Hydroxyl Radical Is Not the Main Reactive Species Involved in the

Francois Samson-Thibault , Guru S. Madugundu , Shanshan Gao , Jean Cadet ..... Tim Hofer , Arnold Y. Seo , Mercedes Prudencio , Christiaan Leeuwenburg...
0 downloads 0 Views 118KB Size
Download PDF
Chem. Res. Toxicol. 2003, 16, 191-197

191

Hydroxyl Radical Is Not the Main Reactive Species Involved in the Degradation of DNA Bases by Copper in the Presence of Hydrogen Peroxide Sandrine Frelon, Thierry Douki,* Alain Favier, and Jean Cadet* Laboratoire “Le´ sions des Acides Nucle´ iques”, Service de Chimie Inorganique et Biologique, UMR 5046, CEA/DSM/De´ partement de Recherche Fondamentale sur la Matie` re Condense´ e, CEA-Grenoble, 38054 Grenoble Cedex 9, France Received October 22, 2002

Copper is an important biological metal that tightly binds to DNA. Its reaction with endogenously generated hydrogen peroxide may thus lead to the formation of DNA damage. To gain insights into the underlying mechanisms, a comparative study of the damage produced within isolated DNA upon exposure to γ-radiation in aqueous solution, a source of hydroxyl radicals, and incubation with Cu(I) or Cu(II) complexes in the presence of hydrogen peroxide was carried out. Several relevant base modifications were quantified by HPLC-tandem mass spectrometry. It was first shown that addition of copper ions only slightly modified the profile of radiation-induced lesions within DNA. However, the distribution of base modifications was drastically different upon incubation of DNA with Cu(I) or Cu(II) complexes in the presence of H2O2. Indeed, guanine degradation products were produced in much higher yield than lesions of the other bases. These observations are rationalized in terms of the occurrence of one electron oxidation with Cu(I) complexes, as confirmed by the study of the degradation of free thymidine. In contrast, the formation of the sole 8-oxo-7,8-dihydroguanine upon incubation of DNA with Cu(II) ions and H2O2 strongly suggests the production of singlet oxygen as the predominant reactive oxygen species.

Introduction The causative role of ROS1 in deleterious biological processes, such as aging, carcinogenesis, and numerous diseases, has received increasing attention in the past decade. ROS might be of endogenous origin, such as cell aerobic metabolism or inflammation. In addition, ROS are produced upon exposure to various chemical and physical agents. Among the latter, ionizing radiation is known to yield radical species, including hydroxyl radical (•OH), hydrogen atom, and solvated electron, through the ionization of water molecules. Superoxide anion (O2•-), constitutively present in cells because of leakage from the respiratory chain in mitochondria, is also proposed to be a major source of hydroxyl radicals (1). Indeed, O2•may dismutate into hydrogen peroxide, either spontaneously or through metal- and enzyme-catalyzed pathways. Reaction of hydrogen peroxide with superoxide anion, the Haber-Weiss reaction 1, leads to the production of •OH. However, the latter process is only efficient when catalyzed by reduced transition metal ions that trigger the Fenton reaction 2 (2-6). The latter process involves the reduction of hydrogen peroxide by a transition metal ion. * To whom correspondence should be addressed: T.D.: Tel: (33)4 38 78 31 91. Fax: (33)4 38 78 50 90. E-mail: [email protected]. J.C.: Tel: (33)4 38 78 49 87. Fax: (33)4 38 78 50 90. E-mail: [email protected]. 1 Abbreviations: 5-FordUrd, 5-formyl-2′-deoxyuridine; 5-HMdUrd, 5-(hydroxymethyl)-2′-deoxyuridine; 8-oxodAdo, 8-oxo-7,8-dihydro-2′deoxyadenosine; 8-oxodGuo, 8-oxo-7,8-dihydro-2′-deoxyguanosine; FapyAde, 4,6-diamino-5-formamidopyrimidine; FapyGua, 2,6-diamino-4hydroxy-5-fomamidopyrimidine; HPLC-MS/MS, HPLC coupled to tandem mass spectrometry; oPhe, 1,10-phenanthroline; ROS, reactive oxygen species; ThdGly, 5,6-dihydroxy-5,6-dihydrothymidine (thymidine glycols).

In a subsequent step, reduction of higher oxidation state metal ions by superoxide anion (or another reducing agent) regenerates active reduced transition metal ions. In vitro, Mn+ can be Ti(III) or Co(II) (7) but iron and, to a lesser extent, copper are the most likely promoters of • OH radicals in vivo (8).

O2•- + H2O2 f O2 + OH- + •OH

(1)

Mn+ + H2O2 f Mn+1 + OH- + •OH

(2)

The reactive species involved in metal-driven endogenous oxidative stress can damage biomolecules such as DNA. The resulting lesions include strand breaks, abasic sites, and modified bases (9-11). Copper may be of particular importance in the induction of oxidative DNA damage. Indeed, cations of the latter metal form stable complexes with DNA (association constant Cu(I)/DNA, 109 M; Cu(II)/DNA, 104 M (12)). Cu2+ ions may undergo intercalation (13) as well as complexation to purine bases (12, 14, 15). Therefore, Cu+ and Cu2+ ions present within cellular DNA may be involved in the oxidative degradation of DNA, either by initiating the production of ROS or by modifying the evolution of initially produced radicals of DNA components. The actual nature of the DNA damaging species produced upon reaction of copper with hydrogen peroxide is still a matter of debate. Previous studies have shown that following complexation of copper with the DNA molecule, DNA-copper-hydroperoxo (Cu(I)OOH) species (16) were produced by reaction with reactive oxygen compounds. In contrast, other works, based on either use of radical scavengers or

10.1021/tx025650q CCC: $25.00 © 2003 American Chemical Society Published on Web 01/18/2003

192

Chem. Res. Toxicol., Vol. 16, No. 2, 2003

Figure 1. Structure of the studied modified bases (dR, 2-deoxyribose).

chemiluminescence measurements (17-21), led to the conclusion that singlet oxygen was produced, together with low amounts of hydroxyl radical. The determination of the spectrum of modified DNA bases (9, 10) is likely to provide insights into the chemical nature of the involved degrading species. Indeed, singlet oxygen specifically generates 8-oxodGuo (22) while •OH leads to the formation of a wide array of base damage (10) (Figure 1). One electron oxidation also gives rise to a specific pattern of DNA base damage. Indeed, migration of the positive charges along DNA (23-25) leads to the predominant degradation of guanine, which exhibits the lowest oxidation potential among DNA components. In addition, the redox conditions play a critical role in the final damage distribution, as shown by the effect of the presence of oxygen (26, 27) and reducing species (28) on the relative yields of guanine lesions. We thus compared the distribution of final base damage upon exposure of isolated DNA to •OH arising from the γ-radiolysis of water with that obtained upon degradation of DNA bases by the reactive species produced upon reaction of copper/ oPhe complexes with hydrogen peroxide. Cu(oPhe)2+ complex has been extensively studied for its artificial nuclease activity (29-32). However, only limited information is available on the damage induced to the DNA bases under the latter conditions. In the present work, we first determined the yield of formation of 8-oxodGuo, FapyGua, 8-oxodAdo, FapyAde, thymidine glycols (ThdGly), 5-HMdUrd, and 5-FordUrd by HPLC-MS/MS within DNA samples exposed to γ-radiation in the presence and the absence of added copper ions. The level of the seven modified bases was then measured within DNA samples exposed to the reactive species produced upon reaction of hydrogen peroxide with Cu+ or Cu2+/oPhe complexes. The comparison of base damage distributions provided information on the oxidation mechanisms involved under the latter conditions.

Experimental Section Chemicals. Nuclease P1 (Penicillium citrium), calf spleen phosphodiesterases I and II, and calf thymus DNA were obtained from Sigma (St. Louis, MO). Alkaline phosphatase was purchased from Roche Diagnostics (Mannheim, Germany). EDTA was from Interchim (Montluc¸ on, France). Water was deionized with a Millipore Milli-Q system (Molsheim, France). Oxidation of DNA and Thymidine (Thd). An aerated aqueous solution of calf thymus DNA (0.5 mg/mL) with or without added metallic ion (0-100 µM CuCl or CuSO4) in the presence or absence of oPhe (0-100 µM) was exposed to the γ-rays emitted by a 60Co source immersed in a water pool. The dose rate was 20 Gy/min as determined by poly(methyl methacrylate) dosimetry. Air was continuously bubbled through the solution during the irradiation. Increasing doses ranging be-

Frelon et al. tween 40 and 100 Gy were applied. Experiments were repeated in the presence of 5 mM Tris. Hydrogen peroxide (final concentration 0-100 µM) was added to an aerated aqueous solution of calf thymus DNA (0.5 mg/mL). The resulting mixture was then treated with freshly made solutions of either Cu(I)Cl or Cu(II)SO4 (0-100 µM) in the presence or absence of oPhe (0-100 µM). After 1 h of incubation, DNA was precipitated by addition of sodium chloride and ethanol. After it was centrifuged, the pellet was dissolved in 100 µL of 0.1 mM desferroxamine aqueous solution. Exposure of DNA to the oxidizing species produced by H2O2 in the presence of Cu(I) or Cu(II) mixed or not with oPhe was repeated with DNA solutions containing 5 mM Tris. Oxidation of DNA by Cu(II)/H2O2 (100 µM) was also performed in D2O. For this purpose, 500 µL of DNA solution was concentrated to 50 µL under vacuum. D2O (500 µL, 99.9%) was added. The sample was again concentrated and diluted in D2O. Thd (1 mM) was treated using the same protocol with the omission of the final precipitation step. The concentrated solutions of copper salts, oPhe, and Tris as well as the DNA solution were all set at pH 7. Analysis of the Oxidized Nucleosides. Aliquots of the DNA solution (100 µL, 50 µg) were digested by incubation with 0.004 U of phosphodiesterase II and 5 U of nuclease P1 diluted in its buffer (300 mM sodium acetate, 1 mM ZnSO4, pH 5.3). Additional phosphodiesterase buffer (10 µL, 200 mM succinic acid, 100 mM CaCl2, pH 6) was added, and the sample was left for 2 h at 37 °C. Then, 0.003 U of phosphodiesterase I and 5 U of alkaline phosphatase in 10 µL of the alkaline phosphatase buffer (500 mM Tris, 1mM EDTA, pH 8.5) were added. The mixture was subsequently incubated for 2 h at 37 °C. Samples were then transferred into HPLC injection vials. Isotopically labeled internal standards (25 pmol of each of [2-amino-1,3,7,915N ]-8-oxodGuo, [6-amino-1,3,7,9-15N ]-8-oxodAdo, [5-(hydroxym5 5 ethyl)-13C,1,3-15N2]-5-HMdUrd, [5-formyl-13C,1,3-15N2]-5-FordUrd, and [1,3-15N2,5-methyl-13C]ThdGly) were added to the samples. The latter was then injected onto an HPLC Uptisphere ODB (5 µm, 150 mm × 2 mm id) octadecylsilyl silica gel column (Interchim). The elution was achieved at a flow rate of 0.2 mL/ min in the gradient mode using a 2 mM aqueous solution of ammonium formate containing increasing amounts of acetonitrile. Methanol was added at the outlet of the column at a flow rate of 0.2 mL/min. The resulting eluent was then directed toward a L 4000 Merck-Hitachi UV detector set at 280 nm to monitor the elution of normal nucleosides. Oxidized nucleosides were quantified online by a API 3000 tandem mass spectrometer (Perkin-Elmer/SCIEX, Thornhill, Canada) operating in the multiple reactions monitoring mode as previously described (33). Analysis of FapyAde and FapyGua. Aliquots of the DNA solution (100 µL, 50 µg) were digested in the presence of 100 pmol of [4,6-diamino-15N2,5-formylamino-13C]FapyAde and [2,6diamino,5-formylamino-15N3]FapyGua. This was achieved by incubation for 2 h at 37 °C in the presence of 10 U of nuclease P1 suspended in 10 µL of an aqueous solution of 30 mM NH4OAc and 0.1 mM ZnSO4 at pH 5.5. After the digestion was completed, the enzyme was precipitated by addition of 50 µL of chloroform. Then, the aqueous layer was transferred into HPLC injection vials, frozen, and finally lyophilized. The resulting dry mixture of nucleotides was dissolved into 60 µL of 88% formic acid. The sample was left for 20 min at room temperature. Then, formic acid was removed under vacuum. A mixture of acetonitrile and 10 mM ammonium formate (90:10 v/v) was added to the dry residue. Samples were then analyzed by HPLC-MS/MS in the MRM mode (33). An Hypersil NH2 (5 µm, 150 mm × 2 mm id) silica gel column (Interchim) was used in the isochratic mode with a mixture of 10 mM ammonium formate and acetonitrile ([90:10] v/v) at a flow rate of 200 µL/min.

Results Set Up of the Reaction Conditions. Copper ions are known to efficiently bind to DNA. Therefore, the amount of copper present within commercially available calf

Copper/Hydrogen Peroxide-Induced DNA Damage

Figure 2. Distribution of base damage within isolated DNA exposed to γ-radiation in the presence or the absence of copper(I)/phenanthroline complexes. Results (expressed in lesions/106 bases per Gy) represent the slope ( standard error of the slope of the linear regression of the level of lesion with respect to the dose.

thymus DNA was determined. Using flame absorption spectrometry, a concentration of 2 µg/L (0.03 µM) was found in the solutions containing 0.5 mg/mL of DNA. This was far below the concentration of copper added in the present work (up to 100 µM) and was therefore considered as negligible. Unsuccessful attempts were made to reduce the copper content in isolated DNA, either by using a chelex cation exchange resin or by dialysis. However, the content in iron was drastically reduced by chelex treatment (from 55 to 15 µg/L) that was then applied before all experiments. The amount of copper present in the other chemicals used (water, phenanthroline) was always below the detection limit of the flame absorption spectrometry measurement. Comparison of the distribution of base damage induced upon γ-irradiation on one hand and under Fenton reaction conditions on the other hand required the two experimental protocols to be as similar as possible. Therefore, we decided to use Cu(I) complexes to trigger Fenton reaction rather than Cu(II) ions in the presence of a reducing agent. Handling CuCl solution led to the use of thoroughly argon-degassed deionized water for the preparation of the solutions of phenanthroline complexes. In addition, the latter chelating agent was solubilized prior to being mixed with cuprous chloride. Using this protocol, the reaction of Cu+ with molecular oxygen was limited and the reduced copper ions were rapidly stabilized in complexes. The resulting solutions were then added to aerated DNA samples (0.5 mg/mL) containing increasing amounts of hydrogen peroxide (0-100 µM). It might be added that the experiments were all carried out in the presence or the absence of oPhe. The latter compound was found to drastically modify the H2O2mediated DNA damage in the presence of Cu+ but not of Cu2+. Effect of Copper Ions and Tris on the γ-Radiation-Induced Degradation of DNA Bases. As previously reported (33, 34), exposure of isolated DNA to the • OH radicals produced upon radiolysis of water in aerated aqueous solution led to the formation in significant yields of the seven selected degradation products of adenine, guanine, and thymine (Figure 2). Addition of the cuprous ion/phenanthroline complex to the DNA solution prior to irradiation had a low influence on the distribution of induced modified DNA bases. The overall rate of forma-

Chem. Res. Toxicol., Vol. 16, No. 2, 2003 193

Figure 3. Distribution of base damage within isolated DNA exposed to γ-radiation in the presence or the absence of 5 mM Tris. Results (expressed in lesions/106 bases per Gy) represent the slope ( standard error of the slope of the linear regression of the level of lesion with respect to the dose.

Figure 4. Increase in the level of modified bases within isolated DNA incubated in the presence of 100 µM of Cu+/oPhe and increasing concentrations of hydrogen peroxide.

tion of oxidized bases remained roughly unchanged. However, an increase in the ratio between the yield of formation of 8-oxopurines and their related Fapy derivatives was observed. For guanine, the corresponding values were 1.7 and 2.7 upon irradiation in the absence and the presence of 100 µM Cu(oPhe)2+, respectively. In addition, the level of 5-FordUrd was slightly higher with respect to that of 5-HMdUrd upon addition of the copper complex. Exposure of an aerated aqueous solution of isolated DNA was also performed in the presence of 5 mM Tris. In agreement with the hydroxyl radical scavenging properties of the latter compound, the yield of formation of the lesions was found to be 10 times lower than in the absence of Tris (Figure 3). DNA Base Damage Induced by the Cu(oPhe)2+/ H2O2 System. Isolated DNA in aerated aqueous solution was incubated in the presence of Cu(oPhe)2+ and hydrogen peroxide. For a given amount of copper complex, the level of each of the modified bases was found to be proportional to H2O2 concentration (Figure 4). Therefore, reported results represent the value of the latter slope. The main observation dealt with the overwhelming formation of 8-oxodGuo and FapyGua while the other lesions were produced in low yields (Figure 5). This was

194

Chem. Res. Toxicol., Vol. 16, No. 2, 2003

Figure 5. Distribution of bases damage within isolated DNA incubated with copper and hydrogen peroxide. Results (expressed in lesions/106 bases per µM H2O2) represent the slope ( standard error of the slope of the linear regression of the level of lesion with respect to the hydrogen peroxide concentration.

observed at concentration of Cu(I) complex of either 10 (data not shown) or 100 µM. Under the latter conditions, modification products of guanine represented 82% of the overall amount of quantified lesions. The corresponding value was only 51% upon γ-irradiation. Addition of 5 mM Tris to the DNA solution did not decrease the yield of lesion. DNA Base Damage Induced by H2O2 in the Presence of Cu+, Cu2+, and Cu2+/oPhe. When hydrogen peroxide treatment of DNA in the presence of Cu+ was repeated without oPhe, guanine was still found to be the major target, even though the contribution of thymine and adenine oxidation products was slightly higher than in the presence of Cu(oPhe)2+ (35 and 18%, respectively). However, in contrast to the results obtained with the latter oxidizing system, exposure of DNA to H2O2 and free Cu+ ions led to the formation of 8-oxodGuo in much higher yield than FapyGua (Figure 5). A ratio of 9.5 was observed between the yield of the two guanine degradation products while the corresponding values were 2.7 and 1.4 in DNA solution containing 100 µM of Cu(oPhe)2+ exposed to γ-radiation and hydrogen peroxide, respectively. The pattern of base modification observed upon incubation of isolated DNA with a mixture of Cu2+ ions, oPhe, and hydrogen peroxide was similar to that determined upon oxidation by Cu+/H2O2. Indeed, the proportion of guanine modification products was 70% while the ratio between the yields of 8-oxodGuo and FapyGua was 9.3 under the former conditions. DNA was also oxidized by free Cu2+ ions in the presence of H2O2. A distribution of base damage similar to that obtained with Cu2+/oPhe was observed. DNA was also exposed to Cu2+/H2O2 in D2O. An increase in the yield of 8-oxodGuo by a factor of 7 was observed (Figure 6). Oxidation of DNA by hydrogen peroxide in the presence of free Cu+, free Cu2+, and Cu2+/ oPhe was repeated in the presence of 5 mM Tris. The yield of formation of the lesions was not significantly modified. Oxidation of Thd. A series of experiments was then performed with Thd as a substrate under aerated conditions (Figure 7). γ-Irradiation of aerated solutions of the free nucleoside led to the overwhelming formation of Thd glycols. This result was not modified by the addition of Cu(oPhe)2+. In contrast, oxidation of Thd by a mixture of Cu(oPhe)2+ and hydrogen peroxide led to the formation

Frelon et al.

Figure 6. Distribution of oxidized nucleosides within isolated DNA exposed to the oxidizing species produced upon reaction of Cu2+ ions with hydrogen peroxide. Results (expressed in lesions/106 bases per µM H2O2) represent the slope ( standard error of the slope of the linear regression of the level of lesion with respect to the hydrogen peroxide concentration.

Figure 7. Relative contribution of Thd glycols and the methyloxidation products of Thd under different oxidizing conditions.

of methyl oxidation products in a global yield similar to that of ThdGly. Reaction of Thd with H2O2 and Fe2+/ EDTA instead of Cu(oPhe)2+ gave rise to the same pattern of lesions than upon exposure to γ-radiation. In contrast, UVA menadione photosensitization led to the formation of 5-HMdUrd and 5-FordUrd in significant yields as the result of the transient formation of the Thd radical cation and its subsequent deprotonation (35).

Discussion The distribution of radical-induced modified bases is known to be very sensitive to a series of parameters involved in the degradation reaction (10). First, the nature of the oxidizing species is expected to play a major role. For instance, hydroxyl radical (•OH) reacts with all nucleobases at several positions of the purine and pyrimidine rings. This leads to the formation of at least 60 DNA base lesions, including hydroperoxides and diastereoisomers of several products, identified in monomeric model systems (10). In contrast, singlet oxygen specifically reacts with guanine, leading, in DNA, to the overwhelming formation of 8-oxodGuo (22). One electron oxidation processes represent an intermediate situation.

Copper/Hydrogen Peroxide-Induced DNA Damage

At the nucleoside level, the initial formation of the radical cation of the four bases leads to the generation of a wide range of modified bases. Most of them are also produced by initial attack of •OH radicals, even though in different relative yields. However, when generated within DNA, positive holes migrate toward the sites of lowest ionization potential, namely, guanine residues, preferably located at the 5′-end of GG doublets (23-25). This predominant degradation of guanine upon one electron oxidation has been shown by visualizing alkali labile sites (23, 24, 36, 37) and quantifying sites sensitive to the action of repair enzymes (38, 39) in oxidized oligonucleotides of defined sequence. Direct quantification of modified bases by chromatographic assays has led to similar conclusions in isolated calf thymus DNA (40, 41). In addition to the chemical nature of the oxidizing agent, other reaction parameters have to be considered. These include oxygen, pH, reducing species, and structural factors. Most of these parameters are likely to modulate the chemical reactions of initially produced radicals. This may be illustrated by the chemical reactions of the neutral reducing 8-hydroxy-7,8-dihydro-7-yl radical of purine bases arising from either addition of •OH to the C8 position or hydration of the guanine or adenine radical cation. Upon oxidation, the latter purine radicals are converted into the related 8-oxo derivative while the corresponding 5-formamidopyrimidine lesion is produced upon reduction. This is clearly shown by the effect of oxygen and reducing species. Another example is the evolution of the 5-peroxymethyl radical of thymine that mostly yields 5-FordUrd and 5-HMdUrd upon dismutation and reduction, respectively. Copper ions are among the cellular components that may interfere with DNA radical chemistry. Therefore, we investigated the effect of the addition of Cu(oPhe)2+ complexes on the distribution of γ-radiation-induced base damage in aerated aqueous solution of DNA. Only minor changes were observed. The main effect of adding Cu+ ions was an increase in the ratio between 8-oxopurines and their related Fapy derivatives. Similarly, the ratio between the yield of 5-FordUrd and the yield of 5-HMdUrd was found to increase. As mentioned above, these observations are indicative of an increased rate of oxidation of the purine and thymine radicals involved in the formation of the damaged bases. A likely oxidant present in the system is the Cu2+ ion, which may be produced upon reaction of Cu+ with •OH or molecular oxygen. It should be pointed out that the initial base radicals are likely to be the same in the presence or the absence of Cu(oPhe)2+ since the overall formation of lesions at either guanine, thymine, or adenine is constant. In particular, the conformational changes that may have been induced by the addition of Cu+ or Cu2+ (12, 42) do not seem to drastically affect the initial events involved in the •OHmediated degradation of DNA bases. Therefore, the presence of Cu+ or Cu2+ cannot account for the drastically different base damage distribution that was observed upon incubation of isolated DNA in aerated aqueous solution in the presence of Cu(oPhe)2+ complex and hydrogen peroxide. Indeed, under the latter conditions, the two main degradation products were found to be 8-oxodGuo and FapyGua. This preferential degradation of guanine strongly suggests the occurrence of a predominant one electron oxidation reaction. The associated hole migration process toward guanine moieties would favor the formation of the guanine radical cation

Chem. Res. Toxicol., Vol. 16, No. 2, 2003 195

(43) and, thus, of the related degradation products. The lack of decrease in the yield of 8-oxodGuo upon addition of Tris allows us to rule out a major contribution of hydroxyl radical in the formation of the latter lesion upon exposure of DNA to Cu(oPhe)2+/H2O2. Indeed, Tris was found to be an effective •OH scavenger upon γ-irradiation. In particular, Tris did not give rise to secondary radicals able to oxidize guanine as previously reported for DMSO (34, 44). With an oxidizing system involving copper ions and H2O2, a favored degradation of guanine could alternatively be accounted for by a preferential complexation of Cu+ to the latter base (15, 45), leading then to the site specific production of nonscavengeable hydroxyl radicals. However, even though •OH is highly reactive and only diffuses on a very short distance, it seems unlikely that it would not react with neighboring bases even if copper ions are bound to guanines. Production of •OH is likely to be partially involved in the Cu(oPhe)2+/H2O2-mediated degradation of DNA pathways that could account for the formation, in low yields, of thymine and adenine lesions. It should be mentioned that similar distributions of DNA base damage induced by Cu+/H2O2 and hydroxyl radicals have been previously reported (46, 47). However, these works were based on the use of a GC/MS assay, the reliability of which has been questioned (48). Observation of a one electron oxidation of nucleobases by the Cu(oPhe)2+/H2O2 system may be also accounted for by the specific interaction of the latter complex with DNA. Indeed, Cu(oPhe)2+ is known to bind to the minor groove of DNA (49) and is thus in close vicinity of the substrate. As a result, one electron oxidation processes might be favored over classical Fenton reaction pathways involving production of •OH. Therefore, the oxidizing properties of the Cu(oPhe)2+/H2O2 system were studied with Thd, a free nucleoside with which no intercalation is possible. This particular DNA component was chosen because it is known that hydroxyl radical preferentially adds to the C5-C6 double bond while they abstract a hydrogen atom of the methyl group with a lower efficiency (10). In contrast, the radical cation of Thd has been shown to undergo hydration, giving rise mostly to the 6-hydroxy-5,6-dihydrothymid-5-yl radical on one hand, and deprotonation of the methyl group on the other hand (Figure 8). The ratio between these two competitive reactions is approximately 70:30 at neutral pH (35). When Thd was exposed to the oxidizing species produced upon reaction of H2O2 with Cu(oPhe)2+, the yield of methyl oxidation products was almost as high as that of Thd glycols. Interestingly, this result was not accounted for by the involvement of a metal ion in the oxidation process. Indeed, exposure of Thd to the species produced upon reaction of the Fe2+/EDTA complex with hydrogen peroxide led to the same distribution of degradation products upon γ-irradiation, in agreement with the involvement of •OH as the oxidizing species. Altogether, these observations provide further evidence for the ability of Cu(oPhe)2+ to trigger one electron oxidation processes in the presence of hydrogen peroxide. The underlying mechanism is not elucidated. It may be proposed that reaction of H2O2 with the Cu(oPhe)2+ complex leads to the formation of a phenanthroline carbon-centered radical. The latter species would react with molecular oxygen to yield a peroxyl radical that may induce one electron oxidation processes. However, other pathways such as the production of unidentified metallic species may also be involved.

196

Chem. Res. Toxicol., Vol. 16, No. 2, 2003

Frelon et al.

proposed upon complexation of Cu(II) with DNA. The resulting species might lead to the formation of DNACu-hydroperoxo complex and then to the in situ generation of singlet oxygen (16, 18, 51). Further evidence for this pathway was provided by the observation of a much lower ratio between the yield of strand breaks and 8-oxodGuo upon degradation by the Cu(II)/H2O2 system than either γ-radiolysis or treatment with Fe(II)/H2O2 (52), in agreement with the fact that 1O2 does not induce frank DNA strand breaks (53).

Conclusion

Figure 8. Formation of the initial radicals of Thd upon •OH addition and one electron oxidation of Thd (dR, 2-deoxyribose).

The accurate determination of the distribution of base damage together with the lack of significant decrease in the yield of lesions upon addition of Tris also allowed us to rule out a major contribution of hydroxyl radicals in the DNA damage induced by hydrogen peroxide and either free Cu+ and Cu2+ ions or Cu2+/phenanthroline mixture. The great similarities between the results obtained under these conditions might be explained by the low stability of free Cu+ ions that are likely to undergo oxidation into Cu2+ upon reaction with the molecular oxygen present in the DNA solution. In addition, oPhe did not significantly modify the distribution of base damage. In all cases, guanine was found to be a preferential target. Formation of oxidation products of thymine and adenine was also observed but in much lower relative yields than upon exposure to the •OH generated by γ-radiolysis. The latter observation suggests that hydroxyl radical is produced in low yield upon reaction of Cu2+ with H2O2. However, the major degradation product was then 8-oxodGuo. Interestingly, the ratio between the yield of FapyGua and the yield of 8-oxodGuo was much lower than upon exposure to γ-radiations and to the Cu(oPhe)2+/H2O2 system. This result strongly suggests that the formation of 8-oxodGuo within DNA upon reaction of Cu2+ with H2O2 only slightly involves the guanine reducing radical that is the common guanine precursor of 8-oxodGuo and FapyGua upon addition of • OH and one electron oxidation. Therefore, a likely candidate for the oxidizing species involved in the Cu2+/ oPhe/H2O2 system is singlet oxygen. Indeed, 1O2 reacts with guanine to yield a 4,8-endoperoxide that further rearranges to specifically yield 8-oxodGuo within DNA. Neither FapyGua nor other identified guanine lesions could be detected within 1O2-treated isolated DNA (22). Further evidence for the production of singlet oxygen by reaction of Cu2+ with hydrogen peroxide was provided by the increase in the yield of 8-oxodGuo when DNA was solubilized in deuterium oxide instead of water. Indeed, the lifetime of 1O2 is 10 times longer in D2O than in H2O (50). This increased value is expected to favor the formation of singlet oxygen-mediated DNA damage. The formation of singlet oxygen has actually already been

The present data show that hydroxyl radicals contribute to DNA base damage induced by copper/oPhe complexes in the presence of hydrogen peroxide only to a minor extent. These observations confirm previous works mostly focused on the formation of DNA strand breaks and damage to 2-deoxyribose moieties (for a review see 11). Degradation of DNA bases by Cu(Phe)2+ and hydrogen peroxide was found to be rather mediated by one electron oxidation. Moreover, Cu2+, as a free ion or as part of complexes, generates singlet oxygen as the predominant reactive species upon reaction with H2O2. Interestingly, we recently reported that the Fenton reaction triggered by ferrous ions leads to a distribution of base damage strongly indicative of a major role of hydroxyl radical (54). The major differences in oxidation pathways triggered by the two transition metals have to be considered in the estimation of their deleterious cellular effects.

Acknowledgment. The work was partly supported by a grant from Commite´ de Radioprotection (Electricite´ de France). The authors thank Josiane Arnaud and Ve´ronique Ducros (CHU, Grenoble, France) for the determination of metal concentrations.

References (1) Halliwell, B., and Gutteridge, J. M. C. (1985) Role of transition metals in superoxide-mediated toxicity. Superoxide Dismutase 3, 45-82. (2) Walling, C. (1982) The nature of the primary oxidants in oxidations mediated by metal ions. in Oxidases and Related Redox Systems (King, T. E., Mason, H. S., and Morrison, M., Eds.) pp 85-97, Pergamon Press, Oxford, U.K. (3) Goldstein, S., and Czapski, G. (1985) Superoxide dismutase activity of some copper phenanthroline complexes and the mechanism of the oxidation of various cuprous complexes by H2O2. In Superoxide and Superoxide Dismutase in Chemistry (Rotilio, G. E., Ed.) pp 64-66, Elsevier Scientific, New York. (4) Johnson, G. R. A., Nazhat, N. B., and Saadalla-Nazhat, R. A. (1985) Reaction of the aquocopper(I) ion hydrogen peroxide: evidence against hydroxyl free radical formation. J. Chem. Soc. Chem. Commun., 407-408. (5) Sutton, H. C., and Winterbourn, C. C. (1989) On the participation of higher oxidation states of iron and copper in Fenton reactions. Free Radical Biol. Med. 6, 53-60. (6) Qian, S. Y., and Buettner, G. R. (1999) Iron and dioxygen chemistry is an important route to initiation of biological free radical oxidations: an electron paramagnetic resonance spin trapping study. Free Radical Biol. Med. 26, 1447-1456. (7) Halliwell, B., and Gutteridge, J. M. C. (1984) Oxygen toxicity, oxygen radicals, transition metals and disease. Biochem. J. 219, 1-14. (8) Halliwell, B., Gutteridge, J. M. C., and Blake, D. (1985) Metal ions and oxygen radical reactions in human inflammatory joint disease. Philos. Trans. R. Soc. London 311, 659-671. (9) Breen, A. P., and Murphy, J. A. (1995) Reactions of oxyl radicals with DNA. Free Radical Biol. Med. 18, 1033-1077. (10) Cadet, J., Berger, M., Douki, T., and Ravanat, J.-L. (1997) Oxidative damage to DNA: formation, measurement, and biological significance. Rev. Physiol. Biochem. Pharmacol. 131, 1-87.

Copper/Hydrogen Peroxide-Induced DNA Damage (11) Pogozelski, W. K., and Tullius, T. D. (1998) Oxidative strand scission of nucleic acids: route initiated by hydrogen abstraction from the sugar moiety. Chem. Rev. 98, 1089-1107. (12) Trumbore, C. N., Ehrlich, R. S., and Myers, Y. N. (2001) Changes in DNA conformation induced by gamma irradiation in the presence of copper. Radiat. Res. 155, 453-465. (13) Sagripanti, J. L., Goering, P. L., and Lamanna, A. (1991) Interaction of copper with DNA and antagonism by other metals. Toxicol. Appl. Pharmacol. 110, 477-485. (14) Burda, J. V., Sponer, J., Leszczynski, J., and Hobza, P. (1997) Interaction of DNA base pairs with various metal cations (Mg2+, Ca2+, Sr2+, Ba2+, Cu+, Ag+, Au+, Zn2+, Cd2+, and Hg2+): nonempirical ab initio calculations on structures energies, and nonadditivity of the interaction. J. Phys. Chem. B 101, 9670-9677. (15) Burda, J. V., Sponer, J., and Hobza, P. (1996) Ab initio study of the interaction of guanine and adenine with various mono- and bivalent metal cations (Li+, Na+, K+, Rb+, Cs+, Cu+, Ag+, Au+, Mg2+, Ca2+, Sr2+, Ba2+, Zn2+, Cd2+, and Hg2+). J. Phys. Chem. 100, 7250-7255. (16) Schweigert, N., Acero, J. L., Von Gunten, U., Canonica, S., Zehnder, A. J. B., and Eggen, R. I. L. (2000) DNA degradation by the mixture of copper and catechol is caused by DNA-copperhydroperoxo complexes, probably DNA-Cu(I)OOH. Environ. Mol. Mutagen. 36, 5-12. (17) Ding, A. H., and Chan, P. C. (1984) Singlet oxygen in coppercatalyzed lipid peroxidation in erythrocyte membranes. Lipids 19, 278-284. (18) Yamamoto, K., and Kawanishi, S. (1989) Hydroxyl free radical is not the main active species in site-specific DNA damage induced by copper (II) ion and hydrogen peroxide. J. Biol. Chem. 264, 15435-15440. (19) Zaidi, R., and Hadi, S. M. (1992) Strand scission in DNA by gossypol and Cu(II): Role of Cu(I) and oxygen-free radicals. J. Biochem. Toxicol. 7, 213-217. (20) Ma, W.-J., Cao, E.-H., and Qin, J.-F. (1999) The involvement of singlet oxygen in copper-phenanthroline/H2O2-induced DNA base damage: A chemiluminescent study. Redox Rep. 4, 271-276. (21) Gilbert, B. C., Silvester, S., Walton, P. H., and Whitwood, A. C. (1999) DNA damage via intercalation of copper complexes and activation by ascorbate and peroxides: direct EPR evidence for hydroxyl radical formation and reaction. J. Chem. Soc., Perkin Trans. 2, 1891-1895. (22) Ravanat, J.-L., Saint-Pierre, C., Di Mascio, P., Martinez, G. R., Medeiros, M. H., and Cadet, J. (2001) Damage to isolated DNA mediated by singlet oxygen. Helv. Chim. Acta 84, 3702-3709. (23) Schuster, G. B. (2000) Long-range charge transfer in DNA: transient structural distortions control the distance dependence. Acc. Chem. Res. 33, 253-260. (24) Giese, B. (2000) Long-distance charge transport in DNA: the hopping mechanism. Acc. Chem. Res. 33, 631-636. (25) Treadway, C. R., Hill, M. G., and Barton, J. K. (2002) Charge transport through a molecular π-stack: double helical DNA. Chem. Phys. 281, 409-428. (26) Berger, M., and Cadet, J. (1985) Isolation and characterization of the radiation-induced degradation products of 2′-deoxyguanosine in oxygen-free aqueous solutions. Z. Naturfosch. 40b, 1519-1531. (27) Douki, T., Martini, R., Ravanat, J.-L., Turesky, R. J., and Cadet, J. (1997) Measurement of 2,6-diamino-4-hydroxy-5-formamidopyridine and 8-oxo-7,8-dihydroguanine in isolated DNA exposed to gamma radiation in aqueous solution. Carcinogenesis 18, 23852391. (28) Douki, T., Spinelli, S., Ravanat, J.-L., and Cadet, J. (1999) Hydroxyl radical-induced degradation of 2′-deoxyguanosine under reducing conditions. J. Chem. Soc., Perkin Trans. 2, 1875-1880. (29) Chen, C. H., and Sigman, D. S. (1986) Nuclease activity of 1,10phenanthroline-copper: sequence-specific targeting. Proc. Natl. Acad. Sci. U.S.A. 83, 7147-7151. (30) Chen, C. H., Mazumder, A., Constant, J. F., and Sigman, D. S. (1993) Nuclease activity of 1,10-phenanthroline-copper. New conjugates with low molecular weight targeting ligands. Bioconjugate Chem. 4, 69-77. (31) Oyoshi, T., and Sugiyama (2000) Mechanism of DNA strand scission induced by (1,10-phenathroline)copper complex: major direct DNA cleavage is not through 1′,2′-dehydronucleotide intermediate nor β-elimination of forming ribonolactone. J. Am. Chem. Soc. 122, 6313-6314. (32) Chen, C.-H. B., Milne, L., Landgraf, R., Perrin, D. M., and Sigman, D. S. (2001) Artificial nucleases. ChemBioChem 2, 735-740. (33) Frelon, S., Douki, T., Ravanat, J.-L., Pouget, J.-P., Tornabene, C., and Cadet, J. (2000) High-performance liquid chromatographytandem mass spectrometry measurement of radiation-induced

Chem. Res. Toxicol., Vol. 16, No. 2, 2003 197

(34) (35)

(36) (37)

(38)

(39)

(40) (41) (42) (43)

(44)

(45)

(46) (47)

(48) (49)

(50) (51)

(52)

(53)

(54)

base damage to isolated and cellular DNA. Chem. Res. Toxicol. 13, 1002-1010. Douki, T., Bretonnie`re, Y., and Cadet, J. (2000) Protection against radiation-induced degradation of DNA bases by polyamines. Radiat. Res. 153, 29-35. Decarroz, C., Wagner, J. R., Van Lier, J. E., Krishna, C. M., Riesz, P., and Cadet, J. (1986) Sensitized photooxidation of thymidine by 2-methyl-1,4-naphthoquinone. Characterization of the stable photoproducts. Int. J. Radiat. Biol. 50, 491-505. Hall, D. B., Holmlin, R. E., and Barton, J. K. (1996) Oxidative DNA damage through long-range electron transfer. Nature 382, 731-735. Kelley, S. O., Holmlin, R. E., Stemp, E. D. A., and Barton, J. K. (1997) Photoinduced electron transfer in ethidium-modified DNA duplexes: dependence on distance and base stacking. J. Am. Chem. Soc. 119, 9861-9870. Melvin, T., Cunniffe, S., Papworth, D., Roldanarjona, T., and O’Neill, P. (1997) Irradiation of DNA with 193 nm light yields formamidopyrimidine-DNA glycosylase (Fpg) protein-sensitive lesions. Photochem. Photobiol. 65, 660-665. Angelov, D., Spassky, A., Berger, M., and Cadet, J. (1997) Highintensity UV laser photolysis of DNA and purine 2′-deoxyribonucleosides: formation of 8-oxopurine damage and oligonucleotides strand cleavage as revealed by HPLC and gel electrophoresis studies. J. Am. Chem. Soc. 119, 11373-11380. Douki, T., and Cadet, J. (1999) Modification of DNA bases by photosensitized one-electron oxidation. Int. J. Radiat. Biol. 75, 571-581. Douki, T., Angelov, D., and Cadet, J. (2001) UV Laser photolysis of DNA: effect of duplex stability on charge-transfer efficiency. J. Am. Chem. Soc. 123, 11360-11366. Savoye, C., Sabattier, R., Charlier, M., and Spotheim-Maurizot, M. (1996) Sequence-modulated radiosensitization of DNA by copper ions. Int. J. Radiat. Biol. 70, 189-198. Candeias, L. P., and Steenken, S. (2000) Reaction of HO• with guanine derivatives in aqueous solution: formation of two different redox-active OH-adduct radicals and their unimolecular transformation reactions. Properties of G(-H)•. Chem. Eur. J. 6, 475-484. Milligan, J. R., Ng, J. Y.-Y., Wu, C. C. L., Aguilera, J. A., Ward, J. F., Kow, Y. W., Wallace, S. S., and Cunningham, R. P. (1996) Methylperoxyl radicals as intermediates in the damage to DNA irradiated in aqueous dimethyl sulfoxide with gamma rays. Radiat. Res. 146, 436-443. Sorokin, V. A., Valeev, V. A., Gladchenko, G. O., Sysa, I. V., Blagoi, Y. P., and Volchok, I. V. (1996) Interaction of bivalent copper, nickel, manganese ions with native DNA and its monomers. J. Inorg. Biochem. 63, 79-98. Blakely, W. F., Fuciarelli, A. F., Wegher, B. J., and Dizdaroglu, M. (1990) hydrogen peroxide-induced base damage in deoxyribonucleic acid. Radiat. Res. 121, 338-343. Dizdaroglu, M., Rao, G., Halliwell, B., and Gejewski, E. (1991) Damage to the DNA bases in mammalian chromatin by hydrogen peroxide in the presence of ferric and cupric ions. Arch. Biochem. Biophys. 285, 217-324. Cadet, J., D’Ham, C., Douki, T., Pouget, J.-P., Ravanat, J.-L., and Sauvaigo, S. (1998) Facts and artifacts in the measurement of oxidative base damage to DNA. Free Radical Res. 29, 541-550. Kuwabara, M., Yoon, C., Goyne, T., Thederhan, T., and Sigman, D. S. (1986) Nuclease activity of 1,10-phenathroline-copper ion: reaction with CGCGAATTCGCG and its complexes with netropsin and EcoRI. Biochemistry 25, 7401-7408. Merkel, P. B., Nilsson, R., and Kerans, D. R. (1972) Deuterium effects on singlet oxygen lifetimes in solutions. A new test of singlet oxygen reactions. J. Am. Chem. Soc. 94, 1030-1031. Li, Y., and Trush, M. A. (1993) DNA damage resulting from the oxidation of hydroquinone by copper: role for a Cu(II)/Cu(I) redox cycle and reactive oxygen generation. Carcinogenesis 14, 13031311. Kennedy, L. J., Moore, K., Jr., Caulfield, J. L., Tannenbaum, S. R., and Dedon, P. C. (1997) Quantitation of 8-oxoguanine and strand breaks produced by four oxidizing agents. Chem. Res. Toxicol. 10, 386-392. Di Mascio, P., Wefers, H., Do-Thy, H.-P., Lafleur, M. V., and Sies, H. (1989) Singlet oxygen causes loss of biological activity in plasmid and bacteriophage DNA and induces single-strand breaks. Biochim. Biophys. Acta 1007, 151-157. Frelon, S., Douki, T., Favier, A., and Cadet, J. (2002) Comparative study of base damage induced by gamma radiation and Fenton reaction in isolated DNA. J. Chem. Soc., Perkin Trans. 1, 2866-2870.

TX025650Q