Document yx51JNr39BQq88nYeO9gqOED

Role of Oxidants in DNA Damage Hydroxyl Radical Mediates the Synergistic DNA Damaging Effects of Asbestos and Cigarette Smoke J. H. Jackson, I. U. Schraufstatter, P. A. Hyslop, K. Vosbeck, R. Sauerheber, S. A. Weitzman, and C. G. Cochrane Department ofImmunology, Scripps Clinic and Research Foundation, La Jolla, California 92037; Reese Stealy Research Institute, San Diego, California 92101; and Northwestern University, Chicago, Illinois 60611 Abstract The mechanism by which cigarette smoking and asbestos ex posure synergistically increase the incidence of lung cancer is unknown. We hypothesized that cigarette smoke and asbestos might synergistically increase DNA damage. To test this hy pothesis we exposed isolated bacteriophage PM2 DNA to cig arette smoke and/or asbestos, and assessed DNA strand breaks as an index of DNA damage. Our results supported our hypothesis. 7812% of the DNA exposed to both cigarette smoke and asbestos developed strand breaks, while only 9.87.0 or 4.333% of the DNA exposed to cigarette smoke or asbestos, respectively, developed strand breaks under the conditions of the experiment. Our experimental evidence suggested that cigarette smoke and asbestos synergistically increased DNA damage by stimu lating `OH formation. First, significant amounts of *OH were detected by electron paramagnetic resonance (EPR) in DNA mixtures containing both cigarette smoke and asbestos, but no *OH was detected in mixtures containing cigarette smoke alone or asbestos alone. Second, the *OH scavengers, dimethylsulfoxide (DMSO), mannitol, or Na benzoate decreased both `OH detection by EPR and strand breaks in DNA mixtures exposed to cigarette smoke and asbestos. Third, the H202 scavenger, catalase, and the iron chelators, 1,10-phenanthroline and desferrithiocin, decreased both *OH detection and strand breaks in DNA mixtures exposed to cigarette smoke and asbestos. These latter findings suggest that iron contained in asbestos may catalyze the formation of #OH from H202 generated by cigarette smoke. In summary, our study indicates that cigarette smoke and asbestos synergistically increase DNA damage and suggests that this synergism may involve OH production. Introduction Numerous studies have demonstrated that oxidants cause DNA damage, and it has been speculated that this DNA dam age could ultimately lead to carcinogenesis (1-6). Further more, since a recent study demonstrated that cigarette smoke generates oxidants (7) and causes DNA damage in cultured cells (8), it has been speculated that oxidants might be respon- Address reprint requests to Dr. Jackson, Department of Immunology, IMM12, Scripps Clinic and Research Foundation, 10666 North Torrey Pines Road, La Jolla, CA 92037. Dr. Vosbeck's current address is Ciba-Geigy Ltd., Basel, Switzerland. Receivedfor publication 20 March 1987. J. Clin. Invest. The American Society for Clinical Investigation, Inc. 0021-9738/87/10/1090/06 $2.00 Volume 80, October 1987, 1090-1095 sible for the increased incidence of lung cancer seen among cigarette smokers. Even more striking than the association be tween cigarette smoking and lung cancer, however, is the asso ciation between cigarette smoking, asbestos exposure, and lung cancer. Specifically, asbestos exposed cigarette smokers have a 50-90 times greater incidence of lung cancer (9) while asbestos exposed nonsmokers and non-asbestos-exposed ciga rette smokers have only a 5 and 10 times, respectively, greater incidence of lung cancer than non-asbestos-exposed, non smoking individuals (9, 10). The mechanism of this synergy between cigarette smoking and asbestos exposure is unknown. Previous studies have indicated, however, that cigarette smoke generates superoxide anion (02) and hydrogen peroxide (H202; 7), and that iron stimulates the production of hydroxyl radical (`OH) from 02 and H202 (11). Since `OH is the specific oxidant thought to be responsible for cigarette smoke-me diated DNA damage (8), and because asbestos contains a large amount of iron and can stimulate `OH production from H202 (12), we hypothesized that asbestos would synergistically in crease the amount of damage seen in DNA exposed to ciga rette smoke and that this increased damage might be due to stimulation of *OH formation. In order to test this hypothesis we exposed isolated DNA to cigarette smoke, asbestos, *OH scavengers, and/or iron chelators, and assessed the degree of DNA damage and *OH production in our reaction mixtures. Methods Preparation of smoke phosphate-buffered saline (PBS). Smoke PBS was prepared according to the method of Nagata et al. (7). Briefly, smoke from one commercial filter cigarette was bubbled through 6 ml of PBS, pH 7.4 for 5 min. 10 yX of this smoke-PBS was then added to reaction mixtures, as outlined below. Measurement ofDNA damage. DNA strand breaks were measured according to a modification of the method of Lown (13). Briefly, this assay detects strand breaks in closed covalently circular (CCC)1 DNA by assessing the degree of ethidium bromide fluorescence in DNA mixtures exposed to alkaline and heat denaturing conditions. When double stranded, CCC PM2 DNA develops strand breaks, it is con verted into open circular (OC) or linear DNA. OC and linear DNA become single stranded when exposed to heat and alkaline denaturing conditions, but CCC DNA (which contains no strand breaks) remains double stranded under these same conditions. Since ethidium bromide preferentially binds to double stranded DNA, one can assess the rela tive amounts of nonbroken, double stranded, CCC DNA and broken, OC or linear DNA, by measuring the degree of ethidium bromide fluorescence in reaction mixtures. Reaction mixtures for measurement of DNA damage consisted of 0.8 Mg ofPM2 bacteriophage CCC DNA (Boehringer Mannheim, Indi anapolis, IN), along with various combinations of0.05 mg of crocido- 1. Abbreviations used in this paper: CCC, closed covalently circular, DFT, desferrithiocin; DMPO, 5,5-dimethyl-1-pyroline-V-oxide; EPR, electron paramagnetic resonance; OC, open circular, PHEN, 1,10phenanthroline; SOD, superoxide dismutase. 1090 Jackson, Schraufstatter, Hyslop, Vosbeck, Sauerheber, Weitzman, and Cochrane lite asbestos (U.I.C.C., reference standard sample, kindly supplied by Dr. V. Timbrell and Dr. J. C. Wagner, Pneumoconiosis Research Unit; Medical Research Council; Penarth, UK), 10 m1 of smoke PBS, 65 mM hydrogen peroxide (Fisher Scientific Co., Inc., Fair Lawn, NJ), 30 mM FeS04 (Sigma Chemical Co., St. Louis, MO), 100 mM dimethylsulfoxide (DMSO; Sigma Chemical Co.), 100 mM mannitol (Calbiochem-Behring Corp., La Jolla, CA), 100 mM Na benzoate, 4 mM 1,10-phenanthroline (PHEN; both from Sigma), 4 mM desferrithiocin (DFT, kindly supplied by Ciba Geigy Ltd., Basel, Switzerland), 3.5 Mg of catalase, and/or 3.5 Mg of superoxide dismutase (SOD, both from Sigma) in a final volume of 100 m1 of PBS, pH 7.4. Reaction mixtures were incubated for 1 h at room temperature, and were then centrifuged at 10,000 gX 10 s to pellet the asbestos particles. (The asbestos particles interfered with our fluorescence measurements.) Supernatants (50 mO were transferred to glass tubes containing 500 m1 of fluorescence assay solution (0.5 Mg/ml ethidium bromide, 0.5 mM EDTA, 20 mM tripo tassium phosphate, pH 11.8; all from Sigma). Mixtures were heated at 96C for 4 min, placed on ice, and equilibrated to 25 C. Fluorescence was measured on a fluorescence spectrophotometer (model 650-15; Perkin-Elmer Corp., Hitachi-Perkin-Elmer Instruments, Mountain View, CA) using an excitation wavelength of 525 and an emission wavelength of 600. All fluorescence values were obtained after heating at 96C. The percentage of DNA developing strand breaks was calcu lated according to the formula: % DNA developing strand breaks = [1 - [fluorescence of injured DNA/fluorescence of noninjured DNA]] X 100. Measurement of'OH production. *OH production was detected as previously described (12), using electron paramagnetic resonance (EPR) spectroscopy and the spin trap 5,5-dimethyl-1 -pyroline-V-oxide (DMPO). DMPO was purified by charcoal filtration according to the method of Buettner et al. (14). Reaction mixtures for measurement of 'OH were identical to reaction mixtures for measurement of DNA damage except for the addition of 100 mM DMPO (Aldrich Chemical Co., Milwaukee, WI) to all reaction mixtures. In addition, 6.5 M eth anol (U.S. Industrial Chemical Co., New York) was added to some of the reaction mixtures, in order to prove that the EPR signal generated was due to *OH production. Immediately after the addition of all reaction components, mixtures were aspirated into 5-in. glass capillary tubes and inserted into an EPR spectrophotometer (E-109 X-band, Varian Associates, Palo Alto, CA) operating at 100 kHz field modula tion, 2 G modulation amplitude, 10 mW microwave power level, 3.2 X 104 gain setting, 0.064 second time constant and 50 G/min scan speed. The EPR spectra were single scanned. Statistical analysis. Results were analyzed using one-way analysis of variance with Student-Newman-Keuls multiple comparison proce dure, and two-tailed Student's t test. Results DNA damage. 9.87.0, 4.33.3, or 5.35.5% of the DNA exposed to cigarette smoke alone, asbestos particles alone, or FeS04 alone, respectively, developed strand breaks (Fig. 1). In contrast, 7812 or 8612% of the DNA exposed to both ciga rette smoke and asbestos, or cigarette smoke and FeS04, re spectively, developed strand breaks (Fig. 1). Similarly, 6.55.3% ofthe DNA exposed to H202 alone developed strand breaks while 246.3 or 258.1% of the DNA exposed to both H202 and asbestos or H202 and FeS04, respectively, devel oped strand breaks (Fig. 1). The #OH scavengers, DMSO, mannitol, or Na benzoate, the H202 scavenger, catalase, and the iron chelators, PHEN or DFT all prevented strand breaks in DNA mixtures exposed to cigarette smoke and asbestos (Fig. 2). Specifically, 7812% of the DNA exposed to cigarette smoke and asbestos developed strand breaks, while only 0, 5.56.4, 1.53, 4.87.1, 4.05.7 and 0% ofthe DNA exposed to cigarette smoke and asbestos in the presence of DMSO, mannitol, Na benzoate, catalase, PHEN, or DFT, respectively, developed strand breaks. The 02 scavenger, SOD, or heat inactivated catalase did not decrease DNA strand breaks (data not shown). The fluorescence values of uninjured, control DNA, were not significantly different in the presence or absence of DMSO, mannitol, Na benzoate, catalase, SOD, PHEN, or DFT (data not shown). 'OHproduction. DNA mixtures containing cigarette smoke and asbestos generated 'OH. Specifically, DNA mixtures con taining cigarette smoke, asbestos, and the spin trap, DMPO, generated the characteristic four-line spectrum ofthe hydroxyl radical adduct of DMPO (DMPO-OH; 15; Fig. 3 A). Similarly, DNA mixtures containing either cigarette smoke and FeS04, H202 and asbestos, or H202 and FeS04 also generated DMPO-OH (Fig. 3 B, C, and D, respectively). Mixtures con taining DNA, cigarette smoke, asbestos, FeS04, or H202 alone did not generate detectable amounts of DMPO-OH (Fig. 3 E, F, G, H, and /, respectively). When various 'OH scavengers (ethanol, DMSO, mannitol, or Na benzoate), H202 scavengers (catalase), or iron chelators (PHEN or DFT) were added to DNA mixtures containing cigarette smoke, asbestos, and DMPO, the DMPO-OH signal was either diminished or completely abolished. Specifically, Asbestos FeSCU Figure 1. Development of DNA strand breaks after exposure to var ious agents. Reaction mixtures con sisted of 0.8 Mg of PM2 DNA, 0.05 mg of crocidolite asbestos, 10 m1 of smoke PBS, 65 mM H202, and/or 30 mM FeS04 in a final volume of 100 m1 of PBS, pH 7.4. Reactions were incubated at room tempera ture for 60 min. Values are meansSD. Each value represents four separate experiments per formed in triplicate. Values for ciga rette smoke + asbestos, cigarette smoke + FeS04, H202 + asbestos, or H202 + FeS04 are significantly greater (P < 0.05) than the sums of the values of their respective com ponents. DNA Damaging Effects ofAsbestos and Cigarette Smoke Mediated by Oxidants 1091 % DNA Developing Strand Breaks + DMSO + Mannitol Na Benzoate Catalase Phenanthroline +OFT Figure 2. Inhibition of cigarette smoke and asbestos-mediated DNA strand breaks by 'OH scavengers, catalase, or iron chelators. Reaction mixtures and conditions were iden tical to those in Fig. 1, except for the addition of 100 mM DMSO, 100 mM mannitol, 100 mM Na benzoate, 3.5 Mg catalase, 4 mM 1,10-phenanthroline, or 4 mM DFT. Values are meansSD. Each value represents four separate ex periments performed in triplicate. Values for mixtures containing DMSO, mannitol, Na benzoate, cat alase, PHEN, or DFT are all signifi cantly less (P < 0.05) than values for mixtures without these additions. when ethanol was added to reaction mixtures, the DMPO-OH spectrum was no longer observed and was replaced by the characteristic spectrum of the alpha-hydroxyethyl radical ad duct of DMPO (15, Fig. 4 B). Similarly, when DMSO was added to reaction mixtures, the DMPO-OH signal was mark edly decreased and was replaced in large part by the character istic signal of the methyl radical adduct of DMPO (DMPOCH3; 16; Fig. 4 C). Mannitol slightly decreased the magnitude of the DMPO-OH signal (Fig. 4 D), Na benzoate significantly decreased the magnitude of the DMPO-OH signal, (Fig. 4 E), f g 4i Figure 3. *OH production in various DNA mixtures. Reac tion mixtures are identical to those in Fig. 1, except for the addition of 100 mM DMPO. Tracings are EPR spectra ob tained immediately after the addition of all reaction compo nents. (A) DNA + cigarette smoke + asbestos, (B) DNA + cigarette smoke + FeS04, (C) DNA + H202 + asbestos, (D) DNA + H202 + FeS04, (E) DNA alone, (F) cigarette smoke alone, (G) asbestos alone, (H) FeS04 alone, (/) H202 alone. r g tlJu J||||itLk|l1||Jdj Figure 4. Inhibition of *OH production by *OH scavengers, catalase, or iron chelators. Re action mixtures are identical to those in Fig. 2, except for the addition of 100 mM DMPO. Tracings are EPR spectra of DNA mixtures containing cig arette smoke, asbestos, and the following components: (A) no additional components, (5) ethanol (6.5 M), (C) DMSO, (D) mannitol, (E) Na ben zoate, (F) catalase, (G) PHEN, (H) DFT. (*, DMPO-OH sig nal; closed arrows, DMPOalpha-hydroxyethyl radical sig nal; and open arrows, DMPOCH3 signal.) 1092 Jackson, Schraufstatter, Hyslop, Vosbeck, Sauerheber, Weitzman, and Cochrane and catalase, PHEN, and DFT completely abolished the DMPO-OH signal (Fig. 4 F-H, respectively). SOD did not significantly decrease the magnitude of the DMPO-OH signal (data not shown). Ethanol, DMSO, mannitol, Na benzoate, catalase, SOD, PHEN, or DFT alone did not generate detect able EPR signals (data not shown). Discussion Our results indicate that asbestos particles synergistically in crease the amount ofstrand breaks in isolated DNA exposed to cigarette smoke. Several lines of evidence suggest that asbestos causes this synergy by stimulating *OH production. First, the hydroxyl radical adduct of DMPO (DMPO-OH) was detected by EPR in DNA mixtures containing cigarette smoke and asbestos but was not detected in DNA mixtures containing cigarette smoke alone or asbestos alone. Since DMPO-OH can occasionally be artifactually produced through a mechanism that is not dependent on 'OH (15), we had to verify that DMPO-OH was specifically due to 'OH in our DNA mixtures. The accepted method to perform this veri fication involves the use of the secondary *OH trap, ethanol. Ethanol reacts with *OH to produce alpha-hydroxyethyl radi cals (15), which can then form adducts with DMPO. When, therefore, DMPO-OH formation is due to the spin trapping of 'OH, ethanol addition inhibits DMPO-OH formation, and causes DMPO-alpha-hydroxyethyl radical formation (15). Since addition of 6.5 M ethanol to our DNA mixtures con taining cigarette smoke, asbestos, and DMPO prevented DMPO-OH formation and caused DMPO-alpha-hydroxyethyl radical formation, we conclude that DMPO-OH was specifically due to 'OH in our DNA mixtures. This con clusion was further supported by our studies with DMSO. Al though the purpose of adding DMSO to our reaction mixtures was to prevent DNA damage, the chemical properties of DMSO also made it useful in confirming 'OH production. Since DMSO reacts with 'OH to form *CH3, and 'CH3 can react with DMPO to form DMPO-CH3 (15, 16), the formation of DMPO-CH3 in our DNA mixtures containing DMSO, ciga rette smoke, asbestos, and DMPO was further evidence for 'OH formation. It should be noted that although we were un able to detect DMPO-OH in our DNA mixtures containing cigarette smoke alone, it is possible that small quantities of *OH were produced but were below the levels detectable by the EPR. Previous investigators (17, 18) using much larger quan tities of aqueous extracts of cigarette tar have demonstrated that metals in cigarette tar can stimulate *OH production from H202 generated by cigarette tar. Furthermore, our observation that 9.87.0% of our DNA exposed to cigarette smoke alone developed strand breaks that could be prevented by the *OH scavengers DMSO, mannitol or Na benzoate (100 mM, data not shown), also suggests that small quantities of 'OH were generated in our DNA mixtures exposed to cigarette smoke alone. Second, several different *OH scavengers inhibited the de tection of'OH and also prevented strand breaks in DNA mix tures containing cigarette smoke and asbestos. As can be seen in Fig. 4, there was a marked difference between the ability of the various *OH scavengers to inhibit the DMPO-OH signal. (DMSO > Na benzoate > mannitol). Furthermore, although all the *OH scavengers very effectively inhibited DNA strand breaks, none of them completely inhibited the DMPO-OH signal. These apparent discrepancies can be easily explained. Inhibition of DNA damage required that the added scavengers (DMSO, mannitol, Na benzoate) be able to effectively com pete with DNA for 'OH. In contrast, inhibition of the DMPOOH signal required that the added scavengers be able to effec tively compete with DMPO for 'OH. Moreover, since the ve locity of the reaction of DMPO or the various scavengers with 'OH depends on their rate constants and molar concentra tions, and since the concentration of DMPO in our reaction mixtures was the same as the concentration of the scavengers, one might expect that the ability of the scavengers to compete with DMPO for `OH would correlate with their relative rate constants for reaction with *OH. The bimolecular rate con stants for the reaction of DMSO, Na benzoate, mannitol, or DMPO with 'OH are 7 X 109, 3.3-3.8 X 109, 1 X 109, and 3.4 X 109 M_1 s_1, respectively, (15, 19-21). Be cause we found that DMSO was the most effective inhibitor of the DMPO-OH signal, whereas mannitol was the least effec tive inhibitor of the DMPO-OH signal, the ability of the var ious scavengers to inhibit the DMPO-OH signal appeared to correlate with their bimolecular rate constants for reaction with *OH. Third, the H202 scavenger, catalase, (but not heat-inacti vated catalase) decreased the amount of*OH detected and also decreased the amount of strand breaks in DNA mixtures con taining cigarette smoke and asbestos. Cigarette smoke has pre viously been demonstrated to generate 02 and H202 (7). Spe cifically, 02 and H202 can be generated from polyphenols (such as catechols, catechol derivatives, benzopyrene metabo lites, or hydroquinones) that are present in cigarette smoke (7, 8, 22, 23). Once generated, 02 and H202 can, in the presence of iron, be converted into `OH. This conversion is thought to occur via the iron-catalyzed modified Haber-Weiss reaction depicted below: Oi + Fe3+ 02 + Fe2+ H202 + Fe2+ -- 'OH + Fe3+ + OH" (11) In the presence of asbestos, the following reaction has been suggested to occur: H202 + (asbestos)-Fe2+ - *OH + (asbestos)-Fe3+ (12). It is probable, therefore, that catalase pre vented DNA strand breaks by scavenging H202, thereby pre venting its conversion into *OH. This observation is important because it suggests that although cigarette smoke contains many components that could potentially interact with asbestos (or iron) and synergistically increase DNA damage, it is likely that H202 is the component in cigarette smoke that is responsi ble for this effect. This premise is further supported by the observation that the 02 scavenger, SOD, did not decrease the amount of *OH detected or the amount of strand breaks in DNA mixtures containing cigarette smoke and asbestos, and the observation that the addition of reagent H202 to DNA mixtures containing asbestos or FeS04 also synergistically in creased *OH production and strand breaks. The ability of H202 and asbestos to damage DNA has also been reported by Kasai et al. (24). These latter findings may suggest that `OH forma tion in our system could involve Fenton chemistry rather than the modified Haber-Weiss reaction, or that agents in cigarette smoke (other than 02) could reduce Fe3+ to Fe2+. Fourth, the iron chelators DFT (25) and PHEN prevented *OH detection and strand breaks in DNA mixtures containing cigarette smoke and asbestos. In the Fenton reaction or modi- DNA Damaging Effects ofAsbestos and Cigarette Smoke Mediated by Oxidants 1093 fled Haber-Weiss reaction, iron stimulates the conversion of 02 and H202 into 'OH. Crocidolite asbestos has been shown, by neutron activation analysis, to contain 27% iron (26). This iron is not a contaminant, but rather is an integral part of the fibrous silicate lattice structure of asbestos (12). Since it is known that certain iron chelators (such as PHEN) can inhibit the ability of iron to catalyze 'OH production (27), it is possible that DFT and PHEN prevented 'OH detection and strand breaks in DNA mixtures containing cigarette smoke and as bestos by chelating the iron in the asbestos. This premise is supported by the studies of Weitzman et al. (12). Specifically, they demonstrated that asbestos could stimulate *OH produc tion in the presence of H202, and this 'OH production was inhibited by the iron chelator desferroxamine. Because the iron chelators employed in our study are not totally specific for iron, we performed additional studies in which FeS04 was substituted for asbestos in DNA mixtures containing cigarette smoke. FeS04 also synergistically increased the amount of'OH detected and the amount of strand breaks in DNA mixtures containing cigarette smoke. It appears, therefore, that although we cannot exclude the participation of other metal contami nants in our system, it is likely that the iron contained in the asbestos particles is responsible for a significant amount of our observed results. Taken in toto, our results suggest that asbestos may syner gistically increase isolated DNA strand breaks by stimulating 'OH production from oxidants generated by cigarette smoke. Several in vivo studies have demonstrated that asbestos fibers can be found within the cytoplasm and/or nucleus of viable alveolar macrophages, type 1 epithelial cells, type II epithelial cells, fibroblasts, and/or endothelial cells after as bestos inhalation (28, 29). Similarly, asbestos fibers have been found within the cytoplasm and/or nucleus of viable fibroblast or macrophages exposed to asbestos in tissue culture (30, 31). Moreover, it has been demonstrated that asbestos can adsorb benzopyrene (as well as other polycyclic aromatic hydrocar bons) onto its surface (32, 33). It appears, therefore, that in addition to being able to reach key intracellular targets, as bestos particles might also facilitate the transport of oxidant generators to these targets. Since intracellular scavengers could significantly decrease or even prevent the diffusion of extra cellularly generated oxidants to the nucleus, the potential abil ity of benzopyrene adsorbed asbestos fibers to deliver both an oxidant generator and a metal catalyst to critical intracellular targets could be important, because it would allow 'OH pro duction to occur in close proximity to cellular DNA. Finally, since iron can leach out of asbestos fibers in vitro (data not shown), it is possible that iron contained in extracellular or intracytoplasmic asbestos particles might leach out of the as bestos particles and diffuse or be transported into the nucleus where it also could stimulate 'OH production and DNA dam age. It appears, therefore, that although the aim of our study was to demonstrate that asbestos and cigarette smoke syner gistically increased in vitro DNA damage by stimulating 'OH production, a similar mechanism could exist and be relevant in vivo. In addition, since cigarette smoke or asbestos stimu lates neutrophils and macrophages to accumulate in the lung, and since neutrophils, macrophages, and tracheal epithelial cells have been demonstrated to release 02 and/or H202 in response to asbestos fibers (34, 35), it is also possible that asbestos may increase DNA damage by stimulating 'OH pro duction from cell-derived oxidants. Recent studies have suggested that DNA strand breaks (or other forms of DNA damage) caused by active oxygen species may be involved in tumor promotion and malignant transfor mation (1-6). Although it is likely that the majority of DNA damage that occurs in a human body is efficiently repaired, it is possible that some damaged DNA could occasionally either escape repair or be incorrectly repaired. Ifthese errors in repair accumulate over a period of time, it is conceivable that these errors could ultimately contribute to carcinogenesis. Ob viously, further studies will be required to elucidate the exact role of DNA damage in carcinogenesis. Nevertheless, the present study suggests that stimulation of'OH-mediated DNA damage could help explain the synergistically increased inci dence of lung cancer commonly observed in cigarette smokers exposed to significant amounts of asbestos particles. Acknowledgments We acknowledge the excellent secretarial help of Monica Bartlett. This work was supported in part by funds from U. S. Public Health Service (HL-23584, AI-17354, and HL-16411), by a fellowship from the Parker B. Francis Foundation to Dr. Jackson, and by a fellowship from the American Heart Association to Dr. Hyslop. References 1. Schraufstatter, I. U., D. B. Hinshaw, P. A. Hyslop, R. G. Spragg, and C. G. Cochrane. 1986. Oxidant injury ofcells. DNA strand breaks activate polyadenosine diphosphate-ribose polymerase and lead to de pletion of nicotinamide adenine dinucleotide. J. Clin. Invest. 77:1312-1320. 2. Weitzman, S. A., A. B. Weitberg, E. P. Clark, and T. P. Stossel. 1985. Phagocytes as carcinogens: malignant transformation produced by human neutrophils. Science (Wash. DC). 227:1231-1233. 3. Zimmerman, R., and P. Cerutti. 1984. Active oxygen acts as a promoter of transformation in mouse embryo C3H/10T1/2/C18 fi broblasts. Proc. Natl. Acad. Sci. USA. 81:2085-2087. 4. Bimboim, H. C. 1982. DNA strand breakage in human leuko cytes exposed to a tumor promoter, phorbol myristate acetate. Science (Wash. DC). 215:1247-1249. 5. Copeland, E. S. 1983. Free radicals in promotion--a chemical pathology study section workshop. Cancer Res. 43:5631-5637. 6. Cerutti, P. A. 1985. Prooxidant states and tumor promotion. Science (Wash. DC). 227(4685):375-378. 7. Nakayama, T., M- Kodama, and C. Nagata. 1984. Generation of hydrogen peroxide and superoxide anion radical from cigarette smoke. Gann. 75:95-98. 8. Nakayama, T., M. Kaneko, M. Kodama, and C. Nagata. 1985. Cigarette smoke induces DNA single-strand breaks in human cells. Nature (Lond.). 314:462-464. 9. Hammond, E. C., I. J. Selikoff, and H. Seidman. 1979. Asbestos exposure, cigarette smoking and death rates. Ann. NY Acad. Sci. 330:473-490. 10. Smoking and Health: A Report of the Surgeon General. 1979. U. S. Department of Health, Education, and Welfare, DHEW Publi cation No. (PHS) 79-50066. Government Printing Office, Washing ton, DC. 11. McCord, J. M., and E. D. Day, Jr. 1978. Superoxide-dependent production of hydroxyl radical catalyzed by the iron-EDTA complex. FEBS (Fed. Eur. Biochem. Soc.) Lett. 86:139-142. 12. Weitzman, S. A., and P. Graceffa. 1984. Asbestos catalyzes hydroxyl and superoxide radical generation from hydrogen peroxide. Arch. Biochem. Biophys. 228:373-376. 13. Lown, J. W. 1984. Ethidium binding assay for reactive oxygen species generated from reductively activated adriamycin (Doxorubi cin). Methods Enzymol. 105:532-539. 1094 Jackson, Schraufstatter, Hyslop, Vosbeck, Sauerheber, Weitzman, and Cochrane 14. Buettner, G. R., and L. W. Oberley. 1978. Considerations in the spin trapping of superoxide and hydroxyl radical in aqueous systems using 5,5-dimethyl-l-pyroline-l-oxide. Biochem. Biophys. Res. Com mun. 83:69-74. 15. Finkelstein, E., G. M. Rosen, and E. J. Rauckman. 1980. Spin trapping of superoxide and hydroxyl radical: practical aspects. Arch. Biochem. Biophys. 200:1-16. 16. Britigan, B. E., G. M. Rosen, Y. Chai, and M. S. Cohen. 1986. Do human neutrophils make hydroxyl radical. J. Biol. Chem. 261:4426-4431. 17. Cosgrove, J. P., E. T. Borish, D. F. Church, and W. A. Pryor. 1985. The metal-mediated formation of hydroxyl radical by aqueous extracts of cigarette tar. Biochem. Biophys. Res. Commun. 132:390 396. 18. Borish, E. T., J. P. Cosgrove, D. F. Church, W. A. Deutsch, and W. A. Pryor. 1985. Cigarette tar causes single-strand breaks in DNA. Biochem. Biophys. Res. Commun. 133:780-786. 19. Cederbaum, A. I., E. Dicker, E. Rubin, and G. Cohen. 1977. The effect of dimethylsulfoxide and other hydroxyl radical scavengers on the oxidation of ethanol by rat liver microsomes. Biochem. Biophys. Res. Commun. 78:1254-1261. 20. Halliwell, B. 1978. Superoxide dependent formation of hy droxyl radicals in the presence of iron chelates. FEBS (Fed. Eur. Bio chem. Soc.) Lett. 92:321-326. 21. Cederbaum, A. I., E. Dicker, and G. Cohen. 1978. Effect of hydroxyl radical scavengers on microsomal oxidation of alcohol and on assorted microsomal reactions. Biochemistry. 17:3058-3064. 22. Pryor, W. A., B. J. Hales, P. I. Premovic, and D. F. Church. 1983. The radicals in cigarette tar Their nature and suggested physio logic implications. Science (Wash. DC). 220:425-427. 23. Ts'o, P. O. P., W. J. Caspary, and R. J. Lorentzen. 1977. The involvement of free radicals in chemical carcinogenesis. In Free Radi cals in Biology. W. A. Pryor, editor. Academic Press, New York. 3:251-303. 24. Kasai, H., and S. Nishimura. 1984. DNA damage induced by asbestos in the presence of hydrogen peroxide. Gann. 75:841-844. 25. Longueville, A., and R. R. Crichton. 1986. An animal model of iron overload and its application to study hepatic ferritin iron mobili zation by chelators. Biochem. Pharmacol. 35:3669-3678. 26. Timbrell, V. 1970. Characteristics of the International Union against cancer standard reference samples of asbestos. In Pneumocon iosis: Proceedings of the International Conference, Johannesburg. H. A. Shapiro, editor. Oxford University Press, Cape Town. 28-36. 27. Mello, A. C., and R. Meneghini. 1984. In vivo formation of single strand breaks in DNA by hydrogen peroxide is mediated by the Haber-Weiss reaction. Biochim. Biophys. Acta. 781:56-63. 28. Barry, B. E., K. C. Wong, A. R. Brody, and J. D. Crapo. 1983. Reaction ofrat lungs to inhaled chrysotile asbestos following acute and subchronic exposures. Exp. Lung Res. 5:1-22. 29. Brody, A. R., L. H. Hill, and D. B. Warheit 1985. Induction of early alveolar injury by inhaled asbestos and silica. Fed. Proc. 44:2596-2601. 30. Huang, S. L. 1979. Amosite, chrysotile, and crocidolite asbestos are mutagenic in Chinese hamster lung cells. Mutation Res. 68:265 274. ' 31. Johnson, N. F., and R. Davies. 1981. An ultrastructural study ofthe effects ofasbestos fibres on cultured peritoneal macrophages. Br. J. Exp. Pathol. 62:559. 32. Lakowicz, J. R., and J. L. Hylden. 1978. Asbestos-mediated membrane uptake of benzo[a]pyrene observed by fluorescence spec troscopy. Nature (Lond.). 275:446-448. 33. Lakowicz, J. R., and D. R. Bevan. 1979. Effects of asbestos, iron oxide, silica, and carbon black on the microsomal availability of benzo[a]pyrene. Biochemistry. 18:5170-5176. 34. Mossman, B. T., and J. P. Marsh. 1985. Mechanisms of toxic injury by asbestos fibers: Role of oxygen free radicals. In NATO ASI Series. In Vitro Effects of Mineral Dusts. E. G. Beck and J. Bignon, editors. Springer-Verlag, Berlin, Heidelbeig. G3:66-74. 35. Goodglick, L. A., and A. B. Kane. 1986. Role of reactive oxy gen metabolites in crocidolite asbestos toxicity to mouse macrophages. Cancer Res. 46:5558-5566. DNA Damaging Effects ofAsbestos and Cigarette Smoke Mediated by Oxidants 1095