Chem. Res. Toxicol. 1990, 3, 333-339
333
Prostaglandin H Synthase Catalyzed Oxidation of Hydroquinone to a Sulfhydryl-Binding and DNA-Damaging Metabolite Michael J. Schlosser,' Robert D. Shurina, and George F. KalP Department of Biochemistry and Molecular Biology, Jefferson Medical College of Thomas Jefferson University, Philadelphia, Pennsylvania 19107 Received April 6, 1989
Hydroquinone, a metabolite that accumulates in bone marrow following benzene exposure,
was oxidized by prostaglandin H synthase (PHS) to 1,4-benzoquinone, which was measured by HPLC with reductive electrochemistry. Hydroquinone metabolism in the presence of cysteine generated a thiol adduct, which was identified as the monosubstituted cysteine conjugate of hydroquinone by HPLC with oxidative electrochemical and radiochemical detection. The time-dependent formation of both 1,6benzoquinone and the monocysteine-hydroquinone conjugate was monitored spectrophotometrically at 250 and 305 nm, respectively. Monocysteine-hydroquinone was formed at rates similar to 1,Cbenzoquinone formation in reactions without cysteine, suggesting that 1,Cbenzoquinone or its semiquinone intermediate is rapidly binding to sulfhydryls. The PHS-catalyzed activation of hydroquinone to 1,6benzoquinone or its thiol conjugate required the presence of either arachidonic acid or Hz02. The oxidative metabolism of hydroquinone also resulted in the formation of a reactive product(s) that irreversibly bound to DNA. This binding was time dependent and did not occur in the presence of heat-inactivated PHS. Metaboliteb) generated during hydroquinone oxidation also induced single-strand breaks in Bluescript plasmid DNA. The PHS/arachidonic acid catalyzed metabolism of hydroquinone to 1,4-benzoquinone and to product(s) that bound to sulfhydryls and DNA and caused strand breaks in DNA was prevented by indomethacin, an inhibitor of P H S cyclooxygenase. Because prostaglandin synthesis is elevated in bone marrow following benzene exposure and inhibitors of P H S cyclooxygenase prevent benzene-induced myelotoxicity, the activation of hydroquinone by P H S represents a possible mechanism for benzene's effects.
I ntroductlon Benzene is a widely used industrial chemical and environmental pollutant known to cause an increased incidence of aplastic anemia, lymphocytopenia, and leukemia in occupationally exposed workers ( 1 , 2 ) . Hydroquinone, a hepatic metabolite of benzene (3),concentrates in bone marrow following exposure of rats to benzene ( 4 ) . Administration of hydroquinone to mice is known to induce micronuclei in bone marrow ( 5 , 6 ) ,inhibit erythropoiesis (7),and lower the number of functional B-lymphocytes removed from marrow and spleen (8). Hydroquinone also causes sister chromatid exchanges in cultured lymphocytes ( 9 ) , results in DNA strand breaks ( l o ) ,and inhibits the synthesis of DNA and RNA (11-13). It is likely that many if not all these effects are mediated by the hydroquinone oxidation products, 1,Cbenzosemiquinone and 1,4-benzoquinone, which readily bind to thiols, proteins, and nucleic acids (14-19). Oxidation of hydroquinone to 1,4-benzosemiquinone and 1,4-benzoquinonecan occur both enzymatically and nonenzymatically. Hydroquinone autoxidizes at neutral pH to form 1,6benzosemiquinonewhich can disproportionate to 1,6benzoquinone (20,21). Horseradish peroxidase (HRP)' also catalyzes the oxidation of hydroquinone, yielding 1,Cbenzoquinone via 1,6benzosemiquinone at rates much faster than autoxidation (20,211,and has been shown to activate hydroquinone to metabolites that bind irreversibly to cellular protein (22). Leukocyte peroxidase from either neutrophils or macrophages also activates
* Author to whom correspondence should be addressed. Present address: Experimental Toxicology, Pharmaceutical Division, CIBA-GEIGY, 556 Morris Ave., Summit, NJ 07901.
hydroquinone to protein-binding species (22,231. Recent studies from our laboratory have shown that the majority of the protein-binding species generated during the macrophage peroxidase catalyzed activation of hydroquinone could be trapped with cysteine and identified as the monocysteine conjugate of 1,4-benzoquinone (23). Since neoplastic diseases arising from benzene exposure appear to originate in tissues containing peroxidases, hydroquinone oxidized to 1,6benzosemiquinone and 1,4benzoquinone by mammalian peroxidases may be important in the myelotoxic as well as the genotoxic effects of benzene (24,27). The peroxidase activity of prostaglandin H synthase (PHS)is known to oxidize a wide range of xenobiotics to one-electron oxidation products (25) and has been implicated in several toxic effects of benzene (26-28). We and others have demonstrated that PHS activity is increased in bone marrow following benzene exposure (26,28)and that PHS inhibitors (NSAIDs) administered to mice prevent benzene-induced bone marrow cell depression and micronucleus formation (27,28),as well as benzene's inhibition of myeloid cell development and stromal cell supported myelopoiesis (26, 28). The macrophage-rich bone marrow stroma regulates hematopoiesis and synthesizes prostaglandins whose formation is catalyzed by PHS (29),an enzyme contained in macrophages (30). Our laboratory recently demonstrated that phenolic metabolites of benzene are converted to reactive intermediates by macrophage peroxidase and PHS (23,31). Thus, it is not unexpected that macrophages are altered morphologically Abbreviations: HRP, horseradish peroxidase; kb, kilobase pairs;
MPO,myeloperoxidase;NSAIDs,nonsteroidal antiinflammatory drugs; PHS, prostaglandin H synthase.
Q893-228~/90/ 27Q3-Q333$Q2.5Q/Q0 1990 American Chemical Society
334 Chem. Res. Toxicol., Vol. 3, No. 4, 1990
and inhibited functionally in rats and mice following benzene exposure (32-34). Hydroquinone is also toxic in vitro to bone marrow stromal cells and to marrow-derived macrophages (26,35, 36). The stromal macrophage appears to selectively bioactivate hydroquinone by a peroxidase-dependent mechanism (36). Because hydroquinone is an excellent reducing cosubstrate for the peroxidase component of PHS (37),it is not surprising that hydroquinone bioactivation by macrophages can be driven by small amounts of arachidonic acid and inhibited by the NSAID indomethacin (23). The above results suggest that PHS at target sites may activate the benzene metabolite, hydroquinone, to compound(s) capable of reacting with cellular macromolecules. The study presented here investigates the PHS-catalyzed oxidation of hydroquinone and examines the sulfhydryl-binding and DNA-damaging effects of the reactive metabolite generated during this reaction. Also presented are results comparing the activation of hydroquinone by PHS to those of other purified peroxidase systems.
Experimental Procedures Materials. Hydroquinone, l,4-benzoquinone, trichloroacetic acid (TCA), T4 polynucleotide kinase, X phage DNA, and HPLC-grade acetonitrile and methanol were purchased from Fisher Scientific (Pittsburgh, PA). Arachidonic acid was from Nu-Check-Prep (Elysain, MN). ["CIHydroquinone (22 mCi/ mmol; >97% pure) was obtained from Amersham (Arlington Heights, IL). Aquasol-2, Protosol, and [ C X - * ~ P ] ~ C was T Pfrom Du Pont/NEN (Boston, MA). Hydrogen peroxide (30%), hematin, calf thymus DNA, and L-cysteine hydrochloride were purchased from Sigma (St. Louis, MO). Bluescript plasmid was a product of Stratagene, La Jolla, CA. Nick translation reagents and enzymes were purchased from Bethesda Research Laboratories. All other chemicals were of the highest quality available. Enzymes. PHS (54OOO units/mg of protein) was obtained from Oxford Biomedical (Oxford, MI). Cyclooxygenase activity was determined prior to experiments by monitoring the incorporation of molecular oxygen into arachidonic acid by use of a Yellow Springs oxygen monitor. One unit of PHS was defined as the amount of enzyme that exhibits an initial consumption of 1nmol of 02/min a t pH 7.0 at 37 "C in an assay containing 100 pM arachidonic acid and 500 pM phenol. Purified human leukocyte myeloperoxidase (MPO; 32 units/mg of protein) from purulent sputum was purchased from Elastin Products (Pacific, MO). One unit of MPO is the amount of enzyme that decomposes 1 pmol of Hz02/min a t pH 7.0 a t 25 "C with 4-aminoantipyrine as the H+ donor. Horseradish peroxidase type VI (HRP; 325 units/mg of solid) was obtained from Sigma; one unit forms 1 mg of purpurogallin in 20 s from pyrogallol a t pH 6.0 at 20 OC. 1,4-BenzoquinoneDetection. The PHS-catalyzed oxidation of hydroquinone was determined by measuring 1,l-benzoquinone production. HPLC analysis of 1,4-benzoquinone was performed by using a Beckman 5-pm C18 analytical column (4.6 mm X 25 cm) preceded by a guard column and reductive electrochemical detection as described by Lunte and Kissinger (38).The electrochemical detector consisted of a carbon polymer working electrode and a Ag/AgCl, reference electrode. The PHS-catalyzed oxidation of hydroquinone to 1,4-benzoquinonealso was measured by monitoring 1,4-benzoquinone at 250 nm on a Beckman DU spectrophotometer. This wavelength represents near maximum for l,4-benzoquinone absorbance and a nadir for hydroquinone absorbance. An extinction coefficient of 15.4 mM-l cm-' was used for 1,4-benzoquinone absorbance a t 250 nm. Monocysteine-Hydroquinone Detection. The detection of a reactive intermediate generated from the PHS-catalyzed oxidation of hydroquinone was determined by trapping the reactive intermediate with cysteine. We have previously demonstrated that 1,4-benzoquinone and/or 1,4-benzosemiquinone will react with cysteine to yield the monocysteine conjugate of hydroquinone, which was characterized and quantified by using HPLC with radiochemical and oxidative electrochemical detection (23). In
Schlosser et al. the present study, monocysteinehydroquinonewas detected both by using HPLC with oxidative electrochemical detection and by following its formation spectrophotometrically at 305 nm (e = 4.03 mM-' cm-'). This wavelength allows for an adequate absorbance of monocysteinehydroquinone with minimal interference from hydroquinone. Incubation Conditions. Reaction mixtures consisted of 0.1 M sodium/potassium phosphate buffer, pH 7.0, with 100 pM hydroquinone, 5 @/mL PHS, 1pM hematin, and 100 pM of either arachidonic acid or HzOz. Reaction mixtures were preincubated with either indomethacin or its vehicle, ethanol (1% final concentration), and then the reactions were initiated by adding hydroquinone followed by arachidonic acid or HzOz. In some incubations, 100 pM cysteine was added at the beginning of the reaction. When PHS was replaced by other peroxidase systems, MPO (0.3 pg/mL) and HRP (0.05 pg/mL) were used at concentrations that gave rates of 1,4-benzoquinone formation comparable to that of PHS/HzOz-catalyzed reactions. For HPLC analyses, reaction mixtures were incubated for 2 min at 37 OC. For the detection of 1,4-benzoquinone,an aliquot of the reaction mixture was injected directly onto the HPLC system at the end of the 2-min incubation. For HPLC analysis of the monocysteinehydroquinone conjugate, 2-min incubations were terminated by adding TCA (5% final concentration) and freezing in dry ice. Reaction mixtures were thawed and extracted three times with 1.5 volumes of ethyl acetate to remove unreacted hydroquinone and 1,4-benzoquinone,and the aqueous phase was evaporated to dryness under vacuum. The resulting residue was dissolved in 200 pL of water and filtered prior to HPLC analysis. Recovery of monocysteinehydroquinone through extraction was 89%. Spectrophotometric analyses of 1,4-benzoquinone and monocysteinehydroquinone formation were monitored for 1.5 min at 25 "C. A base-line absorbance was established prior to initiating the reaction with arachidonate or HzOz. Reactions were preincubated for 1 min with indomethacin or 1%ethanol vehicle. Absorbance occurring in the absence of arachidonic acid/H202 or enzyme was subtracted from the complete system. In the case of monocysteine-hydroquinone,decreases in absorbance due to hydroquinone disappearance (
