A Single Cyclic p-Benzoquinone Adduct Can Destabilize a DNA

J. Sági, A. Chenna, B. Hang, and B. Singer*. Donner Laboratory, Lawrence Berkeley National Laboratory, University of California, Berkeley, California...
0 downloads 0 Views 94KB Size
Download PDF
Chem. Res. Toxicol. 1998, 11, 329-334

329

A Single Cyclic p-Benzoquinone Adduct Can Destabilize a DNA Oligonucleotide Duplex J. Sa´gi,† A. Chenna, B. Hang, and B. Singer* Donner Laboratory, Lawrence Berkeley National Laboratory, University of California, Berkeley, California 94720 Received October 31, 1997

p-Benzoquinone (p-BQ), a stable metabolite of the human carcinogen benzene, forms tworing benzetheno exocyclic base adducts with C, A, and G bases in DNA. As a part of a project for studying the biological effect of the p-BQ adducts, we report here on the first biophysical characterization of oligodeoxyribonucleotide duplexes containing a single site-specific p-BQC, using thermal denaturation and circular dichroism (CD). We find that the thermal and thermodynamic stabilities of the control duplex are reduced by p-BQ-C. The Tm value decreases by 12.6 °C at the duplex concentration of 1.5 µM and the ∆G° by 10.2 kcal/mol. The latter was determined from the concentration dependence of the Tm values. The destabilization has little dependence on the nature of the opposite base. This reduction is higher than that of the single base mismatches studied (-4.9 to -7.9 kcal/mol) and is close to that observed with an adjacent double mismatch-containing duplex (-11.3 kcal/mol). The overall B-conformation of the duplex with a p-BQ-C is, however, only slightly altered, relative to the parent duplex, as shown by CD spectra. The p-BQ-C-containing duplex has been found recently to be a good substrate for the major human AP endonuclease as compared to an abasic site-containing duplex [Hang, B., et al. (1997) Biochemistry 36, 15411-15418]. We now find that the thermodynamic properties and the localized conformational changes of a p-BQ-C-containing duplex are apparently related to those reported for a duplex containing an abasic site.

Introduction Benzene is classified as a human carcinogen that has been shown to cause hematological disorders. The major effect is acute myeloid leukemia (1). Benzene is not only an occupational hazard but is also present in cigarette smoke, gasoline, and automobile exhaust. Therefore, a large population is continuously exposed to it (1-3). Pathways and metabolites responsible for the genotoxic effects have not yet been fully explored. In one pathway benzene is metabolized by cytochrome P450 to benzene oxide which is further converted to various derivatives. One biologically important major metabolite is p-benzoquinone (p-BQ1), an oxidation product of p-hydroquinone (4-7). p-BQ is also a metabolite of the widely used drug acetaminophen (8). Reaction of p-BQ with DNA in vitro has been shown to result in the formation of three exocyclic benzetheno base derivatives (9). More recently, the 2′-deoxyribose derivatives of these exocyclic adducts, p-BQ-C, p-BQ-A, and p-BQ-G, have been synthesized on a larger scale, converted to the phosphoramidite derivatives, and incorporated into defined 25-nucleotide-long oligodeoxyribonucleotides (10, * To whom correspondence should be addressed. Tel: (510) 6420637. Fax: (510) 486-6488. † Permanent address: Central Research Institute for Chemistry, Hungarian Academy of Sciences, P.O. Box 17, H-1525 Budapest, Hungary. 1 Abbreviations: p-BQ, p-benzoquinone; p-BQ-C, 3′′-hydroxy-3,N4benzethenocytosine; p-BQ-A, 3′′-hydroxy-1,N6-benzethenoadenine; pBQ-G, 3′′-hydroxy-1,N2-benzethenoguanine; AP, abasic or apurinic/ apyrimidinic; HAP1, human AP endonuclease; A, 1,N6-ethenoadenine; CD, circular dichroism.

11). The p-BQ-C-containing duplex was found to be a substrate for the human AP endonuclease (HAP1), as well as for Escherichia coli exonuclease III and endonuclease IV (12). A new, unusual mechanism for cleavage by a human AP endonuclease was described for this oligonucleotide, in which the adduct remained attached to the 5′ terminus of the 3′ fragment (12). While the AP site is the preferred substrate for HAP1, cleavage of the p-BQ-C oligonucleotides requires the same catalytic center as the AP site as determined from parallel repair data using six mutant HAP1 proteins (13). The effects on repair of p-BQ-C are likely to contribute to the mutagenic potential of benzene and may be related to the effect of p-BQ-C on the secondary structure and conformation of the oligonucleotide duplex. To gain insight into the structural properties of these 25-mer duplexes, we initiated this study. In addition, duplexes containing mismatches of normal bases were also examined for changes in the thermal and thermodynamic stabilities and the global conformation.

Materials and Methods Oligodeoxyribonucleotides. Both the unmodified and the p-BQ-C adduct-containing 25-mer oligodeoxyribonucleotides used in this work have been synthesized, purified, and analyzed earlier (10) for biological studies (12, 13). The concentration of the single strands was calculated from the absorbance and the sequence by using the DNA/Oligo Quantitation Software of a Beckman DU 7400 diode-array spectrophotometer. Contribution of the single p-BQ-C adduct was taken into account as that of a purine (A) nucleotide. The double strands were constructed by mixing equimolar amounts of the non-self-complementary single strands and are listed in Table 1.

S0893-228x(97)00199-9 CCC: $15.00 © 1998 American Chemical Society Published on Web 03/07/1998

330 Chem. Res. Toxicol., Vol. 11, No. 4, 1998

Sa´ gi et al.

Table 1. Effect of p-BQ-C and Normal Base Mismatches on the Thermal and Thermodynamic Stability of the Oligodeoxyribonucleotide Double Helices

a Complete sequence of the 25-mer control duplex. b Partial sequence of the 25-mer control duplex. The C* stands for p-BQ-C. C* is in position 8 except for duplex 3 in which C* is in position 14. Single base mismatches are also in position 8. In duplex 11, the CC replaces GT in positions 11 and 12. c The Tm values of duplexes 4-6 and 10 were measured at 1.5 µM duplex concentration. The Tm values at 1.5 µM for the other duplexes are taken from the 1/Tm-ln Ct plots (Figure 3).

Thermal Melting Experiments. Thermal transition profiles were measured in the above Beckman spectrophotometer equipped with the Beckman normal 6-cell holder with internal thermometer and the Peltier temperature controller. The buffer used was 0.1 M NaCl, 0.01 M sodium phosphate (pH 7.0), and 0.1 mM EDTA in order to measure the effects under a nearphysiological condition. Cuvettes of 1- and 0.1-cm path length were used that contained 0.4- and 0.1-mL samples, respectively. Silicon oil (0.96 g/mL and 200 centistokes; Sigma, St. Louis, MO) was layered on top of the solutions to avoid evaporation. Linear heating was used from 20 to 100 °C. Ramp rate was 0.2 °C/ min in the range of (20 °C of the Tm of the samples and 0.5 °C/min at the rest. Absorption values were collected at 0.5 °C intervals at 260 or 280 nm. The 320- or 350-nm values collected simultaneously were used for background correction for the natural duplexes and the p-BQ-C-containing samples, respectively.

Thermodynamic data were obtained from the concentration dependence of the Tm values (14). Duplex concentrations (Ct) ranged from 1 to 75 µM. For annealing, the 75 µM solution was heated to 90 °C for 3 min, allowed to cool to 20 °C, and subsequently diluted with the buffer to the duplex concentrations used. The Tm value is defined as the maximum of the firstderivative curve calculated with the use of the Beckman Tm analysis software from the smoothed absorption versus temperature profiles. These values were shown to be within 1 °C of those determined at one-half of the total hyperchromicity after baseline correction (15-17). Thermodynamic values listed in Table 1 are the averages of two to three determinations from separate samples except for that of the unmodified, control duplex which is the average of five samples. The van’t Hoff transition enthalpy (∆H°) and entropy (∆S°) change values were calculated from the least-squares-fitted slope and intercept

p-BQ-C-Induced Structural Change in DNA

Figure 1. Structure of the p-benzoquinone-derived benzetheno exocyclic adduct of cytosine (C) base. values of the 1/Tm versus ln Ct plots, based on the following equation for bimolecular association of two non-self-complementary strands (14): 1/Tm ) (R/∆H°) ln Ct + (∆S° - R ln 4)/ ∆H°. The transition free-energy change at 25 °C, ∆G°25, was calculated by the standard thermodynamic equation: ∆G° ) ∆H° - T∆S°. The mean value of reproducibility of the Tm values was (0.5 °C; thus that of the ∆Tm’s is (1 °C. Reproducibility of the ∆H°, ∆S°, and ∆G° values was 3-6%. In the case of the duplexes containing mismatch bases, A, T, and C, opposite p-BQ-C only the Tm values at 1.5 µM duplex concentration were determined. Therefore, all Tm values are compared in Table 1 at this concentration. Circular Dichroism. CD spectra were recorded in a Jasco J-600 spectropolarimeter interfaced to an IBM PC, at room temperature in a 1-cm path length cuvette of 0.7 mL. Oligodeoxyribonucleotide duplexes of 2-5 µM total strand concentrations were prepared by equimolar mixing and annealing under the conditions used for the Tm measurements. Absorption values at 260 nm ranged from 0.45 to 1.0. Repeated scans were recorded with 20 nm/min, the time constant was 2 s, and the bandwidth was 1 nm. The spectra of the duplexes were corrected for the buffer.

Results Thermal Melting Studies. Both the duplexes containing unmodified bases, with or without a mismatch, and those containing a single site-specifically inserted p-BQ-C adduct (Figure 1) exhibit monophasic absorption-temperature melting profiles (Figure 2). The 1/Tmln Ct plots were determined from 1 to 75 µM with 7 of the 11 duplexes studied. The remaining four structures were measured at Ct ) 1.5 µM. Therefore, Tm and ∆Tm values of all samples were compared at Ct ) 1.5 µM in Table 1. The duplexes containing a single site-specifically incorporated p-BQ-C base in one strand show significantly decreased Tm values (Table 1, duplexes 2, 3) as

Chem. Res. Toxicol., Vol. 11, No. 4, 1998 331

compared to the unmodified duplex (Table 1, duplex 1). The effect on the Tm of p-BQ-C placed into different sequence environments is the same, -12.6 °C. When any of the other three natural bases was incorporated opposite the p-BQ-C in position 8 (Table 1, duplexes 4-6) the destabilization effects are very similar to each other and values are within the experimental error. The ∆Tm’s are -10.8 to -11.6 °C and are close to that measured with G opposite p-BQ-C, -12.6 °C (Table 1, duplex 2). Single mismatches with unmodified C in position 8 (Table 1, duplexes 7-9) also destabilize duplexes to varying extents, depending on the nature of the mismatch. Only the double mispair (Table 1, duplex 11) decreases the Tm more than p-BQ-C. Figure 3 shows the 1/Tm-ln Ct plots of the oligonucleotide duplexes used. All duplexes show a concentration dependence of the Tm values indicating that molecularity of the transitions is greater than 1. Figure 3 also shows that all plots are linear in the concentration range studied (1-75 µM). Parameters characterizing energetics of the thermal transition, enthalpy, entropy, and freeenergy change values are in Table 1. Data show that the p-BQ-C adducts reduce both the transition enthalpy and entropy (Table 1, duplexes 2, 3). The resulting decreases in free-energy changes are large (-10.2 kcal/ mol, as an average) and are close to the ∆∆G° value observed with the duplex containing the double mispair (-11.3 kcal/mol). Single base mispairs reduce enthalpy and entropy less (the ∆∆G° values are -4.9 to -7.9 kcal/ mol). The observed ∆∆G° values follow the trend of the ∆Τm’s. CD Profiles. The effect of a p-BQ-C adduct on duplex conformation was studied by CD spectra, as shown in Figure 4. The CD profiles are only slightly altered when compared to that of the unmodified duplex. A p-BQ-C base decreases the positive maximum value of the control and causes a 1-2-nm red shift of both the positive λmax and the through values (panel A). The λmin values are not affected and the negative maxima are not changed in the same way by the p-BQ-C in different positions. The main change with the C‚A and the C‚C mismatches was the increase of the positive maximum value of the control duplex (panel B). Changing two base pairs of the control duplex, the GT/AC pairs in positions 11 and 12 of duplex 1 (Table 1) for CC/GG pairs, gives a similar alteration in the spectrum (panel C) as the adjacent C‚A,C‚C double mismatch which increases the positive maximum and decreases the negative, compared to the control duplex.

Figure 2. Absorption (260 nm) versus temperature melting profiles of 25-mer oligodeoxyribonucleotide duplexes of Ct ∼ 1.5 µM in 0.1 M NaCl, 0.01 M sodium phosphate (pH 7.0), and 0.1 mM EDTA. The numbers in the figure refer to the duplexes listed in Table 1: 1, unmodified control; 2, duplex containing a single p-BQ-C in position 8.

332 Chem. Res. Toxicol., Vol. 11, No. 4, 1998

Sa´ gi et al.

Figure 3. 1/Tm versus ln Ct plots, used for the calculation of thermodynamic parameters of the thermally induced helix-to-coil transitions (Ct is the total strand concentration of the duplex). The numbers in the figure refer to the duplexes listed in Table 1: 11, duplex with a C‚A,C‚C adjacent double mismatch; 2, p-BQ-C adduct in position 8; 3, p-BQ-C adduct in position 14; 9, C‚C mismatch; 8, C‚T mismatch; 7, C‚A mismatch; 1, unmodified control duplex.

Discussion Thermal Stability. Benzene is metabolized by one pathway to p-BQ (4-7) which forms benzetheno exocyclic adducts with C, A, and G bases in DNA (9). One facet of the current work is to facilitate the understanding of repair and mutagenic mechanisms resulting from p-BQ adducts (12, 13). In this paper we present the results of a biophysical investigation which is, to our knowledge, the first study which describes the effect of a p-BQ adduct on the structural and conformational properties of an oligodeoxyribonucleotide duplex. For this investigation, duplexes of the same sequence and length were used as for biochemical experiments (12, 13). The 25-mer duplexes containing a site-specifically incorporated single p-BQ-C in one strand show significantly decreased Tm values as compared to that of the unmodified duplex. Thermal destabilization is -12.6 °C in buffered 0.1 M NaCl (pH 7.0) (Table 1). When each of the four natural bases is inserted opposite p-BQ-C in position 8, destabilization is almost as great with an A, T, or C as with G opposite, respectively (Table 1, duplexes 2, 4-6). Thus, the effect on stability has little dependence on the nature of the opposite base. This is not surprising in the light of the impaired base recognition property through hydrogen bonding, but unexpected if stacking is considered. A similar finding was described for 1,N2-propanoguanine by Plum et al. (18). Alternatively, 1,N6-ethenoadenine (A) was found by Leonard et al. (19) to have a preferred opposite base, G, in syn conformation instead of the anti form in the B-DNA, with one-half the -∆Tm value than when placed opposite the other three bases. The A‚G(syn) pair has been shown to be connected via three non-Watson-Crick-type hydrogen bonds (19). The destabilizing effect of p-BQ-C with any of the four bases opposite is higher than that of the same mismatches opposite C or the T‚G wobble pair (Table 1,

duplexes 7-10). It is well-known that base mismatches generally destabilize DNA (20-24), except for a few specific pairs, such as the GA‚GA tandem double mispair (25). The single mismatches constructed here also decrease the Tm and ∆G°, and the magnitude of these effects depends on the nature of the mismatch. Thermodynamic Stability. The equation used for calculating the enthalpy and entropy values from the 1/Tm versus ln Ct plots (see Materials and Methods) is most appropriate for duplexes composed of 12-nucleotidelong or shorter single strands, which can have purely bimolecular, two-state (duplex to random coil) transitions (14). However, the equation is also described to be applicable for longer duplexes that may not have strictly two-state transitions, such as 14-18-mer duplexes (15, 26, 27) and also 21-25-mer duplexes (27-29). Among the apparent criteria of the two-state model during melting are the monophasic melting curves and the linearity of the 1/Tm versus ln Ct plots (15, 28, 30). The 25-mer duplexes studied here showed monophasic melting transitions (Figure 2), and the 1/Tm versus ln Ct plots were linear (Figure 3). Thus, we used the same 25-mer duplex structures for the thermodynamic investigations that were used for the biochemical studies (12, 13). We also believe that the two-state model can provide a reasonable approximation for the differential thermodynamic parameters of duplexes that may not have strictly two-state transition resulting from the length (15, 2629). The Tm values measured at Ct ) 1.5 µM are in good correlation with the free-energy changes, ∆G°, determined from the 1/Tm-ln Ct plots. The significant decrease of the ∆G°25 of the control duplex on the effect of a p-BQ-C adduct originates from the decrease of both the enthalpy and the entropy (Table 1). The lower transition enthalpy refers to the loss of base stacking and pairing. The lower entropy reflects a less constrained structure

p-BQ-C-Induced Structural Change in DNA

Figure 4. CD curves of the unmodified control, the p-BQ-C adduct, and the mismatches-containing duplexes in 0.1 M NaCl, 0.01 M sodium phosphate (pH 7.0), and 0.1 mM EDTA at room temperature. The control duplex (Table 1, duplex 1) is shown as a solid line (s). Panel A: - - -, the duplex containing p-BQ-C in position 8 (Table 1, duplex 2); -‚-, duplex containing p-BQ-C in position 14 (Table 1, duplex 3). Panel B: - - -, C‚A mismatch (Table 1, duplex 7); -‚-, C‚C mismatch (Table 1, duplex 9). Panel C: -‚-, C‚C,C‚A adjacent double mismatchcontaining duplex (Table 1, duplex 11); - - -, and a second control to this, containing C‚G,C‚G base pairs in positions 11 and 12.

of the duplex around the adduct. Enthalpy-entropy compensation effects are observed with all duplexes (not shown), and these effects have been described as a common phenomenon (31-33). The differential thermodynamic values observed with the p-BQ-C-containing duplexes (Table 1) may reflect conformational changes that affect more than five base pairs. This is in apparent contradiction with CD profiles which show only slightly altered global conformation. The different single base mispairs studied here reduce ∆G° much less than the p-BQ-C adduct. The C‚A mismatch, as would be predicted (34), causes the smallest changes, the C‚C the largest. On the basis of the differential free energy change, ∆∆G°, the p-BQ-C adduct has a large negative effect on duplex stability that is 60% higher than that of the average of the different single base mismatches studied, and the impact on the thermodynamic stability is similar to that of an adjacent double mispair.

Chem. Res. Toxicol., Vol. 11, No. 4, 1998 333

CD Profiles. The effect of the p-BQ-C adduct on duplex conformations was evaluated from CD spectra, as shown in Figure 4, panel A. Compared to the spectrum of the unmodified control duplex, all the alterations seen in the spectra of the p-BQ-C-containing duplexes are small and similar in extent to those caused by the different single base mismatches or the double base mismatch (panels B and C), although the direction of the changes is opposite on the positive side. Despite the significant thermodynamic destabilization of the duplexes by p-BQ-C, the CD spectra show only slight modifications of the overall B-conformation of the control duplex. However, there may be larger localized conformational changes at or around the adduct. Implications for Endonuclease Repair. The p-BQC-containing 25-mer duplex was found to be a substrate for the human AP endonuclease (HAP1) (12, 13). This is unexpected since the known substrates for this enzyme are structurally unrelated to p-BQ-C and are similar to an AP site (35). The preferred substrate for HAP1 is an AP site-containing duplex (13, 35). A greatly reduced thermal and thermodynamic stability (∆∆G°25 ) -6.5 kcal/mol) was described earlier for an AP site-containing 13-mer duplex by Vesnaver et al. (21). Here we find a larger decrease of ∆G°25 (-10.2 kcal/mol, as an average) for the unusual substrate, the p-BQ-C-containing 25-mer duplex. Both the AP site (21) and the p-BQ-C exert a much greater negative effect on the duplex thermodynamic stability than can be predicted by calculation from the loss of the nearest-neighbor interactions, since with both the AP site (36, 37) and the p-BQ-C only localized conformational changes are observed. Thus, the common features known at present of the two structurally unrelated substrates of HAP1 are that the duplex stability is greatly reduced and that conformational changes are localized around the lesion in both cases.

Conclusion In this study, a single p-BQ adduct of C was sitespecifically inserted into two different sequence positions of a 25-nucleotide-long oligodeoxyribonucleotide that was annealed to a normal complement. The p-BQ-C significantly reduces the thermal and thermodynamic stabilities of the parent double helix regardless of the opposite base. Single normal base mismatches are much less destabilizing, and the effect of p-BQ-C is similar to that of an adjacent double mismatch. The overall B-conformation of the duplex is, however, only slightly altered by the adduct, as determined by circular dichroism. Thus, the main finding of this study is that incorporation of p-BQ-C in a duplex yields an energetically destabilized but only locally altered structure. The two structurally unrelated substrates of the human AP endonuclease, an AP site- and a p-BQ-C-containing duplex, are apparently thermodynamically related, and effects of the lesions on conformation are localized in both cases.

Acknowledgment. This work was supported by NIH Grant CA 47723 and Grant ES 07363 (to B.S.) and was administered by the Lawrence Berkeley National Laboratory under Department of Energy Contract DE-AC0376SF00098. The authors thank Dr. I. Tinoco, Jr., for the use of his CD facility and Dr. Barbara Dengler for technical assistance.

334 Chem. Res. Toxicol., Vol. 11, No. 4, 1998

References (1) IRAC. (1987) IARC Monographs on the Evaluation of the Carcinogenic Risks to Humans, Suppl. 7, International Agency for Research on Cancer, Lyon, France. (2) Goldstein, B. D. (1977) Benzene toxicity: a critical evaluation: hematotoxicity in humans. J. Toxicol. Environ. Health Suppl. 2, 69-105. (3) Wallace, L. (1996) Environmental exposure to benzene: an update. Environ. Health Perspect. 104, 1129-1136. (4) Guengerich, F. P., and Macdonald, T. L. (1990) Mechanism of cytochrome P-450 catalysis. FASEB J. 4, 2453-2459. (5) Guengerich, F. P., Kim, D. H., and Iwasaki, M. (1991) Role of human cytochrome P-450 IIE1 in the oxidation of many low molecular weight cancer suspects. Chem. Res. Toxicol. 4, 168179. (6) Snyder, R., Jowa, L., Witz, G., Kalf, G., and Rushmore, T. (1987) Formation of reactive metabolites from benzene. Arch. Toxicol. 60, 61-64. (7) Snyder, R., and Hedli, C. C. (1996) An overview of benzene metabolism. Environ. Health Perspect. 104, 1165-1171. (8) Pascoe, G. A., Calleman, C. J., and Baillie, T. A. (1988) Identification of S-(2,5-dihydroxyphenyl)-cysteine and S-(2,5-dihydroxyphenyl)-N-acetyl-cysteine as urinary metabolites of acetaminophen in the mouse. Evidence for p-benzoquinone as a reactive intermediate in acetaminophen metabolism. Chem.-Biol. Interact. 68, 85-98. (9) Bodell, W. J., Levay, G., and Pongracz, K. (1993) Investigation of benzene-DNA adducts and their detection in human bone marrow. Environ. Health Perspect. 99, 241-244. (10) Chenna A., and Singer, B. (1995) Large scale synthesis of p-benzoquinone-2′-deoxycytidine and p-benzoquinone-2′-deoxyadenosine adducts and their site-specific incorporation into DNA oligonucleotides. Chem. Res. Toxicol. 8, 865-874. (11) Chenna, A., and Singer, B. (1997) Synthesis of a benzene metabolite adduct, 3′′-hydroxy-1,N2-benzetheno-2′-deoxyguanosine, and its site-specific incorporation into DNA oligonucleotides. Chem. Res. Toxicol. 10, 165-171. (12) Hang, B., Chenna, A., Fraenkel-Conrat, H., and Singer, B. (1996) An unusual mechanism for the major human apurinic/apyrimidinic (AP) endonuclease involving 5′ cleavage of DNA containing a benzene-derived exocyclic adduct in the absence of an AP site. Proc. Natl. Acad. Sci. U.S.A. 93, 13737-13741. (13) Hang, B., Rothwell, D. G., Sagi, J., Hickson, I. D., and Singer, B. (1997) Evidence for a common active site for cleavage of an AP site and the benzene-derived exocyclic adduct, 3,N4-benzethenodC, in the major human AP endonuclease. Biochemistry 36, 15411-15418. (14) Marky, L. A., and Breslauer, K. J. (1987) Calculating thermodynamic data for transitions of any molecularity from equilibrium melting curves. Biopolymers 26, 1601-1620. (15) Otokiti, E. O., and Sheardy, R. D. (1997) Effect of base pair A/C and G/T mismatches on the thermal stabilities of DNA oligomers that form B-Z junctions. Biochemistry 36, 11419-11427. (16) Bolli, M., Trafelet, H. U., and Leumann, C. (1996) Watson-Crick base-pairing properties of bicyclo-DNA. Nucleic Acids Res. 24, 4660-4667. (17) Gaffney, B. L., and Jones, R. A. (1989) Thermodynamic comparison of the base pairs formed by the carcinogenic lesion O6methylguanine with reference both to Watson-Crick pairs and to mismatched pairs. Biochemistry 28, 5881-5889. (18) Plum, G. E., Grollman, A. P., Johnson, F., and Breslauer, K. J. (1992) Influence of an exocyclic guanine adduct on the thermal stability, conformation, and melting thermodynamics of a DNA duplex. Biochemistry 31, 12096-12102. (19) Leonard, G. A., McAuley-Hecht, K. E., Gibson, N. J., Brown, T., Watson, W. P., and Hunter, W. N. (1994) Guanine-1,N6-ethenoadenine base pairs in the crystal structure of d(CGCGAATT(A)GCG). Biochemistry 33, 4755-4761.

Sa´ gi et al. (20) Plum, G. E., and Breslauer, K. J. (1994) DNA lesions. A thermodynamic perspective. In: DNA Damage, Effects on DNA Structure and Protein Recognition. Ann. N. Y. Acad. Sci. 726, 45-56. (21) Vesnaver, G., Chang, C.-N., Eisenberg, M., Grollman, A. P., and Breslauer, K. J. (1989) Influence of abasic and anucleoside sites on the stability, conformation, and melting behavior of a DNA duplex: correlations of thermodynamic and structural data. Proc. Natl. Acad. Sci. U.S.A. 86, 3614-3618. (22) Arnold, F. H., Wolk, S., Cruz, P., and Tinoco, I., Jr. (1987) Structure, dynamics, and thermodynamics of mismatched DNA oligonucleotide duplexes d(CCCAGGG)2 and d(CCCTGGG)2. Biochemistry 26, 4068-4075. (23) Modrich, P. (1987) DNA mismatch correction. Annu. Rev. Biochem. 56, 435-466. (24) Aboul-ela, F., Koh, D., and Tinoco, I., Jr. (1985) Base-base mismatches. Thermodynamics of double helix formation for dCA3XA3G + dCT3YT3G (X,Y ) A,C,G,T). Nucleic Acids Res. 13, 48114824. (25) Ke, S.-H., and Wartell, R. M. (1996) The thermal stability of DNA fragments with tandem mismatches at a d(CXYG).d(CY′X′G) site. Nucleic Acids Res. 24, 707-712. (26) SantaLucia, J., Jr., Allawi, H. T., and Seneviratne, P. A. (1996) Improved nearest-neighbor parameters for predicting DNA duplex stability. Biochemistry 35, 3555-3562. (27) Lesnik, E., and Freier, S. M. (1995) Relative thermodynamic stability of DNA, RNA and DNA:RNA hybrid duplexes: relationship with base composition and structure. Biochemistry 34, 10807-10815. (28) Amosova, O., George, J., and Fresco, J. R. (1997) Effect of the 1-(2′-deoxy-β-D-ribofuranosyl)-3-nitropyrrole residue on the stability of DNA duplexes and triplexes. Nucleic Acids Res. 25, 19301934. (29) Wang, S., Booher, M. A., and Kool, E. T. (1994) Stabilities of nucleotide loops bridging the pyrimidine strands in DNA pyrimidine-purine-pyrimidine triplexes: special stability of the CTTTG loop. Biochemistry 33, 4639-4644. (30) Aramini, J. M., van de Sande, J. H., and Germann, M. W. (1997) Spectroscopic and thermodynamic studies of DNA duplexes containing R-anomeric C, A, and G nucleotides and polarity reversals: coexistence of localized parallel and antiparallel DNA. Biochemistry 36, 9715-9725. (31) Petruska, J., and Goodman, M. F. (1995) Enthalpy-entropy compensation in DNA melting thermodynamics. J. Biol. Chem. 270, 746-750. (32) Searle, M. S., and Williams, D. H. (1993) On the stability of nucleic acid structures in solution: enthalpy-entropy compensations, internal rotations and reversibility. Nucleic Acids Res. 21, 20512056. (33) Blasko, A., Minyat, E. E., Dempcy, R. O., and Bruice, T. C. (1997) Fidelity of binding of the guanidinium nucleic acid (DNG) d(Tg)4T-azido with short strand DNA oligomers (A5G3A5, GA4G3A4G, G2A3G3A3G2, G2A2G5A2G2). A kinetic and thermodynamic study. Biochemistry 36, 7821-7831. (34) Hunter, W. N., Brown, T., Anand, N. N., and Kennard, O. (1986) Structure of an adenine-cytosine base pair in DNA and its implications. Nature 320, 552-555. (35) Singer, B., and Hang, B. (1997) What structural features determine repair enzyme specificity and mechanism in chemically modified DNA? Chem. Res. Toxicol. 10, 713-732. (36) Kalnik, M. W., Chang, C. N., Grollman, A. P., and Patel, D. J. (1988) NMR studies of abasic sites in DNA duplexes: Deoxyadenosine stacks into the helix opposite the cyclic analogue of 2-deoxyribose. Biochemistry 27, 924-931. (37) Wang, K. Y., Parker, S. A., Goljer, I., and Bolton, P. H. (1997) Solution structure of a duplex DNA with an abasic site in a dA tract. Biochemistry 36, 11629-11639.

TX9701992