Chemoprotective properties of phenylpropenoids, bis(benzylidene)cycloalkanones, and related Michael reaction acceptors: Correlation of potencies as phase 2 enzyme inducers and radical scavengers

Article abstract of DOI:10.1021/jm980424s

Induction of phase 2 enzymes (e.g., glutathione transferases, NAD(P)H:quinone reductase, glucuronosyltransferases, epoxide hydrolase) is a major strategy for reducing the susceptibility of animal cells to neoplasia and other forms of electrophile toxicity. In a search for new chemoprotective enzyme inducers, a structure-activity analysis was carried out on two types of naturally occurring and synthetic substituted phenylpropenoids: (a) Ar- CH=CH-CO-R, where R is OH, OCH₃, CH₃, or Ar, including cinnamic, coumaric, ferulic, and sinapic acid derivatives, their ketone analogues, and chalcones; and (b) bis(benzylidene)cycloalkanones, Ar-CH=C(CH₂)(n)(CO)C=CH-Ar, where n = 5, 6, or 7. The potencies of these compounds in inducing NAD(P)H:quinone reductase activity in murine hepatoma cells paralleled their Michael reaction acceptor activity (Talalay, P.; De Long, M. J.; Prochaska, H. J. Proc. Natl. Acad. Sci. U.S.A. 85, 1988, 8261-8265). Unexpectedly, the bis(benzylidene)cycloalkanones also powerfully quenched the lucigenin- derived chemiluminescence evoked by superoxide radicals. Introduction of o- hydroxyl groups on the aromatic rings of these phenylpropenoids dramatically enhanced their potencies not only as inducers for quinone reductase but also as quenchers of superoxide. These potentiating o-hydroxyl groups are hydrogen-bonded, as shown by moderate down field shift of their proton NMR resonances and their sensitivities to the solvent environment. The finding that the potencies of a series of bis(benzylidene)cycloalkanones in inducing quinone reductase appear to be correlated with their ability to quench superoxide radicals suggests that the regulation of phase 2 enzymes may involve both Michael reaction reactivity and radical quenching mechanisms.

Full text of DOI:10.1021/jm980424s

J . Med. Chem. 1998, 41, 5287-5296
5287
Ch em op r otective P r op er ties of P h en ylp r op en oid s,
Bis(ben zylid en e)cycloa lk a n on es, a n d Rela ted Mich a el Rea ction Accep tor s:
Cor r ela tion of P oten cies a s P h a se 2 En zym e In d u cer s a n d Ra d ica l Sca ven ger s†
Albena T. Dinkova-Kostova,‡ Chitrananda Abeygunawardana,§,| and Paul Talalay*,‡
Departments of Pharmacology and Molecular Sciences and of Biological Chemistry, The J ohns Hopkins University School of
Medicine, 725 North Wolfe Street, Baltimore, Maryland 21205
Received J uly 21, 1998
Induction of phase 2 enzymes (e.g., glutathione transferases, NAD(P)H:quinone reductase,
glucuronosyltransferases, epoxide hydrolase) is a major strategy for reducing the susceptibility
of animal cells to neoplasia and other forms of electrophile toxicity. In a search for new
chemoprotective enzyme inducers, a structure-activity analysis was carried out on two types
of naturally occurring and synthetic substituted phenylpropenoids: (a) Ar-CHdCH-CO-R,
where R is OH, OCH3, CH3, or Ar, including cinnamic, coumaric, ferulic, and sinapic acid
derivatives, their ketone analogues, and chalcones; and (b) bis(benzylidene)cycloalkanones, Ar-
CHdC(CH2)n(CO)CdCH-Ar, where n ) 5, 6, or 7. The potencies of these compounds in inducing
NAD(P)H:quinone reductase activity in murine hepatoma cells paralleled their Michael reaction
acceptor activity (Talalay, P.; De Long, M. J .; Prochaska, H. J . Proc. Natl. Acad. Sci. U.S.A.
85, 1988, 8261-8265). Unexpectedly, the bis(benzylidene)cycloalkanones also powerfully
quenched the lucigenin-derived chemiluminescence evoked by superoxide radicals. Introduction
of o-hydroxyl groups on the aromatic rings of these phenylpropenoids dramatically enhanced
their potencies not only as inducers for quinone reductase but also as quenchers of superoxide.
These potentiating o-hydroxyl groups are hydrogen-bonded, as shown by moderate downfield
shift of their proton NMR resonances and their sensitivities to the solvent environment. The
finding that the potencies of a series of bis(benzylidene)cycloalkanones in inducing quinone
reductase appear to be correlated with their ability to quench superoxide radicals suggests
that the regulation of phase 2 enzymes may involve both Michael reaction reactivity and radical
quenching mechanisms.
In tr od u ction
to toxicity, by clonal selection for resistance, and by
overexpression of genes that code for protective phase
2 enzymes. Recently, transgenic mice lacking quinone
reductase or glutathione transferase were shown to be
more susceptible to quinone toxicity or carcinogenesis,
respectively.5,6 Our interest in the coordinate induction
of phase 2 enzymes in tissues arises from the fact that
this strategy reduces susceptibility to carcinogens and
the incidence of malignancy.
Mammalian cells have evolved elaborate biochemical
mechanisms for protecting DNA and other macromol-
ecules against damage by electrophiles and reactive
oxygen species that arise both from endogenous meta-
bolic processes and from exogenous sources. A particu-
larly important line of cellular defense against these
types of toxicities is the coordinate induction of phase
2 enzymes (e.g., glutathione S-transferases, EC 2.5.1.18;
NAD(P)H:quinone reductase, EC 1.6.99.2) which are
principally concerned with the metabolic deactivation
of these toxic agents, as well as elevation of glutathione
levels. These transcriptionally regulated inductions are
easily evoked by administration of a wide variety of
natural and synthetic chemical agents.1 Much persua-
sive evidence now indicates that elevation of phase 2
enzymes affords protection against the toxic and neo-
plastic effects of electrophiles and oxidative stress.2-4
Such protection has been confirmed by demonstrating
that cells subjected to chemical induction are resistant
Extensive efforts to elucidate the molecular mecha-
nism of induction of phase 2 enzymes have focused on
identifying the cis-acting regulatory sequences, the
trans-acting transcriptional regulatory factors, and the
chemical structures of the inducers.1,2,4,7-11 A simple
assay for determining inducer potency developed in our
laboratory involves measurement of quinone reductase
(a prototype for phase 2 detoxication enzymes) in
murine hepatoma cells grown in 96-well microtiter
plates.12,13 This system reliably predicted the enzyme
inducer activity of compounds in animal tissues. At least
eight chemically distinct classes of compounds are
inducers of quinone reductase in this system.1,2 Most
of these agents are monofunctional since they elevate
phase 2 enzymes selectively, without activating the
transcription of the Ah (Aryl hydrocarbon)-dependent
cytochromes P450.14 These chemical classes include (a)
o- and p-diphenols and phenylenediamines that can
undergo oxidations to quinones or quinonimines, re-
† Abbreviations and trivial names: quinone reductase, NAD(P)H:
(quinone acceptor) oxidoreductase, EC 1.6.99.2, also known as DT
diaphorase; CD value, concentration of an inducer required to double
the specific activity of quinone reductase in Hepa lclc7 murine
hepatoma cells.
* Corresponding author: Paul Talalay. Phone: (410)-955-3499.
Fax: (410)-502-6818. E-mail: ptalalay@welchlink.welch.jhu.edu.
‡ Department of Pharmacology and Molecular Sciences.
§ Department of Biological Chemistry.
| Present address: Merck & Co., Inc., West Point, PA 19486-0004.
10.1021/jm980424s CCC: $15.00 © 1998 American Chemical Society
Published on Web 11/25/1998

5288 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 26
Dinkova-Kostova et al.
spectively; (b) quinones; (c) various other Michael reac-
tion acceptors; (d) isothiocyanates; (e) hydroperoxides;
(f) vicinal dimercaptans; (g) trivalent arsenicals; (h) 1,2-
dithiole-3-thiones; and (i) divalent heavy metals. The
inducer potencies of these compounds, expressed as the
concentrations required to double quinone reductase-
specific activities in murine hepatoma cells (CD values),
span more than 4 orders of magnitude. There is no
obvious common chemical or structural theme that can
easily explain the activities and potencies of all of these
classes of inducers. Their structural diversity and
universal chemical reactivity suggest that passive ligan-
ding to a structurally complementary receptor is un-
likely to account for their mode of action. Many of the
inducers are electrophiles, nearly all can react with
sulfhydryl groups by oxidation or electrophilic attack,
and inducer potency appears to correlate with their
ability to participate as substrates for glutathione
transferases.15 Whereas all inducers are chemically
reactive, it is curious that some are powerful oxidants
(e.g., peroxides, hydroperoxides) and can cause oxidative
damage, others are obviously antioxidants (e.g., BHA,
BHT, 2,3-dimercaptopropanol), and a third group is
apparently unable to participate directly in oxidoreduc-
tion reactions (e.g., isothiocyanates). It is possible that
these inducers act by different mechanisms, but recent
work from several laboratories16,17 suggests that oxi-
dants as well as antioxidants and intracellular redox
state can regulate gene expression and specifically
transcription of phase 2 enzymes.
dimeric Michael reaction acceptors and have been
shown to exhibit antineoplastic activity.28 One of the
earliest reports on the synthesis of such compounds was
published in 1912,29 and subsequently these compounds
have attracted considerable interest in various and
sometimes intriguing ways. Some members of this group
have found pharmaceutical application as components
of choleretic drugs, e.g., 2,6-bis(3-methoxy-4-hydroxy-
benzylidene)cyclohexanone also known as cyclovalone.30
Others have been specifically developed as very sensi-
tive pH indicators.31 Borden32 at Eastman Kodak syn-
thesized a large number of bis(benzylidene)cycloal-
kanones as materials for condensation into light-
sensitive phenolic polymers suitable for photolithographic
purposes. Konrad and Mo¨ller33 at Henkel subsequently
developed these substances as sunscreens for the UV-A
range (320-405 nm). Most recently, these agents were
rediscovered by Markaverich and his colleagues as
esterase-insensitive ligands for nuclear type II estrogen
receptors and as inhibitors of cell proliferation.28,34
In the present study, several bis(benzylidene)cycloal-
kanone derivatives were synthesized and tested as
inducers of quinone reductase with two specific ques-
tions in mind: (a) What is the significance of the
substituents on the aromatic rings for the inducer
activity? and (b) Is the size of the alkanone ring
important? In order to answer these questions a detailed
structure-activity study was undertaken. First, we
examined several “partial molecules”, all of which belong
to the class of plant phenylpropenoids and their deriva-
tives, i.e., cinnamic acids and methyl cinnamates; cou-
maric acids and methyl coumarates; coumarins; and
chalcones. The lessons that we learned from these
“partial molecules” subsequently enabled us to identify
bis(benzylidene)cycloalkanones with extraordinary high
quinone reductase inducer potency.
A number of phenolic antioxidants, some of which are
commonly used as food additives (e.g., 2(3)-tert-butyl-
4-hydroxyanisole (BHA), 3,5-di-tert-butyl-4-hydroxytolu-
ene (BHT)) have been shown to protect against carcino-
genesis and mutagenesis.35-39 The relationship between
their antioxidant properties and their ability to serve
as chemoprotective agents is probably not a coincidence,
since free radical-mediated processes are believed to be
associated with the occurrence of many human diseases,
including cancer.40 This prompted us to examine the free
radical-scavenging capacity of several quinone reductase
inducers obtained in the course of this study. We found
that among these compounds some of the most potent
inducers display free radical-scavenging activity and
that there is good correlation between their potencies
as inducers of phase 2 enzymes and their free radical-
scavenging capacity.
Michael reaction acceptors (i.e., olefins or acetylenes
that are conjugated to electron-withdrawing groups) are
a major group of inducers with potencies paralleling the
strength of the electron-withdrawing groups and con-
sequently their rate of reaction with nucleophiles.1
Interestingly, and perhaps not merely by coincidence,
Michael reaction acceptor groups are present in many
natural products; for example, the plant phenylpro-
panoids and phenylpropenoids and analogous com-
pounds perform critical physiological functions in the
defense strategies of all vascular plants.18 Moreover,
representatives of this class, e.g., cinnamic, ferulic,
caffeic, coumaric, and sinapic acids, lignans, and chal-
cones, as well as coumarins and flavonoids, have been
reported to display a wide variety of pharmacological
properties, e.g., antioxidant,19,20 anticancer,21-23 anti-
mutagenic,24 and antimalarial25 activities. One cannot
help wondering whether all of these pharmacological
properties depend on distinct mechanism(s), or whether,
and perhaps more probably, there is a common mecha-
nistic link between these properties that has hitherto
eluded identification. Search for this missing link is
important, especially since many of these plant con-
stituents eventually become part of the human diet, and
some are even ingested in substantial quantities, ap-
proaching 1 g/day.26 In addition, epidemiological and
migrational studies have revealed a correlation between
high-fiber diet and low incidence of several types of
cancer, e.g., stomach, colon, breast, and prostate can-
cers,27 which has led to the suggestion that certain plant
phenylpropanoids and their derivatives may serve as
natural cancer-protective substances.
Resu lts a n d Discu ssion
P h en ylp r op en oic Acid s a n d Th eir Ester s a s
In d u cer s of Qu in on e Red u cta se. Earlier studies on
the phase 2 enzyme inducer potencies of Michael reac-
tion acceptors of the type phenyl-CHdCH-CO-R had
shown that these potencies paralleled the electron-
withdrawing powers of the R groups: i.e., ketones (R )
alkyl or aryl) were more potent than esters (R )
O-alkyl), and carboxylic acids (R ) OH) were barely
active.1 In further studies in this laboratory,15 a series
Bis(benzylidene)cycloalkanones are structurally re-
lated to plant phenylpropenoids. They can be viewed as

Protective Properties of Phenylpropenoids
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 26 5289
Ta ble 1. Potencies of Phenylpropenoic Acids and Their Methyl
Esters as Inducers of Quinone Reductase in Murine Hepatoma
Cellsa
of methyl esters of cinnamic acid substituted with
electron-withdrawing or electron-donating groups on the
aromatic ring were synthesized. Their potencies in
inducing quinone reductase activity revealed a general
trend: electron-withdrawing substituents at the meta-
position enhanced the inducer activity, the methyl esters
of 3-bromo-, 3-nitro- and 3-chlorocinnamates being the
most active. Moreover, the order of potency correlated
linearly with the Hammett σ and σ- values of these
substituents. The chemical approach for methylation
that was used,15 namely, refluxing of methanolic solu-
tions of the corresponding cinnamic acids in the pres-
ence of boron trifluoride etherate, did not, however,
permit synthesis of the methyl esters of cinnamic acids
bearing additional aromatic hydroxyl substituents. Be-
cause hydroxylated phenylpropenoids and their deriva-
tives are ubiquitous in nature and there are many
reports of their antioxidant, antiinflammatory, antimu-
tagenic, and antitumor activities, it was important to
obtain the methyl esters of these hydroxylated cin-
namates.
By means of a facile and selective method, involving
the refluxing of methanolic solutions of the correspond-
ing cinnamic acids in the presence of small amounts of
a cation-exchange resin, we obtained selective methy-
lation of the carboxyl groups of the four most naturally
abundant substituted phenylpropenoic acids and tested
these compounds for their ability to induce quinone
reductase activity in cultured Hepa 1c1c7 murine
hepatoma cells.
As expected, the arylpropenoic acids themselves, i.e.,
cinnamic (1a ), caffeic (5a ), ferulic (6a ), and sinapic (7a )
acids (Table 1), were inactive as inducers in agreement
with their weak Michael acceptor reactivity, in confir-
mation of previous observations.1,41 Their methyl esters
(1b, 6b, and 7b) were weak inducers (CD ) 82-93 µM),
except for methyl caffeate (5b), which was moderately
potent (CD ) 22 µM) (Table 1).
a
The CD values are shown in parentheses.
Ta ble 2. Potencies of Coumarin Derivatives as Inducers of
Quinone Reductase in Murine Hepatoma Cellsa
Among coumaric (hydroxycinnamic) acids (Table 1),
p-coumaric (2a ) and m-coumaric (3a ) acids were also
inactive as inducers. Surprisingly, however, o-coumaric
acid (4a ) showed moderate inducer potency (CD ) 19
µM). The corresponding methyl esters had inducer
potencies similar to those of the methyl esters of the
cinnamic acids, again the only exception being methyl
o-coumarate (4b), which was a slightly more potent
enzyme inducer (CD ) 15 µM). The last finding suggests
that the relatively weak Michael acceptor activity of
o-coumaric acid (4a ) was compensated by the potent
contribution of the o-hydroxyl moiety. These results
provided the first demonstration that the presence of a
hydroxyl group at the ortho-position on the aromatic
ring of a Michael reaction acceptor (e.g., cinnamic acid
derivative) increases its inducer potency markedly. The
differences in inducer potencies between the acids and
methyl esters are unlikely to be due to solubility
differences alone for two reasons: (i) the p- and m-
hydroxycinnamic acids are inactive, whereas the ortho-
isomer is active; large differences in solubilities among
these isomers are unlikely; and (ii) there is a very small
inducer potency difference between o-methyl coumarate
(CD ) 15 µM) and o-coumaric acid (CD ) 19 µM).
a
The CD values are shown in parentheses.
tives were examined (Table 2). These represent another
large and diverse group of widely distributed plant
metabolites. Coumarins are biosynthesized from cin-
namic acid precursors in which the aromatic rings are
hydroxylated at the ortho-position, the side-chain double
bond undergoes light-catalyzed trans-cis isomerization,
and lactonization gives rise to the basic benzopyran-2-
one (cis-o-coumaric acid lactone) group of the coumarin
skeleton.42 Coumarin (8) itself is a very weak inducer
of quinone reductase (CD ∼ 500 µM),1 as is 7-hydroxy-
coumarin (9), the major metabolite of coumarin in
humans.43-45 3-Acetylcoumarin (10) is slightly more
potent (CD ∼ 200 µM). When the acetyl group was
replaced by a hydroxyl function, as in 3-hydroxycou-
marin (11), however, a remarkable increase of more
than 300-fold in inducer potency was observed (CD )
1.5 µM). It should be pointed out that this is the only
coumarin derivative examined here for which keto-enol
H yd r oxyla t ed Cou m a r in s a s In d u cer s of
Qu in on e Red u cta se. Next, several coumarin deriva-

5290 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 26
Dinkova-Kostova et al.
Ta ble 3. Potencies of Phenylbutenones and Chalcones as
Inducers of Quinone Reductase in Murine Hepatoma Cellsa
F igu r e 1. Concentration dependence of induction of quinone
reductase in murine hepatoma cells grown in 96-well micro-
titer plate wells by the following bis(benzylidene)cycloal-
kanones: (O) 2,5-bis(benzylidene)cyclopentanone (22); (b) 2,6-
bis(benzylidene)cyclohexanone (23); (4) 2,7-bis(benzylidene)-
cycloheptanone (24); (+) 2,5-bis(2-hydroxybenzylidene)cyclo-
pentanone (26); (3) 2,6-bis(2-hydroxybenzylidene)cyclohex-
anone (27); (]) 2,6-bis(2-methoxy benzylidene)cyclohexanone
(28). Each data point represents the average of eight replicates,
with standard deviation between 5% and 10% of its ordinate
value. The graph demonstrates the dramatic increase in
potency resulting from ortho-hydroxylation (compare 26 and
27 with 22 and 23).
group at the 2-position increased the inducer potency
approximately 3 times (2-hydroxychalcone (16), CD )
12 µM). This was consistent with the findings in the
methyl cinnamate series, where a ring hydroxyl group
had a strong impact on inducer potency only when
located at the ortho-position. A 2′-hydroxyl group (i.e.,
in 17) had a similar enhancing effect (CD ) 9.8 µM).
Moreover, the simultaneous presence of hydroxyl groups
at both 2- and 2′-positions (see 18, CD ) 4.7 µM) led to
further enhancement of inducer potency. Introduction
of a third hydroxyl group at the 4′-position (as in 2,2′,4′-
trihydroxychalcone (21) provided no further increase in
inducer potency (CD ) 4.8 µM). Interestingly, in the
chalcone series, a hydroxyl group at the 4′-position (19)
led to a 2-fold increase in inducer potency (CD ) 16 µM).
Recently 2-hydroxychalcone (16) (CD ) 12 µM) and 2′-
hydroxychalcone (17) (CD ) 9.8 µM) were shown to
inhibit proliferation of HeLa cells and were the most
potent among many other chalcone and related fla-
vonoid derivatives tested.50 In addition, it is noteworthy
that the related chalcone isoliquiritigenin (4,2′,4′-trihy-
droxychalcone (20), CD ) 11 µM), which occurs natu-
rally in bark and wood of some leguminous trees, was
found to be a very strong antitumor agent.46
Bis(ben zylid en e)cycloa lk a n on es a s In d u cer s of
Qu in on e Red u cta se. With these lessons from the
naturally existing “partial molecules” and their syn-
thetic analogues, we explored the structurally related
bis(benzylidine)cycloalkanones. They contain two Michael
reaction acceptor moieties making them good candidates
as potential quinone reductase inducers and chemopro-
tective agents. The unsubstituted compounds were
synthesized according to Rumpel,30 and their inducer
potencies were compared (Figure 1, Table 4). We found
that although the cyclopentanone derivative 22 was the
a
The CD values are shown in parentheses.
tautomerism of the molecule can occur. However, the
precise reason for the extraordinary increase in inducer
potency of 11 remains unclear.
Ch a lcon es a s In d u cer s of Qu in on e Red u cta se.
Another group of plant products that are structurally
and metabolically related to the aforementioned phe-
nylpropenoids are the chalcones, which may be regarded
as phenyl ketone analogues of cinnamic acid. The
rationale for examining these compounds was twofold:
ketones are more potent Michael acceptors than are
esters, and numerous reports in the literature describe
their antitumor,12,46,47 antiinflammatory,19,24 and anti-
mutagenic48 activities. Notably, structure-activity stud-
ies have revealed the absolute requirement for an
olefinic function for these biological activities.19,49 Fur-
ther, as part of a previous study1 it had been shown that
trans-4-phenyl-3-buten-2-one (12) (the methyl ketone
analogue of cinnamic acid) was a more potent inducer
(CD ) 15 µM) than was the corresponding ester, i.e.,
methyl cinnamate (1b ) (CD ) 82 µM), in agree-
ment with the order of reactivity of Michael reac-
tion acceptors. In the present study, we found that the
ortho-hydroxylated derivative of trans-4-phenyl-3-buten-
2-one (13) was 3 times more potent as an inducer (CD
) 5.3 µM) than was the nonhydroxylated compound 12
(Table 3).
The same correlation was found among the conven-
tional chalcones (Table 3). Hydroxyl substitution at
position 4 did not affect the inducer potency of chalcones
(compare chalcone (14), CD ) 31 µM, and 4-hydroxy-
chalcone (15), CD ) 33 µM). In contrast, a hydroxyl

Protective Properties of Phenylpropenoids
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 26 5291
Ta ble 4. Potencies of Bis(benzylidene)cycloalkanones as
Inducers of Quinone Reductase in Murine Hepatoma Cells and
as Scavengers of Superoxide Radicals Generated by the
Oxidation of Xanthine by Xanthine Oxidasea
oxybenzylidene)cyclohexanone (28), led to a large de-
crease in potency (CD ) 2.2 µM) to a level comparable
to that of the unsubstituted cyclohexanone derivative
23 (CD ) 2.9 µM) (Table 4). Although the cytotoxicities
of the bis(benzylidene)cycloalkanones have not been
extensively investigated, no detectable reduction in cell
mass was observed for compounds 22, 23, and 26-28
under conditions used to determine the CD values (up
to a concentration of 20 µM).
F r ee Ra d ica l-Sca ven gin g Activity of Qu in on e
Red u cta se In d u cer s. With these extremely potent
inducers in hand, we next sought a link between ability
to elevate quinone reductase levels and potential free
radical-scavenging activity. The rationale for these
experiments was based on the recognized role of free
radicals in the different stages of carcinogenesis. Fur-
thermore, many of the “partial molecule” inducers are
well-established antioxidants, and there is recent inter-
est in the antioxidant function of â-diketones, especially
those related to curcumin (diferuloylmethane).51
For this part of the study we utilized the oxidation of
xanthine by xanthine oxidase as a system to generate
superoxide and the inhibition of the lucigenin-derived
chemiluminescence as a measure of the free radical-
scavenging activity.52,53 Again, it was important to start
with the “partial molecules”, e.g., the chalcones. Al-
though it was not surprising to find that these com-
pounds are good antioxidants, it was quite unexpected
that free radical-scavenging activity in the xanthine/
xanthine oxidase system correlated with inducer po-
tency in the murine hepatoma system. For example,
chalcone (14) doubles the specific activity of quinone
reductase at a concentration of 31 µM and inhibits
chemiluminescence by 50% at 6.0 µM. Addition of a
hydroxyl group at the ortho-position, as in 17, which
increases the inducer potency 3-fold (CD ) 9.8 µM), also
leads to an increase in free radical scavenging, inhibit-
ing chemiluminescence by 50% at 2.5 µM. Simultaneous
presence of hydroxyl groups on both aromatic rings at
ortho-positions, which decreases the CD value to 4.7 µM
(see 18), leads to a proportional decrease in the concen-
tration needed for 50% inhibition of the chemilumines-
cence to 1.0 µM.
The next aim was to determine whether this correla-
tion is also valid for our most potent group of quinone
reductase inducers. Figure 2 represents a typical profile
of the lucigenin-derived chemiluminescence in the pres-
ence of increasing concentrations of 2,5-bis(2-hydroxy-
benzylidene)cyclopentanone (26). Table 4 shows the
inhibition of chemiluminescence (based on area under
the curve) at 50 nM concentration of the corresponding
bis(benzylidene)cycloalkanone. As can be seen, all the
bis(benzylidene)cycloalkanones examined in this study
are not only extremely potent quinone reductase induc-
ers but also exceptionally strong free radical scavengers.
Moreover, inducer potency correlates with free radical-
scavenging activity. Among the seven compounds (Table
4) for which measurements of both potency of induction
and quenching of luminescence at 50 nM are available,
the rank order of potencies in both assays are virtually
identical. To confirm that the observed inhibition of
chemiluminescence is due to free radical scavenging
only and not to inhibition of xanthine oxidase, we
compared the rates of enzymatic formation of uric acid
a
The potency order ranks for each type of measurement are
given in parentheses and are in good agreement. The values for
scavenging activity are expressed as percent quenching of lumi-
nescence of lucigenin at a 50 nM concentration of each compound.
least potent in this series (CD ) 16 µM), it was
comparable in potency to trans-4-phenyl-3-buten-2-one
(12) (CD ) 15 µM). The analogues containing the larger
six-membered cycloalkanone ring were considerably
more potent, but there was no significant difference in
potency between 2,6-bis(benzylidene)cyclohexanone (23)
(CD ) 2.9 µM) and 2,7-bis(benzylidene)cycloheptanone
(24) (CD ) 3.3 µM). Substituents at positions 3 and 4
of the aromatic rings were found to contribute further
to the inducer potency, which is exemplified by 2,6-bis-
(3-methoxy-4-hydroxybenzylidene)cyclohexanone (25)
(CD ) 0.9 µM).
We next examined whether an aromatic hydroxyl
group at position 2 would significantly affect inducer
potency, as observed in the plant phenylpropenoids we
had tested above. Thus, 2,5-bis(2-hydroxybenzylidene)-
cyclopentanone (26) and 2,6-bis(2-hydroxybenzylidene)-
cyclohexanone (27) were synthesized according to the
procedures of Borsche and Geyer.29 When tested as
quinone reductase inducers, 26 and 27 were found to
be dramatically more potent inducers (leading to at least
12 times increases in enzyme specific activity without
any measurable cytotoxicity): 2,6-bis(2-hydroxyben-
zylidene)cyclohexanone (27) (CD ) 280 nM) and 2,5-
bis(2-hydroxybenzylidene)cyclopentanone (26) (CD ) 75
nM) (Figure 1, Table 4). We confirmed that this dra-
matic increase is due specifically to the 2-hydroxyl group
by showing that its methylation as in 2,6-bis(2-meth-

5292 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 26
Dinkova-Kostova et al.
F igu r e 3. Designation of protons in 2,5-bis(2-hydroxyben-
zylidene)cyclopentanone (26). See also Table 5.
ring of 2,5-bis(benzylidene)cyclopentanone (22) at δ 3.14
and the proton at δ 7.61 (designated e in Table 5 and
Figure 3) indicated that these three protons are in
spatial proximity, and therefore the stereochemistry is
E-E. Similarly, the cross-peak observed between the
methylene protons of the cycloalkanone ring of 2,5-bis-
(2-hydroxybenzylidene)cyclopentanone (26) at δ 3.04
and the proton at δ 7.5 (designated e in Table 5 and
Figure 3) established that in this case also the geometry
was E-E. Similar results were obtained when 2,6-bis-
(benzylidene)cyclohexanone (23) and 2,6-bis(2-hydroxy-
benzylidene)cyclohexanone (27) were examined, thereby
establishing that the cyclohexanone derivatives also
have the E-E geometry.
B. Con for m a tion . Interestingly, it became clear from
the same experiment that the o-hydroxyl groups of bis-
(benzylidene)cycloalkanones are pointing “up,” i.e., away
from the methylenic protons (R and â in Figure 3) of
the cycloalkanone ring. Although this orientation was
expected because of steric hindrance, this conformation
was established unambiguously for 26 by 2D-NOESY
(CDCl3/DMSO-d6) through the observation of cross-
peaks between the o-hydroxyl proton (δ 9.49) and the
vinyl proton (δ 7.94), as well as between the o-hydroxyl
proton and the proton at δ 7.18 (designated b in Table
5 and Figure 3). In contrast, no cross-peak was observed
F igu r e 2. Concentration-dependent inhibition by 2,5-bis(2-
hydroxybenzylidene)cyclopentanone (26) of lucigenin-derived
chemiluminescence arising from superoxide generated by the
oxidation of xanthine by xanthine oxidase; A, no inhibitor; B,
50 nM; C, 100 nM; D, 500 nM; E, 1 µM; F, 10 µM.
from xanthine in the absence and presence of 20 µM
concentrations of the corresponding bis(benzylidene)-
cycloalkanones and found that these values were es-
sentially the same.
Str u ctu r e an d Con for m ation of Bis(ben zyliden e)-
cycloa lk a n on es. A. Molecu la r Geom etr y. Three
geometric isomers of the bis(benzylidene)cycloalkanones
can be envisaged, i.e., E-E, E-Z, and Z-Z.54 On the
basis of the symmetrical patterns of their 1H NMR
spectra, isomer E-Z was eliminated as a possibility. The
stereochemistry (E-E or Z-Z) was subsequently es-
tablished by 2D-NOESY experiments in a mixture of
CDCl3/DMSO-d6. Thus, the detection of a cross-peak
between the methylene protons of the cyclopentanone
Ta ble 5. Proton NMR Assignments and Chemical Shifts (ppm) of Bis(benzylidene)cyclopentanone and Bis(benzylidene)cyclohexanone
and Their Ortho-Hydroxylated Analogues

Protective Properties of Phenylpropenoids
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 26 5293
Ta ble 6. 13C NMR Chemical Shifts (ppm) of Carbon Atoms of Bis(benzylidene)cyclopentanone and Bis(benzylidene)cyclohexanone
and Their Ortho-Hydroxylated Analogues
between the o-hydroxyl proton and the methylene
protons (designated R and â in Table 5 and Figure 3) of
the cyclopentanone ring (δ 3.04) or between the o-
hydroxyl proton and the proton at δ 6.94 (designated d
in Table 5 and Figure 3).
potency as inducers of QR upon ortho-hydroxylation:
200- and 15-fold in 26 and 27, respectively. This result
would not be expected if the inducer potency depended
solely on the electrophilicity of the â-carbon of the vinyl
group. A similar paradoxical finding was the observation
that the inducer potencies of various cinnamates did not
correlate with the electrophilicity of the â-vinyl carbon
of substituted cinnamates, as determined by 13C NMR.15
The other differences between the 13C resonances as-
sociated with hydroxylation were confined to a +25 ppm
shift in the carbon atoms carrying the hydroxyl groups,
a -13 ppm shift in the carbon atoms flanking the
hydroxylated carbons, and a -10 ppm shift of the
4-carbon of the aromatic ring.
Since interaction with intracellular nucleophile(s),
e.g., thiols, would be expected to occur via Michael
addition on the vinyl carbon atoms, the presence of
o-hydroxyl groups must make a significant contribution
to the rate of addition of S- anion, although the exact
mechanism remains unknown. A parallel can be drawn,
however, with the antitumor cytotoxic sesquiterpene
lactones from Compositae isolated by Kupchan and
colleagues,55,56 in which a hydroxyl or O-acyl group
adjacent to the methylene function of these R-methyl-
ene-γ-lactones enhanced the rate of cysteine addition
enormously. This analogy is probably not surprising,
because both R-methylenecyclopentanone and R-meth-
ylene-γ-lactone have the same Michael reaction accep-
tor, namely, the electrophilic R-methylene carbonyl.
In unrelated studies, Rodriguez et al.57 have drawn
attention to the importance of molecular accessibility
factors in the regulation of the reactivity of R,â-unsatur-
ated carbonyl systems with sulfhydryl nucleophiles.
Further information on the conformations of these
1
compounds was provided by the H NMR resonances of
the phenolic protons of 2,5-bis(2-hydroxybenzylidene)-
cyclopentanone (26) and 2,6-bis(2-hydroxybenzylidene)-
cyclohexanone (27) (Table 5). These phenolic protons are
probably involved in weak hydrogen bonds (most likely
with the solvent or between two or more solute mol-
ecules), since their chemical shifts are observed near
9-10 ppm. In addition, these are the only protons of
26 and 27 whose chemical shifts are significantly
affected by the nature of the solvent; e.g., for 2,5-bis(2-
hydroxybenzylidene)cyclopentanone (26) the shift was
9.09 ppm in acetone-d6, 9.49 ppm in a mixture of CDCl3/
DMSO-d6 (2:8, by volume), and 10.38 ppm in DMF-d7.
C. Electr on Den sity a t Vin yl Ca r bon Atom s.
Since it had been previously established that the degree
of potency of many quinone reductase inducers paral-
leled their reactivity as Michael reaction acceptors,1 we
next used 13C NMR spectroscopy in conjunction with
HSQC (heteronuclear single quantum correlation) spec-
troscopy to determine the effect of the aromatic 2-hy-
droxyl groups on the electron density of the vinyl
carbons, which are the presumed points of attack by
intracellular nucleophiles. Significant upfield shifts of
the 13C resonances of the vinyl carbons of 2,5-bis(2-
hydroxybenzylidene)cyclopentanone (26) (by 5.7 ppm)
and 2,6-bis(2-hydroxybenzylidene)cyclohexanone (27)
(by 3.3 ppm) were observed in comparison to the
chemical shift of the vinyl carbons of the unsubstituted
analogues 22 and 23, respectively (Table 6). Surpris-
ingly, these increases in electron density at the vinyl
carbons were associated with very large increases in
Con clu sion s
The presence of hydroxyl groups at the ortho-posi-
tion(s) on the aromatic ring(s) of the phenylpropenoids

5294 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 26
Dinkova-Kostova et al.
(ppm) 7.61 (d, 1H, J ) 16 Hz, ph-CHd), 6.75 (s, 2H, arom),
6.32 (d, 1H, J ) 16 Hz, dCH-CO-), 5.73 (s, 1H, OH), 3.90 (s,
6H, CH3O), 3.78 (s, 3H, CH3). Methyl caffeate (5b): δ (ppm)
7.56 (d, 1H, J ) 16 Hz, ph-CHd), 7.05-6.84 (m, 3H, arom),
6.25 (d, 1H, J ) 16 Hz, dCH-CO-), 5.48 (s, 1H, OH), 5.36 (s,
1H, OH), 3.77 (s, 3H, CH3). Methyl ferulate (6b): δ (ppm) 7.60
(d, 1H, J ) 16 Hz, ph-CHd), 7.03-6.90 (m, 3H, arom), 6.28
(d, 1H, J ) 16 Hz, dCH-CO-), 5.89 (s, IH, OH), 3.91 (s, 3H,
CH3O), 3.77 (s, 3H, CH3).
2,6-Bis(benzylidene)cyclohexanone (23) and 2,5-bis(ben-
zylidene)cyclopentanone (22) were synthesized according to the
procedure of Rumpel.30 For the synthesis of 23, equimolar
amounts of cyclohexanone (0.98 g, 0.01 mol) and benzaldehyde
(1.08 g, 0.01 mol) were mixed and 3 drops of concentrated HCl
was added. The reaction was allowed to proceed for 40 h at
room temperature with stirring until solid product was formed.
The latter was crystallized from cold CH3COOH/H2O (1:1) to
yield long bright-yellow needles (1.9 g, 70%). The same
procedure, replacing cyclohexanone for cyclopentanone, was
used for the synthesis of 22. 1H and 13C NMR spectroscopic
constants of 22 and 23 are given in Tables 5 and 6.
and bis(benzylidene)cycloalkanones examined leads to
significant enhancement of their potencies both as
quinone reductase inducers and as free radical scaven-
gers. In the case of the substituted phenylpropenoids,
this increase ranges from 3-fold for the ketones (e.g.,
compare the CD values for trans-4-phenyl-3-buten-2-
one (12) with 2-hydroxy-trans-4-phenyl-3-buten-2-one
(13) and chalcone (14) with 2-hydroxychalcone (16)) to
5-fold for the methyl esters of phenylpropenoic acids
(e.g., compare methyl cinnamate (1b) with methyl
o-coumarate (4b)). Moreover, this effect is additive only
when two hydroxyl groups are simultaneously present
at ortho-positions on both aromatic rings, e.g., as in 2,2′-
dihydroxychalcone (18). The most pronounced enhance-
ment is within the class of bis(benzylidene)cycloal-
kanones, where the difference between the potencies of
the hydroxylated and the nonhydroxylated derivatives
can be dramaticsgreater than 200 times (e.g., compare
2,5-bis(benzylidene)cyclopentanone (22) with 2,5-bis(2-
hydroxybenzylidene)cyclopentanone (26)). Taken to-
gether, our findings identify two previously unrecog-
nized properties of this class of compounds that are
exhibited at very low concentrations and without any
detectable cytotoxicity, which make them auspicious
candidates as chemoprotective agents: (a) the ability
to induce quinone reductase (a prototype of phase 2
detoxication enzyme) and (b) the ability to serve as
potent free radical scavengers. That the potencies of
these compounds in both processes parallel each other
suggests that these properties may be functionally
related.
The hydroxylated bis(benzylidene)cycloalkanones were syn-
thesized according to the procedure of Borsche and Geyer.29
For the synthesis of 2,6-bis(2-hydroxybenzylidene)cyclohex-
anone (27), cyclohexanone (5.0 g, 0.05 mol) and salicylaldehyde
(12.1 g, 0.1 mol) were mixed and 37.5 mL of absolute ethanol
was added. This was followed by the dropwise addition of 30
mL of 20% NaOH, with constant stirring. The reaction was
allowed to proceed for 48 h at room temperature, 200 mL of
H2O was added, and the mixture was neutralized by gently
bubbling CO2 through it. The resulting solid material was
filtered and crystallized from warm acetone to yield yellow
1
crystals (10 g, 67%). H and 13C NMR spectroscopic constants
for 26 and 27 are given in Tables 5 and 6.
NMR Sp ectr oscop y. Proton NMR spectra were obtained
in CDCl3 (unless otherwise specified) at 300 MHz on a Bruker
spectrometer at 25 °C. 13C NMR spectra were obtained in a
solvent mixture of CDCl3/DMSO-d6 (2:8) at 400 MHz on a
Varian VXR 400 at 30 °C. Two-dimensional NMR was per-
formed in the same solvent mixture on a Varian Unity Plus
600 NMR spectrometer at 30 °C. Chemical shift frequencies
were relative to tetramethylsilane.
Exp er im en ta l P r oced u r es
Ma ter ia ls. Xanthine, xanthine oxidase (grade I) from
buttermilk, and lucigenin were obtained from Sigma (St. Louis,
MO). The structures of all test compounds are shown in Tables
1-4.
Qu in on e Red u cta se Activity Assa y. Specific activity of
quinone reductase was determined in Hepa 1c1c7 murine
hepatoma cells grown in 96-well microtiter plates according
to minor modifications of the described procedure.12,13,15,58 Cells
(10 000/well) were grown for 24 h in medium supplemented
with 10% heat- and charcoal-treated fetal bovine serum and
exposed to serial dilutions of inducers for 48 h before specific
activities of quinone reductase were measured. All compounds
were dissolved in dimethyl sulfoxide and diluted so that the
final concentration of solvent was 0.1% by volume in all wells.
The specific activities of quinone reductase were normalized
to the protein concentrations of the cell extracts. For this
purpose, 20-µL aliquots of the cell lysates were transferred to
corresponding wells of a duplicate plate, 300 µL of bicincho-
ninic acid reagent was added to each well, and the absorptions
at 550 nm were determined according to the procedure of
Bradford.59 The concentration required to double the specific
activity of quinone reductase (CD value) was used as a
measure of inducer potency.
Deter m in a tion of F r ee Ra d ica l-Sca ven gin g Activity.
Assays were carried out at 37 °C in plastic tubes in reaction
mixtures gassed with air, which contained in a final volume
of 1.00 mL: Dulbecco’s phosphate-buffered saline, pH 7.4, 5
µM lucigenin (1 µL of a 5 mM solution in H2O), 0.0025 unit of
xanthine oxidase (8.5 µL of a solution containing 0.33 unit/
mL of chromatographically purified milk xanthine oxidase;
Sigma, grade I, specific activity 0.69 unit/mg protein), and 1
µL of the test compound in dimethyl sulfoxide, to give final
concentrations of 50 nM-10 µM.52 After gentle mixing, each
assay tube was placed in the holder of Berthold Biolumat
model LB9505 luminometer (Bad Wildbad, Germany). Six
samples were assayed at one time. After obtaining a stable
Cinnamic acid (1a ), methyl cinnamate (1b), coumarin (8),
7-hydroxycoumarin (9), trans-4-phenyl-3-buten-2-one (12),
2-hydroxy-trans-4-phenyl-3-buten-2-one (13), chalcone (14),
4-hydroxychalcone (15), 4′-hydroxychalcone (19), 2,7-bis(ben-
zylidene)cycloheptanone (24), and 2,6-bis(2-methoxybenzyli-
dene)cyclohexanone (28) were obtained from Aldrich (Milwau-
kee, WI); caffeic (5a ), ferulic (6a ), and sinapic (7a ) acids and
p-(2a ), m-(3a ), and o-(4a ) coumaric acids were from Lancaster
(Windham, NH); 3-acetylcoumarin (10), 3-hydroxycoumarin
(11), 2-hydroxychalcone (16), 2′-hydroxychalcone (17), 2,2′-
dihydroxychalcone (18), 4,2′,4′-trihydroxychalcone (20), and
2,2′,4′-trihydroxychalcone (21) were from Indofine (Somerville,
NJ ). All other test compounds were synthesized as described
below.
Ch em ica l Syn th eses. Methyl esters of cinnamic and other
arylpropenoic acids were prepared by a general procedure
involving the refluxing of methanolic solutions (10 mL) of the
corresponding acids (usually 1 g) in the presence of a small
amount (0.5 mL) of Dowex 50W-8X cation-exchange resin for
2-3 h. The resin was removed by filtration, the solvent was
evaporated in a vacuum, and the product was crystallized from
suitable mixtures of hexane and acetone. The identities of the
compounds were established by 1H NMR spectroscopy (300
MHz, CDCl3). Methyl o-coumarate (4b): δ (ppm) 8.01 (d, 1H,
J ) 16 Hz, ph-CHd), 7.48-6.84 (m, 4H, arom), 6.62 (d, 1H,
J ) 16 Hz, dCH-CO-), 5.80 (s, 1H, OH), 3.83 (s, 3H, CH3).
Methyl m-coumarate (3b): δ (ppm) 7.61 (d, 1H, J ) 16 Hz,
ph-CHd), 7.33-6.79 (m, 4H, arom), 6.40 (d, 1H, J ) 16 Hz,
dCH-CO-), 5.57 (s, 1H, OH), 3.80 (s, 3H, CH3). Methyl
p-coumarate (2b): δ (ppm) 7.65 (d, 1H, J ) 16 Hz, ph-CHd),
7.52-6.72 (m, 4H, arom), 6.28 (d, 1H, J ) 16 Hz, dCH-CO-),
5.58 (s, 1H, OH), 3.79 (s, 3H, CH3). Methyl sinapate (7b): δ

Protective Properties of Phenylpropenoids
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 26 5295
(9) Favreau, L. V.; Pickett, C. B. Transcriptional Regulation of the
Rat NAD(P)H:Quinone Reductase Gene. Identification of Regu-
latory Elements Controlling Basal Level Expression and Induc-
ible Expression by Planar Aromatic Compounds and Phenolic
Antioxidants. J . Biol. Chem. 1991, 266, 4556-4561.
(10) Friling, R. S.; Bergelson, S.; Daniel, V. Two Adjacent AP-I
Binding Sites Form the Electrophile-Responsive Element of the
Murine Glutathione S-Transferase Ya Subunit Gene. Proc. Natl.
Acad. Sci. U.S.A. 1992, 89, 668-672.
(11) Prestera, T.; Holtzclaw, W. D.; Zhang, Y.; Talalay, P. Chemical
and Molecular Regulation of Enzymes that Detoxify Carcino-
gens. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 2965-2969.
(12) Prochaska, H. J .; Santamaria, A. B. Direct Measurement of
NAD(P)H:Quinone Reductase from Cells Cultured in Microtiter
Wells: A Screening Assay for Anticarcinogenic Enzyme Induc-
ers. Anal. Biochem. 1988, 169, 328-336.
baseline (over the period of 1 min), the reaction was initiated
by injection of 20 µL of a xanthine solution to give a final
concentration of 56 µM, and the chemiluminescence was
monitored for an additional 4 min. Test compounds were
dissolved in dimethyl sulfoxide or acetonitrile to give a final
concentration of the organic solvent of 0.1% by volume. The
luminescence data were obtained as integrated areas under
the curve (during the 4-min interval), and inhibition of
chemiluminescence was expressed as a percentage of control
area.
In h ibition of Xa n th in e Oxid a se Activity. The activity
of xanthine oxidase was determined spectrophotometrically by
measuring the enzymatic conversion of xanthine to uric acid
at 25 °C. Each assay mixture contained in a final volume of
1.00 mL: 0.0025 unit of purified milk xanthine oxidase, 20
µmol of the test compound in dimethyl sulfoxide (final con-
centration 0.1% by volume in the reaction), and Dulbecco’s
phosphate-buffered saline, pH 7.4. Absorbance was monitored
at 295 nm, and after the baseline had stabilized, the reaction
was started by addition of 2.0 µL of a 25 mM solution of
xanthine to give a final concentration of 50 µM; then the linear
increase in absorbance was recorded.
(13) Prochaska, H. J .; Santamaria, A. B.; Talalay, P. Rapid Detection
of Inducers of Enzymes that Protect Against Carcinogens. Proc.
Natl. Acad. Sci. U.S.A. 1992, 89, 2394-2398.
(14) Prochaska, H. J .; Talalay, P. Regulatory Mechanisms of Mono-
functional and Bifunctional Anticarcinogenic Enzyme Inducers
in Murine Liver. Cancer Res. 1988, 48, 4776-4782.
(15) Spencer, S. R.; Xue, L.; Klenz, E. M.; Talalay, P. The Potency of
Inducers of NAD(P)H:(Quinone Acceptor) Oxidoreductase Paral-
lels their Efficiency as Substrates for Glutathione Transferases.
Structural and Electronic Correlations. Biochem. J . 1991, 273,
711-717.
Ack n ow led gm en t. These studies were supported by
a Postdoctoral Fellowship of the Cancer Research
Foundation of America (to A.T.D.-K.); by gifts from the
Lewis and Dorothy Cullman Foundation, from Charles
B. Benenson, and from other Friends of the Brassica
Chemoprotection Laboratory; and by a Program-Project
Grant (PO1 CA 44530) from the National Cancer
Institute, Department of Health and Human Services.
The authors are grateful for chemical advice from
Yoshitake Ichikawa (who suggested the ion-exchange
resin-catalyzed ester synthesis) and Gary H. Posner and
to Charles Long for performing the 13C NMR measure-
ments. The chemiluminescence experiments were car-
ried out with the advice and assistance of Michael A.
Trush and Yunbo Li. We thank Douglas G. Borden for
discussions on bis(benzylidene)cycloalkanones and for
providing an unpublished report from the Eastman
Kodak Co. on the synthesis of several of these com-
pounds. We thank Qinjian Zhao for stimulating discus-
sions and also Herman Adlercreutz and Kristiina Wa¨-
ha¨la¨ for their kind gift of compound 25.
(16) Zheng, M.; A° slund, F.; Storz, G. Activation of the OxyR Tran-
scription Factor by Reversible Disulfide Bond Formation. Science
1998, 279, 1718-1721.
(17) Pinkus, L.; Weiner, L. M.; Daniel, V. Role of Oxidants and
Antioxidants in the Induction of AP-1, NF-κB, and Glutathione
S-Transferase Gene Expression. J . Biol. Chem. 1996, 271,
13422-13429.
(18) Lewis, N. G.; Kato, M. J .; Lopes, N.; Davin, L. B. Lignans:
Diversity, Biosynthesis, and Function. In Chemistry of the
Amazon. Biodiversity, Natural Products, and Environmental
Issues; Seidl, P. R., Gottlieb, O. R., Kaplan, M. A. C., Eds.; ACS
Symposium Series 588; American Chemical Society: Washing-
ton, DC, 1995; pp 135-167.
(19) Batt, D. G.; Goodman, R.; J ones, D. G.; Kerr, J . S.; Mantegna,
L. R.; McAllister, C.; Newton, R. C.; Nurnberg, S.; Welch, P. K.;
Covington, M. B. 2′-Substituted Chalcone Derivatives as Inhibi-
tors of Interleukin-1 Biosynthesis. J . Med. Chem. 1993, 36,
1434-1442.
(20) Ballesteros, J . F.; Sanz, M. J .; Ubeda, A.; Miranda, M. A.; Iborra,
S.; Paya´, M.; Alcaraz, M. J . Synthesis and Pharmacological
Evaluation of 2′-Hydroxychalcones and Flavones as Inhibitors
of Inflammatory Mediators Generation. J . Med. Chem. 1995, 38,
2794-2797.
(21) Wattenberg, L. W.; Coccia, J . B.; Lam, L. K. T. Inhibitory Effects
of Phenolic Compounds on Benzo(a)pyrene-Induced Neoplasia.
Cancer Res. 1980, 40, 2820-2823.
(22) Iwata, S.; Nishino, T.; Nagata, N.; Satomi, Y.; Nishino, H.;
Shibata, S. Antitumorigenic Activities of Chalcones. I. Inhibitory
Effect of Chalcone Derivatives on 32Pi-Incorporation into Phos-
pholipids of HeLa Cells Promoted by 12-O-Tetradecanoyl-
Phorbol 13-Acetate (TPA). Biol. Pharm. Bull. 1995, 18, 1710-
1713.
(23) Wattenberg, L. W. Chalcones, myo-Inositol and Other Novel
Inhibitors of Pulmonary Carcinogenesis. J . Cell. Biochem. 1995,
S22, 162-168.
(24) Shibata, S. Anti-tumorigenic Chalcones. Stem Cells 1994, 12,
44-52.
(25) Li, R.; Kenyon, G. L.; Cohen, F. E.; Chen, X.; Gong, B.;
Dominguez, J . N.; Davidson, E.; Kurzban, G.; Miller, R. E.;
Nuzum, E. O.; Rosenthal, P. J .; McKerrow, J . H. In vitro
Antimalarial Activity of Chalcones and Their Derivatives. J .
Med. Chem. 1995, 38, 5031-5037.
(26) Middleton, E. The Flavonoids. TIPS 1984, 5, 335-338.
(27) Adlercreutz, H. Western Diet and Western Diseases: Some
Hormonal and Biochemical Mechanisms and Associations. Scand.
J . Clin. Lab. Invest. 1990, 50 (S201), 3-23.
(28) Markaverich, B. M.; Schauweker, T. H.; Gregory, R. R.; Varma,
M.; Kittrell, F. S.; Medina, D.; Varma, R. S. Nuclear Type II
Sites and Malignant Cell Proliferation: Inhibition by 2,6 Bis-
benzylidine Cyclohexanones. Cancer Res. 1992, 52, 2482-2488.
(29) Borsche, W.; Geyer, A. U¨ ber Tricyclische Benzopyryliumverbin-
dungen. Ann. 1912, 393, 29-60.
(30) Rumpel, W. Verfahren zur Herstellung von Divanillyden-Cyclo-
hexanon. Austrian Patent No. 180258, 1954.
(31) Razdan, B. K.; Sugden, J . K. Some pH Indicators Derived from
Cyclopentanone. Chem. Ind. 1970, 21, 685-686.
Refer en ces
(1) Talalay, P.; De Long, M. J .; Prochaska, H. J . Identification of a
Common Chemical Signal Regulating the Induction of Enzymes
that Protect Against Chemical Carcinogenesis. Proc. Natl. Acad.
Sci. U.S.A. 1988, 85, 8261-8265.
(2) Prestera, T.; Zhang, Y.; Spencer, S. R.; Wilczak, C. A.; Talalay,
P. The Electrophile Counterattack Response: Protection Against
Neoplasia and Toxicity. Adv. Enzyme Regulat. 1993, 33, 281-
296.
(3) Talalay, P. The role of Enzyme Induction in Protection Against
Carcinogenesis. In Cancer Chemoprevention; Wattenberg, L.,
Lipkin, M., Boone, C. W., Kelloff, G. J ., Eds.; CRC Press: Boca
Raton, FL, 1992; pp 469-478.
(4) Primiano, T.; Sutter, T. R.; Kensler, T. W. Antioxidant-Inducible
Genes. Adv. Pharmacol. 1997, 38, 293-328.
(5) Radjendirane, V.; J oseph, P.; Lee, Y.-H.; Kimura, S.; Klein-
Szanto, A. J . P.; Gonzalez, F. J .; J aiswal, A. K. Disruption of
the DT Diaphorase (NQO1) Gene in Mice Leads to Increased
Menadione Toxicity. J . Biol. Chem. 1998, 273, 7382-7389.
(6) Henderson, C. J .; Smith, A. G.; Ure, J .; Brown, K.; Bacon, E. J .;
Wolf, C. R. Increased Skin Tumorigenesis in Mice Lacking pi
Class Glutathione S-transferases. Proc. Natl. Acad. Sci. U.S.A.
1998, 95, 5275-5280.
(7) Hayes, J . D.; Pulford, D. J . The Glutathione S-transferase
Supergene Family: Regulation of GST and the Contribution of
the Isozymes to Cancer Chemoprotection and Drug Resistance.
Crit. Rev. Biochem. Mol. Biol. 1995, 30, 445-600.
(8) Wasserman, W. W.; Fahl, W. E. Comprehensive Analysis of
Proteins which Interact with the Antioxidant Responsive Ele-
ment: Correlation of ARE-BP-1 with the Chemoprotective
Induction Response. Arch. Biochem. Biophys. 1997, 344, 387-
396.
(32) Borden, D. G. Design of Photocrosslinkable Polyesters with
Specific Absorptions. J . Appl. Polm. Sci. 1978, 22, 239-251.
(33) Konrad, G.; Mo¨ller, H. Neue Lichtschutzsubstranzen fu¨r den UV-
A-Bereich. Parfu¨merie Kosmetik 1983, 64, 317-318, 320-322.

5296 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 26
Dinkova-Kostova et al.
(34) Markaverich, B. M.; Varma, M.; Densmore, C. L.; Tiller, A. A.;
Schauweker, T. H.; Gregory, R. R. Nuclear Type II [3H]Estradiol
Binding Sites in MCF-7 Human Breast Cancer Cells; Binding
Interactions with 2,6-bis([3,4-Dihydroxyphenyl]-methylene)-cy-
clohexanone Esters and Inhibition of Cell Proliferation. Int. J .
Oncol. 1994, 4, 1291-1300.
(48) Edenharder, R.; von Petersdorff, I.; Rauscher, R. Antimutagenic
Effects of Flavonoids, Chalcones and Structurally Related
Compounds on the Activity of 2-Amino-3-methylimidazo[4,5-f]-
quinoline (IQ) and Other Heterocyclic Amine Mutagens from
Cooked Food. Mutat. Res. 1993, 287, 261-274.
(49) Tsuchiya, H.; Sato, M.; Akagiri, M.; Takagi, N.; Tanaka, T.;
Iinuma, M. Anti-Candida Activity of Synthetic Hydroxychal-
cones. Pharmazie 1994, 49, 756-758.
(35) Wattenberg, L. W. Inhibition of Carcinogenic and Toxic Effects
of Polycyclic Hydrocarbons by Phenolic Antioxidants and Ethox-
yquin. J . Natl. Cancer Inst. 1972, 48, 1425-1430.
(50) Ramanathan, R.; Das, N. P.; Tan, C. H. Inhibitory Effect of
2-Hydroxychalcone and other Flavonoids on Human Cancer Cell
Proliferation. Int. J . Oncol. 1993, 3, 115-119.
(36) Wattenberg, L. W.; Speier, J . L.; Kotake, A. Effects of Antioxi-
dants on Metabolism of Aromatic Polycyclic Hydrocarbons. Adv.
Enzyme Regul. 1976, 14, 313-322.
(37) Batzinger, R. P.; Ou, S.-Y. L.; Bueding, E. Antimutagenic Effects
of 2(3)-tert-Butyl-4-hydroxyanisole and of Antimicrobial Agents.
Cancer Res. 1978, 38, 4478-4485.
(51) Sugijama, Y.; Kawakishi, S.; Osawa, T. Involvement of the
â-diketone Moiety in the Antioxidative Mechanism of Tetrahy-
drocurcumin. Biochem. Pharmacol. 1996, 52, 519-525.
(52) Li, Y.; Zhu, H.; Kuppusamy, P.; Roubaud, V.; Zweier, J . L.;
Trush, M. A. Validation of Lucigenin (Bis-N-methylacridinium)
as a Chemilumigenic Probe for Detecting Superoxide Anion
Radical Production by Enzymatic and Cellular Systems. J . Biol.
Chem. 1998, 273, 2015-2023.
(53) Liochev, S. I.; Fridovich, I. Lucigenin Luminescence as a Measure
of Intracellular Superoxide Dismutase Activity in Escherichia
coli. Proc. Natl. Acad. Sci. U.S.A 1997, 94, 2891-2896.
(54) George, H.; Roth, H. J . Photoisomerisierung und cyclo-1,2-
Addition R,â-ungesa¨ttigter Cyclanone. Tetrahedron Lett. 1971,
43, 4057-4060.
(55) Kupchan, S. M.; Fessler, D. C.; Eakin, M. A.; Giacobbe, T. J .
Reactions of Alpha Methylene Lactone Tumor Inhibitors with
Model Biological Nucleophiles. Science 1970, 168, 376-378.
(56) Kupchan, S. M.; Eakin, M. A.; Thomas, A. M. Tumor Inhibitors.
69. Structure-Cytotoxicity Relationships among the Sesquiter-
pene Lactones. J . Med. Chem. 1971, 14, 1147-1152.
(57) Rodriguez, A. M.; Enriz, R. D.; Santagata, L. N.; J a´uregui, E.
A.; Pestchanker, M. J .; Giordano, O. S. Structure-Cytoprotective
Activity Relationship of Simple Molecules Containing an R,â-
Unsaturated Carbonyl System. J . Med. Chem. 1997, 40, 1827-
1834.
(38) Benson, A. M.; Cha, Y.-N.; Bueding, E.; Heine, H. S.; Talalay,
P. Elevation of Extrahepatic Glutathione S-Transferase and
Epoxide Hydratase Activities by 2(3)-tert-Butyl-4-hydroxyani-
sole. Cancer Res. 1979, 39, 2971-2977.
(39) Wattenberg, L. W. Inhibition of Neoplasia by Minor Dietary
Constituents. Cancer Res. 1983, 43, 2448S-2453S.
(40) Halliwell, B.; Gutteridge, J . M. C. Role of Free Radicals and
Catalytic Metal Ions in Human Disease: an Overview. Methods
Enzymol. 1990, 186, 1-85.
(41) Prochaska, H. J .; De Long, M. J .; Talalay, P. On the Mechanisms
of Induction of Cancer-Protective Enzymes: a Unifying Proposal.
Proc. Natl. Acad. Sci. U.S.A. 1985, 82, 8232-8236.
(42) Brown, S. A. Biochemistry of the Coumarins. In Recent Advances
in Phytochemistry, Biochemistry of Plant Phenolics; Swain, T.,
Harborne, J . B., van Sumere, C. F., Eds.; Plenum Press: New
York, 1979; Vol. 12, pp 249-286.
(43) Shilling, W. H.; Crampton, R. F.; Longland, R. C. Metabolism
of Coumarin in Man. Nature 1969, 221, 664-665.
(44) Van Iersel, M. L. P. S.; Henderson, C. J .; Walters, D. G.; Price,
R. J .; Wolf, C. R.; Lake, B. G. Metabolism of [3-14C]Coumarin
by Human Liver Microsomes. Xenobiotica 1994, 24, 793-803.
(45) Fentem, J . H.; Fry, J . R. Metabolism of Coumarin by Rat, Gerbil
and Human Liver Microsomes. Xenobiotica 1992, 22, 357-367.
(46) Yamamoto, S.; Aizu, E.; J iang, H.; Nakadate, T.; Kiyoto, I.;
Wang, J . C.; Kato, R. The Potent Anti-Tumor-Promoting Agent
Isoliquiritigenin. Carcinogenesis 1991, 12, 317-323.
(58) Fahey, J . W.; Zhang, Y.; Talalay, P. Broccoli Sprouts: an
Exceptionally Rich Source of Inducers of Enzymes that Protect
Against Chemical Carcinogens. Proc. Nati. Acad. Sci. U.S.A..
1997, 94, 10367-10372.
(47) Makita, H.; Tanaka, T.; Fujitsuka, H.; Tatematsu, N.; Satoh,
K.; Hara, A.; Mori, H. Chemoprevention of 4-Nitroquinoline
1-Oxide-Induced Rat Oral Carcinogenesis by the Dietary Fla-
vonoids Chalcone, 2-Hydroxychalcone, and Quercetin. Cancer
Res. 1996, 56, 4904-4909.
(59) Bradford, M. M. A Rapid and Sensitive Method for the Quan-
titation of Microgram Quantities of Protein Utilizing the Prin-
ciple of Protein-Dye Binding. Anal. Biochem. 1976, 72, 248-254.
J M980424S