Content uploaded by Vladimir A Kostyuk
Author content
All content in this area was uploaded by Vladimir A Kostyuk
Content may be subject to copyright.
Reactive oxygen species (ROS) which include oxy-
gen-centered radicals (О2
, •OH, NO•, RO•, ROO•) and
non-radical molecules (hydrogen peroxide, singlet oxy-
gen, hypochloric acid) are produced by cells as a result of
aerobic metabolism. ROS are involved in the organism’s
normal vital activity, including phagocytosis, regulation
of cell proliferation, intracellular signalization, and syn-
thesis of biologically active compounds and ATP [1]. With
an insufficiency of the antioxidant protective system or
under an intense influence of radical-initiating factors
(ionizing radiation, hard ultraviolet radiation, xenobi-
otics, mineral dust), ROS are hyperproduced and oxida-
tive stress develops. Oxidative stress is a specific feature in
pathogenesis of various diseases, including cardiovascular
diseases, diabetes, tumors, rheumatoid arthritis, and
epilepsy [1]. Although in many cases oxidative stress is
not the cause but a symptom of a disease, there is good
evidence for preventive and therapeutic effects of natural
antioxidants including flavonoids (quercetin, rutin, green
tea catechins) [2-4]. Because of their low redox potential
(0.23 < E7< 0.75 V), flavonoids can reduce highly oxi-
dized free radicals with redox potential values of 2.13-
1.0 V (О2
, •OH, NO•, RO•, ROO•) [5]. Moreover,
flavonoids can suppress the production of ROS due to
inhibition of redox enzymes (monooxygenase, cyclooxy-
genase, lipoxygenase, xanthine oxidase, NADH oxidase)
[1, 2] and also bind ions of variable valence metals which
are involved in generation of oxygen radicals by Fenton’s
reaction [6]. The biological activities of flavonoids (anti-
inflammatory, anti-allergic, anti-carcinogenic) is
believed to be mainly due to their antioxidative proper-
ties, but the contribution of antiradical mechanisms to
their biological effects remains unclear. We found earlier
a pronounced antioxidative effect of the flavonols
quercetin and rutin under conditions of microsomal lipid
peroxidation (LPO) in vitro [6, 7] and their ability to
interact with oxygen-anion radical and thus protect
phagocytizing cells against asbestos-induced damage [8,
9]. This work presents a comparative study of antioxidant
properties of flavonoids with similar structure: catechins
epicatechin gallate and epigallocatechin gallate, flavonols
quercetin and rutin, and a flavononol dihydroquercetin.
We also attempted to establish the role of antiradical
mechanisms in the cytoprotective effect of the com-
pounds studied under conditions of oxidative stress with
asbestos-induced damage to phagocytizing cells.
Biochemistry (Moscow), Vol. 68, No. 5, 2003, pp. 514-519. Translated from Biokhimiya, Vol. 68, No. 5, 2003, pp. 632-638.
Original Russian Text Copyright © 2003 by Potapovich, Kostyuk.
0006-2979/03/6805-0514$25.00 ©2003 MAIK “Nauka/Interperiodica”
* To whom correspondence should be addressed.
Comparative Study of Antioxidant Properties
and Cytoprotective Activity of Flavonoids
A. I. Potapovich and V. A. Kostyuk*
School of Biology, Belorussian State University, pr. F. Skoriny 4, Minsk 220050, Belarus;
fax: 375 (172) 77-5535; E-mail: kostyuk@bsu.by
Received February 14, 2002
Revision received May 15, 2002
AbstractѕAntioxidant properties and cytoprotective activity of flavonoids (rutin, dihydroquercetin, quercetin, epigallocate-
chin gallate (EGCG), epicatechin gallate (ECG)) were studied. All these compounds inhibited both NADPH- and CCl4-
dependent microsomal lipid peroxidation, and the catechins were the most effective antioxidants. The I50 values calculated
for these compounds by regression analysis were close to the I50 value of the standard synthetic antioxidant ionol (2,6-di-tert-
butyl-4-methylphenol). The antiradical activity of flavonoids to О2
was studied in a model photochemical system. Rate con-
stants of the second order reaction obtained by competitive kinetics suggested flavonoids to be more effective scavengers of
oxygen anion-radicals than ascorbic acid. By competitive replacement all flavonoids studied were shown to be chelating
agents capable of producing stable complexes with transition metal ions (Fe2+, Fe3+ , Cu2+). The flavonoids protected
macrophages from asbestos-induced damage, and the protective effect increased in the following series: rutin < dihydro-
quercetin < quercetin < ECG < EGCG. The cytoprotective effect of flavonoids was in strong positive correlation with their
antiradical activity to О2
.
Key words: antioxidants, flavonoids, oxidative stress, reactive oxygen species, asbestos, rutin, quercetin, epicatechins
ANTIOXIDANT PROPERTIES AND CYTOPROTECTIVE ACTIVITY OF FLAVONOIDS 515
BIOCHEMISTRY (Moscow) Vol. 68 No. 5 2003
MATERIALS AND METHODS
Superoxide-dependent reduction of p-nitrotetrazoli-
um chloride (85 µM) was performed in 0.175 M phosphate
buffer (pH 7.8) containing 0.06 mM EDTA, 0.6 mM
N,N,N′,N′-tetramethylethylene diamine, and 6µM
riboflavin [10]. The light source was an LD-20 daylight
lamp (20 W) placed at the distance of 20 cm from the
specimens. After the illumination the reaction was
stopped by introduction into the samples of 0.02 ml of
superoxide dismutase (SOD, 5µg/ml), and the absorp-
tion of the specimens was determined at 515 nm.
Lipid peroxidation was induced by addition of
0.3 mM NADPH and was performed at 37°C. During
CCl4-initiated LPO the incubation medium contained
0.05 M phosphate buffer (pH 7.4), 0.02 M KCl, 0.6 mM
EDTA, and 3.4 mM CCl4dissolved in alcohol at the
final concentration of ethanol 2%, and also microsomal
protein (1.2 mg/ml). In the course of the NADPH-
dependent LPO the incubation medium contained
0.05 M phosphate buffer (pH 7.4), 0.02 M KCl, 10 µM
FeSO4, and microsomal protein (1.2 mg/ml).
Microsomes were isolated from rat liver by differential
centrifugation at 105,000g using a VAC 601 centrifuge
(Germany). The protein content was determined by the
Lowry method. To determine contents of LPO products,
the samples (1 ml) were supplemented with 0.5 ml of
30% TCA and 2.5 ml of 0.5% 2′-thiobarbituric acid
(TBA), the mixture was kept for 15 min in a boiling
water bath, and it was then centrifuged to remove the
denatured protein and the absorption was determined at
532 nm.
Peritoneal macrophages were prepared by a modifi-
cation of a published method [11]. Suspension of
macrophages in isotonic phosphate buffer (pH 7.3) con-
tained 5·106cells/ml. Oxidative stress in the macrophages
was induced by addition of aqueous suspension of
asbestos at the final concentration of 3 mg/ml. The sam-
ples were incubated at 37°C for 20 min, and the degree of
cell damage was determined by release of lactate dehy-
drogenase (LDH).
Lactate dehydrogenase activity was determined spec-
trophotometrically by the rate of NADH utilization in the
enzyme-catalyzed back reaction of pyruvate conversion
to lactate.
Reduced glutathione content was determined using
Ellman’s reagent [12].
To assess and compare antiradical,antioxidant,and
cytoprotective properties of individual chemical com-
pounds, the parameter I50 was used equal to the concen-
tration of antioxidant corresponding to 50% inhibition of
processes under study. The I50 values were calculated from
the dose-effect dependency by regression analysis.
The following reagents were used: epigallocatechin
gallate (EGCG), epicatechin gallate (ECG), rutin,
quercetin, dihydroquercetin, riboflavin, sodium pyru-
vate, superoxide dismutase, NADH, thiobarbituric acid
from Sigma (USA); p-nitrotetrazolium chloride,
N,N,N′,N′-tetramethylethylene diamine from Reanal
(Hungary); chrysotile asbestos [Mg6Si4O10(OH)8] (the
length of the fibers was 5-10 µM) was from the Tuva
deposit (Russia).
RESULTS AND DISCUSSION
Study of antioxidative activity of flavonoids.
Antioxidative properties of the flavonoids (Table 1) were
studied on initiation of NADPH- and CCl4-dependent
LPO in microsomal membranes of rat liver. The
NADPH-dependent LPO was more likely initiated by
ADP-perferryl-ion (ADP-Fe2+-O2↔ADP-Fe3+-О2
)
which was generated with involvement of the NADPH-
dependent flavoprotein and could introduce the activated
oxygen into molecules of polyunsaturated fatty acids and
destroy hydroperoxides produced [13]. In the case of
CCl4-dependent LPO, the initiation stage included the
metabolic activation of CCl4in the microsomal electron
transport chain with involvement of cytochrome P450
and production of free radical intermediates including
CCl3O2
•which could initiate LPO in the absence of iron
ions [14].
All flavonoids studied effectively inhibited both
NADPH- and CCl4-dependent LPO, and the degree of
inhibition monotonically increased for all flavonoids with
increase in their concentration, and this allowed us to cal-
culate the I50 values by regression analysis (Table 2) and
correctly use these values to comparatively assess the
antioxidant effects. It should be noted that the antioxi-
dant effects of ECG and EGCG, which displayed the
strongest antioxidative properties, were comparable to the
effect of ionol (2,6-di-tert-butyl-4-methylphenol), a syn-
thetic antioxidant of hindered phenols.
Antiradical properties of flavonoids to oxygen anion-
radical. The interaction of antiradical agents with oxygen
anion-radical can be characterized by the rate constant of
the second order reaction. In addition to direct methods
of determination of this parameter based on measuring
the rate of changes in the О2
concentration, the method
of competitive kinetics is widely used. In this case a con-
stant level of oxygen anion-radical is provided by chemi-
cal О2
•-generating systems, and the inhibition of reac-
tions of О2
-dependent reduction (oxidation) of a test
substance is determined relative to a standard. As an О2
-
generating system xanthine oxidase is usually used and
cytochrome c is used as a test substance. But flavonoids
are known to inhibit xanthine oxidase [2], and quercetin
and some other flavonoids can directly reduce
cytochrome c. Therefore, in the present study a
riboflavin-containing photosystem [10] was used for gen-
eration of oxygen anion-radical and the antiradical effect
of flavonoids was assessed by inhibition by these com-
516 POTAPOVICH, KOSTYUK
BIOCHEMISTRY (Moscow) Vol. 68 No. 5 2003
pounds of the О2
-dependent reduction of p-nitrotetra-
zolium chloride to diformazan.
Our findings suggested that flavonoids should be
effective scavengers of oxygen anion-radical. Based on I50
values and with superoxide dismutase as a standard, the
rate constants of the second order reaction were calculat-
ed for interaction of flavonoids (FL) with oxygen anion-
radical (Table 2):
kFL/kSOD = I50 (SOD)/I50 (FL),
kFL = kSOD · I50 (SOD)/I50 (FL),
where kSOD = 2·109M–1·sec–1 [1].
Note that the rate constant of the quercetin reaction
with О2
determined by us by competitive kinetics (Table
2) is close to the value obtained by other authors by pulse
radiolysis and EPR-spectrometry (0.9·105M–1·sec–1)
[15]. Our findings showed that all flavonoids studied,
except rutin, were more effective scavengers of oxygen
anion-radical than the water-soluble antioxidant ascorbic
acid. The high antiradical activity of flavonoids to oxygen
anion-radical is obviously due to the reactivity of hydrox-
yl groups in the m-position of ring A and o-position of
Group
Flavans (catechins)
Flavonols
Flavanonols
Structure Compound under study
EGCG,
[3-gallate, 3′,4′,5′,5,7-(OH)5]
ECG, [3-gallate, 3′,4′,5,7-(OH)4]
quercetin, [3,3′,4′,5,7-(OH)5]
rutin,
[3-rutinoside, 3′,4′,5,7-(OH)4]
dihydroquercetin,
[3,3′,4′,5,7-(OH)5]
Table 1. Flavonoids under study and their structures
Flavonoid
Rutin
Dihydroquercetin
Quercetin
Epicatechin gal-
late
Epigallocatechin
gallate
Ionol
Ascorbic acid
NADPH-
dependent
LPO (I50, µM)
19.0
38.0
4.5
2.5
2.0
1.1
—
k×105,
M–1·sec–1
0.5
1.5
1.7
3.5
5.4
—
0.7
CCl4-
dependent
LPO (I50,
µM)
160.0
110.0
6.0
2.5
2.5
1.1
—
Table 2. Antioxidant effects of flavonoids (I50) on initia-
tion of LPO in liver microsomes and rate constants of
flavonoid reactions with oxygen anion-radical
ANTIOXIDANT PROPERTIES AND CYTOPROTECTIVE ACTIVITY OF FLAVONOIDS 517
BIOCHEMISTRY (Moscow) Vol. 68 No. 5 2003
ring B, and the antiradical activity of flavonoids increases
with increase in the number of hydroxyl groups in their
structure. Thus, the highest activity was found for cate-
chins EGCG and ECG which due to gallation (addition
of a gallic acid residue at C3) acquired additional hydrox-
yl groups. On the contrary, because glycosylation blocked
chemically active groups of quercetin, its glycoside rutin
displayed a significant (more than threefold) decrease in
antiradical activity, and this is in agreement with data of
other authors [5].
Study of chelating properties of flavonoids. The
antioxidant effect of flavonoids can be associated with the
binding of ions of variable valence metals which play the
key role in initiation of free radical reactions [1]. The
ability of some flavonoids to produce stable complexes
with metal ions can be detected by differential spec-
troscopy [6]. Figure 1a presents absorption spectra of
quercetin and its metallocomplexes which are produced
by interaction of the flavonoid with trivalent iron ions.
Similar absorption spectra are recorded for the complex-
ing of quercetin and rutin with Cu2+ and Fe2+.
The chelating properties of dihydroquercetin and
catechins, the interaction of which with metal ions was
not accompanied by pronounced spectral changes, were
studied by competitive replacement. In these experiments
we determined the ability of dihydroquercetin, ECG, and
EGCG to displace rutin from its complexes with metals.
The ability of EDTA, which is one of most active chela-
tors, to displace rutin and quercetin from their complexes
with metals was also determined. By differential spec-
troscopy and competitive replacement quercetin was
found to be the strongest chelator of metal ions of the
flavonoids studied. In particular, quercetin was not dis-
placed from the complex with Fe3+ by a twofold excess of
EDTA (Fig. 1a). In total, the ability of flavonoids for com-
plexing with transition metal ions is described as follows:
Thus, all flavonoids studied were chelating agents
and could bind ions of variable valence metals producing
Fig. 1. Typical changes in the absorption spectra of quercetin on complexing with trivalent iron ions (a) and during oxidation in the system
containing peritoneal macrophages and asbestos (b). a) The absorption spectrum of quercetin (15 µM) in 0.9% NaCl (1), the differential
spectrum of Fe3+-quercetin complex against 0.9% NaCl containing Fe3+ (2), the differential spectrum of Fe3+-quercetin complex just on
addition of EDTA (30 µM) against 0.9% NaCl containing Fe3+ and EDTA (3), the same as spectrum 3but 15 min after the addition of
EDTA (4). b) Peritoneal macrophages (2·106cells/ml) were incubated at 37°C in isotonic phosphate buffer (pH 7.3) containing chrysotile
asbestos (2 mg/ml) and 60 µM quercetin. The samples were centrifuged, and the absorption spectrum of the supernatant fluid diluted 2.5-
fold was recorded: baseline spectrum of quercetin (1), spectrum of quercetin after 20 min of incubation (2), spectrum of quercetin after
20 min of incubation in the presence of SOD (100 µg/ml) (3).
1
0.25
0
2
250
3
4
А
350 450 550
ab
0.2
0.1
0
А
1
3
2
300 350 400 450 500
λ, nm λ, nm
Metal
Fe2+
Fe3+
Cu2+
Complexing ability of flavonoids
quercetin > rutin >> ECG = EGCG > dihydro-
quercetin
quercetin > EGCG = ECG > rutin >> dihydro-
quercetin
rutin > EGCG = ECG = dihydroquercetin
518 POTAPOVICH, KOSTYUK
BIOCHEMISTRY (Moscow) Vol. 68 No. 5 2003
stable complexes. However, there was no direct depend-
ence between the inhibition of microsomal LPO by
flavonoids (Table 2) and their chelating properties, and
this suggested that the chelating of metal ions insignifi-
cantly contributed to the antioxidant effect of flavonoids.
Ability of flavonoids to prevent oxidative damage to
cells was studied using a model of asbestos-induced
oxidative stress in isolated phagocytizing cells (peritoneal
macrophages). The incubation of peritoneal
macrophages with chrysotile asbestos fibers was earlier
shown [11] to sharply increase the production of oxygen
anion-radical by the NADPH oxidase complex (respira-
tory burst) that resulted in damage and lysis of the phago-
cytizing cells. In our experiments incubation for 20 min
of peritoneal macrophages with chrysotile asbestos fibers
(3 mg/ml) resulted in a sharp decrease in the level of
reduced glutathione (from 7.3 ± 2.0 to 3.1 ± 1.5 µM, p <
0.01) and in significant damage and lysis of the cells that
was manifested by a decrease in the number of
macrophages from (5.5 ± 0.6)·106to (1.0 ± 0.2)·106
cells/ml (p < 0.01) and by the release of a cytoplasmic
enzyme LDH into the incubation medium (86 ± 12%, p <
0.01).
All flavonoids studied prevented the asbestos-
induced damage of peritoneal macrophages which was
assessed by entrance of LDH into the incubation medi-
um. The protective effect can be characterized by I50
value equal to the flavonoid concentration providing the
50% decrease in the degree of cell damage. Table 3 pres-
ents I50 values calculated by regression analysis from the
corresponding dose-effect dependences. It is more con-
venient to describe the cytoprotective effect (CPE) in
arbitrary units, with CPE of the most effective cytopro-
tector epigallocatechin gallate taken as the unit and CPE
of other flavonoids calculated by the formula:
CPEFL = 1·(I50 (EGCG)/I50 (FL)).
Data presented in Table 3 show that catechins
EGCG and ECG protect the cells against lysis signifi-
cantly more effectively than quercetin and dihydro-
quercetin and more than 20-fold more effectively than
rutin. Comparison of the CPE values and rate constants
of the flavonoid reactions with oxygen anion-radical by
correlation analysis shows a direct correlation between
the cytoprotective effect of flavonoids studied and their
antiradical activity toward О2
(Fig. 2), and both these
parameters increase in the following series: rutin <
dihydroquercetin < quercetin < ECG < EGCG.
However, there is no correlation between the cytopro-
tective effect and the chelating properties of the
flavonoids.
It is known that oxidation by oxygen anion-radical of
flavonols quercetin and rutin is accompanied by specific
changes in their spectra [8]. Similar changes were record-
ed in the spectra of quercetin (Fig. 1b) and rutin in the
presence of asbestos-activated peritoneal macrophages.
On addition into the incubation medium of the enzyme
SOD (100 µg/ml) catalyzing the dismutation of О2
the
oxidation of rutin was inhibited virtually completely and
the oxidation of quercetin was inhibited by 50% (Fig. 1b).
These findings show that flavonoids can scavenge oxygen
anion-radical generated during the asbestos-induced res-
piratory burst in peritoneal macrophages that results in
inhibition of Fenton’s reaction and other processes with
involvement of О2
which cause damage and lysis of the
cells.
Fig. 2. Dependence between the cytoprotective effect and anti-
radical properties of flavonoids: rutin (1), dihydroquercetin
(2), quercetin (3), ECG (4), EGCG (5). The correlation coef-
ficient is 0.89 ± 0.24; the regression coefficient is 0.14 ± 0.04,
p= 0.042.
1
0.9
0
2
0
3
4
CPE
2 4 6
k×105, M–1·sec–1
5
0.6
0.3
8
Flavonoid
Rutin
Dihydroquercetin
Quercetin
Epicatechin gallate
Epigallocatechin gallate
I50, µM
275
150
39
12
10
CPE*, arbitrary units
0.036
0.067
0.26
0.80
1.00
Table 3. Values of I50 and cytoprotective effects of
flavonoids on asbestos-induced damage to peritoneal
macrophages
The CPE of the most effective cytoprotector EGCG is taken as the
unit and CPE of other compounds are calculated by the formula:
CPEFL = 1·(I50 (EGCG)/I50 (FL)).
*
ANTIOXIDANT PROPERTIES AND CYTOPROTECTIVE ACTIVITY OF FLAVONOIDS 519
BIOCHEMISTRY (Moscow) Vol. 68 No. 5 2003
Thus, the studied flavonoids are shown to have high
antioxidative activity and prevent the asbestos-induced
damage of phagocytizing cells. These findings confirm
the key role of О2
in the asbestos-induced damage of
phagocytizing cells and suggest that the cytoprotective
effect of flavonoids is due to their antiradical activity
toward oxygen anion-radical.
REFERENCES
1. Halliwell, B., and Gutteridge, J. M. C. (1989) Free Radical
in Biology and Medicine, 2nd ed., Clarendon Press, Oxford
University Press, Oxford.
2. Korkina, L. G., and Afanas’ev, I. B. (1997) Adv.
Pharmacol., 38, 151-163.
3. Ho, C.-T., Chen, Q., Shi, H., Zhang, K.-Q., and Rozen,
R. T. (1992) Prevent. Med., 21, 520-525.
4. Hof, K. H., Wiseman, S. A., Yang, C. S., and Tijburg, L. B.
(1999) Proc. Soc. Exp. Biol. Med., 220, 203-209.
5. Pietta, P. G. (2000) J. Nat. Prod., 63, 1035-1042.
6. Afanas’ev, I. B., Dorozhko, A. I., Brodskii, A. V., Kostyuk,
V. A., and Potapovitch, A. I. (1989) Biochem. Pharmacol.,
38, 1763-1769.
7. Kostyuk, V. A., Potapovich, A. I., Tereshchenko, S. M.,
and Afanas’ev, I. B. (1988) Biokhimiya, 53, 1365-1370.
8. Kostyuk, V. A., and Potapovitch, A. I. (1989) Biochem. Int.,
19, 1117-1124.
9. Kostyuk, V. A., Potapovich, A. I., Speransky, S. D., and
Maslova, G. T. (1996) Free Rad. Biol. Med.,21, 487-493.
10. Beauchamp, C., and Fridovich, I. (1971) Analyt. Biochem.,
44, 276-287.
11. Korkina, L. G., Suslova, T. B., Cheremisina, Z. P., and
Velichkovsky, B. T. (1988) Stud. Biophys., 126, 99-104.
12. Habeeb, A. F. S. A. (1972) Meth. Enzymol., 25, 457-464.
13. Svingen, B. A., Buege, J. A., O’Neal, F. O., and Aust, S. D.
(1979) J. Biol. Chem., 254, 5892-5899.
14. Waller, R. L., Glende, E. A., Jr., and Recknagel, R. O.
(1983) Biochem. Pharmacol., 1983, 1613-1617.
15. Bors, W., Heller, W., Michel, C., and Saran, M. (1990)
Meth. Enzymol., 186,343-354.
