Anti- and pro-oxidant effects of quercetin in copper-induced low density lipoprotein oxidation Quercetin as an effective antioxidant against pro-oxidant effects of urate Paulo Filipe 1,2 , Josiane Haigle 3 , Joa ˜ o Nuno Silva 1,2 , Joa ˜ o Freitas 1,2 , Afonso Fernandes 1 , Jean-Claude Mazie ` re 4 ,Ce ´ cile Mazie ` re 4 , Rene ´ Santus 3,5 and Patrice Morlie ` re 3,5 1 Centro de Metabolismo e Endocrinologia and 2 Clinı ´ ca Dermatolo ´ gica Universita ´ ria, Faculdade de Medicina de Lisboa, Hospital de Santa Maria, Lisbon, Portugal; 3 Muse ´ um National d’Histoire Naturelle, RDDM, Photobiologie – INSERM U.532, Paris, France; 4 Laboratoire de Biochimie, Universite ´ de Picardie Jules Verne, U Amiens, Ho ˆ pital Nord, Amiens, France; 5 Institut de Recherche sur la Peau, INSERM U.532, Ho ˆ pital Saint-Louis, Paris, France We recently reported that, depending on its concentration, urate is either a pro- or an antioxidant in Cu 2+ -induced low- density lipoprotein (LDL) oxidation. We also previously demonstrated an antioxidant synergy between urate and some flavonoids in the Cu 2+ -induced oxidation of diluted serum. As a result, the effect of the flavonoid quercetin on the Cu 2+ -induced oxidation of isolated LDL has been studied either in the presence or absence of urate. We demonstrate that, like urate, quercetin alone, at low concentration, exhibits a pro-oxidant activity. The pro-oxidant behavior depends on the Cu 2+ concentration but it is not observed at high Cu 2+ concentration. When compared with urate, the switch between the pro- and the antioxidant activities occurs at much lower quercetin concentrations. As for urate, the pro-oxidant character of quercetin is related to its ability to reduce Cu 2+ with the formation of semioxidized quercetin and Cu + with an expected yield larger than that obtained with urate owing to a more favorable redox potential. It is also shown that the pro-oxidant activity of urate can be inhibited by quercetin. An electron transfer between quercetin and semioxidized urate leading to the repair of urate could account for this observation as suggested by recently published pulse radiolysis data. It is anticipated that the interactions between quercetin–Cu 2+ –LDL and urate, which are tightly controlled by their respective concentra- tion, determine the balance between the pro- and antioxidant behaviors. Moreover, as already observed with other anti- oxidants, it is demonstrated that quercetin alone behaves as a pro-oxidant towards preoxidized LDL. Keywords: antioxidant; copper; flavonoid; low-density lipo- protein; oxidative stress; pro-oxidant; urate. It is generally accepted that diets rich in fruit and vegetables protect against cardiovascular diseases [1,2], certain types of cancer [3], and perhaps against other pathological condi- tions. This protection has been attributed to the anti- oxidants present in plants including various phenolic compounds such as flavonoids. The biological, preventive, and therapeutic properties of flavonoids have been exten- sively studied. It has recently been shown that some flavonoids can cross the intestinal barrier into the blood- stream [4,5]. However, detected levels are extremely low in relation to the ingested amounts. Low-density lipoprotein (LDL) oxidative modification is currently thought to be a key process in the atherogenesis [6–9]. The in vitro inhibitory effects of flavonoids on LDL peroxidation induced by copper(II) ions, azo derivatives and macrophages are well established. The radical scaven- ging properties of phenolic compounds depend on their redox potentials [10–12]. In addition, the antioxidant action is related to their ability to chelate transition metal ions [13–16]. The flavonoid antioxidant efficacy in biological systems also depends on the partition coefficient between the lipophilic and the aqueous phases [17,18], the binding to macromolecules [19,20], and the interaction with other antioxidants [21–28]. In most reports, the concentration of phenolic com- pounds to produce half of the maximum inhibition lies in the 1–10 l M range. With the exception of ascorbate, the antioxidant synergy between flavonoids and other extra- cellular antioxidants has not been fully explored. The interaction of flavonoids with ascorbate was first des- cribed in 1936 [29,30], when it was found that the extracts of Hungarian red pepper contained an ascorbate-protect- ive factor, named vitamin P. Later on, this factor was identified as a mixture of flavonoids, and its effect was interpreted as the result of the antioxidant action of these compounds. Correspondence to P. Morlie ` re, Muse ´ um National d’Histoire Naturelle, RDDM, Photobiologie – INSERM U.532, Case Postale 26, 43 rue Cuvier, 75231 Paris cedex 05, France. Fax: + 33 1 40793716, Tel.: + 33 1 40793884, E-mail: morliere@mnhn.fr Abbreviations: LDL, low-density lipoprotein; LPO, lipid peroxidation; MDA, malondialdehyde; MM-LDL, minimally modified low-density lipoprotein; SOD, superoxide dismutase; TBARS, thiobarbituric acid-reactive substances. (Received 30 December 2003, revised 3 March 2004, accepted 25 March 2004) Eur. J. Biochem. 271, 1991–1999 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2004.04111.x It is well known that urate, one of the most important plasma antioxidants resulting from the breakdown of purines, inhibits LDL peroxidation [31]. The mechanisms of this effect include the ability of urate to scavenge reactive oxygen and nitrogen species, and to chelate catalytic metal ions [32–34] (reviewed in [35]). However, data are lacking on the interaction of urate with other antioxidants. As a consequence we have investigated the effect of dietary flavonoids, associated or not with urate, on the oxidation of human LDL induced by copper(II) ions. This investigation appears sensible as it was recently demonstrated [36] that, at low concentration, urate may behave as a pro-oxidant. Moreover, in an earlier report it was suggested that flavonoids and urate could act in synergy to protect plasma lipoproteins against the oxidative stress [37]. The main goal of this work was therefore to evaluate whether and how urate and flavonoids interact during the Cu 2+ -induced LDL oxidation. Because of its largely recognized anti- oxidative properties, quercetin was chosen as a model flavonoid in this study. Materials and methods Materials, solvents and routine equipment Quercetin dihydrate and sodium urate were obtained from Sigma Chemical Co. (Saint-Louis, MO, USA). The high- performance liquid chromatography (HPLC) columns were supplied by Merck (Darmstadt, Germany) and HPLC grade solvents were purchased from Carlo Erba (Val de Reuil, France). All other chemicals were of the highest purity available from Sigma or Merck companies. Preparation and treatment of LDL Serum samples were obtained from healthy volunteers. LDL (d ¼ 1.024–1.050) were prepared by sequential ultra- centrifugation according to Havel et al. [38] and dialyzed against pH 7.4, 5 m M Tris buffer containing 50 m M NaCl and 0.02% EDTA. Protein determination was carried out by the technique of Peterson [39]. Unless specifically stated in the text, these LDL preparations were used within 2–3 weeks. Just before experiments, LDL were dialyzed twice for 8 and then 16 h against 1 L of pH 7.4, 10 m M phosphate buffer to remove ETDA. Then, LDL were diluted to a final concentration of 0.15 mgÆmL )1 (300 n M ). To 800 lL of these LDL solutions, were added 50 lLofa stock solution of urate in pH 7.4, 10 m M phosphate buffer and/or 10 lL of a stock solution of the studied flavonoid in pH 7.4, 10 m M phosphate buffer or ethanol. LDL solutions without urate and/or flavonoid but containing matching solvent volumes were similarly prepared. Then, all these LDL solutions were diluted to a final volume of 950 lL with phosphate buffer and were incubated at 37 °Cfor 15 min. Lipid peroxidation (LPO) was triggered by adding 50 lLofaCuCl 2 solution in pH 7.4, 10 m M phosphate buffer preheated at 37 °C to obtain a final concentration of Cu 2+ of 5 or 175 l M .AfterCu 2+ addition, the formation of conjugated dienes and malondialdehyde (MDA) and the consumption of urate and carotenoids were measured, as described below, either after a 1 h incubation at 37 °Corat various times during incubation. Conjugated diene determination Conjugated diene formation was monitored by second derivative spectroscopy (220–300 nm) based on an earlier described methodology [40]. In short, 80 lLofthesample were diluted 10-fold with pH 7.4, 10 m M phosphate buffer before spectrum recording. The second derivative spectrum was subtracted from the second derivative spectrum of the matching control sample without cop- per(II). The increase in conjugated dienes expressed in relative unit was obtained from the amplitude of the 254 nm peak. Malondialdehyde measurement The simultaneous determination of free MDA and urate was performed by HPLC using a LiChrospher100 NH 2 column [41]. After incubation, solutions were mixed with an equal volume of acetonitrile, centrifuged at 12 000 g for 5 min and frozen at )80 °C until HPLC measurement. The supernatants were isocratically eluted during 25 min with a mobile phase consisting of pH 7.4, 54 m M Tris-HCI and acetonitrile (30 : 70, v/v). The flow rate was 1.0 mLÆmin )1 and the absorption was monitored at 270 nm. The MDA peak was identified by comparison with a reference chromatogram of free MDA, freshly prepared by acid hydrolysis of 1,1,3,3-tetraethoxypropane stock solution. The MDA concentration of this standard solution was determined assuming a molar absorbance of 13 700 M )1 Æcm )1 at 245 nm. This solution was then diluted with pH 7.4, 54 m M Tris-HCI buffer to obtain MDA concentrations in the 1–10 l M range and then, mixed with acetonitrile (1 : 1, v/v) before HPLC. Consumption of carotenoids The basal carotenoid content of LDL preparations was spectrophotometrically determined after extraction [42]. To this end, 0.25 mL of water, 1 mL of ethanol and 2 mL of hexane were added to 0.25 mL of LDL. The hexane upper phase (2 mL) was collected and the visible absorption spectrum (350–600 nm) was recorded. The concentration of total carotenoids was determined using an average extinction coefficient of 140 000 M )1 Æcm )1 at 448 nm based on a calculation from the four main carotenoids in human plasma, a-carotene, b-carotene, b-cryptoxanthin and lycopene [43,44]. Change in caro- tenoid concentration during LDL oxidative treatment was monitored by second derivative absorption spectros- copy (400–550 nm) through the measure of the amplitude of the second derivative spectrum between 489 and 516 nm. Urate consumption As mentioned above, urate was determined by HPLC, simultaneously with MDA. The urate peak in chromato- grams was identified by comparison with reference chro- matograms of freshly prepared standard urate solutions in the 1–20 l M range. The concentration of urate in the samples was calculated from the peak area compared to that of standard solutions. 1992 P. Filipe et al. (Eur. J. Biochem. 271) Ó FEBS 2004 Results MDA production as a function of the quercetin concentration After incubation for 15 min at 37 °C with various concen- trations of quercetin, LDLs were exposed to 175 l M or 5 l M CuCl 2 . One hour after Cu 2+ addition, the extent of LPO was estimated from MDA measurements as shown in Fig. 1. At high Cu 2+ concentration (175 l M ),theMDA production decreases with increasing quercetin concentra- tion (Fig. 1A). It is worth noting that quercetin concentra- tions as low as 0.5 l M already significantly decreases the MDA production. At low Cu 2+ concentration (5 l M )a pro-oxidant effect is observed at low quercetin concentra- tions (< 2 l M ) whereas quercetin becomes antioxidant at higher concentration (‡ 2 l M ). The switch between the pro- and antioxidant activities occurs at  1.5 l M of quercetin. These experiments were also performed in the presence of urate. They were carried out as described above, except that LDL were simultaneously incubated with various concen- trations of quercetin and 10 l M urate prior to Cu 2+ addition. As can be seen in Fig. 1, urate alone at 10 l M is pro-oxidant, with an amplification of about 200% at high Cu 2+ concentration (175 l M ) and about 600% at low Cu 2+ concentration (5 l M ) in agreement with reference [36]. At high Cu 2+ concentration and concentrations of quercetin as low as 0.5 l M , the production of MDA drops to negligible values much lower than those obtained in the absence of both urate and quercetin. At low Cu 2+ concentration, quercetin also decreases the MDA level but the full inhibition requires larger concentrations than those needed at high Cu 2+ concentration. Interestingly, we may note that the pro-oxidant effect observed with low concentrations of quercetin alone did not add to the pro-oxidant effect of 10 l M urate. On the contrary, such low concentrations lowered the MDA production. For example, 0.75 l M quercetin or 10 l M urate alone enhances the MDA production by factors of about 250 and 600%, respectively. When added together they significantly decrease the MDA level to 60%. Data profiles similar to those shown in Fig. 1 were obtained when LPO was estimated by the production of thiobarbituric reactive substances or conjugated dienes (data not shown). Time courses of MDA and conjugated diene formation The data presented in Fig. 1 have been obtained at a constant time (1 h) after the addition of Cu 2+ . Kinetic analyses may prove to be helpful in understanding the effects observed under Ôstatic conditionsÕ. Our attention has been focused on the effect of low concentrations of quercetin. Depending on the Cu 2+ concentration this flavonoid may be pro-oxidant in the absence of urate or may inhibit the pro-oxidant effect of urate. The Cu 2+ - induced LPO in LDL was evaluated by monitoring the formation of MDA (Fig. 2A,C) and of conjugated dienes (Fig. 2B,D) in the presence and absence of 10 l M urate at high (175 l M ) (Fig. 2A,B) and low (5 l M ) (Fig. 2C,D) Cu 2+ concentrations. With regard to the MDA formation at high Cu 2+ concentration, Fig. 2A fully confirms the pro- oxidant activity of low urate concentrations with enhanced MDA formation and the moderate protection brought by 0.5 l M quercetin at short times after Cu 2+ addition. In the presence of both urate and quercetin, the MDA formation is strongly slowed down and occurs about 90 min after Cu 2+ addition. Similar effects are observed with the formation of conjugated dienes (Fig. 2B). Thus, the time course of their formation (Fig. 2B,D) exhibits the classical shape charac- terized by a lag time followed by a linear increase until a maximum is reached. Then they slowly decay [45]. When separately added, 10 l M urate decreases the lag-time whereas 0.5 l M quercetin slightly increases it. On the other hand, when simultaneously added, quercetin and urate have a strong antioxidant activity with a lag-time for conjugated diene formation lasting for about 2 h. At low Cu 2+ concentration, the kinetic data in Fig. 2C are in agreement with those obtained under Ôstatic condi- tionsÕ in Fig. 1A; that is to say 10 l M urate strongly enhances MDA production whereas 0.5 l M quercetin also increases it but to a lesser extent. Higher quercetin concentrations, namely 2 l M , inhibit the MDA formation Fig. 1. Effect of quercetin on LDL oxidation induced by 175 l M (A) or 5 l M Cu 2+ (B) in the absence or in the presence of 10 l M urate. In (A) and (B), LDL at 0.12 mgÆmL )1 (240 n M ) in pH 7.4, 10 m M phosphate buffer were incubated for 15 min at 37 °C with various concentrations of quercetin, with or without urate. Then, 175 l M (A) or 5 l M (B) of CuCl 2 were added and the mixture was further incubated at 37 °Cfor1hbeforeMDA assay. In (A) and (B), controls in the absence of Cu 2+ yielded non-detectable or negligible levels of MDA (data not shown). Data were normalized to those obtained in the absence of quercetin and urate (taken as 100%) and are the mean ± SD of at least three experiments performed with independent LDL preparations. Ó FEBS 2004 Flavonoids and urate in copper-induced low-density lipoprotein oxidation (Eur. J. Biochem. 271) 1993 up to about 2 h. Finally, as shown in Fig. 2(C), slightly pro-oxidant concentrations of quercetin (0.5 l M ) protect from the pro-oxidant effect of 10 l M urate. Larger quercetin concentrations (e.g. 2 l M ) completely inhibit the pro-oxidant effect of 10 l M urate as no MDA is formed during 3 h. The same conclusions can be drawn with regard to the formation of conjugated dienes at low Cu 2+ concentrations (Fig. 2D). Again, when separately added, 10 l M urate shortens the lag time, in agreement with a pro-oxidant behavior whereas 0.5 l M quercetin exhibits a moderate pro- oxidant behavior. When added together 0.5 l M quercetin partly inhibits the pro-oxidant effect of 10 l M urate. At higher concentration (2 l M ), quercetin strongly delays the appearance of the propagation phase when alone but completely abolishes the pro-oxidant effect of 10 l M urate. Time courses of carotenoids and urate consumption The consumption of carotenoids was measured as an index of the overall consumption of the endogenous antioxidants of LDL during their Cu 2+ -induced oxidation. As shown in Fig. 3(A,C), carotenoids are consumed in parallel to MDA and conjugated diene formation. As already reported [36], at low Cu 2+ concentration, 10 l M urate accelerates the carotenoid consumption (Fig. 3C). At high Cu 2+ concen- tration, the consumption of carotenoids is practically unchanged by the presence of this pro-oxidant concentra- tion of urate (Fig. 3A). As observed with the formation of conjugated dienes (Fig. 2B) or of MDA (Fig. 1A), 0.5 l M quercetin behaves as an antioxidant and induces a delay in the carotenoid consumption (Fig. 3A). At low Cu 2+ concentration, 0.5 l M quercetin becomes pro-oxidant with enhanced formation of MDA and conju- gated diene, as presented above. Accordingly no delay in the carotenoid consumption is observed (Fig. 3C). However, the protective effect of quercetin is recovered at higher concentrations, namely 2 l M (Fig. 3C). In the presence of a pro-oxidant concentration of urate (e.g. 10 l M ),0.5l M quercetin, which is pro-oxidant in the absence of urate, markedly slows down the carotenoid consumption (Fig. 3C). Increasing the quercetin concentration to 2 l M leads to strong protection of the carotenoids, this protection being even better than with quercetin alone (Fig. 3C). Interestingly, at this low Cu 2+ concentration, urate, which is rapidly consumed when added alone, is slightly protected by 0.5 l M quercetin and fully protected by 2 l M quercetin (Fig. 3D). At high Cu 2+ concentration, 0.5 l M quercetin delays the carotenoid consumption more efficiently that quercetin alone (Fig. 3A) and also protects urate (Fig. 3B). Effect of quercetin on the copper-induced lipid peroxidation in preoxidized LDL In order to evaluate the effect of quercetin on the Cu 2+ - induced LPO in pre-oxidized LDL, LDL were first incuba- tedwithCu 2+ and then quercetin was added after LPO started. Quercetin concentrations were chosen such as Fig. 2. Kinetic profiles of MDA (A,C) and conjugated diene (B,D) formation in LDL oxidation induced by 175 l M (A,B) or 5 l M (C,D) Cu 2+ . LDL at 0.12 mgÆmL )1 (240 n M ) in pH 7.4, 10 m M phosphate buffer was incubated for 15 min at 37 °C with or without quercetin, and with or without 10 l M urate. Then, 175 l M or 5 l M of CuCl 2 were added and the mixture was further incubated at 37 °C. MDA: Conjugated dienes were measured at various intervals after Cu 2+ addition. Note that time zero corresponds to the shortest time after addition of Cu 2+ inallsamples,e.g.1min. Control experiments in the absence of Cu 2+ yielded non detectable or negligible levels of MDA and of conjugated dienes (data not shown). Data are the mean ± SD of at least three experiments performed with independent LDL preparations. 1994 P. Filipe et al. (Eur. J. Biochem. 271) Ó FEBS 2004 quercetin behaves as an antioxidant in un-oxidized LDL, i.e. 0.75 l M and 2 l M for high and low Cu 2+ concentra- tions, respectively. By contrast, when quercetin is added 30 min after the oxidation started, LPO was higher than that obtained in the absence of quercetin (Fig. 4). In other words, under these conditions quercetin becomes pro- oxidant. As a second model of slightly preoxidized LDL, we used LDLs that were kept in the dark at 4 °C in the presence of EDTA for 5–8 weeks. Such conditions are described in the literature as yielding the so-called minimally modified LDL (MM-LDL) [46,47]. As can be seen in Table 1, in the presence of quercetin, lag times for conjugated diene formation are shorter than those measured in its absence. In addition, quercetin accelerates the consumption of carotenoids in MM-LDL treated with Cu 2+ (Table 1). This definitely means that under these conditions quercetin is no longer an antioxidant but behaves as a pro-oxidant. Discussion Over the 15 past years, the oxidation of LDL has been widely studied in order to understand better the role of LDL oxidation in vivo in pathological situations, such as ather- ogenesis [6–9]. For this purpose, in vitro models have been developed including the oxidation of LDL by Cu 2+ to which much attention has been paid [48]. However, the exact mechanisms relating Cu 2+ redox change to the LPO in LDL are not yet clearly established [49]. In the presence of pre-existing traces of hydroperoxides (LOOH), Cu 2+ and Cu + may be either reduced or oxidized, generating hydroperoxyl (LOO • ) or alkoxyl (LO • ) radicals that in turn can trigger the LPO propagation phase. It is currently admitted that Cu 2+ reduction to Cu + is required for triggering LPO in LDL [50], but the exact nature of the Fig. 3. Kinetic profiles of carotenoid (A,C) and urate (B,D) consumption in LDL oxidation induced by 175 l M (A,B) or 5 l M (C,D) Cu 2+ . Experi- mental conditions are those of Fig. 2. Urate and carotenoids were measured at various intervals after Cu 2+ addition. Control experiments in the absence of Cu 2+ (·), yielded no significant loss of carotenoids or urate. Note that time zero in (B) and (D) corresponds to the shortest time after addition of Cu 2+ in all samples, e.g. 1 min. Data are expressed as a percentage of the value obtained before Cu 2+ addition and are the mean ± SD of at least three experiments performed with independent LDL preparations. Fig. 4. Effect of quercetin on Cu 2+ -induced LDL oxidation in preoxi- dized LDL. CuCl 2 (175 or 5 l M ) was added to LDL solutions at 0.12 mgÆmL )1 (240 n M ) in pH 7.4, 10 m M phosphate buffer preheated for 15 min at 37 °C, and the mixture was further incubated at 37 °C for 1 h before MDA assay. Quercetin was added either 15 min before Cu 2+ addition or 30 min after Cu 2+ addition. Quercetin concentra- tionswere0.75or2 l M for Cu 2+ concentrations equal to 175 or 5 l M , respectively (see text). Data were normalized to those obtained in the absence of quercetin (taken as 100%) and are the mean ± SD of at three experiments performed with independent LDL preparations. Ó FEBS 2004 Flavonoids and urate in copper-induced low-density lipoprotein oxidation (Eur. J. Biochem. 271) 1995 involved reductants in LDL, such as pre-existing LOOH, tryptophan residues or even a-tocopherol, is still questioned [51–59]. Their progressive involvement has been suggested by Perugini et al. [57]. It has also been suggested but never demonstrated that • O 2 – , generated by Cu 2+ oxidation may be involved [60]. Our main goal was to investigate the interplay between urate and flavonoids in the Cu 2+ -induced oxidation of LDL. Such a work was undertaken because we recently reported, in mimicking oxidative stress in diluted plasma [37], that flavonoids could act in synergy with urate, one of the major plasma antioxidant. Antioxidant properties of flavonoids have been widely studied within the last 20 past years. A peculiar attention has been paid to the antioxidant effect of flavonoids towards the oxidation of LDL whose oxidative modification is thought to be a key process in atherogenesis. The antioxidant protection conferred by phenolic compounds is believed to be caused by a combi- nation of their binding to critical sites on LDL, their metal chelation properties and their free radical scavenging activities. Their overall antioxidant efficacy in biological systems also depends on their partition between lipophilic and aqueous phases, their binding to other biomolecules and their interaction with other antioxidants. Quercetin was chosen as a representative flavonoid in this study as it is rather universally found in plants and it has been the subject of numerous studies. As a result, we first characterized its antioxidant properties in our LDL oxidation model. It must be pointed out that besides the commonly used low Cu 2+ concentration (5 l M here), we also performed experiments with a physiologically unrealistic high Cu 2+ concentration (175 l M ). Such a high Cu 2+ concentration was used to overcome the chelating ability of quercetin. Moreover the use of both low and high Cu 2+ concentrations allows discussing the present data in the light of our earlier reports [36,37]. It must be underlined that the quercetin concentra- tions that were used here (0.25–2 l M ) are biologically realistic. Indeed, it has been shown that the concentration of quercetin derivatives in plasma reached about 0.4 [4], 0.6 [5] and 0.8 l M [61], 2–3 h after the ingestion of a quercetin-rich meal (about 50–90 mg of ingested quercetin). These values are about an order of magnitude higher than the baseline concentration. In addition to their well established antioxidant proper- ties, flavonoids, as several other antioxidants, to be pro- oxidant under certain circumstances. In the Cu 2+ -induced LDL oxidation model, catechins [62] and a flavonoid extract [63] share this pro-oxidant ability with caffeic acid [64], ascorbate [63] or urate [36,65,66]. This pro-oxidant behavior was described when the antioxidant is added at various times after LPO started, i.e. when some lipid peroxides are already formed. Accordingly, this pro-oxidant behavior is also encountered in Cu 2+ -induced oxidation of slightly oxidized LDL. We observed that quercetin also exhibits this pro-oxidant capacity. Indeed, the production of an exacerbated amount of MDA (Fig. 4) demonstrates that quercetin is antioxidant when present before LPO started but becomes pro-oxidant when introduced 30 min after addition of Cu 2+ . It is also illustrated in Table 1 reporting that, as opposed to native LDL, quercetin accelerates the formation of conjugated dienes (shortened lag time) and the consumption of endogenous carotenoids (shortened half time) in MM–LDL. It is accepted that this pro-oxidant behavior is related to the availability of hydroperoxides and that ÔantioxidantsÕ promote LPO by increasing the concen- tration of catalytic Cu + because of their ability to reduce Cu 2+ . More importantly, our data demonstrate that a pro- oxidant behavior of quercetin can be observed in the absence of pre-existing hydroperoxides, i.e. working with native LDL with quercetin present before LPO started. The presence of abnormally high levels of hydroperoxides in our LDL preparations, that could explain the data reported above, has been ruled out as discussed in [36,] This is observed at low Cu 2+ concentration and it is illustrated on Figs 1 and 2C in terms of MDA production and on Fig. 2D in terms of conjugated diene formation. Interestingly, this pro-oxidant behavior is observed at low concentration of ÔantioxidantÕ. In view of the MDA formation profile on Fig. 1, the switch from the pro-oxidant behavior to the classical antioxidant properties occurs with 1–2 l M querce- tin. This is confirmed with the conjugated diene formation, which is slightly promoted by 0.5 l M quercetin but strongly abolished by 2 l M quercetin (Fig. 2D). As to the consump- tion of endogenous carotenoids, the pro-oxidant behavior is still possibly observed at low quercetin concentration (0.5 l M ) while 2 l M quercetin definitely induces a strong protective effect. We recently reported such a behavior with another well-known physiological antioxidant, e.g. urate [36]. We suggested that this pro-oxidant behavior was related to the ability of urate to reduce Cu 2+ , leading to • UH – and Cu + . We brought evidence that, at high Cu 2+ concentration, • O 2 – was involved probably formed by the oxidation of Cu + by O 2 . We suggested that the concomitant formation of • UH – and • O 2 – could allow a reaction between these species, thus leading to some kind of • O 2 – activation. Table 1. Effect of quercetin on the lag time for conjugated diene formation and on the half time for carotenoid consumption in Cu 2+ -treated MM-LDL. Data in parentheses correspond to those obtained with native LDL. Detailed experimental conditions are those of Figs 2 and 3. Quercetin concentrations were 0.75 and 2 l M for Cu 2+ equal to 175 and 5 l M , respectively. Lag times before conjugated diene formation were estimated as the intercept of the linear part of the kinetics of diene formation shown in Fig. 2(B,D) with x-axis. Half times for carotenoid consumption were obtained from the kinetic of carotenoid consumption shown in Fig. 3(A,C). Conditions Lag time (min) Half time (min) Cu 2+ ¼ 175 l M Cu 2+ ¼ 5 l M Cu 2+ ¼ 175 l M Cu 2+ ¼ 5 l M Without quercetin  5( 24)  17 ( 33)  18 ( 27)  24 ( 45) With quercetin ( 39) a  6 (> 120)  14 ( 50)  12 (> 150) a Too short to be measured. 1996 P. Filipe et al. (Eur. J. Biochem. 271) Ó FEBS 2004 At low Cu 2+ concentration, the exact mechanism was not elucidated, but an • O 2 – -independent mechanism was pro- posed, still involving the reduction of Cu 2+ by urate. This intriguing behavior has to be related to complex urate– Cu 2+ –LDL interactions, which are governed by their respective concentrations. In the view of the analogous behaviors observed here with quercetin and previously with urate, it may be hypothesized that similar mechanisms are involved. We already provided evidence that quercetin was able to reduce Cu 2+ [37]. When comparing the present data with those obtained with urate, it is interesting to note that at low Cu 2+ concentration the switch between the pro- oxidant and antioxidant behavior occurs at about 1 l M with quercetin and 200 l M with urate [36]. As we pointed out in our earlier report dealing with urate, the pro-oxidant mechanism observed at low concentration is still operative at high concentration but is no longer observed because, at high concentration, the antioxidant activity (chelation, radical scavenging) prevails. Thus, the lower concentration for switching between pro- and antioxidant properties for quercetin as compared to urate may reflect the better overall antioxidant activity of quercetin. Conversely, it may also reflect a pro-oxidant ability larger for quercetin than for urate, thus requiring lower concentrations to be observable. As a matter of fact, at pH 7, the standard redox potential of the couple • QH – ,H + /QH 2 – (0.33 vs. NHE [12,67]) is higher than that of the couple • UH – ,H + /UH 2 – (0.59 vs. NHE [34]). The reduction of Cu 2+ to Cu + (E° ¼ 0.167 vs. NHE) is thermodynamically unfavorable but less unfavorable for quercetin than for urate. Thus the production of equal amounts of catalytic Cu + from urate or quercetin will require lower quercetin concentrations. Finally, as mentioned in the introduction, we studied the interaction of urate and quercetin in the Cu 2+ -induced LDL oxidation model. As stated above, urate at moder- ately low concentrations behaves as a pro-oxidant at low and high Cu 2+ concentrations. This is illustrated with 10 l M urate in Figs 1 and 2(B,C) showing an overpro- duction of MDA. It is also clearly shown with an accelerated conjugated diene formation (Fig. 2B,D) and carotenoid consumption especially at low Cu 2+ concen- tration (Fig. 3C). At high Cu 2+ concentration, the addi- tion of 0.5 l M quercetin, which, alone, is moderately antioxidant, not only inhibits the pro-oxidant effect of urate, but exerts an overall protective effect on the LDL oxidation larger than that obtained with quercetin alone. Under these conditions, the urate destruction is slowed down by the presence of quercetin (Fig. 3C). At low Cu 2+ concentration, a somewhat similar behavior is observed. Thus, 0.5 l M quercetin which, in this case, is pro-oxidant when alone, partly inhibits the pro-oxidant action of urate (Figs 1B, 2C,D and 3C). Larger quercetin concentrations, namely 2 l M , completely inhibit the pro-oxidant action of urate leading to full overall protection. It is interesting to mention that 2 l M quercetin inhibits the urate consump- tion up to 3 h after Cu 2+ addition (Fig. 3D). This powerful antioxidant interaction of quercetin and urate can be partially explained by the interception of reactive species, chelation of transition metal ions and/or regener- ation of urate from its radical form. The recycling of urate is possible as at pH 7, the redox potential of the • UH – , H + /UH 2 – couple (0.59 V vs. NHE) is higher than that of the • QH – ,H + /QH 2 – couple (0.33 V vs. NHE) [12,34,67] and allows the reaction of quercetin with • UH – .Using pulse radiolysis, we recently provided good evidence for this reaction [68]. We showed, upon addition of quercetin, an increase in the decay rate of the transient absorption of • UH – accompanied by a growth of the transient absorp- tion of the semireduced quercetin ( • QH – ) demonstrating the regeneration of urate by quercetin. At high Cu 2+ concentration, where • O 2 – is believed to be involved in the pro-oxidant action of urate [37], the reaction of • UH – with quercetin competes with its reaction with • O 2 – and impedes the pro-oxidant activity of urate to occur. At low Cu 2+ concentration, another unidentified mechanism was sug- gested where • O 2 – is not involved, but where Cu + is necessary [37]. As a consequence of the reaction of quercetin with • UH – , it turns out that • UH – wouldalsobe involved in the pro-oxidant action of urate. Interestingly, at high Cu 2+ concentration, in the presence of 0.5 l M quercetin and 10 l M urate, the LDL oxidation measured in terms of MDA and conjugated diene formation drops to a level below that observed in the absence of urate (Figs 1A and 2A,B). The same concept applies to the carotenoid consumption shown in Fig. 3A. Though we have no definite interpretation for such an observation, we may suppose that in the absence of quercetin the pro- oxidant activity of urate masks its antioxidant activity as mentioned above, whereas, in the presence of quercetin, the pro-oxidant activity of urate disappears and its antioxidant activity is revealed. Conclusions The results presented here on the Cu 2+ -induced oxida- tion of LDL show that quercetin, as other oxidants, can be pro-oxidant towards slightly oxidized LDL and, at low concentration, can behave like urate as a pro-oxidant towards native LDL. This study also demonstrates that quercetin at low and pro-oxidant concentrations is no longer pro-oxidant in the presence of pro-oxidant urate concentrations. Under these conditions, it even protects against the pro-oxidant activity of urate. As to the mechanism, it is suggested that quercetin, because of its appropriate redox potential can regenerate urate by reducing the • UH – radical. Accordingly, quercetin inhibits the pro-oxidant activity of urate associated with the presence of the • UH – radical. However, some points still remain un-understandable. Indeed, during the • UH – repair, • QH – would be generated, which may be associ- ated with the pro-oxidant activity of quercetin at low concentration. It is quite obvious that a detailed know- ledge of these mechanisms deserves further investigations. It may be anticipated that the intricate relationship between antioxidant and pro-oxidant action of antioxi- dants, such as quercetin or urate, is closely related to the spatial and temporal interactions of the various actors, as emphasized by our recent study on the repair of semioxidized urate by quercetin bound to HSA [69]. In other words, the interactions between quercetin, Cu 2+ , LDL, and urate when present, which are closely controlled by their respective concentrations, may be determinant in the balance between the pro- and antioxidant behaviors. Ó FEBS 2004 Flavonoids and urate in copper-induced low-density lipoprotein oxidation (Eur. J. Biochem. 271) 1997 Acknowledgements This work was partly supported by an exchange grant from INSERM and GRICES. C. M. and J C. M. thank the ÔUniversite ´ de Picardie Jules VerneÕ and the ÔMiniste ` re de la Recherche et de la TechnologieÕ for financial support. References 1. Hertog, M.G., Feskens, E.J., Hollman, P.C., Katan, M.B. & Kromhout, D. (1993) Dietary antioxidant flavonoids and risk of coronary heart disease: the Zutphen Elderly Study. Lancet 342, 1007–1011. 2. Knekt, P., Jarvinen, R., Reunanen, A. & Maatela, J. (1996) Fla- vonoid intake and coronary mortality in Finland: a cohort study. Br.Med.J.312, 478–481. 3. Kuo, S M. (1997) Dietary flavonoid and cancer prevention: evi- sence and potential mechanism. Crit.Rev.Oncog.8, 47–69. 4. Manach, C., Morand, C., Crespy, V., Demigne ´ ,C.,Texier, Re ´ ge ´ rat, F. & Re ´ me ´ sy, C. (1998) Quercetin is recovered in human plasma as conjugated derivatives which retain antioxidant prop- erties. FEBS Lett. 426, 331–336. 5. Hollman, P.C.H., Gaag, M.V.D., Mengelers, M.J.B., van Trijp, J.M.P., de Vries, J.H.M. & Katan, M.B. (1996) Absorption and disposition kinetics of the dietary antioxidant quercetin in man. Free Radic. Biol. Med. 21, 703–707. 6. Parthasarathy,S.,Printz,D.J.,Bloyd,D.,Joy,L.&Steinberg,D. (1986) Macrophage oxidation of low density lipoprotein generates a modified form recognized by the scavenger receptor. Arterio- sclerosis 6, 505–510. 7. Steinbrecher, U.P., Lougheed, M., Kwan, W.C. & Dirks, M. (1989) Recognition of oxidized low density lipoprotein by the scavenger receptor of macrophages results from derivatization of apolipoprotein B by products of fatty acid peroxidation. J. Biol. Chem. 264, 15216–15223. 8. Steinberg, D., Parthasarathy, S., Caew, T.E., Khoo, J.C. & Wit- zum, J.L. (1989) Beyond cholesterol: modifications of low-density lipoprotein that increase its atherogenicity. N.Engl.J.Med.320, 915–924. 9. Witzum, J.L. (1994) The oxidation hypothesis of atherosclerosis. Lancet 344, 793–795. 10. Bors, W., Michel, C. & Saran, M. (1994) Flavonoid antioxidants: rate constants for reactions with oxygen radicals. In Methods in Enzymology: Oxygen Radicals in Biological Systems (Packer, L., ed.), Part D, pp. 420–429. Academic Press, San Diego, CA. 11. Jovanovic, S., Steenken, S., Rosic, M., Marjanovic, B. & Simic, M. (1994) Flavonoids as antioxidants. J. Am. Chem. Soc. 116, 4846–4851. 12. Jovanovic,S.,Steenken,S.,Simic,M.&Hara,Y.(1998)Anti- oxidant properties of flavonoids: reduction potentials and electronic transfer reactions of flavonoid radicals. In Flavonoids in Health and Disease (Rice-Evans, C. & Packer, L., eds), pp. 137– 161. Marcel Dekker, New York. 13. Morel, G., Lescoat, P., Cogrel, O., Sergent, O., Pasdeloup, N., Brissot, P. & Cillard, J. (1993) Antioxidant and iron-chelating activities of the flavonoids catechin, quercetin and diosmetin on iron-loaded rat hepatocyte cultures. Biochem. Pharmacol. 45, 13–19. 14.Moran,J.,Klucas,R.,Grayer,R.,Abian,R.J.&Becana,M. (1997) Complexes of iron with phenolic compounds from soybean nodules and other legume tissues: prooxidant and antioxidant properties. Free Radic. Biol. Med. 22, 861–870. 15. Brown, J., Khodr, R., Rider, R.C. & Rice-Evans, C.A. (1998) Structural dependence of flavonoid interactions with Cu 2+ ions: implications for their antioxidant properties. Biochem. J. 330, 1173–1178. 16. Yoshino, M. & Murakami, K. (1998) Interaction of iron with polyphenolic compounds: application to antioxidant character- ization. Anal. Biochem. 257, 40–44. 17. Chimi, R., Cillard, J., Cillard, P. & Rahmani, M. (1991) Peroxyl and hydroxyl radical scavenging activity of some natural phenolic antioxidants. J. Am. Oil Chemists Soc. 68, 307–312. 18. Salah, N., Miller, N., Paganga, G., Tijburg, L., Bolwell, G. & Rice-Evans, C. (1995) Polyphenolic flavanols as scavengers of aqueous phase radicals and as chain-breaking antioxidants. Arch. Biochem. Biophys. 332, 339–346. 19. Tikkanen, M., Wa ¨ ha ¨ la ¨ , K., Ojala, S., Vihma, V. & Adlercreutz, H. (1998) Effect of soybean phytoestrogen intake on LDL oxidation resistance. Proc. Natl Acad. Sci. USA 95, 3106–3110. 20. Boulton, D., Walle, U.K. & Walle, T. (1998) Extensive biding of the bioflavonoid quercetin to human plasma proteins. J. Pharm. Pharmacol. 50, 243–249. 21. Clemetson, C. & Andersen, L. (1966) Plant polyphenols as anti- oxidants for ascorbic acid. Ann. NY Acad. Sci. 136, 339–378. 22. Harper, K.A., Morton, A.D. & Rolfe, E.J. (1969) The phenolic compounds of black currant juice and their protective effect on ascorbic acid. III. The mechanism of ascorbic acid oxidation and its inhibition by flavonoids. J. Food Technol. 4, 255–267. 23. Sorata, Y., Takahama, U. & Kimura, M. (1988) Cooperation of quercetin with ascorbate in the protection of photosensitized lysis of human erythrocytes in the presence of hematoporphyrin. Photochem. Photobiol. 48, 195–199. 24. Hodnick, W.F., Milosavlievic, E.B., Nelson, J. & Pardini, R. (1988) Electrochemistry of flavonoids: relationships between redox potential, inhibition of mitochondrial respiration, and production of oxygen radicals by flavonoids. Biochem. Pharmacol. 37, 2607– 2611. 25. Negre-Salvayre, A., Affany, A., Hariton, C. & Salvayre, R. (1991) Additional antilipoperoxidant activities of alpha-tocopherol and ascorbic acid on membrane like systems are potentiated by rutin. Pharmacology 42, 262–272. 26. Kandaswami, C., Perkins, E., Soloniuk, D.S., Drzewiecki, G. & Middleton, E. (1993) Ascorbic acid-enhanced antiproliferative effect of flavonoids on squamous cell carcinoma in vitro. Anti- Cancer Drug 4, 91–96. 27. Laranjinha, J., Vieira, O., Madeira, V. & Almeida, L. (1995) Two related phenolic antioxidants with opposite effects on vitamin E content in low density lipoproteins oxidized by ferrylmyoglobin: consumption vs regeneration. Arch. Biochem. Biophys. 323, 373–381. 28. Lotito, S. & Fraga, C. (1998) (+)-Catechin prevents human plasma oxidation. Free Radic. Biol. Med. 24, 435–441. 29. Rusznyak, S. & Szent-Gyo ¨ rgyi, A. (1936) Vitamin P: flavonols as vitamins. Nature 138,27. 30. Bentsath, A., Rusznyak, S. & Szent-Gyo ¨ rgyi, A. (1936) Vitamin nature of flavones. Nature 138, 798. 31. Esterbauer, H., Gebicki, J., Puhl, H. & Ju ¨ rgens, G. (1992) The role oflipid peroxidation and antioxidants in oxidative modification of LDL. Free Radic. Biol. Med. 13, 341–390. 32. Davies, K.J.A., Sevanian, A., Muakkassah-Kelly, S.F. & Hoch- stein, P. (1986) Uric acid iron ion complexes: a new aspect of the antioxidant functions of uric acid. Biochem. J. 235, 747–754. 33. Buxton, G., Greenstock, C.L., Helman, W. & Ross, A.B. (1988) Critical review of rate constants for reactions of hydrated elec- trons, hydrogen atoms and hydroxyl radicals (ÆOH/ÆO – ) in aque- ous solution. J. Phys. Chem. Ref. Data 17, 513–886. 34. Simic, M.G. & Jovanovic, S.V. (1989) Antioxidation mechanisms of uric acid. J. Am. Chem. Soc. 111, 5778–5782. 35. Becker, B.F. (1993) Towards the physiological function of uric acid. Free Radic. Biol. Med. 14, 615–631. 36. Filipe, P., Haigle, J., Freitas, J.P., Fernandes, A., Mazie ` re, J C., Mazie ` re, C., Santus, R. & Morlie ` re, P. (2002) Anti- and 1998 P. Filipe et al. (Eur. J. Biochem. 271) Ó FEBS 2004 pro-oxidant effects of urate in copper-induced low-density lipoprotein oxidation. Eur. J. Biochem. 269, 5474–5483. 37. Filipe, P., Lanc¸ a, V., Silva, J.N., Morlie ` re, P., Santus, R. & Fernandes, A. (2001) Flavonoids and urate antioxidant interplay in plasma oxidative stress. Mol. Cell. Biochem. 221, 79–87. 38. Havel, R.J., Eder, H.A. & Bragdon, J.H. (1955) The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum. J. Clin. Invest. 34, 1345–1353. 39. Peterson, G.L. (1977) Simplification of the protein assay method of Lowry et al. which is more generally applicable. Anal. Biochem. 83, 346–356. 40. Corongiu, F.P., Banni, S. & Dessi, M.A. (1989) Conjugated dienes detected in tissue lipid extracts by second derivative spectro- photometry. Free Radic. Biol. Med. 7, 183–186. 41. Esterbauer, H., Lang, J., Zadravec, S. & Slater, T.F. (1984) Detection of malonaldehyde by high performance liquid chromatography. In Methods in Enzymology: Oxygen Radicals in Biological Systems (Packer, L., ed.), pp. 319–328. Academic Press, Orlando, FL. 42. Behrens, W.A., Thompson, J.N. & Made ` re, R. (1982) Distribu- tion of alpha-tocopherol in human plasma lipoproteins. Am. J. Clin. Nutr. 35, 691–696. 43. Thurnham, D.I., Smith, E. & Flora, P.S. (1988) Concurrent liquid- chromatography assay of retinol, a-tocopherol, b-carotene, a-carotene, lycopene and b-cryptoxanthin in plasma, with toco- pherol acetate as internal standard. Clin. Chem. 34, 377–381. 44. Windholz, M. (1976) The Merck Index. An Encyclopedia of Chemicals and Drugs, 9 edn. Merck, Inc, Rahway, NJ. 45. Patel, R.P. & Darley-Usmar, V.M. (1999) Molecular mechanisms of the copper dependent oxidation of low-density lipoprotein. Free Rad. Res. 30,1–9. 46. Berliner, J.A., Territo, M.C., Sevanian, A., Ramin, S., Kim, J.A., Bamshad, B., Esterson, M. & Fogelman, A.M. (1990) Minimally modified low density lipoproteins stimulates monocyte endothelial interactions. J. Clin. Invest. 85, 1260–1266. 47. Carrero, P., Ortega, H., Martı ´ nez-Botas, J., Go ´ mez-Coronado, D. &Laruscio ´ n, M.A. (1998) Flavonoid-induced ability of minimally modified low-density lipoproteins to support lymphocyte pro- liferation. Biochem. Pharmacol. 55, 1125–1129. 48. Rice-Evans, C., Leake, D., Bruckdorfer, R. & Diplock, A.T. (1996) Practical approaches to low density lipoprotein oxidation: whys, wherefores and pittfalls. Free Radic. Res. 25, 285–311. 49. Halliwell, B. (1995) Oxidation of low density lipoproteins: ques- tions of initiation, propagation and the effect of antioxidants. Am. J. Clin. Nutr. 61, 670S–677S. 50. Esterbauer, H. & Jurgens, G. (1993) Mechanistic and genetic aspects of susceptibility of LDL to oxidation. Curr. Opin. Lipid. 4, 114–124. 51. Proudfoot, J.M., Croft, K.D., Puddey, I.B. & Belli, L.J. (1997) The role of copper reduction by a-tocopherol in low-density lipoprotein oxidation. Free Radic. Biol. Med. 23, 720–728. 52. Perugini, C., Seccia, M., Albano, E. & Bellomo, G. (1997) The dynamic reduction of Cu (II) to Cu (I) and not Cu (I) availability is a sufficient trigger for low density liporpotein oxidation. Biochim. Biophys. Acta 1347, 191–198. 53. Giessauf, A., Steiner, E. & Esterbauer, H. (1995) Early destruction of tryptophan residues of apolipoprotein B is a vitamin E-independent process during copper-mediated oxidation of LDL. Biochim. Biophys. Acta 1256, 221–232. 54. Maiorino, M., Zamburlini, A., Roveri, A. & Ursini, F. (1995) Copper-induced lipid peroxidation in liposomes, micelles and LDL: which is the role of vitamin E? Free Radic. Biol. Med. 18, 67–74. 55. Yoshida, Y., Tsuchida, J. & Niki, E. (1995) Interaction of a-tocopherol with copper and its effects on lipid peroxidation. Biochim. Biophys. Acta 1200, 85–92. 56. Kontush, A., Meyer, S., Finck, B., Kohlschutter, A. & Beisiegel, U. (1996) a-Tocopherol as a reductant for Cu (II) in human lipoproteins. Triggering role of in the initiation of lipid peroxida- tion. J. Biol. Chem. 271, 11106–11112. 57. Perugini, C., Seccia, M., Bagnati, M., Cau, C., Albano, E. & Bellomo, G. (1998) Different mechanisms are progressively recruited to promote Cu (II) reduction by isolated human low density lipoprotein undergoing oxidation. Free Radic. Biol. Med. 25, 519–528. 58. Lynch, S.M. & Frei, B. (1995) Reduction of copper not iron, by human low density lipoprotein. J. Biol. Chem. 270, 5158–5163. 59. Patel, R., Svistunenko, D., Wilson, M.T. & Darley-Usmar, V. (1997) Reduction of Cu (II) by lipid hydroperoxides: implications for the copper-induced oxidation of low-density lipoprotein. Biochem. J. 322, 425–433. 60. Pecci, L., Montefoschi, G. & Cavallini, D. (1997) Some new details on the copper–hydrogen peroxide interaction. Biochem. Biophys. Res. Commun. 235, 264–267. 61. McAnlis, G.T., McEneny, J., Pearce, J. & Young, I.S. (1999) Absorption and antioxidant effects of quercetin from onions, in man. Eur. J. Clin. Nutrit. 53, 92–96. 62. Yamanaka, N., Oda, O. & Nagao, S. (1997) Green tea catechins suh a(–)-epicatechin and (–)-epigallocatechin accelerate Cu 2+ - induced low density lipoprotein in propagation phase. FEBS Lett. 401, 230–234. 63. Otero, P., Viana, M., Herrera, E. & Bonet, B. (1997) Antioxidant and prooxidant effects of ascorbic acid, dehydroascorbic acid and flavonoids on LDL submitted to different degrees of oxidation. Free Radic. Res. 27, 619–626. 64. Yamanaka, N., Oda, O. & Nagao, S. (1997) Prooxidant activity of caffeic acid, dietary non-flavonoid phenolic antioxidant, on Cu 2+ -induced low density lipoprotein oxidation. FEBS Lett. 405, 186–190. 65. Abuja, P.M. (1999) Ascorbate prevents prooxidant effects of urate in oxidation of human low density lipoprotein. FEBS Lett. 446, 305–308. 66. Bagnati, M., Perugini, C., Cau, C., Bordone, R., Albano, E. & Bellomo, G. (1999) When and why a water-soluble antioxidant becomes pro-oxidant during copper-induced low-density lipoprotein oxidation: a study using uric acid. Biochem. J. 340, 143–152. 67. Jovanovic, S.V., Steenken, S., Hara, Y. & Simic, M.G. (1996) Reduction potentials of flavonoid and model phenoxy radicals: which ring in flavonoids is responsible for antioxidant activity? J. Chem. Soc. Perkin Trans. 2, 2797–2504. 68. Santus, R., Patterson, L.K., Filipe, P., Morlie ` re, P., Hug, G.L., Fernandes, A. & Mazie ` re, J C. (2001) Redox reactions of the urate radical/urate couple with the superoxide radical anion, the tryptophan neutral radical and selected flavonoids in neutral aqueous solutions. Free Radic. Res. 35, 129–136. 69. Filipe, P., Morlie ` re,P.,Patterson,L.K.,Hug,G.L.,Mazie ` re, J C., Freitas, J.P., Fernandes, A. & Santus, R. (2004) Oxygen–copper (II) interplay in the repair of semi-oxidized urate by quercetin bound to human serum albumin. Free Rad. Res. 38, 295–301. Ó FEBS 2004 Flavonoids and urate in copper-induced low-density lipoprotein oxidation (Eur. J. Biochem. 271) 1999 . Anti- and pro-oxidant effects of quercetin in copper-induced low density lipoprotein oxidation Quercetin as an effective antioxidant against pro-oxidant. be determinant in the balance between the pro- and antioxidant behaviors. Ó FEBS 2004 Flavonoids and urate in copper-induced low- density lipoprotein oxidation