Differential Effects of DNA Supercoiling onRadical-Mediated DNA Strand Breaks

William A. LaMarr,† Kathleen M. Sandman,‡ John N. Reeve,‡ andPeter C. Dedon*,†

Division of Toxicology, 56-787, Massachusetts Institute of Technology, 77 Massachusetts Avenue,Cambridge, Massachusetts 02139, and Department of Microbiology, The Ohio State University,

484 West 12th Avenue, Columbus, Ohio 43210

Received May 6, 1997X

Supercoiling is an important feature of DNA physiology in vivo. Given the possibility thatthe reaction of genotoxic molecules with DNA is affected by the alterations in DNA structureand dynamics that accompany superhelical tension, we have investigated the effect of torsionaltension on DNA damage produced by five oxidizing agents: γ-radiation, peroxynitrite, Fe2+/EDTA/H2O2, Fe2+/H2O2, and Cu2+/H2O2. With positively supercoiled plasmid DNA preparedby a recently developed technique, we compared the quantity of strand breaks produced bythe five agents in negatively and positively supercoiled pUC19. It was observed that strandbreaks produced by γ-radiation, peroxynitrite, and Fe2+/EDTA/H2O2 were insensitive to DNAsuperhelical tension. These results are consistent with a model in which chemicals thatgenerate highly reactive intermediates (e.g., hydroxyl radical), but do not interact directly withDNA, will be relatively insensitive to the changes in DNA structure and dynamics caused bysuperhelical tension. In the case of Fe2+ and Cu2+, metals that bind to DNA, only Cu2+/H2O2proved to be sensitive to DNA superhelical tension. Strand breaks produced by Cu2+/H2O2 inthe positively supercoiled substrate occurred at lower Cu concentrations than in negativelysupercoiled DNA. Furthermore, a sigmoidal Cu2+/H2O2 damage response was observed in thenegatively supercoiled substrate but not in positively supercoiled DNA. The results with Cu2+

suggest that the redox activity, DNA binding orientation, or DNA binding affinity of Cu1+ orCu2+ is sensitive to superhelical tension, while the results with the other oxidizing agentswarrant further investigation into the role of supercoiling in base damage.

Introduction

DNA damage resulting from exposure to endogenousreactive oxygen species and other radicals plays a rolein a variety of biological processes such as mutagenesis,aging, and carcinogenesis (1, 2). Even though there isconsiderable understanding of radical-mediated DNAlesions in vitro, the role of genomic organization indefining the location, quantity, and chemistry of DNAdamage remains elusive. To better understand how DNAphysiology affects DNA damage, we have examined onefeature of genomic organization, DNA superhelical ten-sion, and the role it plays in determining strand breaksproduced by γ-radiation, peroxynitrite, Fe2+/EDTA/H2O2,Fe2+/H2O2, and Cu2+/H2O2.Supercoiling is an important aspect of DNA physiology

in both prokaryotic and eukaryotic cells and appears tobe involved in gene expression (3), DNA replication (3),and DNA repair (4). While the Escherichia coli genomehas a superhelical density (σ) of ∼-0.05 (supercoils/helical turn), the genomes of Drosophila melanogasterand human cells have no net superhelical tension (5).However, as first proposed in the twin supercoileddomain model of Liu and Wang (6), the act of transcrip-tion generates localized torsional tension in genes: nega-tive superhelical tension in the 5′-ends and positivesupercoiling in the 3′-ends (6-9).

The possibility that supercoiling-induced alterations inDNA structure and dynamics could affect the interactionof oxidizing agents with DNA served as the premise forthe present studies. Torsional tension can affect DNAstructure in several ways, including changes in helicalrepeat (twist) and writhing of the DNA helix (10) and,in the case of negative supercoiling, alteration of DNAsecondary structure to form Z-DNA, cruciforms, andunpaired regions (11, 12). Although the interaction ofintercalators with negatively supercoiled DNA is welldefined (8, 13), the effect of superhelical tension on theactivity of other DNA-binding and DNA-damaging chemi-cals has not been thoroughly explored.To study the role of supercoiling in DNA damage, we

have developed a technique to prepare highly positivelysupercoiled plasmid DNA (14). The DNA has a super-helical density of ∼+0.04, while negatively supercoiledplasmid DNA, isolated from E. coli, has an averagedensity of ∼-0.05. These levels of supercoiling aresimilar to those estimated to exist in vivo (3, 15).Using positively and negatively supercoiled DNA sub-

strates, we have investigated the effect of superhelicaltension on the quantity of DNA strand breaks producedby five oxidizing agents: γ-radiation, peroxynitrite, Fe2+/EDTA/H2O2, Fe2+/H2O2, and Cu2+/H2O2. Fe and Cu arephysiologically important metals that have been impli-cated in DNA damage in vivo (e.g., see refs 16, 17). Inthe presence of hydrogen peroxide, both metals generatehydroxyl radical-like species through Fenton chemistry,which for Cu2+ appears to involve an initial reduction toCu1+ (18-20):

* Author to whom correspondence should be addressed: e-mail,pcdedon@mit.edu.

† Massachusetts Institute of Technology.‡ The Ohio State University.X Abstract published in Advance ACS Abstracts, September 15, 1997.

1118 Chem. Res. Toxicol. 1997, 10, 1118-1122

S0893-228x(97)00072-6 CCC: $14.00 © 1997 American Chemical Society

Supercoiling-dependent deformation of DNA structuremay affect the binding of these metals and thus the DNAdamage they produce. For example, Cu1+ and Cu2+ bindpreferentially to the N7 of guanine, and at least Cu2+ iscapable of forming chelation complexes with the O6 inthe same guanine (21) and with other nearby bases (21,22). This may account for the formation of 8-oxoG (23,24) and strand breaks (25, 26) at runs of guanines (27).The effects may be different for Fe since it is less redoxactive and binds to DNA with lower affinity than Cu (28).The EDTA complex of Fe2+ presents a unique op-

portunity to compare Fe bound to DNA to Fe free insolution. As with free Fe2+, Fe2+/EDTA has been shownto generate a hydroxyl radical-like species in the presenceof hydrogen peroxide (29, 30). However, unlike free Cuand Fe, the negative charge of the Fe2+/EDTA complexprevents it from binding to DNA phosphates or bases (30,31), and the resulting hydroxyl radical-like species mayexist in the fluid phase. This could account for the lackof sequence selectivity of strand breaks generated by thecomplex (30, 31). In this respect, Fe2+/EDTA may besimilar to ionizing radiation in its sensitivity to DNAsupercoiling.In addition to a minor contribution from direct interac-

tion with DNA, ionizing radiation reacts with water toproduce hydroxyl radicals and hydrogen atoms by ho-molytic fission of oxygen-hydrogen bonds and to producehydrated electrons (29, 32). Radiation-induced hydroxylradical, like that produced by Fe2+/EDTA/H2O2, causesboth strand breaks (33, 34) and base damage (35). Bothγ-radiation and Fe2+/EDTA/H2O2 are sensitive to localDNA conformation (29, 36), but the effect of superhelicaltension on radiation-induced DNA damage has beencontroversial (37-39).Nitric oxide, an important mediator of many biological

processes (40), reacts with superoxide to form peroxy-nitrite (ONOO-) with a rate constant near the diffusion-controlled limit (41). Peroxynitrite and its conjugate acid,peroxynitrous acid (ONOOH), are highly reactive atphysiological pH and are capable of producing basedamage (42, 43) and strand breaks in DNA (43, 44).However, peroxynitrite appears to have a differentreactivity than hydroxyl radical (23, 45).We have found that only the Cu2+/H2O2 system is

sensitive to DNA supercoiling. A lack of sensitivity tosupercoiling for γ-radiation, peroxynitrite, and Fe2+/EDTA is consistent with damage caused by a species thatdoes not bind to DNA prior to inducing damage, whilethe lack of effect with Fe2+/H2O2 suggests that Fe2+ bindsto DNA by a different mechanism than Cu ions.

Experimental Procedures

Chemicals. All chemicals were reagent grade. Peroxynitritesolution was generated from a reaction of ozone with sodiumazide as described by Pryor et al. (46); the product wascharacterized spectrophotometrically in 0.1 N NaOH [ε302 )1670 M-1 cm-1 (46)] and by its ability to cause the nitration ofL-tyrosine (46). Caution should be exercised in the handling ofperoxynitrite as it is a strongly oxidizing species.Preparation of Supercoiled pUC19 Substrates. Positive

supercoiling of plasmid pUC19 was achieved using a recombi-nant archaeal histone HMf (rHMf) synthesized in E. coli from

the cloned hmfB gene from Methanothermus fervidus as de-scribed elsewhere (14). The DNA had an average superhelicaldensity of ∼+0.04 (number of supercoils/helical turn), asdetermined by two-dimensional gel electrophoresis (14).Negatively supercoiled pUC19 DNA was isolated from DH5R

E. coli with an average density of ∼-0.05. This DNA wassubjected to the same treatment as the positively supercoiledDNA except for the addition of topoisomerase-containing chickenblood extract; sham reactions were instead performed with thedilution and storage buffer used with the chicken blood extract.Both plasmid substrates were quantitated using the Hoechstdye 33258 (Sigma) fluorescence quantitation assay (47).In both DNA substrates, care was taken to remove trace

quantities of metals and other contaminants that could interferewith the radical-mediated DNA damage. To remove tracequantities of nonspecifically bound, positively charged peptides,the DNA was adjusted to 2 M NaCl and incubated at roomtemperature for 2 h. The DNA was then dialyzed at 4 °Cagainst Chelex-treated phosphate buffer (50 mM, pH 7.4)containing the metal-chelating agent diethylenetriaminepen-taacetic acid and finally dialyzed against phosphate buffer aloneto remove the chelator.Treatment of Supercoiled pUC19 with DNA-Damaging

Agents. Damage reactions were performed in 25 µL reactionvolumes with 30 µg/mL pUC19 plasmid DNA at ambienttemperature for 30 min. γ-Radiation and peroxynitrite damagereactions were carried out in 50 mM potassium phosphate buffer(pH 7.4). DNA was irradiated with 0-17.6 Gy in a 60Co γ-sourceat 4 Gy/min, and the irradiated DNA was left at ambienttemperature for approximately 30 min before further processing.Peroxynitrite in 100 mMNaOH was diluted with 10 mMNaOHimmediately before addition of 1 µL aliquots to the DNAsolution. Fe2+/EDTA/H2O2, Fe2+/H2O2, and Cu2+/H2O2 damagereactions were carried out in 50 mM phosphate buffer (pH 7.4)with 1 mM H2O2. Fe2+ and Cu2+ were added to a solution ofDNA in phosphate buffer followed by addition of H2O2 to 1 mM;the Fe2+/EDTA reaction was performed as described elsewhere(48). Since oxidation of deoxyribose causes abasic sites as wellas direct strand breaks, the damaged DNA was treated withputrescine (100 mM, pH 7, 1 h, 37 °C) to convert abasic sites ofall types to strand breaks (49-52). As shown in previousstudies, putrescine treatment does not appear to react withoxidized bases to cause formation of abasic sites and strandbreaks (23).Following treatment, DNA samples were purified by passage

over G-50 Sephadex spin columns. The DNA was then relaxedwith chicken blood extract, to remove any superhelical-depend-ent biases in subsequent electrophoresis or probing; this treat-ment did not affect the number of strand breaks in the DNAsubstrates (data not shown). Finally, the DNA was purified byproteinase K digestion (100 µg/mL; 1% SDS; 2 h at 37 °C)followed by phenol/chloroform and chloroform/isoamyl alcoholextractions and ethanol precipitation.Gel Electrophoresis. DNA from each sample was resolved

on a 1.5% agarose gel (Tris-borate-EDTA) containing 60 µMchloroquine (Sigma). The gels were washed extensively in waterto remove the chloroquine and then stained with 0.5 µg/mL EtBrfor UV photography. The gels were exposed to short wave UVlight for an additional 2 min to nick the DNA and rule outsuperhelical-dependent effects on hybridization efficiency. Thegels were subsequently dried on a conventional gel dryer for 45min at ambient temperature and 45 min at 60 °C.Quantitation of Plasmid Topoisomers. Gels were probed

with a pUC19 random primer probe according to the protocolof Lueders and Fewell (53), and the radioactivity associated witheach plasmid form was determined on a phosphorimager (Mo-lecular Dynamics). The quantity of strand breaks was calcu-lated from the proportion of supercoiled DNA (form I) convertedto nicked DNA (form II) and linear DNA (form III).

Results and Discussion

Given the biological importance of DNA supercoilingand the changes in DNA conformation that accompany

2Cu2+ + H2O2 f 2Cu1+ + O2 + 2H+ or

2Cu2+ f Cu1+ + Cu3+

Cu1+/Fe2++ H2O2 f Cu2+/Fe3+ + OH- + •OH

DNA Supercoiling and Radical-Mediated DNA Damage Chem. Res. Toxicol., Vol. 10, No. 10, 1997 1119

superhelical tension, we have investigated the effect ofDNA supercoiling on the damage produced by γ-radia-tion, peroxynitrite, Fe2+/EDTA/H2O2, Fe2+/H2O2, andCu2+/H2O2. These five agents all produce DNA damageby free radical processes but differ in the mechanismsby which they associate with and damage DNA.The DNA substrates for these studies consisted of a

2686-base pair plasmid that possessed a high degree ofeither negative or positive supercoiling. The negativelysupercoiled plasmid is estimated to have σ ∼ -0.05; thisrepresents approximately -15 supercoils/plasmid (∆Lk∼-15). The positively supercoiled plasmid has an aver-age σ ∼ 0.04, with a maximal ∆Lk of +17 (14). To ensurethe removal of contaminants that could interfere withthe radical-induced DNA damage, the DNAwas subjectedto extensive dialysis against Chelex-treated phosphatebuffer containing a metal chelator. The purity of theDNA is indicated by the absence of detectable nicking ofthe plasmid DNA in the presence of 1 mM H2O2 alone(data not shown). Furthermore, the negatively super-coiled DNA was subjected to the same conditions usedto prepare the positively supercoiled DNA. Finally, whilewe show all of the data, we were careful to drawconclusions only from the strand break data that fellwithin “single-hit conditions”, which correspond to∼30-40 strand breaks in 106 bases of pUC19 DNA accordingto a Poisson distribution (49). Above this level of damage,the number of strand breaks is underestimated becauseadditional nicks in an already nicked plasmid cannot bedetected by topoisomer analysis.The results of the supercoiling studies are shown in

Figures 1 and 2. As expected for genotoxins that havelittle affinity for DNA, strand breaks produced by per-oxynitrite, Fe2+/EDTA/H2O2, and γ-radiation were foundto be insensitive to DNA superhelical tension (Figure 1).This insensitivity is likely due to the high reactivity ofthe hydroxyl radical-like species thought to be responsiblefor the DNA damage. In any case, the results indicatethat negative and positive supercoiling do not create sitesof high reactivity for these agents.The present results with radiation are consistent with

the findings of Milligan et al. (39) who observed no effectof supercoiling on radiation-induced strand breaks. How-ever, their results and the results of the present studydiffer from the contradictory results of Miller et al. (38)and Roots et al. (37). Roots et al. concluded that therewas 10-15-fold more damage in isolated salmon spermDNA compared to negatively supercoiled SV40 DNA (37),while Miller et al. observed a direct relationship betweenradiation-induced strand breaks and negative super-helical density in plasmid pIBI30 (38). In both studies,the purity of the DNA was not addressed, and in the caseof Miller et al., residual ethidium bromide from prepara-tion of supercoiled DNA could interfere with radiation-induced DNA damage. We have taken great care toensure that both supercoiled DNA substrates were freefrom contaminants that could either enhance or inhibitthe DNA damage. Furthermore, the results with theother oxidizing agents (Figures 1 and 2) lend support toa model in which chemicals that do not bind to DNA butgenerate highly reactive oxidizing species (e.g., hydroxylradical) will be relatively insensitive to the changes inDNA structure and dynamics caused by superhelicaltension.The observed lack of supercoiling effects on strand

breaks does not rule out small, local changes in thereactivity of DNA due to superhelical tension, such asthe variation in hydroxyl radical-mediated strand break

formation as a function of minor groove width (29, 36).Our results do suggest, however, that there are nosupercoiling-dependent "hotspots" for pUC19 DNA dam-age produced by Fe2+/EDTA/H2O2, γ-radiation, and per-oxynitrite.The two agents that bind to DNA, Cu and Fe, show

different sensitivities to DNA supercoiling. Fe2+/H2O2

produced an equal number of strand breaks in bothsubstrates (Figure 2), which suggests that supercoiling-dependent changes in DNA structure and dynamics donot greatly affect the binding of Fe to DNA. With Cu2+/H2O2, however, strand breaks in the positively super-coiled substrate occurred at lower Cu concentrations thanin negatively supercoiled DNA. Furthermore, the sig-moidal damage response of the negatively supercoiledsubstrate, which has been observed in other studies (23,

Figure 1. Strand breaks produced by γ-radiation, peroxy-nitrite, and Fe2+/EDTA/H2O2 are not sensitive to DNA super-coiling. Samples of negatively and positively supercoiled pUC19were treated with γ-radiation, peroxynitrite, and Fe2+/EDTA/H2O2, as described in Experimental Procedures, and gel-resolvedplasmid topoisomers were quantified by phosphorimager analy-sis of probed gels; representative experimental results are shownin the graphs. The average ratio of strand breaks in negativelysupercoiled DNA to strand breaks in positively supercoiled DNAis shown for each agent (n ) 3-5).

1120 Chem. Res. Toxicol., Vol. 10, No. 10, 1997 LaMarr et al.

54), was absent in positively supercoiled DNA (Figure2). The similarity of the damage profiles at high Cuconcentrations (>3 µM) is likely due to the fact that thelevel of strand breaks exceeds single-hit conditions.There are several explanations for the supercoiling-

dependent changes in DNA damage produced by Cu2+/H2O2. First, supercoiling may simply alter the affinityof Cu2+ or Cu1+ for DNA, with higher binding affinitycorrelated with higher levels of DNA damage. Highconcentrations of Cu2+ (equivalent to ∼50 µM under thepresent conditions) are known to denature DNA by amechanism that may involve unwinding of the helixamong other changes in DNA structure (e.g., see ref 55).This may explain the observed effects of supercoiling onCu-induced DNA damage if the formation of DNA-unwinding Cu complexes is inhibited in positively su-percoiled DNA, as occurs with intercalators in positivelysupercoiled DNA. However, even at extremely highconcentrations (g1 mM), Levy and Hecht observed thatCu2+ did not affect the topology of negatively supercoiledDNA (56), so the question of Cu-induced unwinding ofDNA remains controversial.Alternatively, if the reduction of Cu2+ to Cu1+ occurs

only with unbound Cu2+, then an inverse relationshipmay exist between Cu2+ DNA binding affinity and DNAdamage. While the present studies do not address basedamage, it has been proposed that Cu bound to high-

affinity sites on bases, probably the N7 of guanine (57),causes base lesions while unbound Cu or Cu bound tolower affinity sites (e.g., backbone phosphate) producesstrand breaks (54). In light of this model, the observedincrease in strand breaks in positively supercoiled DNAmay be accompanied by reduced base damage if positivesuperhelical tension limits access to high-affinity bindingsites on bases. Finally, it is possible that strand breaksand base damage increase simultaneously due to im-proved binding site access for both lesions.Wang and Van Ness have observed that Cu2+/ascor-

bate/H2O2 produces site-specific, negative supercoiling-dependent strand breaks in plasmid DNA (58). Ourresults suggest that such local changes in DNA reactivitycould represent one component of a complex set ofsupercoiling-dependent changes in the reaction of Cuwith DNA. It is possible that negative supercoilingcreates sites that are hyperreactive for Cu-induced strandbreak formation in the setting of an overall lowerreactivity compared to positively supercoiled DNA. Stud-ies are underway to assess the sequence selectivity ofboth base damage and strand breaks produced by Cuunder different states of DNA supercoiling.In conclusion, we have observed differential effects of

superhelical tension on the reactivity of DNAwith severaloxidizing agents. Strand breaks produced by Cu2+/H2O2

were found to occur at lower Cu concentrations inpositively supercoiled DNA, while γ-radiation, peroxyni-trite, Fe2+/EDTA/H2O2, and Fe2+/H2O2 did not showsignificant global changes in reactivity with DNA indifferent states of supercoiling. The results warrantfurther study to define the effects of superhelical tensionon both the local and global reactions of oxidants withbase and sugar moieties of DNA.

Acknowledgment. The authors are grateful to Ken-neth Moore, Jr. (MIT) for the synthesis of peroxynitrite.This work was supported by NIH grants CA64524(P.C.D.), CA72936 (P.C.D.), and GM53185 (J.N.R.); theSamuel A. Goldblith Career Development Professorship(P.C.D.); and NIEHS training grant ES07020 (W.A.L.).

References

(1) Halliwell, B., Gutteridge, J. M. C., and Cross, C. E. (1992) Freeradicals, antioxidants, and human disease: Where are we now?J. Lab. Clin. Med. 119, 598-620.

(2) Ames, B. N., Shigenaga, M. K., and Hagen, T. M. (1993) Oxidants,antioxidants, and the degenerative diseases of aging. Proc. Natl.Acad. Sci. U.S.A. 90, 7915-7922.

(3) Sinden, R. R. (1994) DNA Structure and Function, AcademicPress, San Diego.

(4) Koo, H.-S., Classen, L., Grossman, L., and Liu, L. F. (1991) ATP-dependent partitioning of the DNA template into supercoileddomains by Escherichia coliUvrAB. Proc. Natl. Acad. Sci. U.S.A.88, 1212-1216.

(5) Sinden, R. R., Carlson, J. O., and Pettijohn, D. E. (1980) Torsionaltension in the DNA double helix measured with thrimethylpso-ralen in living E. coli cells: Analogous measurements in insectand human cells. Cell 21, 773-783.

(6) Liu, L. F., and Wang, J. C. (1987) Supercoiling of the DNAtemplate during transcription. Proc. Natl. Acad. Sci. U.S.A. 84,7024-7027.

(7) Ljungman, M., and Hanawalt, P. C. (1995) Presence of negativetorsional tension in the promoter region of the transcriptionallypoised dihydrofolate reductase gene in vivo. Nucleic Acids Res.23, 1782-1789.

(8) Ljungman, M., and Hanawalt, P. C. (1992) Localized torsionaltension in the DNA of human cells. Proc. Natl. Acad. Sci. U.S.A.89, 6055-6059.

(9) Jupe, E. R., Sinden, R. R., and Cartwright, I. L. (1995) Specializedchromatin structure domain boundary elements flanking a Droso-phila heat shock gene locus are under torsional strain in vivo.Biochemistry 34, 2628-2633.

Figure 2. Strand breaks produced by Cu2+/H2O2, but not Fe2+/H2O2, are sensitive to DNA supercoiling. Samples of negativelyand positively supercoiled pUC19 were treated with Cu2+/H2O2and Fe2+/H2O2, as described in Experimental Procedures, andgel-resolved plasmid topoisomers were quantified by phos-phorimager analysis of probed gels; representative experimentalresults are shown in the graphs. The average ratio of strandbreaks in negatively supercoiled DNA to strand breaks inpositively supercoiled DNA is shown for each agent (n ) 3-5).

DNA Supercoiling and Radical-Mediated DNA Damage Chem. Res. Toxicol., Vol. 10, No. 10, 1997 1121

(10) Vologodskii, A. V., and Cozzarelli, N. R. (1994) Conformationaland thermodynamic properties of supercoiled DNA. Annu. Rev.Biophys. Biomol. Struct. 23, 609-643.

(11) Lilley, D. M. J. (1982) Dynamic and sequence-dependent DNAstructure as exemplified by cruciform extrusion from invertedrepeats in negatively supercoiled DNA.Cold Spring Harbor Symp.Quant. Biol. 47, 101-112.

(12) Wang, J., Peck, L., and Becherer, K. (1982) DNA supercoiling andits effects on DNA structure and function.Cold Spring Harb. Sym.47, 85-91.

(13) Bauer, W. R., and Gallo, R. (1990) Physical and topologicalproperties of closed circular DNA. In Chromosomes: Eukaryotic,Prokaryotic, and Viral (Adolph, K. W., Ed.) pp 87-126, CRCPress, Inc., Boca Raton, FL.

(14) LaMarr, W. A., Sandman, K. M., Reeve, J. N., and Dedon, P. C.(1997) Large scale preparation of positively supercoiled DNAusing the archaeal histone HMf. Nucleic Acids Res. 25, 1660-1662.

(15) Lee, M.-S., and Garrard, W. T. (1991) Positive DNA supercoilinggenerates a chromatin conformation characteristic of highly activegenes. Proc. Natl. Acad. Sci. U.S.A. 88, 9675-9679.

(16) Ozawa, T., Ueda, J.-i., and Shimazu, Y. (1993) DNA single strandbreakage by copper(II) complexes and hydrogen peroxide atphysiological conditions. Biochem. Mol. Biol. Int. 31, 455-461.

(17) Imlay, J. A., and Linn, S. (1988) DNA damage and oxygen radicaltoxicity. Science 240, 1302-1309.

(18) Burkitt, M. J. (1994) Copper-DNA adducts. Methods Enzymol.234, 66-79.

(19) Prutz, W. A., Butler, J., and Land, E. J. (1990) Interaction ofcopper(I) with nucleic acids. Int. J. Radiat. Biol. 58, 215-234.

(20) Halliwell, B., and Gutteridge, J. M. C. (1989) Free Radicals inBiology and Medicine, Clarendon Press, Oxford.

(21) Zimmer, C., Luck, G., Fritzsche, H., and Triebel, H. (1971) DNA-copper(II) complex and the DNA conformation. Biopolymers 10,441-463.

(22) Geierstanger, B. H., Kagawa, T. F., Chen, S.-L., Quigley, G. J.,and Ho, P. S. (1991) Base-specific binding of copper(II) to Z-DNA.The 1.3-Å single crystal structure of d(m5CGUAm5CG) in thepresence of CuCl2. J. Biol. Chem. 266, 20185-20191.

(23) Kennedy, L. J., Moore, K., Jr., Caulfield, J. L., Tannenbaum, S.R., and Dedon, P. C. (1997) Quantitation of 8-oxoguanine andstrand breaks produced by four oxidizing agents. Chem. Res.Toxicol. 10, 386-392.

(24) Aruoma, O. I., Halliwell, B., Gajewski, E., and Dizdaroglu, M.(1991) Copper-ion-dependent damage to the bases in DNA in thepresence of hydrogen peroxide. Biochem. J. 273, 601-604.

(25) Sagripanti, J. L., and Kraemer, K. H. (1989) Site-specific oxidativeDNA damage at polyguanosines produced by copper plus hydro-gen peroxide. J. Biol. Chem. 264, 1729-1734.

(26) Rodriguez, H., Drouin, R., Holmquist, G. P., O'Connor, T. R.,Boiteux, S., Laval, J., Doroshow, J. H., and Akman, S. A. (1995)Mapping of copper/hydrogen peroxide-induced DNA damage atnucleotide resolution in human genomic DNA by ligation-medi-ated polymerase chain reaction. J. Biol. Chem. 270, 17633-17640.

(27) Tkeshelashvili, L. K., McBride, T., Spence, K., and Loeb, L. A.(1991) Mutation spectrum of copper-induced DNA damage. J.Biol. Chem. 266, 6401-6406.

(28) Dizdaroglu, M., Rao, G., Halliwell, B., and Gajewski, E. (1991)Damage to DNA bases in mammalian chromatin by hydrogenperoxide in the presence of ferric and cupric ions. Arch. Biochem.Biophys. 285, 317-324.

(29) Dixon, W. J., Hayes, J. J., Levin, J. R., Weidner, M. F., Dom-browski, B. A., and Tullius, T. D. (1991) Hydroxyl radicalfootprinting. Methods Enzymol. 208, 380-413.

(30) Tullius, T. D. (1987) Chemical snapshots of DNA: using hydroxylradical to study the structure of DNA and DNA-protein com-plexes. Trends Biochem. Sci. 12, 297-300.

(31) Celender, D. W., and Cech, T. R. (1990) Iron(II)-ethylenediamine-tetraacetic acid catalyzed cleavage of RNA and DNA oligo-nucleotides: similar reactivity toward single- and double-strandedforms. Biochemistry 29, 1355-1361.

(32) von Sonntag, C. (1987) The Chemical Basis of Radiation Biology,Taylor & Francis, New York.

(33) Yamamoto, O., Fuji, I., and Ogawa, M. (1985) Difference in DNAstrand break by gamma- and beta-irradiations: an in vitro study.Biochem. Int. 11, 217-223.

(34) Henle, E. S., Roots, R., Holley, W. R., and Chatterjee, A. (1995)DNA strand breakage is correlated with unaltered base releaseafter gamma irradiation. Radiat. Res. 143, 144-150.

(35) Dizdaroglu, M. (1985) Formation of an 8-hydroxyguanine moietyin deoxyribonucleic acid on gamma-irradiation in aqueous solu-tion. Biochemistry 24, 4476-4481.

(36) Hayes, J. J., Kam, L., and Tullius, T. D. (1990) Footprintingprotein-DNA complexes with γ-rays.Methods Enzymol. 186, 545-549.

(37) Root, R., Henle, E., Holley, W. R., and Chatterjee, A. (1991)Measurements of nucleic bases released after gamma irradiationof DNA in solution in air. Radiat. Res. 125, 288-292.

(38) Miller, J. H., Nelson, J. M., Ye, M., Swenberg, C. E., Speicher, J.M., and Benham, C. J. (1991) Negative supercoiling increases thesensitivity of plasmid DNA to single-strand break induction byX-rays. Int. J. Radiat. Biol. 59, 941-949.

(39) Milligan, J. R., Arnold, A. D., and Ward, J. F. (1992) The effect ofsuperhelical density on the yield of single-strand breaks inγ-irradiated plasmid DNA. Radiat. Res. 132, 69-73.

(40) Schmidt, H. H. H. W., and Walter, U. (1994) NO at work. Cell78, 919-925.

(41) Beckman, J. S., Beckman, T. W., Chen, J., Marshall, P. A., andFreeman, B. A. (1990) Apparent hydroxyl radical production byperoxynitrite: Implications for endothelial injury from nitric oxideand superoxide. Proc. Natl. Acad. Sci. U.S.A. 87, 1620-1624.

(42) Yermilov, V., Rubio, J., and Ohshima, H. (1995) Formation of8-nitroguanine in DNA treated with peroxynitrite in vitro andits rapid removal from DNA by depurination. FEBS Lett. 376,207-210.

(43) Inoue, S., and Kawanishi, S. (1995) Oxidative DNA damageinduced by simultaneous generation of nitric oxide and superox-ide. FEBS Lett. 371, 86-88.

(44) Salgo, M. G., Stone, K., Squadrito, G. L., Battista, J. R., and Pryor,W. A. (1995) Peroxynitrite causes DNA nicks in plasmid pBR322.Biochem. Biophys. Res. Commun. 210, 1025-1030.

(45) Goldstein, S., Squadrito, G. L., Pryor, W. A., and Czapski, G.(1996) Direct and indirect oxidations by peroxynitrite, neitherinvolving the hydroxyl radical. Free Radical. Biol. Med. 21, 965-974.

(46) Pryor, W. A., Cueto, R., Jin, X., Koppenol, W. H., Ngu-Schwemlein,M., Squadrito, G. L., Uppu, P. L., and Uppu, R. M. (1995) Apractical method for preparing peroxynitrite solutions of low ionicstrength and free of hydrogen peroxide. Free Radical Biol. Med.18, 75-83.

(47) Cesarone, C. F., Bolognesi, C., and Santi, L. (1979) Improvedmicrofluorometric DNA determination in biologic material using33258 Hoechst. Anal. Biochem. 100, 188-197.

(48) Price, M. A., and Tullius, T. D. (1992) Using hydroxyl radical toprobe DNA structure. Methods Enzymol. 212, 194-218.

(49) Dedon, P. C., Salzberg, A. A., and Xu, J. (1993) Exclusiveproduction of bistranded DNA damage by calicheamicin. Bio-chemistry 32, 3617-3622.

(50) Yu, L., Golik, J., Harrison, R., and Dedon, P. (1994) Thedeoxyfucose-anthranilate of esperamicin A1 confers intercalativeDNA binding and causes a switch in the chemistry of bistrandedDNA lesions. J. Am. Chem. Soc. 116, 9733-9738.

(51) Povirk, L. F., and Houlgrave, C. W. (1988) Effect of apurinic/apyrimidinic endonucleases and polyamines on DNA treated withbleomycin and neocarzinostatin: specific formation and cleavageof closely opposed lesions in complementary strands. Biochemistry27, 3850-3857.

(52) Lindahl, T., and Andersson, A. (1972) Rate of chain breakage atapurinic sites in double-stranded deoxyribonucleic acid. Biochem-istry 11, 3618-3623.

(53) Lueders, K. K., and Fewell, J. W. (1994) Hybridization of DNAin dried gels provides increased sensitivity compared withhybridization to blots. BioTechniques 16, 66-67.

(54) Drouin, R., Rodriguez, H., Gao, S.-W., Gebreyes, Z., O'Connor, T.R., Holmquist, G. P., and Akman, S. A. (1996) Cupric ion/ascorbate/hydrogen peroxide-induced DNA damage: DNA-boundcopper ion primarily induces base modifications. Free RadicalBiol. Med. 21, 261-273.

(55) Liebe, D. C., and Stuehr, J. E. (1972) Copper(II)-DNA denatur-ation. I. Concentration dependence of melting temperature andterminal relaxation time. Biopolymers 11, 145-166.

(56) Levy, M. J., and Hecht, S. M. (1988) Copper(II) facilitatesbleomycin-mediated unwinding of plasmid DNA. Biochemistry 27,2647-2650.

(57) Kagawa, T. F., Geierstanger, B. H., Wang, A. H., and Ho, P. S.(1991) Covalent modification of guanine bases in double-strandedDNA. The 1.2-A Z-DNA structure of d(CGCGCG) in the presenceof CuCl2. J. Biol. Chem. 266, 20175-20184.

(58) Wang, Y., and Van Ness, B. (1989) Site-specific cleavage ofsupercoiled DNA by ascorbate/Cu(II). Nucleic Acids Res. 17,6915-6926.

TX970072C

1122 Chem. Res. Toxicol., Vol. 10, No. 10, 1997 LaMarr et al.