Free Radical Intermediates during Peroxidase ... - Science Direct

ARCHIVES

OF BIOCHEMISTRY

AND BIOPHYSICS

Vol. 269, No. 2, March, pp. 423-432,1989

Free Radical Intermediates during Peroxidase Oxidation of 2-t-Butyl-4-methoxyphenol, 2,6-Di-t-butyl-4-methylphenol, and Related Phenol Compounds MASSIMO VALOTI,’ HERBERT J. SIPE, JR.; GIAMPIETRO SGARAGLI,l AND RONALD P. MASON3 Laboratory

of Molecdar

Biophysics, National Institute of Environmental Health Sciences, P.O. Box 12233, Research Triangle Park, North Carolina 27709 Received September 6,1988

2-t-butyl-4-methoxyphenol (BHA) and 2,6-di-t-butyl-4-methylphenol (BHT) are widely used antioxidant food additives that are generally recognized as safe by the Food and Drug Administration. Previously reported studies have suggested that the ip LDsO of BHA may be as much as 2 orders of magnitude lower than its oral LDsO. Metabolic activation of BHA to reactive intermediates possibly may be responsible for this result and may be related to other reported toxic effects. BHT has been reported to cause haemorrhagic lung damage and possible hepatocarcinogenicity in test animals. The present studies report investigations by electron spin resonance spectroscopy of free radical metabolites of BHA, BHT and related compounds. The primary, unstable phenoxy free radical of BHA has been generated by oxidation with horseradish peroxidase and hydrogen peroxide and detected by ESR spectroscopy. A scheme has been proposed for the peroxidatic oxidation of BHA. The ESR spectrum of the di-BHA dimer, one product of BHA oxidation, has been observed, analyzed, and reported. ESR studies have been extended to other phenol derivatives structurally related to BHA and suspected to be substrates for peroxidase. Similarly it has been found that BHT and structurally related phenols are substrates for peroxidation by horseradish peroxidase and hydrogen peroxide. In agreement with previous chemical and biochemical studies, it has been found that ortho-disubstituted phenols are oxidized to more stable phenoxy radicals than are orthomonosubstituted phenols. The ESR hyperfine coupling constants for the phenoxy radicals studied are in agreement with those for similar radicals produced by chemical oxidation. Attention has been drawn to the biochemical and toxicological implications of these and related studies of BHA and BHT peroxidation. o 1989AeademicPress.Inc. BHA4 (2-t-butyl-4-methoxyphenol), a widely used antioxidant food additive, has a number of potentially important phar‘Permanent address: Universiti degli Studi di Siena, Istituto di Scienze Farmacologiche, Via Piccolomini 170, Siena, Italy. * Permanent address: Department of Chemistry, Hampden-Sydney College, Hampden-Sydney, VA 23943. 3 To whom correspondence should be addressed. 4 Abbreviations used: BHA, 2-t-butyl-l-methoxyphenol; BHT, 2,6-di-t-butyl-4-methylphenol; di-BHA,

macological and toxicological properties. It has been indicated as a protective agent against carcinogenesis and toxicity since it effectively inhibits tumor formation induced by a variety of carcinogens (1,2) and it protects animals against the acute toxic effects of a variety of compounds (3-5). BHA can also be harmful, for long-term 2,2-dihydroxy-3,3’-di-t-butyl-5,5’-dimethoxydiphenyl; DTBHA, 3,5-di-t-butyl-4-hydroxyanisole; TTP, 2,4,6tri-t-butyl-phenol.

423

0003-9861189 $3.00 Copyright All rights

Q 1989 by Academic Press, Inc. of reproduction in any form reserved.

424

VALOTI

administration of this compound in pellet diet has been shown to promote the formation of forestomach carcinoma in male and female rats (6) and papilloma in Syrian golden hamster (7). Nevertheless, at lower concentrations BHA exhibits a very low toxicity in mammals when administered orally (WHO Food Additive Series No. 5, 1974), and it is generally recognized as safe by the Food and Drug Administration. However, additional information on the elementary toxicology of BHA has been provided recently by the unexpected demonstration that the ip LDsOis about 2 orders of magnitude lower than the oral LD5,, (8). Furthermore, BHA was observed to cause hemorrhagic lung damage in rats (9). Although the above toxic properties of BHA have been well described, little is known about the exact mechanism of how this compound causes toxicity and/or carcinogenicity. It is possible that these toxic effects result from a metabolic activation of this compound to reactive intermediates. In this context, of particular interest is the capacity of BHA to undergo peroxidatic oxidation in vitro. Previous work has shown that BHA is oxidized by horseradish or mammalian peroxidase systems to the corresponding dimer, di-BHA (10). Recently, rat intestine peroxidase purified to electrophoretic homogeneity has been shown to catalyze this reaction as efficiently as the oxidation of guaiacol(l1). Interest in peroxidases as systems which metabolically enhance the toxicity of xenobiotics arose from the observation that they catalyze the oxidation of phenol or amine derivatives to reactive species which can bind to both proteins and DNA (12-15). Furthermore, Rahimtula (16) has shown that BHA undergoes both mixedfunction oxidase-dependent oxidation with liver microsomes and peroxidatic oxidation with sheep seminal microsomes, horseradish peroxidase, and liver microsomes, to yield a variety of products including formaldehyde, di-BHA, and polar metabolites as well as reactive intermediates that bind irreversibly to proteins. BHT (2,6-di-t-butyl-4-methylphenol) is also widely used, sometimes in combina-

ET AL.

G MOWl.ATlGN

FIG. 1. ESR spectra of BHA. (A) Complete system of 1.39 mM BHA, 1.0 mM HaOa, and 20 pg/ml horseradish peroxidase, in Tris-HCl buffer, pH 8.0, incubation 12.5% (v/v) ethanol and bubbled with N2 before aspiration into the ESR flat cell. Instrument conditions: sweep rate, 25 G/100 s; microwave power, 20 mW; modulation amplitude, 1.0 G; time constant, 1.0 s; gain, 2.0 X 106. (B) Same as A, but 2 mM H202 and same instrument conditions except: sweep rate, 25 G/ 200 s; modulation amplitude, 0.5 G; gain, 4.0 X 106. (C) Computer simulation of BHA ESR spectrum (a; = 5.33 G (lH), oficHs = 1.87 G (3H), op5 = 0.70 G (lH), a?’ = 0.30 G (1H)).

tion with BHA, as an antioxidant food additive. Although generally accepted as safe for such uses, BHT has been shown to cause lung damage in mice, hemorrhagic death in rats, and liver necrosis in mice and rats (17-20). Currently there is also concern about possible hepatocarcinogenicity of BHT (21, 22). Experiments designed to study the effects of BHT on the carcinogenicity of chemical agents have given mixed results (23). In previous work (lo), ESR spectroscopy was used to observe free radical intermedi-

PHENOXY 25

FREE

RADICAL

I

0.5

10

1.5

2.0

25

1/ HRP (units/ml)

Horseradish peroxidase (type VI, activity 250-330 units rng-‘) was used as received from Sigma Co. (St. Louis, MO). Phenolic derivatives 3,5-di-t-butyl-4-hydroxyanisole (DTBHA) and 2,4,6-tri-t-butylphenol (TTP) from Aldrich Chemical Co. were recrystallized once from ethanol prior to study; BHA and 2,6-dit-butyl-p-cresol (butylated hydroxytoluene, BHT) from Fluka A.G. were recrystallized once, before use, from petroleum ether and ethanol, respectively. InstmLmentation. ESR spectra were measured at room temperature using either an IBM/Bruker ERZOOD or a Varian E-109 ESR spectrometer fitted with a TM,,, cavity and large volume flat cell and employing the modified Gilford rapid sampler previously described (26). Computer simulation of the experimental ESR spectra were carried out on a Hew-

FIG. 2. Horseradish peroxidase dependence of the phenoxy free radical concentration derived from BHA. BHA 1.38 mM, 12.5% (v/v) ethanol, 1 mM H202 in Tris-HCI buffer, pH 8, 100 mM was assayed in the presence of different enzyme concentrations. Sample amplitudes were measured with 1 G amplitude modulation, time constant 1 s and gain 4 X 106.

ates during peroxidative oxidation of BHA. Because of the long incubation times employed, those studies did not observe the initial BHA radical and reported a complex ESR spectrum without analysis. Also reported was an ESR spectrum for the di-BHA dimer produced by the BHA radical coupling reaction. The aim of the present investigation was to detect the initial free radical intermediate generated during peroxidatic oxidation of BHA. In order to characterize this radical fully, a comparative study was undertaken on a series of phenol derivatives structurally related to BHA as possible substrates of peroxidases which might generate free radical intermediates when incubated with horseradish peroxidase and HZOz.In particular, since phenols with unsubstituted ortho positions are known to undergo dimerization reactions when oxidized (24), ESR studies of ortho-substituted analogs of BHA were desirable. MATERIALS

AND

METHODS

Muteri&. The dimer di-BHA (2,2’-dihydroxy-3,3’di-t-butyl-5,5’-dimethoxydiphenyl) of BHA was synthesized by the method of Hewgill and Hewitt (25).

425

METABOLITES

COMPLETE SYSTEM

C

-DlBHA

E

-H,O:

FIG. 3. ESR spectra of DTBHA. (A) Complete system of 1.06 mM DTBHA, 1.0 mM H202. and 10 Fg/ml horseradish peroxidase (HRP), in Tris-HCl buffer, pH 8.0; incubation 12.5% (v/v) ethanol and bubbled with N, before aspiration into the ESR flat cell. Instrument conditions: microwave power, 20 mW; sweep rate, 25 G/200 s; modulation amplitude, 0.25 G; time constant, 1.0 s; gain, 5.0 X 105. (B) Computer simulation of DTBHA radical ESR spectrum (agcH3 = 1.70 G (3H), a$’ = 0.63 G (2H)). (C) Same as A, but without DTBHA. (D) Same as A, but without HRP. (E) Same as A, but without H,O,.

VALOTI ET AL.

COMPLETE SYSTEM

Kinetics of radical jbrmation. The formation of BHA radical was monitored by observing the growth of one of its ESR lines as a function of time for varying enzyme concentrations. Enzyme activity was assayed by the method of Chance and Mahely (Z?), which used guaiacol as hydrogen donor as previously described (II), so that solutions of enzyme by weight per unit volume could be expressed as activity per unit volume. RESULTS

In a system containing horseradish peroxidase and H202 in Tris-HCl buffer, pH SIMULATION

A

E

COMPLETE

-H*O

FIG. 4. ESR spectra of TTP. (A) Complete system of 0.95 mM TTP, 1.0 mM Hz02, and 10 Kg/ml horseradish peroxidase (HRP), in Tris-HCl buffer, pH 8.0; incubation 50% (v/v) ethanol and bubbled with N2 before aspiration into the ESR flat cell. Instrument conditions: sweep rate, 25 G/500 s; microwave power, 20 mW; modulation amplitude, 0.25 G; time constant, 1.0 s; gain, 5.0 X 105. (B) Computer simulation of TTP ESR spectrum (atiGmtBu = 0.36 G (9H), aa = 1.63 G (2H)). (C) Same as A, but without TTP. (D) Same as A, but without HRP. (E) Same as A, but without HKh.

lett-Packard HP-900-310 microcomputer using a modified version of Varian’s isotropic simulation programming package. Imubatim conditions. Peroxidatic oxidation of different phenol compounds was carried out at room temperature in a reaction mixture of 4 ml containing (final concentration) lo-20 pg/ml horseradish peroxidase; 50 or 100 mM Tris-HCl buffer, pH 8; and 1 mM H202 (initiator of the reaction). Substrate concentrations were 1.39 mM BHA or 1.06 mM DTBHA in 12.5% (v/v) ethanol and 0.95 mM TTP or 1.1 mM BHT or 43 pM di-BHA in 50% (v/v) ethanol. ESR spectra were recorded immediately after addition of HzOz and bubbling for 1 min with Nz.

C

-BtlT

D

-HRP

E

-HZ01

FIG. 5. ESR spectra of BHT. (A) Complete system of 1.1 mM BHT, 1.0 mM HzOz, and 10 pg/ml horseradish peroxidase (HRP), in Tris-HCl buffer, pH 8.0; incubation 50% (v/v) ethanol and bubbled with NZ before aspiration into the ESR flat cell. Instrument conditions: sweep rate, 50 G/200 s; microwave power, 20 mW; modulation amplitude, 1.0 G; time constant, 1.0 s; gain, 2.0 X 106. (B) Computer simulation of BHT ESR spectrum (agH3= 11.61G (3H), aa” = 1.54 G (ZH)). (C) Same as A, but without BHT. (D) Same as A, but without HRP. (E) Same as A, but without HzOz.

PHENOXY

FREE

RADICAL

METABOLITES

427

di-BHA r ,

COMPLETE

SYSTEM

I

-di-BHA

-HRP

-Hz02

FIG. 6. ESR spectrum of di-BHA. (A) Complete system of 43 pM di-BHA, 1.0 InM H202, and 20 pg/ ml horseradish peroxidase (HRP), in Tris-HCl buffer, pH 8.0; incubation 50% (v/v) ethanol and bubbled with Nz before aspiration into the ESR flat cell. Instrument conditions: sweep rate, 25 G/ 500 s; microwave power, 20 mW; modulation amplitude, 0.5 G; time constant, 2.0 s; gain, 2.0 x 106. (B) Same as A, but without di-BHA. (C) Same as A, but without HRP. (D) Same as A, but without

8, BHA was easily oxidized by horseradish peroxidase to generate a free radical. The ESR spectrum of the primary free radical detected at high amplitude (1 G) modulation (Fig. 1A) consisted of a doublet of quartets. Low amplitude (0.5 G) modulation allowed better resolution resulting in the separation of two nonequivalent meta protons, although at a poorer signal-tonoise ratio (Fig. 1B). The assignment of hyperfine coupling constants (see Table I) was confirmed by computer simulation (Fig. 1C). The formation of the free radical was dependent upon the presence of BHA, horseradish peroxidase, and HzOz. The dependence of the BHA free radical concentration on enzyme concentration was also determined by monitoring one of the lines in the ESR spectrum. As shown in Fig. 2, BHA radical concentration during the first minute of incubation was linear in the square root of horseradish peroxidase concentration at least through 5.6 units ml-‘. In order to provide more information which might help to analyze the ESR spec-

trum of BHA in more detail, other phenol derivatives, structurally similar to BHA, were employed as substrates for the peroxidase. During peroxidatic oxidation these compounds were shown to generate free radicals and their corresponding ESR spectra were analyzed. The analysis of the BHA ESR spectrum was based on hyperfine coupling constants obtained originally from the much more intense ESR spectrum of DTBHA radical (Fig. 3A). This species was produced in much higher steady-state concentration under experimental conditions since substitution of both ortho positions by t-butyl groups prevented the destruction of the DTBHA radical by ortho dimerization. The radical formed during peroxidatic oxidation of DTBHA showed a spectrum characterized by a quartet of triplets due to the equivalent meta protons and to the three protons of the para-methoxy group, respectively. Similarly, TTP was readily oxidized by the horseradish peroxidase/H,02 system

428

VALOTI ET AL.

COMPLETE

SYSTEM

I

1

C

OEUTERATED

SUFFER

5 wusa 1

D

SIMULATION

FIG. 7. ESR spectra of di-BHA. (A) Complete system of 43 FM di-BHA, 1.0 mM HaOa, and 20 pg/ml horseradish peroxidase (HRP), in Tris-NC1 buffer, pH 8.0; incubation 50% (v/v) ethanol and bubbled with Nz before aspiration into the ESR flat cell. Instrument conditions: sweep rate, 25 G/200 s; microwave power, 20 mW; modulation amplitude, 0.5 G; time constant, 0.5 s; gain, 1.0 x 106.(B) Computer simulation of di-BHA ESR spectrum (a:” = 0.60 G (lH), a$‘““~= 0.90 G (6H)). (C) Complete system as in A, but all solutions prepared using either Da0 or CHaCHaOD and Tris buffer prepared with D20 and DCl. Instrument conditions same as A except: gain, 2.5 X 10’. (D) Computer simulation of ESR spectrum C (an = 0.92 G (f33)).

to produce a high steady-state radical concentration and strong ESR spectrum (Fig. 4A). The spectrum was characterized by a triplet splitting from two meta protons and a smaller, unresolved splitting presumably from t-butyl protons. A good computer simulation (Fig. 4B) could be achieved assuming 9 equivalent hydrogens

caused the poorly resolved hyperfine splitting but not by assuming either 18 or 27. Thus, the partially resolved hyperfine splitting was assigned to the nine hydrogens of the para t-butyl group of TTP radical. A free radical intermediate of BHT also was shown to be produced by peroxidatic oxidation. The ESR spectrum of this radical is reported in Fig. 5A; there, the pattern of a quartet of triplets is evident, indicating the presence of the para-methyl group and of two equivalent protons in the meta position. The coupling constants for the quartet and triplet are comparable to those found by Lambelet et al. (28) when BHT was incubated with a previously oxidized chicken fat liquid fraction. No signal was obtained when either DTBHA or TTP was incubated in the absence of H,Os or horseradish peroxidase. Interestingly, however, when BHT was incubated with horseradish peroxidase without hydrogen peroxide, a weak signal from BHT free radical was still generated (Fig. 5E). This signal was suppressed completely by the addition of 20 pg/ml catalase to the incubation prior to horseradish peroxidase addition (results not shown), implying autoxidation of BHT forms traces of Hz02. Furthermore, the radical decay of these three phenol derivatives was slower as compared to that of BHA. The corresponding ESR signal, in fact, was observed for many minutes. The peroxidatic oxidation product of BHA, di-BHA, was shown in a previous study (29) to be a substrate for peroxidases. In the present study, the formation of a free radical intermediate of di-BHA during peroxidatic oxidation was observed. The corresponding ESR spectrum is shown in Fig. 6A. The computer simulation (Fig. 7B) of this spectrum was performed by using the same hyperfine coupling constants for this radical found by Hewgill and Legge (30) in a study where diBHA was chemically oxidized with silver oxide in trichloromethane. The spectrum was characterized by seven lines due to the six equivalent protons of the two methoxy groups. The doublet due to one hydroxy proton was not well resolved under these conditions. However, when di-BHA was in-

PHENOXY

FREE

RADICAL

cubated with peroxidase and hydrogen peroxide in a buffer composed of Tris-DCl, D20 and CH3CH20D (Fig. 7C), an ESR spectrum was observed showing a simple septet with better defined, sharper lines. Deuteration removed the hydrogen splitting of the -OH group, thus allowing the resolution of the septet originating from the six equivalent protons of the two methoxy groups. The hyperfine coupling constants measured from the ESR spectra of the phenolic free radicals are summarized in Table I. The BHA phenoxy radical had an &ho coupling constant of 5.33 G, typical of phenoxy radicals. Species observed in the BHA series and BHT series of compounds all have similar hyperfine couplings from analogous positions although, not surprisingly, there are differences between the two series. Interestingly, the relatively simple ESR spectrum observed for the BHA dimer radical (Figs. 6A, 7A, 7C) is consistent with that of the BHA monomer radical (Fig. 1A). In a first approximation, delocalization of the radical’s spin density throughout two aromatic rings would decrease by half the coupling constants while doubling the numbers of equivalent nuclei undergoing hyperfine interaction with the spin density. Thus the agcH3 = 1.87 G of the BHA phenoxyl radical becomes 0.90 G for di-BHA, while upd5 = 0.70 and 0.30 G of the BHA phenoxyl radical would become 0.35 and 0.15 G for diBHA and are not resolved.

2

HRP

-2

WH3)3

429

METABOLITES DISCUSSION

In previously reported work (lo), the initial BHA radical was not observed, probably because of its instability compared to the long incubation periods adopted in that study. In that study, long-term incubation of BHA in the presence of horseradish peroxidase produced an ESR spectrum identical with that produced by electrolysis of di-BHA. The ESR spectrum reported previously for di-BHA was qualitatively similar to Fig. 6 in our present study, although hyperfine coupling constants were not reported. The primary phenoxy free radical of BHA, generated during its peroxidatic oxidation, has been clearly detected in the present study by using ESR spectroscopic methods. Furthermore, this study was extended to other phenol derivatives structurally related to BHA and suspected to be substrates for peroxidase. DTBHA, TTP, and BHT when incubated with peroxidase and HzOz generated the corresponding free radicals which were detected by ESR spectroscopy. The spectral analysis of these radicals made it possible to define the structure of the BHA-derived radical (Table I), which is characterized by a doublet of quartets. From the present results it is now possible to define the mechanism of peroxidatic oxidation of BHA (Scheme 1):

WH3)3

““““~W”&

Hz02

OCHs

OCH3

OCH3

OCHl

[I 3

0’

OH WH3)3

HRP H202

SCHEMEI

430

VALOTI

ET AL.

TABLE

I

HYPERFINE COUPLING CONSTANTS FOR PHENOXY RADICALS FROM PHENOL COMPOUNDS

Name

RI

R2

a3 and a5

LW

1.87 f 0.02 (3 H)

0.70 f 0.02 (1 H) 0.30 f 0.02 (1 H)

0.50

H

DTBHA

t-butyl

OCH,

--a

1.70 f 0.02 (3 H)

0.63 f 0.02 (2 H)

0.30

TTP

t-butyl

t-butyl

-a

0.36 f 0.02 (9 H)

1.63 f 0.02 (2 H)

0.38

BHT

t-butyl

CHB

-0

11.61 f 0.02 (3 H)

1.54 + 0.02 (2 H)

0.40

0.60 f 0.02 (OH)

0.90 f 0.02 (6 H)

-a

0.50

-’

0.92 + 0.02 (6 H)

-.-a

0.30

Q OH C,CH,,. OCH3

5.33 + 0.02 (1 H)

aRz

BHA

di-BHA

OCH3

aRl

:’ OCH, (OD)*

a Unresolved. * Tris-DCl, DzO, CH,CH,OD.

According to the above scheme, two molecules of the primary phenoxy free radical [Z] react together to yield the dimer [3]. The square-root dependence of the BHA steady-state radical concentration on the enzyme concentration (Fig. 2) is consistent with a second-order radical decay process (31). Furthermore, we have shown here that di-BHA is also oxidized to a free radical intermediate [4] when incubated with peroxidase and HzOz, giving rise perhaps to the formation of not yet identified compound(s). The spectral analysis of this radical has offered some information on its structure. In particular, the experiment carried out in deuterated buffer allowed the assignment of hyperfine coupling constants of the methoxy protons in positions 3 and 3’ and of the phenolic proton as well. Furthermore, the signal due to the phenolit proton should indicate the formation of hydrogen bonding between the two phenolit oxygens, thus suggesting that the two oxygens are in a cis configuration. The formation of similar intramolecular hydrogen bonding between an 0’ and an OH group was reported to occur with naphtha-

zarin during air oxidation in a weakly basic solution (32) and with daunomycin and related compounds in a xanthine/xanthine oxidase system (33). As to the primary free radical intermediate of BHT, its formation during incubation with peroxidases and HzOz has been recently hypothesized and suspected to be involved in the general toxicity exhibited by BHT (34, 35). The relatively long lived BHT and related radicals have also been reported to decay by dimerization paths (36,37). However, in contrast to the stable di-BHA dimer, the BHT dimer, 4-(2,6-di-tbutyl-4-methylphenoxy)-2,6-di-t-butyl-4methyl-Z,&cyclohexadiene-l-one, exists in equilibrium with BHT radical, thus ensuring a strong BHT ESR signal. At present, the possible involvement of a reactive free radical intermediate generated during peroxidatic oxidation of BHA in its toxicity toward mammals can be only a matter of conjecture. It is worth outlining that BHA given ip to rats or mice has been shown to be selectively toxic to the intestine, causing a delayed (24-48 h) impairment of smooth muscle contractility

PHENOXY

FREE

RADICAL

leading to death of animals in a few days (38,39). Such delayed toxicity may possibly result from a metabolic toxic activation of this compound at the intestine wall. In this context it is of particular interest that BHA was shown to undergo peroxidatic oxidation in vivo. This transformation, in fact, has been demonstrated to occur in everted intestinal preparations (40) and in the rat following the oral administration of BHA (41). In the latter study, the intestine was suggested to be the major site where the transformation of BHA into diBHA occurs. Interestingly, a recent study has shown that this metabolic transformation took place at the rat intestinal wall when BHA was given into the peritoneal cavity (39). ACKNOWLEDGMENTS This work has been partially supported by CNR, Roma, Italy (Contract 87.00545.04). Universiti di Siena and NIEHS are gratefully acknowledged for supporting travel expenses of M.V.

REFERENCES 1. WATTENBERG, L. W., JERINA, D. M., LAM, L. K. T., AND YAGI, H. (1979) J. Natl. Cancer Inst. 62, 1103-1106. 2. WATTENBERG, L. W., AND SPARNINS, V. L. (1979) J. Natl. Cancer Inst. 63,219-222. 3. WATTENBERG, L. W. (1972) J. Nat1 Cancer Inst. 48,1425-1430. 4. KIM, H. L., ANDERSON, A. C., TERRY, M. K., AND BAILEY, E. M., JR. (1981) Res. Commun. Chem. Pathol. Pharmacol. 33,365-368. 5. CUMMING, R. B., AND WALTON, M. F. (1973) Food Cosmet. ToxicoL 11,547-553. 6. ITO, N., FUKUSHIMA, S., HAGIWARA, A., SHIBATA, M., AND OGISO, T. (1983) J. Natl. Cancer Inst. 70,343-349. 7. ITO, N., FUKUSHIMA, S., IMAIDA, K., SAKATA, T., AND MASUI, T. (1983) Gann 74,459-461. 8. DELLA CORTE, L., AND SGARAGLI, G. (1984) PharmacoL Res. Commun. 16,1041-1047. 9. TAKAHASHI, O., SAKAMOTO, Y., AND HIRAGA, K. (1985) Z’micol. Lett. 27,15-25. 10. SGARAGLI, G., DELLA CORTE, L., PULITI, R., DE SARLO, F., FRANCALANCI, R., GUARNA, A., DoLARA, P., AND KOMARYNSKY, M. (1980) Birr them. Pharmacol. 29,763-769. 11. VALOTI, M., DELLA CORTE, L., TIPTON, K. F., AND SGARAGLI, G. (1988) Biochem. J., 250,501-507.

METABOLITES

431

12. EPE, B., AND METZLER, M. (1985) Chem. BioL Interact. 56,351-361. 13. SUBRAHMANYAM, V. V., AND O’BRIEN, P. J. (1985) Xenobiotica 15,873-885. 14. SMART, R. C., AND ZANNONI, V. G. (1984) Mol. Phamnacol. 26,105-111. 15. AUCLAIR, C., DUGU& B., MEUNIER, B., AND PAOLETTI, C. (1986) Biochemistry 25,1240-1245. 16. RAHIMTULA, A. (1983) Chem. BioL Interact. 45, 125-135. 17. TAKAHASHI, O., AND HIRAGA, K. (1978) ToxicoL AppL Pharmacol. 43,399-406. 18. MARINO, A. A., AND MITCHELL, J. T. (1972) Proc. Sot. Exp. BioL Med. 140,122-125. 19. NAKAGAWA, Y., TAYAMA, K., NAKAO, T., AND HIRAGA, K. (1984) Biochem. PharmacoL 33, 26692674. 20. MIZUTANI, T., NOMURA, H., NAKANISHI, K., AND FUJITA, S. (1987) Toxicol. Appl. PharmacoL 87, 166-176. 21. INAI, K., KOBUKE, T., NAMBU, S., TAKEMOTO, T., Kou, E., NISHINA, H., FUJIHARA, M., YONEHARA, S., SUEHIRO, S., TSUYA, T., HORIUCHI, K., AND TOKUOKA, S. (1988) Japan J. Cancer Res. (Gann) 79,49-58. 22. OLSEN, P., MEYER, O., BILLE, N., AND W~RTZEN, G. (1986) Food Chem. ToxicoL 24,1-12. 23. WHO (1987) Food Additive Series, No. 21, pp. 2546, Cambridge Univ. Press, New York. 24. SHIGA, T., AND IMAIZUMI, K. (1975) Arch. Biothem. Biophys. 167,469-479. 25. HEWGILL, F. R., AND HEWITT, D. G. (1967) J. Chem. Sot. C, 726-730. 26. MASON, R. P. (1984) in Methods in Enzymology (Packer, L., Ed.), Vol. 105, pp. 416-422, Academic Press, San Diego. 27. CHANCE, B., AND MAHELY, A. C. (1955) in Methods in Enzymology (Colowick, S. P., and Kaplan, N. O., Eds.), Vol. 2, pp. 764-775, Academic Press, New York. 28. LAMBELET, P., DUCRET, F., SAUCY, F., SAVOY, M. C., AND L~LIGER, J. (1987) J. Chem. SOC., Faraday Trans. 183,141-149. 29. VALOTI, M., TIPTON, K. F., DELLA CORTE, L., AND SGARAGLI, G. P. (1985) Boll. SOC.Ital. Biol. SOW. 61,979-982. 30. HEWGILL, F. R., AND LEGGE, F. (1983) J. Chem. Sot. Perkin Trans. II, 653-659. 31. YAMAZAKI, I. (1977) in Free Radicals in Biology (Pryor, W. A., Ed.), Vol. 3, pp. 183-218, Academic Press, New York. 32. FREED, J. H., AND FRAENKEL, G. K. (1963) J. Chem. Phys. 38,2040-2041. 33. SCHREIBER, J., MOTTLEY, C., SINHA, B. K., KALYANARAMAN, B., AND MASON, R. P. (1987) J. Amer. Chem. Sot. 109,348-351. 34. THOMPSON, D. C., CHA, Y. N., AND TRUSH, M. A. (1986) in Biological Reactive Intermediates

432

35. 36. 37. 38.

VALOTI ET AL.

(Kocsis, J. J., Jollow, D. J., Witmer, C. M., Nelson, J. O., and Snyder, R., Eds.), Vol. III, pp. 301-309, Plenum Press, New York. THOMPSON,D.C., AND TRUSH,M.A.(~~~~) Food Chem. ToxicoL 24,1189-1195. R~~EGGE,D.,ANDFIscHER,H.(~~~~)J. Chem. SK Faraday Trans. 1,84,3187-3205. BECKER,H.-D.(1965)J. Org. Chem. 30,982-989. FRANCHI-MICHELI, S., DELLA CORTE,L., PULITI, R., GIOVANNINI,M. G.,ZILLETTI,L.,MARCONI,

M., AND SGARAGLI,G. (1980)Boll. Sot. Ital. BioL Sper. 56,2521-2524. 39. DELLA CORTE,L., BIANCHI, L., NINCI, R., SGARAGLI, G.,AND VALOTI,M.(~~~~) Brit. J. Pharmacd, 92, supp. I, 732 P. 40. DELLA CORTE,L., GIOVANNINI, M. G., AND SGARAGLI,G. (1984)Arch ToxiwL %mL 7,307-310. 41. GUARNA,A.,DELLACORTE,L.,GIOVANNINI,M.G., DE SARLO,F., AND SGARAGLI,G.(1983)DmLg Metab. I&p. 11,581-584.