Document VjjGnJok9o220KdKb05BoQ8Y4

DownloadSearchable PDFRandom document
TOXICOLOGY AND APPI IED PHARMACOLOGY 59, 187- 195 (1981) l UHIBIT A Proposed Mechanism of Benzene Toxicity: Formation of Reactive I ntermed i ates from PoIy phenoI MetaboIitesi WlLLlAM F. GKEENLEE,"' JAMES D. s.A N D JAMES Bus5 Cheriiicul Indrrsiry Iirsriiirie vf Toxicology, Depurtmenis of Paihology urid Biocheinicul Toxicology, Rcsetrrcli Triungle Purk. North Curolino 27709 z Rrcrived July 14, 1980; trcceprrd Jrinrtury 12, 1981 A Proposed Mechanism of Benzene Toxicity: Formation of Reactive Intermediates from Polyphenol Metabolites. GREENLEEW, . F., SUN,J . D.,AND Bus, J . S . (1981). Toxicol. Appl. Phurmacol. 59, 187-195. Male Fischer-344 rats were given 100 pCi (14 mg/kg) ["C]catechol or ["C]hydroquinone by injection into the lateral tail vein. For a period of at least 24 hr, soluble radioactivity associated with either compound was retained in the bone marrow, but not in the liver or thymus. The amount ofcovalently bound radioactivity increased with time in all tissues examined and was significantly depressed in liver, white blood cells, and bone marrow i n rats pretreated with Aroclor 1254, a regimen which protects against benzene toxicity. Potential enzymatic and nonenzymatic activation pathways for catechol, hydroquinone, and other known benzene metabolites were examined. In air-saturated 50 mbi phosphate buffer (pH 7.4) at 3 7 T , only hydroquinone and 1,2,4-benzenetriol autoxidized. The oxidation product of hydroquinone had an uv absorption maximum (248 nm) identical to that of benzoquinone. With 250 units superoxide dismutase, hydroquinone autoxidation increased fivefold, whereas the oxidation of 1,2,4-benzenetriol was inhibited (4% of control). Epinephrine autoxidation, an indirect measure of superoxide anion generation, was stimulated by 1,2,4-benzenetriol and hydroquinone, but was barely detectable in the presence of catechol. Of the compounds studied, only benzoquinone augmented the oxidation of NADPH by a 3000g rat bone marrow supernatant. These data support a mechanism for benzene toxicity in which the formation of potentially cytotoxic metabolites, semiquinone, and quinone oxidation products and superoxide radicals, result from autoxidation of at least two polyphenol metabolites of benzene, hydroquinone, and I ,2,4-benzenetriol. Chronic exposure to concentrations of benzene ranging from 25 to 1000 ppm results in a progressive degeneration of the hemopoietic system (Carpenter er NI., 1944; Hough and Freeman, 1944) and immune I Presented in part at the August Meeting of the American Society for Pharmacology and Experimental Therapeutics, August 17-2 I , 1980. CIIT Postdoctoral Fellow. Present address: Laboratory 0 1 Toxicology, Harvard School of Public Hesrlth, 665 Huntington Avenue, Boston, Mass. 021 15. ' Present address: ITRI, Lovelace Biomedical and Environmental Research Institute, P.O. Box 5890, Albuquerque, N.M. 87115. To whom all correspondence should be sent. dysfunction (Goldwater, 1941; Irons and Moore, 1980). A number of studies indicate that the expression of benzene toxicity requires metabolism of the parent compound to one or more toxic species (for reviews see Snyder and Kocsis, 1975; Laskin and Goldstein, 1977;Cohen rr ul., 1978).Several metabolites including phenol, catechol, hydroquinone, I ,2,4-benzenetriol, and benzene dihydrodiol have been identified in the urine of rats given ["Clbenzene (Parke and Williams, 1953a,b; Timbrel1 and Mitchell, 1977). Andrews rf al. (1977) detected high concentrations of benzene metabolites in the bone marrow of mice given 880 mg/ I87 0041-008x/81/080187-09502.00/0 Copyright 0 19x1 by Academic Presb. IOU. All riglilr of reproduction 111 any lurni I-escrvcd. ! 188 GREENLEE, SUN, AND BUS kg [3H]benzene. The appearance of these metabolites in the bone marrow correlated temporally with their disappearance from the liver. Rickert et al. (1979) reported that hydroquinone and catechol were retained in the bone marrow of rats exposed to 500 ppm benzene for 6 hr and studies on the disposition of 14C-labeledbenzene metabolites using whole body autoradiography indicate that radioactivity associated with hydroquinone or catechol, but not phenol, concentrate in the bone marrow and lymphoid organs (Greenlee et ai., 1980). Polyphenol metabolites of benzene such as hydroquinone will spontaneously autoxidize to quinones (Mason, 1979). The formation of semiquinone free radical intermediates, quinones, and superoxide anion (0;d)uring this oxidative process may be critical events leading to cytotoxicity. In this report findings on the disposition and covalent binding of catechol and hydroquinone are presented and discussed with respect to potential enzymatic and nonenzymatic pathways leading to the formation of reactive intermediates. METHODS Animals. Male Fischer-344 rats, weighing 250 g, were obtained from Charles River Breeding Laboratories (Wilmington, Mass.). The rats were housed in stainless-steel suspended wire cages, exposed to a daynight light cycle of 12 hr, and given food (Wayne Lab Blox, Allied Mills, Chicago, Ill.) and wateradlibitrtm. Disposition of "C-labeled hydroquinone and catechol. Five days prior to the administration of radiolabeled substrates, rats were given a single 500 mdkg ip injection of a 50% solution of Aroclor 1254 (a kind gift from Monsanto Co.) in corn oil or corn oil vehicle alone. [14C]Catechol(100 $3, 14 mg/kg) or ["Clhydroquinone (100 pCi, 14 mg/kg) were given by a single injection into the lateral tail vein tocontrol and Aroclorpretreated animals. Both radiolabeled compounds (specific activity, 37.9 mCi/mmol) were purchased from Midwest Research Institute (Kansas City, Mo.).The purity of each radiolabeled metabolite was >98% as assessed by thin-layer chromatography _using two solvent systems, benzene:methanol:aceticfrcid (30: 15:I) and ethyl acetate:benzene ( I : I), and by high-pressure liquid chromatography (Radial Pak A column contain- ing octadecylasilane) using a 12-min linear gradient (flow rate, 2.4 mumin) from water to methanol. All solutions were prepared using cold (4C) phosphatebuffered saline immediately before dosing. Rats from the control and Aroclor-pretreated groups were sacrificed 2 and 24 hr after dosing with each radiolabeled metabolite. Liver, spleen. and thymus were homogenized in 4 vol 10 mM phosphate buffer (pH 7.0). Bone marrow was aspirated from both femurs using 2 ml of the same buffer and homogenized in a Kontes glass-glass (3 mi) homogenizer. Total radioactivity was determined by solubilizing 100 p l of each homogenate in 1 mi of Protosol overnight at ambient temperature. Each sample was then neutralized with 60 pl of glacial acetic acid and IO ml of ACS scintillation cocktail (Amersham Corp., Arlington Heights, Ill.) were added. Radioactivity was quantified in a Searle Mark I11 scintillation. counter. Quenching was corrected by automatic external standardization. "C efficiency in the various samples ranged from 90to 95%. Total radioactivity covalently bound was measured by exhaustive extraction as described by Irons et a/. (1980) with the following modifications: to 0.5 ml of each sample an equal volume of 20% TCA was added. The resulting precipitate was then washed sequentially, four times each, with 1 ml of 40, 60, 80, and 100% methanol. After each wash, the precipitate was sedimented by centrifugation and resuspended in fresh extraction solvent. After the final wash with 100% methanol, the precipitate was dissolved in 250 pl 10 N NaOH at ambient temperature for 12 hr. The solubilized mixture was acidified with 200 p i 16 N HNO, and radioactivity quantified in 10 ml of ACS. Total protein concentration was measured by the method of Peterson (1977). Preparation of bone mnrroic' Iwrnogenate. Bone marrow was aspirated from each femur of untreated rats with I ml of SO m M phosphate buffer (pH 7.4). Aspirates from 5 to 50 animals were pooled, passed two times through a Idgauge needle. two times through a 21-gauge needle, and the cells disrupted using a Teflon-glass homogenizer. The homogenate was centrifuged at 3000g for, IO min. The resulting supernatant was diluted 1:5 with phosptiate buffer and used as an enzyme source in studies on NADPH oxidation. All procedures were carried out at 4C. Measurement of NADPH oxidation. For studies on NADPH oxidation and autoxidation of polyphenols, an Aminco DW-2A spectrophotometer equipped with anaerobic cells with plunger assemblies (American Instrument Co.. Silver Spring, Md.) was used. Incuba- tion mixtures contained 63 mg of protein and an initial concentration of 1.6 mM NADPH in a total volume of 2.5 ml of 50 mM phosphate buffer (pH 7.4). After a I-min preincubation period, the reaction was started by the addition of 5 pmol (final concentration equivalent to 2 mM) of substrate and incubated for 5 min at - -( Live Thyr Bone -WBC h 37C 400 I Cllvel For a cuvet ured menta value Air/ phenc and b strate the ra sec. S 0.45 n buffer anaerc dation quanti (Misra Spectrc triol, 1 chased Wisc.) and N Compa Stflri analysi analyze of signi Dispo. Calc Malt (14 rng PROPOSED MECHANISM O F BENZENE TOXICITY 189 2nt TABLE I All ;le- DISPOSITION OF ACID-SOLUBLE RADIOACTIVITY IN CONTROL A N D AROCLOR-PRETREATREADTS GIVEN ["C)CATECHOL OR ["C]-HYDROQUINONE Ips io- ["C]Catechol" ["C] Hydroquinone" :re 2 hr 24 hr 2 hr 24 hr >H Jrs Organ Control Arodor Control Aroclor Control Aroclor Control Aroclor la ~ ~~~~~ io- Liver 279 f 18 I88 z I 1 45 t l b 49 f 10' 203 f 31 I l l 2 12 27 f Zb 17 t I b ich Thymus 149 t 35 336 I34 43 f 2b 44 t 4' 163 z 27 207 t 37 38 t Zb 45 t Ib 3nt ith Bone marrow WBC' 335 f 72 - 370 f 9 - 252 t 19 I45 2 3 1% f 43 I26 t 13 235 f 17 215 t 21 -- 156 t Jb 129 t 7 122 t 23 99 t 10 in- * Values are expressed as dpmdnig protein and represent the mean t SE of tissues from three animals. I S , Value for given trratnient group at 24 significantly different ( p < 0.05) from corresponding group at 2 hr. , a White blood cells. ,as 'C 37C. NADPH oxidation was monitored at AOD 340- none by injection into the lateral tail vein. %. 400 nm using the split-beam mode. The reference The distribution of soluble radioactivity (not .ed 111. of cuvette contained all components except the substrate. For anaerobic measurements N, was bubbled into both cuvettes for 3 min and NADPH oxidation was meas- precipitable by TCA, see Methods) and covalently bound radioactivity (measured :d. ured under a N, atmosphere. Absorbance measure- after exhaustive extraction) was monitored l Y t ments were converted to concentration units using a in various tissues from control rats and rats )% value for B of 6200 M-'cm .'. pretreated with Aroclor 1254, a regimen di.sh )YO Atcioxidurion of polyphrnols. The autoxidation of phenol, catechol, 1,7,4benzenetriol, hydroquinone and benzoquinone was monitored by measuring sub- which protects against benzene toxicity (Greenlee et a / . , 1980). Two hours after I N strate and product absorption wavelength maxima in dosing with either compound, the amount of lu- the range 200-350 nm at a scan speed of 200 nm/ soluble radioactivity measured in liver, 0 3 sec. Substrate (final concentration equivalent to 0.10 to thymus, and bone marrow ranged from 100 tal of 0.45 mM) was added to 2.5 ml of 50 mM phosphate to 300 dpm/mg protein [Table 1). N o dif- buffer (pH 7.4) using a plunger assembly fitted to an anaerobic cell. The generation of 0, from the autoxi- ference was observed in control versus ne dation of benzene metabolites was monitored by Aroclor-pretreated animals. After 24 hr, the ed quantitating adrenochrome formation from epinephrine amount of soluble radioactivity measured 4). (Misrd and Fridovich, 1972) at 480 nm in a Cary 219 in liver and thymus was markedly lower ed spectrophotometer (Palo Alto, Calif.). 1,2,4-Benzene- than at 2 hr, whereas in bone marrow no Sh a triol, benzoquinone, and hydroquinone were purchased from Aldrich Chemical Company (Milwaukee, statistically significant difference was found as Wisc.). Catechol, epinephrine, superoxide dismutase, at the two time points, except for control :r- and NADPH were obtained from Sigma Chemical rats given ["C]hydroquinone (Table 1). ed Company (St. Louis. Mo.). Even for the latter group, the amount of sol- n. Srnrisricirl citictlysis. The data were analyzed via uble radioactivity detected at 24 hr was 66% on Is. analysis of variance, with difference among means analyzed with Duncan's multiple range test. The level that detected at 2 hr. These findings indi- of significance chosen was p < 0.05. cate that soluble radioactivity associated rh with ['-'C]catechol and [*T]hydroquinoneis Jn ,a- RESULTS retained in the bone marrow and are in agreement with a previous study by Rickert ial et (11. (1979) showing a selective retention of of these metabolites in the bone marrow of a :d rats exposed to benzene. a- Male Fischer-344 rats were given 100 pCi After the administration of [L'C]catechol at (14 mg/kg) [l"C]catechol or ['Tlhydroqui- or [14C]hydroquinoneto control rats, the 190 GREENLEE. S U N , AND BUS TABLE 2 DISPOSITION OF COVALENTLY BOUND RADIOACTIVITY IN CONTROL A N D AROCLOR-PRETREATED RATS G IVEN ["C]CATECHOL OR [ " C ~ H Y D R O Q U I N O N E I "CICatec hol" ["CjHydroquinone" 2 hr 24 hr 2 hr 24 hr Organ Control Aroclor Control Aroclor Control Aroclor Control Aroclor Liver Thymus Bone marrow W BC' 210 f 21 406 c 42 424-f 68 121 f 49 390 f 61 198 k 62 - 268 f. 22 347 f 41 715 f 27" I059 f 74 82 f 13b 209 f 4 291 2 73' 497 f 21" 16Ok 14 556 f 142 363-f. 48 30 f. l W 262 c 18 296-f 4 239 c 19" 330 f 17 558 f. 10'' 726 f 122 46 f 5* 289 f. 14 349 f. 49 547 f. I5 '' Values are expressed a s d p d m g protein and represent the mean 2 S E of tissues (same RS those in Table I ) from three animals. Aroclor value at 24 hr significantly different ( p C 0.05)from control value at 24 hr. Aroclor value at 2 hr significantly different ( p C 0.05)from control value at 2 hr. Control value at 24 hr significantly different ( p C 0.05)from control value at 2 hr. ' White blood cells. amount of covalently bound radioactivity measured in the bone marrow after 24 hr was significantly greater than the amount detected after 2 hr (Table 2). Pretreatment with Aroclor resulted in a significant decrease in the amount of covalently bound radioactivity measured at 24 hr in both the bone marrow and peripheral white blood cells. In the liver covalently bound radioactivity associated with [* 4 C ] ~ a t e ~ hwoals significantly depressed after 24 hr and with [14C]hydroquinoneafter both 2 and 24 hr in Aroclor-pretreated rats. No significant differences in the temporal pattern of covalent binding were observed in the thymus. Autoxidation of Polyphenols The potential for autoxidation of catechol and hydroquinone as well as other benzene metabolites, phenol, and 1,2,4benzenetriol was determined at various pH values ranging from 6.5 to 9.5. At physiologic pH (7.4) hydroquinone and 1,2,4benzenetriol spontaneously autoxidized to products with ultraviolet absorption maxima of 248 and 267 nm, respectively (Table 3). No decomposition was observed anaerobicalb. The ultraviolet absorption maximum of the hydroquinone autoxidation product was identical to that of benzoquinone (248 nm) and both species decomposed under alkaline conditions. In the presence of 250 units of superoxide dismutase, the rate of autoxidation of hydroquinone was stimulated fivefold, whereas the rate of autoxidation of I ,2,4-benzenetriol was markedly inhibited to a value less than 4% of control (Table 3). At pH values greater than or equal to 8.5, the autoxidation of both hydroquinone and 1,2,4benzenetriol was quite rapid and was essentially complete within 1 min after the addition of either compound to the buffer (data not shown). Under the same experimental conditions, phenol and catechol did not spontaneously autoxidize at any of the pH values examined. The ability of the various metabolites to stimulate the formation of adrenochrome from epinephrine was determined as an indirect measure of the generation of 0; (Misra and Fridovich, 1972). As shown in Table 4, adrenochrome formation was stimulated by 1,2,4-benzenetriol (0.298 pmoV min) and hydroquinone (0.0042 pmol/min), but was barely detectable in the presence of catechol (0.0004 pmolhin). Addition of superoxide dismutase inhibited 1,2,4-benzenetriol and catechol stimulated adrenochrome formation, while it resulted in a two- fo ne AI. no N/ stu ox, roi the nm aer I: Pro oxic tion mat C -( , 1,2,4(29 Hydrc (292 - n bt 'C "I 41s - \roclor b + Sb 4 + 14 9 f 49 7 1 IS I ) from both ondits of IxidafiveJn of bited !le 3). 3 8.5, ? and I was r the suffer peri- did f the es to rome n in- . 0; In in jtimmol/ nin), ence in of benenotwo- PROPOSED MECHANISM OF BENZENE TOXICITY 191 fold increase in hydroquinone-induced epinephrine autoxidation. Augmentation of N A D P H Oxidntion The effect of benzoquinone, hydroquinone, catechol, and 1,2,4-benzenetriol on NADPH oxidation by rat bone marrow was studied. Only benzoquinone augmented the oxidation of NADPH by a 3000g bone marrow supernatant (Table 4). Anaerobically, the total amount of NADPH oxidized (141 nmol) was twice the amount oxidized aerobically (68.5 nmol) (Table 4). DISCUSSION It has been suggested that the chronic production of small amounts of benzene oxide in bone marrow may result in reactions with critical macromolecules, ultimately leading to bone marrow depression or leukemia (Snyder et a l . , 1978). Studies on the covalent binding of benzene to liver microsomal protein, however, have indi- cated that 8 metabolite of phenol and not benzene oxide was responsible for the observed binding (Tunek et af., 1978). Several recent lines of evidence have indicated that benzene-induced bone marrow toxicity is associated with two polyphenol metabolites, hydroquinone and catechol. In a study of the metabolism of ["Clbenzene in isolated rat femurs, it was reported that 3% of the radioactivity was recovered as benzene metabolites. The bone marrow/blood concentration ratios for both hydroquinone and catechol were approximately 400 (Irons et a l . , 1980). I n vivo, hydroquinone and catechol were found to be retained at steady-state concentrations for at least 18hr in the bone marrow of rats following a 6-hr exposure to 500 ppm benzene (Rickert et nl., 1979). In studies on the disposition of '"C- TABLE 3 RATES OF FORMATION A N D SPECTRAL CHARACTERISTICS OF AUTOXIDATION PRODUCTS OF 1,2,4-BENZENETRIOLA N D HYDROQUINONE" Compound ( A m a x I nm) Concentration (mM) Incubation conditions Amax of product(s) Rate o f product formation" 1,2,4-Benzenetriol 0.40 Air 267 (290) 0.40 Nitrogen - 0.10 Air 267 0.10 Air + 62.5 U SOD' 267 0.10 Air + 125 U SOD 267 0.10 Air + 250 U SOD 267 AOD/minb 0.87 2 0.04 0 0.55 t 0.01 0.21 2 0.01 0.08 2 0.01 0.02 2 0.01 Hydroquinone (292) nrnoUrnind 0.10 Air 248 2.2 2 0.2 0.40 Air 248 4.8 2 0.3 0.40 Air + 62.5 U SOD 248 13.6 t 0.9 0.40 Air + 125 U SOD 248 17.2 2 0 0.40 Air + 250 U SOD 248 23.9 z 1.9 0.40 Nitrogen 0 Autoxidation rates are the mean t SE of triplicate determination. emax of autoxidation product not known. Superoxide dismutase. Based on a E,,,,, value for benzoquinone of 16.3 mM-l cm-' (pH 7.4). 192 GREENLEE, SUN, AND BUS TABLE 4 S TIMULATION OF NADPH OXIDATION A N D ADRENOCHROME FORMATION B Y B ENZOQUINONE, 1,2,4-BENZENETRIOL. HYDROQUINONE, AND CATECHOL Incubation conditions Air atmosphere Nitrogen atmosphere Compound" Benzoquinone I ,2,4-Benzenetriol + SOD" Hydroquinone + SOD Catechol + SOD Benzoquinone 1,2,4-Benzenetriol Hydroquinone Catechol Total NADPH oxidizedb (nmol) 68.5 f 28.0 (3) 0 (3) - 0-(3) 0 (3) - 141 '.48 0 0 0 (8) Adrenochrome formationC (pmoUmin) - 0.298 f 0.007 (4) 0.120 t 0.002 (4) 0.0042 rt O.OOO1 (4) 0.0080 t 0.0002 (4) 0.0004 f O.oo00 (4) nd' Compounds were present at a concentration of 0.1 m M in a total incubation volume of 3.0 ml. * NADPH oxidation (1.6 mM starting concentration) was monitored for 5 min at AOD 340-400 nm in the presence of 3000g rat bone marrow supernatant (63 mg total protein). Values shown are the mean 2 SE. The number of determinations made are given in parentheses. E Epinephrine present at 0.5 mM. The values shown were calculated using an emax = 4.01 x IF M-' cm-' and were corrected for background epinephrine oxidation. All values represent the mean f SE of four determinations. Superoxide dismutase; 25 pg protein, 62.5 U per 3-ml incubation. No adrenochrome formation detected. labeled benzene metabolites using whole body autoradiography, radioactivity associated with hydroquinone or catechol, but not phenol, was found to concentrate in bone marrow and lymphoid organs. Further, the level of radioactivity in these tissues was reduced in rats pretreated with Aroclor 1254, a regimen which protects against benzene toxicity (Greenlee et ai., 1980). The results in this study further support the hypothesis that polyphenol metabolites of benzene may be important in benzeneinduced cytotoxicity. Following administration of a single iv dose of [*4C]catecholor [14C]hydroquinone,soluble radioactivity was retained in bone marrow, but not liver or thymus, for at least 24 hr. Covalently bound radioactivity increased with time in all tissues examined, with the largest increase observed in bone marrow. Covalent binding was depressed in liver, white blood cells, and bone marrow by Aroclor pretreatment, which protects against benzene myelotoxicity. The observations of this study suggested that covalent binding of hydroquinone and or catechol in the bone marrow may be an important event in the expression of toxicity. Thus, potential pathways for the activation of these compounds and subsequent metabolites such as 1,2,4-benzenetriol were examined. At physiologic pH, hydroquinone and 1,2,4-benzenetrioI (a putative metabolite of catechol) spontaneously autoxidized as detected by the appearance of ultraviolet absorption products. A proposed metabolic scheme is presented in Fig. 1. For the triol, the sequence, 9 .-+ 10 + 11, is modeled after studies on the autoxidation of 6-hydroxydopamine (Graham, 1978).The pathway shown for the oxidation of hydro- quint the a and benz from and/( romc serv: ment ['TI, signi: is PO' ['TI( tion auto: Au nism benzi and q et al. catec dopa, presu ucts . chror al., 1 row I 05,: matic gener activi provi for ac none It c gener taboli of bel ulatio benze by su produ vich, of hy lation creasc may t scribe PROPOSED MECHANISM OF BENZENE TOXICITY 193 quinone (4 + 5 -+ 6) is based on studies of the autoxidation of durohydroquinone (James 1 and Weissberger, 1938). It is proposed that ' benzene-induced cytotoxicity may result from binding of highly reactive semiquinone and/or quinone metabolites to cellular macromolecules, which is supported by the observation of binding after in vivo treat- I ment with both ['"Clhydroquinone and ' ['4C]catechol. Although catechol did not significantly autoxidize at physiologic pH, it is possible that the in vivo binding seen with [L4C]catechoml ay originate from the formation of 1,2,4-benzenetriol, which readily 0 00 qH1 1 OH - - - - - -7 - - - - --c 8 old! ~ Autoxidation may not be the only mecha- nism whereby polyhydroxy metabolites of benzene can be activated to semiquinone OH and quinone metabolites. Studies by Dybing COVALENT BINOWG TO et (11. (1976) have demonstrated that several CELLULAR MACROMOLECULES '~IM, t l -n- catechols such as a-methyldopa, dopa, and dopamine are metabolically activated to OH presumed semiquinone andor quinone prod- ucts via 0; produced by microsomal cytochrome P-450. A recent study (Hanson et '.? has shown that rabbit bone marrow microsomes are capable of generating 05, as determined by adrenochrome formation. Thus, the presence of endogenously generated 0; might also be expected to 6& FIG. 1. Proposed oxidative pathways for various benzene metabolites. I , benzene; 2, benzene oxide; 3, phenol; 4, hydroquinone; 5 , semiquinone; 6 , benzoquinone; 7, benzene dihydrodiol; 8, catechol; 9, 1,2.4-benzenetriol; 10, semiquinone of benzenetriol; ' ' 9 2-hydroxybenzoquinone. activate benzene metabolites and would ested and le an provide, for example, a direct mechanism would displace the equilibrium to the right: for activation of catechol to radical or qui- c +none products. hydroquinone 2 0 , e icity. ation It cannot be excluded, however, that 0; generated by autoxidation of benzene me- benzoquinone + 201 + 2H+. 3- superoxide dismutase me- , tabolites may also be important component H202 + 0, were of benzene-induced cytotoxicity. The stim- )qui- dation of epinephrine oxidation by 1,2,4- The enhanced cooxidation of epinephrine stive benzenetriol and catechol, and its inhibition induced by superoxide dismutase can not 1 au- by superoxide dismutase, suggests 0; as a clearly be attributed to an increased flux of :e of product of metabolite autoxidation (Frido- O,, however, in that semiquinone and/or osed 3 vich, 1975). In contrast, the autoxidation quinone autoxidation products may also be For of hydroquinone and subsequent stimu- capable of oxidizing epinephrine directly. I , is lation of epinephrine oxidation were in- Superoxide anion is known to be cytodion creased by superoxide dismutase, which toxic, presumably mediated via formation of The may be accounted for in the reaction de- hydrogen peroxide and hydroxyl radicals dro- scribed below in which removal of 0; (Fridovich, 1978). In rabbit bone marrow i$ l. 0 t t d I I94 GREENLEE, SUN, AND BUS the activity of superoxide dismutase, an enzyme which functions to detoxify 0; (Fridovich, 1978), is less than 10% of the activity in liver tissue (Hansen et af., 1978). Thus, bone marrow may be particularly sensitive to 0;-induced tissue injury. It should be noted that the mechanism of toxicity of several other chemicals may be similar to that proposed here for benzene. The neurotoxic agent 6-hydroxydopamine is thought to destroy dopaminergic neurons via 0;and/or semiquinone formation (Sachs and Jonsson, 1975; Graham el af., 1978). Likewise, the diabetogenic agent alloxan appears to be reduced to di- aluric acid in the p cell of the pancreas, with subsequent autoxidation producing O;, HzOZ,and OH (Cohen and Heikkila, 1974). Finally, the herbicide paraquat may produce pulmonary damage by 0; generated from a cyclic single-electron reduction-oxidation of the parent molecule (Bus et a l . , 1976).It is particularly interesting that the specific target organ toxicity of these agents correlates with a selective uptake and/or retention of the agent in the target tissue (Cohen et al., 1976; Hammarstrom and Ullberg, 1966; Rose et af., 1974), which is analogous to the retention of the proposed cytotoxic metabolites of benzene in bone marrow. A second source of semiquinone and 0; can result from a one-electron transfer from NADPH to benzoquinone (6 + 5 , Fig. 1). to be catalyzed by a 3000g rat bone marrow supernatant (Table 4). Under aerobic conditions, the one-electron reduction product would be reoxidized to benzoquinone with formation of 0;. Anaerobically, the total amount of NADPH oxidized was twice the amount oxidized aerobically (Table 4), suggesting the two-electron reduction sequence 6 45 + 4 (Fig. 1). In summary, the data in this report support a mechanism for benzene toxieity in which the formation of potentially cytotoxic products result from the autoxidation of at least two of the reported in vivo polyphenol metabolites of benzene, hydroquinone, and 1,2,4-benzenetriol. ACKNOWLEDGMENT We are grateful to John P. Chism for his excellent technical assistance with certain aspects of this work. REFERENCES ANDREWS, L. S..LEA,E.W., WITMER, C. M.,Kocsis, J. J., A N D SNYDER, R . (1977). Effects of toluene on the metabolism, disposition and hemopoietic toxicity of [aH]benzene. Biochem. Phnrmacol. 26, 293-300. s.,BUS, J. AUST. S. D.,A N D GIBSON, J. E. (1976). Paraquat toxicity: Proposed mechanism of action involving lipid peroxidation. Environ. Health. Perspecf. 16, 139-146. CARPENTER. C. D.. SHAFFER, C . B., WEIL, C. S., A N D SMITH, H. F., JR., (1944). Studies on the inhalation of 1:3 butadiene; with a comparison of its narcotic effect with benzol, toluol, and styrene, and a not as the elimination of styrene by the human. J. Ind. Hyg. Toxicol. 28, 69-78. COHENG. ., AND H E I K K I L AR., E. (1974). The generation of hydrogen peroxide, superoxide radical and hydroxyl radical by bhydroxydopamine, dialuric acid and related cytotoxic agents. J . Bioi. Chem. 249, 2447-2452. COHENG. . , HEIKKILA, R. E., ALLISB, ., CABBATF,., DEMBIECD. ., MACNAMEED, ., MYTILINEOUC.. A N D WINSTON, B. (1976). Destruction of sympathetic nerve terminals by 6-hydroxydopamine: Pro- tection by I-phenyl-3-(2-thiazolyl)-2-thiourea. diethyldithiocarbamate, methimazole, cysteamine, ethanol and n-butanol. J. Pharmncol. Exp. Ther. 199, 336-352. s.,COHENH. . FREEDMAN, M . L., A N D GOLDSTEIN, B. D. (1978). The problem of benzene in our environment: clinical and molecular considerations. Amer. 1. Med. Sei. 275, 124-136. DYBING.E., NELSON, S. D . , MiTciiELL. J. R., SASAMEH, . A., A N D GILLETTEJ,. R. (1976). Oxidation of a-methyldopa and other catechols by cytochrome F45cgenerated superoxide anion: Possible mechanism of methyldopa hepatitis. Mol. Phormacol. 12, 91 1-920. FRIDOVICH, I. (1975). Superoxide dismutases. Annu. Rev. Biochem. 44, 147-157. FRIDOVICH, I. (1978). The biology of oxygen radicals. Science 201, 875-880. GOLDWATER, L. J. (1941). Disturbances in the blood followin 26, 957GRAHAM, cholamii toxic qu GRAHAM, JR., ANI versus c nism of and rela toma cel GREENLEE (1980). phenol, whole-bt act., in 1 HAMMARS uptake c Nurim f HANSEN2. M. R. ( marrow : 85, 783HOUGH, C toxicity benzene 2%-306 IRONS, R. RICKERT covalent Biol. lnir IRONS, R. short-ter subpopu P-lymph( Paihol. I JAMES, T .I processe quinone. LASKINS.. toxicity: viron. H I MASON, R PROPOSED MECHANISM OF BENZENE TOXICITY GRAHAM, D. C . (1978). Oxidative pathways for catecholamines in the genesis of neuromelanin and cytotoxic quinones. Mol. Phartnucol. 14, 633-643. GRAHAM, D. G . , T IFFANY, S . M., BELL, W. R., JR., A N D GUTKNECHT, W. F. (1978). Autoxidation versus covalent binding of quinones as the mechanism of toxicity of dopamine, 6-hydroxydopamine, and related compounds toward C1300 neuroblastoma cells in vifro. Mol. Phurttiuco/. 14, 644-653. w.GREENLEE, F., GROSSE, . A., A N D IRONS, R. D. (1980). A Study of the disposition of 14C-labeled phenol, catechol and hydroquinone in rat using whole-body autoradiography. Chem. -Bio/. Inter- HAMMARSTROLM., ,A N D ULLBERCS,. (1966). Specific uptake of labeled alloxan in the pancreatic islets. Nuture (Lotidoti) 212, 708-709. son, Bend, and Philpot. eds.), p. 151. Elsevier/ North-Halland, AmsterdadNew York. MISw, H. P., A N D FRIDOVICIH. ,(1972). The univalent reduction of oxygen by reduced flavins and quinones. J. Biol. Chem. 247, 188-192. PARKED, . V.. A N D WILLIAMS, R. T. (1953a). De- toxication. XLIX. Metabolism of benzene containing "C-benzene. Biochem. J. 54, 231-238. PARKED, . v.. A N D WILLIAMS, R. T. (1953b). De- toxication. LIV. The metabolism of benzene. (a) The formation of phenyl glucuronide and phenyl sulfuric acid from [14C]benzene.(b) The metabolism of ["Clphenol. Biochem. J . 55, 337-343. PETERSON, G . L. (1977). A simplification of the protein assay method of Lowry et al. which is more generally applicable. Anal. Btocliern. 83, 346-356. RICKERT, D.E., BAKER, T. S . , BUS, J . S., BARROW, ROSE, M. S., S MITH, L. L., A N D WYATT, I. (1974). toxicity of commercial benzene and a mixture of benzene toluene, and xylene. J . f n d . H y g . 26, IRONS, R. D., DENT, J. G . , BAKER, T. S . , A N D RICKERTD,. E. (1980). Benzene is metabolized and covalently bound in bone marrow in situ. Chern.Biol.Interact. 30, 241-245. IRONS, R. D., A N D MOORE, B. J. (1980). Effect of short-term benzene administration on lymphocyte subpopulations in the rabbit: Evidence of a selective fi-lymphocyte sensitivity. Res. Commun. Chem. Parho/. Pharmucol. 27, 147-155. JAMES, T. H., A N D WwwvatGER, A. (1938). Oxidation processes. XI. The autoxidation of durohydroquinone. J . Amer. Chetn. Soc. 60, 98-104. LASKIN,S . , AND GOLDSTEINB, . D. (1977). Benzene toxicity: a critical evaluation. Arch. Toxicol. Environ. Health. Suppl. 2. MASON, R. P. (1979). Free radical metabolites of paraquat into rat lung. Nutitre [London) 252, 314- 315. SACHS,c., AND JONSSON, G. (1975). Mechanisms of action of6-hydroxydopamine. Biochem. Pharmucol. 24, 1-8. SNYDER, R., AN D Kocsls, J . (1975). Current concepts of chronic benzene toxicity. CRC Crit. Rev. Tomcol. 3, 265-288. SNYDER, R., LEE, E. W., A N D Kocsis, J. J. (1978). Bone marrow depressant and leukemogenic actions of benzene. LiJe Sci. 21, 1709-1722. TIMBELLJ., A., A N D MITCHELL, J. R. (1977). Toxicity-related changes in benzene metabolism iti vivo. Xenobioticu 7 , 415-423. TUNEKA. ., PLAIT, K. L., BENTLEY,P., A N D OESCH, F. (1978). Microsomal metabolism of benzene to species irreversibly binding to microsomal protein and effects of modifications of this metabo- Iism. Mol. Phurmucol. 14, 920-929.