Document 3QjdLekg6e1K3o4y8KgEJqyJy

Volume 281, number 1.2, 9- 19 FEBS 09577 . 1991 Federation of European Biochemlcal Societles 00145793/91/$3.50 ADo,'VfS0014579391002873 Review Letter bra b DNA dama n.-d.erivec Its mechanism and measurement in mammalian systems ADril 1991 Barry Halliweil' and Okezie I. Aruoma'. Division of Pulmonary Medicine. UC Davis Medical Center. Sacramento. CA 95817. USA and 2Departnient of Biochemistry. University of London King's College. Strand Campus. London WCZR 2LS. U K Received 11 December 1990; revised version received 22 January 1991 When cells are exposed to oxidative stress, D N A damage frequently occurs. The molecular mechanismscausing this damage may include activation of nucleases and direct reaction of hydroxyl radicals with the DNA. Several oxygen-derivedspecies can attack DNA. producing distinctive patterns of chemical modification. Observation of these patterns and measurement of some of the products formed has been used to determine the role of different oxygen-derived species in D N A cleavage reactions, to assess the extent of oxidative damage to D N A in vivo and to investigate the mechanism of DNA damage by ionizing radiation and chemical carcinogens. DNA; Mutation; Oxygen radical; Hydroxyl radical; Oxidative stress I . INTRODUCTION It is well-established that aerobic organisms constantly produce small amounts of reactive oxygen species', including superoxide radical (01).hydrogen peroxide (HtOz) and hypochlorous acid (HOCl), the latter being generated by the enzyme myeloperoxidase in neutrophils [l-31. Exposure of living organisms to background levels of ionizing radiation leads to homolytic fission of oxygen-hydrogen bonds in water to produce hydroxyl radical, 'OH [4]. Hydroxyl radical can also be generated when HZOZcomes into contact with certain transition metal ion chelates, especially those of iron and copper [ 5 ] . In general, the reduced forms of these metal ions (Fe2', Cu') produce 'OH at a faster rate upon reaction with HzO2 than the oxidized forms (Fe3', Cu") and so reducing agents such as O r and ascorbic acid can often accelerate 'OH generation by metal ion/H202 mixtures [ 5 ] . However, both Cu" [6,7] and certain Fe3'-complexes (especially Fe3'-nitrilotriacetic acid) do generate some 'OH upon raction with HzO2 [8-lo]. In the case of ferric-NTA Correspondence uddress: 6 . Halli\~ell, Division of Pulmonary lledicine. UC Davis Medical Center, 4301 X Street. Sacramento, CA 95517, USA 'Reocrive osvgen species' is a collective term used in the biomedical literature that includes oxygen-containing radicals (such as O f , 'OH. ROO, R01') and non-radical species that can produce oxygencontaining radicals during their reactions (HrO:. 03, singlet Or'Ag, HOCI). The term 'reactive' is relative; for example, O r is very much less chemically reactive than is 'OH. complexes, 01 is somehow involved in the 'OH formation in the presence of H202, since it is almost completely inhibited by the OF-scavenging enzyme superoxide dismutase [9,10]. Aerobes have evolved antioxidant defences to protect themselves against the reactive oxygen species' generated in vivo. These defences include enzymes (such as superoxide dismutase, catalase and glutathione peroxidase), low molecular mass agents (examples being a-tocopherol and ascorbic acid) and proteins that bind metal ions in forms unable to accelerate free radical reactions [1,2,11-131. Oxidarive stress results when reactive oxygen species are not adequately removed. This can happen if antioxidants are depleted and/or if the formation of reactive oxygen species is increased beyond the ability of the defences to cope with them [2]. Subjecting cells to oxidative stress can result in severe metabolic dysfunctions, including peroxidation of membrane lipids, depletion of nicotinamide nucleotides, rises in intracellular free CaZ+ions, cytoskeleral disruption and DNA damage. The latter is often measured as formation of single-strand breaks, double strand breaks or chromosomal aberrations. Methods for measuring DNA strand breaks have recently been discussed [14-161. Indeed, DNA damage has been almost invariably observed in a wide range of mammalian cell types exposed to oxidative stress [16-441. The systems used to impose oxidative stress upon cells have included exposure to elevated oxygen concentrations [42], incubation with enzymes that generate reactive oxygen species (such as xanthine oxidase plus its substrates, xanthine or hypoxanthine [18,22-27]), Pitblished b! Elsevrer Science Pitblislrers 8.V . NOTICE: This material may be prateMd by mw'aht caw (Title 17. United States Code). 9 Volume 281. number 1 2 FEBS LETTERS April 1991 direct addition to the cells of Hz02 [18,28-32,43,44, 44a,b], of organic hydroperoxides [33-36,44a,4S], or of compounds whose metabolism by the cell results in increased intracellular generation of 01and H z 0 ~(such as paraquat and menadione [19,21]), and co-incubation of the cells with activated phagocytes such as macrophages [38] and neutrophils [18,39-411. Activated neutrophils and macrophages generate 01 and H202: in addition, neutrophils produce HOC1 [3,46]. Neutrophils do not produce 'OH unless a source of transition metal ions is added to the incubation mixture, i.e. neutrophils d o not themselves appear to contain any form of metal ion catalyst that will convert H2Oz into 'OH [47-491. Oxidative stress [50-541 and DNA damage [SI] also occur when some mammalian cells are exposed to tumor necrosis factor. The DNA damage produced in human cells by exposure to cigarette smoke [55,5Sa,S6], asbestos [57,S8], ozone [60,61] or to certain carcinogenic metals, such as nickel [59], has also been suggested to involve reactive oxygen species. 2. POSSIBLE MECHANISMS OF DNA DAMAGE INDUCED BY OXIDATIVE STRESS Why does oxidative stress cause DNA damage? In the case of externally-generated reactive oxygen species (e.g. when cells are incubated with HzOz, activated phagocytes or xanthine oxidase plus its substrates) damage is usually inhibited by adding catalase, showing that HzOzis needed. Superoxide dismutase (SOD) does not usually inhibit much, which could mean either that 01 is not involved in the DNA damage, or that SOD does not enter cells easily. That the latter interpretation is correct in at least one cell system is shown by the observations that SOD can protect hepatocytes from the toxicity of Hz02 or t-butylhydroperoxide under conditions where it does enter the cells [62,63]. However, neither 0 2 nor H ~ 0 u2ndergoes any chemical reaction with DNA, as measured by strand breakage [64-661 or by chemical changes in the deoxyribose, purines or pyrimidines [9,67,68]. Hence, DNA damage by oxidative stress cannot involve direct attack of 01 or of HzO2 upon the DNA. Two explanations of the DNA damage have been advanced (Fig. 1). First, it is possible that the damage is due to 'OH radical formation [30]. Thus, it is envisaged that H202, which crosses biological membranes easily [69], can penetrate to the nucleus and react with ions of iron or copper to form 'OH. Because of the high reactivity of 'OH and its resultant inability to diffuse significant distances within the cell [69], this mechanism is only feasible if the 'OH is generated from H202 by reaction with metal ions bound upon or very close to the DNA. One possibility is that these metal ions might always be present bound to the DNA in vivo. For example, copper ions are thought to be present in chromosomes 1701 and copper ions are very effective in promoting HzOz-dependent damage to isolated DNA [6,7] and to DNA within chromatin [71] in vitro. A second possibility is that the metal ions might be released within the cell as a result of oxidative stress, and then bind to the DNA [72]. Thus, just as oxidative Stress causes rises in intracellular free ca2+,it may cause rises in intracellular free iron andlor copper ions that could bind to DNA and make it a target for oxidative damage [72-741. It has recently been shown that same chelators of the carcinogenic metal nickel [59] also react with HzO2 to cause 'OH-dependent damage to isolated DNA. Mixtures of cobalt(I1) ions and H202 which are thought to produce *OH [7SI, again damaged DNA in a way characteristic of attack by 'OH (Nackerdien, Rao, Halliwell, and Dizdaroglu, in preparation. A second explanation of the ability of oxidative stress to cause DNA damage is that it triggers off a series of metabolic events within the cell (76,771 that lead to activation of nuclease enzymes, which cleave the DNA backbone. There has been much debate recently concerning the suggestion that oxidative stress causes rises in intracellular free Ca", which might fragment DNA by activating Ca'+-dependent endonucleases [21,37,78] in a mechanism resembling that of apoptosis ('programmed cell death'). An example of apoptosis is the killing of immature thymocytes by glucocorticoid hormones, which activate a self-destructive process that apparently involves Ca2+-dependent DNA fragmentation [79,80]. These two mechanisms (DNA damage by 'OH or by activation of nucleases) are not mutually exclusive, i.e. they could both take place. Indeed, there is evidence consistent with both of them. Their relative importance may depend on the cell type used and on how the OXidative stress is imposed. For example, chelating agents that bind iron ions into chelates unable to generate 'OH (such as desferrioxamine [81], desferrithiocin [18], or phenanthroline [82]) can often protect cells against DNA damage and other toxic effects of oxidative stress [30,31,34,38,73,83-85]. The effects of desferrioxamine are variable, since in general it does not cross cell membranes readily, although it appears to enter some cell types (such as hepatocytes) more readily than it enters others. Jonas et al. [29] showed that the toxicity of HzOz to epithelial cells is greatly diminished at 4C but it can be increased again by adding ascorbic acid: this effect is not seen if cells are pre-treated with desferrioxamine. Their observations are consistent with a mechanism of cell damage that depends on H2Or and reduced iron ions: at high temperatures normal metabolism may provide a reductant (such as OF), whereas at low temperatures ascorbate can replace it. Suggestive evidence that Fenton-type reactions can occur within bacterial cells has been presented [87-891, although this is not necessarily relevant to mammalian systems. 8-Hydroxyguanine (8-OH-Gua) was increased in amount in the DNA of P388 D1 cells after exposure 10 Volume 281. number 1.2 A Fenton Chemistry FEBS LETTERS Oxidative stress -----$ .OH generation upon DNA by reaction o H202 with transition metal ions already bound to DNA T ii binding of ions co DNA iron ions within the STRAND BREAKAGE DNA BASE MODIFICATION DEOXYRIBOSE FRAGMENTAT ION April 1991 -B Nuclease Activation Oxidative stress binding by endoplasmic reticulum. DNA FRAGMENTATION Inhibition of plasma membrane Ca2+-extrusion systems. Release of Ca2+ from mitochondria. T -I I II \ Rise in intracellular + Endonuclease activation Fig. 1. Hypotheses to explain DNA damage resulting from exposing cells IO oxidative stress. to HzOz [18]: 8-OH-Gua can arise from attack of `OH upon guanine (see next section) but should not be produced as a result of nuclease action. DNA isolated from P388 D 1 cells after exposing them to H202did not show a regular pattern of fragmentation, such as might be expected from nuclease attack [18]. Increases in 8-OHGua have also been observed in DNA from other cells subjected to oxidative stress [90,91]. Thymine glycol, another product that can result from attack of `OH upon DNA (see next section) has been reported to be formed in the DNA of yeast cells after exposure to high concentrations of Hz02 (921 and in murine tumor cells after exposure to tumor necrosis factor [51]. Treatment of murine hybridoma cells with HzOz caused a pattern of chemical changes in the DNA bases that is characteristic of attack by `OH (931. However, the evidence for metabolic changes produced in cells by oxidative stress is also strong [21,37,76-79). iMenadione and other quinones (which `redox cycle' within cells to give 01and H202)appear to produce DNA strand breaks in hepatocytes by Ca"-dependent activation of an endonuclease. DNA damage could be inhibited by preventing the rise in Ca*+using Ca"-chelators (37,941. Oxidative stress can also sometimes activate and/or cause changes in the subcellular location of PKC (protein kinase C) [95-971. Exposure of mouse epidermal JB6 cells to HtOz ap- peared to cause a Ca"-dependent translocation of PKC to the plasma membrane [95] whereas menadione activated PKC both in these cells and in rat hepatocytes [95,96] without producing translocation. Cantoni et ai. [98] found that the Ca'--chelator quin 2 inhibited 11 Volume 281, number 1,2 FEBS LETTERS HzOz-induced DNA strand breakage in CHO cells, although it did not inhibit iron ion-dependent 'OH generation in vitro under their reaction conditions (its effect on copper ions was not examined). Of course, even if an iron ion-quin 2 complex is capable of catalyzing 'OH formation, it could still protect by removing metal ions from the vicinity of the DNA, so that any 'OH generated no longer attacks this molecule. Nicotera et al. [45] found that desferrioxamine prevented a sustained rise in intracellular free Ca" in hepatocytes exposed to f-butylhydroperoxide. Fig. 2 shows that both types of experimental result may be accommodated by proposing that changes in the availability of calcium ions may depend upon, or give rise to, changes in the availability of iron or copper ions. Clearly, attempting to elucidate the mechanism of DNA damage by the use of free radical scavengers or metal ion chelators added to the outside of cells is unlikely to give unambiguous answers. Let us see what can be learned from the techniques of molecular biology and analytical chemistry. 3. REACTIVE OXYGEN SPECIES AS MUTAGENS AND CARCINOGENS Oxidative stress, imposed by a variety of mechanisms (including increased 0 2 concentrations [98a]), has been convincingly shown to be mutagenic to bacteria [69,99-1041. For example, E. coli mutants lacking SOD activity show greatly-enhanced rates of spontaneous mutation [99]. Similar mutagenic effects have been shown in a range of mammalian cell types [42,105-1071 subjected to oxidative stress. Moraes et al. [lOS] studied the pattern of mutations obtained in a gene of a shuttle plasmid when simian cells transfected with this plasmid were exposed to Hz02. Both single base changes and deletions were observed. The majority of base changes were at GC base pairs, the GC-+ATbase transition be- ing predominant. Treatment of the plasmid with &oz in vitro before transfection did not produce a n increased number of mutations (unless iron ions were added [109]), consistent with the inability of Hz02 to react directly with DNA, as demonstrated by chemicd studies [9,64-681. Can mutations induced by oxidative stress lead to cancer? Ionizing radiation is well-known to be both mutagenic and carcinogenic [4,110,1111. Since much of the cell damage caused by such radiation involves 'OH production by homolytic fission of the oxygenhydrogen bonds in water, then 'OH can probably be classified as a complete carcinogen. Base-pair changes and some frameshifts are the commonest mutations observed in cells exposed to ionizing radiation [110,1111. Chemical changes in the DNA bases, singleand double-strand breaks and enhanced expression of certain proto-oncogenes [4,112,1131 have also been observed. However, the precise relationship between these different events and the development of cancer is uncertain. Thus, the chemical changes in DNA might themselves somehow lead to cancer [114]. An unrepaired lesion in DNA might be by-passed in an error-prone fashion. Resynthesis of DNA after excision repair might conceivably introduce errors. There are many steps between a healthy cell and a malignant tumor. Cancer biologists have often referred to at least three stages: initiation (an irreversible change in DNA), promotion (probably involving changes in gene expression) and progression (further changes in DNA leading to the eventual production of a malignant tumor). Both Zimmerman and Cerutti [115] and Weitzman et al. [116] showed that a clone of C3H mouse fibroblasts exposed to activated human neutrophils or to hypoxanthine plus xanthine oxidase underwent malignant transformation. Nassi-Calo et al. [1171 showed that H202 also transformed these cells, an action prevented by the chelating agent o-phenanthroline. srress / ENDONUCLLGE ACTIVATION I 1intracellular I I 7 Il A \ 2A 2B rises in intracellular -$ T binding :o DNA Lree ironicopper ions -OH generation DEOXYRI30SE DEGRADATION Fig. 2. A combined hypothesis. Rises in intracellular free iron or copper ion concentrations could be a consequence of rises in Ca' (IB.ZB) or vice versa (1A. ZA). 12 Volume 281, number 1.2 FEBS LETTERS April 1991 The ability of oxidative Stress to induce transformation has also been shown in human lung fibroblasts [118], and the ability of H2Oz to cause pre-neoplastic changes in hamster tracheal explants has been described [119]. Increased expression of the proto-oncogenes c-jos and ~ - m y chas been shown in several mammalian cell types exposed to oxidative stress [120-1221. Several tumor promoters, such as phorbol myristate acetate, are powerful activators of Oi and HzO2 production by phagocytes [123]. Although most attention has been paid in the literature to the action of reactive oxygen species as promoters of carcinogenesis [115,120,121,123], their ability to damage DNA and produce alterations in gene ex- pression implies that they could be involved in all stages of carcinogenesis [124,127,129]. I t has been argued [42,130-132] that continuous damage to DNA by free radical mechanisms is a significant cause of cancer in humans, an observation that might explain results of the epidemiological investigations, which show inverse correlations in humans between plasma concentrations of certain antioxidants and the incidence of cancer [ 131,133). However, several antioxidants can alter the metabolism of procarcinogens, favoring metabolic pathways that do not result in formation of ultimate carcinogens [13I , 1341. Hence a protective effect of antioxidants does not necessarily mean that oxidative stress leads to the cancers in question. Of course, some carcinogens might act by imposing an oxidative stress on their target cells (Section 4), and reactive oxygen species have been claimed to be capable of converting some procarcinogens into ultimate carcinogens [134,135]. In addition, it must be borne in mind that a high plasma level of such antioxidants as ascorbic acid and vitamin E may simply be an index of a good diet, which protects against many diseases. DNA damage resulting from oxidative stress (or from any other mechanism) need not necessarily lead to cancer. Low levels of damage may be efficiently repaired with a minimal risk of error. High levels of oxidative stress may lead to cell death, so that initiated cells do not remain in the organism. Thus, an intermediate level of damage is most likely to predispose to malignancy, which may explain the close association of chronic inflammation (involving phagocytic production of Or and HtOz) with malignancy in such human diseases as ulcerative colitis, Crohn's disease and reflux esophagitis (reviewed in [127]). Cerutti et al. [122] showed that one difference between a clone of mouse epidermal cells that was promotable by xanthine/xan:hine oxidase and a non-promotable clone was that the latter had lower levels of SOD and catalase and was more sensitive to killing by reactive oxygen species. Thus, increased antioxidant defences, by protecting against cell death resulting from oxidative stress, might conceivably (and ironically) sometimes lead to increased cancer [136,1371. 4. CHARACTERIZATION AND GENETIC EFFECTS OF CHEMICAL CHANGES PRODUCED IN DNA BY REACTIVE OXYGEN SPECIES What types of chemical change can be produced in DNA by reactive oxygen species? Superoxide and HzO? do not react with DNA unless transition metal ions are present to allow `OH formation [9,64-681. Radiation chemists have carried out many studies of the effects of `OH, generated by ionizing radiation, upon DNA. This radical is so reactive that it can attack all components of the DNA (reviewed in [4,138-1401). Thus, 'OH abstracts hydrogen atoms from deoxyribose, giving sugar radicals that can fragment in various ways. Reactions of deoxyribose-derived radicals can lead to the release of purine and pyrimidine bases from the DNA (producing abasic sites), and to strand breaks. Some of the altered sugars that remain attached to DNA can be split to give strand breaks by incubation with alkaline solutions; these are the socalled `alkali-labile sites' [4,15,141]. Chemical changes to the purine and pyrimidine bases have also been characterized in detail (reviewed in [138-1401). Thus, `OH can add on to guanine residues at C-4, C-5, and C-8 positions. For example, addition of `OH to C-8 of guanine produces a radical adduct that has several possible fates. It can be reduced to 8-hydroxy-7,8dihydroguanine, oxidized to 8-hydroxyguanine, or undergo ring opening followed by one-electron reduction and protonation to give 2,6-diamino-4-hydroxy-Sformamidopyrimidine, usually abbreviated as FapyGua (Fig. 3). Similarly, `OH can add on to C4, C5, or C8 of adenine residues: among other fates, the C-8 `OH adduct radical can be converted into 8-hydroxyadenine by oxidation, or undergo ring-opening followed by oneelectron reduction to give 5-formamido-4,6-diaminopyrimidine (FapyAde). Fig. 4 shows the structures of some of these compounds. Pyrimidines in DNA are also attacked to give multiple products. Thus, thymine can form cis and frans thymine glycols (5,6-dihydroxy-6hydrothymines), 5-hydroxy-5-methylhydantoin,5,6-dihydrothymine and 5-hydroxymethyluracil. Cytosine can form several products, including cytosine glycol and 5,6-dihydroxycytosine (Fig. 4). When whole cells or isolated chromatin are exposed to ionizing radiation, cross-links can occur between DNA bases and amino acid residues in nuclear proteins [139,142-1481. Thus, thymine-tyrosine [145], thyminealiphatic amino acid [143-1491, and cytosine-tyrosine [146] links have been identified in isolated calf-thymus chromatin subjected to 7-irradiation. Treatment of chromatin with Fe"-chelates and H.02 also produces DNA-protein cross-links [1501 and such links have been detected in cellular DNA after exposure of the cells to ozone [60]. Molecular biologists have examined the likely 13 Volume 281, number 1,2 a FEBS LETTERS April 1991 R G80H' C Fig. 3. Some of the products that can result from attack of hydroxyl radicals upon guanine. Adapted with permission from reference [138].G80H' is the radical formed by attack of 'OH upon C8 of guanine. It can undergo ring opening followed by reduction and protonation to give FapyGua (route a), oxidation to 8-OH-Gua (route b) or reduction to give 7,8-dihydro-8-hydroxyguanine,which can be converted to FapyGua (route c). physiological effects of these various lesions in DNA. situation with other reactive oxygen species is less clear- 8-Hydroxyguanine (and, by inference, %hydroxy- cut at present. Singlet oxygen is able to produce limited adenine) might lead to mutations by inducing mis- strand breakage in isolated DNA [153,154], and its ; reading of the base itself and of the adjacent bases ability to modify the DNA bases is also limited [154a]. [110,111,151,152]. Thymine glycol might have some Thus, IM. Dizdaroglu and H. Sies (personal com- mutagenic action and it can be lethal if not removed munication) found small amounts of 8-OH-Gua and from the DNA before replication [110,1111, Ring- FapyGua but no other significant base changes in DNA fragmented bases are thought to block DNA replication exposed to singlet 0 2 generated by the thermal decom- [110,1111. Abasic sites, which can result from direct at- position of an endoperoxide. Exposure to illuminated tack of 'OH, can also be mutagenic in vivo [110,111]. Methylene blue causes formation of 8-hydroxyguanine It is clear that 'OH produces multiple changes in [155] and of some strand breaks [156] in DNA but the DNA whereas 01 and HzOt have no effect, but the HN Nq!A;species responsible was not identified, except for the HN~H 0 oAN H H oAN H H 5-hydroxy-5-methylhydantoin 5-hydroxyhydantoin cytosine glycol 5-hydroquracil 5-hydroxymethyluracil 5 - hydroxycytosine thymine glyco1 5.6-dihydroxyuracil 5.6-dihydroxycytosine 4.6-diamino-5 formamidopyrimidine 8-hydroxyadenine 2.6-diamfno-4-hydroxy5- Eo-idopyrinidine 8-hydroxyguanine Fig. 4. Some of the end-products that result from attack of hydroxyl radicals upon the bases of DNA. 14 volume 281, number 1.2 FEBS LETTERS April 1991 observation that scavengers of 'OH did not protect. However,.the excited state of the photosensitizing dye Rose bengal has itself been claimed to cleave DNA 11571 and, i f 'OH were generated by Methylene blue bound to DNA, then 'OH scavengers would not be expected (0 protect. Illuminated riboflavin, which generates single1 0 2 , was reported not to produce 8-hydroxyouanine in DNA [158]. Thus, singlet 0 2 certainly does not induce the extensive pattern of DNA base modification produced by 'OH. Peroxidizing lipids have been reported to damage DNA [159-163a] but peroxidizing lipids produce a range of reactive oxygen species including 'OH, H202, singlet oxygen, peroxyl radicals, and alkoxy1 radicals [69,164] and the exact contributions of these species to the DNA damage observed need to be determined [ 16j,1661. Lipid peroxides also decompose to give a huge range of products 1166-1681 including carbonyl compounds, such as malondialdehyde and the unsaturated aldehyde 4-hydroxy-2-trans nonenal [1681 which has been shown to be mutagenic to mammalian cells [1691. If these aldehydes are generated in the vicinity of DNA, they may be able to combine with it to form distinctive products [165,1691. Thus, malondialdehyde reacts with adenine, cytosine, and guanine [161,17Oa] and a guanine-MDA adduct has been identified in human urine [171]. The product of reaction of hydroxynonenal with deoxyguanosine has also been characterizcd [172]. Humans are constantly exposed to background levels of ionizing radiation, which will generate some 'OH in vivo. This radical may also arise by reaction of metal ions with H202 in vivo [5]. Thus, it is not surprising to find that repair systems have evolved to remove at least some of the lesions in DNA that can result from attack of 'OH and other reactive oxygen species (reviewed in 1173,1741). Single-strand breaks are usually quickly repaired; indeed, they are generated as intermediates in the repair of other lesions (see below). 8-Hydroxyguanine is slowly removed from cellular DNA, but the repair mechanism is unknown [173]. Several lesions, including thymine glycol, are probably removed in human cells by action of a DNA glycosylase, which cuts the base-deoxyribose bond to give an abasic site. This site is recognized by an endonuclease activity (on the same enzyme), which nicks the strand at the abasic site. The damaged part of the strand is removed, followed by resynthesis of the DNA and re-joining of the strand by a DNA ligase enzyme. Glycosylases that recognize hydroxymethyluracil and ring-opened purines in DNA have also been described in mammalian cells [ 110,111,173,174]. Modified DNA bases and nucleosides (base-deoxyribose) have been detected in the urine of humans and other mammals. Thus, 8-hydroxyadenine, 7-methyl-8 -hydroxyguanine, thymine glycol, thymidine glycol, hydroxymethyluracil, 8-hydroxyguanine and 8-hydro- xydeoxyguanasine have been detected in mammalian urine [132,175-1781. The presence of these products in urine suggests that oxidative damage to the DNA bases does occur in vivo and that repair systems are active to cleave modified bases from DNA. However, it is possible that some excreted bases originate from the diet or from the metabolism of the gut flora, and that DNA released from dead and dying cells within an organism undergoes rapid oxidative damage (since cell disruption can increase free radical reactions [69,72,164]). Hence, one must be cautious in using the amounts of modified DNA bases excreted from the body as an index of the extent of repair of oxidative DNA damage in healthy cells. 5. ,MEASUREMENT OF BASE-DERIVED PRODUCTS AS A PROBE FOR THE MECHANISM AND EXTENT OF DNA DAMAGE The development of an HPLC technique, coupled with highly-sensitive electrochemical detection, for the measurement of 8-hydroxy-deoxyguanosine has led to a series of pioneering studies in which measurement of this product has been used to gain information about free radical damage to DNA in intact celIs and whole organisms (91,114,179-1811. Antibody techniques fur the measurement of such products as 8-hydroxyguanine and thymine glycol have also long been available (e.g. see [182-1841). Ames et al. 1175-1771 have used the urinary excretion of products derived from guanine and thymine as an index of free radical damage to DNA in vivo, and have attempted to draw conclusions about changes in the rate of such damage as a function of age in mammals. The amount of 8-hydroxyguanine in the DNA from certain sub-populations of rat liver mitochondria was found to be considerably higher than that in nuclear DNA, leading to proposals about the role of mitochondria in aging and carcinogenesis [185,186]. Exposure of some cells to oxidative stress has been reported to lead to formation of 8-hydroxyguanine in the DNA [22,90,91,114]. For example, treatment of Ehrlich ascites cells with the carcinogen 4-nitroquinoline 1-oxide led to an increased content of 8-hydroxyguanine in DNA [1871. Intra-peritoneal injection of ferric-nitrilotriacetic acid (which reacts with HzOz to give 'OH [9,lOJ)into rats produced a significant rise in the 8-hydroxyguanine content of kidney DNA [1881. Carcinogenic peroxisome proliferators [189,190], acetoxime [191], 2-nitropropane [191] and, in one study, a choline-deficient diet [191a], have all been reported to result in increased amounts of 8-hydroxyguanine in DNA in vivo in mammals. These studies have certainly produced qualitative evidence for oxidative damage to DNA in vivo, although care must be used in interpreting the data [191b]. For example, nitroquinolines have been sug- IS - P Volume 2S1. number 1,2 FEBS LETTERS April 1991 gested to form 8-hydroxyguanine in DNA by a Table I mechanism that involves direct reaction of the ultimate How reaction conditions can alter the end products derived from at. carcinogen with DNA, rather than by oxygen radical generation [192]. One must also be extremely cautious tack of hydroxyl radicals generates by different systems upon punne bases in the DNA of isolated chromatin in attempting to use measurement of any one product as a quantirarive measure of DNA base damage by reactive oxygen species. When `OH attacks DNA bases, radicals are formed that can react in various ways depending on the conditions used (Fig. 3 shows an example). Thus, attack of 'OH upon guanine can lead to Systems used to generate 'OH H~Ormeralions' Air sarurared solurions: Ratios of 8-OH-Ade to 8-OH-Gua to FapyAde FapyCua formation of 8-hydroxyguanine by oxidation of the C-8 `OH adduct radical, but this radical can lead to other HZOz/Fe"/ascorbate H20z/Fe3'-EDTA 1 . 1 5.9 4.2 8.6 products as well, depending on the reaction conditions. Thus, variable amounts of 8-hydroxyguanine can result from attack of the same amount of `OH upon guanine in DNA, and so changes in 8-hydroxyguanine levels do H202/Fe3'-EDTA/ascorbate H20z/Fe"-NTA H~O~/Fe"-NTA/ascorbare HzOz/Cu" HzO~/Cu"/ascorbate 0.5 I .5 1 .? 18.6 11.1 2.2 8.3 5.3 48.1 31.5 not necessarily mean changes in the amount of free radical attack upon DNA. To take some examples, Ionizing radiarion: solutions saturated with . iron-ion dependent systems generating `OH led to Argon ...<I . ,. substantial formation of FapyGua as well as Air o s 5 0.57 1.8 3.5 ... . ...,,,..... I. ., .. i.-...I!..--r. . .r. * ~ I .', 8-hydroxyguanine in DNA [67], whereas systems containing copper ions and HzO2 greatly favored 8-hydroxyguanine production [6,71,193]. When isolated, mammalian chromatin was irradiated in aqueous suspension, the relative amounts of 8-hydro- Nitrous oxide Nitrous oxide and oxygen 0.8 3.4 0.75 4.5 Results were obtained with mammalian chromatin in aqueous suspension. Calculations by courtesy of Dr. .M. Dizdaroglu. Data selected from [71.194]. ;:,e xypurines and formamidopyrimidines generated 5 depended upon the radical environment provided by the gases used to saturate the aqueous solution [194]. For nucleases should leave the purines and pyrimidines unaltered (Figs. 1 and 2). For studies on DNA example, the presence of oxygen favoured the forma- modification, extraction of chromatin from cells for tion of 8-hydroxypurines [194,196]. Table I summarizes analysis is preferable to extraction of DNA, since it some of the results obtained. Products derived from minimizes the loss of extensively-fragmented DNA, and pyrimidines can similarly be affected by changes in of DNA that has become covalently cross-linked to pro- reaction conditions [195,196]. tein. A complete characterization of damage to DNA by GC/MS-SIM has been used to characterize the reactive oxygen species can be achieved by the techni- damage done to DNA by various reactive oxygen que of gas chromatography/mass spectrometry (review- species. Hydroxyl radical appears to produce a ed in [194a.197,198]), which may be applied to DNA uniquely-extensive pattern of base modifications itself or to DNA-protein complexes such as chromatin. (multiple products from all four bases), and this pattern The DNA or chromatin are hydrolyzed and the pro- seems to be a `fingerprint' for `OH, i.e., it can be used ducts converted to volatile derivatives, which are to identify `OH as a damaging species [6,9,68, separated by gas chromatography and identified by 71,139,143,146,152,193,194,196,198,201]. For exam- mass spectrometry. High sensitivity of detection can be ple, measurement of damage to the DNA bases by achieved by operating the mass spectrometer in the GC/MS-SIM has been used to show that the strand selected ion monitoring (SIM) mode. In this mode, the cleavage produced in isolated DNA by treatment with a mass spectrometer is set to monitor several ions derived copper-ion phenanthroline chelate probably involves by fragmentation of a particular product during the `OH [193], whereas DNA cleavage by a bleomycin-iron time at which this product is expected to emerge from ion complex is not mediated by `OH [1991. G U M S has the GC column. The G U M S - S I M technique is being also been used to identify 8-OHGua, FApyAde, used in the authors' laboratories to examine the S-OHAde and FApyGua in neoplastic tissues [200], to mechanism by which DNA is damaged in cells subjected show that the damage done to the bases in isolated to oxidative stress. Thus, if damage is due to `OH DNA by activated human neutrophils is most likely due generation, then products characteristic of 'OH attack to `OH generated by reactions involving metal ions in should be detected (Fig. I ) , as has been observed in the reaction mixture [201], to measure adducts of car- murine hybridoma cells treated with HzOz [93] and in cinogens with proteins in vivo in attempts to assess car- primate tracheal epithelial cells exposed to ozone cinogen exposure [202] and to characterize the changes (Aruoma, Halliwell and Wu, in preparation). By con- produced in plasmid DNA by treating it with potassium trast. cleavage of the DNA backbone by the action of permanganate [203]. The authors believe that such 16 volume 281, number 1,2 FEBS LETTERS April 1991 *fingerprinting'of DNA damage is one way forward in investigating the role of reactive oxygen species in damaging DNA in vivo, and assessing the contribution [hat such' species make to the mechanism of action of carcinogens and to the increased cumulative risk of cancer with age [132]. 4ckno wledgemenrs: The authors are grateful to the .Medical Rssearch Council, Arthritis and Rheumatism Council and Association for International Cancer Research for research support. Drs. Lars Breimer and Miral Dizdaroglu are particularly thanked for their helpful comments upon the manuscript. We thank Dr. ,Viral Dizdaroglu and Prof. Helmur Sies for providing unpublished informanun. Dr. Aruoma was a visiting scientut (1988-89) at the National In,[itUte of Standards and Technology, Gaithersburg, MD, USA. REFERENCES [ I ] Fridovich, I.(1989) Annu. Rev. Pharm. Tox. 23, 239-257. [2] Sies, H., ed. (1991) Oxidative Stress: Oxidants and Antioxidants, Academic Press, New York. [31 Weiss, S.J. (1989) New Engl. J. Med. 320, 365-376. (41 Von Sonntag, C. (1987) The Chemical Basis of Radiarion Biology. Taylor and Francis, London. [j] Halliwell. B. and Gutteridge, J.M.C. (1990) Methods Enzymol. 86, 1-85. (61 Aruoma, 0.1.. Halliwell, B., Gajewski, E. and Dizdaroglu, M. (1991) Biochem. J.. 273, 601-604. [7] Sagripanti, J.L. and Kramer, K.H. (1989) J. Biol. Chem. 264, 1729-1734. [SI Inoue, S. and Kawanishi, S. (1987) Cancer Res. 47. 6522-6527. [9] Aruoma, 0.1..Halliwell, B., Gajewski, E. and Dizdaroglu. M. (1989) J. Bioi. Chem. 264, 20509-20512. [IO] Gutteridge, J.M.C. (1990) Free Radical Res. Commun. 9, 119-125. [l I] Halliwell, B. and Gutteridge, J.1M.C. (1990) Arch. Biochem. Biophys. 280, 1-8. 1121 Wayner. D.D.M.. Burton, G.W. and Ingold, K.U. (1986) Biochim. Biophys. Acta 884. 119-123. (131 Frei, B., England, L. and Ames, B.N. (1989) Proc. Natl. Acad. Sci. USA 86, 6377-6381. (141 Birnboim, H.C. (1990) Methods Enzymol. 186, 550-555. [IS] Ahnstrom, G. (1988) Int. J. Radiat. Bioi. 54, 695-707. (161 Bhusate, L.L., Herbert, K.E. and Perrett, D. (1990) Biochem. Soc. Trans. 18, 676-677. (171 Schraufstatter, I.U., Hinshaw, D.B.. Hyslop, P A . , Spragg, R.G. and Cochrane. C.G. (1986) J. Clin. Invest. 77, 13 12-1320. (181 Schraufstatter, 1.U.. Hyslop. P.A., Jackson. J.H. and Cochrane, C.G. (1988) J. Clin. Invest. 82. 1040-1050. (191 Eklow-Lasrbom, L.. Rossi, L., Thor, H. and Orrenius. S. (1986) Free Radical Res. Commun. 2, 57-68. (201 Starke. P.E.. Hock. J.B. and Farber, I.L. (1986) J. Biol. Chem. 261, 3006-3012. [21) Orrenius, S., McConkey, D.J., Bellomo, G. and Nicorera, P. (1989) Trends Pharm. Sci. IO, 281-285. (221 Weirberg, A.B. (1989) Mutat. Res. 216, 197-201. [23] Paul, L.A.. Fulton. A.M. and Heppner, C.H. (1989) Mutat. Res. 215, 223-234. 1241 Phillips, B.J., James, T.E.B. and Anderson, D. (1984) Murat. Res. 126, 265-271. [ Z j ] Iwata, K.. Shibuya, H.. Ohkawa, Y. and Inui. N. (1984) Toxicol. Lett. 22, 75-81. [26] Hall, A.H. Jr.. Eanes. R.Z., Waymack, P.P., Jr. and Patterson, R.M.(1988) Mutat. Res. 198, 161-168. [27] Carson, D.A.. Seto, S. and Wasson, D.B.(1986) J. Exp. .Med. 163, 746-751. [18] Olson, M.J. (1988) J . Tox. Environ. Hlth. 23. 107-423. [29] Jonas, S.K.. Riley. P A . and Willsun, R.L. (1989) Biochem. J . 264, 651-655. Mello Filho. A.C.. Hoffrnann. M.E. and Meneghini, R. (1984) Biochem. J. 218, 273-275. Kleiman, N.J., Wang. R.R. and Spector. A. (1990) Mutar. Res. 240, 35-45. Link. E.M. and Riley, P..4. (1988) Biochem. J. 249. 391-399. Ochi. T. (1989) Murat. Res. 212. 243-248. Coleman, J.B., Gilfor. D. and Farber. J.L. (1986) hlol. Pharmacol. 50, 261-262. Nakayama, T.. Kaneko. M.and Kodama. IM. (1986) Aric. Bioi. Chem. 50. 261-262. Norris. K.H. and Hornsby, P.J. (1988) in: Oxygen Radicals in Biology and Medicine [Simic, 1M.G.. Taylor, K.A., Ward, J.F. and von Sonntag. C.. eds.) Plenum, New York. pp. 461-466. Dypbukt, J.M.. Thor, H. and Nicotera. P. (1990) Free Radical Res. Commun. 8, 346-354. Chong. Y .C., Heppner. G.H., Paul, L A . and Fulton, A,%[. (1989) Cancer Res. 49, 6652-6657. Weitberg. A.B., ~VeitmanS. .A., Destrempes. M., Latt. S.A. and Stossel, T.P. (1983) New Engl. J. Med. 308, 26-29. Weitberg, AB.. Weitzman, S.A., Clark. E.P. and Stossel, T.P. (1982) J. Clin. Invesr. 75. 1835-1841. Shacrer, E.. Beecham, E.J., Covey, J.M., Kohn, K.W. and Potter, M. (1988) Carcinogenesis 9, 2297-2304. Cille, J.J.P., Yullaan. E., Vijg, J., L e p a . A.L.. Arwert. F. and Joenje. M. (1989) Murat. Res. 219, 17-28. Canroni, 0..Cattabeni, E.. Stocchi, V., lleyn. R.E..Cerurti, P. and Murray, D. (1989) Biochim. Biophys. Acta 1014. 1-7. Prise, K.IM..Davies. S. and Michael. B.D.(1989) Int. J. Rad. Bioi. 55, 583-592. Sansrrom, B.E. and .Marklund, S.L. (1990) Biochem. J. 271, 17-23. Spragg. R.G. (1991) Am. J. Respir. Cell. Mol. Biol. 4, 4-10. Nicorera, P.. McConkey, D., Svensson, S.A.. Bellomo, G. and Orrenius. S. (1988) Toxicology 52, 55-63. Curnutte, J.T. and Babior. B.M. (1987) Adv. Human Genet. 6. 229-297. Thomas. P.J.. Shirley, P.S., Hedrick. C.C. and De Chareler, L.R. (1986) Biochemistry 25, 8042-8048. Britigan, B.E.. Coffman. T.J. and Buetmer, G.R. (1990) J. Bioi. Chem. 263. 2650-2656. Kaur. H., Fagerheim, 1.. Grootveld, 41.. Puppo. A. and Halliwell, B. (1988) Anal. Biochem. 172. 360-367. Larrick, J.W. and Wright. S.C. (1990) FASEB J. 4. 3215-3223. Zimmerman, R.J.. Chan, A. and Leadon, S.A. (1987)Cancer Res. 49, 1644-1648. Wong. G. and Goeddel, D. (1988) Science 242, 941-944. Shaffer, J.B., Treanor, C.P. and Del Vecchio. P.J. (1990) Free Rad. Biol. .Wed. 8, 497-502. Hauser. G.J.. McIntosh. J.K., Travis, 1V.D. and Rosenbery, S.A. (1990) Cancer Res. 50. 3503-3508. Borish, E.T., Cosgrove, J.P., Church, D.F.. Deutsch. W.A. and Pryor. W..k (1985) Biochem. Biophys. Res. Commun. 133, 780-786. Kiyosawa. H.. Suko. M.. Okudaira. H.. Murara. K.. LMiyamoto. T.. Chung, .M.H.. Kasai. H. and Nishimura. S. (1990) Free Radical Res. Commun. 11. 23-27. Nakayama, T.. Kaneko. M., Kodama. 41. and Nagata, C. (1985) Nature 314, 462-44. Weitzman, 5.4. and Graceffa. P. (1984) Arch. Biochem. Biophys. 228. 373-376. Jackson, J.H.. Schraufstatter, I.U.. Hyslop. P.A.. Vosbeck. K.. Sauerheber. R.. Weitzman. S A . and Cochrane. C.G. (1987) J . Clin. Invest. 80, 1090-1095. Kawanishi. S., lnoue. S. and Yamamoto. K. (1989) Carcinogenesis 10. 223 1-2235. Van der Zee. J., Van Beck. E., Dubbslman. T..M..X.R. and Van Steveninck, J . (1987) Biochem. J . 147. 69-72. Borek, C.. Zaider. M..Ong,A., Mason. H. and \vitz. G . (1986) Carcinogenesis 7 . 1611-1613. 17 Volume 281, number 1,2 FEBS LETTERS April 1991 (621 Kyle, M.E.. Nakae, D.. Sakaida, I., Miccadei, S. and Farber, J.L. (1988) J . Biol. Chem. 263. 3784-3789. [63l Nakae. D., Yoshiji, H., Amanuma, T.. Kinugasa, T., Farber, J.L. and Konishi. Y. (1990) Arch. Biochem. Biophys. 279, 3 15-319. 1641 Brawn, M.K. and Fridovich, I. (1981) Arch. Biochem. Biophys. 206, 414-419. 1651 Lesko, S.A., Lorentzen. R.J. and Tso, P.O.P. (1980) Biochemistry 19. 3023-3028. [661 Rowley, D.A. and Halliwell. B. (1983) Biochim. Biophys. Acta 761, 86-93. [a71 Aruoma, 0.1.. Halliwell. B. and Dizdaroglu, M. (1989) J. Biol. Chem. 264, 13024-13028. (681 Blakely. W.F., Fuciarelli, A.F., Wegher, B.J. and Dizdaroglu, M. (1990) Radiat. Res. 121. 338-343. I691 Halliwell, B. and Gutteridge. J.M.C. (1989) Free Radicals in Biology and ,Medicine, second edition. Clarendon Press, Oxford. [701 Prutz. W.A.. Butler, J. and Land, E.J. (1990) Int. J. Rad. Biol. 58. 215-234. [711 Dizdaroglu, M.,Rao, G., Halliwell, B. and Gajewski, E. (1991) Arch. Biochem. Biophys. 285, 317-324. [72] Halliwell, B. (1987) FASEB J . 1, 358-364. [73] Sakaida, I., Kyle, M.E. and Farber, J.L. (1990) Mol. Pharmacol. 37, 435-442. [741 Starke, P.E., Gilbertson. J.D. and Farber. J.L. (1985) Biochem. Biophys. Res. Commun. 133, 371-379. [75] Moorhouse, C.P.. Halliwell. B., Grootveld, M. and Gutteridge, J.,M.C. (1985) Biochim. Biophys. Acta 843, 261-268. (761 Birnboim, H.C. and Kanabus-Kaminska, M. (1985) Proc. Natl. Acad. Sci. USA 82, 6820-6824. (771 Birnboim, H.C. (1988) Biochem. Cell Biol. 66, 374-381. [781 Farber, J.L.(1990) Chem. R a . Tox. 3. 503-508. [78al Orrenius, S.. McConkey. D.J., Jones. D.P. and Nicotera. P. (1988) IS1 Atlas Sci. Pharmacol. 1, 319-324. (791 McConkey, D.J.. Hartzell, P., Nicotera, P. and Orrenius, S. (1989) FASEB 1. 3, 1843-1849. [801 Arends, M.J.. Morris, R.G. and Wyllie, A.H. (1990) Am. J. Path. 136, 593-608. [81] Halliwell, B. (1989) Free Rad. Biol. Med. 7, 645-651. [82l Halliwell. B. (1978) FEBS Lett. 96,238-242. [83] Meneghini, R. (1988) Mutat. Res. 195. 215-230. [841 Kyle, M.E., Nakae, D., Sakaida, I., Serroni, A. and Farber. J.L. (1989) Biochem. Pharmacol. 38, 3797-3805. [85] Starke, P.E. and Farber, J.L. (1985) J. Biol. Chem. 260. 10099-10104. [86] Cannon, D.E., Varani, J., Phan, S.H.. Ward, J.H., Kaplan, J., Till, G.O.. Simon, R.H.. Ryan, U.S.and Ward, P.A. (1987) Lab. Invest. 57, 37-44. [871 Imlay, J.A. and Linn, S. (1988) Science 240, 1302-1309. (881 Brandi, G., Cattabeni. F., Albano. A. and Canroni, 0. (1989) Free Radical Res. Comrnun. 6, 47-55. [89l Hassett, D.J., Bean, K., Biswas. G. and Cohen, M.S. (1989) Free Radical Res. Cornrnun. 7, 83-87. [901 Floyd, R.A.. Watson, J.J., Harris, J.. West, .M. and Wong, P.K. (1986) Biochem. Biophys. Res. Commun. 137. 841-846. [911 Kasai, H., Crain, P.F., Kuchino, Y., Nishimura, S., Oorsuyama, A. and Tanooka, H. (1986) Carcinogenesis 7, 1849-185 1. (921 Kaneko, M., Leadon, S.A. and Ito. T. (1988) Mutat. Res. 207, 17-22. [93] Dizdaroglu, M.. Nackerdien, Z., Chao. B.C.. Gajewski, E. and Rao, G. (1991) Arch. Biochem. Biophys, 285, 388-390. [94] McConkey, D.J., Hartzell, P.. Nicotera, P.. Wyllie, A.H. and Orrenius, S. (1988) Toxicol. Lett. 42, 123-130. [95) Larsson. R. andcerutti, P. (1989) Cancer Res. 49,5627-5632. [96l Kass, G.E.N., Duddy. S.K. and Orrenius. S. (1989) Biochem. J . 260. 499-507. (971 Gopalakrishna, R. and Anderson, W.B. (1989) Proc. Natl. Acad. Sci. USA 86, 6758-6762. 18 (981 Cantoni. O., Sesrili. P.. Cattabeni. F.. Bellomo, G.. Pou, s Cohen, M. and Cerutti, P. (1989) Eur. J. Biochem. 18;: 209-212. [98a] Joenje, H. (1989) Murat. Res. 219, 193-208. (991 Touati, D. (1989) Free Radical Res. Commun. 8, 1-9. [IO01 Moody, C.S. and Hassan. H.M. (1982) Proc. Natl. Acad. sei, USA 79, 2855-2859. [loll Storz. G., Christman, M.F., Sies, H. and Ames, B.N. (1983 Proc. Natl. Acad. Sci. USA 84, 8917-8921. [IO21 Levin. D.E.. Hollstein, M.,Christman. M.F., Schwiers, E.A. and Ames, B.N. (1982) Proc. Natl. Acad. Sci. USA 79, 7445-7449. (1031 Barak, .M., Ulitzur, S. and Merzbach. D. (1983) Mutat. Res, 121, 7-16. [IO41 Abril, N. and Pueyo, C. (1990) Environ. Molec. Mutagen. 15. 186189. [IO51 Weirzman. S.A. and Stossel, T.P. (1981) Science 212, 546-548. [I061 Weitzman. S.A. and Stossel, T.P. (1984) Cancer Lett. 22, 337-342. [IO71 Hsie. A.W., Recio. L., Katz, D.S., Lee, C.Q., Wagner, .M. and Schenley, R.L. (1986) Proc. Natl. Acad. Sci. USA 83, 9616-9620. [I081 Moraes, E L . , Keyse, S.M. and Tyrrell, R.M. (19%) Carcinogenesis 11, 283-293. [IO91 Moraes, E.C.. Keyse. S.M.. Pidoux, M. and Tyrrell. R.M. (1989) Nucleic Acids Res. 17, 8301-8312. [I101 Breimer, L.H. (1988) Br. J. Cancer 57. 6-18. (11I] Breimer, L.H. (1990) mol. Carcinogenesis 3, 188-197. [112] Ziegler-Skylakakis, K. and Andrea, U. (1987) Mutat. Res. 192, 65-67. [l 131 Yarnashina, K., iMiller, B. and Heppner, G.H. (1986) Cancer Res. 46, 2396-2400. [114] Floyd. R.A. (1990) Carcinogenesis 11, 1447-1450. [ I 151 Zimmerman. R. and Cerutti, P. (1984) Proc. Natl. Acad. Sci. USA 81, 2085-2087. (1161 Weirzman. S.A. Weitberg. A.B., Clark, E.P. and Stossel, T.P. (1985) Science 227, 1231-1233. [117] Nassi-Calo, L., iMello-Filho, A.C. and Meneghini, R. (1989) Carcinogenesis IO, 1055-1057. 11181 Weitberg, A.B. and Corvese, D. (1990) Biochem. Biophys. Res. Commun. 169, 70-74. [ I 191 Radosevich. C.A. and Weitzman, S.A. (1989) Carcinogenesis IO. 1943-1946. [I201 Sherman, M.L., Datta, R., Hallahan. D.E., Weichselbaum, R.R. and Kufe. D.W. (1990). Proc. Natl. Acad. Sci. USA 87, 5663-5666. [121] Shibanuma, M., Kuroki, T . and Nose, K. (1988)Oncogene 3. 17-21. [I221 Cerutti, P., Larsson, R., Krupitza, G.. Muehlmatter. D.. Crawford, D. and Amstad, P. (1989) Mutat. Res. 214, 81-88. [123] Ccrutti, P. (1985) Science 227, 375-381. [I241 Perchelert, J.P.and Perchellet, E.M. (1989) Free Rad. Biol. Med. 7. 377408. [125] Kensler, T.W. and Taffe. B.G. (1986) Adv. Free Rad. Biol. iMed. 2, 347-387. [I261 Goldstein. B.D.. Czerniecki, B. and Witz. G. (1989) EnvironHealth Perspect. 81. 55-57. [I271 Weitzman, S.A. and Gordon, L.I. (1990) Blood 76, 655-663. [I281 Halliwell. 8. and Gurteridge. J.M.C. (1989) Free Radicals in Biology and Medicine, Second Edition. Clarendon Press, Oxford, Chapter 8. [I291 Stich, H.F. and Anders, F. (19S9) Murat. Res. 214, 47-61. [I301 Totter. J.R. (1980) Proc. Natl. Acad. Sci. USA 77. 1763-1767. [I311 Ames. B.N. (1983) Science 221, 1256-1264. [I321 Ames. B.N. (1989) Free Radical Res. Commun. 7. 121-128. [I331 Gey. K.F., Stahelin, H.B. and Brubacher, G.B. (1988) in: Medical. Biochemical and Chemical Aspects of Free Radicals. In Proceedings of the 4th Biennial General Meeting of the Society for Free Radical Research (Hayaishi. 0.. Niki. E., Kondo. 41. and Yoshikawa, T. eds.) Elsevier, Amsterdam. pp. 377-384. Volume 281, number 1,2 FEBS LETTERS April 1991 [I341 Gower, J.D. (1988) Free Rad. Biol. Med. 5 , 95-111. (1351 Reed, G.A. (1987) Chem. Phys. Lipids W, 127-148. [I361 Okamoto, H. (1985) Bioessays 2. 15-22. 11371 Boothman. D.A. and Pardee. A.B. (1989) Proc. Natl. .&cad. Sci. USA 96. 4963-4967. [I381 Steenken, S. (1989) Chem. Rev. 89, 503-520. [139] Dizdaroglu. .M. (1986) Biotechniques 4, 536-546. [I401 Teoule. R. (1987) Int. J. Radiat. Biol. 51. 573-589. [I411 Beesk, F., Dizdaroglu, M., Schulte-Frohlinde, D. and Von Sonntag. C. (1979) Int. J. Rad. Biol. 36. 565-576. 11421 Mee. L.K. and Adelstein. S.J. (1981) Proc. Natl. Acad. Sci. USA 78, 2194-2198. (1431 Gajewski, E.. Fuciarelli, A.F. and Dizdaroglu. M. (1988) Int. J. Radiat. Biol. 54. 445-459. [I441 Dizdaroglu. AM.and Gajewski, E. (1989) Cancer Res. 49, 3463-3467. [I451 Dizdaroglu. M.,Gajewski. E., Reddy, P. and Marpolis, S.A. (1989) Biochemistry 28, 3625-3628. [I461 Gajewski, E. and Dizdaroglu, M. (1990) Biochemistry 29. 977-980. [I471 Cress, A.E. and Bowden, G.T. (1983) Radiat. Res. 95. 6 10-6I S . (1481 Oleinick, Y.L.. Chiu, S., Ramakrishman, N. and Xue. L. (1987) Br. J. Cancer 55 (suppl. VIII), 135-140. [I491 Gajewski. E., Fuciarelli. E.F. and Dizdaroglu, M. (1988) Int. J. Radiat. Biol. 54. 455-459. [I501 Lesko. S.O., Drocourt. J.L. and Yang. S.U. (1982) Biochemistry 21, 5010-5015. [I511 Kuchino. Y., JMori,F., Kasai, H.. Inoue, H., Iwai. S., Miura, K., Ohtsuka. E. and Nishimura, S. (1987) Nature 327, 77-79. [152] Wood, M.L., Dizdaroglu, IM.. Gajewski, E. and Essigmann, J.M. (1990) Biochemistry 29, 7024-7032. (1531 Di Mascio. P., Wefers, H., Do-Thi, H.P.. Lafleur, M.V.M. and Sies, H. (1989) Biochim. Biophys. Acta 1007, 151-157. [I541 Blazek, E.R.. Peak, J.G. and Peak, M.J. (1989) Photochem. Photobiol. 49, 607-613. [154ai Muller. E., Boiteaux. S., Cunningham, R.P. and Epe, B. (1990) Nucleic Acids Res. 18. 5969-5973. [I551 Schneider, J.E., Price, S., Maidt, L., Gutreridge, J.M.C. and Floyd, R.X. (1990) Nucleic Acids Res. 18. 631-635. [I561 Epe, B., Hegler, J. and Wild, D. (1989) Carcinogenesis 10, 2019-2024. (1571 Peak, J.M., Peak, J.G., Foote, C.S.andKrinsky, N.1. (1984) J. Photochem. 25, 309-315. [I581 Floyd, L A . . West, M.S., Eneff. K.L. and Schneider, J.E. (1990) Free Rad. Biol. Med. 8, 327-330. [I591 Brambilla. G., Martelli. A. and Marinari, U.M. (1989) Mutar. Res. 214, 123-127. [I601 Kasai, H. and Nishimura. S. (1988) in: Medical, Biochemical and Chemical Aspects of Free Radicals (Hayaishi, 0.. Niki, E., Kondo, .M. and Yoshikawa, T.eds.) Elsevier. Amsterdam, 1021-1023. [I611 Vaca, C.E., Wilhelm, J. and Harms-Ringdahl, M. (1988) Mutat. Res. 195, 137-149. (1621 Inouy. S. (1984) FEBS Lett. 172, 231--234. 11631 Hruszkewycz, A M . and Bertgold, D.S. (1990) Mut. Res. 244, 123-128. (J63al Hruszkewwycz, A.M. (1988) Biochem. Biophys. Res. Commun. 153, 191-197. [ I 6 4 1 Curreridge, J.M.C. and Halliwell, B. (1990) Trends Biochem. Sci. 15, 129-135. (1651 Morita, J.. Veda, K., Nakai, K., Baba. Y. and Komano. T. (1983) Agric. Biol. Chem. 17, 2977-2979. (1661 Frankel. E.N.. Neff, W.E., Brooks, D.D. and Fujimoro. K. (1987) Biochim. Biophys. Acta 919, 239-244. [I671 Frankel. E.N. (1987) Chem. Phys. Lipids 44, 73-85. 1168) Esterbauer, H., Zollner, H. and Schaur. R.J. (1988) IS1 Atlas Sci. Biochcm. 1, 311-317. [I691 Brambilla. G.. Sciaba, L., Faggin. P.. Maura, A., Marinari. U.M.. Fe:ro. M. and Esrerbauer, H. (1986) Mutat. Res. 171, 169-176. [I701 Stone. K . , Ksebari, M.B. and ~Marnetr,L.J. (1990) Chem. Res. To\. 3. 33-38. [170a] Stone, K.. Utieblo. A. and Marnett. L.J. (1990) Chem. Res. Tox. 3. 467-472. [1711 Hadley, M. and Draper, H.H. (1990) Lipids 25. 81-85. [I721 Sodum. R.S. and Chung. F.L. (1988) Cancer Res. 48, 320-323. [I731 Breimer, L.H. (1991) Free Radical Res. Commun..in press. 11741 Doetsch, P.W.,Hamilton, K.K., Rapkin. L.. Okenquist, S.A. and Lenz. J. (1990) in: Ionizing Radiation Damage to DNA: Molecular Aspects (Wallace, S.S. and Painter, R.B. eds.) Wiley-Liss, New York, pp. 109-125. [I751 Cathcart, R., Schwiers. E.,Saul, R.L. and Ames, B.N. (1984) Proc. Natl. Acad. Sci. USA 81. 5633-5637. (1761 Adelman, R., Saul. R.L. and Ames. B.N. (1988) Proc. Natl. Acad. Sci. USA 85, 2706-2708. (1771 Frasa. C.C., Shigenaka. W.K., Park, J.W., Degan, P. and Ames, B.N. (1990) Proc. Natl. .4cad. Sci. USA 87, 45334537. [I781 Stillwell, W.G.. Xu. H.X., Adkins. J.A., Wishnok, J.S. and Tannenbaum, S.R. (1989) Chem. Res. Toxicol. 2, 94-99, [I791 Floyd, R.A., Watson, J.J., Wong, P.K., Altmiller. D.H. and Rickard. R.C. (1986) Free Radical Res. Commun. 1, 163-172. [le01 Floyd. R.A., West. M.S., Eneff, K.L.. Schneider, J.E., Wong, P.K.. Tingey, D.T. and Hogsett, W.E. (1990) Anal. Biochem. 188, 155-158. [I811 Floyd. R.A., West, M.S., Hogsett, W.E. and Tingey, D.T. (1989) Plant Physiol. 91. 644-647. [l82l Kasai, H. and Nishimura. S. (1986) Environ. Health Perspect. 67, 111-116. [I831 Kaneko, ,M. and Leadon, S.A. (1986) Cancer Res. 46,71-75. [I841 Hubbard, K., Huang, H., Laspia. M.F., Ide. H.. Erlanger. B.F. and Wallace, S.S. (1989) Radiat. Res. 118, 257-268. [I851 Richter, C.. Park, J.W. and Ames. B.N. (1988) Proc. Natl. Acad. Sci. USA 85, 6465-6467. [I861 Richrer, C. (1988) FEBS Lett. 241. 1-5. [I871 Kohda, K., Tada, .U.,Kasai. H., Nishirnura. S. and Kawazoe, Y. (1986) Biochem. Biophys. Res. Commun. 139, 626-632. [1881 Umemura, T., Sai, K.. Takagi. A.. Hasegawa, R. and Kurokawa. Y. (1990) Carcinogenesis 11. 345-347. [I891 Kasai. H.. Okada, Y., Nishimura, S.. Rao. H.S. and Reddy. J.K. (1989) Cancer Res. 49, 2603-2605. [I901 Hussain, N.S.. Conaway, C.C., Guo. N., Asaad, W. and Fiala, E.S. (1990)Carcinogenesis 11, 1013-1016. [I911 Takagi, A., Sai, K., Umemura. T.. Hasegawa, R. and Kurokawa, Y. (1990) Jpn. J. Cancer Res. 81, 213-215. [191a] Hinrichsen. L.1.. Floyd, R.A. and Sudilovsky, 0. (1990) Carcinogenesis 11, 1879-1881. [191b] Hegi, M.E., Ulrich. D., Sagelsdorf. P.. Richter, C. and Lutz, W.K. (1990) Mutat. Res. 138. 325-329. [I921 Kohda. K., Tada. tM.. Hakura. A.. Kasai, H. and Kawazoe. Y. (1987) Biochem. Biophys Res. Commun. 149, 1141-1148. 11931 Dizdaroglu, M., Aruoma, 0.1. and Halliwell. B. (1990) Biochemistry 29, 8447-845 1. [194a] Dizdaroglu, M. and Gajewski, E. (1990) Methods Enzymol. 186. 530-544. -[I951 Tofigh. S. and Frenkel, K. (1989) Free Rad. Biol. Med. 7, 13 I 143. [I961 Fuciarelli, A.F.. Wegher. B.J., Blakely. W.F. and Dizdaroglu, IM. (1990) Int. J . Radiat. Biol. 58. 397-415. (1971 Dizdaroglu, M. (1985) Anal. Biochem. 144. 593-603. [198] Dizdaroglu, M. (1991) Free Rad. Biol. >led.. in press. [I991 Gajewski, E., Aruoma. 0.1.. Dizdaroglu, M. and Halliwell. B. (1991) Biochemistry, in press. [ZOO] ;Malins, D.C., Ostrander. G.K., Haimanor. R. and Williams. P. (1990) Carcinogenesis 11. 1045-1047. [20l] Jackson, J.H.. Cajewski. E.. Schraufstatter. I.U.. Hyslop. P.A., Fuciarelli, A.F.. Cochrane. C.G. and Dizdaroglu. M . (1989) J . Clin. Invest. 84. 1644-1649. [202) Skipper, P.L. and Tannenbaum, S.R. (1990) Carcinogenesis 1 1 , 507-518. [203] Akman, S.A.. Doroshow. J.H. and Dizdaroglu. M . (1990) Arch. Biochem. Biophys. 282. 102-205. 19