Efficiency and Mechanism of the Antioxidant Action of trans-Resveratrol and Its Analogues in the Radical Liposome Oxidation

nlopim

tvrfltadhmfitdtp

dgeP

Archives of Biochemistry and BiophysicsVol. 391, No. 1, July 1, pp. 79–89, 2001doi:10.1006/abbi.2001.2388, available online at http://www.idealibrary.com on

Efficiency and Mechanism of the Antioxidant Actionof trans-Resveratrol and Its Analogues in theRadical Liposome Oxidation

Sandra Stojanovic, Helmut Sprinz, and Ortwin Brede1

Research Unit Time-Resolved Spectroscopy, University of Leipzig, Permoserstrasse 15, D-04318 Leipzig, Germany

Received January 29, 2001, and in revised form April 3, 2001; published online June 4, 2001

trans-Resveratrol (trans-3,5,4*-trihydroxystilbene) is aonflavonoid polyphenol reported to exert different bio-

ogical activities, among them inhibition of the lipid per-xidation, scavenging of the free radicals, inhibition of thelatelet aggregation, and anticancer activity as the most

mportant. In order to enlighten the radical-scavengingechanism of trans-resveratrol, stationary g-radiolytic ex-

periments in liposomes and pulse radiolytic experimentsin aqueous solutions were performed. Applying the sta-tionary g-radiolysis together with the subsequent productanalysis, reactions of lipid peroxyl radicals, LOO•, withrans-resveratrol and other natural antioxidants were in-estigated. It was found that trans-resveratrol was a betteradical scavenger than vitamins E and C but similar to theavonoids epicatechin and quercetin. The comparison ofhe radical-scavenging effects of trans-resveratrol and itsnalogues trans-4-hydroxystilbene and trans-3,5-dihy-roxystilbene revealed that trans-resveratrol and trans-4-ydroxystilbene showed almost the same effect and wereore efficient than trans-3,5-dihydroxystilbene. Thesendings indicate greater radical-scavenging activity ofrans-resveratrols para-hydroxyl group than its meta-hy-roxyl groups. Using the pulse radiolysis, reactions ofrans-resveratrol and its analogues with trichloromethyl-eroxyl radicals, CCl3OO•, were studied. Spectral and

kinetic properties of the observed transients showedgreat similarity between trans-resveratrol and trans-4-hy-

roxystilbene which seems to confirm that para-hydroxylroup of trans-resveratrol scavenges free radicals moreffectively than its meta-hydroxyl groups. © 2001 Academic

ress

Key Words: resveratrol; lipid peroxidation; antioxi-dants; radical scavenging; vitamin E; vitamin C; epi-catechin; quercetin; pulse radiolysis.

1
To whom correspondence should be addressed. Fax: 149-341-235-2317. E-mail: [email protected].
0003-9861/01 $35.00Copyright © 2001 by Academic PressAll rights of reproduction in any form reserved.

It is well known that lipid peroxidation representsone of the basic mechanisms in the cell and tissuedamage thus leading to various diseases, e.g., athero-sclerosis, inflammation, and cancer (1, 2). Lipid peroxi-dation (peroxidation of polyunsaturated fatty acid res-idues in lipids) is a chain-reaction process (Eq. [1]–]5])which can be induced by different free-radical sources(R•) such as enzymatic reactions, ionizing irradiation,and UV light. However, lipid peroxidation can be in-hibited in the presence of various antioxidants whichcan act at different levels. The chain-breaking antioxi-dants (e.g., phenolic antioxidants, ArOH) are one of theantioxidant types which inhibit the process of lipidperoxidation by scavenging the free radicals such aslipid peroxyl radicals, LOO•, converting them intolong-lived and less reactive ArO• radicals (Eq. [6]).

Initiation R• 1 LH3 RH 1 L• [1]

Propagation L• 1 O23 LOO• [2]

LOO• 1 LH3 LOOH 1 L• [3]

L• 1 O23 LOO• [4]

Termination L•/LOO•3 nonradical products

[5]

Antioxidant action ArOH 1 LOO•3 ArO•

1 LOOH. [6]

Polyphenolic compounds also belong to the chain-breaking antioxidants besides tocopherols (vitamin Eas the representative) (3). One of them is resveratrol, a
nonflavonoid polyphenol which recently attracted
79

ord

c

80 STOJANOVIC, SPRINZ, AND BREDE

great interest in the science due to the “French para-dox” (despite high fat intake, mortality from coronaryheart disease is lower in some regions of France than inthe other developed countries due to the regular wineconsumption) (4). It has been reported that besidesinhibition of the lipid peroxidation (5–8), resveratrolalso inhibits platelet aggregation (9–12) and exhibitsanticancer (13–17) and anti-inflammatory activity (18).

Resveratrol (3,5,49-trihydroxystilbene) exists in twoisomeric forms as trans- and cis-resveratrol (Fig. 1).Although present in 72 plant species (19), grapes andwine are the most important foodstuff containing res-veratrol. Concentration of trans-resveratrol in redwines ranges from 0.1 to 10 mg dm23 (20, 21) whereasin white wines it is lower than 0.1 mg dm23 (22). Theccurrence of cis-resveratrol in grapes has not beeneported but it has been detected in wines (0.1–3.0 mgm23 in red wines (20) and ,0.1 mg dm23 in white

wines (23)). Due to its light sensitivity there has beenlittle information on cis-resveratrols properties (24,25).

It has been demonstrated that trans-resveratrol sup-presses lipid peroxidation both by chelation of copper(7, 26) and by scavenging of the free radicals (7, 26–29). However, its radical-scavenging mechanism hasnot been established so far; thus it is the subject ofinvestigation in the present study.

In the first part of this study, lipid peroxidation ofegg yolk liposomes alone and in the presence of variousnatural antioxidants has been induced using station-ary g-radiolysis. In this way reactions of lipid peroxylradicals, LOO•, with trans-resveratrol and other natu-ral antioxidants were performed and subsequent prod-uct analysis by high-performance liquid chromatogra-phy (HPLC) and iodometric technique enabled compar-

FIG. 1. Isomeric forms of resveratrol. (a) trans-Resveratrol; (b)is-resveratrol.

ison between radical-scavenging effects of the applied

antioxidants. Apart from trans-resveratrol, antioxi-dants included vitamins E and C, the flavonoids epi-catechin and quercetin, and the trans-resveratrols an-alogues trans-4-hydroxystilbene and trans-3,5-dihy-droxystilbene.

To characterize the elementary reactions of trans-resveratrol’s antioxidant action, in the complementarypart of this study pulse radiolytic experiments withtrichloromethylperoxyl radicals, CCl3OO•, were per-formed. It is known that CCl3OO• radicals are involvedin different pathogeneses, especially lipid peroxidation(30), and they can be successfully generated by pulseradiolysis of carbon tetrachloride, CCl4, in aeratedaqueous solution containing 2-propanol, (CH3)2CHOH,and acetone, (CH3)2CO (Eqs. [7–12] (31).

OH• 1 ~CH3!2CHOH3 H2O 1 ~CH3!2C•OH [7]

H• 1 ~CH3!2CHOH3 H2 1 ~CH3!2C•OH [8]

H2O V e aq2 , OH•, H•. . . [9]

e aq2 1 ~CH3!2CO 3 ~CH3!2CO•2 -|0

H1

~CH3!2C•OH

[10]

~CH3!2C•OH 1 CCl4 3

~CH3!2CO 1 CCl3• 1 H1 1 Cl2 [11]

CCl3• 1 O23 CCl3OO•. [12]

In the presence of phenols, trichloromethylperoxylradicals should react according to Eq. [13] producingphenoxyl radicals (32, 33).

CCl3OO•1ArOH3 CCl3OOH 1 ArO•. [13]

Due to the presence of three hydroxyl groups, trans-resveratrol offers several possibilities to CCl3OO• at-tack. In order to analyze in detail the involvement ofthe different phenolic groups, reactions of CCl3OO•

were carried out also with the trans-resveratrol ana-logues trans-4-hydroxystilbene and trans-3,5-dihy-
droxystilbene (Fig. 2). Spectral and kinetic properties

mou

tHetit

Masp

(

1sBldewa

t

wwdf1mft

moie

ttCtlpA

d

81trans-RESVERATROL AND RADICAL LIPOSOME OXIDATION

of the observed transients have been acquired usingthe time-resolved spectroscopy and their comparisonenabled the verification of the radical-scavengingmechanism of trans-resveratrol.

MATERIALS AND METHODS

Chemicals. Egg yolk L-a-phosphatidylcholine (EYPC),2 trans-res-veratrol, and L-ascorbic acid-6-palmitate were obtained from Sigma(Deisenhofen, Germany). (2)-Epicatechin, quercetin dihydrate, andDL-a-tocopherol were purchased from Aldrich (Deisenhofen, Ger-

any), whereas potassium iodide and hydrogen peroxide (30%) werebtained from Merck (Darmstadt, Germany). All substances weresed without further purification.Analogues of trans-resveratrol were synthesized according to

he literature: trans-4-hydroxystilbene according to Friedrich andenning (34) and trans-3,5-dihydroxystilbene according to Bach-

lor et al. (35). The purity of both trans-4-hydroxystilbene andrans-3,5-dihydroxystilbene was checked by determining the melt-ng points and by NMR. Results were in a good agreement withhe literature (34 –36).

Following solvents were applied: methanol and acetic acid fromerck, carbon tetrachloride (Aldrich, Deisenhofen, Germany) and

cetone and 2-propanol from Riedel-de-Haen (Seelze, Germany). Allolvents were of the highest purity available. Aqueous solutions wererepared using Millipore Milli-Q-Plus filtered water.Argon (99.998%) and oxygen (99.998%) were purchased from Linde

Leuna, Germany) and used without further purification.Liposomes. Liposomes were prepared by ultrasonic treatment of

% lipid suspensions (EYPC in 2 mM NaCl solution, pH 6) under atream of argon at 4°C with an ultrasound homogenizer (Bandelin,erlin, Germany). Incorporation of the antioxidants (1 mM) into

iposomes was achieved by the following method: antioxidants wereissolved in methanol, dried at 40°C on a rotary evaporator, and thenvacuated for 1 h under high vacuum. Liposomes were added after-ard and the vessels were shaken overnight at a room temperaturend finally saturated with argon.g-Radiolysis. g-Irradiation experiments were performed using

he Panorama 60Co irradiation source (mean energy 1.25 MeV) at theInstitute for Surface Modification (IOM) in Leipzig, Germany. The

2 Abbreviations used: EYPC, egg yolk L-a-phosphatidylcholine;

FIG. 2. Analogues of trans-resveratrol. (a) trans-4-Hydroxystil-bene; (b) trans-3,5-dihydroxystilbene.

cAPH, 2,29-azobis(2-amidinopropane)dihydrochloride; DPPH, 1,1-iphenyl-2-picrylhydrazil.

liposome samples (with and without antioxidants) were irradiated atthe room temperature with the dose rate of 1.0 kGy h21 and thesamples were saturated with air during the irradiation process.

Pulse radiolysis. Pulse radiolytic studies were carried out usingthe ELIT1M accelerator generating 15-ns pulses of 1 MeV electronswith a dose of 50 Gy (37). Solutions of trans-resveratrol and itsanalogues in the mixture of water:2-propanol:acetone (60:30:10%,v:v:v) were prepared just before the experiments and then 40 mMcarbon tetrachloride was added into the solution. The samples weresaturated with air during the measurements and the experimentswere performed in a closed circulation system due to the volatility ofcarbon tetrachloride. The dose per pulse was determined before eachexperiment by measuring the absorption of the hydrated electron at580 nm.

HPLC. HPLC was performed on a Hewlett–Packard HP-1050system with a Supelco LC-18-DB column (25 3 4.6 cm) and a diodearray UV-visible detector. Measurements were carried out at a con-stant column temperature (40°C) and constant flow rate (1 mlmin21). Chromatographic conditions were 50% methanol and 50%

ater with 0.01% triethylamine as solvent A and 100% methanolith 0.01% triethylamine as solvent B. The elution program wasuring the first 18 min solvent A was maintained at 100%, gradientrom A to B in 2 min, and for the next 20 min solvent B was kept at00%. Five minutes later, gradient rose to A and solvent A wasaintained constant for the last 5 min of elution. Deterioration of the

atty acid residues in liposomes was monitored at 207 nm whereashe lipid hydroperoxide formation was observed at 233 nm.

Iodometric technique for the hydroperoxide determination. Thisethod is based on the oxidation of the iodide ions, I2 by hydroper-

xide, ROOH, and absorbance measurement of the formed tri-iodideons, I3

2 (Eq. [14] and [15]). The whole procedure has been describedlsewhere (38).

ROOH 1 2H1 1 2I23 ROH 1 H2O 1 I2 [14]

I2 1 I2 ^ I32 [15]

Knowing the value for the extinction coefficient of I32 at 358 nm

(«358 5 2.97 z 104 M21 cm21), the quantity of the hydroperoxide can becalculated from the Eq. [16] where A and A 0 represent absorbancesof the sample and of the reference sample respectively (the volume of3 ml was used for all samples).

n(ROOH) 53 z ~ A 2 A0!

1000 z «3585 101 z DA [nmol] [16]

RESULTS

Determination of the Free-Radical-Scavenging Effi-ciency of trans-Resveratrol

For this purpose control liposomes (without antioxi-dants) and liposomes containing one of the antioxidantswere g-irradiated at different doses. In the first set ofexperiments radical-scavenging effect of trans-resvera-rol was compared with the radical-scavenging effects ofhe classic chain-breaking antioxidants, vitamins E and. HPLC product analysis showed that the highest con-

ent of lipid hydroperoxide, LOOH, was in the controliposomes and it increased with the dose (Fig. 3). In theresence of antioxidants, lipid peroxidation was signifi-
antly suppressed. The lowest LOOH content was ob-

w

ddcas

wept

in 1


served in the liposomes containing trans-resveratrol andit was found to be unchanged throughout the irradiationprocess. In contrast to this, the LOOH content in lipo-somes containing vitamin E (a-tocopherol) or vitamin Cderivative (ascorbic acid-6-palmitate3) increased with thedose. For the highest applied dose of 2 kGy, the LOOHcontent in the presence of resveratrol amounted to only5% of the LOOH content in the control liposomes,whereas the LOOH content in the presence of vitamins Eor C reached ;35 and ;63% of the control sample, re-spectively. In other words, trans-resveratrol inhibitedlipid peroxidation with approximately 95% efficiency, vi-tamin E with ;65% efficiency, and vitamin C derivative

ith ;37% efficiency.These results could be well confirmed using the io-

ometric technique (Fig. 4). In the presence of antioxi-ants, the LOOH content was much lower than in theontrol samples and trans-resveratrol showed the bestntioxidant effect whereas the vitamin C derivativehowed the lowest.In the second set of experiments trans-resveratrolas compared with the flavonoid polyphenols epicat-chin and quercetin. Flavonoid polyphenols are alsoresent in red wine but in much higher concentrationhan trans-resveratrol (1–3 g dm23 for the total fla-

vonoids) (28); thus it has been suggested that the ben-eficial effect of red wine on human health could be due

3

FIG. 3. HPLC method for the lipid hydroperoxide characterization

Instead of water-soluble ascorbic acid its lipid-soluble derivativeascorbic acid-6-palmitate was used.

to the flavonoids instead of trans-resveratrol (6). Toprove this point, a comparison of the antioxidant effectof trans-resveratrol with those of the two main fla-vonoid representatives, epicatechin and quercetin, wasperformed. As expected, HPLC product analysis (Fig.5) revealed that liposomes without antioxidants werethe most susceptible to the lipid peroxidation. On theother hand, when one of the polyphenols (trans-res-veratrol, epicatechin, or quercetin) was present, lipidperoxidation was strongly suppressed and it was main-tained constant throughout the irradiation process. Allthree polyphenols showed similar inhibiting effect—forthe highest used dose of 2 kGy trans-resveratrol ex-erted ;95% of inhibition, epicatechin ;96%, and quer-cetin ;93%. Similar results were obtained also by io-dometric method (data not shown).

In order to establish the influence of the spatialposition of the hydroxyl groups on the radical-scaveng-ing effect of trans-resveratrol, in the third set of exper-iments trans-resveratrol was compared with its ana-logues trans-4-hydroxystilbene and trans-3,5-dihy-droxystilbene. This was also of interest under theaspect of antifungal properties reported for both oftrans-resveratrols analogues (39–41). HPLC productanalysis revealed the following: in contrast to the con-trol liposomes where LOOH content was the highest,all three stilbenes successfully inhibited lipid peroxi-dation (Fig. 6). trans-Resveratrol and trans-4-hydrox-ystilbene showed almost the same effect while trans-

% EYPC liposomes with and without addition of 1 mM antioxidants.

3,5-dihydroxystilbene suppressed lipid peroxidation

m

yd


less efficiently. For the highest applied dose of 2 kGy,trans-resveratrol suppressed lipid peroxidation with;96%, trans-4-hydroxystilbene with ;94%, and trans-3,5-dihydroxystilbene with ;78% efficiency. The iodo-

etric method gave similar results—trans-resveratroland trans-4-hydroxystilbene were more efficient thantrans-3,5-dihydroxystilbene (data not shown).

FIG. 4. Results of the iodometric determination of the lipid h

FIG. 5. HPLC method for the lipid hydroperoxide determination. 1

Pulse Radiolysis Experiments

As previously mentioned in the introduction, trichlo-romethylperoxyl radicals, CCl3OO•, react with phenolsforming phenoxyl radicals (32, 33). For the reason ofclarity, the reaction of trans-4-hydroxystilbene withCCl3OO• has been described first. Pulse radiolysis of

roperoxide content (1% EYPC liposomes, 1 mM antioxidants).

% EYPC liposomes, addition of the different 1 mM polyphenols.

da

oxi


the acidic solution (pH 5) of trans-4-hydroxystilbeneyielded a transient absorption spectrum with only oneabsorption maximum at 415–420 nm (Fig. 7). Since itis known that CCl3OO• absorbs only weakly in thenear-UV region (31), the absorption maximum at 420nm is attributed to the phenoxyl radical of trans-4-hydroxystilbene. The bimolecular decay of the phe-noxyl radical proceeded with 2k 18 5 5.3 z 108 M21 s21

(inset in Fig. 7).After reaction between CCl3OO• and trans-3,5-dihy-

droxystilbene, a transient spectrum with two absorp-tion maxima at 365 and 440 nm was obtained (Fig. 8).Phenomenologically speaking, both peaks were presentat shorter times; however, at longer times (.100 ms)the peak at 440 nm disappeared whereas the peak at365 nm slightly increased. The time profiles taken at360 and 440 nm imply that the transient absorption at440 nm decays while the absorption at 360 nm exhibitsa very delayed buildup (inset in Fig. 8). Tentativelyanalyzed, this indicates the presence of two differentspecies, a rapidly formed one possessing two maximaand a delayed one with the absorption around 360 nm.

When trans-resveratrol was exposed to the CCl3OO•

attack, the transient at 420 nm with a shoulder around370 nm was the only one observed as the primaryradical product (Fig. 9). However, some 100 ms afterthe electron pulse, the species at 420 nm slowly disap-peared whereas the transient with the absorption max-imum around 370 nm remained. Similarly to trans-3,5-

ihydroxystilbene, the comparison of the time profilest 420 and 360 nm for trans-resveratrol shows that

while the species at 420 nm decays, the species at 370

FIG. 6. HPLC method for the determination of the lipid hydroper

nm follows a partially delayed buildup (inset in Fig. 9).

The rate constants for the reactions of CCl3OO• withtrans-4-hydroxystilbene, trans-3,5-dihydroxystilbene,and trans-resveratrol were determined by monitoringthe buildups at 415, 440, and 420 nm, respectively(values are given in Table I).

DISCUSSION

g-Radiolytic Measurements

As it can be seen from Figs. 1–4, unprotected lipo-somes were the most susceptible to the peroxidationprocess and the lipid hydroperoxide content increasedproportionally with the applied dose. This pronouncedeffect can be well-distinguished from the low but notnegligible amount of LOOH which was noticed in non-irradiated liposomes indicating that the peroxidationoccurred already during the ultrasonic preparation ofliposomes due to the oxidative stress (42), despite un-dertaken preventive measures.

In the presence of any of the applied antioxidants,lipid peroxidation was significantly reduced.

In comparison to vitamins E and C, trans-resveratrolinhibited lipid peroxidation much more effectively thusshowing to be a better radical scavenger. The lipid-soluble derivative of vitamin C showed the lowest rad-ical-scavenging efficiency. The obtained results are inaccordance with the literature (5) where it was statedthat trans-resveratrol suppressed the lipid peroxida-tion more successfully than vitamin E. The inferiorityof vitamin E (a-tocopherol) versus trans-resveratrolcould be explained taking into consideration that vita-min E contains less hydroxyl groups than trans-res-

des. 1% EYPC liposomes, addition of the different 1 mM stilbenes.

veratrol.

0%


When polyphenolic antioxidants were used, both fla-vonoid polyphenols epicatechin and quercetin and non-flavonoid trans-resveratrol showed a remarkable anti-oxidant effect, but our results are different from thosereported in the literature (5, 26, 29, 43). Frankel et al.(5) reported that epicatechin and quercetin inhibitedcopper-catalyzed peroxidation of human low-densitylipoprotein (LDL) twice better than trans-resveratrol.On the other hand, reports of Fremont et al. (26, 43)and Sanchez-Moreno et al. (29) showed that trans-resveratrol inhibited copper-catalyzed peroxidation ofLDL much better than epicatechin and quercetin but itwas a less powerful free-radical scavenger than thesetwo flavonoids (AAPH and DPPH were used as free-radical generators).

The discrepancy between the literature and the re-sults in this study is probably due to the differences inthe experimental techniques.

The comparison of trans-resveratrol with its ana-logues revealed that all three stilbenes efficiently sup-pressed formation of the lipid hydroperoxides, buttrans-resveratrol and trans-4-hydroxystilbene werefound to be more reactive than trans-3,5-dihydroxystil-

FIG. 7. Transient spectra taken at different times in the pulse radair-saturated solution containing 30% 2-propanol, 10% acetone, and 6trans-4-hydroxystilbene.

bene. The fact that trans-resveratrol and trans-4-hy-

droxystilbene show almost the same antioxidant effectindicates that the radical-scavenging activity of trans-resveratrol depends on the position of the hydroxylgroups. Therefore, for trans-resveratrol it can be con-cluded that its para-hydroxyl group dominates in theradical-scavenging efficiency whereas its meta-hy-droxyl groups show only minor reactivity.

Study of the Elementary Reactions in the Radical-Scavenging Mechanism of trans-Resveratrol

To learn more about the details of trans-resveratrolsantioxidant action, pulse radiolytic investigation oftrans-resveratrol including its analogues was under-taken. For this purpose CCl3OO•, instead of LOO•,were used as model peroxyl radicals.

In the reaction with the simplest analogue, trans-4-hydroxystilbene, CCl3OO• radicals oxidized the para-hydroxyl group producing the phenoxyl radical exhib-iting an absorption maximum at 420 nm (Eq. [17]). Thereaction took place with a rate constant k 17 5 (1.4 60.1) z 108 M21 s21. The obtained phenoxyl radical de-cayed according to Eq. [18] with the rate constant

sis of 2 mM trans-4-hydroxystilbene in the presence of 40 mM CCl4;water, pH 5; dose 50 Gy per pulse. (Inset) decay at 415 nm for 2 mM

ioly

2k 18 5 5.3 z 108 M21 s21.

3


2 ArO•3 nonradical products. [18]

FIG. 8. Transient spectra acquired at different times after pulse radiolysis of 2 mM trans-3,5-dihydroxystilbene in the presence of 40 mMCCl4 in aerated solution (30% 2-propanol, 10% acetone, 60% water, pH 5); dose 50 Gy per pulse. (Inset) corresponding time profiles taken at60 and 440 nm.

af4

( ter3


In the case of trans-3,5-dihydroxystilbene the situa-tion is more complicated due to the existence of twohydroxyl groups. Knowing that at pH 5 trans-3,5-dihy-droxystilbene exists in its neutral form (pK a1 5 7.7 andpK a2 5 10.3) and also from the resemblance to thebsorption spectrum of the phenoxyl radical derivedrom resorcinol (44), the directly formed absorption at

FIG. 9. Transient spectra at different times taken in the pulse radair-saturated solution of 30% 2-propanol, 10% acetone, and 60% wa60 and 420 nm.

40 nm is ascribed to the phenoxyl radical of trans-3,5-

react with oxygen (Eq. [20]). Stating that the obtained

dihydroxystilbene. The synchronously appearing ab-sorption at 360 nm is attributed to the keto radicalform, being a mesomeric form of the phenoxyl radical(Eq. [19]). A similar occurrence has been also observedin the case of phloroglucinol (1,3,5-trihydroxybenzene)(45). The calculated rate constant for the formation ofthe phenoxyl radical derived from trans-3,5-dihydrox-

sis of 2 mM trans-resveratrol in the presence of 40 mM CCl4, pH 5); dose 50 Gy per pulse. (Inset) corresponding time profiles taken at

ystilbene was k 19 5 (1.1 6 0.1) z 107 M21 s21.

Whereas the phenoxyl form dismutates according toEq. [18], the keto form as a C-centered radical is able to

peroxyl radical (Eq. [20]) absorbs also in the near-UVregion, the very delayed buildup could be explained.

ioly

Hence the mesomeric forms seem to follow two compet-

6

1

r


ing pathways, reactions [18] and [20]. The estimatedvalue for k 20 is ;4 z 103 s21.

The transients obtained after reaction betweentrans-resveratrol and CCl3OO• pointed out to the sim-ilarity between trans-resveratrol and both of its ana-

TABLE I

Rate Constants for the Reactions of CCl3OO• withtrans-Resveratrol and Its Analogues

Compound k/M21 s21

trans-4-Hydroxystilbene (1.4 6 0.1) 3 108

trans-3,5-Dihydroxystilbene (1.1 6 0.1) 3 107

trans-Resveratrol (9.6 6 1.0) 3 107

logues. Therefore, the absorption at 420 nm is mainly

der Grenzflachen” at the University of Leipzig for granting S.S. thePh.D. scholarship.

attributed to the phenoxyl radical derived from thepara-hydroxyl group. However, compared to trans-4-hydroxystilbene, the lower absorption of the trans-res-veratrols species at 420 nm and its different spectralshape (shoulder at 370–380 nm) hints to the additionalinvolvement of the two hydroxyl groups in the meta-position. Thus the transient spectrum of trans-resvera-trols species should be interpreted as dominantlycaused by the para-phenoxyl radical superimposedwith the two mesomeric radical forms derived from themeta-hydroxyl groups, analogous to trans-3,5-dihy-droxystilbene (Eq. 21). The delayed buildup at 370 nmpoints out to the probable reaction of the keto radicalform with O2. Knowing that at pH 5 trans-resveratrolis present in the solution in its neutral form (pK a1 5.4, pK 5 9.4, pK 5 10.5), the following reaction
a2 a3
mechanism can be proposed:

Therefore, the phenoxyl radical [21a] decays bimolecu-larly according to Eq. [18], whereas the keto radical[21b] decays through reaction with O2 as described byEq. [20] for trans-3,5-dihydroxystilbene.

The obtained rate constant for the reaction ofCCl3OO• with trans-resveratrol was k 21 5 (9.6 6 1.0) z

07 M21 s21 and it is close to the value obtained fortrans-4-hydroxystilbene, which indicates that thepara-hydroxyl group of trans-resveratrol is more reac-tive in radical scavenging than its meta-hydroxylgroups. This is also in accordance with the g-radiolyticesults.

ACKNOWLEDGMENTS

The authors thank Mrs. Ingeburg Blaue for her excellent technicalassistance, Dr. Jurgen Reinhardt (IOM, Leipzig) for making theg-radiolytic experiments possible, and GK “Physikalische Chemie

REFERENCES

1. Yagi, K. (1982) Lipid Peroxides in Biology and Medicine, Aca-demic Press, New York.

2. Dargel, R. (1992) Exp. Toxicol. Pathol. 44, 169–181.3. Bravo, L. (1998) Nutr. Rev. 56, 317–333.4. Siemann, E. H., and Creasy, L. L. (1992) Am. J. Enol. Vitic. 43,

49–52.5. Frankel, E. N., Waterhouse, A. L., and Kinsella, J. E. (1993)

Lancet 341, 1103–1104.6. Frankel, E. N., Waterhouse, A. L., and Teissedre, P. L. (1995) J.

Agric. Food Chem. 43, 890–894.7. Fremont, L., Belguendouz, L., and Delpal, S. (1999) Life Sci. 64,

2511–2521.8. Blond, J. P., Denis, M. P., and Bezard, J. (1995) Sci. Aliment. 15,

347–358.9. Pace-Asciak, C. R., Hahn, S., Diamandis, E. P., Soleas, G., and

Goldberg, D. M. (1995) Clin. Chim. Acta 235, 207–219.10. Bertelli, A. A. E., Giovannini, L., Gianessi, D., Migliori, M.,

Bernini, W., Fregoni, M., and Bertelli, A. (1995) Int. J. TissueReac. 17, 1–3.

1

11

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

3

3

3

3

3

3

3

3

3

4

44

44


11. Das, D. K., Sato, M., Ray, P. S., Maulik, G., Engelman, R. M.,Bertelli, A. A. E., and Bertelli, A. (1999) Drugs Exp. Clin. Res. 25,115–120.

12. Ruf, J. C. (1999) Drugs Exp. Clin. Res. 25, 125–131.3. Jang, M., Cai, L., Udeani, G. O., Slowing, K. V., Thomas, C. F.,

Beecher, C. W. W., Fong, H. H. S., Farnsworth, N. R., Kinghorn,A. D., Mehta, R. G., Moon, R. C., and Pezzuto, J. M. (1997)Science 275, 218–220.

4. Surh, Y. J. (1999) Mutat. Res. 428, 305–327.5. Jang, M., and Pezzuto, J. M. (1999) Drugs Exp. Clin. Res. 25,

65–77.6. Garcia-Garcia, J., Micol, V., de Godos, A., and Gomez-Fernan-

dez, J. C. (1999) Arch. Biochem. Biophys. 372, 382–388.7. ElAttar, T. M. A., and Virji, A. S. (1999) Anti-Cancer Drugs 10,

187–193.8. Rotondo, S., Rotilio, D., Cerletti, C., and Gaetano, G. (1996)

Thromb. Haemostasis 76, 813–821.9. Wamhoff, H., Richardt, G., Schneider, V., and Tulke, A. (1998)

Chem. Zeit 32, 87–93.0. Goldberg, D. M., Karumanchiri, A., Ng, E., Yan, J., Diamandis,

E. P., and Soleas, G. J. (1995) J. Agric. Food Chem. 43, 1245–1250.

1. Lamuela-Raventos, R. M., Romero-Perez, A. I., Waterhouse,A. L., and Torre-Boronat, M. C. (1995) J. Agric. Food Chem. 43,281–283.

2. Soleas, G. J., Diamandis, E. P., and Goldberg, D. M. (1997) Clin.Biochem. 30, 91–113.

3. Soleas, G. J., Dam, J., Carey, M., and Goldberg, D. M. (1997) J.Agric. Food Chem. 45, 3871–3880.

4. Trela, B. C., and Waterhouse, A. L. (1996) J. Agric. Food Chem.44, 1253–1257.

5. Basly, J. P., Marre-Fournier, F., Le Bail, J. C., Habrioux, G., andChulia, A. J. (2000) Life Sci. 66, 769–777.

6. Belguendouz, L., Fremont, L., and Linard, A. (1997) Biochem.Pharmacol. 53, 1347–1355.

7. Belguendouz, L., Fremont, L., and Gozzelino, M. T. (1998) Bio-chem. Pharmacol. 55, 811–816.

4

8. Fauconneau, B., Waffo-Teguo, P., Huguet, F., Barrier, L., Decen-dit, A., and Merillon, J. M. (1997) Life Sci. 61, 2103–2110.

9. Sanchez-Moreno, C., Larrauri, J. A., and Saura-Calixto, F.(1999) Food Res. Int. 32, 407–412.

0. Slater, T. F. (1982) in Free Radicals, Lipid Peroxidation andCancer (McBrien, D. C. H., and Slater, T. F., Eds.), pp. 243–274,Academic Press, London.

1. Packer, J. E., Wilson, R. L., Bahnemann, D., and Asmus, K.-D.(1980) J. Chem. Soc. Perkin Trans. 2, 296–299.

2. Neta, P., Huie, R. E., Maruthamuthu, P., and Steenken, S.(1989) J. Phys. Chem. 93, 7654–7659.

3. Alfassi, Z. B., Huie, R. E., and Neta, P. (1993) J. Phys. Chem. 97,7253–7257.

4. Friedrich, K. H., and Henning, H. G. (1959) Chem. Ber. 92,2944–2952.

5. Bachelor, F. W., Loman, A. A., and Snowdon, L. R. (1970) Can.J. Chem. 48, 1554–1557.

6. Ali, M. A., Kondo, K., and Tsuda, Y. (1992) Chem. Pharmacol.Bull. 40, 1130–1136.

7. Bertram, D., Brede, O., Helmstreit, W., and Mehnert, R. (1975)Z. Chem. 15, 125–135.

8. Gebicki, J. M., and Guille, J. (1989) Anal. Biochem. 176, 360–364.

9. Langcake, P., and Pryce, R. J. (1977) J. Chem. Soc. Chem.Commun. 7, 208–210.

0. Calderon, A. A., Pedreno, M. A., Ros Barcelo, A., and Munoz, R.(1990) J. Biochem. Biophys. Methods 20, 171–180.

1. Rennerfelt, E. (1943) Sv. Bot. Tidskr. 37, 83.2. Gabrielska, J., Sarapuk, J., and Przestalski, S. (1995) Z.

Naturforsch. 50c, 561–564.3. Fremont, L. (2000) Life Sci. 66, 663–673.4. Adams, G. E., and Michael, B. D. (1967) Trans. Faraday Soc. 63,

1171–1180.
5. Wang, D., Gyorgy, I., Hildenbrand, K., and von Sonntag, C.
(1994) J. Chem. Soc. Perkin Trans. 2, 45–55.

Documents

Efficiency and Mechanism of the Antioxidant Action of trans-Resveratrol and Its Analogues in the Radical Liposome Oxidation