3,4,4′-Trihydroxy-trans-stilbene, an analogue of resveratrol, is a potent antioxidant and cytotoxic agent

Article abstract of DOI:10.3109/10715762.2011.629199

Resveratrol (3,5,4′-trihydroxy-trans-stilbene) is a naturally occurring polyphenol widely distributed in food and dietary plants. This phytochemical has been intensively studied as an efficient antioxidant and anticancer agent, and a variety of substituted stilbenes have been developed in order to improve the potency of resveratrol. In this work, we described the synthesis of 3,4,4-trihydroxy-trans-stilbene (3,4,4′-THS), an analogue of resveratrol, and studied its antioxidant and cytotoxic activity in vitro. 3,4,4-THS was much more efficient than resveratrol in protecting against free radical-induced lipid peroxidation, photo-sensitized DNA oxidative damage, and free radical-induced hemolysis of human red blood cells. More potent growth inhibition in cultured human leukemia cells (HL-60) was also observed for 3,4,4-THS. The relationship between the antioxidant efficiency and cytotoxic activity was discussed, with the emphasis on inhibition of the free radical enzyme ribonucleotide reductase by antioxidants. The result that this subtle structure modification of resveratrol drastically improves its bioactivity provides important strategy to develop novel resveratrol-based molecules.

Full text of DOI:10.3109/10715762.2011.629199

Free Radical Research, November–December 2011; 45(11–12): 1379–1387
ORIGINAL ARTICLE
3,4,4¢-Trihydroxy-trans-stilbene, an analogue of resveratrol,
is a potent antioxidant and cytotoxic agent
WENQING CAI*, LIANGWEI ZHANG,YANLIN SONG, BAOXIN ZHANG, XUEMEI CUI,
GUANMING HU & JIANGUO FANG
State Key Laboratory of Applied Organic Chemistry, Lanzhou University, Lanzhou,Gansu, China
Abstract
Resveratrol (3,5,4¢-trihydroxy-trans-stilbene) is a naturally occurring polyphenol widely distributed in food and dietary
plants. This phytochemical has been intensively studied as an efficient antioxidant and anticancer agent, and a variety of
substituted stilbenes have been developed in order to improve the potency of resveratrol. In this work, we described the syn-
thesis of 3,4,4¢-trihydroxy-trans-stilbene (3,4,4¢-THS), an analogue of resveratrol, and studied its antioxidant and cytotoxic
activity in vitro. 3,4,4¢-THS was much more efficient than resveratrol in protecting against free radical-induced lipid peroxi-
dation, photo-sensitized DNA oxidative damage, and free radical-induced hemolysis of human red blood cells. More potent
growth inhibition in cultured human leukemia cells (HL-60) was also observed for 3,4,4¢-THS. The relationship between
the antioxidant efficiency and cytotoxic activity was discussed, with the emphasis on inhibition of the free radical enzyme
ribonucleotide reductase by antioxidants.The result that this subtle structure modification of resveratrol drastically improves
its bioactivity provides important strategy to develop novel resveratrol-based molecules.
Keywords: free radicals, resveratrol, antioxidant, lipid peroxidation, DNA oxidation
Introduction
red wine has been suggested to link to the low inci-
dence of heart diseases in some regions of France, the
so-called “French paradox”, i.e., despite high fat
intake, mortality from coronary heart diseases is lower
due to the regular drinking of wine [16,17]. In addi-
tion, resveratrol has recently been shown to play an
important role in the prevention of cancer, and its
anticancer activity is, at least partly, related to the
antioxidant effect [18]. Therefore, the past several
years have witnessed intense research devoted to the
biological activity, especially the antioxidant and anti-
cancer activity, of this molecule [15,18–22].
Structural modification of natural products is one
efficient way to increase their activity and lower the
side effects. Dozens of molecules have been devel-
oped based on the structure of resveratrol in recent
years, and some hydroxystilbenes with improved
activity have been discovered [21,23–30]. In our pre-
vious work, we found that the anticancer activity for
hydroxystilbenes correlates well with their antioxidant
The free radical-mediated autoxidation of organic
molecules by molecular oxygen is a double-edged
sword and has both positive and negative aspects.
It has been applied for the production of chemicals
such as terephthalic acid and cyclohexanol. On the
other hand, the recent development of free radical
biology and medicine has provided a large body of
evidence that oxidative stress arising from reactive
oxygen species (ROS) is implicated in a number of
biomacromolecular damage and human diseases
[1–3]. And small molecule antioxidants, such as vita-
min C [4], and other food-derived phytochemicals
[5–8] or artificial molecules [9,10], may have benefi-
cial effects in preventing against these damage and
the related diseases [11].
Resveratrol (3,5,4¢-trihydroxy-trans-stilbene, Figure
1) is a naturally occurring phytoalexin present in grapes
and other dietary plants. This “magic” molecule has
various biological functions [12–15]. Its presence in
*W. Cai is currently a PhD candidate in the Sanford–Burnham Institute, La Jolla, USA.
Correspondence: Jianguo Fang, Lanzhou University, 222 South Tianshui Road, Lanzhou, 730000 China. Tel: ϩ86-931-8912500;
Fax: ϩ86-931-8915557. E-mail: fangjg@lzu.edu.cn
ISSN 1071-5762 print/ISSN 1029-2470 online © 2011 Informa UK, Ltd.
DOI: 10.3109/10715762.2011.629199

1380 W. Cai et al.
activity [26]. In this research, we reported the
synthesis of 3,4,4¢-trihydroxy-trans-stilbene (3,4,4¢-
THS, Figure 1), an analogue of resveratrol, and the
antioxidant activity as well as the cytotoxicity of this
molecule. Simply changing 5-hydroxy in resveratrol
to 4-hydroxy efficiently increases the protective
effects against free radical-induced lipid peroxidation,
photo-sensitized DNA oxidative damage, and free
radical-induced hemolysis of human red blood cells
(HRBC). 3,4,4¢-THS inhibits cultured human leuke-
mia (HL-60) cell proliferation more efficiently than
does resveratrol. The findings that this subtle struc-
ture modification drastically improves the bioactivity
of resveratrol provide important strategy to develop
novel resveratrol-orientated molecules.
4-Methoxybenzyl chloride (10 ml) was heated with
triethyl phosphite (12 ml) under refluxing condition
until the evolution of the gas had ceased.The reaction
mixture was distilled under vacuum to give the dieth-
ylbenzylphosphonate. The diethylbenzylphosphonate
(1.4 g) was added to 15 ml of dry dimethylformamide
containing 0.5 g of sodium methoxide at room tem-
perature. To this was added 3,4-dimethoxybenzalde-
hyde (0.85 g), and the reaction was allowed to proceed
with stirring for 10 h at room temperature.The reac-
tion mixture was poured into water–methanol (2:1),
and the precipitated stilbene was collected by filtra-
tion and washed with water. The crude product was
recrystallized from ethanol–water to give the pure
3,4,4¢-trimethoxystilbene (1.25 g, 92%).
Excess pyridine hydrochloride (3 g) and 3,4,4¢-
trimethoxystilbene (0.5 g) were mixed and heated to
190–195°C for 4 h. The hot dark syrup was poured
into 20 ml of 2 M HCl, and the reaction mixture was
extracted with ether (4 ϫ 20 ml).The ether layer was
dried with anhydrous MgSO4, and then ether was
removed under reduced pressure. The residual was
purified by column chromatography on silica gel
eluted with petroleum ether-acetone (2:1) to afford
3,4,4¢-trihydroxystilbene 0.20 g (43%). M. P. 242–
244°C (literature 242–243°C [28]). MS (HP-5988A)
Materials and methods
Anisole, 3,4-dimethoxyaldehyde, 2,2¢-Azobis(2-
amidinopropane hydrochloride) (AAPH), 3-(4,5-di-
methylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
(MTT),5,5¢-Dithiobis(2-nitrobenzoicacid)(DTNB),
and 3,5-dimethoxyaldehyde were purchased from
Aldrich. Methylene blue (MB), agarose, ethidium
bromide and plasmid pBR322 DNA were purchased
from Sigma. RPMI 1640 medium, and penicillin/
streptomycin were from Invitrogen. All solvents used
in organic synthesis were purified and dried by stan-
dard techniques prior to use. All other chemicals were
of analytical grade and used upon received. Melting
points were measured on a Kofler apparatus and were
uncorrected.
1
m/e 228 (Mϩ); H NMR (aceton-d , 400 MHz,
Bruker) δ 6.80-6.83 (m, 3H), 6.88 (s, 61H), 6.89 (d,
1H, J ϭ 16.3 Hz), 7.02 (d, 1H, J ϭ 16.3 Hz), 7.05
(d, 1H), 7.37 (m, 2H).
Resveratrol was synthesized in the same manner as
3,4,4 -THS using 3,5-dimethoxybenzaldehyde instead
of 3,4-dimethoxybenzaldehyde as a starting material.
M. P. 254–257°C (literature 256–259°C [28]). MS m/e
228 (Mϩ); 1H NMR (aceton-d 6, 400 MHz) δ 6.22 (t,
1H), 6.49 (d, 2H), 6.81 (m, 2H), 6.86 (d, 1 H, J ϭ 16.4
Hz), 6.98 (d, 1H, J ϭ 16.4 Hz), 7.37 (m, 2H).
Synthesis of 3,4,4¢-THS and resveratrol
3,4,4¢-THS and resveratrol were prepared by the
method according to literature [21,28] with slight
modification as shown in Scheme 1.
3,4,4¢-THS: Anisole (55 ml) was dissolved in dry
benzene (220 ml) and 20 g paraformaldehyde was
added to the solution. Dry hydrogen chloride was
bubbled into the solution with stirring over a period
of 1.5 h at 0°C. After continuing for 2 h at 0°C, the
reaction mixture was washed with ice-water three
times, ice cold NaHCO3 solution (2%) three times.
The organic layer was then separated and dried over
anhydrous Na2CO3. The 4-methoxybenzyl chloride
was obtained by distilling under vacuum.
Inhibition of free radical-induced lipid peroxidation
High performance liquid chromatography (HPLC)
was used to determine the linoleic acid peroxides[21].
Linoleic acid was dissolved in t-butanol/water (3:2,
v/v) and its peroxidation was initiated by addition of
AAPH at 37°C. Linoleic acid hydroperoxides, which
have a typical absorbance at 234 nm, were detected
by HPLC analysis using a Gilson liquid chromato-
graph with a ZORBAX ODS reversed phase column
Figure 1. Chemical structures of resveratrol and 3,4,4¢-THS.

Resveratrol analogue antioxidation 1381
Scheme 1. Synthesis of 3,4,4¢-THS.
Assay for photo-sensitized oxidative DNA damage
(6 ϫ 250 mm Du Pont instruments) and eluted with
methanol-propan-2-ol (3:1 v/v). A Gilson model 116
UV detector was used to monitor the total linoleic
acid hydroperoxides at 234 nm. Every determination
was repeated, and the experimental deviations were
within Ϯ 10%.
The system for exposure of DNA to photosensitiza-
tion was in principle similar to that described earlier
[33,34]. Briefly, the system consisted of a 300 W
Tungsten Halogen lamp which was placed 25 cm
away from cuvettes containing 0.1–0.2 μg pBR322
DNA in the presence of indicated concentration of
MB in 20 μl of 0.15 M phosphate buffer solution
(PBS, pH 7.4).The cuvettes were kept in a water bath
that was maintained at 37°C.The light source and the
cuvettes were separated by a glass plate to cut off the
light the wavelength of which is less than 400 nm.
3,4,4 -THS and resveratrol were added to the incuba-
tion and exposed to photosensitization under air
atmosphere. Electrophoretic separation of different
forms of DNA was achieved on 1.0% agarose gel
using 40 mM Tris-acetate / 1 mM EDTA buffer (pH
8.3). DNA bands were stained with ethidium bro-
mide and photographed under UV light.
Determination of the antioxidants consumption
The procedure was the same as described above for
determination of linoleic acid hydroperoxides, except
that a Gilson model 142 electrochemical detector set
at ϩ 700 mV versus SCE was used for monitoring
resveratrol, 3,4,4¢-THS and α-tocopherol (TOH).
The column was eluted with methanol-propan-2-ol-
formic acid (80:20:1, v/v/) containing 50 mmol L-1 of
sodium perchlorate as supporting electrolyte.
Assay for free radical-induced hemolysis of HRBC
The method used for determining the hemolysis of
HRBC essentially followed the published procedure
[31,32] with slight modification. HRBC were sepa-
rated from heparinized blood that was drawn from a
healthy donor. The blood was centrifuged at 2000
rpm for 10 min to separate the HRBC from plasma
followed by washing the HRBC three times with PBS
at 4°C.
The 5% suspension of HRBC in PBS was incu-
bated under air atmosphere at 37°C for 5 min, into
which an aqueous solution of AAPH was added to
initiate hemolysis. The reaction mixture was shaken
gently while being incubated at 37°C. The extent of
hemolysis was determined spectrophotometrically as
described previously [32]. Aliquots of the reaction
mixture were taken out at appropriate time intervals,
diluted with 0.15 M NaCl, and centrifuged at 2000
rpm for 10 min at 4°C.The percentage hemolysis was
determined by measuring the absorbance of the
supernatant at 540 nm and compared with that of
complete hemolysis by treating the controlled HRBC
suspension with distilled water. In case of antioxida-
tion experiments resveratrol or 3,4,4¢-THS was added
and incubated before addition of AAPH. Every exper-
iment was repeated three times, and the results were
reproducible within 10% deviation
MTT Assay for growth inhibition in cultured
HL-60 cells
HL-60 cells were grown as suspension cultures in
RPMI 1640 medium supplemented with 10% (v/v)
fetal calf serum and antibiotics (penicillin/streptomy-
cin). Resveratrol and 3,4,4¢-THS were dissolved in
DMSO as stock solutions. Before addition to the cul-
ture medium, these stock solutions were diluted with
PBS and the final concentration of DMSO in the
medium is 0.1 %.To assay the antiproliferative effect
of resveratrol or 3,4,4¢-THS, 5000 cells in 90 μl of
medium were plated to each well in 96-well plates.
The medium was supplemented with 10 μl of PBS
containing various concentrations of compounds.
Four replicate wells were used for each concentration
of both compounds. After treatment of cells at 37°C
for 24 h or 48 h, 10 μl of MTT solution (2 mg/ml in
PBS) was added to each well, and the plates were
incubated for additional 4 h at 37°C.The plates were
centrifuged and the medium within the wells was
aspirated. The intracellular formazan dye crystals
were dissolved by addition of 100 μl of DMSO to each
well and were incubated overnight at room temperature
in the dark with constant shaking. The absorbance of
formazan at 562 nm was measured using a microplate

1382 W. Cai et al.
reader. The percentage of growth inhibition was cal-
culated by comparison with control PBS-treated cells
in the presence of 0.1% DMSO.
Consumption of antioxidants
In order to rationalize the antioxidant mechanism
under the inhibiting the linoleic acid peroxidation, the
decay of antioxidants was determined by HPLC sep-
aration of the reaction mixture followed by electro-
chemical detection of resveratrol, 3,4,4¢-THS, and
TOH. Representative results are illustrated in Figure
2B. The decay of TOH was approximately linear in
the absence of other antioxidants (line a in Figure
2B), in accordance with the previous reported kinetic
behavior for antioxidation reactions.When resveratrol
was added, the decay of TOH didn¢t change (line b
in Figure 2B).While 3,4¢4¢-THS was added, the decay
of TOH became slower, but was still linear (line c).
Resveratrol didn’t decay either with or without TOH
Results
Protection of linoleic acid from free radical-induced
peroxidation
The primary peroxidation products of linoleic acid
are hydroperoxides formed by oxygen addition at
C-9 and/or C-13 positions with either trans, cis- or
trans, trans-diene stereochemistry [35]. They show
characteristic ultra-violet absorption at 235 nm
which was used to monitor the formation of the
total hydroperoxides formed during the peroxida-
tion after separation of the reaction mixture by high
performance liquid chromatography (HPLC). A set
of representative kinetic curves of the total hydroper-
oxides formation during the peroxidation of linoleic
acid in solution is shown in Figure 2A. The con-
centration of the hydroperoxides increased fast and
linearly with time in the absence of antioxidants
upon AAPH-initiation, demonstrating fast peroxi-
dation of the substrate (line a in Figure 2A). The
peroxides formation was inhibited remarkably by
addition of 3,4,4¢-THS (line e) with a inhibition
time (tinh) of around 50 min, but not by resveratrol
(line b in Figure 2A) under the experimental condi-
tions. After the inhibition period the rate of
hydroperoxide formation turned faster which is close
to the original rate of propagation, demonstrating the
exhaustion of the antioxidant.
No synergistic effect of resveratrol and 3,4,
4¢-THS in the presence of α-tocopherol
α-Tocopherol (TOH), the most abundant and active
form of vitamin E, is a well-known and the principal
lipid-soluble chain-breaking antioxidant in plasma
and erythrocytes. Its synergistic antioxidative effect
withotherantioxidantshasbeenwelldocumented[36].
Therefore, we want to know if TOH could also inter-
act synergistically with resveratrol or 3,4,4¢-THS in
solution. TOH showed typical antioxidant behavior
against linoleic acid peroxidation (line c in Figure 2A)
as reported previously with a tinh of around 30 min
[21]. Addition of 3,4,4¢-THS together with TOH
remarkably prolonged the inhibition period with a tinh
of around 80 min (line f in Figure 2A), which almost
equals to the sum of the inhibition times caused by
TOH and 3,3,4¢-THS individually, and showed no
synergistic antioxidation effect. Addition of resvera-
trol together with TOH has little effect on the inhibi-
tion period of the latter (line d in Figure 2A),
confirming resveratrol is not a efficient antioxidant in
homogenous solution.
Figure 2. AAPH-induced formation of lipid hydroperoxides
(LOOH) and inhibition by antioxidants. (A). The peroxidation of
linoleic acid (LH, 0.1 M) was initiated by addition of AAPH (10
mM) in tert-butyl alcohol-water (3:2 v/v) at 37°C (line a), and
inhibited by resveratrol (18.6 μM, line b),TOH (15.0 μM, line c),
resveratrol plus TOH (line d), 3,4,4¢-THS (18.6 μM, line e), and
3,4,4¢-THS plus TOH (line f). (B). Consumption of antioxidant
(AH) during the prevention of AAPH (10 mM)-induced linoleic
acid (0.1 M) peroxidation in tert-butyl alcohol-water (3:2 v/v) at
37°C. [TOH]0 ϭ 15 μM [AH]0 ϭ 18.6 μM. a: decay of TOH; b:
decay of TOH in the presence of resveratrol; c: decay of TOH in
the presence of 3,4,4¢-THS; d: decay of 3,4,4¢-THS in the presence
of TOH; e: decay of 3,4,4¢-THS; f: decay of resveratrol in the
presence of TOH; g: decay of resveratrol.

Resveratrol analogue antioxidation 1383
(lines g and f), supporting that it’s not an efficient
antioxidant under the experimental condition. The
decay of 3,4,4¢-THS in the presence of TOH was
slower compared that withoutTOH ((lines d and e in
Figure 2B).
Protection of HRBC from free radical-induced hemolysis
AAPH decomposes at 37°C to produce initiating free
radicals [21]. Since the lipid peroxidation is a free
radical chain reaction, and one initiating radical can
induce up to twenty propagation reactions [21], the
HRBC membrane is quickly damaged, and eventually
leading to hemolysis. As shown in Figure 3, in the
absence of AAPH the HRBC were stable, and little
hemolysis took place within 4 h (line d). Addition of
AAPH induced, after an inhibition period, fast hemo-
lysis (line a).This inhibition stems from the action of
native antioxidants,e.g.vitamin E and/or ubiquinol-10
presented in the HRBC membranes. Addition of res-
veratrol or 3,4,4¢-THS into the HRBC suspension
significantly increased the intrinsic inhibition time of
the HRBC (lines b and c). 3,4,4¢-THS gives much
longer inhibition time than does resveratrol (line c),
indicating that 3,4,4¢-THS is a more efficient free
radical scavenger.
Protection of DNA from photosensitized
oxidative damage
Figure 4. Protection of plasmid DNA from singlet oxygen-induced
oxidative damage by resveratrol and 3,4,4¢-THS. (A) Agarose gel
electrophoretic pattern showing strand breaks in plasmid pBR322
DNA as a function of MB concentrations. DNA was exposed
to light for 30 min. Lane 1¢control (without MB); Lanes 2–8:
3.12, 6.25, 12.5, 25, 50, 100, 200 μM MB. Protection against MB
(100 μM ) plus light-induced plasmid pBR322 DNA strand breaks
by resveratrol (B) or 3,4,4¢-THS (C). The DNA was exposed to
light for 30 min. Lane 1: control; Lanes 2–5: 0, 15, 30, 60 μM
resveratrol. (D) Quantitative analysis of protective effects by
resveratrol and 3,4,4¢-THS. Band density from panel B and panel
C was quantified by densitometry, and was expressed as the
percentage of the control (mean Ϯ SD).
MB is a photosensitizer for generating ROS and has
been commonly used to investigate cellular and geno-
toxic effects in a variety of organisms. Singlet oxygen
(1O2) is one of the major ROS generating in this sys-
tem [34]. Figure 4A shows a photograph resulting
from agarose gel electrophoresis of plasmid pBR322
DNA exposed to light in the presence of MB under
air atmosphere. The plasmid DNA was easily oxida-
tive damaged induced by photosensitization, showing
a dose-dependent manner. Exposure of the plasmid
DNA to MB and light increased the relative intensity
of the open circular form (Form II) and decreased
that of supercoiled form (Form I) (lanes 1 and 2).
When the concentration of MB reached to 100 μM
under a 30-min photosensitization, the supercoiled
form of DNA completely disappeared (lane 7).
Figures 4B and C show protective effects of res-
veratrol and 3,4,4¢-THS against photosensitized oxi-
dative DNA damage, and quantitative results were
shown in Figure 4D based on the band density from
3 independent experiments. As seen from the figures,
both 3,4,4¢-THS and resveratrol can protect plasmid
pBR322 DNA from oxidative damage with different
Figure 3. Inhibition of AAPH-induced hemolysis of HRBC by
resveratrol or 3,4,4¢-THS.The hemolysis was initiated by addition
of 51.6 mM AAPH (a) and inhibited by addition of 10 μM
resveratrol (b), or 10 μM 3,4,4¢-THS (c).There was no hemolysis
of HRBC in the experimental conditions with out AAPH (d). Every
experiment was repeated three times and dada were expressed as
mean Ϯ SE.

1384 W. Cai et al.
Initiation: AAPH ® 2R· ϩ N2
R· ϩ LH ® RH ϩ L·
(2)
(3)
(4)
(5)
levels. As antioxidant concentration accrues, the pro-
tective effect increases. Although 15 μM of resveratrol
has little effect, the same concentration of 3,4,4¢-THS
can significantly protect DNA from the photosensi-
tized oxidative damage (lane 3 in Figures 4B and C).
Under other concentrations used in the experiments,
3,4,4¢-THS is also more efficient than resveratrol to
protect plasmid pBR322 DNA from damage.
Propagation: L· ϩ O2 ® LOO·
LOO· ϩ LH ® LOOH ϩ L·
Termination: LOO· ϩ LOO· ® molecular products (6)
Antioxidation: LOO· ϩ AH ® LOOH ϩ A·
(7)
MB is a photo sensitizer for generating ROS and has
been routinely used to investigate cellular and geno-
toxic effects in a variety of organisms. Upon irradiation
by light, singlet oxygen (1O2) is generated as one of the
major ROS in this system. Singlet oxygen is evolved in
mammalian cells under normal and various pathophys-
iological conditions [1,40]. Due to its relatively long
half-life, in the range of 10 to 50 μs in aqueous systems,
1O2 is capable of travelling appreciate distance in the
cellular milieu causing considerable damage to the
cellular molecules.
Inhibition of cancer cells growth
We previously reported that the induction of cell
apoptosis by hydroxystilbenes is correlated well with
their antioxidant potency in different cancer cell lines
[26]. Therefore, we compared the anticancer activity
of these two molecules by measuring their ability to
inhibit HL-60 cells proliferation. As shown in Figure
5, both resveratrol and 3,4,4¢-THS, time- and dose-
dependently inhibit HL-60 cells growth under the
experimental conditions. 3,4,4¢-THS, with IC50
around 30 μM after 48 h incubation, is more potent
than resveratrol to prevent cancer cells proliferation.
Our data demonstrate that 3,4,4¢-THS protects
biomolecules from either peroxyl radicals- or O2-
1
induced oxidative damage more efficiently than does
resveratrol. The only structural difference between
the two molecules is one hydroxyl position on the
benzene ring (Figure 1). This subtle structure modi-
fication remarkably improves the antioxidant potency
of resveratrol, and thus provides information to
develop novel antioxidants with higher activity derived
from resveratrol. Resveratrol has no effect against free
radical-induced lipid peroxidation under the experi-
mental conditions. In stark contrast, 3,4,4¢-THS
inhibits lipid peroxidation efficiently. This difference
should be due to the lower reactivity of resveratrol
toward free radicals [21,27]. In line with the above
results, 3,4,4¢-THS decays gradually during the pro-
cess of lipid peroxidation, while resveratrol doesn’t
(Figure 2B). In our previous work, 3,4,4¢-THS syn-
ergistically prevents free radical-induced lipid peroxi-
dation with TOH in micelles [21], however, no such
effect was observed in homogeneous solution. The
synergistic antioxidant effect between polyphenols
andTOH is believed due to the regeneration ofTOH
radicals by polyphenols, and thus retards the con-
sumption of TOH [21,41,42]. We reasoned that the
lipid peroxidation in homogeneous solution is a pure
chemical reaction, which is quite different from that
in micelles, where the TOH radicals is fairly
stable and accumulate with polyphenols in a certain
microenvironment, and hence the TOH regeneration
reaction could took place [21]. However, there is no
such a microenvironment in homogeneous solution,
and thus no TOH regeneration happens. To verify
this, we monitored the decay of antioxidants in the
process of lipid peroxidation. As shown in Figure 2B,
although the decay of TOH in the presence of 3,4,4¢-
THS is slower than that without 3,4,4¢-THS, there is
no obvious lag phase, which means no TOH regen-
eration. We also noticed the same pattern for the
decay of 3,4,4¢-THS, i.e. its consumption became
Discussion
3,4,4¢-THS and resveratrol were synthesized with ref-
erence to the modified Wittig reaction. Diethylben-
zylphosphonate, which was prepared from methoxy
substituted benzyl chloride and triethyl phosphite,
reacts with methoxy substituted benzaldehyde to
afford methoxylstilbene. The methyl-protected res-
veratrol and 3,4,4¢-THS further reacted with pyridine
hydrochloride, a demethylating reagent, to yield the
desired compounds.The synthesis route was exempli-
fied in Scheme 1.This procedure gives exclusively the
trans-isomer, and the yield depends predominantly on
the demethylation reaction.
Peroxyl radicals can be generated in vivo due to the
presence of lipid hydroperoxides (LOOH, the pri-
mary products of lipid peroxidation). For example,
metal ions can decompose LOOH to generate lipid
peroxyl radicals following the reaction shown in equa-
tion 1:1
LOOH ϩ Fe3ϩ ® LOO· ϩ Fe2ϩ ϩ Hϩ
(1)
To mimic the effect of peroxyl radicals in vivo, AAPH,
a water-soluble azo compound, is widely used as an
artificial source of peroxyl radicals [21,37–39].Ther-
mal decomposition of AAPH in the aqueous solution
produces an initiating radical (R·) which can attack
the polyunsaturated lipids (LH) to induce lipid per-
oxidation (Eqs. 2–6) [21]. On the other hand, if anti-
oxidant (AH), such as vitamin E, are present, it would
react with the chain propagating peroxyl radicals to
stop the peroxidation (Eq. 7), hence inhibiting lipid
peroxidation.

Resveratrol analogue antioxidation 1385
meta-dihydroxy structure, which makes it unfavor-
able reacting with free radicals. The proposed anti-
oxidant mechanism for scavenging peroxy radicals
by 3,4,4¢-THS is illustrated in Scheme 2. Singlet
oxygens, unlike peroxy radicals, have low reactivity
towards phenolic compounds. Quenching singlet
oxygen back to triplet by unsaturated molecules like
β-carotene and lycopene is a general pathway to
detoxify singlet oxygen [45,46]. And thus, the pro-
tective effect against singlet oxygen-induced DNA
damage by resveratrol and 3,4,4¢-THS likely follows
the same mechanism. This is further supported by
the recent work by Celaje et al. [47], which demon-
strated that physical quenching of singlet oxygen by
resveratrol accounts for 75% of the total removal of
singlet oxygen.
The results described in this paper support our pre-
vious hypothesis that the cytotoxic potency of hydrox-
ystilbenes is correlated well with their antioxidant
activity, i.e. compounds with higher antioxidant activ-
ity can suppress cancer cells proliferation more effi-
ciently. Resveratrol was reported an inhibitor of DNA
synthesis in mammalian cells via inhibition of the
ribonucleotide reductase (RNR) [48].This inhibition
is caused due to the scavenging tyrosyl free radicals
within RNR.The tyrosyl radical is a key intermediate
in reduction of ribonucleotides into the correspond-
ing deoxyribonucleotides [49], which are the basic
building blocks for DNA synthesis. In this work, our
results demonstrated that 3,4,4¢-THS is a more potent
antioxidant than resveratrol in preventing ROS-in-
duced oxidative damage. So it is rational to expect
that 3,4,4¢-THS is a stronger inhibitor for RNR than
resveratrol.This hypothesis is supported by the recent
publications where resveratrol analogues were
reported to inhibit deoxyribonucleoside triphosphates
synthesis and direct inhibitors of RNR [50,51]. Since
the proliferation of cancer cells is crucially dependent
on the constant DNA synthesis, the better inhibitor
of RNR should suppress DNA synthesis more severely.
This accounts for the better activity of 3,4,4¢-THS
towards cancer cells than resveratrol, and might be
applied to understand that why dietary antioxidants
usually have anticancer activity. Taken together, inhi-
bition of RNR by antioxidants through quenching
tyrosyl radicals is a possible pathway to explain the
anticancer property of antioxidants, and it still needs
much more evidence to evaluate and understand
the relationship between antioxidant activity and
anticancer effect.
Figure 5. Cytotoxicity of resveratrol and 3,4,4¢-THS in inhibition
of HL-60 cells proliferation. HL-60 cells were treated with either
resveratrol or 3,4,4¢-THS for 24 h (panel A) or 48 h (panel B), and
the cell viability was determined using the MTT assay as described
in materials and methods. Values are the mean Ϯ SD.
slower when TOH is present. Together, these results
indicate thatTOH and 3,4,4¢-THS counteract lipid per-
.
oxyl radicals (LOO, Eq. 7) independently in the solu-
tion, and there is no chance for TOH regeneration.
Why 3,4,4¢-THS is a more powerful antioxidant
than resveratrol? The recent works from our group as
well as other groups reported that trans-stilbene com-
pounds bearing ortho-dihydroxyl and/or para-dihydroxyl
functionalities possess remarkably higher antioxidant
activity than those bearing no such functionalities
[21,24,26,27,43,44]. These are understood as the fol-
lows. First, molecules bearing ortho-dihydroxyl group
have lower bond energy for H-O, which would make
them to react easily with ROS to form ortho-semiqui-
none radical intermediates through direct hydrogen-
abstraction; Secondly, the ortho-dihydroxy would make
the ortho-semiquinone intermediates fairly stable via an
intramolecular hydrogen bond and the 4¢-hydroxyl
group can stabilize the semiquinone radical intermedi-
ate by resonance through the trans-double bond; And
thirdly, this ortho-semiquinone intermediate is easily
to be further oxidized to finally stable product ortho-
quinone [27]. Resveratrol, not like 3,4,4 -THS, has a
In summary, we demonstrated that simple struc-
tural modification of resveratrol by changing the
5-hydroxy group to the 4-position remarkably
increases its antioxidant activity and cytotoxicity.
These findings contribute new data to enrich the struc-
ture-activity relationships of hydroxystilbenes, and
shed lights in modifying natural products for better
activity.

1386 W. Cai et al.
Scheme 2. Proposed antioxidative mechanism of 3,4,4¢-THS.
[11] Ghosh N, Ghosh R, Mandal SC. Antioxidant protection: A
Declaration of interest
promising therapeutic intervention in neurodegenerative dis-
ease. Free Radic Res 2011;45(8):888–905.
The authors report no conflicts of interest. The
authors alone are responsible for the content and
writing of the paper.This work was supported by the
National Natural Science Foundation of China
(21002047), the Ministry of Education of China
(20100211110027) and Lanzhou University (the
Fundamental Research Funds for the Central Uni-
versities, lzujbky-2009-73).
[12] Velioglu-Ogunc A, Sehirli O,Toklu HZ, Ozyurt H, Mayadagli
A, Eksioglu-Demiralp E, et al. Resveratrol protects against
irradiation-induced hepatic and ileal damage via its anti-
oxidative activity. Free Radic Res 2009;43(11):1060–1071.
[13] Somoza V. MAGIC-OL Resveratrol. Mol Nutr Food Res
2005;49(5):373.
[14] Milne JC, Lambert PD, Schenk S, Carney DP, Smith JJ,
Gagne DJ, et al. Small molecule activators of SIRT1 as ther-
apeutics for the treatment of type 2 diabetes. Nature 2007;
450(7170):712–716.
[15] Baur JA, Pearson KJ, Price NL, Jamieson HA, Lerin C, Kalra
A, et al. Resveratrol improves health and survival of mice on
a high-calorie diet. Nature 2006;444(7117):337–342.
[16] Constant J. Alcohol, ischemic heart disease, and the French
paradox. Coron Artery Dis 1997;8(10):645–649.
[17] Kopp P. Resveratrol, a phytoestrogen found in red wine. A
possible explanation for the conundrum of the ‘French para-
dox’? Eur J Endocrinol 1998;138(6):619–620.
[18] Jang M, Cai L, Udeani GO, Slowing KV,Thomas CF, Beecher
CW, et al. Cancer chemopreventive activity of resveratrol, a
natural product derived from grapes.Science 1997;275(5297):
218–220.
[19] Pervaiz S, Holme AL. Resveratrol: its biologic targets and
functional activity. Antioxid Redox Signal 2009;11(11):
2851–2897.
[20] Danz ED, Skramsted J, Henry N, Bennett JA, Keller RS.
Resveratrol prevents doxorubicin cardiotoxicity through mito-
chondrial stabilization and the Sirt1 pathway. Free Radic Biol
Med 2009;46(12):1589–1597.
References
[1] Halliwell B, Gutteridge J. Free Radicals in Biology and Med-
icine. 4th ed. USA: Oxford University Press; 2007.
[2] Coyle JT, Puttfarcken P. Oxidative stress, glutamate, and neu-
rodegenerative disorders. Science 1993;262(5134):689–695.
[3] Lin MT, Beal MF. Mitochondrial dysfunction and oxidative
stress in neurodegenerative diseases. Nature 2006;443(7113):
787–795.
[4] Rice ME. Ascorbate regulation and its neuroprotective role in
the brain. Trends Neurosci 2000;23(5):209–216.
[5] Martin HD, Kock S, Scherrers R, Lutter K, Wagener T,
Hundsdorfer C, et al. 3,3 -Dihydroxyisorenieratene, a natural
carotenoid with superior antioxidant and photoprotective
properties. Angew Chem Int Ed Engl 2009;48(2):400–403.
[6] Balsano C, Alisi A. Antioxidant effects of natural bioactive
compounds. Curr Pharm Des 2009;15(26):3063–3073.
[7] Aggarwal BB, Shishodia S. Molecular targets of dietary agents
for prevention and therapy of cancer. Biochem Pharmacol
2006;71(10):1397–1421.
[21] Fang JG, Lu M, Chen ZH, Zhu HH, Li Y, Yang L, et al.
Antioxidant effects of resveratrol and its analogues against the
free-radical-induced peroxidation of linoleic acid in micelles.
Chemistry 2002;8(18):4191–4198.
[8] Konishi T. “From herb to kitchen and bedside: food factors
are pharmacological molecules with antioxidant activity”.
Free Radic Res 2011;45(8):863.
[22] Holthoff JH, Woodling KA, Doerge DR, Burns ST, Hinson
JA, Mayeux PR. Resveratrol, a dietary polyphenolic phyto-
alexin, is a functional scavenger of peroxynitrite. Biochem
Pharmacol 2010;80(8):1260–1265.
[23] Kang SS, Cuendet M, Endringer DC, Croy VL, Pezzuto
JM, Lipton MA. Synthesis and biological evaluation of a
library of resveratrol analogues as inhibitors of COX-1,
COX-2 and NF-kappaB. Bioorg Med Chem 2009;17(3):
1044–1054.
[9] Nam TG, Rector CL, Kim HY, Sonnen AF, Meyer R,
Nau WM, et al. Tetrahydro-1,8-naphthyridinol analogues
of alpha-tocopherol as antioxidants in lipid membranes
and low-density lipoproteins. J Am Chem Soc 2007;129(33):
10211–10219.
[10] Muller A, Cadenas E, Graf P, Sies H. A novel biologically
active seleno-organic compound—I. Glutathione peroxidase-
like activity in vitro and antioxidant capacity of PZ 51 (Ebse-
len). Biochem Pharmacol 1984;33(20):3235–3239.

Resveratrol analogue antioxidation 1387
[24] Fang JG, Zhou B. Structure-activity relationship and mecha-
nism of the tocopherol-regenerating activity of resveratrol
[38] Rubbo H, Radi R, Anselmi D, Kirk M, Barnes S, Butler J,
et al. Nitric oxide reaction with lipid peroxyl radicals spares
alpha-tocopherol during lipid peroxidation. Greater oxidant
protection from the pair nitric oxide/alpha-tocopherol than
and its analogues.
11458–11463.
J Agric Food Chem 2008;56(23):
[25] Heynekamp JJ, Weber WM, Hunsaker LA, Gonzales AM,
Orlando RA, Deck LM, et al. Substituted trans-stilbenes,
including analogues of the natural product resveratrol, inhibit
the human tumor necrosis factor alpha-induced activation of
transcription factor nuclear factor kappaB. J Med Chem
2006;49(24):7182–7189.
alpha-tocopherol/ascorbate.
10812–10818.
[39] YangY, Song ZG, Liu ZQ. Synthesis and antioxidant capaci-
ties of hydroxyl derivatives of cinnamoylphenethylamine in
protecting DNA and scavenging radicals. Free Radic Res
2011;45(4):445–453.
J Biol Chem 2000;275(15):
[26] Cai YJ, Wei QY, Fang JG, Yang L, Liu ZL, Wyche JH, et al.
The 3,4-dihydroxyl groups are important for trans-resveratrol
analogs to exhibit enhanced antioxidant and apoptotic activi-
ties. Anticancer Res 2004;24(2B):999–1002.
[27] Shang YJ, Qian Y, Liu XD, Dai F, Shang XL, Jia WQ, et al.
Radical-scavenging activity and mechanism of resveratrol-
oriented analogues: influence of the solvent, radical, and sub-
stitution. J Org Chem 2009;74(14):5025–5031.
[28] Roberti M, Pizzirani D, Simoni D, Rondanin R, Baruchello
R, Bonora C, et al. Synthesis and biological evaluation of
resveratrol and analogues as apoptosis-inducing agents. J Med
Chem 2003;46(16):3546–3554.
[29] Mazue F, Colin D, Gobbo J,Wegner M, Rescifina A, Spatafora
C, et al. Structural determinants of resveratrol for cell prolif-
eration inhibition potency: experimental and docking studies
of new analogs. Eur J Med Chem 2010;45(7):2972–2980.
[30] de Lima DP, Rotta R, Beatriz A, Marques MR, Montenegro
RC, Vasconcellos MC, et al. Synthesis and biological evalua-
tion of cytotoxic properties of stilbene-based resveratrol ana-
logs. Eur J Med Chem 2009;44(2):701–707.
[40] Piette J. Biological consequences associated with DNA oxida-
tion mediated by singlet oxygen. J Photochem Photobiol B
1991;11(3–4):241–260.
[41] Iglesias J, Pazos M, Andersen ML, Skibsted LH, Medina I.
Caffeic acid as antioxidant in fish muscle: mechanism of syn-
ergism with endogenous ascorbic acid and alpha-tocopherol.
J Agric Food Chem 2009;57(2):675–681.
[42] Niki E. Interaction of ascorbate and alpha-tocopherol. Ann N
Y Acad Sci 1987;498:186–199.
[43] Frombaum M, Therond P, Djelidi R, Beaudeux JL,
Bonnefont-Rousselot D, Borderie D. Piceatannol is more
effective than resveratrol in restoring endothelial cell dimeth-
ylarginine dimethylaminohydrolase expression and activity
after high-glucose oxidative stress. Free Radic Res 2011;45
(3):293–302.
[44] Hung LM, Chen JK, Lee RS, Liang HC, Su MJ. Beneficial
effects of astringinin, a resveratrol analogue, on the ischemia
and reperfusion damage in rat heart. Free Radic Biol Med
2001;30(8):877–883.
[45] Di Mascio P, Kaiser S, Sies H. Lycopene as the most efficient
biological carotenoid singlet oxygen quencher. Arch Biochem
Biophys 1989;274(2):532–538.
[31] YamamotoY, Niki E, KamiyaY, Miki M, Tamai H, Mino M.
Free radical chain oxidation and hemolysis of erythrocytes by
molecular oxygen and their inhibition by vitamin E. J Nutr
Sci Vitaminol (Tokyo) 1986;32(5):475–479.
[32] Fang JG, Lu M, Ma LP,Yang L, Wu LM, Liu ZL. Protective
effects of resveratrol and its analogues against free radical-
induced oxidative hemolysis of red blood cells. Chinese
Journal of Chemistry 2002;20(11):1313–1318.
[46] Foote CS, Denny RW. Chemistry of singlet oxygen. VII.
Quenching by beta-carotene.
90(22):6233–6235.
J Am Chem Soc 1968;
[47] Celaje JA, Zhang D, Guerrero AM, Selke M. Chemistry of
trans-Resveratrol with Singlet Oxygen: [2 ϩ 2] Addition,
[4 ϩ 2] Addition, and Formation of the Phytoalexin Moracin
M. Org Lett 2011;13(18):4846–4849.
[48] Fontecave M, Lepoivre M, Elleingand E, Gerez C, Guittet O.
Resveratrol, a remarkable inhibitor of ribonucleotide reduct-
ase. FEBS Lett 1998;421(3):277–279.
[49] Holmgren A, Sengupta R.The use of thiols by ribonucleotide
reductase. Free Radic Biol Med 2010;49(11):1617–1628.
[50] Fritzer-Szekeres M, Savinc I, Horvath Z, Saiko P, Pemberger
M, Graser G, et al. Biochemical effects of piceatannol in
human HL-60 promyelocytic leukemia cells—synergism with
Ara-C. Int J Oncol 2008;33(4):887–892.
[51] Saiko P, Pemberger M, Horvath Z, Savinc I, Grusch M,
Handler N, et al. Novel resveratrol analogs induce apoptosis
and cause cell cycle arrest in HT29 human colon cancer cells:
inhibition of ribonucleotide reductase activity. Oncol Rep
2008;19(6):1621–1626.
[33] Epe B, Pflaum M, Boiteux S. DNA damage induced by pho-
tosensitizers in cellular and cell-free systems. Mutat Res
1993;299(3–4):135–145.
[34] Kumar SS, Ghosh A, Devasagayam TP, Chauhan PS. Effect of
vanillin on methylene blue plus light-induced single-strand breaks
in plasmid pBR322 DNA. Mutat Res 2000;469(2):207–214.
[35] Porter NA, Caldwell SE, Mills KA. Mechanisms of free radical
oxidation of unsaturated lipids. Lipids 1995;30(4):277–290.
[36] Goupy P,Vulcain E, Caris-Veyrat C, Dangles O. Dietary anti-
oxidants as inhibitors of the heme-induced peroxidation of
linoleic acid: mechanism of action and synergism. Free Radic
Biol Med 2007;43(6):933–946.
[37] Chao CC, MaYS, Stadtman ER. Modification of protein sur-
face hydrophobicity and methionine oxidation by oxidative
systems. Proc Natl Acad Sci U S A 1997;94(7):2969–2974.
This paper was first published online on Early Online on
28 October 2011.