
with ethyl iodoacetate (7) in the presence of potassium
fluoride on alumina. Hydrolysis of the product (4ac)
gave compound 4ad. Compounds 4ae and 4af (Table
1) were obtained in a similar manner from 1. This
method failed to produce 4ag (Table 1). Rather, 4ag was
prepared by treating probucol with â-propiolactone
under basic conditions as shown in Scheme 5.16 Cou-
pling of 4ad with â-alanine ethyl ester in the presence
of 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hy-
It seems that the introduction of four or more hydroxy
groups to the side chain of probucol ether derivatives

6422 Journal of Medicinal Chemistry, 2004, Vol. 47, No. 25
Meng et al.
Scheme 2a
a Reagents and conditions: (a) triphenylphosphine, diethyl azodicarboxylate, THF, reflux; (b) 1 N NaOH, EtOH, reflux; (c) AcOH,
THF, Et3N, reflux; (d) iminodiacetic acid or glycine, Et3N, reflux; (e) MeOH, K2CO3, rt; (f) Bu4NF, THF, rt; (g) 1. MeOH, AcOH, H2O,
reflux; 2. MeOH, K2CO3, rt; (h) NaOH, H2O, THF, rt; (i) LAH, THF, rt; (j) NaBH4, THF, rt.
decreases the potency on VCAM-1 expression. Com-
pound 4u, having one more hydroxy group on the side
chain than 4o or 4p, showed a higher IC50 value (13
µM). Although the stereochemistry of 4u on the side
chain is different from that of 4o or 4p, compound 4x,
having five hydroxy groups on the side chain and
showing an IC50 value of 15 µM, confirms that reduction
of lipophilicity beyond a certain point decreases the
inhibitory potency of the compounds. The hyperbolic
relationship between lipophilicity reflected by calculated
logP (ClogP) of these hydroxylated compounds and their
in vitro efficacy on inducible VCAM-1 expression is
depicted in Figure 1.
A cyclic carbohydrate-like residue on the side chain
of probucol derivatives is well tolerated and the trend
of potency depending on lipophilicity remains. Com-
pound 4r with two hydroxy groups and a lactone residue
on the side chain showed an IC50 value of 13 µM. When
the lactone in 4r is reduced to lactol 4t, the potency
increases to 8 µM because there are three hydroxy
groups on the side chain now. When the hydroxy groups
on the side chain of compound 4w are protected with
acetyl groups (compound 4v) the potency is almost
eliminated (Table 1), confirming the importance of
hydroxy groups in modulating the lipophilicity of com-
pounds.
(compounds 4ae and 4af), the potency decreases (IC50
) 11 and 23 µM, respectively). The correlation between
lipophilicity and efficacy of these closely related car-
boxylic compounds is depicted in Figure 2; compound
4ag is in the optimal range of lipophilicity for best
potency, compounds 4ad and 4ae are slightly off the
range, and 4af is too lipophilic to be very potent.
Compound 3, with the same length of side chain but an
extra carbonyl group and hence less lipophilicity com-
pared to 4ae, is more potent than the latter (Table 1).
This further supports the lipophilicity-potency hypoth-
esis discussed above. Although compound 4ag is the
most potent of the four probucol derivatives of simple
carboxylic acids, compounds 4ad and 4ae were consid-
ered for further exploration due to potential metabolic
instability of and in vivo production of acrylic acid from
4ag.
It is worth noting that when a double bond is
introduced to the side chain of compound 4ag, the
potency dropped from IC50 ) 7 µM to 30 µM (4ab). The
change of lipophilicity alone cannot explain such a
dramatic decrease in potency. Presumably, the intro-
duction of a double bond rendered configurational
changes in the molecule and, hence, altered the binding
pattern of the molecule to the biological target. There-
fore, the lipophilicity-potency relationship discussed
above is only valid among structurally similar com-
pounds and special configuration precedes lipophilicity
in relationship with potency of compounds. The fact that
the curves in Figure 1 and Figure 2 stay apart instead
of superimposing with each other confirms that the
lipophilicity-efficacy relationship reported herein is
only valid within closely related series of compounds.
The trend between potency and lipophilicity observed
with the hydroxylated probucol derivatives is also seen
with ether-linked carboxylic derivatives. Compound 4ad
showed an IC50 value of 10 µM (Table 1). When the side
chain of 4ad is extended by one methylene unit (com-
pound 4ag), the potency is increased (IC50 ) 7 µM).
However, when the side chain of 4ag is further extended

Phenolic Antioxidants as Inhibitors of VCAM-1 Expression
Journal of Medicinal Chemistry, 2004, Vol. 47, No. 25 6423
Table 1. Inhibiting Profile of Compounds on VCAM-1 Expression
a All IC50 numbers reflect an average of at least two determinations, and standard deviations are given for compounds with three or
more determinations; “NE50” and “NE100” stand for “no effect at 50 µM” and “no effect at 100 µM,” respectively.
Probucol does not belong to either of the series of Figure
1 or Figure 2 and, therefore, cannot be predicted by
either curve.
nal hydroxy group of compound 4u is oxidized to a
carboxylic acid (4s), the potency dropped from IC50
)
13 µM to 35 µM, presumably due to the decrease in
lipophilicity and changes in configuration. Moreover,
when 4ab or 4ai is esterified the potency is completely
lost (4y or 4ah) due to increases in lipophilicity. When
an amide group is inserted into the side chain of 4l, the
potency (compound 4am) increases from IC50 of 13 µM
to 6 µM, possibly due to the decrease in lipophilicity
brought in by the amide group. Compounds 4am and
4aj are both amides and equipotent, but one has a
When an amide group is introduced into the side
chain of compound 4ad, the potency remained (com-
pounds 4aj). Further addition of a methyl group to the
nitrogen (compound 4ak) and a methylene unit into the
side chain (compound 4ai) did not affect the potency
much (Table 1). This may mean that the optimal range
of lipophilicity for potency is not a very narrow one in
this subseries of probucol derivatives. When the termi-

6424 Journal of Medicinal Chemistry, 2004, Vol. 47, No. 25
Meng et al.
Scheme 3a
a Reagents and conditions: (a) Et3N, THF, reflux; (b) LAH, THF, rt; (c) LiOH, THF, H2O, rt.
Scheme 4a
a Reagents and conditions: (a) KF on alumina, DMF; (b) LiOH, THF, MeOH, H2O; (c) â-alanine ethyl ester hydrochloride, Et3N, EDCI,
CH2Cl2.
Scheme 5
terminal carboxy group and the other has two terminal
hydroxy groups. This again shows that neither a carboxy
nor a hydroxy group is essential for inhibitory activity
on VCAM-1 expression; rather, it is the correct range
of lipophilicity that matters.
Too much hydrophilicity is detrimental to the inhibi-
tory efficacy on VCAM-1 expression in the carboxylic
derivatives of probucol, as well. Compared to compound
4ae, compound 4s with three extra hydroxy groups on
the side chain is more hydrophilic and less potent in
the in vitro assay with an IC50 value of 35 µM (Table
1). Compound 4g has an IC50 value of 16 µM. When an
additional carboxymethyl group is introduced to the
molecule (4f), the potency is eliminated possibly due to
a decrease in lipophilicity. Similarly, compound 4al,
with an additional propionic group than 4aj, is dramati-
cally less potent than the latter (Table 1). Compound
4ad shows an IC50 of 10 µM on the inhibition of VCAM-1
expression. When the carboxy group of 4ad is replaced
with a cyano group or an N-hydroxyamidino group, the
potency is completely lost (compound 4an, no effect at
50 µM) or dramatically decreased (compound 4ao, IC50
) 26 µM), presumably due to changes of lipophilicity.
Figure 1. Correlation between calculated logP and in vitro
inhibitory effect on inducible VCAM-1 expression of hydrox-
ylated ether compounds.
In our hands probucol exhibited no effect on VCAM-1
expression at concentrations up to 100 µM. Others have
reported that probucol (1) in vivo reduced the level of
basal VCAM-1 expression and prevented its upregula-
tion.17 Since probucol is known to inhibit the oxidation
of LDL and oxidized LDL may induce VCAM-1 expres-

Phenolic Antioxidants as Inhibitors of VCAM-1 Expression
Journal of Medicinal Chemistry, 2004, Vol. 47, No. 25 6425
Table 3. Inhibitory Profile of Selected Compounds on
LPS-Induced Secretion of Cytokines from HPBMN Cellsa
compd TNF-R (IC50, µM) IL-1â (IC50, µM) IL-6 (IC50, µM)
1
>40
4.0 ( 1.0
>40
>40
1.7 ( 0.9
4o
4p
4w
4x
5a
7.7 ( 0.6
3.3 ( 0.3
1.9 ( 0.7
1.8 ( 0.3
8.0 ( 0.0
b
1.2 ( 0.3
1.6 ( 0.6
1.0 ( 0.3
2.2 ( 1.6
>10c
>10c
>10c
a All IC50 numbers reflect an average of at least three deter-
minations. b No dose response was observed, but the compound
was as potent as 4p at 10 µM. c Unable to test at higher
concentrations due to insolubility of the compound.
properties as discussed above, these results further
suggest the involvement of a redox pathway in VCAM-1
expression.
Cytokines are extracellular signaling proteins pro-
duced by many cell types playing a central role in
human immune response, and can be categorized as
either pro-inflammatory or antiinflammatory in action.
TNF-R, IL-1â and IL-6 are major pro-inflammatory
cytokines implicated in the pathogenesis of numerous
diseases. The expression of these pro-inflammatory
cytokines is also redox-regulated.19 Selected compounds
were tested for their inhibitory effects on the LPS-
induced secretion of these three cytokines from hPBMCs
in vitro. As shown in Table 3, compounds 4o, 4p, 4w,
and 4x exhibited potent inhibitory effects on the secre-
tion of TNF-R (IC50 ) 1.8-4.0 µM) and IL-6 (IC50 ) 1.0-
1.7 µM). On the inhibition of IL-1â secretion, compound
4x showed an IC50 value of 2.2 µM while compounds
4o and 4p had an IC50 value around 8 µM. Compound
4w did not show a dose response in the assay, but was
as potent as 4p on the inhibition of IL-1â secretion at
10 µM. Compounds 4o, 4p, 4w, and 4x have a free
phenol group in their structures, are antioxidants, and
are inhibitors of VCAM-1 expression as discussed above.
On the other hand, compound 5a does not have a free
phenol group, is not an antioxidant, and does not inhibit
VCAM-1 expression (Table 1). Similarly 5a did not show
a potent inhibitory effect on the secretion of any of the
three cytokines (Table 3). Therefore, the inhibitory
effects of the compounds on cytokine secretion could also
be related to their antioxidant effects. Probucol (1),
though a strong antioxidant, was completely ineffective
in inhibiting the secretion of any of the three cytokines
(Table 3). This further indicates that properly designed
monosubstituted probucol derivatives can enter cells or
have access to cell membranes and affect intracellular
signaling more readily than probucol itself. In support
of these contentions, it has been found that there is a
significant difference in cell uptake between probucol
(1) and 3.20
Figure 2. Correlation between calculated logP and in vitro
inhibitory effect on inducible VCAM-1 expression of ether-
linked carboxylic compounds.
Table 2. Antioxidant Profile of Selected Compounds against
Oxidation of N-Benzoyl Leucomethylene Bluea
compd
IC50 (µM)
1
14 ( 2
17 ( 2
15 ( 5
10 ( 2
12 ( 3
>125
4o
4p
4w
4x
5a
a A literature protocol was followed;18 all IC50 numbers reflect
an average of at least three determinations.
sion, the in vivo inhibitory effect of probucol on VCAM-1
expression could well be due to its inhibitory effect on
the oxidation of LDL rather than any direct effect on
VCAM-1 expression.17
Although they could be defined in various ways, many
antioxidants are inhibitors of lipid peroxidation. Se-
lected compounds from this program were assayed for
their ability to inhibit the hemoglobin-catalyzed oxida-
tion of N-benzoyl leucomethylene blue (LMB assay)18
by the peroxide generated from the oxidation of linoleic
acid by 15-lipoxygenase. As shown in Table 2, probucol
(1), compounds 4o, 4p, 4w, and 4x all showed similar
potencies in this assay. However, compound 5a, with
both of the phenol groups of probucol substituted, did
not show any antioxidant property in this assay, con-
firming the phenol group in these compounds as the
source of the antioxidant property. Interestingly, probu-
col, with two phenol groups in each molecule, is not
significantly more potent than its ether derivatives with
only one phenol group. This may mean that after one
of the two phenol groups of probucol is consumed as an
antioxidant, the other one cannot be an antioxidant any
more.
The methylated bovine serum albumin (mBSA)-
induced delayed-type hypersensitivity (DTH) model of
mice is often used for screening compounds with anti-
inflammatory effects.21 Four poly-hydroxylated com-
pounds (4o, 4p, 4w, and 4x) among others (whose
results will be reported separately23) were tested in this
model, dosed subcutaneously at 24 and 2 h prior to
challenge at doses of 50 and 25 mg/kg with cyclosporin
A (CSA) as positive control. As shown in Figure 3, all
four compounds showed significant inhibition (58-69%)
at 50 mg/kg and moderate inhibition (29-46%) at 25
mg/kg. The four compounds tested in the DTH model
As reported previously,11 when both phenol groups of
probucol are acylated and, hence, the antioxidant prop-
erty is eliminated, the product does not show any
inhibitory effect on VCAM-1 expression. Similarly,
monoether derivative 4ad exhibits an IC50 value of 10
µM, while its diether analogue 5a is inactive on VCAM-1
expression. Since 5a does not have any antioxidant

6426 Journal of Medicinal Chemistry, 2004, Vol. 47, No. 25
Meng et al.
have dramatically different effects on lipid metabolism
suggests that the two compounds may bind differently
on lipoprotein particles and that stereochemistry plays
a major role.
Based on the overall profile of potent in vitro efficacy
on VCAM-1 expression, scalability of synthesis, toler-
ability in animals and favorable efficacy in models of
inflammation, etc., a few compounds from the series
reported herein were scaled to kilogram quantities and
tested in animal models of chronic inflammatory dis-
eases. Compound 4ad (AGIX-4207) inhibited the onset
of paw edema in a collagen II-sensitized rat model of
RA in a dose-dependent manner when given orally. It
also significantly enhanced the effect of a suboptimal
dose of dexamethasone in this model.23 Compound 4ae
(AGI-1096) inhibited arteriosclerosis in allograft recipi-
ents of rats, suggesting that it may be a promising new
agent for the prevention of allograft rejection in organ
transplantation.24 Detailed pharmacological studies on
these two compounds will be reported separately. Com-
pounds 4ad and 4ae are currently in clinical trials for
the treatment of RA and the prevention of organ
transplant rejection, respectively.
Figure 3. Inhibitory profile of compounds on paw swelling
in the DTH model of inflammation (s.c. dosing; left column:
50 mg/kg, right column: 25 mg/kg dosing, CSA: 20 mg/kg).
Table 4. Effects of Selected Compounds on Cholesterol Levels
in Hypercholesterolemic Hamsters
control group
LDL HDL
treated group
LDL HDL
% changea
compd
∆LDL ∆HDL
4g
4s
4u
4w
4x
241 ( 17 104 ( 7 190 ( 17 146 ( 13 -21
262 ( 84 133 ( 10 175 ( 31 173 ( 12 -33
241 ( 17 104 ( 7 338 ( 10 119 ( 3 + 40
+ 41*
+30
+ 15
300 ( 30 278 ( 13
145 ( 18 153 ( 9
9 ( 7
30 ( 5
18 ( 4
63 ( 5 -97* -77*
94 ( 6 -79
86 ( 9 -92* -21*
-39*
In summary, a series of novel antiinflammatory
compounds has been designed and synthesized starting
from a well-known antioxidant, probucol (1), with the
intention of maintaining the antioxidant properties and
eliminating the potential adverse effects of QTc prolon-
gation and HDL lowering. Many of these compounds
showed potent inhibitory effects on VCAM-1 expression.
A correlation between efficacy and lipophilicity has been
observed. This series of monosubstituted probucol ethers
showed potent antioxidant effects equivalent to probucol
in vitro. Some of the derivatives also inhibited LPS-
induced secretion of pro-inflammatory cytokines TNF-
R, IL-1â, and IL-6 in a concentration-dependent manner
in vitro and, therefore, are potential candidates or lead
compounds for a wide variety of diseases, in which these
cytokines are implicated. Probucol, though as strong an
antioxidant in homogeneous solution, did not show any
effect on the secretion of these cytokines at similar
concentrations, indicating that properly designed de-
rivatives of probucol may enter cells or have access to
cell membranes more readily than probucol itself.
Selected compounds have shown antiinflammatory ef-
fects in an animal model of delayed-type hypersensitiv-
ity. Some compounds have exhibited dramatic lipid-
modulating effects in hypercholesterolemic hamsters.
Especially of interest is the fact that some compounds
elevated HDL levels, which suggests that properly
modified derivatives of probucol can be deprived of the
untoward HDL-lowering effect of probucol itself.
The research program reported herein has led to the
discovery of two clinical candidates, i.e., compound 4ad
for RA and compound 4ae for organ transplant rejection.
By inhibiting VCAM-1 expression, a crucial component
of RA pathology, compound 4ad represents a potentially
novel approach to the treatment of RA. As will be
reported separately, this compound possesses numerous
favorable preclinical pharmacological properties that
suggest it may show therapeutic benefit for RA. Com-
pound 4ae addresses the serious issues of accelerated
arteriosclerotic formation and inflammation at the
allograft site through inhibition of VCAM-1 expression.
4ad 227 ( 16 109 ( 6
a *p < 0.05 determined by ANOVA, followed by Fisher’s PLSD
Post Hoc test.
were well tolerated except for that 4o resulted in some
weight loss at the higher dose.
A few compounds were administered as admixtures
in the chow at an approximate dose of 150 mg/kg/d to
hypercholesterolemic hamsters for lipid-modulating ef-
fects. As shown in Table 4, compounds 4w, 4x, and 4ad
lowered LDL levels dramatically (79-97%). However,
these compounds also lowered HDL levels significantly
(21-77%). Compounds 4g and 4s showed interesting
profiles. They lowered LDL levels by 21-33% and raised
HDL levels by 30-41%. These results bolster our
original proposal that properly modified probucol de-
rivatives could be deprived of the detrimental HDL-
lowering effect of probucol itself. Compounds 4u and 4x
are homologous polyols. Compound 4x, with its side
chain just one carbon unit longer than that of 4u, had
a dramatic LDL-lowering effect, while the latter in-
creased LDL levels. It is worth noting that the two
compounds are also different in stereochemistry on their
side chains, which cannot be ruled out as the possible
cause of the different outcomes. Also interestingly, when
the terminal hydroxyl group of 4u is oxidized to the
corresponding carboxylic acid (4s), the detrimental LDL-
elevating effect is eliminated. However, it cannot be
generalized that a terminal hydroxyl group is not good
for LDL lowering, because 4x showed dramatic LDL-
lowering effect. Rather, certain lipophilicity and struc-
tural configuration could be required for probucol
derivatives to properly bind to LDL or HDL particles
to lower LDL levels or elevate HDL levels.22 Thus,
structural modification of compounds of Table 3 could
lead to drug candidates with dual LDL-lowering and
HDL-elevating effects.
In the mouse DTH model described above, compounds
4p, 4x, and 4w also lowered total cholesterol levels in
a dose-related manner while 4o had little effect. The
fact that compounds 4o and 4p are enantiomers and

Phenolic Antioxidants as Inhibitors of VCAM-1 Expression
Journal of Medicinal Chemistry, 2004, Vol. 47, No. 25 6427
over magnesium sulfate. The solution was poured onto the top
of a silica gel column which was eluted with hexanes/ethyl
acetate (2:1). Compounds 4d (viscous residue, 75 mg, 21%) and
4e (viscous residue, 6.3 mg, 2%) were obtained. 4d: 1H NMR
(CDCl3) δ 7.56 (s, 2H, Ph-H), 7.44 (s, 2H, Ph-H), 5.38 (s, 1H,
Ph-OH), 4.44-4.50 (m, 1H, CHOH), 4.20-4.31 (m, 2H, CH2-
OAc), 3.90 (dd, 1H, PhOCH), 3.81(dd, 1H, PhOCH), 2.09 (s,
3H, OAc), 1.40-1.52 (m, 42H, tert-butyls, S,S′-isopropylidene).
MS m/z: 671 ([M + K]+, 100%). 4e: 1H NMR (CDCl3) δ 7.56
(s, 2H, Ph-H), 7.45 (s, 2H, Ph-H), 5.53-5.46 (m, 1H, CHOAc),
5.39 (s, 1H, Ph-OH), 4.02 (dd, 1H, PhOCH), 3.81-3.93 (m, 1H,
CH2OH, PhOCH), 2.18 (s, 3H, OAc), 1.42-1.49 (m, 42H, tert-
butyls, S,S′-isopropylidene). MS m/z: 650 ([M + NH4]+, 30%),
395 (100%).
[N-Carboxymethyl-(3-{2,6-di-tert-butyl-4-[1-(3,5-di-tert-
butyl-4-hydroxyphenylsulfanyl)-1-methylethylsulfanyl]-
phenoxy}-2-hydroxypropyl)amino]acetic Acid (4f). To a
solution of 4a (538 mg, 0.89 mmol) in DMF (5 mL) were added
iminodiacetic acid (1.18 g, 8.9 mmol) and triethylamine (2 mL,
27 mmol), and the mixture was stirred at reflux overnight.
Upon cooling to room temperature it was poured into water
(100 mL), extracted with dichloromethane (3 × 100 mL), dried
over magnesium sulfate, and evaporated. Silica gel chroma-
tography (dichloromethane/methanol 40:1 to 1:1) gave the title
compound as a viscous residue (121 mg, 19%). 1H NMR (CDCl3)
δ 7.54 (s, 2H, Ph-H), 7.52 (d, 1H, NH), 7.44 (s, 2H, Ph-H), 5.40
(s, 1H, PH-OH), 4.23-4.30 (m, 1H, CHOH), 3.81-3.87 (dd,
1H, PhOCH), 3.65-3.72 (dd, 1H, PhOCH), 3.44 (s, 4H, NCH2-
COOH), 2.91-3.00 [m, 1H, (OH)CHCH2N], 2.40-2.49 [m, 1H,
(OH)CHCH2N], 1.39-1.48 (m, 44H, S,S′-isopropylidene, tert-
butyls, CH2CH2COOH). MS m/z: 618 ([M - 2 (COOH) + H]+,
100%).
(3-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxyphen-
ylsulfanyl)-1-methylethylsulfanyl]phenoxy}-2-hydroxy-
propylamino)acetic Acid (4 g). To a suspension of 4a (0.17
g, 0.28 mmol) in ethanol (10 mL) were added glycine (43 mg,
0.57 mmol) and triethylamine (1 mL, 13.6 mmol). The result-
ant mixture was stirred under nitrogen at reflux overnight
and then evaporated. Silica gel chromatography (dichlo-
romethane/methanol 10:1 to 1:1) gave the title compound as
a white solid (99 mg, 55%), mp 180-184 °C (dec). 1H NMR
(CDCl3) δ 7.52 (s, 2H, Ph-H), 7.43 (s, 2H, Ph-H), 5.37 (s, 1H,
Ph-OH), 4.58 (br. s, 1H, CH2CH(OH)CH2), 3.60-3.84 (m, 5H,
OCH2CH(OH)CH2NHCH2), 3.15-3.37 [m, 2H, OCH2CH(OH)-
CH2NHCH2], 3.09-3.15 (m, 1H, NH), 1.43 (s, 18H, tert-butyls),
1.41 (s, 6H, S,S′-isopropylidene), 1.38 (s, 18H, tert-butyls). MS
m/z: 648 ([M + H]+, 60%), 410 (100%). Anal. (C36H57NO5S2‚
H2O) C, H, N, S.
Used as a monotherapy or in combination with an
immunosuppressant, this compound may become an
effective drug for the prevention of chronic organ
transplant rejection. The overall utility of this approach
for beneficial therapies in chronic inflammatory diseases
remains to be demonstrated with clinical trials.
Experimental Section
Chemistry. Melting points are uncorrected. 1H NMR
spectra were recorded on a Varian 400 MHz or a QE300
spectrometer and chemical shifts are reported in parts per
million (ppm, δ) relative to tetramethylsilane as internal
standard. Mass spectra were obtained on a VG 70S (for EI) or
Micromass Q-TOF (for ES) instrument. Elemental analyses
(C, H, N, S) were performed by Atlantic Microlabs, Norcross,
Georgia. Silica gel 60 (E. Merck, 230-400 mesh) was used for
preparative column chromatography. Calculated logP was
obtained by using LeadScope PC Professional 2.0.11. All
compounds reported herein had 95% or higher purity according
1
to H NMR spectra.
2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-oxiranylmethoxy-
phenylsulfanyl)-1-methyl-ethylsulfanyl]phenol (4a). To
a solution of probucol (2.58 g, 5 mmol) in THF (50 mL) cooled
to 0 °C were added glycidol (0.66 mL, 10 mmol), triphenyl-
phosphine (2.62 g, 10 mmol), and diethyl azodicarboxylate
(1.57 mL, 10 mmol). The resultant mixture was stirred under
nitrogen at reflux for 48 h and then evaporated. Silica gel
chromatography (hexanes/dichloromethane 4:1, 2:1) gave the
title compound as a viscous residue (1.01 g, 35%) which
solidified on standing, mp 139-141 °C. 1H NMR (CDCl3) δ 7.56
(s, 2H, Ph-H), 7.44 (s, 2H, Ph-H), 5.39 (s, 1H, Ph-OH), 4.04
(dd, 1H, PhOCH2), 3.75 (dd, 1H, PhOCH2), 3.36-3.42 (m, 1H,
PhOCH2CH), 2.90 (dd, 1H, epoxide CH2), 2.77 (dd, 1H, epoxide
CH2), 1.40-1.49 (m, 42H, tert-butyls, S,S′-isopropylidene). MS
m/z: 671 ([M + K]+, 100%).
2,6-Di-tert-butyl-4-{1-[3,5-di-tert-butyl-4-(3-ethoxy-2-
hydroxypropoxy)phenylsulfanyl]-1-methyl-ethylsulfanyl}-
phenol (4b) and 3-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-
4-hydroxyphenylsulfanyl)-1-methylethylsulfanyl]phe-
noxy}propane-1,2-diol (4c). To a suspension of 4a (320 mg,
0.56 mmol) in ethanol (10 mL) was added 1 N sodium
hydroxide solution (1.5 mL), and the resultant mixture was
stirred at reflux for 72 h and then evaporated. The residue
was distributed between ethyl acetate (50 mL) and brine (50
mL). The organic phase was washed with brine (50 mL), dried
over magnesium sulfate, and evaporated. Silica gel chroma-
tography (hexanes/dichloromethane 1:1, dichloromethane, and
then dichloromethane/ethyl acetate 4:1) afforded 4c (viscous
solid, 58 mg, 18%), mp 63-65 °C and 4b (off-white solid, 52
mg, 15%), mp 98-100 °C. 4b: 1H NMR (CDCl3) δ 7.55 (s, 2H,
Ph-H), 7.45 (s, 2H, Ph-H), 5.38 (s, 1H, Ph-OH), 4.30-4.39 [m,
1H, CH2CH(OH)CH2], 3.82 (d, 2H, PhOCH2), 3.53-3.65 (m,
4H, CH2OCH2CH3), 1.43-1.46 (m, 42H, tert-butyls and S,S′-
isopropylidene), 1.22 (t, 3H, OCH2CH3). MS m/z: 657 ([M +
K]+, 100%). 4c: 1H NMR (CDCl3) δ 7.56 (s, 2H, Ph-H), 7.44
(s, 2H, Ph-H), 5.37 (s, 1H, Ph-OH), 4.29-4.36 [m, 1H,
CH2CH(OH)CH2], 3.93 [dd, J ) 8, 9 Hz, 1H, OCH2CH(OH)],
3.83 [dd, J ) 3, 11 Hz, 1H, OCH2CH(OH)CH2OH], 3.75 [dd, J
) 4, 9 Hz, 1H, OCH2CH(OH)], 3.66 [dd, J ) 6, 11 Hz, OCH2-
CH(OH)CH2OH], 1.45 (s, 6H, S,S′-isopropylidene), 1.44 (s,
18H, tert-butyls), 1.43 (s, 1H, tert-butyls). MS m/z: 629 ([M +
K]+, 80%), 353 (100%). Anal. (C34H54O4S2‚H2O) C, H, S.
Acetic Acid 3-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-
4-hydroxyphenylsulfanyl)-1-methylethylsulfanyl]phe-
noxy}-2-hydroxypropyl Ester (4d) and Acetic Acid
2-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxyphenyl-
sulfanyl)-1-methylethylsulfanyl]phenoxy}-1-hydroxy-
methylethyl Ester (4e). To a solution of 4a (339 mg, 0.56
mmol) in THF (15 mL) was added acetic acid (10 mL) and the
mixture was stirred for 5 min. Triethylamine (10 mL) was
added and the resultant mixture was stirred at reflux for 72
h and then evaporated. The residue was dissolved in dichlo-
romethane (150 mL), washed with water (3 × 100 mL), dried
2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-{2-[2-(2-meth-
oxyethoxy)ethoxy]ethoxy}phenylsulfanyl)-1-methylethyl-
sulfanyl]phenol (4h). To a solution of probucol (2.0 g, 3.9
mmol) and tri(ethylene glycol) monomethyl ether (1.27 g, 7.7
mmol) in tetrahydrofuran (40 mL) was added triphenylphos-
phine (2.0 g, 7.7 mmol), and the resulting mixture was cooled
to 0 °C. Diethyl azodicarboxylate (1.3 g, 7.7 mmol) was then
added dropwise. The mixture was stirred at 0 °C for 30 min
and allowed to warm to room temperature. The solution was
ultimately warmed to 40 °C and stirred for an additional 2 h.
The reaction mixture was concentrated under reduced pres-
sure to a brown oil. Silica gel chromatography (10-30% ethyl
acetate/hexanes) afforded 0.96 g (40%) of the expected ether
as a viscous yellow oil. Rf 0.25 (30% ethyl acetate/hexanes).
1H NMR (CDCl3) δ 7.53 (s, 2H, Ph-H), 7.45 (s, 2H, Ph-H), 5.36
(s, 1H, Ph-OH), 3.89 (br s, 4H, Ph-OCH2CH2O), 3.67-3.75 (m,
6H, OCH2CH2OCH2), 3.56-3.58 (m, 2H, OCH2), 3.39 (s, 3H,
OCH3), 1.42-1.44 (m, 42H, tert-butyls and S,S-isopropylidene).
MS (ESI) m/z: 685 ([M + Na]+, 100%). HRMS (ESI) calcd for
C
38H62O5S2 (M + K), 701.3676; found, 701.3649. Anal.
(C38H62O5S2) C, H, S.
Acrylic Acid 4-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-
4-hydroxyphenylsulfanyl)-1-methylethylsulfanyl]-
phenoxy}butyl Ester (4i) and 2,6-Di-tert-butyl-4-{1-[3,5-
di-tert-butyl-4-(4-hydroxybutoxy)phenylsulfanyl]-1-
methylethylsulfanyl}phenol (4k). To a solution of probucol

6428 Journal of Medicinal Chemistry, 2004, Vol. 47, No. 25
Meng et al.
(2.58 g, 5 mmol) in THF (50 mL) were added 4-hydroxybutyl
acrylate (1.0 mL, 10 mmol), triphenylphosphine (2.62 g, 10
mmol), and diethyl azodicarboxylate (1.57 mL, 10 mmol). The
resultant mixture was stirred under nitrogen at reflux for 72
h and then evaporated. Silica gel chromatography (hexanes/
dichloromethane 4:1) gave 4i as a brown oily residue (0.92 g,
g, 5 mmol), and 2-ethoxy-1,3-dioxolane-4(S),5(S)-dimethanol14
(0.89 g, 5 mmol). The resultant mixture was stirred at reflux
for 3 h and then evaporated. Silica gel chromatography
(hexanes/ethyl acetate 4:1) gave three parts: diasteroisomer
A (0.36 g), diastereoisomer B (0.22 g) and a mixture of the
two (0.72 g) all as viscous residues (total yield: 39%). Dias-
tereoisomer A: 1H NMR (CDCl3) δ 7.56 (s, 2H, Ph-H), 7.44 (s,
2H, Ph-H), 5.89 [s, 1H, CH(O)3], 5.37 (s, 1H, PhOH), 4.73-
4.81 (m, 1H, PhOCH2CH), 4.19-4.25 (m, 1H, CHCH2OH),
3.80-4.01 (m, 3H, CH2O), 3.62-3.96 (m, 3H, CH2O), 2.80 (dd,
J ) 9, 3, 1H, OH), 1.45 (s, 6H, S,S′-isopropylidene), 1.44 (s,
18H, tert-butyls), 1.43 (s, 18H, tert-butyls), 1.25 (t, 3H,
OCH2CH3). Diasteroisomer B: 1H NMR (CDCl3) δ 7.56 (s, 2H,
Ph-H), 7.44 (s, 2H, Ph-H), 5.92 [s, 1H, CH(O)3], 5.36 (s, 1H,
PhOH), 4.50-4.59 (m, 1H, PhOCH2CH), 4.13-4.30 (m, 3H,
CHCH2OH, CH2O), 3.89-4.00 (m, 3H, CH2O), 3.61 (quad, J
) 7, 2H, OCH2CH3), 1.86 (dd, J ) 6, 8, 1H, OH), 1.41-1.46
(m, 42H, S,S′-isopropylidene, tert-butyls), 1.21 (t, 3H, OCH2CH3).
The mixture obtained above (0.72 g, 1.1 mmol) was dissolved
in methanol (25 mL). Acetic acid (2 mL) and water (1 mL) were
added, and the resultant mixture was stirred at reflux for 3 h
and then evaporated to about 10 mL. It was poured into water
(100 mL) and extracted with dichloromethane (2 × 100 mL).
The organic phase was dried over magnesium sulfate and
evaporated. The residue was dissolved in methanol (15 mL),
and potassium carbonate (0.5 g) was added. The mixture was
stirred at room temperature for 1 h and then poured into 1 N
HCl solution (100 mL). It was extracted with dichloromethane
(2 × 100 mL), dried over magnesium sulfate, and evaporated.
Silica gel chromatography (dichloromethane/ethyl acetate 4:1)
gave the title compound as a white powder (0.45 g, 66%), mp
1
29%). H NMR (CDCl3) δ 7.54 (s, 2H, Ph-H), 7.46 (s, 2H, Ph-
H), 6.42 [dd, 1H, OC(O)CHdCHcis], 6.14 [dd, 1H, OC(O)-
CH)CH2], 5.84 [dd, 1H, OC(O)CHdCHtrans], 5.38 (s, 1H, Ph-
OH), 4.23 [t, 2H, CH2OC(O)], 3.75 (t, 2H, PhOCH2), 1.92-2.00
(m, 2H, PhOCH2CH2), 1.78-1.86 (m, 2H, PhOCH2CH2CH2),
1.46 (s, 6H, S,S′-isopropylidene), 1.45 (s, 18H, tert-butyls), 1.42
(s, 18H, tert-butyls). To a suspension of this compound (0.82
g, 1.28 mmol) in methanol (20 mL) was added potassium
carbonate (0.5 g, 3.6 mmol), and the resultant mixture was
stirred under nitrogen at room temperature overnight. It was
poured into water (50 mL), extracted with dichloromethane
(2 × 50 mL), dried over magnesium sulfate and evaporated.
Silica gel chromatography (hexanes/ethyl acetate 4:1) gave 4k
as a colorless viscous residue (0.52 g, 70%) which solidified on
standing, mp 106-108 °C. 1H NMR (CDCl3) δ 7.54 (s, 2H, Ph-
H), 7.46 (s, 2H, Ph-H), 5.38 (s, 1H, Ph-OH), 3.71-3.77 (m, 4H,
OCH2CH2CH2CH2OH), 1.92-1.99 (m, 2H, PhOCH2CH2),
1.1.66-1.74 (m, 2H, CH2CH2OH), 1.39-1.52 (m, 42H, S,S′-
isopropylidene, tert-butyls). MS m/z: 627 ([M + K]+, 85%), 351
(100%).
2-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxyphen-
ylsulfanyl)-1-methylethylsulfanyl]phenoxymethyl}pro-
pane1,3-diol (4l). To a solution of probucol (1.0 g, 1.9 mmol)
and 3-(tert-butyldimethylsilanyloxy)-2-(tert-butyldimethylsi-
lanyloxymethyl)propan-1-ol25 (1.3 g, 3.9 mmol) in THF (20 mL)
was added triphenylphosphine (1.0 g, 3.9 mmol), and the
resulting mixture was cooled to 0 °C. Diethyl azodicarboxylate
(0.60 g, 3.9 mmol) was then added dropwise. The mixture was
stirred at 0 °C for 30 min and allowed to warm to room
temperature. The solution was ultimately warmed to 50 °C
and stirred for an additional 18 h. The reaction mixture was
concentrated under reduced pressure to a brown oil and
subjected to silica gel chromatography (100% hexanes then
1-5% CH2Cl2/hexanes) to afford 0.33 g (20%) of 2,6-di-tert-
butyl-4-(1-{3,5-di-tert-butyl-4-[3-(tert-butyl-dimethyl-silanyloxy)-
2-(tert-butyl-dimethyl-silanyloxymethyl)-propoxy]phenylsul-
fanyl}-1-methyl-ethylsulfanyl)phenol (4j) as a yellow foam
(mixed with trace amounts of starting probucol). Rf 0.61 (10%
ethyl acetate/hexanes). 1H NMR (CDCl3) δ 7.53 (s, 2H, Ph-H),
7.45 (s, 2H, Ph-H), 5.35 (s, 1H, Ph-OH), 3.76-3.81 (m, 6H,
CH2CH(CH3)3), 2.34-2.38 (m, 1H, CH2CH), 1.41-1.44 (m,
42H, tert-butyls and S,S-isopropylidene), 0.88 9s, 18H, SiC-
(CH3)3], 0.04 [s, 12H, Si(CH3)2]. MS (ESI) m/z: 855 ([M + Na]+,
90%), 595 (100%). To a solution containing the above silyl ether
(0.10 g, 0.12 mmol) in THF (10 mL) was added tetrabutylam-
monium fluoride (1 M in tetrahydrofuran, 0.4 mL), and the
reaction was stirred at room temperature for 2 h. The solution
was ultimately warmed to 35 °C and stirred for an additional
2 h. The reaction mixture was diluted with a 50% aqueous
solution of ammonium chloride (20 mL) and extracted with
methylene chloride (3 × 20 mL). The combined organic layers
were washed with water (2 × 10 mL), brine (1 × 10 mL) and
dried over sodium sulfate, and the solvent was removed under
reduced pressure. The crude residue was purified by silica gel
chromatography (ethyl acetate/hexanes 1:1) to give 0.032 g
(45%) of the title compound as a pale yellow foam. Rf 0.34 (50%
ethyl acetate/hexanes). 1H NMR (CDCl3) δ 7.55 (s, 2H, Ph-H),
7.44 (s, 2H, Ph-H), 5.36 (s, 1H, Ph-OH), 3.83-4.02 (m, 6H,
CH2CH(CH2)2), 2.50-2.56 (m, 1H, CH2CH), 2.03-2.07 (m, 2H,
CH2OH); 1.42-1.44 (m, 42H, tert-butyls and S,S-isopropyl-
idene). MS (EI) m/z: 604 (M+, 0.1%), 279 (100%). HRMS (ESI)
calcd for C35H56O4S2 (M + K), 643.3257; found 643.3255. Anal.
(C35H56O4S2) C, H, S.
1
69-71 °C. H NMR (CDCl3) δ 7.56 (s, 2H, Ph-H), 7.45 (s, 2H,
Ph-H), 5.36 (s, 1H, PhOH), 4.26-4.34 (m, 1H, PhOCH2CH),
4.01 (dd, J ) 9, 10, 1H, CH-O-), 3.72-3.86 (m, 4H,, CH-O-),
2.69-2.77 (m, 2H, OH), 2.15 (br. t, 1H, OH), 1.42-1.47 (m,
42H, S,S′-isopropylidene, tert-butyls). MS m/z: 643 ([M + Na]+,
100%). Anal. (C35H56O5S2‚1/3H2O) C, H, S.
4-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxyphe-
nylsulfanyl)-1-methylethylsulfanyl]-phenoxy}butane-
1,2(R),3(R)-triol (4p). This compound was prepared using the
1
same procedure as for 4o, mp 69-71 °C. H NMR (CDCl3) δ
7.56 (s, 2H, Ph-H), 7.45 (s, 2H, Ph-H), 5.38 (s, 1H, PhOH),
4.25-4.32 (m, 1H, PhOCH2CH), 3.96 (dd, J ) 9, 10, 1H, CH-
O-), 3.72-3.86 (m, 4H, CH-O-), 3.00-3.40 (m, 3H, OH), 1.46
(s, 6H, S,S′-isopropylidene), 1.45 (s, 18H, tert-butyls), 1.43 (s,
18H, tert-butyls). MS m/z: 643 ([M + Na]+, 100%). Anal.
(C35H56O5S2•1/3H2O) C, H, S.
5-O-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxy-
phenylsulfanyl)-1-methylethylsulfanyl]phenyl}-D-ribonic
Acid γ-Lactone (4r). To a solution of probucol (3.0 g, 5.88
mmol) in THF (40 mL) were added triphenylphosphine (1.54
g, 5.88 mmol), diethyl azodicarboxylate (0.93 mL g, 5.88 mmol),
and 2,3-O-(ethoxymethylene)-D-ribonic acid γ-lactone15 (1 g,
4.89 mmol). The resultant mixture was stirred at reflux for 3
h and then evaporated. Silica gel chromatography (hexanes/
ether 4:1) gave 457 mg (13%) of 5-O-{2,6-i-tert-butyl-4-[1-(3,5-
di-tert-butyl-4-hydroxy-phenylsulfanyl)-1-methyl-ethylsulfanyl]-
phenyl}-2,3-O-(ethoxymethylene)-D-ribonic acid γ-lactone as a
1
viscous residue. H NMR (CDCl3) δ 7.57 (s, 2H, Ph-H), 7.44
(s, 2H, Ph-H), 6.02 [s, 1H, CH(O)3], 5.37 (s, 1H, PhOH), 4.98-
5.15 (m, 2H, CH-O-), 4.88 (br. s, 1H, CH-O-), 4.08 (br. s, 2H,
PhOCH2), 3.42-3.78 (m, 2H, OCH2CH3), 1.38-1.49 (m, 42H,
S,S′-isopropylidene, tert-butyls), 1.18-1.36 (m, 3H, OCH2CH3).
To a solution of this compound (450 mg, 0.64 mmol) in
methanol (80 mL) was added a mixture of acetic acid/water
(4:1, 10 mL), and the resultant mixture was stirred at room
temperature overnight. Most of the solvent was evaporated
and the remaining solution was neutralized with 1 N NaOH
solution. It was extracted with dichloromethane, dried, and
evaporated to give the title compound as a white solid (41 mg,
98%), mp 91-93 °C. 1H NMR (CDCl3) δ 7.58 (s, 2H, Ph-H),
7.44 (s, 2H, Ph-H), 5.38 (s, 1H, PhOH), 4.80-4.85 (m, 1H, CH-
O-), 4.69-4.75 (m, 2H, CH-O-), 4.07 (br. s, 2H, PhOCH2), 1.41-
4-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxy-
phenylsulfanyl)-1-methylethylsulfanyl]phenoxy}butane-
1,2-(S),3(S)triol (4o). To a solution of probucol (5.16 g, 10
mmol) in THF (20 mL) cooled to 0 °C were added triphen-
ylphosphine (1.3 g, 5 mmol), diethyl azodicarboxylate (0.8 mL

Phenolic Antioxidants as Inhibitors of VCAM-1 Expression
Journal of Medicinal Chemistry, 2004, Vol. 47, No. 25 6429
1.50 (m, 42H, S,S′-isopropylidene, tert-butyls), 1.18-1.36 (m,
3H, OCH2CH3). MS m/z: 685 ([M + K]+, 100%).
2H, Ph-H), 5.36 (s, 1H, Ph-OH), [5.31 (br. s) and 4.78 (br. s,
1H, anomeric H)], 3.30-4.38 (br. m, 6H, PhOCH2 and glucose
H), 1.38-1.43 (m, 42H, tert-butyls, S,S′-isopropylidene). MS
m/z: 701 ([M + Na]+, 100%). Anal. (C37H58O7S2‚1/3H2O) C, H,
S.
5-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxyphen-
ylsulfanyl)-1-methylethylsulfanyl]phenoxy}-2(R),3(R),4(R)-
trihydroxypentanoic Acid (4s). To a solution of 4r (200 mg,
0.31 mmol) in THF (5 mL) was added 1.2 mL of 1 N NaOH
solution, and the mixture was stirred at room temperature for
3 h. The mixture was adjusted to pH ) 3 with 1 N HCl,
extracted with dichloromethane, dried over magnesium sul-
fate, and evaporated to give the title compound as a viscous
solid (107 mg, 53%), mp 67-69 °C. 1H NMR (CDCl3) δ 7.57 (s,
2H, Ph-H), 7.44 (s, 2H, Ph-H), 5.37 (s, 1H, PhOH), 4.48 (dd, J
) 8, 13 Hz, 1H, CHOH), 4.38 [d, J ) 6 Hz, 1H, CH(OH)COOH],
4.03 (d, J ) 6 Hz, 2H, PhOCH2), 3.95 [dd, J ) 6, 7 Hz, 1H,
CHOH), 1.46 (s, 6H, S,S′-isopropylidene), 1.44 (s, 36H, tert-
butyls). MS m/z: 703 ([M + K]+, 100%).
6-O-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxy-
phenylsulfanyl)-1-methylethylsulfanyl]phenyl}-D-glucitol
(4x). To a solution of 4w (70 mg, 0.10 mmol) in THF (5 mL)
was added sodium borohydride, and the mixture was stirred
under nitrogen at room temperature for 2 h. Saturated
ammonium chloride (2 mL) was added and the mixture stirred
for another 1 h, poured into water (50 mL), and extracted with
dichloromethane (3 × 50 mL). The organic phase was dried
over magnesium sulfate and evaporated. Silica gel chroma-
tography (dichloromethane/methanol 100:12) gave the title
1
compound as a white solid (19 mg, 27%), mp 109-113 °C. H
NMR (CDCl3) δ 7.54 (s, 2H, Ph-H), 7.44 (s, 2H, Ph-H), 5.36 (s,
1H, PhOH), 4.35 (br. m, 1H, CHOH), 3.30-4.10 (m, 7H,
PhOCH2 and CHOH), 1.39 (s, 24H, tert-butyls, S,S′-isopropyl-
idene), 1.38 (s, 18H, tert-butyls). MS m/z: 703 ([M + Na]+,
100%). Anal. (C37H60O7S2‚H2O) C, H, S.
5-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxyphen-
ylsulfanyl)-1-methylethylsulfanyl]phenoxy}pentane-1,2(S),3-
(S),4(R)tetraol (4u) and 5-O-{2,6-Di-tert-butyl-4-[1-(3,5-
di-tert-butyl-4-hydroxyphenylsulfanyl)-1-methyl-ethyl-
sulfanyl]phenyl}-D-ribose (4t). To a solution of 4s (4 g, 6.16
mmol) in THF (150 mL) was added LAH (1 M in ether, 12.3
mL) slowly, and the mixture was stirred at room temperature
for 2 h. Reaction was incomplete. Additional LAH solution (6.1
mL) was added and the mixture stirred for 1 h. A third charge
of LAH solution (2 mL) was added again and stirred for 1 h.
The reaction was carefully quenched with saturated sodium
sulfate (50 mL), and the mixture was stirred overnight. Ether
was added to the mixture, and it was filtered. The filtrate was
dried over magnesium sulfate and evaporated. Silica gel
chromatography (hexanes/ether 1:1 to 100% ether) gave 4r
(980 mg, 25%), 4t (600 mg, 15%), and 4u (600 mg, 15%). 4u:
3-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxyphen-
ylsulfanyl)-1-methylethylsulfanyl]phenoxy}acrylic Acid
Ethyl Ester (4y) and 2,6-Di-tert-butyl-4-[1-(3,5-di-tert-
butyl-4-ethoxyphenylsulfanyl)-1-methylethylsulfanyl]-
phenol (4z). To a solution of probucol (5.16 g, 10 mmol) in
THF (50 mL) were added ethyl propiolate (1.2 mL, 12 mmol)
and triethylamine (7 mL, 50 mmol). The resultant mixture was
stirred under nitrogen at reflux for 72 h. After being cooled to
room temperature, it was poured into brine (100 mL), extracted
with dichloromethame (3 × 100 mL), dried over magnesium
sulfate, and evaporated. Silica gel chromatography (hexanes/
dichloromethane 9:1 to straight dichloromethane) gave com-
pounds 4y (viscous residue, 0.37 g, 6%) and 4z (colorless solid,
509 mg, 9%), mp 138-140 °C. 4y: 1H NMR (CDCl3) δ 7.63 (s,
2H, Ph-H), 7.43 (s, 2H, Ph-H), 6.40 (d, 1H, OCH)CH), 5.39
(s, 1H, Ph-OH), 5.03 (dd, 1H, OCHdCH), 4.26 (quad, 2H,
OCH2CH3), 1.38-1.55 (m, 42H, tert-butyls, S,S′-isopropyl-
idene), 1.31 (t, 3H, OCH2CH3). MS m/z: 615 ([M + H]+, 50%),
377 (100%). 4z: 1H NMR (CDCl3) δ 7.55 (s, 2H, Ph-H), 7.44
(s, 2H, Ph-H), 5.38 (s, 1H, Ph-OH), 3.76 (quad, 2H, OCH2CH3),
1.38-1.51 (s, 45H, tert-butyls, S,S′-isopropylidene, OCH2CH3).
MS m/z: 583 ([M + K]+, 25%), 279 (100%).
1
Off-white solid, mp 105-110 °C. H NMR (CDCl3) δ 7.57 (s,
2H, Ph-H), 7.45 (s, 2H, Ph-H), 5.37 (s, 1H, PhOH), 4.38 (dd, J
) 6, 13, 1H, CH-O-), 4.02 (d, J ) 7, 2H, PhOCH2), 3.82-3.88
(m, 1H, CH-O-), 3.86-3.90 (m, 2H, CH2OH), 3.74 (dd, J ) 6,
7, 1H, CH-O-), 1.45 (s, 6H, S,S′-isopropylidene), 1.44 (s, 36H,
tert-butyls). MS m/z: 689 ([M + K]+, 100%). 4t: Off-white solid,
1
mp 138-141°C. H NMR (CDCl3) δ 7.52 (s, 2H, Ph-H), 7.45
(s, 2H, Ph-H), 5.43 (d, 1H, anomeric H), 5.34 (s, 1H, PhOH),
4.44-4.49 (m, 1H, CH-O-), 4.12-4.19 (m, 2H, CH-O-), 3.87-
3.91 (m, 2H, PhOCH2), 1.37-1.45 (m, 42H, S,S′-isopropylidene,
tert-butyls). MS m/z: 687 ([M + K]+, 100%). Anal. (C36H56O6S2)
C, H, S.
2,6-Di-tert-butyl-4-{1-[3,5-di-tert-butyl-4-(3-hydroxypro-
penyloxy)phenylsulfanyl]-1-methylethylsulfanyl}phe-
nol (4aa). To a solution of 4y (65 mg, 0.1 mmol) in THF (15
mL) was added lithium aluminum hydride (1 mL, 1 M solution
in THF), and the resultant mixture was stirred at room
temperature overnight. The reaction was carefully quenched
with saturated ammonium chloride solution (20 mL) and
stirred for 0.5 h at room temperature. The mixture was
extracted with dichloromethane (3 × 50 mL), dried over
sodium sulfate, and evaporated. Silica gel chromatography
(hexanes/ethyl acetate 4:1) gave the title compound as an oily
6-O-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxy-
phenylsulfanyl)-1-methylethylsulfanyl]phenyl}-1, 2, 3,
4-tetra-O-acetyl-â-D-glucopyranose (4v). To a solution of
probucol (1.8 g, 3.5 mmol) in THF (20 mL) were added 1,2,3,4-
tetra-O-actyl-â-D-glucopyranose (1.0 g, 2.9 mmol), triphen-
ylphosphine (0.92 g, 3.5 mmol), and diethyl azodicarboxylate
(0.55 mL, 3.5 mmol). The resultant mixture was stirred under
nitrogen at reflux for 2 h and then evaporated. Silica gel
chromatography (hexanes/ethyl acetate 4:1) gave the title
compound as an off-white solid (0.92 g, 31%), 149-151 °C. 1H
NMR (CDCl3) δ 7.53 (s, 2H, Ph-H), 7.45 (s, 2H, Ph-H), 5.80 (d,
1H, anomeric H), 5.38 (s, 1H, Ph-OH), 5.33 (dd, 1H, CHOAc),
5.16 (dd, 1H, CHOAc), 4.90 (dd, 1H, CHOAc), 4.15-4.23 (m,
1H, glucose C5-H), 3.88 (dd, 1H, Ph-OCH), 3.74 (dd, 1H, Ph-
OCH), 2.14 (s, 3H, OAc), 2.06 (s, 3H, OAc), 2.03 (s, 3H, OAc),
2.02 (s, 3H, OAc), 1.45 (s, 24H, tert-butyls, S,S′-isopropylidene),
1.38 (s, 18H, tert-butyls). MS m/z: 885 ([M + K]+, 25%), 279
(100%).
1
residue (46 mg, 80%). H NMR (CDCl3) δ 7.61 (s, 2H, Ph-H),
7.45 (s, 2H, Ph-H), 6.00 (d, 1H, OCH)CH), 5.40 (s, 1H, Ph-
OH), 4.84 (dd, 1H, OCHdCH), 4.45 (dd, 2H, CH2OH), 1.35-
1.48 (m, 42H, tert-butyls, S,S′-isopropylidene). MS m/z: 611
([M + K]+, 55%), 279 (100%).
3-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxyphen-
ylsulfanyl)-1-methylethylsulfanyl]phenoxy}acrylic Acid
(4ab). To a solution of 4y (0.16 g, 0.26 mmol) in THF (5 mL)
were added water (2 mL) and lithium hydroxide monohydrate
(42 mg, 1 mmol). The resultant mixture was stirred at reflux
overnight. After being cooled to room temperature, it was
poured into dichloromethane (50 mL), washed with brine, dried
over magnesium sulfate, and evaporated. Silica gel chroma-
tography (hexanes/ethyl acetate 4:1) gave the title compound
as a viscous residue (22 mg, 14%) which solidified on standing,
mp 213-215 °C. 1H NMR (CDCl3) δ 7.63 (s, 2H, Ph-H), 7.44 (s,
2H, Ph-H), 6.52 (d, 1H, OCH)CH), 5.39 (s, 1H, Ph-OH), 5.08
(d, 1H, OCHdCH), 1.47 (s, 6H, S,S′-isopropylidene), 1.44 (s,
18H, tert-butyls), 1.42 (s, 18H, tert-butyls). MS m/z: 625 ([M
+ K]+, 75%), 159 (100%).
6-O-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxy-
phenylsulfanyl)-1-methylethylsulfanyl]phenyl}-D-gluco-
pyranose (4w). To a suspension of 4v (0.68 g, 0.080 mmol)
in methanol (50 mL) was added potassium carbonate (1 g, 7.2
mmol), and the mixture was stirred under nitrogen at room
temperature overnight. It was poured into water (200 mL),
extracted with ethyl acetate (3 × 150 mL), washed with brine
(100 mL), dried over magnesium sulfate, and evaporated. Silica
gel chromatography (dichloromethane/methanol 10:1 to 5:1)
gave the title compound as an off-white solid (0.26 g, 48%),
1
mp 94-99 °C. H NMR (CDCl3) δ 7.52 (s, 2H, Ph-H), 7.44 (s,

6430 Journal of Medicinal Chemistry, 2004, Vol. 47, No. 25
Meng et al.
1
Acetic Acid, [4-[[1-[[3,5-Bis(1,1-dimethylethyl)-4-hy-
droxyphenyl]thio]-1-methylethyl]thio]-2,6-bis(1,1-dimeth-
ylethyl)phenoxy]- (4ad). To a solution of probucol (0.5 g, 0.97
mmol) in dimethylformamide (1.5 mL) were added ethyl
iodoacetate (0.31 g, 1.45 mmol) and 40% potassium fluoride
on alumina (0.7 g, 4.8 mmol). The mixture was stirred for 24
h and then diluted with ether (25 mL), filtered, and washed
with water (2 × 5 mL). The ether layer was dried over MgSO4,
filtered, and concentrated. Silica gel chromatography (hexanes/
ether 5:95) yielded 160 mg of {2,6-di-tert-butyl-4-[1-(3,5-di-tert-
butyl-4-hydroxyphenylsulfanyl)-1-methylethylsulfanyl]phenoxy}-
acetic acid ethyl ester, which was dissolved in THF/H2O/MeOH
(4:1:1) (4 mL), and lithium hydroxide hydrate (50 mg) was
added. The resultant mixture was stirred for 1 h and then
neutralized with 1 N HCl. It was extracted with ether (2 × 10
mL), dried over MgSO4, filtered, and concentrated. Silica gel
chromatography (hexanes/ether 50:50) gave the title compound
as a white solid (90 mg, 16%), mp 164-165 °C. 1H NMR
(CDCl3) δ 7.55 (s, 2H, Ph-H), 7.40 (s, 2H, Ph-H), 5.35 (s, 1H,
Ph-OH), 4.40 (s, 2H, OCH2COOH), 1.43 (s, 6H, S,S′-isopropyl-
idene), 1.41 (s, 9H, tert-butyl), 1.39 (s, 9H, tert-butyl). MS
m/z: 613 ([M + K]+, 60%), 159 (100%). Anal. (C33H50O4S2) C,
H, S.
Butanoic Acid, 4-[4-[[1-[[3,5-Bis(1,1-dimethylethyl)-4-
hydroxyphenyl]thio]-1-methylethyl]thio]-2,6-bis(1,1-di-
methylethyl)phenoxy]- (4ae). This compound was prepared
using a similar procedure as for 4ad using methyl 4-iodobu-
tyrate and KF on alumina. Off-white solid, mp 138-139 °C.
1H NMR (CDCl3) δ 7.53 (s, 2H, Ph-H), 7.45 (s, 2H, Ph-H), 5.37
(s, 1H, Ph-OH), 3.77 (t, J ) 6.8 Hz, 2H, OCH2CH2), 2.55 (t, J
) 7.6 Hz, 2H, CH2COOH), 2.16 (m, 2H, OCH2CH2), 1.44 (s,
24H, tert-butyls and S,S′-isopropylidene), 1.41 (s, 18H, tert-
butyls). MS m/z: 641 ([M + K]+, 100%). Anal. (C35H54O4S2) C,
H, S.
5-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxyphen-
ylsulfanyl)-1-methylethylsulfanyl]phenoxy}pentanoic
Acid (4af). This compound was prepared in similar procedure
as for 4ad. White solid, mp 60-63 °C. 1H NMR (CDCl3) δ 7.54
(s, 2H, Ph-H), 7.46 (s, 2H, Ph-H), 5.37 (s, 1H, Ph-OH), 3.73 (t,
J ) 7 Hz, 2H, PhOCH2), 2.46 (t, J ) 7 Hz, 2H, CH2COOH),
1.87-1.99 (m, 2H, CH2CH2CH2), 1.74-1.83 (m, 2H, CH2CH2-
CH2), (1.45 (s, 24H, tert-butyls and S,S′-isopropylidene), 1.42
(s, 18H, tert-butyls). MS m/z: 616 ([M + K]+, 100%).
3-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxyphen-
ylsulfanyl)-1-methylethylsulfanyl]phenoxy}propionic
Acid (4ag). To a solution of probucol (2.58 g, 5 mmol) in THF
(30 mL) were added potassium tert-butoxide (0.56 g, 5 mmol)
and â-propiolactone (0.9 g, 12.5 mmol), and the mixture was
stirred at room temperature for 1.5 h. Water (100 mL) was
added and the mixture was adjusted to pH ) 1 with 1 N HCl
solution. It was extracted with dichloromethane, washed with
water, dried over magnesium sulfate, and evaporated. Silica
gel chromatography (hexanes/ethyl acetate 4:1) gave the title
compound as a white solid (200 mg, 68%), mp 160-163 °C. 1H
NMR (CDCl3) δ 7.54 (s, 2H, Ph-H), 7.444 (s, 2H, Ph-H), 5.36
(s, 1H, Ph-OH), 4.01 (t, J ) 7 Hz, 2H, PhOCH2), 2.95 (t, J )
7 Hz, 2H, CH2COOH), 1.44 (s, 24H, tert-butyls and S,S′-
isopropylidene), 1.41 (s, 18H, tert-butyls). MS m/z: 588 ([M]+,
0.1%), 279 (100%).
183-184 °C. H NMR (CDCl3) δ 7.56 (s, 2H, Ph-H), 7.43 (s,
2H, Ph-H), 7.33 (t, J ) 6, 1H, NH), 5.37 (s, 1H, Ph-OH), 4.25
(s, 2H, PhOCH2), 4.69 (quad, J ) 6, 2H, NHCH2), 2.68 (t, J )
6, NHCH2CH2), 1.45 (s, 6H, S,S′-isopropylidene), 1.44 (s, 18H,
tert-butyls), 1.38 (s, 18H, tert-butyls). MS m/z: 668 ([M + Na]+,
100%). Anal. (C36H55NO5S2) C, H, N, S.
(2-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxyphen-
ylsulfanyl)-1-methylethylsulfanyl]phenoxy}acetylamino)-
acetic Acid (4aj). To a solution of 4ad (50 mg, 0.087 mmol)
in methylene chloride (0.87 mL) were added glycine ethyl ester
hydrochloride (15.8 mg, 0.11 mmol), 1-(3-dimethylaminopropyl-
3-ethyl carbodiimide hydrochloride (22 mg, 0.11 mmol), and
4-(dimethylamino)pyridine (28 mg, 0.23 mmol). The reaction
mixture was stirred overnight, and then the methylene
chloride was evaporated. The reaction was diluted with ether
(10 mL), washed with water (2 × 3 mL), dried over MgSO4,
filtered, and concentrated. The crude mixture was purified by
silica gel chromatography (ether/hexanes 50:50) to give 50 mg
of (2-{2,6-di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxy-phenyl-
sulfanyl)-1-methyl-ethylsulfanyl]phenoxy}acetylamino)acetic
acid ethyl ester, which was dissolved in THF/H2O/MeOH (2:
1:1, 1 mL), lithium hydroxide monohydrate (15 mg, 0.36 mmol)
was added, and the reaction was stirred for 1 h. The reaction
was neutralized with 1 N HCl and extracted with ether (2 ×
10 mL), dried over MgSO4, filtered, and concentrated to give
the title product as a viscous residue (25 mg, 45%) which
1
solidified, mp 91-94 °C. H NMR (CDCl3) δ 7.56 (s, 2H, Ph-
H), 7.42 (s, 2H, Ph-H), 7.28 (t, 1H, NH), 5.39 (br s, 1H, Ph-
OH), 4.31 [s, 2H, OCH2C(O)], 4.22 (d, J ) 5.2 Hz, 2H,
NHCH2COOH), 1.44 (s, 6H, S,S′-isopropylidene), 1.42 (s, 18H,
tert-butyls), 1.39 (s, 18H, tert-butyls). MS m/z: 654 ([M + Na]+,
100%). Anal. (C35H53NO5S2‚1/2H2O) C, H, N, S.
[(2-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxy-
phenylsulfanyl)-1-methylethylsulfanyl]phenoxy}acetyl)-
methylamino]acetic Acid (4ak). This compound was pre-
pared in a similar procedure as for 4aj. Off-white solid, mp
1
115-119 °C. H NMR (CDCl3) δ 7.58 (s, 2H, Ph-H), 7.45 (s,
2H, Ph-H), 5.37 (s, 1H, Ph-OH), 4.52 (s, 2H, PhOCH2), 4.24
(s, 2H, NCH2), 3.03 (s, 3H, NCH3), 1.47 (s, 6H, S,S′-isopropyl-
idene), 1.44 (s, 18H, tert-butyls), 1.43 (s, 18H, tert-butyls). MS
m/z: 668 ([M + Na]+, 100%). Anal. (C36H55NO5S2) C, H, N, S.
2-(2-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxy-
phenylsulfanyl)-1-methyl-ethylsulfanyl]phenoxy}acetyl-
amino)pentanedioic Acid (4al). This compound was pre-
1
pared in a similar procedure as for 4aj. Viscous residue. H
NMR (CDCl3) δ 7.57 (s, 2H, Ph-H), 7.52 (d, 1H, NH), 7.42 (s,
2H, Ph-H), 5.37 (s, 1H, PH-OH), 4.81-4.88 (m, 1H, NHCH-
COOH), 4.28 (s, 2H, PhOCH2), 2.53-2.59 (m, 2H, CH2CH2-
COOH), 1.39-1.45 (m, 44H, S,S′-isopropylidene, tert-butyls,
CH2CH2COOH). MS m/z: 742 ([M + K]+, 70%), 159 (100%).
Anal. (C38H57NO7S2) C, H, N, S.
2-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxyphen-
ylsulfanyl)-1-methyl-ethylsulfanyl]phenoxy}-N-(2-hydroxy-
1-hydroxymethyl-ethyl)acetamide (4am). To a solution of
4ad (4.87 g, 8.49 mmol) in dichloromethane (200 mL) were
added serinol (0.77 g, 8.49 mmol) and 1-(3-dimethylaminopro-
pyl)-3-ethylcarbodiimide hydrochloride (3.26 g, 17 mmol), and
the mixture was stirred at room temperature overnight. It was
poured into water (100 mL), and the organic phase was dried
over magnesium sulfate and evaporated. Silica gel chroma-
tography (hexanes/ethyl acetate 1:2) gave the title compound
as a white solid (2.95 g, 57%), mp 163-164 °C. 1H NMR
(CDCl3) δ 7.57 (s, 2H, Ph-H), 7.49 (d, J ) 7, 1 H, NH), 7.44 (s,
2H, Ph-H), 5.37 (s, 1H, Ph-OH), 4.28 (s, 2H, PhOCH2), 4.10-
4.29 (m, 1H, NHCH), 3.82-4.00 (m, 4H, CH2OH), 1.45 (s, 6H,
S,S′-isopropylidene), 2.44 (dd, J ) 6, 11, 2H, OH), 1.44 (s, 18H,
tert-butyls), 1.41 (s, 18H, tert-butyls). MS m/z: 670 ([M + Na]+,
100%). Anal. (C36H57NO5S2) C, H, N, S.
3-(2-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxy-
phenylsulfanyl)-1-methylethylsulfanyl]phenoxy}acetyl-
amino)propionic Acid Ethyl Ester (4ah). This compound
was prepared in a similar procedure as for 4aj. Off-white solid,
1
mp 116-117 °C H NMR (CDCl3) δ 7.56 (s, 2H, Ph-H), 7.43
(s, 2H, Ph-H), 7.31 (t, J ) 6, 1H, NH), 5.37 (s, 1H, Ph-OH),
4.24 (s, 2H, PhOCH2), 4.17 (quad, J ) 7, 2H, OCH2CH3), 3.69
(quad, J ) 6, 2H, NHCH2), 2.61 (t, J ) 6, NHCH2CH2), 1.45
(s, 6H, S,S′-isopropylidene), 1.44 (s, 18H, tert-butyls), 1.39 (s,
18H, tert-butyls), 1.28 (t, J ) 7, 3H, OCH2CH3). MS m/z: 696
([M + Na]+, 100%). Anal. (C38H59NO5S2) C, H, N.
{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxy-
phenylsulfanyl)-1-methylethylsulfanyl]phenoxy}-
acetonitrile (4an). To a solution of probucol (2.0 g, 3.9 mmol)
and iodoacetonitrile (1.1 g, 6.8 mmol) in tetrahydrofuran (40
mL) was added potassium fluoride (1.1 g, 7.7 mmol, 40% on
aluminum), and the resulting mixture was stirred at room
3-(2-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxy-
phenylsulfanyl)-1-methylethylsulfanyl]phenoxy}-
acetylamino)propionic Acid (4ai). This compound was
prepared in a similar procedure as for 4aj. Off-white solid, mp

Phenolic Antioxidants as Inhibitors of VCAM-1 Expression
Journal of Medicinal Chemistry, 2004, Vol. 47, No. 25 6431
temperature for 1 h. The reaction mixture was filtered through
a pad of Celite and rinsed with fresh portions of ethyl acetate
(100 mL). The filtrate was washed with a saturated sodium
bisulfite solution (4 × 75 mL) and brine (1 × 75 mL), dried
over sodium sulfate, and concentrated under reduced pressure
to a dark oil. Silica gel chromatography (10% ethyl acetate/
hexanes) afforded 0.35 g (20%) of the title compound as a pale
yellow solid, mp 139-141 °C. Rf 0.27 (10% ethyl acetate/
hexanes). 1H NMR (300 MHz, CDCl3) δ 7.59 (s, 2H, Ph-H),
7.43 (s, 2H, Ph-H), 5.37 (s, 1H, Ph-OH), 4.48 (s, 2H, CH2-
CN), 1.42-1.45 (m, 42H, tert-butyls and S,S′-isopropylidene).
LRMS (ESI) m/z (%) 555 (M + Na, 100). HRMS (ESI) calcd
for C33H49NO2S2 (M + K), 594.2842; found, 594.2841. Anal.
(C33H49NO2S2) C, H, N, S.
LMB Assay. A published procedure was followed (only the
method described in Figure 4 of the reference was used).18
Cytokine Screening. Fresh cryopreserved hPBMCs (1
million cells/mL; Clonetics, Inc.) were pretreated with test
compound for 1 h in lymphocyte growth media-3 (Clonetics,
Inc.), followed by stimulation with 1 µg/mL of LPS for another
2 h in the presence of test compound. Conditioned media was
collected and assayed for secreted TNF-R, IL-6 and IL-1â using
commercially available human ELISA kits (R&D Systems).
Samples were measured in duplicate and data presented as
mean ( standard deviation. Each experiment was repeated
at least three times with similar results.
DTH Model of Inflammation in Mice. Male BALB/c mice
(20-25 g) were sensitized on day 0 by intradermal abdominal
injection of 100 µL of a 1:1 emulsion of mBSA (5 mg/mL) and
Freund’s complete adjuvant. On day 7, the mice were chal-
lenged by injecting 25 µL of mBSA (5 mg/mL) into the footpad
of the right hindpaw, and the left hindpaw was injected with
saline. Mice were treated by subcutaneous injection of test
compound 24 h and 2 h before the challenge. Twenty four
hours after challenge mice were sacrificed by CO2 inhalation.
The feet were removed by cutting just above the heel with
scissors, and the mass of the saline-injected foot was sub-
tracted from the mass of the mBSA-injected foot to determine
the amount of swelling that occurred in the latter. Inhibition
of swelling for test compound group was calculated, taking the
swelling of the vehicle-treated group as 100%.
Hamster Lipid Screening. Male Golden Syrian hamsters
weighing 110-120 g (Charles River Laboratories, Wilmington,
MA) were housed individually with wood chip bedding and soft
nesting material. Water was provided ad libitum via an
automatic watering system. Hamsters were acclimated for at
least 3 days and fed standard rodent chow (Purina 5001)
during that time. Before the start of the study, hamsters were
weighed and distributed into control (n ) 10) and treatment
groups (n ) 5) such that each group had similar average
weights. Prior to dosing with compounds, the hamsters were
made hypercholesterolemic by feeding rodent chow supple-
mented with 0.5% cholesterol and 10% coconut oil (Test Diet
97235, Harlan Teklad, Madison, WI) for one week. Subse-
quently, the hamsters were fed admixtures of Test Diet by first
forming a suspension of the compound in 0.5% methylcellulose/
0.1% Tween-20 (Sigma) with a Polytron homogenizer (PT2100,
Kinematica, Lucerne, Switzerland). An amount of this suspen-
sion was then added to 20 g of powdered chow so as to
administer a dose of test compound to the animals of 150 mg/
kg/d. The chow/compound mixture was then rolled into balls,
transferred to stainless steel bowls, and placed at the bottom
of the cages. Chow was usually entirely consumed by the next
day. The animals were dosed with compounds for two weeks.
At the completion of the study, hamsters were fasted overnight
by removing chow from cheek pouches and transferring the
animals to clean cages. Hamsters were then anesthetized
intraperitoneally with a solution of ketamine (100 mg/kg) and
xylazine (10 mg/kg). Once the animals were fully sedated but
still respiring, blood was collected for lipid analysis by cardiac
puncture or via the abdominal aorta using a 3 mL syringe and
18 gauge needle. Plasma was fractionated by fast phase liquid
chromatography, and cholesterol levels in the different lipo-
protein fractions were determined by an enzymatic assay as
described previously.26 Statistical significance between treated
and control groups was determined by the Dunnett’s test. An
experimental group was considered statistically different than
control when p < 0.05.
2-{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-hydroxy-
phenylsulfanyl)-1-methylethylsulfanyl]phenoxy}-N-
hydroxyacetamidine (4ao). Hydroxylamine hydrochloride
(0.032 g, 0.47 mmol) was added to
a
mixture of
triethylamine (0.047 g, 0.47 mmol) in 50% aqueous ethanol
(0.5 mL), and the mixture was stirred for five minutes.
Compound 4an (0.20 g, 0.36 mmol) in ethanol (2.0 mL)
was added, and the reaction was refluxed for 5 h, cooled
to room temperature, and diluted with water (5 mL). The
resulting precipitate was collected on filter paper and rinsed
with water. The crude material was tritarated with hexanes
to remove any residual probucol starting material and dried
in vacuo to furnish 0.20 g (95%) of the title compound
as a pale yellow solid, mp 185-186 °C. Rf 0.1 (10% ethyl
acetate/hexanes). 1H NMR (300 MHz, CDCl3) δ 7.57 (s,
2H, Ph-H), 7.44 (s, 2H, Ph-H), 5.95 (s, 1H, N-OH), 5.36
(s, 1H, Ph-OH), 4.98 (brs, 2H, NH2), 4.33 (s, 2H, CH2),
1.42-1.46 (m, 42H, tert-butyls and S,S′-isopropylidene).
LRMS (ESI) m/z (%) 589 (M + H, 50), 351 (100). HRMS
(ESI) calcd for C33H52N2O3S2 (M + H), 589.3497; found,
589.3503. Anal. (C33H52N2O3S2) C, H, N, S.
{2,6-Di-tert-butyl-4-[1-(3,5-di-tert-butyl-4-carboxymeth-
oxyphenylsulfanyl)-1-methylethylsulfanyl]phenoxy}-
acetic Acid (5a). This compound was obtained as a side
product in a large-scale synthesis of 4ad, a white solid, mp
250-253 °C. 1H NMR (300 MHz, DMSO-d6) δ 7.48 (s, 4H, Ph-
H), 4.25 (s, 4H, CO-CH2-O), 1.42 (s, 6H, S,S′-isopropylidene),
1.37 (s, 36H, tert-butyls). HRMS (ESI) calcd for C35H52O6S2
(M + K), 671.2842; found, 671.2825. Anal. (C35H52O6S2) C, H,
S.
In Vitro VCAM-1 Assay. Cultured primary human aortic
(HAEC) or pulmonary artery (HPAEC) endothelial cells were
obtained from Clonetics and were used below passage 9. Cells
were seeded in 96-well plates such that they reached 90-95%
confluency by the following day. On the following day cells
were stimulated with TNF-R (1 ng/mL) in the presence or
absence of compounds dissolved in DMSO such that the final
concentration of DMSO was 0.25%. Cells were exposed to
TNF-R and compounds for approximately 16 h. Then cells were
examined under microscope to score for visual sign of toxicity
or cell stress. The media was discarded, and cells were washed
once with Hanks Balanced Salt Solution (HBSS)/phosphate-
buffered saline (PBS) (1:1). Primary antibody against VCAM-1
(0.25 µg/mL in HBSS/PBS + 5% FBS) was added, and cells
were incubated for 30 min at 37 °C. Cells were washed with
HBSS/PBS three times, then secondary antibody Horseradish
Peroxidase (HRP)-conjugated goat anti-mouse IgG (1:500 in
HBSS/PBS + 5% FBS) was added, and cells were incubated
for 30 min at 37 °C. Cells were then washed with HBSS/PBS
four times and 3,3′,5,5′-tetramethylbenzidine (TMB) was
added, and cells were incubated at room temperature in the
dark until there was adequate development of blue color. The
length of time of incubation was typically 5-15 min. Sulfuric
acid (2 N) was added to stop the color development, and the
data were collected by reading the absorbance on a BioRad
ELISA plate reader at OD 450 nm. The results were expressed
as IC50 values (the concentration of test compound required
to inhibit 50% of the maximal response of the control sample
stimulated by TNF-R only). IC50 numbers reflect an average
of at least two determinations.
References
(1) McMurray, R. W. Adhesion Molecules in Autoimmune Disease.
Semin. Arthritis Rheum. 1996, 25, 215-234.
(2) Carter, R. A.; Wicks, I. P. Vescular Cell Adhesion Molecule-1
(CD106): A Multifaceted Regulator of Joint Inflammation.
Arthritis Rheum. 2001, 44, 985-994.
(3) Bevilacqua, M. P.; Nelson, R. M.; Mannori, G.; Cecconi, O.
Endothelial-Leukocyte Adhesion Molecules in Human Disease.
Annu. Rev. Med. 1994, 45, 361-378.

6432 Journal of Medicinal Chemistry, 2004, Vol. 47, No. 25
Meng et al.
(4) Medford, R. M. Antioxidants and Endothelial Expression of
VCAM-1: A Molecular Paradigm for Atherosclerosis. In Car-
diovascular Disease 2; Gallo, L. L., Ed.; Plenum Press: New
York, 1995; pp 121-127.
(17) Fruebis, J.; Gonzalez, V.; Silvestre, M.; Palinski, W. Effects of
Probucol Treatment on Gene Expression of VCAM-1, MCP-1,
and M-CSF in the Aortic Wall of LDL Receptor-Deficient Rabbits
during Early Atherosclerosis. Arteroscler. Thromb. Vasc. Biol.
1997, 17, 1289-1302.
(18) Somers, P. K.; Medford, R. M.; Saxena, U. Dithiocarbamates:
Effects on Lipid Hydroperoxides and Vascular Inflammatory
Gene Expression. Free Radic. Biol. Med. 2000, 28, 1532-1537.
(19) Haddad, J. J.; Saade, N. E.; Safieh-Garabedian, B. Redox
regulation of TNF-R Biosynthesis: Augmentation by Irreversible
Inhibition of γ-Glutamylcysteine Synthetase and the Involve-
ment of an IκB-R/NF-κB-independent Pathway in Alveolar
Epithelial Cells. Cell Signal. 2002, 14, 211-218.
(20) Kunsch, C.; Luchoomun, J.; Grey, J. Y.; Olliff, L. K.; Saint, L.
B.; Arrendale, R. F.; Wasserman, M. A.; Saxena, U.; Medford,
R. M. Selective Inhibition of Endothelial and Monocyte Redox-
Sensitive Genes by AGI-1067: A Novel, Antioxidant and Anti-
inflammatory Agent. J. Pharmacol. Exp. Ther. 2004, 308, 820-
829.
(21) Sarnstrand, B.; Jansson, A.-H.; Matuseviciene, G.; Scheynius,
A.; Pierrou, S.; Bergstrand, H. N, N′-Diacetyl-L-Cystinesthe
Disulfide Dimer of N-acetylcysteinesIs a Potent Modulator of
Contact Sensitivity/Delayed-Type Hypersensitivity Reactions in
Rodents. Pharmacol. Exp. Ther. 1999, 288, 1174-84.
(5) Marui, N.; Offermann, M. K.; Swerlick, R.; Kunsch, C.; Rosen,
C. A.; Ahmad, M.; Alexander, R. W.; Medford, R. M. Vascular
Cell Adhesion Molecule-1 (VCAM-1) Gene Transcription and
Expression Are Regulated through an Antioxidant-Sensitive
Mechanism in Human Vascular Endothelial Cells. J. Clin.
Invest. 1993, 92, 1866-1874.
(6) Meng, C. Q. Probucol (Restenosis). Curr. Opin. Cardiovasc.
Pulm. Renal Invest. Drugs 2000, 2, 294-298.
(7) Barnhart, J. W.; Wagner, E. R.; Jackson, R. L. The Synthesis,
Metabolism, and Biological Activity of Probucol and Its Ana-
logues. In Antilipidemic Drugs: Medicinal, Chemical, and
Biochemical Aspects; Witiak, D. T., Newman, H. A. I., Feller, D.
R., Eds.; Elsevier Science: Amsterdam, 1991; pp 277-299.
(8) Carew, T. E.; Schwenke, D. C.; Steinberg, D. Antiatherogenic
Effect of Probucol unrelated to its Hypocholesterolemic Effect:
Evidence That Antioxidants in vivo Can Selectively Inhibit Low-
Density Lipoprotein degradation in Macrophage-rich Fatty
Streaks and Slow the Progression of Atherosclerosis in the
Watanabe Heritable Hyperlipidemic Rabbit. Proc. Natl. Acad.
Sci. U.S.A. 1987, 84, 7725-7729.
(9) Tardif, J. C.; Cote, G.; Lesperance, J.; Bourassa, M.; Lambert,
J.; Doucet, S.; Bilodeau, L.; Nattel, S.; De Guise, P. Probucol
and Multivitamins in the Prevention of Restenosis after Coro-
nary Angioplasty. Multivitamins and Probucol Study Group.
New Engl. J. Med. 1997, 337, 365-372.
(10) Klein, L. QT-Interval Prolongation Produced by Probucol. Arch.
Intern. Med. 1981, 141, 1102-1103.
(11) Meng, C. Q.; Somers, P. K.; Rachita, C. L.; Holt, L. A.; Hoong,
L. K.; Zheng, X. S.; Simpson, J. E.; Hill, R. R.; Olliff, L. K.;
Kunsch, C.; Sundell, C. L.; Parthasarathy, S.; Saxena, U.;
Sikorski, J. A.; Wasserman, M. A. Novel Phenolic Antioxidants
as Multifunctional Inhibitors of Inducible VCAM-1 Expression
for Use in Atherosclerosis. Bioorg. Med. Chem. Lett. 2002, 12,
2545-2548.
(22) Medford, R. M.; Saxena, U. Methods and Composition to Lower
Plasma Cholesterol Levels. WO 0028332, 2000.
(23) Sundell, C. L.; Suen, K. L.; Kunsch, C.; Moore, A. R. AGIX-
4207: A Novel Antioxidant and Antiinflammatory Compound
Inhibits Progression of Collagen II Arthritis in the Rat. FASEB
J. 2002, 16, A182. Sundell, C. L.; Howard, R.; Karu, K.;
Luchoomun, J.; Qin, F. H.; Sikorski, J.; Stefanopoulos, D.;
Wasserman, M. A.; Kunsch, C. AGIX-4207, a Novel Antioxidant
and Anti-inflammatory Compound: II. Attenuation of Inflam-
mation and Preservation of Joint Structure in Rat Collagen-
induced Arthritis. J. Pharmacol. Exp. Ther., submitted.
(24) Murata, S.; Sundell, C. L.; Lijkwan, M. A.; Balsam, L. B.;
Ha¨mma¨inen, P.; Coleman,C.; York, C.; Luchoomun, J.; Suen, K.
L.; Howard, R,; Somers, P. K.; Morris, R. E,; Robbins, R. C.
Effects of AGI-1096, a Novel Antioxidant Compound with
Antiinflammatory and Anti-proliferative Properties, on Rodent
Allograft Arteriosclerosis. Transplantation 2004, 77, 1491-1500.
(25) Kim, H. S.; Barek, D.; Harden, T. K.; Boyer, J. L.; Jacobsen, K.
A. Acyclic and Cyclopropyl Analogues of Adenosine Bisphosphate
Antagonists of the P2Y1 Receptor: Structure-Activity Relation-
ships and Receptor Docking. J. Med. Chem. 2001, 44, 3092-
3108.
(12) Tardif, J. C.; Gregoire, J.; Schwartz, L.; Title, L.; Laramee, L.;
Reeves, F.; Lesperance, J.; Bourassa, M. G.; L’Allier, P. L.; Glass,
M.; Lambert, J.; Guertin, M. C. Effects of AGI-1067 and Probucol
after Percutaneous Coronary Interventions. Circulation 2003,
107, 552-558.
(13) Mitsunobu, O. The Use of Diethyl Azodicarboxylate and Triphen-
ylphosphine in Synthesis and Transformation of Natural Prod-
ucts. Synthesis 1981, 1-28.
(14) Nicolaou, K. C.; Papahatjis, D. P.; Claremon, D. A.; Magolda, R.
L.; Dolle, R. E. Total Synthesis of Ionophore Antibiotic X-14547A,.
J. Org. Chem. 1985, 50, 1440-1456.
(15) Camps, P.; Cardellach, J.; Font, J.; Prtuno, R. M.; Ponsati, O.
Studies on Structurally Simple R, â-Butenolides-II. Tetrahetron
1982, 38, 2395-2402.
(26) Innis-Whitehouse, W.; Li, X.; Brown, V. W.; Le, N. A. An Efficient
Chromatographic System for Lipoprotein Fractionation Using
Whole Plasma. J. Lipid Res. 1998, 39, 679-690.
(16) Gresham, T. L.; Jansen, J. E.; Shaver, F. W.; Bankert, R. A.;
Beears, W. L.; Prendergast, M. G. â-Propiolactone. VI. Reactions
with Phenols, Thiophenols and Their Salts. J. Am. Chem. Soc.
1949, 71, 661-663.
JM049685U