Redox properties and cytotoxicity of synthetic isomeric mitochondriotropic derivatives of the natural polyphenol quercetin

Article abstract of DOI:10.1002/ejoc.201100573

Mitochondria-targeted redox-active polyphenol derivatives are being developed to affect redox processes in the organelles and as tools either to protect cells from oxidative damage or to precipitate their death. The properties of such compounds may be alt

Full text of DOI:10.1002/ejoc.201100573

FULL PAPER
DOI: 10.1002/ejoc.201100573
Redox Properties and Cytotoxicity of Synthetic Isomeric Mitochondriotropic
Derivatives of the Natural Polyphenol Quercetin
Andrea Mattarei,[a,b] Nicola Sassi,[b,c] Christian Durante,[a] Lucia Biasutto,[b,c]
Giancarlo Sandonà,[a] Ester Marotta,[a] Spiridione Garbisa,[c] Armando Gennaro,[a]
Cristina Paradisi,*[a] and Mario Zoratti[b,c]
Dedicated to Professor Gianfranco Scorrano on the occasion of his 72nd birthday
Keywords: Quercetin / Cyclic voltammetry / DPPH assay / Cytotoxicity / Reactive oxygen species / Triphenylphosphonium
salts
Mitochondria-targeted redox-active polyphenol derivatives
are being developed to affect redox processes in the organ- says, the 7-substituted isomer behaved much like quercetin
elles and as tools either to protect cells from oxidative dam- itself, whereas the 3-substituted isomer was less reactive.
age or to precipitate their death. The properties of such com- Low μ concentrations of either compound increased super-
1,1-diphenyl-2-picrylhydrazyl (DPPH) radical scavenging as-
M
pounds may be altered by substituents introduced to en-
hance mitochondrial delivery. Accordingly, we have charac-
terized the redox behaviour of two isomeric mitochondri-
otropic quercetin-based compounds, 3O- and 7O-[4-tri-
phenylphosphonio)butyl]quercetin iodide. In both cyclic vol-
tammetric determinations of the oxidation potential and in
oxide radical anion formation in cultured cells; both com-
pounds were also cytotoxic. The 7-substituted derivative
proved to be more active in both types of assays, thus emerg-
ing as the isomer of choice for further bioactivity studies and
in efforts to understand the link between redox properties,
pro-oxidant behaviour and cytotoxicity.
Introduction
the possibility of a pro-oxidant behaviour as in the genera-
tion of reactive oxygen species following oxidation of the
polyphenol, often mediated by Fe3+ or Cu2+ ions.[6,13] There
are several literature reports of a pro-oxidant activity for
quercetin.[6,14–19] This behaviour becomes particularly rel-
evant when it gives rise to chain reactions that consume
oxygen and polyphenol and/or cellular antioxidants such as
NAD(P)H or glutathione to produce reactive oxygen spe-
cies (ROS).[20–22] A pro-oxidant activity is thus expected to
be favoured by high polyphenol concentrations, as is indeed
observed.[6,10,23–24] A major issue concerning the efficient in
vivo use of natural polyphenols is their low bioavailability,
since only low nm–μm concentrations are found in plasma
and lymph even after a polyphenol-rich meal. Moreover,
when detected, the polyphenols are predominantly in the
form of conjugates.[25]
To overcome the bioavailability problem and to increase
the effectiveness of polyphenols we are developing synthetic
derivatives according to two strategies. One is based on the
reversible protection of the OH groups to obtain prodrugs
that can act as polyphenol vehicles, more resistant to me-
tabolism during the absorption process and capable of re-
generating the natural compound.[26–28] The second and
more selective approach aims at developing polyphenol an-
alogues capable of concentrating in specific organs and/or
In laboratory experiments, many plant polyphenols exhi-
bit very useful biomedical properties, including prevention
of the onset and inhibition of the growth of several types
of cancer, protection of the cardiovascular system, and the
slowing down of both senescence and progression of neuro-
degenerative diseases.[1–3] These activities are due to the
redox properties of the polyphenols[4] as well as to specific
interactions with signal-transducing proteins and transcrip-
tion factors. Many polyphenols, including quercetin
(3,3Ј,4Ј,5,7-pentahydroxyflavone, Scheme 1) and other fla-
vonoids, are readily oxidized. They can donate an electron
or hydrogen atom and scavenge a range of reactive species
such as hydroxyl radicals, thus acting as antioxidants.[5,6]
These activities often induce a protective effect when cells
are exposed to an oxidative challenge.[7–12] However, the
low oxidation potential of such polyphenols also implies
[a] Department of Chemical Sciences, University of Padova,
via F. Marzolo 1, 35131 Padova, Italy
Fax: +39-049-827-5829
E-mail: cristina.paradisi@unipd.it
[b] CNR Institute of Neuroscience, c/o Department of Biomedical
Sciences, University of Padova
via G. Colombo 3, 35121 Padova, Italy
[c] Department of Biomedical Sciences, University of Padova,
via G. Colombo 3, 35121 Padova, Italy
Eur. J. Org. Chem. 2011, 5577–5586
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5577

C. Paradisi, et al.
FULL PAPER
Scheme 1.
subcellular compartments in which their properties can difference (negative inside) maintained across the plasma
have a major impact. and the inner mitochondrial membranes, these compounds
Specifically, we are developing “mitochondriotrop- accumulate in the cytoplasm and, in particular, within the
ic”[29–31] derivatives of natural polyphenols targeted to the mitochondria. Particularly promising results have been ob-
cytosol and to the mitochondrion, the subcellular compart- tained with MitoQ, a ubiquinone derivative[29,31] and with
ment where they might be most useful in view of their redox mitochondriotropic vitamin E derivatives.[42,43] We have ap-
activity.[32,33]
In most cells, mitochondria are the main site of ROS atives of quercetin (Scheme 1).[32,44]
plied this principle to produce mitochondria-targeted deriv-
production; ROS are partly responsible for aging, neuro-
The rationale of such an approach calls for the delivery
degenerative disorders, and ischemia/reperfusion damage, of compounds with physico-chemical properties and bioac-
and they are an important factor in apoptosis, necrosis, car- tivities similar to those of the parent compound. On the
cinogenesis and mitogenesis.[34,35] Mitochondrial ROS pro- other hand, any modification of the chemical structure im-
duction increases the metastatic potential of cells.[36,37] plies a modification of the properties. For example, querce-
Polyphenols, acting as antioxidants, decrease cell shedding tin conjugates produced in metabolism are much less prone
from cancer spheroids in culture (a model of the metastatic to oxidation than is quercetin.[45,46] This situation may also
process).[38] On the other hand, ROS can induce cell apply to different isomers of a given derivatized quercetin.
death[34] and thus act as anti-cancer agents. Since cancerous In the present case, oxidation of flavonoids with a catechol
cells in many cases have a higher-than-normal basal rate of group on the B ring, like quercetin, is generally considered
ROS production despite their enhanced redox defenses, an to involve the catechol,[47–49] but the rate of the electro-
additional oxidative stress may cause cell death more easily chemical process[50] and the identity of the products eventu-
than in normal cells. Thus, both antioxidant and pro- ally formed[51] depend on the presence of a free OH group
oxidant anti-cancer approaches have been proposed for the at the 3-position. Considering the first oxidation intermedi-
polyphenols.[39–41]
ate, it can be seen (Scheme 2) that for 2, a resonance struc-
Mitochondriotropic derivatives can be obtained by link- ture can be written with a positive charge on C-2, which is
ing a membrane-permeable, positively charged triphenyl- not possible for compound 1.
phosphonio (TPP) moiety to the molecule of interest.[29–31]
This site can therefore be attacked by nucleophiles
Because of their positive charge and the electrical potential (water) with consequent opening of the C ring and forma-
Scheme 2. Resonance structures of oxidized quercetin (Q).
5578
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 5577–5586

Redox Properties and Cytotoxicity of Mitochondriotropic Quercetins
tion of a set of products, which are not accessible if the 3- acetylated derivative are cytotoxic,[32] and proceeded to
OH group is blocked.[52,53] In assays based on the reaction compare Q-3BTPI and Q-7BTPI from this point of view.
with stable radicals in solvents supporting ionization, the Cell death or commitment to it was determined by staining
acidity of the available free OH groups has an important cells with annexin V-FLUOS and propidium iodide and
role in determining the oxidation potential of the com- subsequent fluorescence analysis by FACS.
pound.[49]
A mitochondria-targeting group can, in principle, be at-
tached at any of 10 available positions in quercetin. In en-
visioning future elaborations amenable to reversibility, the
regeneration of the parent compound once inside mito-
chondria, and to avoid destruction by C ring opening, we
initially chose to link the TPP group using a butyl chain
linker to the 3-OH group (Q-3BTPI, Scheme 1).[32,44] As
mentioned, modifications of this group can, however, affect
oxidation potential, reactivity and possibly biological ac-
tivity as well. It is therefore important to assess the extent
to which such a variation might alter the activity relative
to that of quercetin and another regioisomer, carrying the
substituent at position 7, which preserves the 3-OH func-
tionality (Q-7BTPI, Scheme 1).[54] In this paper we report
an improved synthesis for Q-7BTPI, the characterization of
redox properties, by chemical and biochemical assays, of
the two isomeric derivatives Q-3BTPI and Q-7BTPI and a
comparison with quercetin.
Results and Discussion
Synthesis
The mitochondriotropic derivative Q-3BTPI was synthe-
sized as described previously.[32] An improved procedure
was developed for the synthesis of Q-7BTPI, as outlined in
Scheme 3. This new protocol has one less step and gives a
better overall yield than the previously published ap-
proach.[54] Briefly, it involves acetylation of quercetin to af-
ford 3 followed by a one-pot conversion of 3 into 4 by imid-
azole-promoted regioselective alkylation and subsequent
hydrolysis of the remaining ester groups. Q-7BTPI was ob-
tained from 4 in two steps as described in the literature.[54]
Cyclic Voltammetry of Quercetin and Its
Mitochondriotropic Derivatives
The oxidation of phenols can proceed by different
mechanisms, depending on experimental conditions such as
The electrochemical behaviour of mitochondriotropic de-
solvent and pH[55] and often leads to complex sets of prod- rivatives Q-3BTPI and Q-7BTPI and of the parent com-
ucts. At least twenty oxidation products have been iden- pound, quercetin, was investigated to assess the effect of
tified for quercetin.[48] Processes in vivo are likely to involve the chemical modifications introduced. To the best of our
multiple pathways and are difficult to reproduce in vitro. knowledge, this is the first such study involving mitochon-
To compare the redox behaviour of our compounds we driotropic redox-active compounds.
used cyclic voltammetry (CV) and the DPPH reactivity as-
The voltammetric behaviour of quercetin at the glassy
say, one of the most widely used methods to rank molecules carbon electrode is shown in Figure 1a.
on the basis of their ROS scavenging power.[56–58] These
The voltammogram presents three oxidation peaks indi-
two, quite different, experimental approaches represent two cated in the graph as A1, A2 and, barely detectable, A3 at
extremes of possible oxidative reaction conditions. While 0.114, 0.485 and 0.86 V vs. SCE, respectively (Table 1). The
they cannot be considered to fully reflect the complete first voltammetric wave A1 corresponded to the cathodic
range of reactions our compounds may undergo in a bio- counterpart C1, which became clearer if the potential scan
logical setting, coherent results may be taken to provide a was inverted before the second oxidation event A2 took
reliable indication of relative redox reactivity in settings of place (solid line). The first redox event was indicative of a
physiological relevance.
quasi reversible electrochemical process since ΔEp, ex-
For a more biologically relevant readout we compared pressed as the difference between the anodic Ep,a and ca-
the pro-oxidant effects of the compounds as revealed by thodic Ep,c peak potentials, was found to be 80 mV. The
the production of superoxide anion in the cells of a human ΔEp values increased as the scan rate was increased, indicat-
leukemic line. In these experiments we used the fluores- ing that the electrochemical process is dependent both on
cence-activated cell scanner (FACS) method. Furthermore, diffusion and on the rate of electron transfer, which, in this
we built on our recent finding that Q-3BTPI and its per- case, is relatively slow. Peak A2 was irreversible for all scan
Scheme 3. Synthesis of Q-7BTPI. (i) Acetic anhydride (20 equiv.), pyridine, reflux; (ii) Cl(CH2)4Br (3.0 equiv.), Cs2CO3 (1.1 equiv.), imid-
azole (2%), DMF, r.t.; (iii) 3 m HCl/CH3CN (1:2), reflux; (iv) saturated NaI in acetone, reflux; (v) PPh3 (15 equiv.), neat, 90 °C.
Eur. J. Org. Chem. 2011, 5577–5586
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5579

C. Paradisi, et al.
FULL PAPER
standard potential as E° = (Ep,a + Ep,c)/2. The standard
potential value for the first oxidation of quercetin in these
experimental conditions was found to be E° = 0.074 V vs.
SCE, consistent with literature values.[59] The standard po-
tential of quercetin and other flavonoid molecules is often
correlated to their antioxidant activity. Since the latter is of
interest here, we focused on the first oxidation process.
The redox behaviour of Q-7BTPI (Figure 1b) was char-
acterized by three oxidation steps, the first of which (A1)
was quasi-reversible, as shown by the ΔEp value consider-
ably larger than the 60 mV expected for a fully reversible
electron transfer (Table 1). Although the anodic peak po-
tential Ep,a1 is more positive than the corresponding peak
potential of quercetin by 0.029 V, the standard potential E°
(0.072 V vs. SCE) was quite close to that of quercetin
(Table 1). It is worth noting that peak A1 is more stretched
and of lower intensity in the case of Q-7BPTI relative to
quercetin as reflected by the values of Jp, ΔEp1 and ΔEp/2
reported in Table 1. This may be ascribed to the contri-
bution of two different effects: (i) a lower diffusion coeffi-
cient due to a different solvation sphere induced by the
charged phosphonium group, (ii) a slower electron-transfer
rate relative to quercetin. A confirmation of this hypothesis
was provided by the observation that the ΔEp1 values in-
crease with the scan rate much more markedly than in the
case of quercetin suggesting that a slower electron transfer
takes place in Q-7BTPI than in quercetin.
In the case of Q-3BTPI the alkylation of the OH group
caused a positive shift of the first oxidation peak and the
loss of the second peak in the voltammetric pattern. This is
in agreement with the notion that the 3-OH group is critical
to the first electrochemical process and is supported by res-
onance structures, which stabilize the electro-generated in-
termediates.
Furthermore, the first peak A1 was characterized by a
lower chemical reversibility (the ratio between the cathodic
and anodic peak currents is Ͻ 1) even when the scan rate
was reversed just after the A1 peak potential (Figure 1c, so-
lid line). Furthermore, the cathodic peak disappeared at a
scan rate less than 0.01 Vs–1 indicating that chemical revers-
ibility was completely lost (Figure 2).
Figure 1. Cyclic voltammograms of 1 mm: (a) Quercetin, (b) Q-
7BTPI, (c) Q-3BTPI, in 9:1 DMF/TrisHCl buffer solution, 0.1 m
Bu4NBF4 as supporting electrolyte, and with a glassy carbon work-
ing electrode at v = 0.2 Vs–1.
rates tested, ranging from 0.01 to 20 V/s, and so was peak
A3, which appeared as a shoulder of the solvent discharge.
On the cathodic scan (not shown) a quasi-reversible re-
duction process took place at a cathodic potential peak of
–1.685 V vs. SCE. Although the electrochemical oxidation
of quercetin has been extensively investigated using various
solvents and supporting electrolytes,[48,50,51] its oxidation
mechanism is not yet been completely understood. How-
ever, it is commonly accepted that the first oxidation pro-
cess A1 is the reversible bielectronic oxidation of the cat-
echolic moiety of quercetin to yield the corresponding quin-
onic derivative. The subsequent peak A2 has been assigned
However, upon increasing the scan rate the anodic cur-
to the oxidation of the OH group in position 3 of the rent peak decreased, whereas the cathodic counterpart was
C ring, whereas peak A3 is probably due to the oxidation found to increase (Figure 2). This behaviour, associated
of an OH group on the A ring. On the cathodic scan, the with the increase of ΔEp/2 with the scan rate, may be
peak at –1.685 V vs. SCE may be ascribed to the reduction ascribed to an electron transfer/chemical reaction/electron
of the carbonyl group on the C ring.
transfer (ECE) mechanism, involving initial electron trans-
When the electrochemical process is reversible or quasi- fer, a subsequent chemical reaction, and then a second elec-
reversible, cyclic voltammetry allows the estimation of the tron transfer. We envisioned that the first electron transfer
Table 1. Electrochemical data for quercetin and derivatives.
[a]
[a]
[a]
[a]
[a]
[c]
[a]
[a,d]
Ep,a1
Ep,a2
Ep,a3
Ep,c1
ΔEp1
Eo[a,b]
Jp,a1
Ep/2
ΔEp/2
Quercetin
Q-7BTPI
Q-3BTPI
0.114
0.143
0.239
0.485
0.533
–
ca. 0.86
ca. 0.80
0.897
0.034
0.001
0.146
0.080
0.140
0.093
0.074
0.072
0.166
0.467
0.226
0.198
0.065
0.062
0.175
0.049
0.081
0.065
[a] All potentials are expressed in V vs. SCE. [b] The standard potential is determined by Eo = (Ep,a + Ep,c)/2. [c] The peak current density
is expressed in mAcm–2. [d] The voltammetric peak width is expressed as ΔEp/2 = Ep,a – Ep/2 where Ep/2 is the half peak potential.
5580
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 5577–5586

Redox Properties and Cytotoxicity of Mitochondriotropic Quercetins
Figure 3 shows the results of three representative experi-
ments in which DPPH was titrated with quercetin, Q-
7BTPI or Q-3BTPI. The results are displayed as plots of
[(Acontr – Asam)/(Acontr – Ablank)]ϫ100, a quantity which is
taken to represent the percent fraction of DPPH quenched,
as a function of the polyphenol concentration (see the Ex-
perimental Section for details). The steepness of the
quenching curve is directly related to the oxidizing power
of the polyphenol. From the titration curve values for EC50
and n were derived, which represent the concentration of
polyphenol necessary to quench the DPPH absorbance by
50% and the reaction coefficient, respectively; EC50 and n
values derived from our experiments with quercetin, Q-
3BTPI and Q-7BTPI are reported in the inset of Figure 3.
Figure 3 clearly shows that, although quercetin and Q-
7BTPI produce very similar titration curves and reaction
coefficients, Q-3BPTI is less efficient than either of these at
quenching DPPH. The presence of a free OH moiety at
position 3 in the flavonoid is thus important for this reac-
tion as well as for the electrochemical oxidation. Con-
versely, the OH group at position 7 within Q-3BPTI seems
to exert little or no influence. This behaviour, which is con-
sistent with observations made by other groups,[63–67] can
be rationalised considering that the sequential proton loss
electron transfer (SPLET) reaction probably taking place at
the catechol moiety[49] forms a delocalized radical, which –
due to the free OH group in 3 – can lead to the formation
of an o-quinone group in the C ring upon further oxidation
(Figure 1).
Figure 2. Cyclic voltammograms of Q-3BTPI (1 mm Q-3BTPI in
9:1 DMF/TrisHCl buffer solution, 0.1 m Bu4NBF4 as supporting
electrolyte), using a glassy carbon working electrode at scan rates
ranging from 0.01 to 2 Vs–1.
leads to the formation of a radical cation, capable of under-
going a chemical reaction. The chemical reaction is proba-
bly deprotonation of position 4Ј in the electro-generated
radical cation, which ultimately produces a semiquinone
species, that at the same potential can be further oxidized
to the quinone in a net 2-electron process. In the case of Q-
3BTPI the deprotonation reaction is assumed to be slower
due to the lack of the OH group in position 3. As a result,
at sufficiently high scan rates, the deprotonation (chemical
step) may be prevented so that the radical cation produced
upon initial electron transfer oxidation may be reduced at
the electrode before it has a chance to undergo any chemical
reaction.
In this case, the evaluation of the standard potential as
the average of oxidation and reduction peak potentials of
A1 is not straightforward, because at low scan rates the
peak potentials are influenced by the chemical reaction of
the ECE mechanism, while at high scan rates they are influ-
enced by slow electron transfer kinetics that causes the
“stretching” of the peaks. However, the standard potentials
can be roughly estimated at scan rates higher than 1 Vs–1,
while the peak is still fully chemically reversible (E° =
0.166 V vs. SCE). This value is more positive by about
90 mV with respect to both quercetin and Q-7BTPI, re-
flecting a lower oxidation potential of Q-3BTPI.
We next assessed the radical scavenging properties of
quercetin and related derivatives by the DPPH method.
DPPH is a stable free radical with an absorbance maximum
at 520 nm (purple). Its conversion to DPPHH (yellow) is
used as a readout of radical scavenging by an added antiox-
idant such as quercetin[60–62] [Equation (1)].
Figure 3. Reaction of DPPH with quercetin, Q-3BTPI and Q-
7BTPI monitored spectrophotometrically at 520 nm. Data points
are the average of triplicate measurements Ϯ s.d. Inset: EC50 and
reaction coefficient values (averages Ϯ s.d.; N = 3 for each com-
pound).
C ring opening products may then give further reactions
with DPPH, thus accounting for the different stoichiome-
tries observed with quercetin and Q-7BTPI on one hand
and with Q-3BTPI on the other. Further reaction of an ini-
tially formed product to give a higher overall stoichiometry
has been reported, for instance in the reaction of chrysin
with the ABTS [2,2Ј-azinobis(3-ethylbenzothiazoline-6-
sulfonic acid)] radical in the Trolox equivalent antioxidant
(1) capacity (TEAC) assay.[68]
Eur. J. Org. Chem. 2011, 5577–5586
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5581

C. Paradisi, et al.
FULL PAPER
In conclusion, both direct electrochemical oxidation ex- pounds were supplied at the highest concentration used in
periments and radical quenching assays have revealed that cytotoxicity experiments (5 μm).[32] Cyclosporin A (CsA),
Q-7BTPI behaves very similarly to the parent compound the paradigmatic mitochondrial permeability transition
quercetin. In contrast, the regioisomer of Q-7BTPI, Q- (MPT) inhibitor, was included to exclude possible effects
3BTPI is characterized by its higher electrochemical oxi- due to MPT onset. However, we verified that the absence of
dation potential and a lower reactivity with DPPH. Based CsA, or substituting Cyclosporin H – which inhibits MDR
on these findings, Q-7BTPI is preferred over its isomer Q- pumps but not the MPT – for CsA did not have a detectable
3BTPI for the development of mitochondriotropic anti- or effect on the MitoSOX® response to mitochondriotropic
pro-oxidant prodrugs. Conversely, Q-7BTPI may be ex- quercetin derivatives, thus making it unlikely that the MPT
pected to be more prone to oxidative degradation and is upstream of ROS production (not shown).
C ring opening in vivo (Scheme 2) than is Q-3BTPI.
The results are exemplified by a representative experi-
ment presented in Figure 4. In flow cytometry experiments
the light forward- and side-scattered by a cell carries infor-
mation on the size and the internal structure of the cell,
ROS Production
To verify whether the differences between the two iso- respectively. Forward (FSC) and side (SSC) scatter param-
meric compounds would be maintained in a cellular system eters of a healthy cell fall within a certain range, which de-
we monitored superoxide production in cells exposed to Q- pends on the cell type. FSC and SSC change when the cell
3BTPI and Q-7BTPI. Superoxide production was detected becomes apoptotic or necrotic. In our experiments popula-
as fluorescence changes of the indicator MitoSOX® in tions of cells were analyzed that displayed forward- and
FACS experiments with Jurkat lymphocytes. The com- side-scatter parameters characteristic of healthy cells (pop-
Figure 4. Mitochondriotropic quercetin derivatives elicit production of superoxide anion in Jurkat cells; representative experiment. (A)
The scatter parameters of Jurkat cells are not significantly altered by 5 μm Q-3BTPI or Q-7BTPI over a 2 h period. SSC vs. FSC
plots obtained by FACS from suspensions of cells treated with the indicated compounds for 0 or 120 min, as indicated. TPMP is
triphenylmethylphosphonium chloride, used as control also in addition to quercetin. (B) Fluorescence distribution histograms. Cells were
exposed to 5 μm Q-7BTPI in HBSS for the indicated times. 10000 cells were counted in each case, and the fluorescence emitted by the
selected population (circled and in red in panel (A) was plotted. (C) MitoSOX® fluorescence response. Plots of the medians of cell
fluorescence distribution histograms analogous to those shown in panel (B), for 5 μm of the indicated compounds. The median values
plotted for Q-7BTPI are those of the histograms shown in panel (B).
5582
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 5577–5586

Redox Properties and Cytotoxicity of Mitochondriotropic Quercetins
ulations in red, enclosed by a polygon, in Figure 4a), i.e. and to 49.7Ϯ10.2% for treatment with 5 μm Q-7BTPI (N
cells not exhibiting morphological changes associated with = 9). Tukey HSD post-hoc analysis showed a significant
advanced necrosis or apoptosis. Thus, the effects observed difference in the Q-3BTPI vs. control (p = 0.0017) and Q-
are not consequences of the death process, but rather pre- 7BTPI vs. control (p = 0.00017) comparison as well as be-
cede or accompany it, and can be attributed directly to the tween Q-3BTPI and Q-7BTPI treatments (p = 0.0002).
compounds tested. MitoSOX® fluorescence increased sig-
nificantly with both Q-3BTPI and Q-7BTPI, but was mark-
edly higher with Q-7BTPI. Figure 4b shows the fluores-
cence distribution histograms obtained upon cell exposure
to 5 μm Q-7BTPI. Figure 4c plots the median values of his-
tograms of this type for the two quercetin derivatives and
controls. Similar results were obtained in 5 other analogous
experiments. Thus, superoxide generation appeared to re-
flect the redox properties of the compounds; Q-3BTPI and
Q-7BTPI both elicited a strong response, with the response
to Q-7BTPI clearly surpassing that of Q-3BTPI.
Figure 5. Q-7BTPI is more effective than Q-3BTPI at inducing an-
nexin/PI labelling of Jurkat cells. Dots in the lower left quadrant
Superoxide production seemed to be specifically associ-
ated with mitochondriotropic compounds, since quercetin
or the combination of quercetin and a phosphonium salt
did not produce a similar effect. The redox properties and
reactivity of Q-7BTPI, however, were found to be very sim-
ilar to those of quercetin, whereas Q-3BTPI was more re-
sistant to oxidation. An explanation for this behaviour must
therefore consider the environmental conditions under
which oxidation takes place. In other words, the abundant
production of radicals must have to do with the accumu-
lation of Q-7BTPI or Q-3BTPI in mitochondria. The con-
centration of oxidizable species can, itself, be a factor in
determining pro-oxidant behaviour.[69] Mitochondria main-
tain a slightly alkaline pH, which is expected to favour oxi-
dation. They also are rich in copper and iron, which can
serve as mediators in electron transfer processes from the
polyphenolic species to O2. Generally speaking, mito-
chondria are the seat of intense enzymatic redox activity,
which may entrain polyphenols.
represent healthy cells, those in the lower right quadrant cells lab-
elled by annexin V-FLUOS, those in the upper right quadrant cells
labelled by both annexin and propidium iodide, and those in the
upper left quadrant were labeled only by PI. One representative
experiment out of nine performed is shown. The suspension me-
dium was HBSS. 5000 cells were counted for each condition. See
the Experimental Section for details.
Conclusions
Mitochondriotropic quercetin analogs Q-3BTPI and Q-
7BTPI consistently display a similar relationship to each
other no matter what aspect is being examined; the latter is
more easily oxidized, has a higher stoichiometry of reaction
with DPPH, is a more effective inducer of superoxide pro-
duction in cells, and is more cytotoxic. Q-7BTPI thus ap-
pears to be the regioisomer of choice for further work
aimed at (i) understanding the factors underlying cytotoxi-
city of such mitochondriotropic quercetin analogs, and (ii)
exploiting these compounds pharmacologically.
Cytotoxicity of Mitochondriotropic Quercetin Derivatives
Experimental Section
To investigate the possible correlation between the ob-
served oxidative behaviour and cytotoxicity we compared
the effects of Q-3BTPI and Q-7BTPI on cultured Jurkat
cells. Cell death was assessed by staining cells with annexin
V-FLUOS and propidium iodide (PI) and fluorescence
analysis by flow cytometry. Annexin recognizes phosphatid-
ylserine and labels cells, which expose this phospholipid in
the outer leaflet of the plasma membrane – a relatively early
sign of apoptotic death – and those, which have lost plasma
membrane integrity, as happens in advanced apoptosis or
necrosis. Propidium iodide labels only the latter. In nine re-
peated experiments (a typical one is shown in Figure 5) the
fraction of cells positive for annexin (x axis in Figure 5), PI
(y axis in Figure 5) or both was always lowest for the con-
trol sample and higher in the case of Q-7BTPI than in that
of Q-3BTPI. After 3 h of incubation (see the Experimental
Section), the average percentage of cells labelled by annexin,
PI or both annexin and PI increased from 20.9Ϯ2.4% in
Materials and Instrumentation: Quercetin and other commercial
chemicals were reagent grade products purchased from Sigma–Ald-
rich, J. T. Baker and Merck. They were used as received. TLCs were
run on silica gel supported on plastic (Macherey–Nagel Polygram®
SIL G/UV254, silica thickness 0.2 mm) and visualized by UV de-
tection. Flash chromatography was performed on silica gel [Mach-
erey–Nagel 60, 230–400 mesh granulometry (0.063–0.040 mm)] un-
der air pressure. The solvents were analytical or synthetic grade
and were used without further purification. 1H and 13C NMR spec-
tra were recorded with a Bruker AC 250F spectrometer operating
at 250 MHz for 1H NMR and 62.9 MHz for 13C NMR spec-
troscopy. Mass spectra were performed with an MSD SL Trap mass
spectrometer (G2445D SL) with ion trap detector and ESI source.
Synthesis: Q-3BTPI was prepared as previously reported.[32] Q-
7BTPI was synthesized as outlined in Scheme 3, which has one
step less and gives a better overall yield (35%) than the previously
published procedure.[54]
2-(3,4-Diacetoxyphenyl)-4-oxo-4H-chromene-3,5,7-triyl Triacetate
controls to 32.0Ϯ6.8% for treatment with 5 μm Q-3BTPI (3): Compound 3 was synthesized from quercetin according to a
Eur. J. Org. Chem. 2011, 5577–5586
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5583

C. Paradisi, et al.
FULL PAPER
published procedure,[54] except for the purification, which was per-
formed by recrystallization from EtOH. Physical and spectroscopic
data for 3 produced in this work were identical to those previously
reported.[54] Yield: 80%.
the desired concentration (at least seven per determination, within
the range 0–200 μm). After incubation of the samples at room temp.
in the dark for 40 min, absorbance was measured using a Tecan
Infinite 200 plate reader (λ = 520 nm, 10 nm slit). According to
literature procedures,[56–58] the percent fraction of DPPH quenched
during the 40 min incubation was approximated by the algorithm
[(Acontr – Asam)/(Acontr – Ablank)]ϫ100, where Acontr, Asam and
Ablank corresponded to the absorbances of the control solution
(only DPPH), the polyphenol plus DPPH solution, and the blank
solution (solvent system), respectively. According to the general
practice, the results of these assays are presented in terms of the
parameters EC50 and n. EC50 is defined as the polyphenol concen-
tration at which the quantity [(Acontr – Asam)/(Acontr – Ablank)]ϫ100
reaches a value of 50. This point was determined using the linear
7-(4-Chlorobutoxy)-2-(3,4-dihydroxyphenyl)-3,5-dihydroxy-4H-
chromen-4-one (4): Cesium carbonate (1.43 g, 4.4 mmol, 1.5 equiv.)
and 1-bromo-4-chlorobutane (1.51 g, 8.8 mmol, 3.0 equiv.) were
added to a solution of 3 (1.50 g, 2.9 mmol, 1.0 equiv.) in DMF
(30 mL). Imidazole (0.03 g, 2% w/w) was then added and the mix-
ture stirred at room temp. and under N2 for 48 h. After addition
of EtOAc (200 mL), the reaction mixture was washed with 3 m aq.
HCl (5ϫ 100 mL). The organic layer was dried with MgSO4 and
the solvent evaporated under reduced pressure. Without further pu-
rification, the crude product was added to a mixture of acetonitrile
(60 mL) and 3 m aq. HCl (30 mL). The resulting solution was main-
tained at reflux with stirring for 1 h, and then EtOAc (200 mL) and
water (200 mL) were added. The organic layer was washed with 3 m
aq. HCl (3ϫ 200 mL), dried with MgSO4, and solids removed by
filtration. Removal of the solvent under reduced pressure gave a
crude mixture, which was purified by silica gel flash chromatog-
raphy (toluene/MeOH, 4:1) to obtain 4 as a bright yellow solid
(0.77 g, 67% yield). Physical and spectroscopic data matched those
reported in the literature.[54]
best fit of the initial, linear part of the plot [(Acontr – Asam)/(Acontr
blank)]ϫ100 vs. [polyphenol]. The coefficient n is equal to the ra-
tio (mol of DPPH reacted)/(mol of polyphenol added) at EC50.
–
A
Cells: Jurkat T lymphocytes, with a doubling time of 24 h, were
grown in RPMI-1640 medium supplemented with 10 mm HEPES
buffer (pH = 7.4), 10% (v/v) fetal calf serum (Invitrogen), 100 U/
mL penicillin G, 0.1 mg/mL streptomycin, 2 mm glutamine
(GIBCO) and 1% nonessential amino acids (100X solution;
GIBCO).
ROS Production Assays: Superoxide generation in cells was as-
sessed in FACS experiments using a Beckton Dickinson Canto II
flow cytometer and the mitochondriotropic superoxide anion-spe-
cific probe MitoSOX® Red (Invitrogen/Molecular Probes). Mito-
SOX® Red is a derivative of dihydroethidium comprising a tri-
phenylphosphonium group, which determines its accumulation in
energized mitochondria. It reacts with superoxide producing a fluo-
rescent compound. Jurkat cells were washed and resuspended in
Hank’s buffered salt solution (HBSS; in mm units: NaCl 136.9, KCl
5.36, CaCl2 1.26, MgSO4 0.81, KH2PO4 0.44, Na2HPO4 0.34, glu-
cose 5.55, pH = 7.4 with NaOH) at a density of 1.5ϫ106 cells/mL
and loaded with 1 μm MitoSOX® (37 °C, 20 min). CsA (5 μm) was
generally included as an inhibitor of multiple drug resistance
(MDR) pumps, to obtain a higher accumulation ratio for the dye,
to limit the rate of efflux during the experiment and to prevent the
possible onset of the mitochondrial permeability transition (MPT).
After loading, cells were diluted 1:5 in HBSS (plus CsA) and di-
vided into identical aliquots. At time 0 the compound was added,
and data collected after the desired incubation times, by exciting at
488 nm and by measuring fluorescence in the 542–585 nm range.
Data were analysed using BD VISTA software.
7-(4-Triphenylphosphoniobutoxy)-2-(3,4-dihydroxyphenyl)-3,5-di-
hydroxy-4H-chromen-4-one Iodide (Q-7BTPI): This compound was
synthesized from 4 in two steps according to a published pro-
cedure.[54] Physical and spectroscopic data matched those reported
in the literature.[54]
Electrochemistry: All experiments were performed in a DMF/Tris-
HCl, 9:1 (pH = 7.5) buffer solution and using tetrabutylammonium
tetrafluoroborate (Bu4NBF4, Fluka, Ͼ 98%) as the supporting
electrolyte. DMF (Acros Organics, 99%) was treated with anhy-
drous Na2CO3 and doubly distilled at reduced pressure under N2.
Bu4NBF4 was recrystallized from EtOH/water, 2:1 and dried at
70 °C under vacuum. Electrochemical measurements were per-
formed with an EG&G Princeton Applied Research Model 173A
potentiostat connected to a computer by a DAQ National Instru-
ments USB-9215A card for data monitoring and acquisition. Cyclic
voltammetry experiments were carried out in a three-electrode cell
with a glassy carbon disc as working electrode. The counter elec-
trode and the reference electrode were a Pt wire and Ag|AgI|0.1 m
nBu4NI in DMF, respectively. The latter was calibrated after each
experiment against the ferrocinium/ferrocene couple, which in
DMF/Tris-HCl has an E° value of 0.432 V vs. SCE. The potentials
measured against the Ag|AgI|I– reference electrode were converted
into the SCE scale, to which all potentials in the paper are referred.
The working electrodes were built from a 3 mm diameter glassy
carbon (GC) rod (Tokai GC-20) and were cleaned with diamond
paste (0.25 μm) and sonicated in EtOH for 5 min prior to each
experiment. Furthermore, the GC electrode was electrochemically
activated by performing 20 cycles (from –0.6 to 1.8 V vs. SCE at
0.1 Vs–1) in an Na2SO4 0.2 m solution. All experiments were carried
out at 25 °C, under Ar, and with a 1 mm solution of the compound
of interest.
Annexin and Propidium Iodide (PI) Labelling Assays: Cell death or
commitment to it was determined by staining cells with annexin V-
FLUOS (Roche) and propidium iodide (PI) (Sigma) and fluores-
cence analysis by flow cytometry. Cells were washed and resus-
pended at 3ϫ105 cells/mL in HBSS. Aliquots were incubated for
3 h with the desired compounds at 37 °C, 5% CO2. DMSO concen-
tration was 0.2% in all cases. A 200 μL portion of each incubation
was then placed in a test tube, and propidium iodide (final concen-
tration 1 μg/mL) and annexin-V-FLUOS (Roche; cat. no.
11828681001) (1 μL/sample) were added. FACS analysis was car-
ried out after a further 20 min labelling period at 37 °C in the dark.
Data were processed by quadrant statistics using BD VISTA soft-
ware.
1,1-Diphenyl-2-picrylhydrazyl (DPPH) Radical Scavenging Assays:
The assays were conducted using 96-well plates (Greiner Bio-One)
according to the approach described by Fukumoto and Mazza.[60]
Measurements were performed in triplicate (three wells for each
concentration tested). Each well contained 200 μL of a 150 μm
DPPH (i.e. 30 nmol) solution in MeOH/Tris-HCl, 5:1 (pH = 7.4).
At time 0 a 20 μL aliquot of an appropriate mother solution of the
polyphenol in DMSO was added to each well and mixed to produce
Acknowledgments
We thank I. Szabò and A. Angrilli for useful discussions. This work
was supported by grants from the Fondazione Cassa di Risparmio
5584
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 5577–5586

Redox Properties and Cytotoxicity of Mitochondriotropic Quercetins
[31]
[32]
R. A. Smith, M. P. Murphy, Johns Hopkins Med. Discov. Med.
2011, 11, 106–114.
A. Mattarei, L. Biasutto, E. Marotta, U. De Marchi, N. Sassi,
S. Garbisa, M. Zoratti, C. Paradisi, ChemBioChem 2008, 9,
2633–2642.
L. Biasutto, A. Mattarei, E. Marotta, A. Bradaschia, N. Sassi,
S. Garbisa, M. Zoratti, C. Paradisi, Bioorg. Med. Chem. Lett.
2008, 18, 5594–5597.
di Padova e Rovigo (Fondazione CARIPARO) (Excellence grant
“Developing a pharmacology of polyphenols”), from the Univer-
sity of Padova (PRAT CPDA081128/08) and the Italian Associa-
tion for Cancer Research (AIRC) (to S. G.). L. B. was the benefi-
ciary of a University of Padova post-doctoral fellowship.
[33]
[1] S. Quideau, D. Deffieux, C. Douat-Casassus, L. Pouységu, An-
gew. Chem. Int. Ed. 2011, 50, 586–621.
[2] C. G. Fraga, M. Galleano, S. V. Verstraeten, P. I. Oteiza, Mol.
Aspects Med. 2010, 31, 435–445.
[3] N. A. Kelsey, H. M. Wilkins, D. A. Linseman, Molecules 2010,
15, 7792–7814.
[4] M. C. Foti, J. Pharm. Pharmacol. 2007, 59, 1673–1685.
[5] Q. Cai, R. O. Rahn, R. Zhang, Cancer Lett. 1997, 119, 99–107.
[6] G. Cao, E. Sofic, R. L. Prior, Free Radical Biol. Med. 1997, 22,
749–760.
[7] D. Kook, A. H. Wolf, A. L. Yu, A. S. Neubauer, S. G. Pri-
glinger, A. Kampik, U. C. Welge-Lüssen, Invest. Ophthalmol.
Vis. Sci. 2008, 49, 1712–1720.
[8] U. Kaindl, I. Eyberg, N. Rohr-Udilova, C. Heinzle, B. Marian,
Food Chem. Toxicol. 2008, 46, 1320–1326.
[34]
[35]
[36]
L. Biasutto, I. Szabò, M. Zoratti, Antiox Redox Signal, in
press.
S. J. Ralph, S. Rodríguez-Enríquez, J. Neuzil, E. Saavedra, R.
Moreno-Sánchez, Mol. Aspects Med. 2010, 31, 145–170.
K. Ishikawa, K. Takenaga, M. Akimoto, N. Koshikawa, A.
Yamaguchi, H. Imanishi, K. Nakada, Y. Honma, J. Hayashi,
Science 2008, 320, 661–664.
G. Pani, T. Galeotti, P. Chiarugi, Cancer Metastasis Rev. 2010,
29, 351–378.
S. Günther, C. Ruhe, M. G. Derikito, G. Böse, H. Sauer, M.
Wartenberg, Cancer Lett. 2007, 250, 25–35.
D. Trachootham, J. Alexandre, P. Huang, Nat. Rev. Drug Dis-
[37]
[38]
[39]
covery 2009, 8, 579–591.
[40]
[41]
[42]
J. Wang, J. Yi, Cancer Biol. Ther. 2008, 7, 1875–1884.
S. J. Ralph, J. Neuzil, Mol. Nutr. Food Res. 2009, 53, 9–28.
L. F. Dong, V. J. Jameson, D. Tilly, J. Cerny, E. Mahdavian, A.
Marín-Hernández, L. Hernández-Esquivel, S. Rodríguez-
Enríquez, J. Stursa, P. K. Witting, B. Stantic, J. Rohlena, J.
Truksa, K. Kluckova, J. C. Dyason, M. Ledvina, B. A. Salva-
tore, R. Moreno-Sánchez, M. J. Coster, S. J. Ralph, R. A.
Smith, J. Neuzil, J. Biol. Chem. 2011, 286, 3717–3728.
L. F. Dong, V. J. Jameson, D. Tilly, L. Prochazka, J. Rohlena,
K. Valis, J. Truksa, R. Zobalova, E. Mahdavian, K. Kluckova,
M. Stantic, J. Stursa, R. Freeman, P. K. Witting, E. Norberg,
J. Goodwin, B. A. Salvatore, J. Novotna, J. Turanek, M. Led-
vina, P. Hozak, B. Zhivotovsky, M. J. Coster, S. J. Ralph, R. A.
Smith, J. Neuzil, J. Free Radical Biol. Med. 2011, 50, 1546–
1555.
L. Biasutto, N. Sassi, A. Mattarei, E. Marotta, P. Cattelan, A.
Toninello, S. Garbisa, M. Zoratti, C. Paradisi, Biochim. Bio-
phys. Acta 2010, 1797, 189–196.
F. Lodi, R. Jiménez, C. Menendez, P. W. Needs, J. Duarte, F.
Perez-Vizcaino, Planta Med. 2008, 74, 741–746.
J. Terao, Forum Nutr. 2009, 61, 87–94.
V. Krishnamachari, L. H. Levine, P. W. Paré, J. Agric. Food
Chem. 2002, 50, 4357–4363.
A. Zhou, O. A. Sadik, J. Agric. Food Chem. 2008, 56, 12081–
12091.
M. Musialik, R. Kuzmicz, T. S. Pawlowski, G. Litwinienko, J.
Org. Chem. 2009, 74, 2699–2709.
H. P. Hendrickson, A. D. Kaufman, C. E. Lunte, J. Pharm. Bi-
omed. Anal. 1994, 12, 325–334.
L. V. Jørgensen, C. Cornett, U. Justesen, L. H. Skibsted, L. O.
Dragsted, Free Radical Res. 1998, 29, 339–350.
G. Jungbluth, I. Rühling, W. Ternes, J. Chem. Soc. Perkin
Trans. 2 2000, 2, 1946–1952.
[9] L. C. Wilms, J. C. Kleinjans, E. J. Moonen, J. J. Briedé, Toxicol.
in Vitro 2008, 22, 301–307.
[10] G. N. Kim, H. D. Jang, Ann. N. Y. Acad. Sci. 2009, 1171, 530–
537.
[11] K. Min, S. E. Ebeler, Food Chem. Toxicol. 2009, 47, 2716–2722.
[12] R. Mikstacka, A. M. Rimando, E. Ignatowicz, Plant Foods
Hum. Nutr. 2010, 65, 57–63.
[43]
[13] B. Halliwell, Arch. Biochem. Biophys. 2008, 476, 107–112.
[14] M. J. Laughton, B. Halliwell, P. J. Evans, J. R. Hoult, Biochem.
Pharmacol. 1989, 38, 2859–2865.
[15] M. Kessler, G. Ubeaud, L. Jung, J. Pharm. Pharmacol. 2003,
55, 131–142.
[16] P. Filipe, J. Haigle, J. N. Silva, J. Freitas, A. Fernandes, J. C.
Mazière, C. Mazière, R. Santus, P. Morlière, Eur. J. Biochem.
2004, 271, 1991–1999.
[17] Y. Y. Lee-Hilz, A. M. Boerboom, A. H. Westphal, W. J. Berkel,
J. M. Aarts, I. M. Rietjens, Chem. Res. Toxicol. 2006, 19, 1499–
1505.
[18] E. V. Schmalhausen, E. B. Zhlobek, I. N. Shalova, O. Firuzi,
L. Saso, V. I. Muronetz, Food Chem. Toxicol. 2007, 45, 1988–
1993.
[19] M. A. Lea, C. Ibeh, J. K. Deutsch, I. Hamid, C. desBordes,
Anticancer Res. 2010, 30, 3629–3635.
[20] G. Galati, T. Chan, B. Wu, P. J. O’Brien, Chem. Res. Toxicol.
1999, 12, 521–525.
[21] D. Metodiewa, A. K. Jaiswal, N. Cenas, E. Dickancaité, J. Seg-
ura-Aguilar, Free Radical Biol. Med. 1999, 26, 107–116.
[22] Y. Sakihama, M. F. Cohen, S. C. Grace, H. Yamasaki, Toxi-
cology 2002, 177, 67–80.
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
[53]
[54]
[55]
[23] W. Wätjen, G. Michels, B. Steffan, P. Niering, Y. Chovolou, A.
Kampkötter, Q. H. Tran-Thi, P. Proksch, R. Kahl, J. Nutr.
2005, 135, 525–531.
[24] M. Kanadzu, Y. Lu, K. Morimoto, Cancer Lett. 2006, 241,
A. Zhou, S. Kikandi, O. A. Sadik, Electrochem. Commun.
2007, 9, 2246–2255.
250–255.
[25] C. S. Yang, S. Sang, J. D. Lambert, M. J. Lee, Mol. Nutr. Food
Res. 2008, 52 (suppl. 1), S139–S151.
[26] L. Biasutto, E. Marotta, U. De Marchi, M. Zoratti, C. Parad-
isi, J. Med. Chem. 2007, 50, 241–253.
[27] L. Biasutto, E. Marotta, A. Mattarei, S. Beltramello, P. Cal-
iceti, S. Salmaso, A. Bernkop-Schnurch, S. Garbisa, M. Zor-
atti, C. Paradisi, Cell. Physiol. Biochem. 2009, 24, 557–566.
[28] L. Biasutto, E. Marotta, A. Bradaschia, M. Fallica, A. Matta-
rei, S. Garbisa, M. Zoratti, C. Paradisi, Bioorg. Med. Chem.
Lett. 2009, 19, 6721–6724.
[29] M. P. Murphy, R. A. Smith, Annu. Rev. Pharmacol. Toxicol.
2007, 47, 629–656.
[30] A. T. Hoye, J. E. Davoren, P. Wipf, M. P. Fink, V. E. Kagan,
Acc. Chem. Res. 2008, 41, 87–97.
A. Mattarei, L. Biasutto, F. Rastrelli, S. Garbisa, E. Marotta,
M. Zoratti, C. Paradisi, Molecules 2010, 15, 4722–4736.
M. C. Foti, C. Daquino, I. D. Mackie, G. A. Dilabio, K. U.
Ingold, J. Org. Chem. 2008, 73, 9270–9282 and references
therein.
J. S. Choi, M. J. Jung, H. J. Park, H. Y. Chung, S. S. Kang,
Arch. Pharmacal Res. 2002, 25, 625–627.
S. R. Kanatt, R. Chander, P. Radhakrishna, A. Sharma, J.
Agric. Food Chem. 2005, 53, 1499–1504.
Z. Cheng, J. Moore, L. Yu, J. Agric. Food Chem. 2006, 54,
7429–7436.
E. Sergediene, K. Jönsson, H. Szymusiak, B. Tyrakowska, I. M.
Rietjens, N. Cenas, FEBS Lett. 1999, 462, 392–396.
[56]
[57]
[58]
[59]
Eur. J. Org. Chem. 2011, 5577–5586
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5585

C. Paradisi, et al.
FULL PAPER
[60] L. R. Fukumoto, G. Mazza, J. Agric. Food Chem. 2000, 48, [66] O. Firuzi, P. Mladenka, R. Petrucci, G. Marrosu, L. Saso, J.
3597–3604.
Pharm. Pharmacol. 2004, 56, 801–807.
[67] O. Firuzi, A. Lacanna, R. Petrucci, G. Marrosu, L. Saso, Bi-
ochim. Biophys. Acta 2005, 1721, 174–184.
[68] M. J. Arts, G. R. Haenen, H. P. Voss, A. Bast, Food Chem. Tox-
icol. 2004, 42, 45–49.
[69] M. Kanadzu, Y. Lu, K. Morimoto, Cancer Lett. 2006, 241,
250–255.
[61] M. S. Blois, Nature 1958, 181, 1199–1200.
[62] W. Brand-Williams, M. E. Cuvelier, C. Berset, Lebensm. Wiss.
Technol. 1995, 28, 25–30.
[63] G. Sichel, C. Corsaro, M. Scalia, A. J. Di Bilio, R. P. Bonomo,
Free Radical Biol. Med. 1991, 11, 1–8.
[64] K. Ioku, T. Tsushida, Y. Takei, N. Nakatani, J. Terao, Biochim.
Biophys. Acta 1997, 1234, 99–104.
[65] G. R. Haenen, J. B. Paquay, R. E. Korthouwer, A. Bast, Bio-
chem. Biophys. Res. Commun. 1997, 236, 591–593.
Received: April 23, 2011
Published Online: August 17, 2011
5586
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 5577–5586