oxygen activation during peroxidase catalysed metabolism of flavones or flavanones

Chemico-Biological Interactions 122 (1999) 15–25

Oxygen activation during peroxidase catalysedmetabolism of flavones or flavanones

Tom Chan, Giuseppe Galati, Peter J. O’Brien *Department of Pharmacology and Faculty of Pharmacy, Uni6ersity of Toronto, Toronto, Ont.,

M5S 232 Canada

Received 18 December 1998; received in revised form 1 February 1999; accepted 1 April 1999

Abstract

Flavonoids containing phenol B rings, e.g. naringenin, naringin, hesperetin and apigenin,formed prooxidant metabolites that oxidised NADH upon oxidation by peroxidase/H2O2.Extensive oxygen uptake occurred which was proportional to the NADH oxidised and wasincreased up to twofold by superoxide dismutase. Only catalytic amounts of flavonoids andH2O2 were required indicating a redox cycling mechanism that activates oxygen andgenerates H2O2. NADH also prevented the oxidative destruction of flavonoids by peroxi-dase/H2O2 until the NADH was depleted. These results suggest that prooxidant phenoxylradicals formed by these flavonoids cooxidise NADH to form NAD· radicals which thenactivated oxygen. Similar oxygen activation mechanisms by other phenoxyl radicals havebeen implicated in the initiation of atherosclerosis and carcinogenesis by xenobiotic phenolicmetabolites. This is the first time that a group of flavonoids have been identified asprooxidants independent of transition metal catalysed autoxidation reactions. © 1999 El-sevier Science Ireland Ltd. All rights reserved.

Keywords: Oxygen activation; Peroxidase; Flavones; Flavanones; Flavonoids

www.elsevier.com/locate/chembiont

1. Introduction

Flavonoids are dietary phenolic compounds with antioxidant, anti-inflammatory,antimutagenic and anticarcinogenic activities [1]. They may also be anti-atheroscle-rotic as they prevent Cu2+ catalysed low density lipoprotein (LDL) oxidative

* Corresponding author.

0009-2797/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved.

PII: S0009 -2797 (99 )00103 -9

T. Chan et al. / Chemico-Biological Interactions 122 (1999) 15–2516

modification by scavenging reactive oxygen species and eliminating LDL peroxyland alkoxyl radicals [2,3]. Many flavonoid preparations are believed to possess avariety of alleged non-toxic therapeutic effects and are marketed as herbalmedicines or dietary supplements despite having yet to pass controlled clinicaltrials. However, despite all these useful therapeutic properties, flavonols thatcontain catechol or pyrogallol B rings can autooxidise in the presence of transitionmetals to produce reactive oxygen species (ROS) which accelerate LDL oxidationduring the propagation phase [4] or cause DNA strand scission [5]. Carcinogenicitydata for quercetin are conflicting [6] and the mutagenicity found for some flavonols,e.g. quercetin, likely result from autoxidation during the in vitro mutagenicitytesting [7]. This prooxidant activity as a result of autoxidation may not beimportant in vivo, where Cu2+ is largely sequestered.

It is well established that peroxidases oxidise phenols to phenoxyl radicals whichoxidise NADH and cause oxygen activation [8]. In the following, new prooxidantclasses of flavonoids have been identified, i.e. flavones and flavanones containingphenol B rings that on metabolism by peroxidases form prooxidant aryloxy radicalswhich cause extensive oxygen activation in the presence of NADH. This is the firstreport that some flavonoids are metabolically activated to prooxidants in theabsence of a transition metal catalysed autoxidation. This could have biologicalconsequences for peroxidase containing non-hepatic tissues.

2. Materials and methods

Flavonoids, NADH, horseradish peroxidase (HRP) type VI, superoxide dismu-tase (SOD) type 1 and cytochrome c were obtained from Sigma. Phenol wasobtained from Baker.

2.1. Measurement of NADH oxidation and oxygen consumption

The reaction mixtures contained 2 ml 0.1 M Tris–HCl/1.0 mM EDTA buffer(pH 7.4), flavonoid or phenol (25 mM), H2O2 (25 mM) and NADH (200 mM).Reactions were started by the addition of HRP (0.1 mM) and the oxidation ofNADH was followed at 340 nm using a Shimadzu UV-240 spectrophotometer untilthe NADH oxidation was complete. Oxygen consumption in this reaction mixturewas measured with a Clarke type electrode at 20°C.

2.2. Measurement of superoxide radical formation

The generation of superoxide radicals was studied by monitoring superoxidedismutase sensitive cytochrome c reduction by the flavonoid reaction mixture. Thereaction mixtures contained 2 ml 0.1M Tris–HCl/1.0 mM EDTA buffer (pH 7.4),flavonoid or phenol (10 mM), NADH (200 mM), HRP (0.1 mM), H2O2 (10 mM) andcytochrome c (40 mM). The absorbance of the reduced form of cytochrome c wasfollowed at 550 nm using a Shimadzu UV-240 spectrophotometer.

T. Chan et al. / Chemico-Biological Interactions 122 (1999) 15–25 17

Fig. 1. (A) Dependence of total NADH cooxidised by the peroxidase/flavonoid system on H2O2

concentration. Reaction conditions: NADH oxidation was followed at 340 nm in a reaction mixturecontaining 2 ml 0.1 M Tris–HCl/1.0 mM EDTA buffer pH 7.4, HRP (0.1 mM), NADH (200 mM),flavonoid (25 mM) various H2O2 concentrations (0.5–25 mM) as shown. Reactions were started by theaddition of HRP. Values represent mean9S.E. of three separate experiments. Phenol ("), apigenin ()and naringenin (�). (B) Dependence of the rate of peroxidase catalysed NADH oxidation on flavonoidor phenol concentration. Reaction conditions: the rate of NADH oxidation was determined at 340 nmin a reaction mixture containing 2.5 ml 0.1 M Tris–HCl/1.0 mM EDTA buffer pH 7.4, HRP (0.1 mM),NADH (200 mM), H2O2 (25 mM) and varying flavonoid or phenol concentrations as shown. Valuesrepresent mean9S.E. of three separate experiments. Phenol ("), apigenin () and naringenin (�). (C)NADH or GSH prevents apigenin oxidative degradation by peroxidase/H2O2 Reaction conditions:apigenin oxidation was followed at 390 nm in a reaction mixture containing 2.5 ml 0.1 M Tris–HCl/1.0mM EDTA buffer pH 7.4, HRP (0.04 mM), apigenin (25 mM) and H2O2 (50 mM). Values representmean9S.E. of three separate experiments. None ("), NADH 25 mM (), NADH 100 mM (�) andGSH 100 mM ().


2.3. Measurement of fla6onoid oxidati6e degradation

Apigenin oxidative degradation by peroxidase/H2O2 was monitored by followingthe decrease in its maximal absorbance at 390 nm using a Shimadzu UV-240spectrophotometer. The reaction mixtures contained 2 ml 0.1 M Tris–HCl/1.0 mMEDTA buffer (pH 7.4), HRP (0.04 mM), apigenin (25 mM), H2O2 (50 mM) andeither NADH (25, 100 mM) or GSH (100 mM).

3. Results

The data in Fig. 1A and B show that catalytic amounts of apigenin ornaringenin, flavonoids with a phenolic B ring, promoted NADH oxidation byperoxidase and catalytic amounts of H2O2 at a very high efficiency. NADH wascompletely restored on addition of lactate dehydrogenase and pyruvate when theNADH oxidation was complete thereby establishing that the decrease at 340 nmwas due to oxidation of NADH to NAD+ (results not shown). NADH was notoxidised by peroxidase and H2O2 in the absence of apigenin. In the experimentshown in Fig. 1A with 25 mM apigenin and trace amounts of H2O2, a total ofapproximately 1400 mol NADH were oxidised per mole of added H2O2 if thereaction was left for 60 min.

As shown in Fig. 1B, the rate of NADH oxidation was dependent on apigenin ornaringenin concentrations indicating that the rate of peroxidase/H2O2 catalysedoxidation of apigenin was rate limiting. NADH oxidation occurred at a pseudo-first-order rate with respect to apigenin concentration when fixed concentrations ofother reactants were used (Fig. 1B) and pseudo-first-order with respect to NADHconcentration (results not shown).

Fig. 1. (Continued)


The data in Fig. 1C, shows that apigenin in the absence of NADH was rapidlydegraded by peroxidase and H2O2 but not while NADH was present. It can also beseen that apigenin oxidation commenced once the NADH was oxidised. NADHalso prevented naringenin oxidation by peroxidase and H2O2 (results not shown). Acomparison of the prooxidant activity of a variety of flavonoids (structural formu-las shown in Table 1) was not carried out.

As shown in Table 2, other flavones or flavanones but not flavonols containinga phenol B ring at 25 mM concentration also catalysed the complete oxidation of200 mM NADH by 25 mM H2O2 and 0.1 mM peroxidase. The order of effectivenessof these flavonoids with respect to oxygen activation were apigenin\naringenin\hesperetin\naringin\phenol\hesperidin. However, kaempferol and flavanonesor flavonols containing catechol B rings, i.e. luteolin, catechin, rutin, quercetin, andfisetin caused nearly stoichiometric NADH oxidation without accompanying oxy-gen uptake.

The data shown in Table 2 show that the NADH oxidation was accompanied byextensive oxygen uptake. Oxygen consumption was proportional to the NADHoxidised and no oxygen consumption occurred in the absence of flavonoid. Further-more, superoxide dismutase increased the amount of oxygen uptake by naringeninup to twofold without affecting the rate of NADH oxidation suggesting thatsuperoxide dismutase generates H2O2 by dismutation of the superoxide anionradicals generated by the NADH oxidation. A stoichiometry of 0.8–1 mol O2 wasconsumed per mole of NADH oxidised. By contrast cytochrome c, a superoxideradical oxidant, inhibited NADH oxidation and oxygen uptake.

The involvement of superoxide radicals was further examined by monitoring thereduction of cytochrome c. Table 3 shows that the flavonoids studied wereineffective at reducing cytochrome c but readily reduced cytochrome c whenoxidised by the HRP/H2O2/NADH system. The flavonoids did not reduce cy-tochrome c in the absence of HRP/H2O2/NADH. Superoxide dismutase completelyinhibited the reduction of cytochrome c by the latter system with catalytic amountsof apigenin, naringenin, naringin and phenol but not with taxifolin, catechin orrutin. Boiled superoxide dismutase, however, had no effect. No cytochrome creduction occurred in the absence of NADH or HRP/H2O2. Quercetin or myricetincould not be used for this assay as these flavonoids reduced cytochrome c in theabsence of HRP/H2O2/NADH. These results suggest that superoxide radicals areresponsible for cytochrome c reduction by apigenin, naringenin and naringin,whereas semiquinone radicals are the species responsible for cytochrome c reduc-tion by taxifolin, catechin and rutin.

4. Discussion

Only flavones and flavanones containing phenol B rings cooxidised NADH withresulting oxygen activation when oxidised by peroxidase. The prooxidant activity offlavones and flavanones in catalysing NADH oxidation with accompanying oxygenuptake seemed to partly correlate with the one-electron redox potential of their


Table 1Structural formulas of the flavonoids used in this study

Flavonoid subclasses and examples Substituents

7 3% 4%3 5

Fla6onesOHHApigenin OHH OH

OH OHLuteolin OHH OHOHDiosmin H OH ORua OMe

Fla6anonesOH H OHNaringenin H OH

HNaringin H OH ORhb OHOH OMeHesperetin OHH OH

OMeHesperidin H OH ORua OH

Fla6anonolOHTaxifolin OHOH OHOH

Fla6an-3-olOHCatechin OH OH OH OH

Fla6onolsOHRutin ORua OH OHOHOHQuercetin OH OH OH OHH HGalangin OHOH OH

OHKaempferol OH OH OH HOHFisetin OHOH OHH

a Ru, rutinoside (6-0-(6-deoxy-a-L-mannopyranosyl)-b-D-glucopyranosyl).b Rh, rhamnoglucoside (2-0-(6-deoxy-a-L-mannopyranosyl)-b-D-glucopyranosyl).

phenoxyl radicals (Table 1) as only apigenin, naringenin and hesperetin had redoxpotentials higher than the other flavonoids. The one-electron redox potential ofNAD�, H+/NADH has been calculated to be 300 mV [9]. Kaempferol, a flavonolcontaining a phenol B ring, was much less active, presumably because the one-elec-tron redox potential of its phenoxyl radical was not high enough. As only catalyticamounts of apigenin and H2O2 were required to catalyse NADH oxidation,apigenin was oxidised by peroxidase/H2O2 to a product which oxidised NADH and


underwent redox cycling accompanied by extensive oxygen uptake that resulted inH2O2 formation. Apigenin was much more efficient than phenol as a catalyst forNADH oxidation as 1400 mol NADH respectively were oxidised per mole of H2O2

with apigenin using the experimental conditions shown in Fig. 1A.It is believed that peroxidases catalyse a one-electron oxidation of phenol to

phenoxyl radicals which cooxidise NADH to NAD� which in turn rapidly reduceO2 to O2

�.

Table 2Peroxidase catalysed oxygen activation by flavonoids and NADHa

Substrates Oxygen uptake,One-electron redox potential Eo (mV, pH NADH oxida-total (mM)7; PhO�/PhO−) tion, total (mM)

None 291 B1860cPhenol 589319992

Fla6onesApigenin 19892\1000d 88910Luteolin 180d, 180e, 299g 4094 B1Diosmin 492 B1

Fla6anones600d 141911 7797Naringenin

141913195912Naringenin/SODb

4393Naringenin/cy- 7997tochrome cb

194912 6196NaringinHesperetin 8197440d 6095

3593440d, 720fHesperidin 3393

Fla6anonolTaxifolin 379483g B1

Fla6an-3-olCatechin 4294160d, 130e, 570f B1

Fla6onols180d, 600f, 275g 3193 B1Rutin

3394Quercetin B130d, 0.06e, 398g

Galangin 320d, 340e B1291491120d, 170e, 209g B1Kaempferol

Fisetin 1191120d, 140e, 214g B1

a Incubation conditions. The reaction mixture contained in 2 ml 0.1 M Tris–HCl/1.0 mM EDTAbuffer pH 7.4, flavonoid or phenol (25 mM), NADH (200 mM), H2O2 (25 mM). Reactions were startedby the addition of HRP (0.1 mM) and followed until complete. NADH oxidation was followed at 340nm and oxygen uptake was determined with an oxygen electrode. Mean9S.E.M. for three separateexperiments are given.

b Superoxide dismutase (SOD) (1 mM) or cytochrome c (20 mM) were added where indicated.c E7 (mV) [28].d Ep/2 (mV) [29].e E1/2 (V�SCE) [30].f E7 (V�NHE) [16].g E1

o (mV) [31].


Table 3Flavonoid catalysed superoxide radical formation by the H2O2/HRP/NADH systema

Total cytochrome c reductionRate of cytochrome c reduction (nmol/minSubstrates(nmol/ml)per ml)

+SOD +SOD−SOD−SOD

B1None B1 B1 B11392Phenol 391 B1 B1

Fla6oneB1 3094 B1Apigenin 2994

Fla6anonesB12193Naringenin B11893

2693Naringin 692 B1B1

Fla6anonol2892Taxifolin 2496 2094 2692

Fla6an-3-ol20921592Catechin 22921793

Fla6onol1091Rutin 691 591 1292

a The complete system (2 ml) contained: 200 mM NADH, 0.1 mM HRP, 10 mM H2O2, 40 mMcytochrome c and 10 mM substrate in 0.1 M Tris–HCl/1.0 mM EDTA buffer pH 7.4. SOD (1 mM) wasadded where indicated. Cytochrome c reduction was determined by following the increase in absorbanceat 550 nm. Mean9S.E.M. for three separate experiments are given.

PhO�+NADH�PhOH+NAD

�(1)

NAD�+O2�NAD+ +O2

−� (2)

Pulse-radiolysis techniques have found the first-order biomolecular rate constantfor the reaction between NADH and the phenoxyl radicals of phenol to be8.0×107M−1 s−1 [10]. A direct electron transfer mechanism was proposed for thereaction between phenoxyl radicals and NADH that results in the formation ofNAD�. The rate constant for the reaction of NAD� with O2 has been reported tobe 2.0×109M−1 s−1 [11].

Evidence that the oxygen uptake resulted in superoxide radical formation wasthat cytochrome c reduction occurred when flavonoids were activated by HRP/H2O2/NADH and was prevented by superoxide dismutase. This suggests thatsuperoxide radicals were formed and were responsible for the cytochrome creduction. Furthermore, superoxide dismutase markedly increased the oxygen takenup when NADH was oxidised by naringenin or apigenin, presumably as a result ofthe formation of H2O2 from superoxide generated by NADH oxidation. Wepreviously described similar results for the activation of phenol by HRP/H2O2 [12].

Flavonoids containing catechol B rings stoichiometrically oxidised NADH with-out oxygen uptake suggesting that the NADH underwent a two-electron oxidationby the o-quinone product. Furthermore the reduction of cytochrome c by these


flavonoids when activated by HRP/H2O2/NADH was not affected by superoxidedismutase suggesting that redogenic semiquinone radicals and not superoxideradicals were formed. This lack of oxygen uptake and oxygen activation by theseflavonoids is surprising as resorcinol when oxidised by peroxidase cooxidisedNADH with resulting oxygen activation [8] and all of the flavonoids containingcatechol B rings contain a resorcinol A ring. Furthermore, the inactivation ofthyroid peroxidase and lactoperoxidase by flavonoids has been attributed tocovalent binding of A ring resorcinol radicals to catalytic amino acid radical(s) ofperoxidase compound II [13]. Another possibility is that the o-semiquinone radicalsformed from the flavonoids containing catechol B rings disproportionate and aretoo unstable to react with NADH. Thus when the o-semiquinone (oSQ) radicals ofcatechol were stabilised with Zn2+ of Mg2+ the radical reacted with ascorbate orGSH [14,15]. However we have found that no oxygen uptake or oxygen activationoccurred when quercetin or luteolin were activated by HRP/H2O2/NADH in thepresence of 0.2 M ZnCl2. Furthermore GSH conjugates were formed when NADHwas replaced with GSH and little oxygen activation occurred (results not shown).Pulse radiolysis studies have also shown that quercetin or luteolin semiquinoneradicals disproportionate rapidly and also interact with semidehydroascorbateradicals [16]. This suggests that reactions 3–5 could explain the lack of NADHmediated oxygen activation by semiquinone radicals.

oSQ+oSQ�oQ+catechol (3)

oSQ+NAD� X oQ+NADH (4)

oSQ+NAD��catechol+NAD+ (5)

The peroxidase catalysed oxidation products of apigenin or hesperetin are notknown. However in the absence of NADH or other suitable hydrogen donor,naringenin after forming the phenoxyl radical undergoes a complex series ofreactions which result in the oxidative degradation of naringenin including anoxidative attack at C2 and ring A [17]. The prevention of the oxidative degradationof naringenin or apigenin by NADH until the NADH has been oxidised indicatesthat NADH reduces the phenoxyl radical back to naringenin or apigenin.

Phenoxyl radicals have been implicated in benzene induced myelotoxicity as bonemarrow myeloperoxidase catalyses the oxidative activation of phenolic metabolitesof benzene [18] to phenoxyl radicals which caused oxidative DNA damage directlyor via GSH/NADH mediated oxygen activation [19]. Incubation of bone marrowcells or phenolic metabolites of benzene with HL-60 human promyelocytic leukemiccells resulted in oxygen activation, H2O2 formation, oxidative DNA damage andcytotoxicity [20,21]. Phenoxyl radicals have also been implicated as a cause ofatherosclerosis as myeloperoxidase readily catalyses the oxidation of tyrosine, aplasma phenol [22] to the tyrosyl radical that cooxidised the lipids and proteins oflow density lipoproteins (LDL) [23,24]. Myeloperoxidase is also found in humanatherosclerotic tissue [23]. Protein bound o,o %-dityrosine (a tyrosyl radical product)was also found to be markedly increased in LDL isolated from human atheroscle-rotic lesions [25]. Interestingly, this LDL oxidation mechanism unlike previous


mechanisms does not require free transition metal ions (copper or iron) which inthe plasma are present almost exclusively in tightly bound proteins (e.g. ceruloplas-min and transferrin) and do not catalyse oxidative modification of LDL.

The three most prooxidant flavonoids found were naringenin, naringin andapigenin. Naringin constitutes up to 10% of the dry weight of grapefruit concentra-tions and naringin concentrations in grapefruit juice are reported to range from 100to 800 mg/l. Recently there are concerns that naringenin, the major naringinmetabolite, may contribute to the cytochrome P4503A4 inhibition that results indrug interactions with grapefruit juice [26]. Apigenin concentrations in celery arereported to be 108 mg/kg fresh weight [1]. It is also being considered as a skincancer preventive agent in sunscreens [27]. Further research is required to determinethe consequences, if any, for activated oxygen species formation by naringenin,naringin and apigenin in peroxidase containing tissues or for plasma LDLoxidation.

5. Note added in proof

We have now shown that some flavonoids containing phenol B rings also oxidiseglutathione upon oxidation by peroxidase/H2O2 [32].

Acknowledgements

This work was supported by a grant from the National Sciences and EngineeringResearch Council of Canada.

References

[1] C. Manach, F. Regerat, O. Tecier, G. Agullo, C. Demigne, C. Remesy, Bioavailability, metabolismand physiological impact of flavonoids, Nutr. Res. 16 (1996) 517–544.

[2] N. Salah, N.J. Miller, G. Paganga, L. Tyburg, G.P. Bolwell, C. Rice-Evans, Polyphenolic flavonolsas scavengers of aqueous phase radicals and chain-breaking antioxidants, Arch. Biochem. Biophys.322 (1995) 339–546.

[3] N. Yamanaka, O. Oda, S. Nagao, Prooxidant activity of caffeic acid, dietary non-flavonoidphenolic acid, on Cu2+-induced low density lipoprotein oxidation, FEBS. Lett. 405 (1997)186–190.

[4] N. Yamanaka, O. Oda, S. Nagao, Green tea catechins such as epicatechin and epigallocatechinaccelerate Cu2+-induced low density lipoprotein oxidation in propagation phase, FEBS. Lett. 401(1997) 230–234.

[5] M.S. Ahmad, F. Fazal, A. Rahman, S.M. Hodi, J.H. Parish, Activities of flavonoids for thecleavage of DNA in the presence of Cu(II): correlation with generation of active oxygen species,Carcinogenesis 13 (1992) 605–608.

[6] N. Ito, Is quercetin carcinogenic?, Jpn. J. Cancer Res. 83 (1992) 312–314.[7] J. Rueff, A. Laires, J. Gaspair, H. Booba, A. Rodrigues, Oxygen species and the genotoxicity of

quercetin, Mutat. Res. 265 (1992) 75–81.


[8] P.J. O’Brien, Radical formation during the peroxidase-catalysed metabolism of carcinogens andxenobiotics: the reactivity of these radicals with GSH, DNA and unsaturated lipid, Free Radic.Biol. Med. 4 (1988) 169–183.

[9] J. Grodknowski, P. Neta, B.W. Carlson, L. Miller, One electron transfer reactions of the coupleNAD·/NADH, J. Phys. Chem. 87 (1983) 3135–3138.

[10] L.G. Fomi, R.L. Willson, Thiyl and phenoxyl free radicals and NADH. Direct observations ofone-electron oxidation, Biochem. J. 240 (1986) 897–903.

[11] E.J. Land, A.J. Swallow, One-electron reactions in biochemical systems as studied by pulseradioly-sis, Biochim. Biophys. Acta 234 (1971) 34–42.

[12] V.V. Subrahmanyan, P.J. O’Brien, Peroxidase catalysed oxygen activation by arylamine carcinogensand phenol, Chem.-Biol. Interact. 56 (1985) 185–199.

[13] R.L. Divi, D.R. Doerge, Mechanism-based inactivation of lactoperoxidase and thyroid peroxidaseby resorcinol derivatives, Biochemistry 33 (1994) 9668–9674.

[14] B. Kalyanaraman, C.C. Felix, R.C. Sealy, Semiquinone anion radicals of catechol(amines), catecholestrogens and their metal complexes, Environ. Health Perspect. 64 (1985) 185–198.

[15] A. Sadler, V.V. Subrahmanyam, D. Ross, Oxidation of catechol by horseradish peroxidase andhuman leukocyte peroxidase: reactions of o-benzoquinone and o-benzosemiquinone, Toxicol. Appl.Pharmacol. 93 (1988) 62–71.

[16] W. Bors, C. Michel, S. Schikora, Interaction of flavonoids with ascorbate and determination oftheir univalent redox potentials: a pulse radiolysis study, Free Radic. Biol. Med. 19 (1) (1995)45–52.

[17] M. Patzlaff, W. Barz, Peroxidatic degradation of flavanones, Z. Naturforschi. 330 (1978) 675–684.[18] V.V. Subrahmanyan, D. Ross, D.A. Eastmond, M.T. Smith, Potential role of free radicals in

benzene-induced myelotoxicity and leukemia, Free Radic. Biol. Med. 11 (1991) 495–515.[19] D.J. Reed, J.R. Babson, P.W. Beatty, A.E. Boodie, W.W. Ellis, D.W. Potter, HPLC analysis of

nanomole levels of GSH, GSSG and related thiols and disulfides, Anal. Biochem. 106 (1980) 55–62.[20] P. Kolachana, V.V. Subrahmanyan, K.B. Meyer, L. Zhang, M.T. Smith, Benzene and its phenolic

metabolites produce oxidative DNA damage in HL-60 cells in vitro and the bone marrow in vivo,Cancer Res. 53 (1993) 1023–1026.

[21] N.R. Rao, R. Snyder, Oxidative modifications produced in HL-60 cells on exposure to benzenemetabolites, J. Appl. Toxicol. 15 (1995) 403–409.

[22] M. Savenkova, D.M. Mueller, J.W. Heinecke, Tyrosyl radical generated by myeloperoxidase is aphysiological catalyst for the initiation of lipid peroxidation in low density lipoprotein, J. Biol.Chem. 269 (1994) 20394–20400.

[23] A. Daugherty, J.L. Dunn, D.L. Rateri, J.W. Reinecke, Myeloperoxidase, a catalyst for lipoproteinoxidation, is expressed with human atherosclerotic lesions, J. Clin. Invest. 91 (1993) 437–444.

[24] J.W. Heinecke, W. Li, G.A. Francis, J.A. Goldstein, Tyrosyl radical generated by myeloperoxidasecatalyses the oxidative cross-linking of proteins, J. Clin. Invest. 91 (1993) 2866–2872.

[25] C. Leeuwenburgh, J.E. Rasmussen, F.F. Hsu, D.M. Mueller, J.W. Heinecke, Mass spectrometricquantification of markers for protein oxidation by tyrosyl radical in low density lipoprotein isolatedfrom human atherosclerotic intima, J. Biol. Chem. 272 (1997) 3520–3526.

[26] U. Fuhr, Drug interactions with grapefruit juice, Drug Saf. 18 (4) (1998) 251–272.[27] B. Li, D.H. Robinson, D.F. Birt, Evaluations of properties of apigenin and analytic method of

development, J. Pharm. Sci. 86 (1997) 721–725.[28] P. Wardman, Reduction potentials of one-electron couples involving free radicals in aqueous

solution, J. Phys. Chem. Ref. Data 18 (4) (1989) 1637–1755.[29] S.A.B.E. Van Acker, D. Van Den Berg, M.N.J.L. Tromp, D.H. Griffioen, W.P. Van Bennekom,

W.J.F. Van Der Vijgh, et al., Structural aspects of antioxidant activity of flavonoids, Free Radic.Biol. Med. 20 (1996) 331–342.

[30] W.F. Hodnick, E.B. Milosavljevic, J.H. Nelson, R.S. Pardini, Electrochemistry of flavonoids,Biochem. Pharmacol. 37 (1988) 2607.

[31] S.V. Jovanovic, S. Steenken, M. Tosic, M.G. Simic, Flavonoids as antioxidants, J. Am. Chem. Soc.116 (1994) 4846–4851.

[32] G. Galati, T. Chan, P.J. O’Brien, Chem. Res. Toxicol., in press..

oxygen activation during peroxidase catalysed metabolism of flavones or flavanones

Documents