Document ga1L7KRxQ5joQg9knMEeR765Q

.` ERT, !v. L.. A N D r3rative metgbo- t binding to heirunder, and rat. .~c~.J.J.(1981). 3gents on hepatic in ninbow trout. D. R. (1984). jby purified isonaphthotlavone,acol. 33. 3743- . P.. AND HAN>roteinsin poly197-203. -TOXICOLOGY AND APPLIED PHARMACOLOGY 91.85-95 ( 1987) An Interaction of Benzene Metabolites Reproduces the Myelotoxicity Observed with Benzene Exposure c DAVID A. EASTMOND,*MARTYN T.SMITH,* A N D RICHARD D.IRONS? Dcparmient ofBiomcdical and Environmental lli~alfti Sciences,School of Public Ilcaltfi.Uniwrsityof California, Bcrkelc.v,California94720.mid tCfiemica1Inditstgv Institute o/To..ricology. Research TrianglePark, North Carolina27709 Rcccivcd March 11. 1987:acccpted July 7,1987 An Interaction ofBenzene Metabolites Reproduces the MyelotoxicityObservedwith Benzene Exposure. EASTMOND, D. A., SMITHM. . T., AND IRONS. R. D. (1987). Toxicol.Appl. Phartnncol. 91. 85-95. Benzene-induced myelotoxicity can be reproduced by the coadministration of two principal metabolites. phenol and hydroquinone. Coadministration of phenol (75mg/ kg) and hydroquinone (25-75 mg/kg) twice daily to B6C3FI mice for I2 days resulted in a significantloss in bone marrow cellularity in a manner exhibitingadosc-response. One explana- tion for this potentiation is that phenol stimulates the peroxidasedependent metabolism of hydroquinone. Addition of phenol to incubations containing horseradish peroxidase. H202, and hydroquinone resulted in a stimulation of both hydroquinone removal and benzoquinone formation. Stimulationoccurred with phenol as low as 100p~ and with very low concentrations of horseradish peroxidase. When boiled rat liver protein was added to identical incubations containing ["C]hydroquinone. the level of radioactivity rrcovmd as protein bound increased by 37% when phenol was added. Similar results were observed when [`4C)hydroquinonewas incubated in the presence of activated human leukocytes. Hydroquinone binding was increased by approximately 70% in the presence of phenol. Phenol-induced stimulation of hydroquinone metabolism and benzoquinone formation represents a likely explanation for the bone marrow suppression associated with benzene toxicity. B 1987 ACX~IICPI-CS. IW. Chronic exposure to benzene is associated low-level chronic exposure to benzene poses with aplastic anemia, pancytopenia, leuke- a significant health threat (Goldstein. 1953). mia, and other blood dyscrasias in m a n . Although themechanism of toxicity is as of (Goldstein. 1977; IARC, 1982). Nevertheless yet unknown, studies have shown that benthe significanceof exposure to low concentra- zene itself is probably not the actual tosicant tions of benzene (c10 ppm) remains contro- but is converted by the liver to a metabolite(s) versial. This is a serious health concern be- which travels to the bone marrow and exens cause benzene is an extremely important id- its toxic effects (Sawahata CI ai., 1985: Samdustrial chemical, to which 2 million U.S. mett et ai.. 1979; Andrews el ai., 1977). In workers are potentially exposed (NIOSH, vivo metabolic studies following benzene 1974), and a ubiquitous pollutant (USEPA, exposure have identified phenol as the pri1980). Due to the limitations inherent in epi- mary metabolite ofbenzene with hydroquidemiological and animal carcinogenicity none, catechol, Iruns,truns-muconic acid, studies, understanding the mechanism by and I ,2,4-trihydroxybcnzene as significant which benzene exerts its hematopoietic toxic- secondary metabolites (Rusch ct ai.. 1977; ity may be the only way to determine whether Snyder (I' ai., 1981; Gad-El-Karim et ai.. 85 I -86 EASTMOND, SMITII, AND IRONS 1985a). After benzene exposure. phenol con- certain agents. undergo ;1unique type of oxi- centrations within the bone marrow rise and dative rnctabolism known as the "oxidative fall relatively quickly, whereas hydroquinone burst'' which is characterized by a cyanide- and catechol concentrations. although mod- inscnsitive increase in oxygen consumption, erately low initially. rise with time and persist increascd hexose monophosphate shunt ac- at relatively high concentrations (Rickert ct tivity, and a release of oxygen radicals, mye- af..1979; Irons ci nl., 1982; Greenlee et ai.. loperoxidax, and other lysosomal enzymes T98la.b). A perplexing problem is that the (Karnovsky and Robinson, 1981). Recent administration ofthe principal metabolites of studies in our laboratory have shown that iso- benzene has failed to mimic the myelotoxi- lated human myeloperoxidase and PMNs, city characteristic of benzene exposure (Gad- when stimulated to undergo the oxidative El-Karim et al.. 1985b: Tunek et af., 1981; burst, are capable of converting the phenolic Mitchell. 1971; Snyder and Engelsberg, compounds, phenol, I-naphthol. and dieth- 19s'). This failure ofthe known benzene me- ylstilbestrol to reactive protein-binding spe- tabolites to reproduce benzene-induced my- cies (Eastmond efai.. 1986, 1987a.b). These elotoxicity led us to hypothesize that toxicity observations suggest that a secondary peroxi- may be the result of a combination of metab- dase-dependent metabolism of phenol, hy- olites rather than the result of any single me- droquinone. and catechol may be occurring tabolite. within the bone marrow and that this mecha- ...' The isolation of phenol. hydroquinone, nism may be responsible for benzene-in- i i and catechol in the bone marrow following duced myelotoxicity. The object of this study benzene exposure suggested the possibility was to test whether benzene-induced myelo- that the critical metabolic events resulting in toxicity could be reproduced by the coadmin- toxicity may involve an interaction of these istration of two of its metabolites and to phenolic compounds during localized metab- identify a potential mechanism whereby my- olism within the bone marrow. Human bone eloperoxidase may be involved in this inter- marrow contains appreciable levels of myelo- action. peroxidase for which phenol is a known sub- strate and hydroquinone and catechol are probable substrates (Yamazaki, 1958; Him- METHODS melhoch et al.. 1969; Eastmond et af..1986). Evidence for the involvement of peroxidases Chiwiiculs atid ct?=wrc*sP. henol and hydroquinone in the metabolic activation of various xenobiotics such as acetaminophen. diethylstilbestrol, and certain aromatic amines has accumulated in recent years (Moldeus CI 01.. 1982; Metzler. 1984: Wise et a!., 1984). Further- were purchased from Aldrich Chemical Co. (Milwaukee. WI),Hydrogen pcroxide (30%soluiion). horseradishperoxidase (HRP; type VI. 275 U/mg). dextran 70. cytochrome c (type 111). and phorbol 12-myristate 13-acetate (PMA) were obtained from Sigma Chemical Co. (SI. Louis, MO). [2.3.5.6-"C]Hydroquinone (33 mCi/ more, the target organ specificityexhibited by benzene for peroxidase-containing organs during recent animal carcinogenesis bioassays such as the zymbal gland. the harderian mmol) was Purchased from Arnersham. (Arlington Heights. IL). [U-"CIPhenol (30.5 mCi/rnrnol) was ob- tained from Pathfinder Laboratories(St. Louis. h10). AN other chemicals or solvents were generally of the highest gnde available and obtained through local supplien. gland, and mammary glands also suggests a Distilled water was purified using the MilliporeQ system role for peroxidases in the metabolic activation of benzene (Haseman et al.. 1984: Osborne el al.. 1980: Morrison and Allen. 1966; Cavalien and Rogan. 1985). Polymorphonuclear leukocytes (PMNs),when stimulated by from Millipore (Bedford. MA). Atritt1d.s. Male BbC3Fl mice. 4 to 6 weeks old on ar- rival. were obtained from Charles River Breeding L a b ntories (Wilmington. M A ) and held in quarantine for 2 weeks. Animals were housed five to six per cape with hardwood bedding and were given rood (Nlll-07.Ziedcr Brt am pat tre. sol bu I Iro 1 of Xr; etic sus ma lac per (CC I :5 to I cot OX1 r& phr ate drc sha mi: tair Wt [ 141 pcr 19: Pe: the am qu sisi am Ul-f 1 fro SCC EO: bil si0 Ph la1 (h' ICl bu Sic fib ph WT th Sa1 !etype of ?xi. ne "oxidative ~y a cyanidc:onsumption. clte shunt acadicals. myemal enzymes 481). Recent )ownthat iso- and PMNs. :he oxidative the phenolic )1. and dieth-binding spc17a,b). These ndary perosi' phenol. hybe occumng .t this mechabenzene-in:of this study iuced myelo:he coadmin)lites and to Lvhereby myin this inter- I hydroquinone '0.(Mrlwaukee. norscndX per-" Sxtran 70, cyto:state 13-acetate cmical Co. (St. me (22 mCi/ Am. (Arlington mmol) was obLouis. MO). All ly of the highest local suppliers. lipore Q system xeks old on arBreeding Laboquarantine for x per cage with 4IH-07, Ziegler MECHANISM OF BENZENE TOXICITY 87 Bros.. Gardners. PA) and water ad libittun. Serologic ex- binding of hydroquinone equivalents io leukocyte pro- amination conlirmed the absence of antibody titers to tein w s performed aspreviously described (Eastmond et pathologic murine viruses. Groups of six mice d.,1986). treated intraperitoneally twice daily at 6-hr intervalswith Sluristical onaljws. Statistical analyses. unless othersolutions of metabolites freshly preparcd in phosphate- wise stated. were performed using Dunnett's I test for buffered saline as previously described (Wierda and multiple comparisons following one- or two-way analy- Irons. 1982). sesof variance. In experiments where the varianceswen Bonc i?iarrowcc1liilarir.v. It is the previous experience not homogeneous. log transformation of the data was of this laboratory that bone marrow cellularity is more performed prior to analysis. Critical values were deter- sensitive and reliable as a single indicator of hematopoi- mined using a 0.05 probability of type I error. Analysis etic status than peripheral leukocyte counts, which arc of bone marrow cellularitydata consisted of an Ftest for susceptible to a variety of influences in addition to bone equality of variance followed by a pooledI test. marrow toxicity (Irons et al., 1979). Bone marrow cellu- larity was determined by flushing and thoroughly sus- pending bone marrow from left femurs with 4 ml lsoton (Coulter Electronics). A 400-pl aliquot was then diluted RESULTS 150 with lsoton containing0.3 ml Zapoglobin (Coulter) to lyse erythrocytesand counted on a Model ZF Coulter Toxicity studies. The administration of counter. phenol (0-150 mg/kg) twice daily for 12days d\fctahulisinand binding studies with horseradisliperosidase. The standard incubation contained HRP (0.06 &mi). H202 (1 mM). and hydroquinone (75 JIM) in phosphate buffer (0.1M,pH 7.4). lncubationswere initiated by the addition of H202(immediately following hy- did not result in a suppression of bone marrow cellularity. Hydroquinone administration (100 mg/kg, 2X/day) produced a transient mild suppression in cellularity that was droquinone addition) and were performed at 37'C in a evident after 3 days but was essentially the shaking water bath. The reaction was terminated at 2 min by the addition of 5% trichloroacetic acid and maintained on ice until HPLC or protein-binding analyses were performed. Determination of the binding of ["C]hydroquinone or phenol equivalents to protein was performed as previously described (Eastmond cr ai.. same as that of controls after 12 or even 36 days of continued administration (Fig. 1A). Neither compound produced any obvious dose-response (data not shown). However, the concomitant administration of phenol 1986).HPLC analysesby electrochemical detection were performed as described by Eastmond et ai. (1986) with the following modifications. Working potentials of +0.7 and -0.4 V were used for hydroquinone and 1.4-benzoquinone detection. respectively. The mobile phase. consisted of 99% 0.1 M ammonium acetate buffer (pH 4.0) (75 mg/kg)and hydroquinone (25-75mg/kg) twice daily resulted in a significant decrease in bone marrow cellularity. The decrease in cellularity was evident after 12 days of administration and exhibited a clear dose-re- and 1% acetonitrile at a flow rate of 1 ml/min. The col- sposne relationship (Fig. 1B). The adminis- umn was maintained at 50'C. tration of catechol, either alone or in combi- Ldiocrfe emerimenis. Leukocvtes were isolated from the blood of healthy human volunteers by dextran' sedimentation as described by Markert ef ul. ( 1985). Cell - nation with phenol or hydroquinone. did not contribute to the suppression of bone mar- counts were performed using a hemocytometer and via- row cellularity (data not shown). bility was determined to be >95% by trypan blue exclu- Metabolism stiidics. HRP.a model peroxi- sion. Prior to further experimentation. the ability of PMNs to produce superoxide anion radial when stimulated by PMA was determined as previously described (Markert el a/.. 1985). Incubations to determine the protein binding of [14C]hydroquinoneduring the oxidative burst by PMNs were performed as follows: Cell suspen- dase enzyme, was shown to convert hydroquinone to 1,4-benzoquinone in a reaction that was dependent upon HzOz. The addition of phenol to these incubations resulted in a 30% stimulation of hydroquinone removal sionscontaining PMNs (5X 106),[14C]hydroquinone(30 JIM), and PMA ( I pg/ml) in modified Dulbecco's phosphate buffer (pH 7. I ) were incubated at 37'C in a shaking water bath. The reactions were terminated at 20 min by the addition of 5% trichloroacetic acid after which the samples were maintained on ice and centrifuged. The from the incubation (Table I). Hydroquinone removal was accompanied by an increase in 1,4-benzoquinone recovered in the medium in a stoichiometric fashion (Table I). This stimulation was dose rclaicd and ob- 88 EASThlOND. ShllTI 1. AND IRONS - Hydroquinone c 75 mwkg Phmd 40 10 20 10 Day8 01 Adminiatration i 40 1 0 1 25 50 75 008. (mgtkg) - FIG. 1. Influence of phenol on suppression of bone marrow cellularity by hydroquinone. B6C3F1 male mice. 8-10 weeks of age. wece administered phenol. hydroquinone. or both intraperitoneally at the doses indicated. twice daily. 6 hr apart. (A) Hydroquinone alone (0)over 36 days of administration. (B) Hydroquinone(25-75 mg/kg)in the presenceofphenol(75 mg/kg)(@):phenol alone(0)after 12daysofadministraiion. Each point represents the mean femoral cellularity of five or six animals expressed asa percentage of control (untreated) mice which averaged 1.63 X IO' ( 1.02 X IO6 SE)nucleated cells per femur. Asterisk indicatessignificantdifference from controls (p < 0.001). senable with phenol concentrations as low as protein was added to similar incubationscon- 100phi (Fig.2).At low peroxidase concentra- taining [I4C]hydroquinone, the level of radio- tions. the phenol-induced stimulation of hy- activity recovered bound to protein (an indi- droquinone metabolism increased to nearly cation of the formation of a highly reactive 200%. Furthermore, when boiled rat liver product from hydroquinone) was increased by 37%when phenol was added to these incu- bations (Table 2). However. the addition of TABLE I PHESOL-INDUCED STIMULATION OF HORSERADISH PEROXIDASE-CATALY ZED METABOLISM OF HYDROQUI- NONE nonradiolabeled hydroquinone to similar incubations containing ["C]phenol reduced the binding of phenol equivalents by 85% indicating that hydroquinone was acting as a competitive inhibitor of phenol metabolism Hydroquinone Benzoquinone by HRP (Table 3). Both sets of incubations. Treatment remaining (pM) formed (pM) those containing phenol and those without. Complete" +IO m M phenol Negative controlsd 22 r 3b 6 r IC 75k IC 47 2 5* 71 23' 3 2 IC were shown to be peroxidase dependent and required the presence ofboth H2O2and HRP. Leirkocyie experiments. Similar results were observed when freshly isolated human "The incubation and HPLC conditions are as de- PMNs were stimulated by PMA to undergo scribed under Materials and Methods. Mean and standard deviation of five incubations. 'Difeen significantly from complete incubation using a two-tailed Dunnett multiple range test at p c:0.05. dCompOsiteof the complete incubation and incuba- tion + phenol in which H202and HRP were omitted. the oxidative burst. The addilion of phenol to incubations containing [lJC]l~ydroquinone. PMA. and PMNs resulted in an approxi- mately 70% increase in [I4C]hydroquinone equivalents binding to Icukocl.te protein over that bate addi cub: hyd i mg i of P ClUC binc a 12 in t k T: tabc sen. ane met: mye exar ing nala dost repc dam sive sure hlECHANISX1 OF BENZENE TOXICITY - 89 B6C3FI male y at the doses n. (B)Hydro!sofadminisj a percentage mur. Asterisk )cubationscone level of radiorotein (an indihighly reactive I \vas increased .d to these incuthe addition of le to similar inhenol reduced ents by 85% in.vas aciiiig aS a 101 metabolism of incubations. those without. dependent and H,Ozand HRP. Similar results isolated human MA to undergo ion of phenol to ]hydroquinone. in an approxi']hydroquinone yte protein over 100 1 II 0 .6 I 1 I I Phenol Concentration (mM) . 1 10 FIG. 2. Effects of various phenol concentrations on hydroquinone metabolism. Incubation conditions are as described under Materials and hlethods. The data represent the means and standard deviations of three esperiments. The 0and 0. I phenol treatment groupsdiffer significantly(P 0.05:paired I test). that seen when ['4C]hydroquinonewas incu- lated that toxicity was due to a combination bated in the absence of phenol (Table4).The of metabolites rather than any one metabo- addition of PMA to the PMN-containing in- lite. Here we report that the coadministration cubations resulted in a ninefold increase in of phenol and hydroquinone results in the ex- hydroquinone binding (from 0.9 to 6.6 nmol/ tensive bone marrow damage characteristic mg protein) over that observed in the absence of benzene toxicity, whereas the coadminis- of PMA. However when phenol was also in- tration of combinations ofthe other phenolic cluded in the incubations. hydroquinone metabolites had no significant effect on bone binding increased to 10.S nmol/mg protein. marrow cellularity. a 1?-fold increase over the binding observed Metabolism studies employing the model in the absence of PMX. peroxidase enzyme HRP and isolated human PMNs support the hypothesis that the syner- DISCUSSION gistic effects of phenol and hydroquinone in ~.i,*aore due to a localized peroxidase-medi- The failure of the individual benzene metabolites to reproduce the myelotosicity observed following benzene exposure has been an enigma. This failure ofthe known benzew metabolites to reproduce benzene-inducd myelotoxicitv has led some investigators to "examine reactive intermediates formed du; ated metabolism of these compounds within the bone m2rrow.The.presence of phenol in incubations containing HRP induced a significant stimulation in hydroquinone oxidation to 1,Cbenzoquinonc and an increase in binding of hydroquinone to protein. The HRP-catalyzed metabolism ofhydroquinone ing ring opening such as tram,irans-muco- and the binding of [''C]hydroquinone equiv- naldehyde. The administration of near lethal alents to protein was increased by 30-40% by dosesof tramtrans-muconalde hyde has been the addition of phenol into the incubations. reported to result in moderate bone marrow Although a 30-407'0 stimulation of hydroqui- damage (Witz el a/.. 1985) but not the exten- none metabolism may seem insufficient to sive damage that accompanies benzene expo- account for the major differencein bone mar- sure (Green er a/.. 198I). We therefore postu- row damage observed when hydroquinone is 90 EASTMOND. SMITI3. AND IRONS TABLE 2 PHENOL-DEPENDENT STIMULATIONOF [ "C]HYDROQUINONE BINDISGTO RAT LIVER PROTEIN TABLE 3 lSlllUl rlON OF PEROXIDASE-MEDIATEDPHEXOL BINDING BY HYDROQUINONE nmol hydroquinone Sample nmol ["C]phenol Treatment equiv. bound/mg protein size equiv. bound/ mg protein' Perc'entage Complete" 10.4 2 3.2 10 $Phenol Complete -HlO: -HRP +Phenol' -+Phenol H202 +Phenol - HRP 14.4 ? 3.7' 9.4 2 2.9= 2.4 2 1.4d 3.2 ? 0.5d 13.4 5 3.5 2.5 2 1.6/ 3.3 ? 0.9/ +IO Completeb 115 IO0 6 6 Complete 75 PM hydro- -6 quinone Complete HzOz 6 40 27 3SC 23 6 Means of duplicate experiments. 6 'The complete incubation contained HRP (0.06 pg/ "The incubation conditions are as described under Materials and Methods. * Differs significantly from tbe corresponding com- plete incubation using Student's I test at p < 0.05. Mean and standard deviation. Diffen significantly.fromthe complete incubation using a one-railed Dunnett multiple range test at y < 0.05. Phenol (IOmbf) was added lo the complete incuba- ml). Hz02( I mM), ['4Cl~heno(l10 mM). and 0.9 mg/ml boiled rat liver postmitochondrial supernatant in phosphate bufler (0.I M. pH 7.4). The incubations were performed for 2 min at 37'C in a shaking incubator and terminated by the addition of 5% trichloroacetic acid. Protcin binding was determined as described under Ma- terials and Methods. Represents an 85% inhibition from the complete in- cubation after correcting for background binding. tion. 'Differs significantly fromthe +phenol treatment us- ing a one-tailed Dunnett multiple range test at p c 0.05. of metabolic activation of hydroquinone could be occumng within the bone marrow following exposure to benzene. administered separately and with phenol. this The observation that phenol stimulates hy- stimulation can be increased by simple modi- droquinone metabolism rather than acting as fications of the incubation conditions. By re- ducing the concentration of HRP and thereby reducing the amount of metabolism, TABLE 4 the phenol-induced stimulation of hydroqui- PHEXOL-IXDUCED STISIULATION OF HYDROQUI- none metabolism increased from 30 to nearly NOSE BISDINGDURISG THE PMA-STIMULATE0Dx1200%. Similar incubations employing iso- DATIVE BURST OF HUMAN LEUK- OCY-TES lated human myeloperoxidase showed a phenol-induced stimulation of lq-droquinone metabolism of greater than 400% (data not Treatment nmol hydroquinone equiv. bound/mg protein shown). This is particularly relevant to benzene toxicity in light of results reported here and elsewhere that the administration Complete" +IO mbi phenol -PMA~ 6.6 2 0.7" 10.8 ? 2.7' 0.9 t0.3' of hydroquinone by itself can cause mild "The incubation conditions are as described under bone marrow damage (Gad-El-Karim ('t ai.. Materials and Methods. 1985b). The stimulation of hydroquinone metabolism by low concentrations of phcnol and at low peroxidase concentrations, conditions analogous to those occurring in the bone marrow, suggests that this mechanism Mean and standard deviation of four incubations. D X e s significantly from the complete incubation using a two-tailed Dunnett multiple nnge test at p '< 0.05. Incubations containing phenol Nithout PMA showed no dimerence from the -PMA incuhations. al =\ Pi be OX co 19 an it! ca qu inr thi be m; an bir bir dic til-- hyi spt inc mr mz Pa: PrL cln co1 1% StU 19; der drc der Spe leu me Pan re!: tair C)z- arc ant OXi. er L . ED PIIESOL' SE Percenrage 100 35' 3' HRP (0.06pg/ . and 0.9 rnblml natant in phosations were per;incubator and ,\oroacetic acid. ibed under hla- he complete in: binding. ydroquinone lone marrow :imulatcs hyhan acting as F HYDROQUI- \tULATED 0x1- -> -- oquinone /me protein .0.7* I 3.7' :0.3' iescribed under incubations. 4ete incubation range test at p without PMA cubations. -MECHANISM OF BENZENE TOXICITY 91 a competitive inhibitor of the peroxidase en- ai. (1984) have prcscnted evidence that myzyme is somewhat anomalous. However. elotoxicity of benzene is accompanied by an phenol and other phenolic compounds have increased production of thiobarbituric acidbeen previously reported to stimulate the per- reactive products and increased levels of suoxidase-mediated oxidation of a variety of peroxide dismutase in the bone marrow. compounds (Yamazaki. 1958; Danner ef ai.. Each of these results suggest a potential role r 1973;Kalyanaraman iv a/., 1985).The mech- for 02-and free radical generation during anism has been proposed to involve the abil- benzene toxicity such as that which occurs ity of the phenoxy radical (derived enzymati- during the oxidative burst of PMNs. cally from phenol) to directly oxidize hydro- The increased conversion of hydroquinone quinone to the semiquinone radical. thereby to I .Cbenzoquinone by myeloperoxidase in increasing the speed of a rate-limiting step in the presence of phenol and the ability of these the overall oxidation of hydroquinone to 1.4- two compounds to produce myelotoxicity in benzoquinone (Yamazaki, 1958: Kalyanara- vivo provides evidence supporting previous man cf ai., 1985). Consistent with this mech- reports that I ,Cbenzoquinone is the proxianism are the results demonstrating an inhi- mate species responsible for benzene toxicity bition of peroxidase-mediated [14C]phenol (Kracke and Parker, 1934; Greenlee et al., binding by hydroqunione (Table 3)which in- 1981: Irons et ai., 1981; Irons, 1985). 1,4- dicate that hydroquinone acts as a competi- Benzoquinone has been shown to bind to tive inhibitor of phenol oxidation by HRP. protein and DNA and interfere with critical The ability of stimulated PMNsto convert cellular processessuch as microtubule assem- hydroquinone into a reactive protein binding bly and the synthesis of DNA and RNA species and the 70% increase in this binding (Rushmore et al.. 1984; Tunek et al., 1980; induced by phenol suggest that leukoc\-te- Irons et al.. 1981: Irons, 1985: Schwartz et al., mediated metabolism by myeloperoxidase 1985; Post et al.. 1984). In addition, the in may be an important metabolic activation vivo administration of other quinones such as pathway in benzene toxicity. This has been lapachol and Adriamycin have been reported proposed for other xenobiotics as well. in- to cause bone marrow damage (Morrison et cluding estrogens. aqlamines, and phenolic al.. 1970;Carter. 1975).Snyder and Engelsbcompounds (Klebanoff, 1977;Tsuruta e[ ai.. erg ( 1982) were not able to reproduce the re1985: Eastmond et ai., 1987a,b). Recent sults of an earlier report demonstrating mystudies by Trush and co-workers (Trush et ai., elotoxicity following I ,4-benzoquinone ad- 19S5: Cathers and Trush. 1986) have also ministration it? v i w (Kncke and Parker, demonstrated that benzo[a]pyrene-7,8-dihy- - 1934). The m'ost Iikhly explanation for the drodiol is converted in a peroxidase-depen- failure of it? vivo administered I .4-benzoqui- dent reaction to a binding and mutagenic none to mimic the hematopoietic toxicity of species during the oxidative burst of human benzene is due to its reactive properties. Adleukocytes and have suggested that this ministered I .4-benzoquinone reacts rapidly mechanism may be involved in the hemato- with blood and plasma proteins and does not poietic toxicity observed with benzo[a]py- reach the bone marrow in sufficient concenrene administration. The bone marrow con- tration to produce myelotoxicity (Ironsel ai., tains 90% of the body's granulocytic leuko- 1981). cytes (Andrews et ai., 1979), many of which Recently, it has been reported that the adare capable of undergoing the oxidative burst ministration of high doses of thc putative and contain considerable levels of myeloper- benzene metabolite 1rtrns.fratts-muconaldeoxidase (Zakhireh and Root, 1979; Bainton hyde to rodents could result in moderate el a/..1971). Recent experiments by Khan ef bone marrow damage (Witz ef a/.. 1985). Al- . l: 92 EASTMOND. SMITH. AND IRONS I though this findingcould partially explain the added to in vitro incubations or even that of mechanism of benzene toxicity. other me- phenol produced endogenously from ben- tabolites and mechanisms are probably in- zene in vivo. This explanation is supported by volved since numerous in vivo and cellular recent studies by Gilmour el ai.(1986)which metabolic studies have failed to observe indicate that the metabolism of phenol trans.rrarzs-muconaldehyde formation dur- formed from benzene in microsomal inbuba- ipg benzene exposure (Snyder cf al., tions differs from the metabolism of phenol 1981; Rusch et ai.. 1977: Billings. 1985). when added directly to microsomal incuba- trans.trans-Muconic acid. the subsequent ox- tions. Other important considerations could idation product of trans.rrans-muconalde- be the regimen of metabolite administration hyde, has. however. been isolated (Parke and or the subsequent differential distribution of Williams. 1953:Gad-El-Karim et al.. 1985a). these metabolites within the body. The regi- The high reactivity of muconaldehyde and men of administration (Le., twice a day) ap- the numerous cellular defenses within the pears to be important in producing bone liver, such as glutathione and aldehyde dehy- marrow suppression. The daily single admin- drogenase (Brabec, 1981; Goldstein cf al., istration of the metabolites did not result in 1982).make it unlikely that this reactive alde- significant bone marrow toxicity (data not hyde would be able to travel from the interior shown). Even using the combined treatment of the liver to the bone marrow in significant protocol described in this study, daily admin- quantities without being detoxified. The istration results in a gradual deminution of ip administration of large quantities of bone marrow suppression over a 2- to 4-week trans.trans-muconaldehyde would circum- period. However, discontinuous administra- vent hepatic detoxification, thereby allowing tion (3-4 days/week) results in prolonged this reactive aldehyde to reach the bone mar- bone marrow suppression for the duration of row at cytotoxic concentrations. The detec- treatment. Regimen-dependent bone mar- tion ofboth phenol and hydroquinone within row suppression is consistent with previous the bone marrow in vivo following benzene reports that benzene-induced bone marrow exposure (Rickert et al.. 1979),plus our find- toxicity is cell-cycle dependent (Irons et al.. ings that the administration of phenol and 1979; Muirhead et a!.. 1980: Pfeifer and hydroquinone in combination produce ex- Irons. 1981). tensive myelotoxicity, indicate that an inter- In summary. we report that myelotosicity action between phenolic compounds repre- such as that observed following benzene ex- sents a likely alternative mechanism for ben- posure can be reproduced by the coadmin- zene toxicity. isthation of phen51 and-hydroquinone, two One perplexing problem is the inability of major metabolites of benzene. The hemato: administered phenol to produce bone mar- poietic toxicity observed during benzene ex--. row toxicity in vivo (Gad-El-Karim et al.. -/- posure therefore appears to depend primarily _--1985:Snyder and Engelsberg. 1982: Mitchell, U -~ ---1971 ) since the in vitro biotransformation of z ? e to phenol aniydroquinone: (2) a sub- phenol by the cytochrome P-450monooxygenase enzymes has been reported to result in s~n--e--oqnuee-nint stheleecbtoivnee accumulation mar--ro-w;_(3)a of h?&quiz localized in- the formation ofhydroquinone (Tunek er al.. te-ra_ctio-n-o b h m o l and hvdromon-thi 1980; Sawahata and Neal, 1983; Smart and sis,te,mastarobably thmgthaphmnal4n- Zannoni, 1984). A potential explanation for duced stimulation of the myeloperoxidasg- this phenomenon is that the in vivo hepatic de_p_en_d-e_n_t _co_n-v-e-rsion of hydraquinone to metabolism of exogenously administered 1.4-benzoquinone: and (4) the reaction of phenol differs from the metabolism of phenol 1,4-benzoquinone with macromolecules re- sult cez ing oft dro, rest two con hYd Pro n &tic nent Pro? Schc ley. tanu ANC J. - di: Bl ANC (1' rw BAI~ (I' Pt E* BILL tic c?- BRA- In F. NI CAR C1 CAT de bc CAV ca ro DAL PA or even that of sly from benis supported by :I.( 1986) which ;m of phenol jsomal incubadism of phenol )soma1 incubajerations could administration distribution of 3ody. The regi\vice a day) aproducing bone y singleadminid not result in .kit? (data not lined treatment y. daily admindeminution of r a 2- to 4-week )us administra; in prolonged the duration of :nt bone rnar: with previous i bone marrow -it (Irons et al.. 0: Pfeifer and t myelotosicity ng benzene ex- the gadmin.oquinone. two :. The hematong benzene ex.pend primarily ivenion of benlone: (2) a subin of hydroquiI a localized inquinone at this 1 a phenol-in:eloperoxidasedroquinone to he reaction of omolecules re- MECHANISM O F BENZENE TOXICITY 93 suiting in an inhibition of critical cellular pro- cesses. Studies within our laboratories are be- ing conducted to elucidate the precise nature of this interaction between phenol and hy- action product by horseradish peroxidase and hydrogen peroxide. Arch. Biochem. Biophys. 156,759-763. EASTMOND.D. A.. FRENCH, R. C.. Ross, D., AND SMITH, M. T. ( 1987a).Metabolic activation ofdiethylstilbestrol by stimulated human leukocytes. Cancer droquinone, to investigate the genotoxicity Lett. 35.79-86. resulting from the coadministration of these two compounds, and to assess the relative contribution of both tmns,trans-muconaldehyde and 1,4-benzoquinone to myelotoxicity EASTMOND.D. A., FRENCH, R. C.. Rdss. D., AND Shimi. M.T.( 1987b).Metabolicactivation of I-naphtho1and phenol by a simplesuperoxide-generatingsystem and human leukocytes. Chem.-Biol. Interact., in press. produced by exposure to benzene. EASThlOND. D. A. SMITH, M. T.,RUZO, L. 0..AND Ross. D. (1986). Metabolic activation of phenol by human myelopcroxidaseand horseradish peroxidase. ACKNOWLEDGMENTS Thiswork was supported in part by the National Foundation for Cancer Research, the Health Effects Component of the University of California Toxic Substances Program. and a Graduate Research Award from the School of Public Health, University of California. Berkeley. W e thank Come N. Smith for expert technical assistance. Mol. Plrnrmacol. 30,674-679. GADEL-KARIM. M. M., SADAGOPA RAMANUJAM, V.M.. AHMED, A. E., AND LEGATOR, M. s. (1985b). Benzene myeloclastogenicity: A function of its metabolism. Amer.J. Ind. Med. 7,475-484. GAD-EL-ffiRIht, M. M., SADAGOPA RAMANUJAM, V. M., AND LEGATOR, M. s. (1985a). tmam,trnns-Mu- conic acid. an open-chair urinary metabolite of benzene in mice. Quantification by high-pmsurr liquid chromatogtnphy. Xenobiorica 15.2 11-220. GADEL-KARIM. M. M., SADAGOPA RAMANUJAM, REFERENCES V. M.. AHMED, A. E,AND LEGATOR. M. S. (1985b). Benzene myeloclastogenicity: A function of its metab olism. Anier.J.Ind. Med. 7,475-484. ANDREWS, L. s., LEE, E. W.,WITMER, c. M.,KWSIS. GILMOURS., K.,KALF,G.F.,AND SNYDER, R. (1986). J. J.. AND SNYDER. R. (1977). Effectsof toluene on the Comparison of the metabolism of benzene and its me- disposition and hemopoietic toxicity of ('H]benzene. tabolite phenol in rat liver microsomes. Adv. Exp. Biochcln. Piiarrnarol. 26,293-300. hied. Biol. 197,223-235. ANDREW'S. L. s., SASAhE, H. A., A N D GILLETTEJ, . R. GOLDSTEINB.. D. (1977). Hematoxicity in humans. J. (1979). 'H-Benzene metabolism in rabbit bone mar- Toxicol. Environ. Health Stippl. 2,69- 105. row.LVcSci. 25,567-572. GOLDSTEINB.. D. (1983). Benzene is still with us.Amer. BAISTON, D.F., UUYOT, J. L.,AND FARQUHAR. M. G. J. Ind. AfCd. 4,585-587. (1971). The development of neutrophilic polymor- GOLDSTEtN. B. D.,WIT& G..JAVID.J.. AMORUSO, phonuclear leukocytes in human bone marrow. 1. M. A.. ROSSMAN.T.. AND "OLDER. B. (1982). Muco- EXP.\led. 134,907-934. naldehydt. a potential toxic intermediate of benzene BILLIXGSR.. E. (1985). Mechanisms of catechol forma- metabolismJdr. EYPM. ed. Biol. 136A. 331-339. tion from aromatic compounds in isolated rat hepato- GREEN,J. D.. SNYDER. C. A.. ~ B u EJ.,. G O L D S ~ N , cytes. Drug Merab. Dispos. 13.287-290. B. D., AND ALBERT. R. E. ( I98 1) Acute and chronic BRABEC, M. J. (1981). Aldehydes and acetals. In Parry's dose/response effect of benzene inhalation on the pe- InditsrrinlHygieneand Toxicology(G. D. Clayton and ripheral blood, bone marrow. and spleen cells of CD- F. E. Clayton, Eds.), Vol. 2A, pp. 2629-2669. Wiley, 1male mice. To.xico1. Appl. Pharnrncol. 59,704-214. New York. GREENLEE.W. F.. GROSS. E. A.. AND IRONS. R. D. CARTER, S. K.(1975). Adriamycin-a review. J. h'arl. (198la). Relationship between benzene toxicity and Cancer Insr. 55. 1265-1274. the disposition of "C-labeled benzene metabolita in CATHERSL. . E.. AND TRUSHM. .A. (1986). Neutrophil- the rat. Chem.-Biol. Inrernct.33,755-299. derived oxidants as mediators of chemical toxicity in GREENLEE. W.F.. SUN,J. D.. A N D BUS. J. S.(198Ib). A bone marrow. To.ricologist 6, 157. proposed mechanism of benzene tonicity: Formation CAVALIERI. E., AND ROGAN,E. (1985). Role of radical of reactive intermediates from polyphenol m e t a b cationsin aromatic hydrocarbon carcinogenesis.Eiivi- lites. To.vico1. Appl. Pharmacol. 59, 187-195. roii.Health Pmpcct. 64.69-84. HASEMANJ..K.,CRAWFORD, D. D. HUFF. J. E.. BOOR- DANNER. D. J.. BRIGNAC.P. 1. ARCENEAUX. D..A N D MAN. G. A.. AND LlCCONNELL E. E. (1984). Results PATELV, . (1973). The oxidation of phenol and its re- from 86 two-yearcarcinogenicitystudiesconducted by . -94 WSTMOND. SMITH. AND IRONS the NationalToxicology Program. J. Toxicol.Environ. lyzed activation of pancetnmol. Biochem.Pharmacol. Health 14.621-639. 31,1363-1368. HIMMELHOCSH..R.. EVANS, W. H.. MAGE. M. G..AND MORRISON, M.. AND ALLEN. P. z. ( 1966). hctoperoxi- PETERSON. E. A. (1969). Purification of myeloperoxi- d m : Identification and isolation from the harderian dase from the bone marrow of the guinea pig. B i u and lacrimal glands. Science 152, 1626- 1628. chemistry8.914-921. MORRISON, R. K.. BROWN, D.E. OLESON, J. J.. AND IARC (1982). Benzene. IARC hfonogr.Eval. Carcinog. Risk C!reni.hfan29,93-148. -COONEY,D. A. (1970). Oral toxicology Qudies with lapachol. Toxicol.Appl. Pharmacol. 17, 1 1 I. r IRONS. R. D. (1985). Quinones as toxic metabolites of MUIRHEADK, . A., IRONS, R. D..BRUINS, R.. AND benzene. J. To.xico1. Environ.Health 16,673-676. HOIUN, P. K. (1980). A rabbit bone marrow model IRONS, R. D.. GREENLEE. W. F., WIERDA. D.. AND Bus, system for evaluation of cytotoxicity: Characterization J. S. (1982). Relationship between metabolism and of normal bone marrow cell cycle parameten by flow toxicity: A proposed mechanism for the formation of reactive intermediates from polyphenol metabolites. cytometry. J. Hisrochem.Cytochetn.28,526432. NlOSH (1974). Recommended Srandard /or Occupa- Adv. Esp. hied. Bid. 136A. 229-243. IRONS. R. D.. HECK. H. D.. MOORE. B. J., AND MUIR- HEAD, K. A. (1979). Effects of short-term benzene ad- tional Expostireto Benzene. pp. 74-134. US. Dept. of Health. Education, and welfare. Washington, DC. OSBORNE, J. C.. MEIZLER, M., AND NEUMANN. H.G. ministration of bone marrow cell cycle kinetics in the rat. Toxicol.Appl. Phamacol. 51,399-409. ( 1980). Peroxidase activity in the rat zymbal gland and its possible role in the metabolic activation of amino- w.IRONS, R. D.. NEPNN, D.A., A N D PFEIFER, R. (198 I). Inhibition of lymphocyte transformation and nicrotabule assembly by quinone metabolites of ben- stilbenes in the target tissue. Cancerk i t . 8.221-226. PARICE, D. V.. AND WILLIAMS. R. T. (1953). Studies in detoxication. The metabolism of benzene containing zene: Evidence for a common mechanism. J. Rt>ticti- loendothel.Soc. 30,359-37 1. c.KALYANARAMAN. B., FELIX, c.. AND SEALY, R. C. (1985). Semiquinone anion radicals of cate- chol(amine)s. catechol estrogens, and their metal ion complexes. Enriron. Healtir Perspect.64, 185-198. KARNOVSKY. M. J., AND ROBINSON. J. M. (1981). Con- tribution of oxidative cytochemistry to our under- standing of the phagocytic process. In Histochetnrstry: The ii'idenrng Horizons (P. J. Stoward and J. M. Po- lak. Eds.). pp. 46-66, Wiley. New York. ("Clbenzene. Biochem. J. 54.23 1-238. PFEIFER. R. W., AND IRONS, R. D.(1981). Inhibition of lectin-stimulated lymphocyte agglutination and mitogenesisby hydroquinone: Reactivity with intracellular sullhydryl groups. Exp. hfol.Pathol.35,189- 198. POST.G.B. SNYDER. R.. AND L L F . G. F.(1984).Inhi- bition of mRNA synthesis in rabbit bone m m w nuclei in vitro by quinone metabolites of benzene. Chem.-Bioi.Inferact.50,203-21 1. RICKERT, D. E. BAKER, T. S.. Bus, J. S.. BARROW, C. S.,AND IRONS, R. D.(1979). Benzene disposition in the rat after exposure by inhalation. Toxicol.Appl. KHAN, W. A.. GUPTA,A.. SHANKERU, ., AND PANDYA, K.P. (1984). Involvement of iron and free radicals in benzene toxicity. Biochem. Pharmacol. 33, 2009- Pharmacol. 49.4 17-423. RUSCH.G.M., LEONG. B. K.. AND LASUN, S. (1977). Benzene metabolism. J. Tosirol. Enriron. Health 2012. S ~ p p l2.,23-36. KLEBANOFSF..J. (1977). Estrogen binding by leukocytes RUSHMORE, T.,SNYDER, R., AND QLF, G. (1984).Co- during phagocytosis. J. Exp.hied. 145,983-998. valent binding of benzene and its metabolitesto DNA KRACKE.R. R., AND PARKER, F. P. (1934). The etiology in rabbit bone marrow mitochondria itr vitro. Chem.- of granulopenia (agranulocytosis) with particular ref- Biol. Interact.49,133- 154. erence to drugs containing the benzene ring. Amer. J. SAMMEIT. D., LEE. E. W., KoCSIS, J. J.. AND SNYDER. Clin.Pathol.4,45 1-469. R. (1979). Partial hepatectomy reduced both the me- MARKERT. M.. ANDREWSP,. C., AND BABIOR, B. M. tabolism and toxicity of benzene. J. Toxicol.Enriron. (1985). Measurement of 02-production by human Health 5,785-792. neutrophils. The preparation and assay of NADPH SAWAHATA. T..AND NEAL, R. A. (1983). Biotransfor- oxidase-containingparticles from human neutrophils. mation of phenol the hydroquinone and catechol by In Methods in Enzymology (L.Packer, Ed.), Vol. 105, rat liver microsomes. hfol.Pharmacol.23.453460. pp. 358-365. Academic Pms. Orlando, FL. SAWAHATA. T.. RICKERT. D. E.. AND GREENLEE, W. F. METZLERM, . ( 1984). Biochemicaltoxicologyof diethyl- (1985). Metabolism of benzene and is metabolites ir. stilbestrol. Rev. Biochem. Toxicol.6. I9 1-220. bone marrow. In ToxicoloKs ofthe Blood and Bot11 MITCHELL,J. R. (197 I). Mechanism ofbenzene-induced Marrow(R. D. Irons. Ed.). pp. 141-148. Raven Press. aplastic anemia (abstract). Fed. Proc. 30,56 1. New York. c.MOLDEUS. P.. ANDERSSON, B.. RAHIMTULA, A,, AND ~ W A R T Z . S..SNYDER. R.. AND L L F . G.F. (19851 BERGGRENM.. ( 1982).Prostaglandin synthetasc-cata- The inhibition of mitochondrial DNA replication i, ..,PI. Pharnmcnl. ,). bctopcroxi'1 the hardcrian - 162s. 50s.J. J.. ASD -.gy studies uith 1-1 I. ups. R.. A S D marrow model 'h3naerization >metersby flow s. 526-532. rd ,for Ocntpa.J.U S . Dcpt. of iington. DC. : U S I A ~ ~H..-G. .mbd $and and ation of amino:{I. 8.221-226. 453). Studies in zene containing 3. I). Inhibition of ation and mitoirh intncel1ul;ir 5. IS9-19s. F. (19S-l).Inhime marrow nuts of benzene. J. S.. BARROW', icne disposition :. Tosicol. .4ppl. jus. 5. (1977). h i r o n . Healilr .G.(1984).CO- ibolites to DSA in ritro. Clteni.- _..4SD ShYDER. xiboth the me- oxicol. Enriro. ;3). Biotransforand catechol by '. 23,453460. .REESLEE. w. F. s metabolites in 9 l d and Bone 48.Raven Press, LF. G. F.(1985). \ replication in -MECHANISM OF BENZENE TOXICITY 95 i*itroby the metabolitesofbenzene. hydroquinone and TuNEK.A..PLA~I.LL. PRZYBYLSKI, M.,ANDOESCH, f-benzoquinone. Cliet,n.-Biol.Interact. 53.327-350- F. ( 1980). Multi-step metabolic activation of benzene. SMART. R. C..AND ZANNONVI,.G.(1984).DTdiaPhorase and peroxidase influence the covalent binding of Effect of superoxide dismutase on covalent binding to microsomal macromolecules. and identification of the metabolites of phenol. the major metabolite of glutathioneconjugates using high pressureliquid chro- benzene. Mo/.Phamacol. 26,105-I I I. matography and field desorption m a s spectrometry. SNYDER. R.. AND ENGELSBERGB. N. (1982).Effectsof benzene and other chemicals in 72 hr 'Fe utilization C in S\virsalbino mice. Toxicologisf2, I2I. SNYDER. R.. LONGACRES,.L.. WITMER. C. M.. Kocsls, J. J.. ANDRE\\". L. S.. AND LEE, E. W. (1981).Biochemical toxicology of benzene. Rev. Biochon. ToxirOl. 3.123- 154. TRUSH. M. A.. SEED. J. L.,AND KENSLER,T. W. (1985). Oxidantdependent metabolic activation of polycyclic aromatic hydrocarbons by phorbol ester-stimulated human polymorphonuclear leukocytes Possible link between inflammation and cancer. Proc. Natl. Acad. Sci.LS.4 82,5 194-5198. TSURUTA. Y..SUBRAHMANYAV.MV.,. MARSHALL.W., AND OBRIENP. . J. (1985).Peroxidase-mediated irreversible binding of arylamine carcinogens to DNA in Chem.-Biol.Interact.33. 1-17. USEPA (1980).Ambient cl'aler Qiiality Criteriafor Ben- zene, EPA 44O/5-80-O18.U.S. Environmental Protection Agency, Washington. DC. WIERDA. D.. AND IRONS, R. D. (1982).Hydroquinone and catechol reduce the frequency of progenitor B lymphocytes in m o w spleen and bone marrow, Im- munopharmacology4.4l-54. WISE. R. W.. ZENSERT,.V.. KADLUBAR, E E.AND DAVIS, B. B. (1984).Metabolic activation ofcarcino- genic aromatic amines by dog bladder and kidney prostaglandin H synthase. CancerRes.44* 1893-1897. W I G..~b o ,G. S., AND GOLDSTEIN,3. D. (1985). Short-term toxicity of trans.trans-muconaldehyde. Toxicol.Appl. Pharmucol.80.5 I1-5 I 6. YAMAZAKI., (1958).The oxidoreductive feature of intermediates formed in the reaction of peroxidase. In intact polymorphonuclear leukocytes activated by a Proceedings of the International Symposium on En- tumor promoter. Chem-Biol.Interact. 53.25-35. zyme Chemistry Tokyoand KVO~1O957 (K.Ichihara, TUXELA.. OLOFSSONT, ., AND BERLIN, M. (1981). Ed.), pp. 224-229.Pan-PacificP m Tokyo. Toxic effects of benzene and benzene metabolites on ZAKHIREHB.. AND ROOT, R. (1979).Development _mnulopoietic stem cells and bone marrowcellularity of oxidase activity by human bone marrow granulo- in mice. To-t-icolA. ppl. Phannacol. 59, 149-156. cytes. Blood54.429439. i