Document MM8dJvL7nm203MoYmg9VMGKy9

0()26.895X/86/060674-l)6$02.00/() C(J)Yrght c l)y The American Society for lharn,acoIogy All rights ot reproduction in any form reserved. MOl.F.CI1.AR PHARMAcoloGY. 30:674-079 and Experimental Therapeutics Metabolic Activation of Phenol by Human Myeloperoxidase Horseradish Peroxidase and D. A. EASTMOND, M. T. SMITH, L. 0. RUZO, and D. ROSS Department of Biomedical and Environmental Health Sciences (D.A.E., M. T.S.) and Department of Entomology (L.O.R.), University Berkeley, California 94720; and Molecular and Environmental Toxicology Program, School of Pharmacy, University of Colorado, Boulder, Colorado 80309 (DR.) Received May 29, 1 986; Accepted September 1 9, 1986 of California, SUMMARY The oxidation of phenol catalyzed by human myeloperoxidase and horseradish peroxidase resulted in extensive binding of phenol-derived metabolites to boiled rat liver protein. This binding paralleled closely the removal of phenol from the incubations and was inhibited from 83 to 99% by the addition of the antioxidants, ascorbate and glutathione, suggesting that metabolism and bind- ing were occurring via a one-electron oxidation pathway. Meta- bolic studies employing both human myeloperoxidase and horse- radish peroxidase and diphenoquinone resulted in the identification of 4,4'-biphenol as the principal identifiable metabolites. The addition of reduced glutathione to incubations containing horse- radish peroxidase resulted in the formation of two conjugate species. These conjugate species were identified by fast atom bombardment mass spectrometry to be glutathione conjugates of diphenoquinone. The major gluthathione conjugate was iden- tified as 3-(glutathion-S-yl)-4,4'-biphenol by NMR spectroscopy. These results suggest that the formation of highly reactive species through the peroxidase-mediated metabolism of phenol and other phenolic compounds could play an important role in the hematopoietic toxicity observed during chronic benzene ex- posure. Chronic exposure to benzene has been shown to lead to numerous blood and bone marrow disorders including pancy- topenia, aplastic anemia, and leukemia (1, 2). Although the actual mechanism by which benzene exerts its hematopoietic effects appears to be complex, it is generally accepted that metabolic activation is required. Coadministration of toluene, a competitive substrate with benzene for cytochnome P-450 monooxygenase enzymes, and benzene reduced the levels of benzene metabolites in the bone marrow without affecting the benzene levels and resulted in protection against toxicity (3). Several studies have investigated cytochnome P-450 metabo- lism in the bone marrow and found the levels of cytochrome P- 450 and the rates of benzene metabolism to be very low; such levels could not account for all of the accumulation and binding of radiolabeled compounds in the bone marrow (4-9). In addi- tion, Sammett et al. (10) demonstrated that partial hepatectomy This work was supported by the National Foundation for Cancer Research (M. T. S. and D. R.) and National Institute of Environmental Health Sciences Grant POIES00049 (L. R.). This work was submitted by D. A. E. in partial fulfillment of the requirements for the Ph.D. degree in Environmental Health Sciences from the University of California, Berkeley, where he was supported in part l)y the University of California Toxic Substances Program and National Institute for Occupational Safety and Health traineeship grants. These results have been presented in part at the Society of Toxicology LA, March 3-7, 1986. meetings, New Orleans, in rodents also protected against benzene toxicity. These results suggested that metabolism within the liver was essential for toxicity but that toxicity was elicited distal from the liven in the bone marrow. A possible explanation for this is that stable benzene metabolites such as phenol, catechol, and hydroqui- none travel from the liver to the bone marrow and there exert their toxic effects. Intravenous administration of the known benzene metabolites has failed to produce the same pattern of toxic effects observed after the administration of the parent compound (11, 12). Thus, it seems probable that it is not the hepatic metabolites of benzene which induce a myelotoxic response but the products of their further biotransformation within the bone marrow itself. Bone marrow contains appreciable amounts of MPO which could be involved in the localized activation of benzene's me- tabolites (13). The target organ specificity demonstrated by benzene in recent animal cancinogenesis bioassays (Zymbal gland, Hardenian gland, as well as others; Ref. 14) would also indicate a role for peroxidases in the metabolic activation of benzene since these target organs are known to contain signifi- cant penoxidase levels (15, 16). Evidence for the involvement of peroxidases in the one- electron oxidation and metabolic activation of numerous xe- nobiotics has accumulated in recent years (17, 18). The penox- ABBREVIATIONS: MPO, myeloperoxidase; HAP, horseradish peroxidase; GSH, reduced glutathione; TCA, trichloroacetic liquid chromatography; FAB, fast atom bombardment; MS, mass spectrometry; GC, gas chromatography. 674 acid; HPLC, high pressure Metabolic Activation of Phenol by MPO and HRP 675 idase activity of prostaglandin synthase has been implicated in the nephrotoxicity induced by phenacetin (19) and acetamino- phen (20) and in the induction of bladder cancer by nitrofurans (21) and aromatic amines (22). Peroxidases appear to be in- volved in the uterine toxicity and renal carcinogenicity induced by diethylstilbestnol (23) as well as in the induction of Harder- ian gland tumors by benzidine (24) and Zymbal gland tumors by trans-4-aminostilbene (15). Phenol, the principal metabolite of benzene in vivo (1), has been shown to be a good reducing cofactor in the reduction of hydrogen peroxide by penoxidases. Early in vitro studies by Danner et al. (25) and Sawahata and Neal (26) have shown that phenol is metabolized to 2,2'-biphenol, 4,4'-biphenol, and diphenoquinone by HRP. Additional reports have indicated that phenol is converted to DNA- and protein-binding species during peroxidase-mediated metabolism and that the exoge- nous addition of antioxidants and sulfhydryl reagents had a protective effect (27, 28). Recent studies within our laboratory have demonstrated that protein binding ofphenol occurs during the oxidative burst of human neutrophils and that this binding was due to peroxidase-mediated metabolism.' The purpose of this study was to investigate further the penoxidase-mediated metabolism of phenol by human MPO and HRP, to study the time course of protein-binding, and to identify the reactive protein-binding species. Materials and Methods Chemicals and enzymes. Phenol, 2,2'-biphenol, 4,4' -biphenol, and ascorbic acid were purchased from Aldrich Chemical Co., Milwaukee, WI. GSH, HRP type VI (275 units/mg) (EC 1.11.1.7), H2O2 (30% solution), and guaiucol were obtained from Sigma Chemical Co., St. Louis, MO. Catalase (65,000 units/mg) was obtained from Boehringer Mannheim. [`4CJPhenol (ring-UL; 30.5 mCi/mmol) was purchased from Pathfinders Laboratories, St. Louis, MO. [3H]GSH (glycine2:lH; 1000 mCi/mmol) was obtained from New England Nuclear, Boston, MA. All other chemicals or solvents were generally the highest grade available and purchased through local commercial suppliers. Deionized water was purified using a Millipore Q system obtained from Millipore, Bedford, MA. Preparation of human MPO. Human neutrophils were isolated as described by Markert et a!. (29). After sonication by a Braunsonic cell sonicator (for 10 mm at 70 W) and centrifugation (550 x g for 5 mm), the supernatant was assayed for peroxidase activity as described by Klebanoff et a!. (30). This MPO was stored at -20 and, after an initial drop in activity, was found to be relatively stable during the period of experimentation. Standard incubation mixture. The standard incubation mixture consisted of 500 MM phenol, 10 pg/ml HRP or 1.5 units/ml MPO, and 1 mM H2O2 in 100 mM phosphate buffer (pH 7.4). Reactions were initiated by the addition of HO2 and were performed at 3T in a shaking water bath. For the metabolic studies and fraction collections, the incubation was stopped at 2 mm (for HRP) or 3 mm (for MPO) by the addition of 650 units of catalase. In reactions to identify the GSH conjugate, this quenching with catalase was followed 15 sec later by the addition of 5 mM GSH. Protein binding experiments. Rat liver 9000 x g postmitochon- drial supernatant was prepared as described previously (31) and boiled for 30 mm. An aliquot of this preparation containing 0.9 mg of protein was added to the standard incubation mixture and incubated for various time periods up to 30 mm. All treatment additions were performed prior to H2O2 addition. The reaction mixture was quenched by the I D. A. Eastmond, ixihlicat ion. R. C. French, D. Ross, and M. T. Smith, submitted for addition of TCA (5% final concentration). Samples were maintained on ice and centrifuged prior to analysis by HPLC with electrochemical detection. Covalent binding to protein was determined by liquid scm- tillation counting following the procedure of Jollow et al. (32) with the following modifications. The precipitated protein was washed several times with ice-cold ethanol:ether (1:1) in addition to several washes with 5% TCA and methanol:H20 (4:1). To help solubilize the protein, Tris buffer containing 0.25% sodium lauryl sulfate was used in combi- nation with the NaOH treatment. The pH of the solubilized protein was adjusted to neutral pH before aliquots were removed for liquid scintillation counting and protein determination as described by Lowry et al. (33). HPLC with electrochemical detection. The apparatus employed in these studies consisted of an isocratic reverse phase HPLC system (Beckman model 100A) with an amperometric detector (Bioanalytical Systems LC-4A) equipped with a glassy carbon working electrode (BAS TL-5) and an Ag/AgC1 reference electrode. A working potential of +1.0 V was used. The system employed a 25 cm X 4.6 mm id. C-18 column (5 zm), from Sulpelco, and a Rheodyne injector (model 7125) with a lo-Ml injection loop. The mobile phase consisted of 90% 0.1 M ammo- nium acetate buffer (pH 4.0) and 10% acetonitrile with a flow rate of 2 mi/mm. Quantitation of phenol was performed by comparison to standard curves based on area counts as determined by a Vanian Vista 401 chromatography data system. HPLC with UV detection. A gradient HPLC system consisting of Beckman (model 100A) and Altex (model 100) pumps controlled by a Beckman microprocessor (model 420), an Altex sample injector (model 210), a Supelco 25 cm x 4.6 mm C-18 column (5 Mm), a Perkin-Elmer LC-75 spectrophotometric detector (254 nm), and a Perkin-Elmer computing integrator (model M-1) was employed for all metabolic studies and fraction collection. A lOO-Ml loop was used for all studies except those employing fraction collection for FAB-MS, NMR, and peak collection with tnitiated glutathione which employed a 1000-sl loop. The mobile phase used was 90% acetonitrile (A) and 0.1 M ammonium acetate buffer (B; pH 4.0), and the flow rate was 1 ml/min. Conditions were 93% buffer B for the first 10 mm, after which a linear gradient was employed over the next 20 mm to reach a final concentra- tion of 10% buffer B. Gas chromatography/mass spectrometry. Analysis formed on ethyl acetate extracts of the standard incubations near-dryness under N2 and derivatized with diazomethane. was pertaken to The MS experiments utilized a Hewlett-Packard 5985B instrument equipped with an HP1000 computer and a model 5840A gas chromatograph. Separations were accomplished column (10 m) with temperature on a high performance methyl silicone programming (80-240, 20/min) with helium carrier gas (1 mI/mm). Analysis was by selected ion monitoring using chemical ionization with methane (0.8 torr) as ionizing gas. FAB was carried out with xenon at 8 kV on a ZAB-VG instrument with the samples dissolved in a glycerol matrix. Nuclear magnetic resonance. NMR was accomplished on a Banker 300 MHz instrument equipped with an Aspect 3000 computer. Samples were dissolved in D2O and chemical shifts (11) were determined relative to acetonitrile. Statistical analyses. Following one-way analysis of variance on the log transformed data, statistical significance was determined using a one-tailed Dunnett's t test for multiple comparisons. Critical values were calculated using a 0.05 probability of type 1 error. Results The extent of HRP- and MPO-catalyzed oxidation of phenol to reactive species which bind to boiled rat liver protein is shown in Fig. 1. Extensive binding occurred very rapidly and paralleled very closely the removal of phenol from the incuba- tion, indicating that an early reaction product was responsible for the binding. This binding was demonstrated to be H202 and peroxidase dependent and was 83-99% inhibited by the addi- 676 Eastmond et a!. u U ot a to Minutsi Fig. 1. Protein binding (0) and substrate removal (#{14d9u}r)ing the HRP (A)and human MPO (B)-catalyzed oxidation of phenol. Incubation conditions are described in Materials and Methods. The data represent the means and standard deviations of three experiments. tion of ascorbate or GSH to the incubation (Table 1). This binding was also shown to be protein dependent. When boiled protein was omitted from the incubations, the recovered radio- activity after TCA precipitation and solvent washes was less than 5% of that recovered in incubations containing protein. Phenol removal from the incubations was not observed with asconbate and glutathione treatments, suggesting that these compounds were acting as antioxidants by reducing the phe- noxy radical back to phenol. To understand further the peroxidase-mediated metabolic pathways and to identify the binding species, various additional analytical approaches were employed. The HRP- or MPO- catalyzed oxidation of phenol was accompanied by the forma- tion of a yellow chromophone (Xnax 399 nm) and brown polymeric compounds. This yellow chromophore corresponds to the spectrum reported for diphenoquinone (26). Analyses of either of these enzymatic oxidation mixtures employing HPLC with UV detection and monitoring at 399 nm failed to detect a peak corresponding to diphenoquinone. Monitoring at 254 nm, HRP-catalyzed oxidations showed one principal peak produced during the incubation (Fig. 2A). This peak co-chromatographed with a known standard for 4,4'-biphenol. Identical incubations performed with `4C-phenol, in which 1-mm fractions eluting from the HPLC were collected and analyzed by liquid scintil- lation counting, showed that the radioactivity co-eluted pni- manly with 4,4'-biphenol and phenol (Fig. 2A). In control experiments lacking HRP, virtually all of the radioactivity co- eluted with phenol (Fig. 2C). Confirmation of these results using HRP as an enzyme source was obtained by GC-MS. Analysis of the incubation products after extraction in ethyl acetate and denivatization by diazomethane showed one pnin- cipal product which was identical in terms of chromatic reten- tion time and mass spectrum to a denivatized standard of 4,4'- biphenol. Products corresponding to 2,2'-biphenol and diphen- oquinone were also observed, but at levels only barely distin- guishable from background. Similar results, but of a lesser magnitude, were obtained in incubations containing MPO. Since the metabolic profiles for HRP and human MPO ap- peared to be qualitatively identical, further analyses employed only HRP as the enzyme source. In an attempt to trap the binding species, GSH was added to HRP incubations after 2 mm of reaction. An immediate de- crease in the yellow colon of the incubation was observed. Analysis by HPLC with UV detection after the addition of glutathione resulted in a significant decrease in the 4,4'-bi- phenol peak with the formation of two additional 2, Fig. 2B). Incubations using `4C-phenol resulted peaks ( 1 and in a decrease in radioactivity co-eluting with 4,4'-biphenol and increases in radioactivity co-eluting with the two new peaks 1 and 2 (Fig. 2B). Incubations using :lHglutathione HRP, and unlabeled phenol showed major increases in radioactivity associated with the two new peaks. These data suggested that peaks 1 and 2 were glutathione conjugates but did not provide any evidence for the identity of the conjugating species. Authentic 4,4'- biphenol remained unchanged when mixed with glutathione and did not produce the putative glutathione conjugates rep- resented by peaks 1 and 2. When 4,4'-biphenol, HRP, and glutathione were mixed, no removal of 4,4'-biphenol non the appearance of peaks 1 and 2 was observed, whereas identical incubations including H202 in addition to the 4,4'-biphenol, HRP, and glutathione resulted in the formation of the putative TABLE 1 Effect of various treatments on the HRP- and MPO-cata lyzed met abolism of phenol Additions Phenol equivalent bound HRP/H202 Phend remazsng Phenol equivalent bound nmo!/mg nmo!/mI nmoi/mg Nonea 34866 1.40.7b 10019 -H202 5 2C 523 39 7 1c -Peroxidase -Phenol 4 1 1 1 ND" ND 7 1C 1 OC +1 mM Ascorbate 60 24C 402 1 1 2 12 1 C +5 m GSH 3 1c 478 79 1 2 4C a Standard incubation conditions are described in Materials and Methods. a This value was determined after a 5-mm incubation. All other values were determined after 30-mm incubations. C These values differ significantly a ND, not determined. from the control (none) values, p < 0.05. MPO/H202 Phend remaining nmol/mI 410 18 478 30 ND ND 465 24 447 21 Metabolic Activation of Phenol by MPO and HRP A. 677 10 .. slit Giycsiol . M Sodium \Adducts .,.` Glycerol 0 C Adducls - 01 491 and 492 Adducl a) 5 1.4 4 1 49 1 + Glycerol .? E ,,, 1I3 1 / `WI.? LLi 0 `` Ik.J.I + 51.1 ei.3 .1..L..I.. .. .. 513 4 Glycerol "i.` . . I i, 500 550 600 650 M/Z > Eiution Volume (ml) Fig. 2. UV/14C HPLC elution profiles: A, HRP-catalyzed oxidation of phenol; B, a similar incubation with the addition of 5 mM glutathione after 2 mm of reaction; and C, an incubation similar to that in A but lacking HAP. Incubation conditions are described in Materials and Methods. TABLE 2 Profile of the metabolites oxidation of phenol identified during the HRP-catalyzed 14C.Phen 3H-GSH nmol nmol Standard incubationa 4,4'-Biphenol 23.4 Standard incubation + 4,4'-Biphenol Peak 1 a Peak2 lncubations contained 500 M GSHb phenol, 4.7 4.1 4.0 11.8c 11.9 10 pg/mI HAP, and 1 mM H202 in 0.1 M phosphate buffer (pH 7.4). The reactions were stopped at 2 mm by the addition of catalase. a Standard incubation conditions were followed but glutathione (5 mM) was added 1 5 sec after the catalase addition. Catalase (650 units/mI) was added at 2 mm for the 14C incubations, whereas 13,000 units/mI catalase were added at 2 mm for the 3H incubattons. Under similar conditions containing protein, 96-98% of the phenol was removed from the incubations and 60% of the phenol equivalents were recovered as protein bound (see Fig. 1A). C This value was calculated after correction for unmetabolized phenol collected in the same fraction. conjugates and removal of 4,4'-biphenol. These data show that the conjugating species was not 4,4'-biphenol itself but a com- mon peroxidative oxidation product of both 4,4'-biphenol and phenol. This compound was presumably diphenoquinone, which could be easily reduced to 4,4'-biphenol during HPLC and GC-MS analyses. This quinone could then react directly with glutathione to form the two conjugates. A summary of the metabolic profile for the HRP/H202-catalyzed oxidation of phenol is shown in Table 2. The HPLC eluate containing each of the conjugate peaks was frozen and lyophilized before being subjected to FAB-MS and NMR. The identities of peaks 1 and 2 were confirmed as glutathione conjugates of diphenoquinone by FAB-MS and NMR spectroscopy (Fig. 3). The mass spectrum of peak 2 (Fig. 3A) exhibits diagnostic masses for a glutathione addition prod- uct of 4,4'-biphenol as follows: M and M + 1 at m/z 491 and 492, corresponding sodium adducts on NaCl addition at m/z 513 and 514, and a glycerol adduct at m/z 575. In addition, the base peak (not shown) was at m/z 307 (glutathione). Peak 1 gave a similar FAB spectrum. Examination of the downfield region of the NMR spectrum of peak 2 confirms the presence of a substituted 4,4'-biphenol (Fig. 3B) indicating one unsub- B. H \ Ha j , i: Hc H5 : Hb OH Hy HcLH H HbL(LsG OH 8.0 7.0 6.0 PPM (8) Fig. 3. FAB mass spectrum (A) and NMR spectrum of the aromatic region (B) of peak 2, the major glutathione conjugate isolated by HPLC. Analytical conditions are described in Materials and Methods. Coupling constants in Hz are as follows: H = 8.4; H0 = 8.4; Hb = 9.2; and H = 9.2. The upfield region of the NMR spectrum (2-5 ppm) exhibits resonances consistent with the presence of the glutathione moiety. stituted contains singlet tenistic though material, adduct ring (H and H doublets) and that the other ring a substituent at the 3-position since Ha appears as a and H5 (partially obscured) and Hc exhibit the charac- ortho-coupling for the structure shown (Fig. 3B). Al- NMR is not available for peak 1 due to the paucity of it is proposed to be the corresponding glutathione at the 2-position. Discussion The peroxidase-catalyzed binding of phenol to protein and DNA has been reported previously by several investigators (9, 26-28, 34, 35)Y In these studies we have confirmed these reports and shown that extensive phenol binding occurs very rapidly with human MPO as well as with HRP and parallels very closely the removal of phenol from the incubation. The de- pendence of this binding on H202, peroxidase, phenol, and protein demonstrates that the formation of these reactive me- tabolites was peroxidase mediated and indicates that these metabolites were actually protein bound. The inhibition of binding observed when ascorbate or glutathione was included in the incubation could be due to their antioxidant properties on to their ability to act as competitive substrates for the peroxidase enzymes. In recent experiments, Subnahmanyam and O'Brien (34) reported that little direct oxidation of ascor- bate took place in H207 and HRP incubations in the absence 2 D. A. Eastmond. publication. R. C. French, D. Ross, and M. T. Smith, submitted for 678 Eastmond et a!. ofphenol. However, in the presence ofphenol, a rapid oxidation of asconbate took place (34). In addition, other studies have failed to demonstrate an ESR signal from glutathione directly during HRP/H202 incubations but have shown a thiyl radical signal when the phenolic cluded in the incubation compound (36). These diethylstilbestrol results indicate was inthat both ascorbate and glutathione are functioning as antioxidants and are reducing the phenoxy radical back to phenol. In these experiments, 4,4'-biphenol and diphenoquinone were the principal identifiable reaction products, whereas a trace amount of 2,2'-biphenol was detected by GC-MS. Other investigators have generally observed similar results, although 2,2'-biphenol was usually a more prominent metabolic product (25, 26, 34, 35). The reason for these differences in relative ratios of metabolic products is most likely the slight modifica- tions in experimental conditions. In these experiments nela- tively high concentrations of HRP were used with a theoretical excess of H2O2 in order to effect a total removal of phenol. Due to the high affinity of 2,2'-biphenol for HRP and a reported increase in reaction rate in the presence of phenol (25), it is possible that the 2,2'-biphenol was formed rapidly during the incubation and subsequently converted to polymeric products which were not detected in our analyses. The use of the nucleophilic tnipeptide glutathione as an agent to trap binding species during metabolism has been reported previously (37). Our use of glutathione as a trapping agent resulted in the formation of two glutathione diphenoquinone, 3-(glutathion-S-yl)-4,4'-biphenol, conjugates of and an- other product, probably 2-(glutathion-S-yl)-4,4'-biphenol. The identification of 3-(glutathion-S-yl)-4,4'-biphenol as the pnin- cipal glutathione conjugate was based upon MS, radioisotope, and NMR results, whereas the structure of the secondary conjugate, 2-(glutathion-S-yl)-4,4' -biphenol, was tentatively identified based upon MS and nadioisotopic evidence. The conclusive identification of two species as glutathione conju- gates of diphenoquinone represents the first definitive identi- fication of a binding metabolite formed during the peroxidase- mediated metabolism of phenol. Although our HPLC results indicated that, in incubations containing HRP, 96-98% of the phenol was removed in 2 mm, the subsequent trapping diphenoquinone yielded only 6% of the phenol equivalents of as glutathione conjugates. Under these conditions, 60% of the phenol equivalents were recovered as protein bound, indicating that other species formed during HRP/H202-catalyzed metab- olism were responsible for most of the binding. These other species could possibly be the phenoxy or C-centered radicals, further oxidation products of 2,2'-biphenol or 4,4'-biphenol, or other polymerization products. Subrahmanyam and O'Brien (34, 35) have recently studied the HRP/H2O2-catalyzed oxida- tion of phenol and its binding to DNA. Their results indicate that a polymerization product of 2,2'-biphenol is a major DNA- binding species formed from phenol. A summary of the known metabolites formed during peroxidase-mediated metabolism of phenol and possible routes for the formation of binding metab- olites is shown in Fig. 4. The demonstration that highly reactive binding species are formed from phenol during peroxidase-mediated metabolism could be of importance in understanding the hematopoietic toxicity product of benzene since phenol is the principal metabolic formed from benzene in vivo (1) and the formation of DNA- and protein-binding products are often related to cyto- lJl P.rouMase c.:: OH Perozldsss 105 =-[ tt:_ip ] OH HO_4rj.__tl_. OH [ 1HO_O(__.Q.o:]_.__.. : HO OH HO_fl( o Fig. 4. Proposed scheme for the peroxidase-mediated metabolism of phenol and the reaction of diphenoquinone with glutathione to form the two identified glutathione conjugates. toxic and genotoxic effects in cells (38, 39). There have been very few studies to examine the direct cytotoxic and genotoxic effects of the peroxidase-mediated metabolites of phenol. Ni- shioka and Ogasawara (40) reported that 2,2'-biphenol but not 4,4'-biphenol exhibited mutagenicity in Escherichia coli when tested in the DNA repair test. These investigators also reported that neither 2,2'-nor 4,4'-biphenol were mutagenic in strains TA98 and TA100 in the Ames Salmonella reversion assay. Recently, Erexson et al. (41) studied the effects of diphenoqui- none, and 2,2'- and 4,4'-biphenol on the induction of sister chromatid exchange, changes in mitotic indices, and interfer- ence with cell cycle kinetics in human lymphocytes. The 2,2'- and 4,4'-biphenols induced slight but significant increases in sister chromatid exchange frequency, reductions in mitotic activity, and inhibition of cell cycle progression, whereas di- phenoquinone caused only a reduction in mitotic activity. In addition, 4,4' -biphenol and diphenoquinone caused cytotoxicity to the cultured lymphocytes at fairly low concentrations, which suggests that these metabolites could be contributing to bone marrow cytotoxicity in vivo. In summary, we present evidence to show that phenol is converted to highly reactive protein-binding species during metabolism catalyzed by human MPO and by HRP and that the mechanism probably involves the formation of a free radical intermediate. The penoxidase-mediated metabolism of phenol was shown to result in the formation of 4,4'-biphenol and diphenoquinone and that diphenoquinone is one of the binding species produced during metabolism. The identification of the other binding species and the relevance of these findings to the hematopoietic toxicity of benzene in vivo will require further investigation. Acknowledgments We would like to thank drawing the blood. Dr. Janice Yager and Ms. Rosalie Moos-Hollings for References 1. Snyder, R., and J. J. Kocsis. Current concepts of chronic benzene toxicity. Crit. Rev. Toxicol. 3:265-288 (1975). 2. Goldstein, B. D. Hematotoxicity in humans. J. Toxicol. Environ. Health Suppl. 2:69-105 (1977). 3. And.rews, L. S., E. Woo Lee, C. M. Witmer, J. J. Kocsis, and R. Snyder. Effects of toluene on the metabolism, disposition and hemopoietic toxicity of l3H)benzene. Biochem. Pharmacol. 26:293-300 (1977). 4. Andrews, L. S., H. A. Sasame, and J. R. Gillette. 3H-Benzene metabolism in rabbit bone marrow. Life Sci. 25:567-572 (1979). 5. Irons, R. D., J. G. Dent, T. S. Baker, and D. E. Rickert. Benzene is metabolized and covalently bound in bone marrow in situ. Chem. -Bzol. Interact. 30:241-245 (1980). 6. Irons, R. D., W. F. Greenlee, D. Wierda, and J. S. Bus. Relationship between benzene metabolism and toxicity: a proposed mechanism for the formation of reactive intermediates from polyphenol metabolites. 136A:229-243 (1982). 7. Gollmer, L., H. Graf, and V. Ullrich. Characterization oxygenase system in rabbit bone marrow. Biochem. 3602 (1984). Adu. Exp. Med. Biol. of the benzene mono- Phormacol. 33:3597- 8. Irons, R. D. Quinones as toxic metabolites of benzene. J. Toxzcol. Environ. Health 16:673-678 U985). 9. Sawahata, T., D. E. Rickert, and W. F. Greenlee. Metabolism of benzene and its metabolites in hone marrow, (R. D. Irons, ed). Raven Press, 10. Sammett, D., E. W. Lee, J. J. in Toxicology New York, Kocsis, and of the Blood and 141 (1985). R. Snyder. Partial Bone Marrow hepatectomy reduces both metabolism and toxicity of benzene. J. Toxicol. Environ. Health 5:785-792 (1979). 1 1 . Snyder, R., and B. N. Engelsberg. 72 hr 59Fe utilization in Swiss (1982). Effects of benzene and other chemicals albino mice. Toxicologist 2:121, abstr. on 428 12. Mitchell, .J. R. Mechanism of benzene-induced 30:561,abstr. 2044 (1971). 13. Himmeloch, S. R., W. H. Evans, M. G. Mage, aplastic anemia. and E. A. Peterson. Fed. Proc. Purification of myeloperoxidases from the bone marrow of the guinea pig. Biochemistr.y 8:914-921 (1969). 14. Haseman, J. K., D. D. Crawford, J. E. Huff, G. A. Boorman, and E. E. McConnell. the National (1984). Results from 86 two-year carcinogenicity studies conducted by Toxicology Program. J. Toxicol. Environ. Health 14:621-639 15. Osborne, J. C., M. Metzler, and H.-G. Neumann. Peroxidase activity in the rat Zymbal gland and its possible role in the metabolic activation of amino- stilbenes 16. Morrison, in the target tissue. Cancer Lett. 8:221-226 (1980). M., and P. Z. Allen. Lactoperoxidase: identification and isolation from Harderian and lacrimal glands. Science (Wash. D. C.) 152:1626-i628 ( I966). 17. Marnett, L. .J., and T. E. Eling. Cooxidation during prostaglandin biosyn- thesis: a pathway for the metabclic activation of xenobiotics, in Reuzeus in Biochemical Toxicology (E. Hodgson, J. R. Bend, and R. M. Philpot, eds.), Vol. 5. Elsevier, New York, 135 (1983). 18. Eling, T., .J. Boyd, G. Reed, R. Mason, and K. Sivarajah. Xenobiotic metab- olism by prostaglandin endoperoxide synthetase. Drug Metab. Rev. 14:1023- 1053 (1983). 19. Andersson, B., M. Nordenskjold, A. Rahimtula, and P. Moldeus. Prostaglan. din synthetase-catalyzed activation of phenacetin metabolites to genotoxic products. 20. Moldeus, Mol. Pharmacol. P., B. Andersson, 22:479-485 (1982). A. Rahimtula, and M. Berggren. Prostaglandin synthetase-catalyzed activation of paracetamol. Biochem. Pharmacol. 31:1363-1368 (1982). 21. Cohen, S. M., T. V. Zenser, G. Murasaki, S. Fukushima, M. B. Mattammal, N. S. Rapp, and B. B. Davis. Aspirin inhibition of N-l4-(5-nitro-2-furyl)-2. thiazolyllformamide-induced lesions of the urinary bladder correlated with inhibition of metabolism by bladder prostaglandin endoperoxide synthetase. Cancer Res. 4 1 :3355-3359 (1981). 22. Wise, R. W., T. V. Zenser, F. F. Kadlubar, and B. B. Davis. Metabolic activation of carcinogenic aromatic amines by dog bladder and kidney pros- taglandin H synthase. Cancer Res. 44:1893-1897 (1984). 2:3. Metzler, M. Biochemical toxicology of diethylstilbestrol, in Reviews in Bio- Metabolic Activation of Phenol by MPO and HRP 679 chemical Toxicology (E. Hodgson, J. R. Bend, and R. M. Philpot, eds.), Vol. 6. Elsevier, New York, 191 (1984). 24. Tsuruta, Y., V. V. Subrahmanyam, W. Marshall, and P. J. O'Brien. Peroxi- dase-mediated irreversible binding of arylamine carcinogens to DNA in intact polymorphonuclear leukocytes Interact. 53:25-35 (1985). 25. Danner, D. ,J., P. J. Brignac, activated by a tumor promoter. Chem. -Biol. D. Arceneaux, and V. Patel. The oxidation of phenol and its reaction peroxide. Arch. Biochem. product by horseradish Biophy.s. 156:759-763 peroxidase (1973). and hydrogen 26. Sawahata, T., and R. A. Neal. Horseradish peroxidase-mediated oxidation of phenol. Biochem. Biophys. Res. Commun. 109:988-994 27. Smart, R. C., and V. G. Zannoni. DT-Diaphorase the covalent binding of the metabolites of phenol, benzene. Mol. Pharmacol. 26:105-111 (1984). (1982). and peroxidase influence the major metabolite of 28. O'Brien, P. J. One electron oxidation in cell toxicity, in Microsomes and Drug Oxidation. (A. R. Boobis, J. Caldwell, F. De Matteis, and C. R. Elcombe, eds.). Taylor and Francis, London, 284 11985). 29. Markert, M., P. C. Andrews, and B. M. Babior. Measurement of production by human neutrophils. The preparation and assay of NADPH oxidase-containing particles from 105:358-365 (1984). 30. Klebanoff, S. J., A. M. Waltersdorph, human neutrophils. and H. Rosen. Methods Antimicrobial Enz,vmol. activity of myeloperoxidase. Methods Enzymol. 105:399-403 (1984). 31. Ernster, L., P. Siekevitz, and G. E. Palade. Enzyme-structure relationships in the endoplasmic reticulum of rat liver. J. Cell Biol. 15:541-562 (1962). 32. Jollow, D. J., J. R. Mitchell, W. Z. Potter, D. C. Davis, J. R. Gillette, and B. B. Brodie. Acetaminophen-induced hepatic necrosis. II. Role of covalent binding in vivo. J. Pharmacol. Exp. Ther. 187:195-202 (1973). 33. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275 (1951). 34. Subrahmanyam, V. V., and P. J. O'Brien. Peroxidase-catalyzed binding of [U-'4C]phenol 35. Subrahmanyam, to DNA. Xenobiotica 15:859-871 V. V., and P. J. O'Brien. Phenol (1985). oxidation product(s), formed by a peroxidase reaction, that bind to DNA. Xenobiotica 15:873-885 (1985). 36. Ross, D., R. J. Mehlhorn, P. Moldeus, and M. T. Smith. Metabolism of diethylstilbestrol by horseradish peroxidase and prostaglandin-H synthase. J. Biol. Chem. 260:16210-16214 (1985). 37. Ross, D., R. Larsson, K. Norbeck, R. Ryhage, and P. Moldeus. Characteri- zation and mechanism of formation of reactive products formed during peroxidase-catalyzed oxidation of p-phenetidine. Mo!. Pharmacol. 27:277- 286 (1985). 38. Jollow, D. J., J. R. Mitchell, N. Zampaglione, and J. R. Gillette. Bromoben- zene-induced liver necrosis. 3,4-bromobenzene oxide 11:151-169 (1974). Protective role of glutathione as the hepatotoxic metabolite. and evidence Pharmacology for 39. Bresnick, E., and A. Eastman. enceofadducts, and relationship Alkylation of mammalian cell DNA, persist- tocarcinogenesis. I)rugMetab. Rev. 13:189- 205 (1982). 40. Nishioka, H., and H. Ogasawara. tives in bacterial systems. Mutat. Mutagenicity testing for diphenyl Res. 54:248-249 (1978). deriva- 41. Erexson, G. L., J. L. Wilmer, and A. D. Kligerman. Sister chromatid exchange induction in human lymphocytes exposed to benzene and its metabolites in vitro. Cancer Res. 45:2471-2477 (1985). Send reprint requests to: Dr. Martyn T. Smith, Department and Environmental Health Sciences, 322 Warren Hall, University Berkeley, CA 94720. of Biomedical of California,