Structural insights into hydroxycoumarin-induced apoptosis in U-937 cells

Article abstract of DOI:10.1016/j.bmc.2007.11.038

In the present study, we sought to establish the effect of diverse structural-related hydroxycoumarins on the proliferation, cytotoxicity, and induction of apoptosis in promonocytic leukemic cells (U-937). The dihydroxylated coumarins, 7,8-dihydroxy-coumarin and esculetin, induced DNA fragmentation as well as characteristic morphological changes of programmed cell death in U-937 cells. With the aim to perform a structure-activity relationship study, the correlation between the physicochemical properties of the molecules and their pro-apoptotic activity was carried out. Results showed that the presence of two adjacent phenolic hydroxyl groups was the most important factor in terms of the SAR. The exposure of leukemic cells to 7,8-dihydroxy-coumarin evoked a phenoxyl radical generation that was detected by electron spin resonance spectroscopy. The present study suggests that reactive oxygen species generation plays a critical role in dihydroxycoumarin-induced apoptosis in U-937 cells. These findings further suggest that these compounds may have a potential therapeutic role in the treatment of hematological malignancies.

Full text of DOI:10.1016/j.bmc.2007.11.038

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 2665–2675
Structural insights into hydroxycoumarin-induced
apoptosis in U-937 cells
Maria E. Riveiro,a,d,e Albertina Moglioni,b,e Ramiro Vazquez,a,d,e Natalia Gomez,b,e
Graciela Facorro,c Lidia Piehl,c Emilio Rubin de Celis,c
Carina Shayod,e and Carlos Davioa,e,
*
aLaboratorio de Radioiso´topos, Facultad de Farmacia y Bioquı´mica, Argentina
bCa´tedra de Quı´mica Medicinal, Facultad de Farmacia y Bioquı´mica, Argentina
cCa´tedra de Fı´sica, Facultad de Farmacia y Bioquı´mica, Universidad de Buenos Aires, Argentina
dLaboratorio de Patologı´a Molecular, Instituto de Biologı´a y Medicina Experimental, Buenos Aires, Argentina
eConsejo Nacional de Investigaciones Cientı´ficas y Te´cnicas, Buenos Aires, Argentina
Received 4 October 2007; revised 10 November 2007; accepted 13 November 2007
Available online 19 November 2007
Abstract—In the present study, we sought to establish the effect of diverse structural-related hydroxycoumarins on the proliferation,
cytotoxicity, and induction of apoptosis in promonocytic leukemic cells (U-937). The dihydroxylated coumarins, 7,8-dihydroxy-cou-
marin and esculetin, induced DNA fragmentation as well as characteristic morphological changes of programmed cell death in U-
937 cells. With the aim to perform a structure-activity relationship study, the correlation between the physicochemical properties of
the molecules and their pro-apoptotic activity was carried out. Results showed that the presence of two adjacent phenolic hydroxyl
groups was the most important factor in terms of the SAR. The exposure of leukemic cells to 7,8-dihydroxy-coumarin evoked a
phenoxyl radical generation that was detected by electron spin resonance spectroscopy. The present study suggests that reactive oxy-
gen species generation plays a critical role in dihydroxycoumarin-induced apoptosis in U-937 cells. These findings further suggest
that these compounds may have a potential therapeutic role in the treatment of hematological malignancies.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
Great efforts have been made in the past years in the
search of new compounds for the treatment of leuke-
mias. Based on their low toxicity, relative low cost,
and their presence in the diet and herbal medicines, it
is highly interesting to examine the pharmacological
behavior and eventual future applications of coumarins.
In previous studies we reported that a group of natural
coumarins shows anti-proliferative and differentiation
activity in the promonocytic U-937 cell line.4 Recently,
we described that the 7,8-dihydroxy-4-methylcoumarin
induces selective apoptosis in U-937 and HL-60 leuke-
mic cells mediated by the activation or inhibition of
the major signaling pathways as the mitogen-activated
protein kinases (MAPKs), phosphoinositide-3-kinase/
Akt signaling cascade (PI3K/Akt), c-myc proto-onco-
gene, and caspase-3.5 In order to gain insight into the
pharmacological properties of the hydroxycoumarins,
we performed a study to understand the relationship be-
tween the structure and the biological activity.
Acute leukemia is a malignant neoplasm of the hemato-
poietic cells characterized by an abnormal proliferation
of precursor cells, decreased rate of apoptosis, and arrest
in cell differentiation.1 This makes induction of
apoptosis in leukemic cells a valuable therapeutic tool.
Unfortunately many leukemias turn refractory to che-
motherapy strongly supporting the necessity of alterna-
tive chemotherapeutic strategies.1 Nowadays, many of
the therapeutic agents used in the treatment of malig-
nant neoplasm including VP16, cisplatin, adriamycin,
and taxol exert their anti-cancer effects by inducing
apoptosis of cancer cells.2,3
Keywords: Hydroxycoumarins; Free radicals; Apoptosis induction;
Leukemia.
*
Natural as well as synthetic coumarins have recently
drawn much attention due to its broad pharmacological
Corresponding author. Tel.: +54 114964 8277x35; fax: +54 114786
2564; e-mail: cardavio@ffyb.uba.ar
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.11.038

2666
M. E. Riveiro et al. / Bioorg. Med. Chem. 16 (2008) 2665–2675
activities.6–8 Many coumarins and their derivatives exert
in vitro and in vivo scavenging of reactive oxygen spe-
cies (ROS), anti-coagulant,9 anti-inflammatory,10 and
anti-viral effects.11 In sensitive neoplasms, coumarins
and derivatives cause significant changes in cell growth
and differentiation.4,5 Furthermore, coumarin deriva-
tives modulate multidrug resistance transporters which
are intimately related with the resistance developed to
diverse chemotherapeutic therapies.12 Recent studies
showed that coumarins display antimicrobial activity
depending on their structure. The presence of a free 6-
OH on the coumarin nucleus is correlated with anti-
fungal activity, whereas a free 7-OH with antibacterial
activity. Moreover, a structure-activity relationship
study of coumarins acting on xanthine oxidase reveals
that at least one hydroxyl group is required for such
activity.6
tant mechanism for their pro-apoptotic properties in
cancer cells.22
The present study evaluates the key structural require-
ments of related hydroxycoumarins to induce apoptosis
in a leukemic cell line. We provide evidence of a clear
relationship between the pro-oxidant activity of hydrox-
ycoumarins and their ability to induce apoptosis. These
findings further suggest that these compounds may have
a potential therapeutic role in the future.
2. Results and discussion
We have recently demonstrated that the treatment with
micromolar concentrations of 7,8-dihydroxy-4-methyl-
coumarin (namely compound 8 in this report) induces
selective apoptosis in human leukemia cells.5 With the
aim to perform a structure–activity relationship (SAR)
study, the correlation between the physicochemical
properties of structurally related hydroxycoumarins
and their pro-apoptotic activity was evaluated.
Furthermore, hydroxycoumarins with an aromatic nitro
group display anti-proliferative and pro-apoptotic ef-
fects in renal carcinoma and leukemic cells.13,14 In addi-
tion, the most cytotoxic compound in lung carcinoma
and colon adenocarcinoma cell lines has at least two po-
lar functional groups on the aromatic ring of the cou-
marinic nucleus.15 The hydroxylated coumarins,
esculetin and scopoletin, exhibit anti-tumoral effects by
inducing programmed cell death in HL-60 cells.16,17
These studies strongly support that the biological activ-
ity and therapeutic applications of hydroxycoumarins
are tightly related to their structure, namely, the pattern
of substitution on the aromatic ring.
Scheme 1 illustrates the chemical synthesis of the cou-
marins that were biologically assayed, whereas Table 1
lists all the compounds analyzed. The coumarin com-
pounds (Scheme 1) were synthesized through the Pech-
mann reaction between a properly substituted phenol
and ethyl acetoacetate. The obtained yields ranged be-
tween 20% and 80%, depending on the starting phenol
used in the reaction. All the synthesized coumarins
had a methyl group in C-4 of the coumarin ring due
Classically dietary phenols and polyphenols are consid-
ered antioxidants, but recent studies show that they also
display pro-oxidant properties.18 Some plant isolated or
synthetic coumarins exhibit significant anti-oxidant
activity based on their ability to inhibit lipid peroxida-
tion and to scavenge reactive species.10,19,11 It has been
reported that dihydroxy-4-methylcoumarins favor the
inhibition of lipid peroxidation by scavenging free radi-
cals through chelation and by preventing the formation
of superoxide and other ROS.20 In a comprehensive re-
view, Hoult and Paya8 describe that dihydroxylated cou-
marins fraxetin, esculetin, 4-methylesculetin, daphnetin,
and 4-methyldaphnetin, are not only effective inhibitors
of Fe3+-ascorbate-dependent microsomal lipid peroxida-
tion and of aqueous alkylperoxyl radicals, but also scav-
engers of superoxide anion radicals. However coumarins
with ortho-dihydroxylation enhance hydroxyl radical
generation in the Fe3+–EDTA–H2O2 deoxyribose sys-
tem, but decrease it in the Fe3+–ascorbate–H2O2 deoxy-
ribose system, supporting that they can chelate iron ions
and also donate electrons, promoting a Fenton type
reaction. These findings support that hydroxylated cou-
marins may either behave as ROS scavengers or pro-oxi-
dant compounds depending on factors such as metal-
reducing potential, metal chelating behavior, pH as well
as solubility.21 Scopoletin induces apoptosis in HL-60
cells, but the effect is prevented by an antioxidant, sug-
gesting that the generation of ROS may contribute to
scopoletin-induced apoptosis.17 Various reports propose
that the pro-oxidant action of plant derived phenolics
rather than their antioxidant action may be an impor-
Scheme 1. General synthesis of coumarins.
Table 1. Structure of natural and synthetic coumarins
R0
R1
R2
R3
R4
R5
1
–H
–H
–H
–H
–H
–H
–H
–H
–H
–H
–H
–H
OH
–H
–H
–H
–H
–H
–H
–H
–H
–H
–H
–H
–H
–H
–H
–H
–H
–H
–H
–H
2
–OH
–OCH3
–H
–OH
–OH
–H
3
–H
–H
4
–CH3
–CH3
–CH3
–CH3
–CH3
–CH3
–CH3
–CH3
–CH3
–H
–CH3
–H
–H
5
–H
–CH3
–H
6
–CH3
–H
–H
7
–OH
–OH
–H
–H
8
–OH
–H
9
–OH
–H
10
11
12
13
–NH2
–OCH3
–OH
–H
–H
–H
–H
–H
–H
–H
–H

M. E. Riveiro et al. / Bioorg. Med. Chem. 16 (2008) 2665–2675
2667
to the ethyl acetoacetate as b-ketoester used in the Pech-
mann reaction. Compounds 1, 2, 3, 12, and 13 without a
methyl group in C-4 were commercial coumarins. The
methylation of the 7-hydroxy group or the change of
the hydroxyl group by an amine group in compound 7
was present in the commercial compounds 11 and 10.
The selection of this type of substitution led us to eval-
uate the influence of one or two hydroxy groups in dif-
ferent positions of the coumarin nucleus, as well as the
effect of the alkylation of the hydroxy group, and its
replacement by an amino or methyl group. The selected
compounds also differed in their substitution in C-4:
methyl, hydrogen or –CF3, groups. Each chemical group
exhibits different electronic properties which may even-
tually impact on the biological activity displayed by
these compounds.
The cytotoxic concentration 50 (CC50) for the tested
coumarins was determined by the Trypan blue dye
exclusion assay. The most potent cytotoxic activity
was observed for coumarins 8 and 2, followed by com-
pound 3 (Table 2). On the other hand, coumarins 7
and 9 as well as the other evaluated compounds did
not affect cell viability at a concentration lower than
2 mM (Table 2). Incubation of U-937 cells with
0.0001% (v/v) DMSO showed that the vehicle failed to
influence cell proliferation or to exhibit cytotoxicity
per se.
Previously, we demonstrated that compound 8 induced
programmed cell death in a concentration and time-
dependent manner in U-937 cells. A ladder-like pattern
of DNA fragments or morphological changes were
clearly observed at concentrations higher than
250 lM.5 A characterized key marker of cellular apop-
tosis is DNA cleavage into nucleosomal size fragments
of 180–200 base pairs or multiples thereof, which are de-
tected by gel electrophoresis as a DNA ladder.23 Figure
1 shows that DNA fragmentation became evident in U-
937 cells treated with 250 lM or 2 mM coumarins 8 or 2
for 24 h. Monohydroxycoumarins as 7, 3 or 9 did not in-
duce DNA fragmentation at 250 lM, but the ladder was
observed at 2 mM. Compound 12, a monohydroxy-
All coumarin analogues were tested for their potential
proliferative and cytotoxic effect on U-937 cells. Cell
growth, assessed by [3H] thymidine incorporation, was
inhibited by the two dihydroxylated coumarins (8 and
2) in a concentration-dependent manner. The inhibitory
concentration 50 (IC50) values for coumarins 8 and 2
were 48.65 1.50 lM and 31.33 1.20 lM, respectively
(Table 2). Coumarins 2 and 8 bearing two free hydroxyl
groups on the aromatic ring were more potent than cou-
marins 3, 7, and 9, which had only one free hydroxyl
aromatic group (IC50 values were 151.50 1.04,
710.90 1.26, and 627.70 1.78 lM, respectively). The
remainder coumarins failed to inhibit cell growth at a
concentration lower than 2 mM (Table 2). Similar re-
sults were obtained with a cellular meter Coulter Z-1
(data not shown).
Table 2. Effect of coumarins on U-937 cell proliferation and
cytotoxicity
Coumarins
CC50 (lM) IC50 (lM)
Means SEM Means SEM
Dihydroxycoumarins
8
2
165.40 1.80
194.10 2.34
48.65 1.50
31.33 1.20
Monohydroxycoumarins
7
>2000
710.90 1.26
151.50 1.04
>2000
3
600.20 8.54
>2000
>2000
12
13
9
>2000
627.70 1.78
>2000
Dimethyl coumarins
4
5
6
>2000
>2000
>2000
>2000
>2000
> 2000
4-Methyl-7-substituted coumarins
10
11
>2000
>2000
>2000
>2000
Coumarin
>2000
>2000
Cell growth of control and treated U-937 cells was assessed by [3H]
thymidine incorporation. Cellular toxicity was evaluated by the Try-
pan blue assay. IC50 and CC50 values were calculated with the equation
for sigmoidal dose response using Prism 4.00 for Windows (GraphPad
Software, San Diego, CA). Data are means SEM (n = 4).
Figure 1. DNA fragmentation analysis. After the incubation of cells
with 250 lM or 2 mM coumarins for 24 h the genomic DNA from
treated and untreated cells was subjected to agarose gel electropho-
resis. Similar results were obtained in three independent experiments.

2668
M. E. Riveiro et al. / Bioorg. Med. Chem. 16 (2008) 2665–2675
coumarin with a strong attractor electron group like tri-
fluoromethyl or 13 with a hydroxyl group in the pyrone
ring, failed to induce apoptosis at any concentration.
The other structurally related coumarins (1, 4, 5, 6, 10,
and 11) did not produce DNA fragmentation at
250 lM or 2 mM (Fig. 1). To further characterize pro-
grammed cell death induced by hydroxycoumarins, we
examined the nuclear morphology of treated and un-
treated cells by the Hoechst/propidium iodide mem-
brane permeability assay.24 Morphological changes as
condensed chromatin, fragment nuclei, and apoptotic
bodies are features of the apoptotic process.25 After
24 h of treatment with these structurally related number
of apoptotic cells was estimated by the appropriate
changes of nuclei stained with Hoechst 33342 observed
under a fluorescence microscope. Only cells treated with
250 lM of coumarin 8 or 2 for 24 h showed a number of
apoptotic cells significantly different with respect to con-
trol cells (Table 3). Figure 2A shows the morphological
changes corresponding to the apoptotic process in cou-
marin 8 treated cells. According to this, May Grun-
wald–Giemsa staining of U-937 cells previously
exposed to 250 lM coumarins for 48 h showed that only
coumarins 8 or 2 induced chromatin condensation, nu-
clear segmentation, and cell shrinkage. In contrast, U-
937 cells treated with compounds 7 or 1 did not exhibit
these characteristic morphological alterations (Fig. 2B).
Similar results were obtained in the presence of the
remainder coumarins (data not shown).
We next attempted to correlate the physicochemical
properties of the molecules with their in vitro activity.
Table 3. Effect of coumarins on U-937 cell apoptosis
Percentage of apoptotic U-937 cells
(means SEM)
Percentage of apoptotic U-937 cells
(means SEM)
U-937 Control cells
Compound 1
Compound 2
Compound 3
Compound 4
Compound 5
Compound 6
3.3 1.5
2.6 0.8
54.3 5.4**
6.3 1.8
4.6 0.8
5.3 1.4
4.3 1.2
Compound 7
Compound 8
Compound 9
Compound 10
Compound 11
Compound 12
Compound 13
5.2 2.4
55.0 3.2**
3.0 1.1
4.0 1.0
4.0 1.2
4.7 1.2
5.3 1.8
Cells were incubated with 250 lM of the assessed coumarins for 24 h.
Apoptotic cells were visualized by the appropriate changes of nuclei stained with Hoechst 33342/propidium iodide dyes and observed under a
fluorescence microscope under UV light at 430 nm and 100 cells were minimally counted. Values are mean SEM (n = 3); **p < 0.01 vs control
group.
Figure 2. Morphological changes in U-937 cells associated with the apoptotic process induced by coumarins. (A) Nuclear staining with Hoechst
33342/propidium iodide dyes. U-937 cells exposed to coumarin 8 for 24 h were observed under a fluorescence microscope at 430 nm. Arrows
represent condensed chromatin, fragmented nuclei, and apoptotic bodies. Each picture is representative of three independent experiments. (B) Cells
were exposed to 250 lM of coumarins 8, 2, 7, and 1 for 48 h. Treated and untreated U-937 cells were then stained with May Grunwald–Giemsa stain,
and cell morphology determined by light microscopy. Arrows represent chromatin condensation, nuclear segmentation, and cell shrinkage. Each
picture is representative of three independent experiments.

M. E. Riveiro et al. / Bioorg. Med. Chem. 16 (2008) 2665–2675
2669
Results showed that the presence of two adjacent phe-
nolic hydroxyl groups was the most important factor
in terms of SAR, whereas the position of the o-dihydr-
oxyl groups in the aromatic nucleus had little effect.
Similar SAR results were previously reported for
hydroxycoumarins as inducers of Cu2+-dependent
DNA strand breakage.26 Moreover we found that
other structural characteristic of coumarins also influ-
enced their pro-apoptotic activity. The methylation of
the 6-OH group in coumarin 3 reduced the biological
activity but the loss of activity was higher in monohy-
droxy-derivatives, as coumarin 7 or 9. We also ob-
served that the derivatives where the phenolic
hydroxyl group was replaced by amino or methyl
groups (compounds 10 and 4–6) failed to exhibit
pro-apoptotic activity in U-937 cells. Furthermore,
modification of the hydroxyl aromatic group by meth-
ylation (coumarin 11) led to a loss of the biological
activity. In accordance with our findings, it has
been reported that coumarin derivatives are potent
inhibitors of 5-hydroxy-6,8,11,14-eicosatetraenoic acid
(5-HETE) formation from arachidonic acid and 12-hy-
droxy-5,8,10-heptadecatrienoic acid in polymorphonu-
clear leukocytes. The inhibitory effect is dependent
on the presence of two adjacent phenolic hydroxyl
groups in the coumarinic structure at positions C-6
and C-7 or C-7 and C-8. On the other hand, mono-
hydroxycoumarins, as 7-hydroxycoumarin and scopo-
ROS has been shown to be an early signal for apopto-
sis.32 To confirm whether radical species were involved
in coumarin 8-mediated apoptosis in U-937 cells, the
production and levels of radical species were measured
by ESR spectroscopy in RPMI 1640 medium in the pres-
ence or absence of cells. The experimental (A) and sim-
ulated (B) ESR spectra of the coumarin 8* radical in
RPMI 1640 cultured U-937 cells are shown in Figure
3. ESR spectrum of the coumarin 8* radical showed a
complex signal of 17 peaks, compatible with the pres-
ence of the phenoxyl radical. Similar ESR spectrum of
the coumarin 8* radical was obtained in RPMI 1640
alone (data not shown). Scheme 2 shows the resonant
structures of the coumarin 8* radical and Table 4 sum-
marizes the hyperfine coupling constant values for the
1
six H under different experimental conditions. For the
simulation of this spectrum, the hyperfine interactions
1
of the six H of the semiquinone form of the coumarin
8* radical were considered. The simulation of the reso-
nant structures in Scheme 2 indicates that the spectrum
shown in Figure 3A is compatible with the phenoxyl
radical of coumarin 8. Given that coumarin 8 is a poly-
phenol, the coumarin 8 radical (QHÅ) may be trans-
formed by fast deprotonation, at pH 7.6 to its more
stable semiquinone form (QÅꢀ).33
RÅ;O2;Active metals
QH2
QHÅ
QHÅ
ð1Þ
ð2Þ
ƒƒƒƒƒƒƒƒƒ!
Q
Åꢀ þ Hþ
ƒ!
letin, show
a
weak inhibition of 5-HETE
formation.27 The presence of a methyl or hydrogen
group in position 4 of the coumarin ring in most of
the derivatives did not influence their pro-apoptotic
activity but the presence of a –CF3 group affected
the biological activity of monohydroxycoumarins (cou-
marins 7 vs 12), likely as the result of the different
electronic properties of this group. On the other hand,
the presence of a vinylic hydroxyl group in the 1,2-
pyrone ring or the coumarin itself did not display
pro-apoptotic activity in leukemic cells.
This semiquinone form (QÅꢀ) in the presence of oxygen
may thus generate superoxide radicals.31
Q
Åꢀ þ O2
Q þ OÅꢀ
ð3Þ
ƒ!
2
The superoxide radicals can promote various secondary
reactions leading to the generation of ROS and other
radical species as shown by the following equations:
It is well established that polyphenolic compounds have
powerful antioxidant properties, through their ability to
scavenge a wide range of reactive oxygen, nitrogen, and
chlorine species. Antioxidant activity of polyphenols
arises from their high reactivity as hydrogen or electron
donors, from their capacity to chelate transition metal
ions, and from the ability of polyphenolic radicals to
stabilize and delocalize the unpaired electron.28,29 In
general, antioxidant activity depends on the number
and positions of hydroxyl groups and other substituents
on the molecule.30 However, polyphenols can also act as
pro-oxidants.31 Under normal growth conditions phe-
noxyl radicals are usually rapidly changed to non-radi-
cal products by polymerization reactions or by
enzymatic (as well as non-enzymatic) radical reduction.
However, the pro-oxidant activity of phenolic com-
pounds depends on their metal reducing properties, che-
lating behavior, and O2-reducing ability. This latter
property depends on the redox potential of the oxidized
species and the lifetime of the phenoxyl radicals. Phe-
noxyl radicals exhibit cytotoxic pro-oxidant activity
when the lifetime of the radicals is prolonged by reso-
nance-stabilization.18
Figure 3. Experimental (A) and simulated (B) ESR spectra of the
coumarin 8* radical in RPMI 1640 in the presence of U-937 cells.
Experimental ESR spectrum was recorded at 20 ꢁC in an X-band ESR
Spectrometer Bruker ECS 106. The spectrometer settings were:
microwave power 10.1 mW, conversion time 2.56 ms, time constant
2.56 ms, modulation frequency 50 kHz, modulation amplitude
0.107 G, gain 2 · 104, resolution 1024 points, sweep width 25 G,
number of scans 100. Simulated spectrum was obtained by the
calculation of the quantum density matrix using the Liouville equation
with the incorporation of the modulated Zeeman effect. The hyperfine
1
interactions of the six H of the semiquinone form of the coumarin 8*
radical were considered for the simulation of this spectrum.

2670
M. E. Riveiro et al. / Bioorg. Med. Chem. 16 (2008) 2665–2675
Scheme 2. Resonant structures of the phenoxyl radical of coumarin 8* and the formation of its semiquinone form. The simulated ESR spectrum of
the semiquinone form was compatible with the experimental ESR spectrum obtained when coumarin 8 was added to RPMI 1640 and to cultured U-
937 cells in RPMI 1640.
Table 4. Hyperfine coupling constants SEM (n = 5) of the six 1H of the coumarin 8* radical semiquinone form in different mediums at pH 7.6: PBS,
RPMI 1640, and U-937 cells in RPMI 1640
Medium
aH1 (G)
aH2 (G)
aCH3(G)
aH6 (G)
PBS 100 mM
RPMI 1640
U-937 cells in RPMI 1640
3.29 0.06
3.32 0.06
3.32 0.05
1.84 0.04
1.64 0.03
1.63 0.03
0.42 0.02
0.41 0.02
0.42 0.02
0.31 0.01
0.30 0.01
0.32 0.01
OÅ2ꢀ þ OÅ2ꢀ þ 2Hþ
H2O2 þ Menacþtive
O þ H O
ð4Þ
ð5Þ
ð6Þ
reacts with oxygen radicals and carbon centered radicals
forming PBN adducts, that can be detected by ESR
spectroscopy.
ƒ!
2
2
2
HOÅ þ HOꢀ þ Meðnþ1Þþ
ƒ!
active
HOÅ þ RH
H O þ RÅ
ƒ!
2
Figure 4 shows the experimental (A) and simulated (B)
ESR spectra obtained from U-937 cells exposed to cou-
marin 8 in RPMI 1640 and incubated in the presence of
the spin trap PBN. Two signals were observed, one cor-
With the aim to confirm the presence of superoxide and/
or other radicals, PBN spin trap was added to cultured
U-937 cells exposed to coumarin 8 (Figs. 4 and 5). PBN
Figure 4. Experimental (A) and simulated (B) ESR spectra of the coumarin 8* radical and the PBN adduct in U-937 cells cultured in RPMI 1640. The
spin trap PBN (100 mM final concentration) was added to the cell culture and both, the phenoxyl radical and the PBN-adduct, signals were
simultaneously recorded in an X-band ESR Spectrometer Bruker ECS 106 at 20 ꢁC. The spectrometer settings were same as those detailed in the
legend to Figure 3 except that the sweep width was 80 G and the number of scans 400. Simulated spectrum of coumarin 8* radical was obtained as
described in the legend to Figure 3. The inset shows the hyperfine coupling constants SEM (n = 7) of the 14N PBN adducts in different experimental
conditions.

M. E. Riveiro et al. / Bioorg. Med. Chem. 16 (2008) 2665–2675
2671
The higher production of phenoxyl radical and PBN ad-
duct in U-937 cells treated with coumarin 8 appeared
roughly as early as 2 h (Fig. 5). PBN adduct shows a
similar kinetics profile to coumarin 8* radical, being an
indirect verification of ROS production. Figure 6 shows
the kinetic of the coumarin 8* radical in RPMI 1640
medium alone and in U-937 cells in RPMI 1640. Both
kinetics were simulated and fitted as previously de-
scribed. The first order rate constants, corresponding
obs:
form:
obs:
decom:
to the formation ðk
Þ and decomposition ðk
Þ of
the coumarin 8* radical calculated in cultured
U-937 cells in RPMI 1640 and the medium alone were:
¼ 1:2 ꢂ 10ꢀ4 sꢀ1
,
k
¼ 1:3 ꢂ 10ꢀ4 sꢀ1 and
obs:
form:
obs:
decom:
k
k
¼ 4:5 ꢂ 10ꢀ5 sꢀ1 and
k
¼ 5:0 ꢂ 10ꢀ5 sꢀ1
,
obs:
form:
obs:
decom:
Figure 5. Kinetic of the coumarin 8* radical (•) and the PBN adduct
(j) in cultured U-937 cells. Cells growing in exponential phase (106
cells/ml) were incubated with 2 mM coumarin 8 in RPMI 1640 at 37 ꢁC
in the presence of the spin trap PBN (100 mM). The ESR signals of the
coumarin 8* radical and the PBN-adduct were followed as function of
time, using the spectrometer settings detailed in the legend of Figure 4.
The signal intensity of the coumarin 8* radical was measured as the
total height of the third low field peak (h+3) in the first derivative
spectrum, whereas that of the PBN adduct was determined as the total
height of the first low field peak (h+1). As the ESR spectra had
equivalent line-shapes and line-widths at different times, the intensity
of the signal was proportional to the radical concentration. Normal-
ized heights were calculated as the percentage relationship between the
height of the ESR signal at a given time and the maximum height value
of the signal obtained in the respective experience. The kinetic curve of
the coumarin 8* radical was fitted according to Eq. 1. Data correspond
to five independent experiments.
respectively. The difference between both constants,
obs
form
k
and kodbecsom, suggests a distinct kinetic behavior in
the formation and decomposition of the coumarin 8*
radical in different conditions (Fig. 6). Whether this dif-
ferent kinetic behavior is implicated in the pro-apoptotic
effect of coumarin 8 is presently unknown and requires
further studies.
Kim and co-workers17 reported that anti-oxidant treat-
ment completely prevents scopoletin-induced apoptosis
in HL-60 promyelocytic cells, suggesting that ROS gen-
eration may play an important role in the pro-apoptotic
activity of this coumarin. As in the present study we ob-
served that hydroxylated coumarins induced apoptosis
in U-937 promonocytic cells, we next evaluated whether
ROS generation was implicated in the pro-apoptotic
activity of these coumarins. The exposure of 10 mM
N-acetyl-L-cysteine-treated U-937 cells to coumarin 8
for 24 h resulted in the suppression of the ESR signal
of the coumarin 8* radical and PBN-adduct generation
(data not shown). To further evaluate the participation
of ROS in coumarin 8-induced apoptosis, we studied
responding to the described coumarin 8* radical and the
other to the PBN adduct. The ESR signal of the PBN
adduct showed the three characteristic peaks of the
hyperfine interaction with the 14N, being the central
peak superimposed with the coumarin 8* radical signal.
The inset in Figure 4 summarizes the hyperfine coupling
constant values for the 14N of the PBN adduct under
different experimental conditions. The high hyperfine
coupling constant values obtained for the 14N of the
PBN adduct suggest that the radical trapped with
PBN was not oxygen centered.34,35
These findings suggest that radical species were gen-
erated from coumarin 8 in RPMI 1640 medium as
well as in U-937 cells cultured in RPMI 1640.
Therefore a time course study of radical species pro-
duction was carried out. The kinetics for the couma-
rin 8* radical and the PBN adduct in cell culture are
shown in Figure 5. In order to plot the values ob-
tained under different experimental conditions the
heights of the coumarin 8* radical ESR signal at dif-
ferent times were normalized to the maximum height
value obtained in each experimental procedure.
Using a computational program, the kinetics of the
coumarin 8* radical was simulated and fitted by
non lineal regression according to the model pro-
posed in Eq. 7.
Figure 6. Coumarin 8* radical kinetic in RPMI 1640 medium (m,
segmented line) and in U-937 cells cultured in RPMI 1640 (•,
continuous line). Coumarin 8 (2 mM) was added to U-937 cells
growing in exponential phase (106 cells/ml) and to RPMI 1640 alone
followed by incubation at 37 ꢁC. The ESR signal of coumarin 8*
radical was followed as function of time, using the spectrometer
settings detailed in the legend of Fig. 4. Signal intensities and
normalized heights were obtained as described in Fig. 5. The kinetics
curves were fitted according to equation Eq. 1. Data correspond to five
independent experiments.
obs:
k
form:
Coumarin 8
Coumarin 8ꢁ radical
ƒƒƒ!
obs:
decom:
k
ESR silent products
ð7Þ
ƒƒƒ!

2672
M. E. Riveiro et al. / Bioorg. Med. Chem. 16 (2008) 2665–2675
the pro-apoptotic effect of this coumarin in the presence
of the radical scavenger NAC by using two different ap-
proaches. U-937 cells were exposed to 250 lM coumarin
8 for 24 h and NAC (10 mM) was added to the cell cul-
ture either every 8 h for 24 h or just once 24 h before the
end of the assay. The first approach fully inhibited the
appearance of a ladder-like pattern of DNA fragments
and blocked coumarin 8-triggered apoptosis in U-937
cells (Fig. 7B). However, the second treatment did not
abolish DNA fragmentation in coumarin 8-treated cells
(Fig. 7A). In the two experimental treatments NAC ex-
erted a protective role, although the effect was further
evident when cells were repeatedly exposed to the radical
scavenger. The underlying reasons for the difference be-
tween the two treatments remain presently unknown.
Taken together, these findings support that ROS and/
or other radical species play a critical role in the induc-
tion of cell death by coumarin 8. In the presence of O2,
transition metals such as Cu2+ and Fe2+ catalyze the re-
dox cycling of phenolics, leading to the formation of
ROS and other organic radicals that can eventually
damage DNA, lipids, and other biological molecules.18
ROS may trigger mitochondrial membrane polarization,
which involves drastic alterations in cell metabolism and
the release of apoptogenic proteins that initiate caspase-
dependent as well as caspase-independent pathways.36,37
However, the intracellular targets of the phenoxyl radi-
cal or other generated ROS that mediate the induction
of programmed cell death by coumarin 8 in U-937 cells
remain to be elucidated.
cells, the generation of phenoxyl radicals was assessed
by ESR spectroscopy. In the case of coumarin 8, the
phenoxyl radical was detected by ESR spectroscopy.
However compounds 7 and 3 induced the formation of
phenoxyl radicals but they were detected at higher con-
centrations and pH values than coumarin 8. As ex-
pected, no phenoxyl radicals were produced by the
remaining coumarins that did not possess a free phenolic
hydroxyl group (1, 4–6, 10, 11, and 13) (data not
shown). In addition, compound 2 failed to produce phe-
noxyl radicals detectable by ESR spectroscopy, likely as
the result of its particular electronical properties. It has
been previously reported that monohydroxycoumarins,
such as 3, induce apoptosis in leukemic cells through
radical generation.17 In the present study, coumarin 3
showed a weaker biological activity as compared to 8,
since monohydroxycoumarins generate less radical spe-
cies. A similar behavior was observed for mono-
hydroxycoumarins
7
and 9. Furthermore, the
replacement of the 7-OH by an amino group (coumarin
10) suppressed both radical generation and the biologi-
cal activity. In compound 12, the presence of a –CF3
group in C-4 inhibited the formation of the phenoxy
radical thus resulting in the loss of biological activity.
These findings support the existence of a direct relation-
ship between the biological activity and the pro-oxidant
properties of these compounds limited by the number of
radical species stabilized by resonance.
Over the last years coumarins and derivatives were
extensively studied and multiple biological effects were
reported for these compounds. In the present work,
structurally related hydroxycoumarins were studied in
order to elucidate the key structural requirements to in-
duce apoptosis in leukemic cells. We showed that the
dihydroxylated coumarin 8 induced apoptosis through
ROS generation in human leukemic cells. To our
knowledge, this is the first report to show that dihydr-
oxylated coumarins at cytotoxic micromolar concentra-
tion induced apoptosis mediated by ROS in U-937
cells. We have recently demonstrated that dihydroxy-
lated coumarins failed to induce apoptosis in normal
cells as mature human monocytic5 or human umbilical
vein endothelial cells (HUVEC) (unpublished observa-
tions). The present study further suggests that dihydr-
oxylated coumarins are likely promising therapeutic
drugs for cancer treatment due to their pro-apoptotic
activity in cancer cells but minimal toxicity in normal
cells.
In order to gain insight into the radical production of
hydroxycoumarins and related compounds in U-937
3. Materials and methods
3.1. Chemicals
RPMI medium 1640, phosphate-buffered saline (PBS),
N-t-butyl-a-phenylnitrone (PBN), N-acetyl-Lcysteine
(NAC), dimethylsulfoxide (DMSO), and commercial
coumarins were obtained from Sigma Chemical Co.
(St. Louis, MO, USA). Fetal calf serum was purchased
from Natocor (Argentina). All commercial chemicals
and solvents were of reagent grade and used without fur-
ther purification unless otherwise specified.
Figure 7. DNA fragmentation analysis. (A) U-937 cells were pre-
treated once with 10 mM NAC (pH 7.0) at 24 h before the end of the
assay. Cells were also co-incubated with 0.1 mM H202 or 2% DMSO
(positive control). Similar results were obtained in three independent
experiments. (B) U-937 cells were exposed to 250 lM coumarin 8 and
10 mM NAC (pH 7.0) was added to the cell culture every 8 h for 24 h.
NAC (10 mM) treated U-937 cells were used as negative control.
Similar results were obtained in three independent experiments.

M. E. Riveiro et al. / Bioorg. Med. Chem. 16 (2008) 2665–2675
2673
3.2. Chemistry
3.3.4. 7-hydroxy-4-methylcoumarin (7)
Yield: 85%.
Standard workup procedures were used for all chemical
reactions, and solvents were dried over MgSO4 before
evaporation. The general procedure for coumarin syn-
thesis was as follows: 6 ml of concentrated sulfuric acid
was placed in a 100-ml flask on ice bath followed by the
dropwise addition under magnetic stirring of a solution
containing 10 mmol phenol in 1.4 ml (11 mmol) of ethyl
acetoacetate. After one hour the reaction mixture was
poured into crushed ice and the precipitate collected
by suction filtration and then washed with cold water.
The crude coumarin was recrystallized from 95% etha-
nol. Reactions were monitored by thin-layer chromatog-
raphy on 0.25 mm silica gel plates (Merck Kieselgel
60F254) and visualized by UV light. All compounds
were purified by recrystallization from ethanol and char-
acterized by IR and nmR spectra as well as by melting
point determination. 1H RMN and 13C RMN data were
recorded on a Bruker Advance 500 spectrometer. Chem-
ical shifts are reported in parts per million (ppm) and
calibrated to the deuterated solvent reference peaks.
Coupling constants are reported in units of hertz (Hz).
Infrared spectra were recorded on a Perkin-Elmer Spec-
trum One apparatus from KBr disks (resolution:
4 cmꢀ1). Melting points were determined with a Thomas
Hoover Unimelt apparatus and are uncorrected.
Mp 184–185 ꢁC (ethanol); lit.40 185 ꢁC.
1H nmR (DMSO-d6) 2.36 (s, 3H), 6.12 (d, 1H,
J = 1.3 Hz), 6.69 (d, 1H, J = 2.63 Hz), 6.79 (dd, 1H,
J = 8.5, 2.3 Hz), 7.59 (d, 1H, J = 8.5), 10.53 (s, 1H).
IR (m, cm-1) 1600.6, 1678.2, 3159.1.
3.3.5. 7,8-Dihydroxy-4-methylcoumarin (8)
Yield: 80%.
Mp 242–244 ꢁC (ethanol); lit.40 243–244 ꢁC.
1H nmR (DMSO-d6) 2.36 (s, 3H), 6.13 (d, J = 0.9 Hz,
1H), 6.82 (d, J = 8.5 Hz, 1H), 7.10 (d, J = 8.7 Hz, 1H).
IR (m, cmꢀ1) 1583.8, 1600.0, 1620.2, 1648.4, 3412.9.
3.3.6. 6-Hydroxy-4-methylcoumarin (9)
Yield: 40%.
Mp 239–241 ꢁC (ethanol); lit.40 241–242 ꢁC.
1HNMR (DMSO-d6) 2.44 (d, J = 1.2 Hz, 3H), 6.42 (d,
J = 1.2 Hz, 1H), 7.10 (s, 1H), 7.11 (dd, J = 7.0 Hz,
2.8 Hz, 1H), 7.30 (dd, J = 6.6 Hz, 2.7 Hz, 1H), 9.8 (s,
1H).
3.3. Compounds
3.3.1. 4,6-Dimethylcoumarin (6)
Yield: 60%.
IR (m, cm-1) 1690.0, 3336.2.
All synthesized and commercial coumarins (Table 1)
were dissolved in 0.01% (v/v) DMSO and stored at
ꢀ20 ꢁC.
Mp 147–149 ꢁC (ethanol); lit.38 149–150 ꢁC.
1H nmR (DMSO-d6) 2.38 (s, 3H), 2.41 (d,
J = 1.3 Hz,1H), 6.37 (d, J = 1.3 Hz, 1H), 7.28 (d,
J = 8.5 Hz, 1H), 7.43 (dd, J = 8.5 Hz,1.5 Hz, 1H), 7.57
(d, J = 1.3 Hz, 1H).
3.4. Cell culture
The U-937 cell line (American Type Culture Collec-
tion, Rockville, MD) was cultured at 37 ꢁC in a
humidified atmosphere with 5% CO2 in RPMI 1640
medium, supplemented with 10% fetal calf serum
and 50 lg/ml gentamicin. The cell suspension was split
at the third day and diluted 1 day before experimental
procedures.
IR (m, cmꢀ1) 1715.8.
3.3.2. 4,7-Dimethylcoumarin (5)
Yield: 65%.
Mp 128–130 ꢁC (ethanol); lit.39 131 ꢁC (ethanol).
3.5. Measurement of the cytotoxicity concentration 50
(CC50)
1H nmR (DMSO-d6) 2.40 (s, 6H), 6.30 (s, 1H), 7.19 (d,
J = 8.5 Hz, 1H), 7.20 (s, 1H), 7.64 (d, J = 8.0 Hz, 1H).
Cells growing in exponential phase were seeded at 105
cells in 1 ml of RPMI 1640 in a 24-well culture plate
and incubated in a 5% CO2 atmosphere. Cells were ex-
posed to different concentrations of coumarins
(0.15 lM to 2.0 mM) or 0.0001% (v/v) DMSO (control
group). After incubation for 48 h, an aliquot of each
treatment was collected and mixed with an equal volume
of 0.4% Trypan blue and incubated for 5 min after
which the number of viable cells was estimated by a
hemocytometer chamber. The cytotoxic concentration
50 (CC50) values were calculated with the equation for
sigmoidal dose response using Prism 4.00 for Windows
(GraphPad Software, San Diego, CA). Assays were car-
IR (m, cmꢀ1) 1704.3, 1720.7.
3.3.3. 4,8-Dimethylcoumarin (4)
Yield: 20%.
Mp 111–113 ꢁC (ethanol); lit.38 114 ꢁC (ethanol).
1H nmR (DMSO-d6) 2.36 (s, 3H), 2.42 (d, J = 1.3 Hz,
3H), 6.39 (d, J = 1.0 Hz, 1H), 7.26 (t, J = 7.6, 1H),
7.50 (d, J = 6.9, 1H), 7.60 (d, J = 7.9 Hz, 1H).
IR (m, cmꢀ1) 1708.5.

2674
M. E. Riveiro et al. / Bioorg. Med. Chem. 16 (2008) 2665–2675
ried out by triplicate in at least three independent
experiments.
chamber. Ten randomly selected fields were acquired
from each treatment and one hundred cells were mini-
mally counted. Assays were carried out by triplicate in
at least three independent experiments.
3.6. Cell growth inhibition assays (IC50)
Cells growing in exponential phase were seeded at 104 cells
in 150 ll of RPMI 1640 in a 96-well culture plate and incu-
bated under a 5% CO2 atmosphere. U-937 cells were ex-
posed to different concentrations of coumarins ranging
from 0.15 lM to 2.0 mM or to 0.0001% (v/v)DMSO (con-
trol group) followed by incubation with 0.5 lCi of
[3H]Methyl-thymidine (Perkin-Elmer, USA), added 12 h
before the end of the experiment and then harvested in
an automatic cell harvester (Nunc, Maryland, USA).
The incorporation of the radioactive nucleotide was mea-
sured in a Pharmacia Wallac 1410 liquid scintillation
counter and expressed as incorporation percentage with
respect to the control group (cells treated with DMSO).
The inhibitory proliferation concentration 50 (IC50) val-
ues were calculated using the equation for sigmoidal dose
response using Prism 4.00 for Windows (GraphPad Soft-
ware, San Diego, CA). Assays were performed by quadru-
plicate in at least four independent experiments.
In addition, cytospin preparations of cell suspensions
were fixed and stained with May Grunwald–Giemsa
and cell morphology determined by light microscopy
(Zeiss, Germany).
3.9. Electron spin resonance (ESR) experiments
U-937 cells growing in exponential phase (106 cells/ml)
were incubated with 2 mM coumarin 8 in RPMI at
37 ꢁC. The electron spin resonance (ESR) signal of the
phenoxyl radical of coumarin 8 was registered along time.
In another experiment the spin trap N-t-butyl-a-phenyl-
nitrone (PBN) (100 mM) was added to the medium and
both, the phenoxyl radical and the PBN-adduct, signals
were detected with time. In both experiments, aliquots
from the media were drawn at different periods of time
(0–24 h) and the ESR spectra recorded at 20 ꢁC in an X-
band ESR Spectrometer Bruker ECS 106 (Brucker
Instruments Inc., Berlin, Germany). The spectrometer
settings were: microwave power 10.1 mW, conversion
time 2.56 ms, time constant 2.56 ms, modulation fre-
quency 50 kHz, modulation amplitude 0.107 G, gain
2 · 104, resolution 1024 points. Spectra were obtained
using a sweep width of 25 G for the characterization of
coumarin 8* radical signals and a sweep width of 80 G
for the detection of both, coumarin 8* radical and PBN-
adduct, signals simultaneously. The number of scans var-
ied from 100 to 400 depending on the signal intensity.
3.7. Detection of DNA fragmentation by gel
electrophoresis
DNA from untreated or coumarin-treated U-937 cells
was analyzed for endonucleolytic DNA damage, using
horizontal agarose gel electrophoresis. Cells were incu-
bated for 24 h in the presence of coumarins or pretreated
with 10 mM N-acetyl-L-cysteine (pH 7.0). Cell aliquots
(3.106 cells) were collected, washed with PBS, and lysed
in a buffer containing 100 mM NaCl, 4 mM EDTA,
50 mM Tris–HCl, 0.5% SDS, pH 8.0, and proteinase
K (100 mg/ml) at 55 ꢁC for 2 h. DNA was then extracted
and purified as described by Parborell et al.41 Electro-
phoresis was performed for 2.30 h at 40 V in 1.8% aga-
rose gels containing ethidium bromide at a final
concentration of 0.1 mg/ml in TBE buffer (89 mM Tris,
89 mM boric acid, 2 mM EDTA at pH 8.0). DNA band-
ing was visualized with an UV transilluminator.
The signal intensity of the coumarin 8* radical was mea-
sured as the total height of the third low field peak (h+3
)
in the first derivative spectrum whereas that of the PBN
adduct was determined as the total height of the first low
field peak (h+1). As the ESR spectra had equivalent line-
shapes and line-widths at different times, the intensity of
the signal was proportional to the radical concentration.
The coumarin 8* radical concentration and the PBN ad-
duct yields were estimated from the total heights of their
respective peaks. The adduct yield corresponds to the
amount accumulated in a fixed period of time. Simulated
spectra were obtained by calculating the quantum den-
sity matrix using the Liouville equation with the incorpo-
ration of the modulated Zeeman effect.42 The kinetics
were simulated by numeric resolution of the differential
equations using the fifth order Kutta–Merson algo-
rithm.43,44 No linear regression fits were performed using
a modification of the Levenberg Marquart algorithm.45
3.8. Evaluation of cellular morphological changes
Cells growing in exponential phase were seeded at 105
cells in 2 ml of RPMI 1640 in a 24-well culture plate
and incubated under a 5% CO2 atmosphere. Cells were
exposed to 0.15 lM to 2.5 mM coumarins or 0.0001%
(v/v) DMSO (control group) and the cellular morphol-
ogy examined after incubation for 24 or 48 h. Cells
(105) were suspended in 100 ll PBS containing 2 lg/ml
Hoechst 33342 nuclear dye (HO, Sigma, St. Louis,
USA) and 12 lg/ml propidium iodide (PI; Sigma, St.
Louis, USA). After incubation at 37 ꢁC for 8 min, cells
were observed under a fluorescence microscope (Zeiss,
Germany) under UV light at 430 nm (HO) and 630 nm
(PI). Apoptotic cells show a high Hoechst 33342 (HO)
staining but a low propidium iodide (PI) staining, since
they initially tend to exclude PI. Necrotic cells are
brightly stained with PI, whereas healthy cells are dimly
stained by HO and unstained by PI.24 The number of
apoptotic cells was estimated by using a hemocytometer
Acknowledgments
We are sincerely grateful to Dr. L. Bianciotti for critical
reading of the manuscript. This study was supported by
grants from the Universidad de Buenos Aires (Grant
UBACyT B0-50); Consejo Nacional de Investigaciones
´
Cientıficas y Tecnicas (PIP 6110), SECYT (PICT-05-
12164), and Fundacion Bunge Born fellowship ‘Jorge
´
´
Oster’ 2004 to the first author.

M. E. Riveiro et al. / Bioorg. Med. Chem. 16 (2008) 2665–2675
2675
22. Hadi, S. M.; Asad, S. F.; Singh, S.; Ahmad, A. IUBMB
Life 2000, 50, 167–171.
References and notes
23. Wyllie, A. H.; Kerr, J. F.; Currie, A. R. Int. Rev. Cytol.
1980, 68, 251–306.
1. Leszczyniecka, M.; Roberts, T.; Dent, P.; Grant, S.;
Fisher, P. B. Pharmacol. Ther. 2001, 90, 105–156.
2. Kaufmann, S. H. Cancer Res. 1989, 49, 5870–5878.
3. Meng, X. W.; Lee, S. H.; Kaufmann, S. H. Curr. Opin.
Cell Biol. 2006, 18, 668–676.
4. Eugenia Riveiro, M.; Shayo, C.; Monczor, F.; Fernandez,
N.; Baldi, A.; De Kimpe, N.; Rossi, J.; Debenedetti, S.;
Davio, C. Cancer Lett. 2004, 210, 179–188.
5. Riveiro, M. E.; Vazquez, R.; Moglioni, A.; Gomez, N.;
Baldi, A.; Davio, C.; Shayo, C. Biochem Pharmacol. 2007,
doi:10.1016/j.bcp.2007.09.025.
6. Borges, F.; Roleira, F.; Milhazes, N.; Santana, L.; Uriarte,
E. Curr. Med. Chem. 2005, 12, 887–916.
24. Zauli, G.; Vitale, M.; Re, M. C.; Furlini, G.; Zamai, L.;
Falcieri, E.; Gibellini, D.; Visani, G.; Davis, B. R.;
Capitani, S., et al. Blood 1994, 83, 167–175.
25. Martinsson, P.; Liminga, G.; Nygren, P.; Larsson, R.
Anticancer Drugs 2001, 12, 699–705.
26. Ma, J.; Jones, S. H.; Hecht, S. M. J. Nat. Prod. 2004, 67,
1614–1616.
27. Kimura, Y.; Okuda, H.; Arichi, S.; Baba, K.; Kozawa, M.
Biochim. Biophys. Acta 1985, 834, 224–229.
28. Blokhina, O.; Virolainen, E.; Fagerstedt, K. V. Ann. Bot.
(Lond.) 2003, 91 Spec. No, 179–194.
29. Rice-Evans, C. A.; Sampson, J.; Bramley, P. M.; Hollo-
way, D. E. Free Radic. Res. 1997, 26, 381–398.
30. Cai, Y. Z.; Mei, S.; Jie, X.; Luo, Q.; Corke, H. Life Sci.
2006, 78, 2872–2888.
31. Roginsky, V.; Barsukova, T.; Loshadkin, D.; Pliss, E.
Chem. Phys. Lipids 2003, 125, 49–58.
7. Curini, M.; Cravotto, G.; Epifano, F.; Giannone, G. Curr.
Med. Chem. 2006, 13, 199–222.
8. Hoult, J. R.; Paya, M. Gen. Pharmacol. 1996, 27, 713–722.
9. Manolov, I.; Topashka-Ancheva, M.; Klouchek, E. Eksp.
Med. Morfol. 1993, 31, 49–60.
10. Fylaktakidou, K. C.; Hadjipavlou-Litina, D. J.; Litinas,
K. E.; Nicolaides, D. N. Curr. Pharm. Des. 2004, 10,
3813–3833.
32. Simon, H. U.; Haj-Yehia, A.; Levi-Schaffer, F. Apoptosis
2000, 5, 415–418.
33. Roginsky, V.; Lissi, E. Food Chem. 2005, 92, 235–254.
34. Buettner, G. R. Free Radic. Biol. Med. 1987, 3, 259–303.
35. Available from: http://EPR.niehs.nih.gov. In NIEHS/
NIH, 2004.
36. Kroemer, G. Biochem. Biophys. Res. Commun. 2003, 304,
433–435.
11. Kostova, I.; Raleva, S.; Genova, P.; Argirova, R. Bioinor.
Chem. Appl. 2005, 2006, 1–9.
12. Kawase, M.; Sakagami, H.; Motohashi, N.; Hauer, H.;
Chatterjee, S. S.; Spengler, G.; Vigyikanne, A. V.; Molnar,
A.; Molnar, J. In Vivo 2005, 19, 705–711.
13. Egan, D.; O’Kennedy, R.; Moran, E.; Cox, D.; Prosser,
E.; Thornes, R. D. Drug Metab. Rev. 1990, 22, 503–
529.
37. Mates, J. M.; Sanchez-Jimenez, F. M. Int. J. Biochem. Cell
Biol. 2000, 32, 157–170.
38. Chakravarti, D.; Dutta, N. J. Indian Chem. Soc. 1940, 17,
65–71.
14. Finn, G.; Creaven, B.; Egan, D. Eur. J. Pharmacol. 2003,
481, 159–167.
39. Osborne, A. G. Tetrahedron 1981, 37, 2021–2025.
40. Rajithaa, B.; Kumara, V. N.; Someshwara, P.; Madhava,
J. V.; Reddy, P. N.; Reddyb, Y. H. General Papers
ARKIVOC 2006, 12, 23–27.
15. Kolodziej, H.; Kayser, O.; Woerdenbag, H. J.; van Uden,
W.; Pras, N. Z. Naturforsch., C 1997, 52, 240–244.
16. Chu, C. Y.; Tsai, Y. Y.; Wang, C. J.; Lin, W. L.; Tseng, T.
H. Eur. J. Pharmacol. 2001, 416, 25–32.
41. Parborell, F.; Irusta, G.; Vitale, A.; Gonzalez, O.; Pecci,
A.; Tesone, M. Biol. Reprod. 2005, 72, 659–666.
42. Kalin, M.; Gromov, I.; Schweiger, A. J. Magn. Reson.
2003, 160, 166–182.
43. Golub, G. H.; Ortega, J. M. Scientific Computing and
Differential Equations: An Introduction to Numerical
Methods; Academic Press: New York, 1992.
44. Press, W.H.; Teukolsky, S.A.; Verrling, W.T.; Flannery,
B.P. Numerical Recipes in Fortran: The Art of Scientific
Computing. Cambridge. UK, 1992.
17. Kim, E. K.; Kwon, K. B.; Shin, B. C.; Seo, E. A.; Lee, Y.
R.; Kim, J. S.; Park, J. W.; Park, B. H.; Ryu, D. G. Life
Sci. 2005, 77, 824–836.
18. Sakihama, Y.; Cohen, M. F.; Grace, S. C.; Yamasaki, H.
Toxicology 2002, 177, 67–80.
19. Whang, W. K.; Park, H. S.; Ham, I.; Oh, M.; Namkoong,
H.; Kim, H. K.; Hwang, D. W.; Hur, S. Y.; Kim, T. E.;
Park, Y. G.; Kim, J. R.; Kim, J. W. Exp. Mol. Med. 2005,
37, 436–446.
20. Sharma, S. D.; Rajor, H. K.; Chopra, S.; Sharma, R. K.
Biometals 2005, 18, 143–154.
21. Decker, E. A. Nutr. Rev. 1997, 55, 396–398.
45. Budil, D. E.; Lee, S.; Saxena, S.; Freed, J. H. J. Magn.
Reson. A 1996, 120, 155–189.