Document QXn1ZRBOgqYYyw4vvz57KoNnE

.. , # ICANCER RESEARCH 48,47624765. September I. 1988] Role of Oxygen Radicals in Induction of DNA Damage by Metabolites of Benzen J. G . Lewis, W. Stewart, and D. 0.Adams Depurfmenf of Pafhology, Duke UniversityMedical Cenfer,Durham. Nonh Carolina27710 ABSTRACT Benzene is strongly suspected of being an animal and human carcina- gen, but the mechanisms by which benzene induces tumors of lymphoid and hematopoietic organs are unknown. Binding studies in rho suggest a very low level of covalent binding to the DNA of bone marrow elements. Since several metabolites of benzene have the potential to undergo autooxidation and thereby generate reactive oxygen intermediates, we have tested the hypothesis that benzene metabolites can induce DNA damage through the generation of oxygen radicals. Hydroquinone (HQX benzoquinone (BQ), catechol, and l,Z,J-benzeneMol (BT) were first tested for their ability to generate 01-at a physioloeical PH. BT,and to a lesser extent H Q , were autooxidized and produced significant quantities of OZ-.No detectable 4-was produced by catechol or BQ. Similarly, BT was very efncient at degrading DNA, and this degradation was inhibited by scavengers of Oz-, H201 and -OH.HQ did not degrade DNA but did induce single- and doublestrand breaks. In contrast to the action of BT, the breakage of DNA by H Q was not inhibited by scaven- gers of reactive oxygen intermediates. The metabolites which did not produce 0 2 - (catechol and BQ) did not induce signifkant breakage of DNA. Taken together, the data support the hypothesis that certain benzene metabolites can induce DNA damage through the production of oxygen radicals; they further suggest that other metabolites may act, via another mechanism, to damage DNA. strand breaks in DNA ofcultured cells, although the by which this occurs remains unexplained (12). In a (I3)* production of oxygen radica of benzene can generate oxy that these metabolites can i mTEW METHODS INTRODUCTION Benzene is a major organic product, solvent, and fuel additive; and, accordingly, humans are widely exposed to benzene (1). Exposure of humans and animalsto benzene can result in severe depression of the bone marrow; aplastic anemia, leukopenia, leukemia, and lymphoma have also been observed (1, 2). Benzene has been classified as a human carcinogen because of reported increases in the incidence of leukemia in workers exposed to high concentrations (3). Studies in animals have shown that benzene is metabolized in the liver to phenol, over a 2- x 30cm column of Chelex 100 (Bio-Rad) and br was included because iron induced breaks in the absence of reactive oxygen intermediates or semiquinone radical interne- DNA in each form was of page charges. This article must therefore be hereby marked adverfucmenf in 'accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Supported in p m by MH Gmu CA 44734 and ES02922. *The abbreviations Used are: CAT.c a t ~ h o l ;HQ,hydroquinone; BQ, bcnZoquinone; BT.1.2.4 benzeneuiol: SOD,superoxide dismutase: ED% concentration causing 50% loss of supercoiled DNA. 2). CAT (Fig. 2) and BQ (data no of cytochrome c. The addition of of cytochrome reduction by bo BT BQ CAT HQ C PST 1 -0c -sc -LIN Fig. 4. Breakage and degradation of 9X-174 supercoiled DNA by benzene metabolites. Type I 9X-174 DNA (0.1 re) was incubated with I mhi solutions of benzene metabolites for 60 min at 25'C and pH 7.4. The supercoiled (SO. open circular (00.and linear (UN)forms were then separated on a 1% agarose gel, stained with ethidium bromide, and visualized with ultraviolet light as previously describcd(l7.18).LansPsTl iscontrol@X-174DNAdigestedwith therestriction endonuclease PSI to yield the linear form. cY -- ~ ~ 05 5 5 sb Log Conccnrralion of ET IpMI Fig. 5. Effect of concentrations on the breakage of 6X-I74 DNA by ET. Type 1 9X-174 DNA (0.1 ~ 0w)as incubated with the indicated concentration of BT for 60 min at 25'C and pH 7.4. The percentage of DNA remaining in the supercoiled form was determined as described in 'Materials and Methods." and double-strand breaks resulting in the complete breakdown of supercoiled DNA to the open circular and linear form (Fig. 4). BQ and CAT caused very little breakage (Fig. 4). Effect of Dose of BT and HQ on Induction of DNA Strand . I, 2,48T Ha eo C A T i r+c race of 0,-generated by I mM solutions of benzene metabolites. 0, Producrion was measured by the SODinhibitable reducuon of cytochrome C pmiouslv dexnbed (14.15). Measuremenu were taken from the linear portion d Ihc cunes shown in Fig. 2. Cofwnns,mean of triplicate samples in which the s.V c 10%. 0. metaboiite alone; metabolite in combination with Fd311. Breaks. When BT and HQ were tested at multiple concentra- tions, strand breakage was induced in a dose-dependent manner (Figs. 5 and 6). BT caused complete loss of supercoiled DNA at concentrations as low as 5.5 x lod M. The EDs0 was 7 X lo-' M (Fig. 5). HQ was less potent and required higher doses (i.e., lo-' M to 1.1 x lom4M) for complete loss of supercoiled DNA (Fig. 6). The EDw for HQ was M (14-fold higher 75 nmol/min in the absence of iron and 17.4 nmol/min in the m n c e of 10 PM FeCl, (Fig. 3); much less 0 2 - was generated by HQ.even in the presence of iron (Fig. 3). In sum, BT at a mncentration of 1 mM can generate sufficient 0 2 - to achieve a concentration ofO2- of over 50 PM in 5 min (1 nmol in 1 ml = than BT). Kinetics of DNA Strand Breaks Induced by HQ and BT. The breakage of DNA by a concentration of 40 PM BQ was very rapid with maximal breakage achieved in less than 2 min (Fig. 7). Breakage by BT at the same concentration was somewhat slower in that 14% of the DNA remained in the supercoiled form at 2 min, and 4% was still in the supercoiled form at 5 aPaclt). of the benzene metabolites to damage DNA, as meas- by the induction of strand breaks. 4X-174 DNA was to 1 m M solutions of BT, HQ, CAT, or BQ. BT caused mmPlete degradation of the DNA (Fig. 4). H Q did not cause min (Fig. 7). After 5 min of exposure to BT, none of the DNA remained in the supercoiled form. Effects of Scavengers on Induction of DNA Strand Breaks. We next determined what portion of breakage of the DNA extensive degradation induced by BT but did cause single- could be blocked by inhibition of reactive species of oxygen. 4763 , .* BENZENE-MEDIATED OXIDATIVE DNA DAMAGE offer effective protection against 40 p~ concentratio (Fig. 9). By contrast, none of the inhibitors offered protection from breakage by H Q (Fig. IO). Albumin, the same concentrations as the radical scavengers, was in in protecting the DNA from breakage (Figs. 8 and 9). DISCUSSION These data indicate H Q and BT, at a physiological \ol! 00 I dl I 1 Log Concentration of HO I m M ) Fig. 6. The effect of concentration of the breakage of OX-174 DNA by HQ. Type 1 OX-174 DNA (0.1 p g ) was incubated with the indicated concentration of HQ for 60 min at 25'C and pH 7.4. The percentage of DNA remaining in the supercoiled form was determined in 'Materials and Methods." not inhibited by scavengers of reactive oxygen inte Taken together, these observations suggest that BT, lite of benzene, induced DNA damage, mostly th t.B T 0--0 HP -a mechanism. That SOD, catalase, and benzoate all inhibited breakag DNA by BT strongly suggests that BT was acting through metal-catalyzed Haber-Weiss reaction, in which 02-both Open C l r c l n 0 IO 20 30 40 50 60 Time of Exposore 10 4 0 p M BTor H Q Irninuler) Fig. 7. The effect of time on the breakage of OX-174 DNA by BT and HQ. Type 1 OX-174 DNA (0.1 p g ) was incubated with 40 piw concentrations of BT or HQ for the indicated times at 25'C and pH 7.4. The pemntage of DNA in the supercoiled form was determined as described 'Materials and Methods." 0 S"p*rco,led O p n Circler Llnrol 80 Control 4 o+p e~r Ipg Albumin Fig. 9. The effect of SOD on the breakage of OX-174 DNA by 40 phi and DH7.4with and without I up of SOD. One ue of albumin was included con&ol for nonspecific competitron of protein wiih the DNA. The mounts DNA in the supereoiled open circular. and linear form were determined described in 'Materials and Methods." Fig. 8. Effect of scavengers on the degradation of 4x147 DNA by 1 mM BT. Type 1 OX-174 DNA (0.1 p g ) was incubated with I mi BT for 30 min at W C , pH 7.4, with and without the indicated scavengers. Albumin was added to control for nonspecific competition of protein with the DNA. The amount of DNA in the supercoiled, open circular, and linear forms was determined as described in -Materials and Methods." % We tested the capacity of SOD,catalase, and benzoate, which scavenge 02-,HzOz, and .OH, respectively, to inhibit the breakage of DNA by BT and HQ. Benzoate and catalase were potent inhibitors of the breakage of DNA by BT and offered effective protection against the complete degradation of DNA induced by 1 mh.1 concentrations of BT (Fig. 8). SOD did not protect against 1 mM concentrations of BT (Fig. 8), but did Fig. 10. The effects of scavengers on the breakage of &Xx-174DNA by I HQ.Type 1 6X-174 (0.1 re) was incubated with 1 miw HQ for 30 min at and pH 7.4 with and without the indicted scavengers. Albumin was added control for n o d i comwtition of Drotein with the DNA. The amoun DNA in the suhrcoiled. ;pen cireulak and linear form was determined described in ',Materials and Methods." 4764 BENZENE-MEDLATED 0XIDATIVE DNA DAMAGE ducesmetal ions and dismutates to form H202(17). H202 could of benzene, the rate of cell division and DNA synthesis is high [hen interact with reduced metal ions to form the very reactive in bone marrow cells, and since metabolic incorporation of I4C hydroxylradical which breaks DNA without regard to sequence has been shown to occur in liver in this time (21), most of the ( o , z ~ ) . There was not, however, complete inhibition of break- counts observed in the marrow may have been localized in age by these scavengers, especially by SOD. This may be due normal bases rather than as DNA adducts. This point requires the lower turnover rate of SOD for its substrate than catalase. further investigation. Studies by Gill and Ahmed (23) also Although the units of activity are not directly comparable, one reported covalent binding of benzene to macromolecules of on determine the turnover rates under ideal conditions. The bone marrow. These studies also had very low counts (<15 5 ~ c i f iacctivity of the SOD used was 3016 units/mg. One pg dpm/femur or 86 dpm for 20 pooled femurs in combined nucleic or a total of 3 units was added in the assays. One unit of SOD acids not purified DNA) and did not control for the possible 5s ~3lculatedunder the standard conditions as described by metabolic incorporation of a few dpm of I4C via the C-1 pool, YlcCord and Fridovitch (15) can scavenge approximately 0.6 and the methods used for the separation of macromolecules of Oz-/rnin. One unit of catalase is defined as that amount were not sufficient to provide purities needed for such low of enzyme which can convert 1 Nmol of H202 to HZO/min. It amounts of radioactivity. Given the bioconcentration of the not possible to add more SOD to the higher concentration, metabolites in the marrow and the low covalent binding to because larger amounts of proteins began inhibiting nonspecific3ily by competing with DNA rather than by their specific scavenger activities. This is evidenced by the fact that albumin began protecting the DNA from breakage when more than 1 DNA in vivo, the data presented here offer an alternative genotoxic mechanism for some of the toxic and carcinogenic actions of benzene. ,,g was added (data not shown). The inability of these scavengers REFERENCES 10 inhibit more completely conversion of the supercoiled DNA 10 open circles may be a result of the sensitivity of the assay. Only a single-strand break is required to relax the supercoiled form of $X-174 DNA to the open circle form (17, 18), but much more extensive damage is required to convert the DNA IO the linear form or completely degrade the DNA (17, 18). High concentrations of BT completely degraded the DNA (Fig. -I)w,hereas lower concentrations caused conversion to open circles (Fig. 5). The scavengers were very efficient at protecting the DNA from conversion to linear form or complete degradation. Alternatively, the semiquinone radical intermediates formed in the autoxidation process may also interact with the DNA and contribute to the breakage. This could be especially true for HQ which was not inhibited by any of the scavengers tested. BQ and CAT induced little if any DNA breakage in these studies, which is in accordance with their inability to generate detectable 02-.Studies by Pellack-Walker and Blumer con- ducted on leukemia cells have shown that BT and BQ, but not HQ.induced alkali label sites or DNA breakage in these cells, and that BT induced breakage of DNA at a rate consistent with the rate of its autooxidation and the generation of 0.' (12). By Lonlrast. BQ induced immediate breakage, even at 4'C. Since B v is already fully oxidized, it could not produce Oz-. Furthermore. enzymatic or nonenzymatic intracellular reduction of BQ CJnnot explain the rapid and potent breakage of DNA in l c u h n i a cells by BQ at 4'C. A recent study suggesting that dlerations in chromatin structure alone in the absence of DNA damage can cause enhanced unwinding and elution of DNA under certain alkaline conditions offers a possible explanation of these findings (19). The observation, that BQ binds avidly to Woskeletal elements and disrupts their assembly (9), suggests that BQ may possibly alter chromatin structure and the elution Fttern of the DNA in this manner. This, however, does not mvidize the studies by Pellack-Walker and Blumer because dtered chromatin structure may itself pose a significant genetic to the host. Previous studies have demonstrated that polyphenol metabolites of benzene concentrate in the marrow of mice exposed to "diolabeled benzene or its metabolites (5). Very little radioac- "iv. however, appears in the DNA of marrow cells in animals cIPosed to "C-labeled benzene (8). In these latter studies, no con*rol was included for the possible incorporation of l4C via Ihec-1 pool into normal bases. Since the measurements were laken22 h after exposure, which allows time for the metabolism 1. Goldstein. B. D. Hematotoxicity in humans. J. Toxicol. Environ. Health, 2 (Suppl.): 1-4, 1977. 2. Snyder, R., and K-is, J. J. Current concepts of chronic benzene toxicity. CRC Crit. Rev. ToXieol, 3: 265-288.1975. 3. RinsLy, R. A., Smith, A. B., Hornung, R., Filloun, T. G., Young, R. J., Okun. A. H., and Landrigan, P. J. Benzene and leukemia. An epidemiologic risk assessment. N. End. J. Med., 316: 1044-1050. 1987. 4. Park D. V., and Williams, R. T. Detoxification. XLIX.Metabolism benzene containing ["CJbenzene. Biochem. J., 54: 231-238,1953. 5. Gmnlee. W. F., Gross, E. A., and Irons, R. D.Relationship between benzene toxicity and the dirposition of "C-labeled benzene metabolites in the rat. Chem.-BioL Interan, 33: 285-299, 1981. 6. Harrison. IC, and Randall, F. W.An application of bone marrow cultures to toxicity and therapeutics. J. Exp. Physiol.. 34: 141, 1948. 7. Nomiyama, K. Studies on poisoning by benzene and its homologues. Toxicity of benzene metabolites to hemopoiesis. Ind. Health, 3: 53,1965. 8. Arfellini, G..G n f i S, Colami. A. Mauullo, M., and Prodi. G. In vivo and in vitro binding of b e n n e to nucleic acids and proteins of various rat tissues and m o w organs Cancer Lett., 28: 159-168,1985. 9. Ptiefer. R. W.. and Irons. R. D. Alteration of lymphocyte function through sulfhydryl-dependent disruption of microtubule assembly. Int. J. Immuno- phannacol., 5: 467470,1983. 10. Cmnlce. W. F.,SUSJ. D, and Bus,J. S.A proposed mechanism of benzene toxicity: formation of reactive intermediates from polyphenol metabolites. Toxicol. Appl. Pharmacol., 59: 187-195, 1981. 11. Khan. W. A., Gupta k,Sharker, U., and Pandya, K.P. Involvement of iron and free radients b e n n e toxicity. Biochem. Phannacol., 33: 2009-2012, 1984. 12. Pellack-Walker. P.. and Blumer, J. L. DNA damage in L5178YS cells following exposure to benzene metabolites. Mol. Pharmacol., 30: 42-47. 1986. 13. Kaneko, M.. and Leader, S.A. Production of thymic glycols in DNA by A'- hydroxymphthyanine as detected by a monoclonal antibody. Cancer Res.. 46: 71-7s,i9a6. 14. Babur, 6. .M.. Kipnes. R. S., and Curnett. J. T. Biological defense mecha- nisms: the production by leukocytes of superoxide, a potential bactericidal agent. J. Clin. Invest.. 52: 714, 1973. IS. IMcCord, J. M. and Fridovich, 0. Superoxide dismutase. J. Biol. Chem., 244: 6049-6055,1%9. 16. Fridovich. I.Superoxide dsmutues. Adv. Enzymol., 58: 61-97.1986. 17. Brown, K., and Fridovkh, J. DNA strand scission by enzymnticallygenerated oxygen radicals. Arch. Biochem. Biophp.. 206: 414-419,1981. 18. Ueda K.. and Kobaypshi, S.,M o r i a J., and Komano. T.Sitespecif%damage c a w d by lipid peroxidation damage. Biochim. Biophys Acta. 824: 341-348, 1985. 19. R u m , P., T a n i n g k , M.. Palp M., Pisano, V.. Petemonte, P.. De Angeli, M. T., Carlone. S., Santi, L., and Parch. S. Characterization of the effects induced in mouse and hamster cells by lithocholic acid. Cancer Res.. 47: 28662874. 1987. 20. Tullius, T. D., and Dombroski, B. A. Hydroxyl radical 'footprinting': highresolution information about DNA-protein contacts and application to X repressor and Crc protein. Prw. Natl. Acad. Sci. USA. 83: 5469-5473.1986. 21. Lewis. J. G.,and Swenberg. J. A. Differential repair of O*-methyl guanine in DNA of rat hepatocytes and nonparenchymal cells. Nature (Lond.), 288: 185-187.1980. 22. McCord. J. IM.. and Day, E. D. Superoxide-dependent production of hydroxyl radical catalyzed by iron-EDTA complex. FEBS Lett., 86: 139-142, 1978. 23. Gill, D. P., and Ahmed, A. E. Covalent binding of "C-benzene to cellular organelles and bone marrow nucleic acid. Biochem. Pharmacol., 30: 1127- 1131, 1981. 4765