Anticancer effects of the metabolic products of the resveratrol analogue, DMU-212: Structural requirements for potency

Article abstract of DOI:10.1016/j.ejmech.2011.03.049

The methoxylated trans-stilbene resveratrol analogue, (E)-3,4,5,4′- tetramethoxystilbene (1), has shown promising antiproliferative activity in in vitro cell line and in vivo models. In vivo 1 gives rise to several metabolic products through demethylation or hydroxylation reactions at the stilbene moiety. In the present study we examined the anticancer activity of 1 and the metabolites (E)-3′-hydroxy-3,4,5,4′-tetramethoxystilbene (2), (E)-4′-hydroxy-3,4,5-trimethoxystilbene (3), (E)-4-hydroxy-3,5,4′- trimethoxystilbene (4) and (E)-3-hydroxy-4,5,4′-trimethoxystilbene (5) by means of cell viability testing, cell cycle analysis, immunostaining and Western blotting. Compounds 1 and 2 exhibited submicromolar toxicity in MCF-7 breast adenocarcinoma and HepG2 hepatoma cells, whereas 3, 4 and 5 were inactive in terms of inhibition of cellular proliferation. Incubation with 1 or 2 at 10 μM for 24 h induced apoptosis and G2/M cell cycle arrest in MCF-7 and HepG2 cells. Immunostaining of MCF-7 cells for β-tubulin in the presence of either 1 or 2 revealed shorter localization of the protein around the nucleus, as compared to control cells. Western blot analyses further demonstrated that treatment with either 1 or 2 at concentrations between 30 and 50 μM for 24 h caused a downregulation in the levels of β-tubulin and cyclin D1 expression and an upregulation in the levels of p53 expression in MCF-7 and HepG2 cells. 2 further increased the ratio of mRNA levels of the apoptosis-related genes Bax/Bcl-xL in both MCF-7 and HepG2 cells in a dose-dependent manner. We conclude that 2 inhibits HepG2 and MCF-7 cellular proliferation by inducing apoptosis and G2/M arrest through p53 and Bax/Bcl-xL upregulation. Our findings further demonstrate that trimethoxy substitutions along with the presence of a methoxy group at position 4′ are necessary for retaining the activity of 1.

Full text of DOI:10.1016/j.ejmech.2011.03.049

European Journal of Medicinal Chemistry 46 (2011) 2586e2595
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Anticancer effects of the metabolic products of the resveratrol
analogue, DMU-212: Structural requirements for potency
,
d
,
b
c
a
Vasilis P. Androutsopoulos a , Ketan C. Ruparelia , Athanasios Papakyriakou , Harilaos Filippakis ,
*
Aristeidis M. Tsatsakis d, Demetrios A. Spandidos a
a Laboratory of Clinical Virology, University of Crete, Medical School, Voutes, 71003 Heraklion, Crete, Greece
b De Montfort University, Leicester School of Pharmacy, Leicester LE1 9BH, UK
c Laboratory of Chemical Biology of Natural Products and Designed Molecules, Institute of Physical Chemistry, NSCR Demokritos, 15310 Agia Paraskevi, Athens, Greece
d Laboratory of Toxicology, University of Crete, Medical School, Voutes, 71003 Heraklion, Crete, Greece
a r t i c l e i n f o
a b s t r a c t
The methoxylated trans-stilbene resveratrol analogue, (E)-3,4,5,40-tetramethoxystilbene (1), has shown
promising antiproliferative activity in in vitro cell line and in vivo models. In vivo 1 gives rise to several
metabolic products through demethylation or hydroxylation reactions at the stilbene moiety. In the
present study we examined the anticancer activity of 1 and the metabolites (E)-30-hydroxy-3,4,5,40-
tetramethoxystilbene (2), (E)-40-hydroxy-3,4,5-trimethoxystilbene (3), (E)-4-hydroxy-3,5,40-trimethox-
ystilbene (4) and (E)-3-hydroxy-4,5,40-trimethoxystilbene (5) by means of cell viability testing, cell cycle
analysis, immunostaining and Western blotting. Compounds 1 and 2 exhibited submicromolar toxicity in
MCF-7 breast adenocarcinoma and HepG2 hepatoma cells, whereas 3, 4 and 5 were inactive in terms of
Article history:
Received 15 November 2010
Received in revised form
17 March 2011
Accepted 22 March 2011
Available online 30 March 2011
Keywords:
Resveratrol
DMU-212
MCF-7 cells
HepG2 cells
Antiproliferation
Apoptosis
inhibition of cellular proliferation. Incubation with 1 or 2 at 10
cell cycle arrest in MCF-7 and HepG2 cells. Immunostaining of MCF-7 cells for
m
M for 24 h induced apoptosis and G2/M
-tubulin in the presence
b
of either 1 or 2 revealed shorter localization of the protein around the nucleus, as compared to control
cells. Western blot analyses further demonstrated that treatment with either 1 or 2 at concentrations
between 30 and 50 mM for 24 h caused a downregulation in the levels of b-tubulin and cyclin D1
expression and an upregulation in the levels of p53 expression in MCF-7 and HepG2 cells. 2 further
increased the ratio of mRNA levels of the apoptosis-related genes Bax/Bcl-xL in both MCF-7 and HepG2
cells in a dose-dependent manner. We conclude that 2 inhibits HepG2 and MCF-7 cellular proliferation
by inducing apoptosis and G2/M arrest through p53 and Bax/Bcl-xL upregulation. Our findings further
demonstrate that trimethoxy substitutions along with the presence of a methoxy group at position 40 are
necessary for retaining the activity of 1.
Cell cycle inhibition
b-Tubulin
Paclitaxel binding site
Molecular modeling
Docking
Ó 2011 Elsevier Masson SAS. All rights reserved.
1. Introduction
pathogenic attack. It is concentrated in high amounts in the skin of
grapes, cranberries, peanuts and other dietary constituents and has
Resveratrol is a naturally occurring phytoalexin, which is
produced in plants in response to stress and environmental
been the focus of extensive investigation due to its chemo-
preventive effect in in vitro and in vivo models. Resveratrol
possesses a wide spectrum of anticancer activities. In vitro it was
found to inhibit the formation of preneoplastic lesions in a 7,12-
dimethylbenz(a)anthracene (DMBA)-induced mouse mammary
organ culture [1], as well as the TCDD-mediated induction of
CYP1A1 mRNA and CYP1A1 activity in human hepatoma cells by
preventing the binding of the AhR to the CYP1A1 promoter [2].
Furthermore, resveratrol affects the cell cycle of cancer cells as it
inhibits the activities of protein kinase C and D [3,4]. It also induces
apoptosis [5] and decreases the activity of the transcription factors
NFkB and AP-1 [6,7]. In animal studies, resveratrol was found to
interfere with the formation of azoxymethane-induced aberrant
Abbreviations:
DMU-212,
(E)-3,4,5,40-tetramethoxystilbene;
DMU-214,
(E)-30-hydroxy-3,4,5,40-tetramethoxystilbene; DMU-281, (E)-40-hydroxy-3,4,5-tri-
methoxystilbene; DMU-291, (E)-4-hydroxy-3,5,40-trimethoxystilbene; DMU-807,
(E)-3-hydroxy-4,5,40-trimethoxystilbene; MTT, 3-(4,5-dimethylthiazole-2-yl)-2,5-
diphenyl tetrazolium bromide; DTT, DL-dithiothreitol; FACS, fluorescence-activated
cell sorting; p53, protein 53.
* Corresponding author. Laboratory of Toxicology, University of Crete, Medical
School, Voutes, 71003 Heraklion, Crete, Greece. Tel.: þ30 2810 394870; fax: þ30
2810 542098.
E-mail addresses: vandrou@med.uoc.gr, vasilis_androutsopoulos@yahoo.com
(V.P. Androutsopoulos).
0223-5234/$ e see front matter Ó 2011 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2011.03.049

V.P. Androutsopoulos et al. / European Journal of Medicinal Chemistry 46 (2011) 2586e2595
2587
cryptic foci in rat colon and decrease adenoma development in the
2. Results
small intestine of ApcMinþ mice [8,9].
Although the anticancer activities of resveratrol are apparent,
the efficacy of this natural compound is limited because of its poor
bioavailability. As in the case of polyhydroxylated flavonoids, the
chemical structure of resveratrol, which contains three free
hydroxyl groups, renders it susceptible to extensive phase-II
conjugation reactions in vivo. One study in humans demonstrated
the potential drawbacks of the poor bioavailability of resveratrol
[10]. To improve the pharmacokinetic properties of resveratrol and
extend its cancer-protecting activity, several synthetic analogues
have been generated and tested in in vitro models. One promising
strategy involves the introduction of methoxy substitutions in place
of the hydroxyl groups of the stilbene structure. Synthesis of
methoxylated analogues of resveratrol has been an active area of
research for the design of novel chemotherapeutic drugs with
improved pharmacological activities. Numerous studies have sup-
ported the conclusion that introduction of methoxy substitutions in
place of hydroxyl groups in the structure of resveratrol improves
the antiproliferative effect of the latter by means of induction of
apoptosis and cell cycle inhibition [5,11].
2.1. NMR and mass spectrometric characterization
2.1.1. (E)-3,4,5,40-Tetramethoxystilbene (1)
White crystals (0.20 g, 28%), m.p. 152e153 ꢀC; m/z [FAB] 301
([M þ 1]þ, 100%), dH (CDCl3) 3.80 (3H, s, OMe) 3.85 (3H, s, OMe),
3.90 (6H, s, 2 ꢁ OMe), 6.75 (2H, s, ArH), 6.93 (1H, d, J ¼ 16.6 Hz, C]
CH), 6.95 (2H, m, ArH), 7.00 (1H, d, J ¼ 16.6 Hz, C]CH), 7.48 (2H, d,
ArH). dC (CDCl3), 55.35, 56.19, 60.97, 103.53, 114.22, 126.65, 127.67,
127.81, 130.10, 133.48, 137.86, 153.47, 159.39; Anal. Calcd C18H20O4:
C, 71.98; H, 6.71. Found C, 71.66; H, 6.75. HRMS found [M þ H]þ
301.1434, C18H21O4 requires [M þ H]þ 301.1434.
2.1.2. 3-Tert-butyldimethylsiloxy-4-methoxybenzaldehyde
precursor of 2
Colourless oil (8.54 g 98%); m/z [FAB] 267 ([M]þ, 100%); dH
(CDCl3), 0.30 (6H, s, Si(CH3)2), 0.9 (9H, s, C(CH3)3), 3.80 (3H, s, OMe),
6.95 (1H, d, ArH), 7.35 (1H, d, ArH), 7.45 (H, dd, ArH), 9.00 (1H, s,
CHO); dC (CDCl3), 18.80, 26.02, 55.86, 55.95, 111.38, 111.62, 120.50,
122.74, 125.38, 126.55, 130.67, 191.22.
The fully methylated analogue of resveratrol, (E)-3,5,40-trime-
thoxystilbene, has been documented to possess strong anti-
angiogenic and antivascular activity [12]. Similarly, the stilbenoid
analogue of resveratrol, (E)-3,4,5,40-tetramethoxystilbene or DMU-
212 (1), possesses enhanced anticancer activity in terms of the
induction of apoptosis and antiproliferation compared to resvera-
trol [5]. In addition, 1 showed improved bioavailability in mouse
liver and plasma compared to resveratrol [13]. 1 escapes glucur-
onidation reactions due to its mehoxy groups and is metabolized
in vivo to four major metabolites (E)-30-hydroxy-3,4,5,40-tetrame-
thoxystilbene or DMU-214 (2), (E)-40-hydroxy-3,4,5-trimethox-
ystilbene or DMU-281 (3), (E)-4-hydroxy-3,5,40-trimethoxystilbene
or DMU-291 (4), and (E)-3-hydroxy-4,5,40-trimethoxystilbene or
DMU-807 (5) [13]. However, whether these metabolites retain the
biological activity of 1 remains unknown.
In order to elucidate the cancer therapeutic potential of 1, we
compared the biological activities of the latter with that of its
metabolites in cancer cell lines. The structures of resveratrol, 1 and
metabolites are shown in Fig. 1. In addition, our study aimed to
identify the structural components of 1 in regards to its metabolic
products, which are required for the antiproliferative activity of this
anticancer agent.
2.1.3. (E)-30-Hydroxy-3,4,5,40-tetramethoxystilbene (2)
White solid (0.50 g 73%); TLC: Rf 0.55 (ethyl acetate/petroleum
ether 4:6), m/z [FAB] 317 ([M þ 1]þ, 100%), dH (CDCl3) 3.90 (3H, s,
OMe), 3.85 (9H, s, 3 ꢁ OMe), 5.70 (1H, bs, OH), 6.7 (2H, s, ArH), 6.80
(1H, d, J ¼ 15 Hz, CH]CH), 6.90 (2H, d, J ¼ 15 Hz, CH]CH), 6.95 (1H,
d, ArH), 7.10 (1H, d, ArH). dC (CDCl3), 53.24, 56.15, 60.49, 61.01,
103.37, 110.73, 111.80, 119.28, 127.06, 130.99, 133.40, 137.67, 145.85,
152.89; HRMS found [M þ 1]þ 317.1386, C18H21O5 requires [M þ 1]þ
317.1384.
2.1.4. (E)-30-Hydroxy-3,4,5,40-tetramethoxystilbene (3)
White solid (0.10 g, 62%), m.p. 152e154 ꢀC; TLC: Rf 0.30 (ethyl
acetate/petroleum ether 2:8); m/z [FAB] 286 ([M þ 1]þ, 75%), dH
(CDCl3) 3.80 (3H, s, OMe), 3.90 (6H, s, 2 ꢁ OMe), 5.00 (1H, bs, OH),
6.70 (2H, s, ArH), 6.82 (2H, d, ArH), 6.85 (1H, d, J ¼ 16 Hz, CH]CH),
6.95 (1H, d, J ¼ 16 Hz, CH]CH), 7.39 (2H, d, ArH); HRMS found
[M þ 1]þ 287.1276, C17H19O4 requires [M þ 1]þ 287.1278; Anal.
Calcd C18H20O5.0.15H2O: C, 70.65; H, 6.38. Found C, 70.56; H, 6.56.
2.1.5. (E)-4-Hydroxy-3,5,40trimethoxystilbene (4)
White solid (0.10 g, 62%); TLC: Rf 0.30 (ethyl acetate/petroleum
ether 2:8); m/z [EI] 286 ([M]þ, 100%), dH (CDCl3) 3.80 (3H, s, OMe),
3.90 (6H, s, 2 ꢁ OMe), 5.50 (1H, bs, OH), 6.70 (2H, s, ArH), 6.82 (2H,
d, ArH), 6.85 (1H, d, J ¼ 14 Hz, CH]CH), 6.90 (1H, d, J ¼ 14 Hz, CH]
CH), 7.43 (2H, d, ArH); HRMS found [M þ 1]þ 287.1279, C17H19O4
requires [M þ 1]þ 287.1278; Anal. Calcd C17H18O4.0.15H2O: C, 70.65;
H, 6.38. Found C, 70.68; H, 6.31.
2.1.6. (E)-3-Hydroxy-4,5,40-trimethoxystilbene (5)
White solid (0.07 g, 98%); m.p. 152e154 ꢀC; TLC: Rf. ¼ 0.35 (ethyl
acetate/petroleum ether 1:1); Mass Spectrum m/z [EI] 287
([M þ 1]þ, 100%), dH (CDCl3, 400 MHz) 3.80 (3H, s, OMe), 3.90 (6H, s,
2 ꢁ OMe), 5.70 (1H, bs, OH), 6.60 (1H, d), 6.75 (1H, d), 6.90 (4H, m),
7.40 (2H, d); 13C NMR (CDCl3)
d: 53.34, 55.92, 61.06, 102.48, 105.96,
114.19, 126.40, 127.69, 127.99, 130.12, 130.24, 133.98, 135.16,149.42,
152.46, 159.35.
2.2. Effects of DMU compounds on proliferation of cancer cell lines
HepG2 and MCF-7
In the present study the antitumor effects of 1 and its metabo-
lites were investigated in cancer cell lines by means of cytotoxic
MTT assays and cell cycle analysis.1 and 2 exhibited submicromolar
Fig. 1. Chemical structures of resveratrol and the synthetic analogues 1, 2, 3, 4 and 5.

2588
V.P. Androutsopoulos et al. / European Journal of Medicinal Chemistry 46 (2011) 2586e2595
toxicity in the MCF-7 breast adenocarcinoma and liver hepatoma
HepG2 cancer cells, whereas 3, 4 and 5 were inactive in terms of
50% inhibition of cellular proliferation (IC50) (Fig. 2). 1 was slightly
2.4. Effect of 1 and 2 on
in MCF-7 and HepG2 cells
b-tubulin and cell cycle protein expression
more potent than 2 with an approximate IC50 of 0.3
m
M. Since 3, 4
Based on the initial observations that treatment with 1 and 2
results in cell cycle arrest of cancerous cells in the G2/M phase and
previous studies that demonstrated the antimitotic activity of 1 [5],
and 5 were considerably inactive in the MTT assay, we focused our
analysis on 1 and 2. In order to investigate which phase of the cell
cycle was more profoundly affected, MCF-7 and HepG2 cells were
the expression of b-tubulin was investigated following treatment of
treated with 10
m
M of 1 or 2 for 24 h, and the percentage of cells in
MCF-7 and HepG2 cells with both compounds by Western immu-
each phase of the cell cycle was measured by flow cytometry. Both
compounds induced G2/M arrest in MCF-7 and HepG2 cells with
a lesser degree of G1 and S phase arrest (Fig. 3). HepG2 cells were
more sensitive to both 1 and 2 treatments than MCF-7 cells. In
order to examine whether 1 and 2 were more active in rapidly
dividing cancer cells, their antitumor effects on HeLa uterine cervix
noblotting and immunofluorescence staining. MCF-7 cells were
treated with 0, 10 and 30
MCF-7 cells at concentrations of 10
b
m
M of 1 or 2 for 24 and 48 h. Incubation of
M for 24 h had no effect on
-tubulin distribution, as determined by immunostaining (data not
m
shown). Treatment of 30 mM 1 or 2 for 24 h decreased the locali-
zation of b-tubulin expression around the nucleus of the MCF-7
adenocarcinoma cells were investigated. Although a 10-
m
M treat-
cells (Fig. 4B). This effect was more profound at the 48 h period
(Fig. 4C). In order to prove that this effect was due to the agents
ment of 1 or 2 resulted in a greater degree of G2/M arrest in HeLa
cells as opposed to MCF-7 or HepG2 cells, the overall cytotoxic
effect, determined by the MTT assay, in the HeLa cells revealed
similar IC50 values with those reported in the latter two cell lines
(data not shown).
acting on
number of live cells, Western immunoblotting was employed to
examine the protein levels of -tubulin in both the MCF-7 and
HepG2 cells following treatment with 1 or 2. The compounds
induced a downregulation of -tubulin expression in the MCF-7
and HepG2 cells at concentrations of 30 M or higher (Figs. 5
b-tubulin rather than a cytotoxic effect decreasing the
b
b
m
2.3. Induction of apoptosis by 1 and 2
and 6). Both 1 and 2 caused a dose-dependent downregulation of
cyclin D1 levels, as well as an upregulation of the tumor-suppressor
protein p53 (Figs. 5 and 6). 1 was somewhat more potent in
inducing the aforementioned changes in both HepG2 and MCF-7
cells.
Initial investigations of cellular morphology using light
microscopy revealed floating 2-treated MCF-7 or HepG2 cells with
condensed cellular membrane. This led us to hypothesize that 2
induces apoptosis in a similar fashion as reported previously for 1
[5]. Cell cycle analysis demonstrated that, in addition to the G2/M
block, both 1 and 2 induced apoptosis of a small fraction of cells, as
determined by the subG1 phase (Fig. 3A and B). This result was
confirmed by DAPI staining which revealed nuclear fragmentation
2.5. Molecular modeling of 1 and 2 docked at the paclitaxel binding
site of b-tubulin
of MCF-7 cells after treatment with concentrations of 30
2 for 24 h (Fig. 4A). Nuclear fragmentation using DAPI staining was
not evident at lower concentrations of 10 M (data not shown).
m
M of 1 or
The trans-tetramethoxystilbene analogue 1, but not resveratrol
(Fig.1), has been recently shown to enhance tubulin polymerization
in vitro [5]. For that reason, compounds 1 and 2 are expected to
m
Fig. 2. The cytotoxicity of compounds 1, 2, 3, 4 and 5 in MCF-7 and HepG2 cells. Compounds 1, 2, 3, 4 and 5 were incubated at a concentration range covering 0.078e20
mM for 96 h
and cell viability measured using the MTT assay, as described in Experimental Protocols. Experiments were performed at least 3 times and error bars indicate S.D. of the mean.

V.P. Androutsopoulos et al. / European Journal of Medicinal Chemistry 46 (2011) 2586e2595
2589
Fig. 4. Induction of nuclear fragmentation and
b-tubulin sortening by 1 and 2 in MCF-
7 cells. Cells were grown on coverslips, treated with compounds for 24 and 48 h and
stained with -tubulin primary antibody. Fluorescence was measured using
a secondary anti-mouse IgG conjugated with AlexaFluor as described in Experimental
Protocols. A) DAPI staining following 30 M treatment of compounds 1 and 2 for
24 h. Red arrows indicate fragmented nuclei. B) Immunostaining for -tubulin at
24 h treatment and at C) 48 h treatment (Bars 10 M). (For interpretation of the
b
m
b
m
references to colour in this figure legend, the reader is referred to the web version of
this article.)
Fig. 3. Cell cycle profile of HepG2 and MCF-7 cells following treatment with 1 or 2. The
cells were treated with Propidium Iodide solution in PBS (50 g/mL) containing RNase
A (100 g/mL) and fluorescence was measured in a Beckman Coulter flow cytometer as
described in Experimental Protocols. A) Representative DNA analysis histograms. B)
Percentage of MCF-7 cells in each phase of the cell cycle measured, following 10 M of
2 treatment. C) Percentage of HepG2 cells in each phase of the cell cycle measured,
following 10
m
m
m
of the ab-tubulin dimer [15]. Compounds 1e5 and resveratrol were
docked at the paclitaxel binding site of b-tubulin (Fig. 7A) using full
m
M of 2 treatment. Error bars are S.D. of the mean for at least n ¼ 3
conformational flexibility and the new free-energy scoring function
of AutoDock4 (see Computational methods). All compounds were
predicted to bind at the paclitaxel site with at least 3 different
conformations (poses) and estimated free-energies of binding at
the range of ꢂ4.8 to ꢂ5.8 kcal/mol, with resveratrol exhibiting the
lowest values. The predicted binding energies correspond to
determinations.
mediate tubulin polymerization through binding at the paclitaxel
site of
trans-stilbene analogues with
lations were performed using the refined crystallographic structure
b
-tubulin [14]. To examine the putative interaction of the
b
-tubulin, molecular docking calcu-

2590
V.P. Androutsopoulos et al. / European Journal of Medicinal Chemistry 46 (2011) 2586e2595
Fig. 5. Effects of 1 on p53, cyclin D1 and
b-tubulin protein expression. MCF-7 and HepG2 cells were treated with 1 and western blotting carried out as described in Experimental
Protocols.
inhibition constants ranging from 50 to 300
m
M, which is in
approximately planar orientation of the phenyl rings of 1. Despite
of the predicted interactions between compounds 1 and 2 with
agreement with the cytotoxic profile of these compounds in HepG2
and MCF-7 cells. However, given that the standard error of the
scoring function is around 2.5 kcal/mol, an absolute ranking of
compounds 1e5 would be highly ambiguous.
critical residues of the paclitaxel site of
b-tubulin, the binding
affinity for -tubulin alone is not sufficient to elucidate their
b
inhibitory potency. This was also observed in a recent study on
resveratrol analogues, in which the authors suggest that many
other factors, such as absorption, distribution and metabolism,
must be taken under consideration to account for the cell prolif-
eration inhibition potency [16].
A
representative pose of the active analogue
a number of contacts with critical residues at the paclitaxel site of
-tubulin (Fig. 7B). Specifically, the methoxy groups at ring A of 1
1 displays
b
exhibit van der Waals contacts with residues Pro360 and Glu27, the
double bond is stacked on Phe272 and ring-B exhibits a “T-shaped”
aromatic interaction with His229 and hydrophobic interactions
with Leu217, Leu230 and Leu275 (Fig. 7B). Interestingly, its
metabolite, DMU-214 (2), is predicted to bind with an equal affinity
that is mediated by an additional hydrogen-bonding interaction
2.6. Effect of 2 on expression of apoptosis-related genes in MCF-7
and HepG2 cells
The apoptotic effect of 1 has been documented in a recent report
[5]. Therefore, we investigated the ability of 2 only, to induce
expression of the pro-apoptotic gene Bax and the anti-apoptotic
gene Bcl-xL by RT-PCR. 2 induced a dose-dependent upregulation
between the 30-OH of ring-B and Thr276 residue of
b-tubulin
(Fig. 7C). At such an orientation, the torsion angle between the
phenyl and ethylene groups of 2 is about 30ꢀ, in contrast to the
Fig. 6. Effects of 2 on p53, cyclin D1 and b-tubulin protein expression. MCF-7 and HepG2 cells were treated with DMU-214 and western blotting carried out as described in
Experimental Protocols.

V.P. Androutsopoulos et al. / European Journal of Medicinal Chemistry 46 (2011) 2586e2595
2591
Fig. 7. Molecular modeling of the trans-stilbenes binding to
b-tubulin. (A) Ribbon representation of
b-tubulin with bound paclitaxel shown using orange sticks for carbon atoms, red
for oxygen and blue for nitrogen. Representative poses of 1 (B) and 2 (C) docked into the paclitaxel binding site and colored according to (A). Residues of
b-tubulin that provide
contacts with the trans-stilbenes are shown with cyan carbons and hydrogen-bonding interactions are marked with dashed lines. Inset is also shown schematically the structural
requirements for the observed antiproliferative activity of 1. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this
article.)
of Bax mRNA levels in both MCF-7 and HepG2 cells. In contrast,
a downregulation of Bcl-xL mRNA levels was noted in MCF-7 cells
following 2 treatment, whereas the mRNA levels of this gene
remained relatively stable in the HepG2 cells (Fig. 8A). The overall
ratio of Bax/Bcl-xL mRNA levels was significantly increased in
A recent study has underlined the importance of methoxylation
in trans-stilbenes in increasing the toxicity towards
Caenorhabditis elegans [19]. The authors examined, amongst other
compounds, resveratrol trimethyl ether, pterostilbene, (E)-3-
hydroxy-5,40-dimethoxystilbene, pinostilbene and desoxy-
rhapontigenin [19]. The latter compounds are based on the struc-
ture of resveratrol and possess from one to three methoxy
substitutions in place of the hydroxyl groups of resveratrol. The
findings are very similar with those obtained in the present study.
The order of potency was as follows: resveratrol trimethyl ether >
(E)-3-hydroxy-5,40-dimethoxystilbene > pter-
ostilbene > desoxyrhapontigenin > pinostilbene. Taken collectively
with our results these data indicate that for maximum toxicity
methoxy substitutions on ring A and one methoxy substitution on
ring-B are required. This may indicate a favored binding orientation
to a putative biological target such as tubulin, or a desired phar-
macodynamic index that increases the bioavailability and bio-
stability of the methoxylated stilbenes.
HepG2 cells following treatment of 20, 30 and 40
mM of 2 for 24 h
and in MCF-7 cells following treatment of 30 and 40
mM (Fig. 8B).
3. Discussion
In the present study we evaluated the anticancer activities of
(E)-3,4,5,40-tetramethoxystilbene (1) and its metabolites. The most
important conclusion drawn by these findings is the structur-
eeactivity relationship that is required to preserve 1 potency in
inhibiting the proliferation of cancer cells. Compounds 3, 4 and 5
differ from 1 by the presence of a hydroxyl group in place of
a methoxy group. Compound 3 is the 40-demethylated analogue of
1, whereas 4 and 5 are the 4- and 3-demethylated analogues of 1,
respectively. Our results demonstrate that the 3,4,40-trimethoxy
substitution is necessary to retain 1 activity in inhibiting cancer cell
proliferation. Hydroxylation at positions 3, 4 or 40 abolishes the
activity of this compound, since 3, 4 and 5 were inactive in terms of
inhibition of cellular proliferation. This conclusion is illustrated as
inset in Fig. 7.
It is generally accepted that methoxylation of flavonoids and
structurally related natural compounds enhances their bioavail-
ability and half life in vivo [17,18]. Walle and colleagues have
demonstrated that methoxylated flavonoids show increased
bioavailability due to decreased glucuronidation and phase-II
conjugation reactions in vivo and in vitro [17,18]. Capping of free
hydroxyl groups with methoxy groups prevents glucuronidation in
liver and intestinal cells. This is one factor that has to be considered
when examining the antiproliferative activity of methoxylated and
hydroxylated stilbenes. Compounds that possess free hydroxyl
groups may present reduced pharmacological and antiproliferative
activities, due to rapid excretion.
Our findings indicate that 1 induced a decrease in
b-tubulin
expression as demonstrated by Western immunoblotting and
immunofluorescence microscopy. Tubulin is composed of a heter-
odimer which consists of two closely related proteins
a and
b
tubulin. Tubulin polymerizes to form microtubules that comprise
the main element for the formation of the mitotic spindle during
mitosis. Anti-tubulin drugs act by interfering with tubulin
assembly. There are two distinct modes by which anti-tubulin
drugs act. The first class of compounds binds preferentially to
polymerized tubulin and enhances its polymerization further [20].
This causes hyperassembly of microtubules and increased stability,
thus altering the dynamics of the mitotic spindle. Microtubules
appear shorter in length and frequently bundled compared to
normal untreated mitotic cells [20]. The second class of compounds
binds preferentially to the a,b-tubulin heterodimer, rather than the
polymerized form of tubulin and inhibit its polymerization [20].
This inhibits the assembly of microtubules and causes their disap-
pearance in both the interphase and mitotically arrested cells [20].

2592
V.P. Androutsopoulos et al. / European Journal of Medicinal Chemistry 46 (2011) 2586e2595
[24e26]. We observed that 2 caused an upregulation of the pro-
apoptotic gene Bax and a downregulation of the anti-apoptotic
gene Bcl-xL (in MCF-7 cells). The anti-apoptotic proteins Bcl-2 and
Bcl-xL are transcriptionally suppressed by p53 and along with the
pro-apoptotic protein Bax control the release of cytochrome c from
the mitochondria to the cytosol and the activation of caspase-3
which is one of the final-executioner proteins involved in the
apoptotic pathway. The ratio of pro-apoptotic to anti-apoptotic
protein levels is a critical determinant of the cell’s fate to enter
apoptosis. 2 increased the Bax/Bcl-xL mRNA levels in both HepG2
and MCF-7 cells, indicating that induction of apoptosis is a key
event for the mechanism of antiproliferative action of this drug.
Compounds 1 and 2 induced G2/M arrest to a greater extent
than apoptosis. This is a common phenomenon that has been
observed previously for structurally related stilbenes [27]. Anti-
cancer agents that disrupt the microtubule apparatus are associ-
ated with G2/M arrest of cancer cells followed by apoptosis [27].
Based on this evidence and the fact that the antiproliferation of 1
and 2 was attributed mainly to G2/M arrest, it is conceivable that
the G2/M block and
b-tubulin downregulation precede the induc-
tion of apoptosis, which is a secondary effect. In addition some
antimitotic agents such as nocodazole, which disrupts microtu-
bules by binding to b-tubulin, induce apoptosis in a p53-indepen-
dent manner [28,29]. Whether 1 and 2 are capable of inducing
apoptosis in the absence of p53 remains ill-defined, although the
results clearly indicate that in wild type p53 cancer cell lines, the
expression of this transcription factor is increased following
treatment by the latter compounds.
Fig. 8. The effect of 2 on the expression of apoptotic related genes Bax and Bcl-xL. 2
was incubated with MCF-7 or HepG2 cells for 24 h and RT-PCR carried out as described
in Experimental Protocols. A) Representative trace of mRNA expression of Bax and Bcl-
xL, following 2 treatment. B) Bax/Bcl-xL mRNA levels in MCF-7 and HepG2 cells.
Results are expressed as means ꢃ S.D. for at least n ¼ 3 determinations. *Statistically
different than control p < 0.05.
Compound 1 was originally designed to be a cytochrome P450
activated prodrug [30]. In vitro and in vivo metabolism of this agent
has revealed the conversion to the five metabolites 2, 3, 4, 5 and
DMU-295 [13]. The major metabolites found in plasma extracts
were DMU-295 (4,40 didesmethyl 1) and 4, whereas in mouse liver
microsomes 2 and 4 predominated [13]. Since DMU-295 occurs by
two demethylation reactions, it is tempting to speculate that it will
have decreased antiproliferative activity compared to 1, as both 3
and 4, the 40 and 4-demethylated analogues of 1, were inactive. In
addition, conversion to DMU-295 cannot proceed without
conversion to either 4 or 3. Collectively with the results by Sale and
colleagues, the data demonstrates that metabolism of 1 in vivo
leads notably to inactivation of its anticancer activity, rather than
activation, since 2 was produced in small quantities in mouse
plasma. In mouse liver microsomes a substantial conversion of 1 to
2 was noted along with the conversion to 4. In combination with
the results presented herein on the activity of 2 in human hepa-
toma HepG2 cells, the study by Sale et al. indicates that 1 could
possess some therapeutic implications for liver cancer, through the
conversion to 2 although further studies are required to confirm its
clinical efficacy in vivo.
Activation or deactivation of anticancer drugs is an essential
step towards their chemotherapeutic efficacy. Cytochrome P450
metabolism activates AQ4N to AQ4N-oxide, which demonstrates in
vivo antitumor activity, whereas hydroxylation of Taxol by cyto-
chrome P450 CYP1B1 results in loss of its anticancer activity
[31,32]. Consequently the successful use of an anticancer drug in
clinical trials requires extensive investigation of its in vivo metab-
olism as well as the determination of the antitumor activity of its
metabolites.
Compounds that belong to the first category include paclitaxel and
epothilones, whereas drugs that fall in the second class include
colchicine, combretastatin A4 and the vinca alkaloids [20]. Com-
bretastatin A4 (the cis-isomer of (E)-3,4,5,40-tetramethoxystilbene,
2) and related analogues mediate their anitumor effects by binding
at the colchicine site of tubulin [21,16]. Recently, the trans-stilbene
analogue 1, but not resveratrol (Fig. 1), has been shown to signifi-
cantly enhance tubulin polymerization in vitro [5]. Therefore,
compounds 1 and 2 are expected to mediate tubulin polymeriza-
tion through binding at the paclitaxel site of b-tubulin [14].
With respect to the observed antiproliferative structur-
eeactivity relationships of the trans-stilbene analogues studied
(Fig. 7 inset) their binding affinity for b-tubulin alone is not suffi-
cient to elucidate their inhibitory potency. As noted in a recent
study of a series of trans and cis resveratrol analogues, many other
factors such as absorption, distribution and metabolism must be
taken under consideration to account for their cell proliferation
inhibition potency [16]. The hydroxy-substitutions of compounds
3e5 probably renders them less lipophilic and more prone to
metabolic conversion and oxidation.
Induction of apoptosis by 2 and 1 was associated with an
upregulation of the cell cycle inhibitor p53. Both MCF-7 and HepG2
cells contain wild type p53 [22,23]. p53 induction is a mechanism
that protects the cell in response to UV radiation or DNA damage
caused by various agents from entering cell division. Apoptosis is
then activated, and the cell does not proceed to the next cell divi-
sion. There are two distinct pathways that lead to the induction of
apoptosis. The extrinsic pathway depends on extracellular signals
transduced through the death receptor complex, whereas the
intrinsic pathway is promoted in response to intracellular stress
from a number of factors. Induction of apoptosis involving modu-
lation of p53 and Bax/Bcl-2 family proteins has been shown to be an
important pathway that applies to several chemopreventive agents
In conclusion, the present study compared the antiproliferative
activity of the resveratrol analogue 1 with that of its metabolites 2,
3, 4 and 5. All of the metabolites with the exception of 2 were
inactive in inhibiting 50% proliferation of MCF-7 and HepG2 cells,
whereas 2 revealed similar potency with that of 1 via induction of
apoptosis and G2/M block. Antimitotic agents that contain methoxy
groups and are based on the chemical structure of resveratrol are

V.P. Androutsopoulos et al. / European Journal of Medicinal Chemistry 46 (2011) 2586e2595
2593
a widely acknowledged class of anticancer compounds. However,
in vivo metabolism of these molecules can give rise to metabolites
with distinct biological properties similar to that of their parent
compound. Our study reinforces the antitumor potential of the
parent compound 1 and provides more evidence on the in vivo
anticancer activity of its metabolic products.
extracted with ethyl acetate (3 ꢁ 20 mL). The combined organic
extracts were washed with water (2 ꢁ 20 mL), brine (2 ꢁ 20 mL)
and dried over anhydrous magnesium sulphate. The solvent was
removed in vacuo. Flash column chromatography (petroleum ether
40e60 with an increasing gradient of ethyl acetate (20e40%))
afforded 1 as white crystals.
4. Conclusion
5.2.2. Synthesis of 3-tert-butyldimethylsiloxy-4-
methoxybenzaldehyde (Step 1)
We have synthesized the anticancer compound (E)-3,4,5,40-tet-
ramethoxystilbene (1) and its metabolites (E)-30-hydroxy-3,4,5,40-
tetramethoxystilbene (2), (E)-40-hydroxy-3,4,5-trimethoxystilbene
(3), (E)-4-hydroxy-3,5,40-trimethoxystilbene (4) and (E)-3-hydroxy-
4,5,40-trimethoxystilbene (5)and compared the anticanceractivities
by conventional biological assays. 2 exhibited similar yet somewhat
lower potency compared to 1, whereas the remaining three
metabolites were considerably inactive. 2 further revealed a similar
mechanism of action with that of 1 via: (i) induction of p53 and Bax/
Bcl-xL upregulation and consequently induction of apoptosis, (ii)
tert-Butyldimethylsilyl chloride (6 g, 39.6 mmol) was added to
a stirred solution of 3-hydroxy-4-methoxybenzaldehyde (5.0 g,
33 mmol) and N,N-diisopropylethylamine (8.5 mL, 50 mmol) in
anhydrous DMF (100 mL) under nitrogen. After 2 h the reaction was
quenched with water (100 mL) and extracted with ethyl acetate
(3 ꢁ100 mL). The combined organic extracts were washed with
saturated sodium bicarbonate (3 ꢁ 50 mL), brine solution
(3 ꢁ 50 mL) and dried over anhydrous magnesium sulphate. The
solvent was removed in vacuo to afford a colourless oil.
induction of G2/M arrest, and (iii)
b-tubulin downregulation. The
5.2.3. Synthesis of 3-(tert-butyldimethylsiloxy)-4, 30,40,50-
structural components responsible for the potency of 1 have been
identified as the 40-methoxy, the 3-methoxy and the 4-methoxy
substitutions. In addition the results presented support the
conclusion that the antitumor activity of 1 in vivo is dependent
mainly on the conversion to the metabolite 2.
tetramethoxystilbene (Step 2)
n-Butyllithium solution (5.5 mL, 14 mmol, 1.6 M in hexanes) was
added dropwise to a solution of 3,4,5-trimethoxybenzyltriphenyl-
phosphonium chloride (6 g, 13 mmol) in anhydrous THF (100 mL)
at ꢂ20 ꢀC under nitrogen. The reaction was stirred for 20 min after
which a solution of 3-tert-butyldimethylsiloxy-4-methoxybenz-
aldehyde (3.33 g, 10 mmol) in anhydrous THF (25 mL) was added
dropwise. The reaction was allowed to warm to room temperature
and stirred for 3 h. The reaction was quenched with ice-water
(100 mL) and extracted with ethyl acetate (3 ꢁ 50 mL). The
combined organic extracts were washed with brine solution
(3 ꢁ 50 mL), dried over anhydrous magnesium sulphate and the
solvent removed in vacuo. Flash column chromatography (SiO2,
pet:ether with an increasing gradient of ethyl acetate (0e2%))
afforded TBDMS-protected DMU-214 as a white solid.
5. Experimental protocols
5.1. Materials and chemicals
MTT, PI, tissue culture reagents and media, Western blotting
lysis buffer, DTT and DAPI were from SigmaeAldrich (Life Science
Chemicals, Athens, Greece). Western blotting reagents were from
Biorad (Biorad Laboratories, Athens, Greece) whereas RT-PCR
reagents from Promega (SB Biotechnology Suppliers, Athens,
Greece). Polyclonal antibody for cyclin D1 and monoclonal anti-
body for p53 were from Santa Cruz Biotechnologies (Heidelberg,
5.2.4. Synthesis of (E)-30-Hydroxy-3,4,5,40-tetramethoxystilbene (2)
(Step 3)
Germany). Monoclonal antibodies for b-tubulin and b-actin were
from SigmaeAldrich (Life Science Chemicals, Athens, Greece).
Secondary antibodies for western blotting were from Santa Cruz
Biotechnologies, whereas for immunofluorescence from Invitrogen
(Antisel SA, Thessaloniki, Greece). Reagents for chemistry synthesis
were purchased from SigmaeAldrich (Poole, UK) or Fisher Scien-
tific (Loughborough, UK).
A solution of tetrabutylammonium fluoride in THF (2.2 mL,
2.2 mmol, 1 M in THF) was added to a stirred solution of 3-(tert-
butyldimethylsiloxy)-4,30,40,50-tetramethoxystilbene (0.94 g, 2.2
mmoles) in THF (20 mL) at room temperature. After 40 min, the
reaction was quenched with water (10 mL) and extracted with ethyl
acetate (3 ꢁ 5 mL). The combined organic extracts were dried over
anhydrous magnesium sulphate and the solvent removed in vacuo.
Flash column chromatography (SiO2, pet:ether (40:60) with an
increasing gradient of ethyl acetate (10e30%)) afforded DMU-214 as
white solid.
The 1H and 13C NMR spectra were recorded on a 400-MHz
super-conducting Bruker Spectrometer (Karlsruhe, Germany) at
30 ꢀC. Mass spectra were recorded on a Micromass Quattro II low
resolution triple quadrupole mass spectrometer (EPSRC National
Mass Spectrometry Service Centre, Swansea UK).
The above procedure was followed similarly for the synthesis of
3, 4, and 5.
5.2. Synthesis of (E)-3,4,5,40-tetramethoxystilbene (1) and
metabolites
5.3. Cell culture
5.2.1. Synthesis of (E)-3,4,5,40-tetramethoxystilbene (1)
MCF-7, HepG2 and HeLa cells were maintained in RPMI-1640
(MCF-7) and DMEM (HeLa and HepG2) containing 10% (v/v) heat-
inactivated (56 ꢀC for 45 min to inactivate complement) fetal calf
serum at 37 ꢀC in 5% CO2/95% air with 100% humidity and were
passaged using trypsin-EDTA. Cultured cells were routinely
passaged every 3e4 days.
A solution of n-butyllithium in hexanes (1.49 mL, 2.4 mmol,
1.6 M in hexane) was added dropwise to a stirred suspension of the
appropriate benzyltriphenylphosphonium chloride (2.4 mmol) in
anhydrous tetrahydrofuran (THF) (20 mL), at ꢂ20 ꢀC, under
nitrogen. The resulting red suspension was stirred for 20 min
at ꢂ20 ꢀC and then the appropriate benzaldehyde (2.4 mmol), in
anhydrous THF (10 mL) was added dropwise. The reaction was
stirred for 1 h at ꢂ20 ꢀC and then allowed to warm to room
temperature and stirred overnight. The progress of the reaction was
monitored by TLC [Rf ¼ 0.69 (ethyl acetate/petroleum ether 4:6)].
The reaction mixture was quenched with ice-water (40 mL) and
5.4. MTT assay
MCF-7, HepG2 and HeLa cells were plated in 96-well flat
bottomed plates and left to grow for 24 h. The following morning
the medium was removed, and compounds 1, 2, 3, 4, 5 were added

2594
V.P. Androutsopoulos et al. / European Journal of Medicinal Chemistry 46 (2011) 2586e2595
at serial dilutions at a concentration range of 0.079e20
for 96 h. Cell viability was measured spectrophotometrically using
MTT as a substrate, as described in a previous study [33].
m
M and left
room temperature for 1.5 h. The membrane was finally exposed to
ECL reagents, and the protein profile was developed on film.
5.8. RT-PCR
5.5. FACS analysis
MCF-7 and HepG2 cells were seeded in T25 flasks and treated
MCF-7 or HepG2 cells were cultured in T75 flasks. The cells were
with DMU-214 at a concentration range of 20e40 mM for 24 h. Total
serum starved for 24 h and then compounds 1 and 2 (10
m
M) were
RNA was extracted using Trizol. Briefly, 1 mL of Trizol reagent was
added to each flask, and the resulting sample was mixed with
200 mL of chloroform and centrifuged at 14,000 rpm for 15 min at
added for 24 h. The cells were washed once with PBS, trypsinized
and collected in a sterile Eppendorf. The pellet was washed again
with PBS, to remove traces of trypsin and medium and was fixed in
70% ethanol for at least 2 h. PI was added to the cellular pellet at
4 ꢀC. The top layer containing the RNA was removed, mixed with an
equal volume of ice-cold isopropanol and stored at ꢂ20 ꢀC over-
night. The next morning the samples were centrifuged at
13,000 rpm for 5 min. Following centrifugation, the supernatant
was removed and the resulting pellet containing the RNA was
further precipitated with the addition of 70% ice-cold ethanol and
centrifuged at 7500 rpm for 10 min. The supernatant was removed,
a concentration of 50 mg/mL containing RNase (100 mg/mL). Fluo-
rescence intensity was measured in a Beckman Coulter flow
cytometer at PMT4. At least 10,000 events were acquired. Data
analysis and acquisition were performed as described previously
[33].
and the RNA was dissolved in 25 mL of sterile H2O. Total RNA was
5.6. DAPI staining and immunofluorescence
checked for purity by spectrophotometry and gel electrophoresis.
RNA was treated with DNase for 45 min at 43 ꢀC in order to avoid
any possible contamination. A negative control sample was also
included in each PCR reaction to rule out any amplification of
contaminating DNA. cDNA was synthesized using a Promega kit,
and PCR was carried out as previously described [35]. Primer
sequences for Bax, Bcl-xL and b2 microglobulin were obtained from
previous studies [36]. Levels of mRNA were quantified by using
Scion Image software and expressed as numerical values relative to
MCF-7 cells were grown in a 24-well plate on coverslips at
a density of 8 ꢁ 104 cells/well. Compounds 1 and 2 were added to
the cells at 10 and 30 mM concentrations for 24 and 48 h. The cells
were washed twice with PBS and 1ꢁ fixation solution (Invitrogen,
Thessaloniki, Greece) was added for 10 min. The fixation solution
was removed and an additional washing step with PBS was per-
formed. The cells were then incubated for 10 min with 1X per-
meabilisation solution (Invitrogen, Thessaloniki, Greece) and then
washed twice with PBS/1%FBS. 30e50 mL of b-tubulin antibody was
added to the coverslips at 1:200 dilution in PBS/1%FBS and the cells
were incubated for 1 h at room temperature. The cells were washed
the corresponding
5.9. Computational methods
The coordinates of -tubulin were taken from the refined model
b2M signal.
gently twice with PBS/1%FBS and 20
mL of secondary anti-mouse
b
ꢀ
IgG conjugated with AlexaFluor were added at a 1:500 dilution in
PBS/1%FBS. The cells were incubated at room temperature for 1 h in
dark and washed twice with PBSe1%FBS. Coverslips were trans-
ferred in clean wells and DAPI working solution (5 mg/mL) was
added at 1:100 dilution for 5 min on the surface of the coverslips in
PBS/1%FBS. Excess DAPI was washed off twice by PBS/1%FBS and
the coverslips were placed on histology slides and stored at 4 ꢀC for
analysis. Fluorescence was measured using a Nikkon microscope as
described previously [34].
of the ab-tubulin dimer at 3.5 A resolution, which was retrieved
from the Protein Data Bank (PDB ID: 1JFF) [15]. The 3D coordinates
of compounds 1e5 were generated from their SMILES representa-
tion using Omega2.1 with standard parameters [37]. Auto-
DockTools1.5.4 was employed for the preparation of b-tubulin and
ligands with the united-atom approximation [38]. Briefly, only
polar hydrogen atoms were retained and Gasteiger atomic charges
were assigned to both protein and ligands. Grid maps were
ꢀ
centered at paclitaxel and comprised 41 ꢁ51 ꢁ41 points of 0.375 A
spacing. Docking calculations were performed using Auto-
Dock4.2.[39]. For each complex, 100 docking calculations were
performed using the Lamarckian genetic algorithm conformational
search with default parameters from Autodock3 [40]. The
maximum number of energy evaluations was set to 25 ꢁ106 and
5.7. Western blotting
MCF-7 or HepG2 cells were plated out in 6-well plates at
a density of 3 ꢁ 105 cells/mL and incubated at 37 ꢀC for 24 h in the
ꢀ
presence of 1 or DMU-214 (0, 30, 40, 50
mM). The medium was
the resulted conformations were clustered with a 2-A tolerance.
removed, and the cells were lysed with 100
m
L of lysis buffer con-
VMD 1.8.7 was used for analysis of the docking results and prepa-
ration of the figures [41]. Calculations were performed using an
x86_64 quad-core workstation running Linux kernel 2.6.29.
taining protease inhibitor cocktail and DL-dithiothreitol (DTT,
1 mM), sonicated on ice for 5 min and finally centrifuged at
13,000 rpm at 4 ꢀC for 10 min. The protein concentration required
for sample loading was estimated at 0.7 mg/mL for each sample,
and the protein lysate was mixed with sample buffer at a 1:1 ratio.
Samples were heated at 100 ꢀC for 5 min and then loaded on an
acrylamide gel at a ratio of 10% for the resolving and 5% for the
stacking gel. Electrophoresis was carried out for 1 h at 120 V, and
the proteins were transferred by wet blotting to a PVDF membrane.
The membrane was incubated in 10% milk/0.05% TBST at room
temperature for 1 h by continuous shaking and further incubation
5.10. Statistical analysis
Data are presented as the mean of at least three independent
measurements and were analysed by the paired t-test and one-way
ANOVA with Dunnet’s post test as indicated, using GraphPadPrism.
Conflict of Interest
The authors declare that they don’t have any conflict of interest.
with primary antibody for
detection, diluted in 5% milk/0.05% TBSTat 4 ꢀC overnight. Antibody
dilutions used were 1:400, 1:200, 1:200 and 1:400 for -tubulin,
p53, cyclin D1 and -actin, respectively. The membrane was
washed three times with 0.05% TBST and incubated with the
secondary antibody against HRP diluted in 5% milk/0.05% TBST at
b-tubulin, p53, cyclin D1 and b-actin
Acknowledgements
b
b
The authors are thankful to Professor Brenton and team at
EPSRC National Mass spectrometry facilities for carrying out anal-
ysis of synthesized compounds. This work was funded by the

V.P. Androutsopoulos et al. / European Journal of Medicinal Chemistry 46 (2011) 2586e2595
2595
[17] X. Wen, T. Walle, Drug Metab. Dispos. 34 (2006) 1786e1792.
[18] T. Walle, Free Rad. Biol. Med. 36 (2004) 829e837.
[19] M.A. Wilson, A.M. Rimando, C.A. Wolkow, BMC Pharmacol. 8 (2008) 15.
[20] E. Hamel, Cell Biochem. Biophys. 38 (2003) 1e21.
Prefecture of Heraklion, Crete, Greece, ELKE University of Crete
grant N. 2590.
[21] N.H. Nam, Curr. Med. Chem. 10 (2003) 1697e1722.
Appendix A. Supplementary material
[22] M. Tsai, J.A. Cook, G.V.R. Chandramouli, W. DeGraff, H. Yan, S. Zhao,
C.N. Coleman, J.B. Mitchell, E.Y. Chuang, Cancer Res. 67 (2007) 3845e3852.
[23] K. Boehme, Y. Dietz, P. Hewitt, S.O. Mueller, Toxicol. Lett. 198 (2010) 272e281.
[24] P. Zheng, L. Chiang, C. Lin, Life Sci. 76 (2005) 1367e1379.
[25] S.K. Katiyar, A.M. Roy, M.S. Baliga, Mol. Cancer Ther. 4 (2005) 207e216.
[26] C. Huang, W. Ma, A. Goranson, Z. Dong, Carcinogenesis 20 (1999) 237e242.
[27] H. Li, W.K.K. Wu, Z. Zheng, C.T. Che, L. Yu, Z.J. Li, Y.C. Wu, K.W. Cheng, J. Yu,
C.H. Cho, M. Wang, Biochem. Pharmacol. 78 (2009) 1224e1232.
[28] M. Casenghi, M. Mangiacasale, M. Tuynder, P. Caillet-Fauquet, A. Elhajouji,
P. Lavia, S. Mousset, M. Kirsch-Volders, E. Cundari, Exp. Cell Res. 250 (1999)
339e350.
[29] Y. Schneider, B. Fischer, D. Coelho, S. Roussi, F. Gosse, P. Bischoff, F. Raul,
Cancer Lett. 211 (2004) 155e161.
[30] G.A. Potter, P.C. Butler, K.C. Ruparelia, T. Ijaz, N.C. Wilsher, E. Wanogho,
H.L. Tan, T.T.V. Hoang, L.A. Stanley, M.D. Burke, Br. J. Cancer 86 (Suppl. 1)
(2002) S117.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.ejmech.2011.03.049.
References
[1] K.P.L. Bhat, D. Lantvit, K. Christov, R.G. Mehta, R.C. Moon, J.M. Pezzuto, Cancer
Res. 61 (2001) 7456e7463.
[2] H.P. Ciolino, P.J. Daschner, G.C. Yeh, Cancer Res. 58 (1998) 5707e5712.
[3] R.S. Haworth, M. Avkiran, Biochem. Pharmacol. 62 (2001) 218e220.
[4] S.J. Slatter, J.L. Seiz, A.C. Cook, B.A. Stagliano, C.J. Buzas, Biochem. Biophys. Acta
1637 (2003) 59e69.
[5] Z. Ma, O. Molavi, A. Haddadi, R. Lai, R.A. Gossage, A. Lavasanifar, Cancer
Chemother. Pharmacol. 63 (2008) 27e35.
[6] Y.J. Surh, K.S. Chun, H.H. Cha, S.S. Han, Y.S. Keum, K.K. Park, S.S. Lee, Mutat.
Res. 480e481 (2001) 243e268.
[7] S. Banerjee, C. Bueso-Ramos, B.B. Aggarwal, Cancer Res. 62 (2002) 4945e4954.
[8] L. Tessitore, A. Davit, I. Sarotto, G. Caderni, Carcinogenesis 21 (2000) 1619e1622.
[9] S. Sale, R.G. Tunstall, K.C. Ruparelia, G.A. Potter, W.P. Steward, A.J. Gescher, Int.
J. Cancer 115 (2005) 194e201.
[10] D.A. Goldberg, J. Yan, G.J. Soleas, Clin. Biochem. 36 (2003) 79e87.
[11] P. Saiko, M. Pemberger, Z. Horvath, I. Savinc, M. Grusch, N. Handler, T. Erker,
W. Jaeger, M. Fritzer-Szekeres, T. Szekeres, Oncol. Rep. 19 (2008) 1621e1626.
[12] M. Belleri, D. Ribatti, S. Nicoli, F. Cotelli, L. Forti, V. Vannini, L.A. Stivala,
M. Presta, Mol. Pharmacol. 67 (2005) 1451e1459.
[31] L.H. Patterson, G.I. Murray, Curr. Pharm. Des. 8 (2002) 1335e1347.
[32] M.C. McFadyen, H.C. McLeod, F.C. Jackson, W.T. Melwin, J. Doehmer,
G.I. Murray, Biochem. Pharmacol. 62 (2001) 207e212.
[33] V. Androutsopoulos, R.R. Arroo, J.F. Hall, S. Surichan, G.A. Potter, Breast Cancer
Res. 10 (2008) R39.
[34] P. Dimitropoulou, R. Caswell, B.P. McSharry, R.F. Greaves, D.A. Spandidos,
G.W. Wilkinson, G. Sourvinos, Eur. J. Cell Biol. 89 (2010) 758e768.
[35] V. Androutsopoulos, N. Wilsher, R.R.J. Arroo, G.A. Potter, Cancer Lett. 274
(2009) 54e60.
[36] G. Soufla, S. Baritaki, S. Sifakis, A. Zafiropoulos, D.A. Spandidos, Int. J. Biol.
Markers 20 (2005) 18e27.
[13] S. Sale, R.D. Verschoyle, D. Boocock, D.J.L. Jones, N. Wilsher, K.C. Ruparelia,
G.A. Potter, P.B. Farmer, W.P. Steward, A.J. Gescher, Br. J. Cancer 90 (2004)
736e744.
[14] E. Nogales, S.G. Wolf, K.H. Downing, Nature 391 (1998) 199e203.
[15] J. Lowe, H. Li, K.H. Downing, E. Nogales, J. Mol. Biol. 313 (2001) 1045e1057.
[16] F. Mazue, D. Colin, J. Gobbo, M. Wegner, A. Rescifina, C. Spatafora, D. Fasseur,
D. Delmas, P. Meumier, C. Tringali, N. Latruffe, Eur. J. Med. Chem. 45 (2010)
2972e2980.
[37] J. Kirchmair, G. Wolber, C. Laggner, T. Langer, J. Chem. Inf. Model. 46 (2006)
1848e1861.
[38] M.F. Sanner, J. Mol. Graph. Model. 17 (1999) 57e61.
[39] R. Huey, G.M. Morris, A.J. Olson, D.S. Goodsell, J. Comput. Chem. 28 (2007)
1145e1152.
[40] G. Morris, D. Goodsell, R. Halliday, R. Huey, W. Hart, R. Belew, A.J. Olson,
J. Comput. Chem. 19 (1998) 1639e1662.
[41] W. Humphrey, A. Dalke, K. Schulten, J. Mol. Graph 14 (1996) 27e38.