Chem. Res. Toxicol. 1992,5, 634-638
634
1,N-Ethenodeoxyguanosine: Properties and Formation in Chloroacetaldehyde-TreatedPolynucleotides and DNA J. T. Kusmierekt and B. Singer* Division of Cell and Molecular Biology, Donner Laboratory, Lawrence Berkeley Laboratory, University of California, Berkeley, California 94720 Received March 26, 1992
1,N-Etheno-2’-deoxyguanosine(l,W-cdGuo), not previously reported as a product of chloroacetaldehyde (CAA) reaction, has been synthesized and characterized. Reaction of deoxyguanosine with CAA in dimethylformamide in the presence of KzC03 led to preparation of pure 1,W-edGuo with 55% yield. pK, values are 2.2 and 9.2. The anionic form of the compound exhibits weak but defined fluorescence; the intensity is similar to that of N,3-etheno-2’-deoxyguanosine (N,3-edGuo) a t neutrality. The stability of the glycosyl bond of 1,W-edGuo (tlIz = 2.3 h a t 37 “C,pH 1)is 10-fold greater than of unmodified deoxyguanosine and a t least one thousand-fold greater than of isomeric N2,3-~dGuo. Reaction of CAA with model polynucleotides indicates that hydrogen bonding of guanine residues in the double-stranded structures is, as expected, an important factor in the formation of 1,W-ethenoguanine. In contrast, the formation of isomeric W,3-ethenoguanine is relatively independent of whether the DNA is single- or double-stranded. In salmon sperm DNA, reacted with CAAat neutrality, the formation of 1,N-ethenoguanine could be demonstrated. However, we find the efficiency of formation of this adduct in double-stranded DNA to be lower than that of all other etheno derivatives.
Introduction Vinyl chloride (VC),’ a known mutagen and carcinogen, has been found to form cyclic adducts in DNA of laboratory animals exposed to this agent ( I ) . This type of base adduct was also found in reactions of the VC metabolites, chloroethylene oxide and chloroacetaldehyde, with DNA in vitro. They include 1,M-ethenoadenine(l,M-eA), 3,N4ethenocytosine (3,N4-eC),and W,3-ethenoguanine (W,3eG), and all of them have been found to miscode (recently reviewed in ref 2). It seems surprising that 1P-ethenoguanine has not been found in these reactions, since there are many r e p o h that a variety of bifunctional agents form analogous lJV2-cyclicadducts with guanine residues in DNA, both in vivo and in vitro. lJV2-Etheno-2’-deoxyguanosine itself has been found as a final product of the reaction of deoxyguanosine after hyperoxide-mediatedaddition of trans-4-hydroxy-2-nonend, a major a,P-unsaturated aldehyde released during lipid peroxidation (3). The a,P-unsaturated aldehydes form 1,Wpropanodeoxyguanosinederivatives in a Michaeltype addition. This type of saturated six-membered ring is formed in the reaction of compounds such as acrolein and crotonaldehyde with guanine residues in DNA both in vitro (4) and in vivo (5, 6). On the other hand, P-dicarbonyl compounds, such as malondialdehyde, cyclize between the N-1 and N2of guanine to form a six-membered aromatic type ring. The formation of the five-membered 1 P - e t h e n o ring occurs in the reaction of haloacetaldehydes with guanosine in both aqueous (8) and nonaqueous solution (9).All + Permanentaddress: Institute of Biochemistry and Biophysics, Polish Academy of Sciences, 36 Rakowiecka St., 02532 Warsaw, Poland. Abbreviations: lfl-tdGuo, lJP-etheno-2’-deoxyguanosine;CAA, chloroacetaldehyde; W,3-sdGuo, lJP-etheno-2’-deoxyguanosine;VC, vinyl chloride;l,W-tA, li\”-ethenoadenine; 3JV4-tC,3,W-ethenocytosine; W,3-tG,W,3-ethenoguanine; poly(dG-dC), alternating poly(deoxyguanosine)-poly(deoxycyt0sine); TLC,thin-layerchromatography;poly(dGdT), alternating poly(deoxyguanosine)-poly(de0xythymidine).
these findings indicate that 1JV2-eGshould also be formed in the reaction of VC metabolites with DNA. In view of the foregoing we have attempted to determine whether 1JV2-eGis one of the reaction products of CAA with DNA.2 We now describe the synthesis of 1JV2-etheno2’-deoxyguanosine (lp-edGuo), as well as present the evidence for formationof 1JV2-eGin polydeoxynucleotides and in DNA reacted with CAA, a vinyl chloride metabolite.
Materials and Methods Chloroacetaldehyde2 (CAA),45% solution in water (-6.9 M) was purchased from Merck (Rahway,NJ). l,P-Etheno-2’-deoxyadenosine and 3,N4-etheno-2’-deoxycytidine were prepared following the procedure of Barrio et al. (IO),originally described for preparation of modified ribonucleosides. 1p-Ethenoadenine was obtained by acid hydrolysis of the modified deoxynucleoside. N2,3-Ethenoguanine was obtained as described by Kugmierek et al. (11). Salmon sperm DNA was from Sigma (St.Louis, MO) and poly(dG-dC) was from Pharmacia (Piscataway, NJ). (dG-dT)sowas synthesized using a DNA synthesizer. This oligonucleotide exhibited a broad thermal transition profile between 5 and 50 “C in 0.3 M sodium cacodylate buffer (pH 7.25) containing 10 mM spermidine (data not shown). The ultraviolet measurements including spectra, pK,, t 1 / 2 determinations, and thermal melting profiles of the polydeoxynucleotides were obtained on a Varian-Cary 219 spectrophotometer equipped with a constant-temperaturecell holder. The corrected fluorescence spectra were determined on a Spex Industries FluoroMax spectrofluorometer. HPLC was performed using an LKB system with a PerkinElmer Model 650-15 fluorescence spectrophotometer and a BioRad UV detector in tandem, using a dual-pen recorder. Cation exchange a t 45 “C was on a 4- X 250-mm A-8 Bio-Rad column (NH4+form) with 0.4 M (pH 5.6 or 9.0) ammonium formate as eluants, at a flow rate of 0.5 mL/min. Relevant retention times 2 Caution: Chloroacetaldehyde i s a hazardous chemical and is only used in a well-uentilated hood. Protectioe clothing and face mask are required.
1992 American Chemical Society
Chem. Res. Toxicol., Vol. 5, No. 5, 1992 635
1,W-Ethenodeoxyguanosine Table I. HPLC Retention Times of Relevant Compoundsa retention time (min) compound pH 5.6 pH 9.0 14 8 2'-deoxy cytidine 6 6 2'-deoxythymidine 32 20 adenine 26 21 guanine 49 37 1,W-ethenoguanine(3) W,3-ethenoguanine 24 14 58 38 1fl-ethenoadenine l,W-etheno-2'-deoxyguanosine (2) 32 22 3JV4--etheno-2'-deoxycytidine 18 15 a See Materials and Methods for details and instrumentation. Figure 2 shows examples of HPLC analyses of DNA and poly(dGdT) reacted with CAA. are given in Table I. Celluloseplates with a fluorescent indicator (Kodak 13254) were used for thin-layer chromatography with 2-propanol/H~O/concentratedammonia (70:20:10) as solvent. Experimental Procedures. (A) l,W-Etheno-2'-deoxyguanosine (Scheme I). To the stirred suspension of 300 mg of 2'-deoxyguanosine.l.5H20(1) (1.02 mmol) and 210 mg of KzC03 (1.5 mmol) in 5 mL of dimethylformamide was added 0.25 mL of 45 % aqueous solution of CAA (ca 1.7 mmol). After 3 h, another portion of K2CO3 (140 mg, 1 mmol) and CAA (0.15 mL, ca. 1 mmol) was added and the reaction was carried on for an additional 12 h at room temperature. TLC showed the presence of 2 (RI, = 0.54) as a major product, some substrate (RI, = 0.31), and several minor products. The brown dark mixture was diluted with water (20 mL) and loaded on a 3.2- X 15-cm column (Dowex 1 X 2 200/400 in HC03- form). The column was eluted with 300 mL of water and then with a linear gradient of 1000 mL of H20 X 1000mL of 0.5 M NH4HC03 in 50% aqueous methanol. Eightmilliliter fractions were collected at a flow rate of about 2.5 mL/ min. Fractions 180-240 (from 0.25 to 0.40 M NH4HC03) containing 5100 0Di:;L units of 2 were pooled and evaporated at 40 "C. The residue was evaporated three more times with water in order to remove NH4HC03 and then heated to about 70 "C to dissolve in a mixture of -50 mL of HzO/methanol (l:l), to which was added concentrated ammonia. The mixture was then treated with charcoal. filtered, and again evaporated. The residue was dissolved in 5 mL of water with addition of 2.5 mL of concentrated ammonia. After dissolving at 100OC, the solution was cooled and kept overnight in a desiccator in the presence of PzO5, to remove excess NH3. The resulting white precipitate was collected by centrifugation, washed with water and then with acetone, and finally dried in a desiccator: yield 164 mg (55%); mp above 360 OC (darkening started >250 "C); UV: (pH 7 ) A, = 227 nm (35 200), A, = 284 nm (11800); (pH 2) A, = 225 nm = 272 nm (8400), A, = 290 nm (8700); (pH 12) (28 goo), A,, A, = 233 nm (30 9001, A, = 280 nm (52001, A,, = 307 nm (8100);Anal. Calcd for C12H13N504.0.25H~O:C, 48.73; H, 4.60; N, 23.68. Found: C, 48.37; H, 4.37; N, 23.81. (B) 1,WEthenoguanosine. The reaction of guanosine with CAA and subsequent workup of reaction mixtures used the same methodology as the synthesis of 2. The yield of lfl-ethenoguanosine varied between 15 and 20%, mp 257-259 OC [lit. mp 252253 OC (8);256-257.5 "C (911. UVspectrawerevirtuallyidentical to the UV spectra of 2. (C) 1,WEthenoguanine (3). The solution of 50 mg of 2 (0.17 mmol) in 4 mL of 0.5 N HC1 was kept at 37 "C. After 18 h TLC showed complete disappearance of 2 (RI,= 0.54) and = 0.30) as a single UV-absorbing product. The formation of 3 (RI, mixture was neutralized by the addition of solid NH4HC03,and the precipitate of 3 was collected by centrifugation. The precipitate was dissolved at 100OC in 5 mL of water with addition of 0.5 mL of concentrated aqueous ammonia. The mixture was heated in a boiling water bath in order to evaporate ammonia until a precipitate appeared (- 1h). After cooling,the precipitate of 3 was again collected by centrifugation, washed with water and then with acetone, and dried in a desiccator: yield 21 mg
Scheme I
Ho-o Help, OH
OH
1
2
3
(70%);mp above 360 "C [lit. mp >290 OC (8)l;U V (pH 7) A,, = 224 and 291 nm; (pH 1)A,, = 220,267, and 296 nm; (pH 13) A, = 235, 263 (sh), and 320 nm. (D) Reaction of Poly(dG-dT) with CAA. A solution of 13 A units of (dG-dT)~o in 150pL of 0.3 M sodium cacodylate buffer containing 10 mM spermidine was heated to 80 "C and then slowly cooled to room temperature. An aliquot of 75 pL of this solution was cooled to 0 "C, 5 pL of 45% CAA was added, and the mixture was stored in a refrigerator (5 "C). After 9 days, 40 pL of the reaction mixture was removed and the reacted polymer precipitated [poly(dG-dT)-11. To the remaining part of the reaction mixture was added another 5 pL of CAA, and the sample was stored for an additional 6 days at 5 "C and then precipitated [poly(dG-dT)-2]. To the remaining half (75 pL) of the original poly(dG-dT) was added a 5-pL solution of 45 % CAA, and this sample was reacted at 50 "C for 1 h [poly(dG-dT)-3]. (E) Reaction of Poly(dG-dC) with CAA. To a solution of 5 A units of poly(dG-dC) in 200 pL of 0.25 M sodium cacodylate buffer (pH 7.25) was added 10 pL of 45% CAA, and the mixture was stored at room temperature (-22 "C) for 27 h [poly(dGdC)-1]. Two other identical reaction mixtures were stored a t either 37 "C for 40 h [poly(dG-dC)-2] or 37 "C for 96 h [poly(dG-dC)-3]. (F) Reaction of DNA with CAA. Four samples of salmon sperm DNA, each containing 4 A units in 150pL of 0.3 M sodium cacodylate buffer (pH 7.251, were reacted at 37 OC as follows: DNA-1 in 0.23 M CAA (5 pL of 45% CAA added) for 4 days, DNA-2 in 0.23 M CAA for 8 days, DNA-3 in 0.46 M CAA (10 pL of 45% CAA) for 4 days, and DNA-4 in 0.46 M CAA for 8 days. (G) Precipitation of Reacted Polymers. The reactions of polydeoxynucleotides and DNA with CAA were stopped by precipitation of reacted polymers. To a sample (40-200 pL) of reacted polymer were added 200 pL of 3 M sodium acetate buffer (pH 5) and 1-1.5 mL of ethanol. After 30 min in an ice bath, the precipitate was centrifuged, washed with ethanol, centrifuged again, and air-dried. Samples were dissolved in an appropriate volume of 10 mM Tris-HC1 buffer (pH 8) and frozen for storage. (H) Acid Hydrolysis of Reacted Polymers. To a solution of 1-3 OD units of reacted polymer dissolved in 50-200 mL of 10 mM Tris-HC1 buffer (pH 8) was added 1 N HC1 to a final concentration of 0.1 N. The sample was heated at 100 "C for 10 min, and then a few crystals of NH4HC03 were added for neutralization. (I) Enzymatic Hydrolysis. To a neutralized sample (after acid hydrolysis) were added 1M Tris-HC1 buffer (pH 9) and 0.1 M MgClz to final concentrations of 0.1 and 0.01 M, respectively. Deoxyribonuclease I (10 mg/mL) was added in a proportion of 2.5 pLi100 pL of depurinated sample and incubated for 3 h a t 37 "C. Snake venom phosphodiesterase (5 mg/mL) was then added in the proportion 2.5 pL/lOO pL and bacterial alkaline phosphatase (40 mg/mL) in the proportion 1 pL/l00 pL. Incubation was continued for 12 h. The addition of phosphodiesterase and phosphatase was repeated, and the samples were incubated for another 12 h prior to analysis.
Results Chemical. T h e reaction of 2'-deoxyguanosine (1) with 45% aqueous CAA in the presence of K&03 in dimethylformamide solution leads t o the formation of 1,Wetheno-2'-deoxyguanosine (2) as the major product which can b e isolated in satisfactory yield (over 50%). The use
636 Chem. Res. Toxicol., Vol. 5, No. 5, 1992
Kugmierek and Singer Table 11. Reactivity of Guanine Residues in Poly(dG-dT) and Poly(dG-dC) Treated with Chloroacetaldehudea __
temp
polynucleotide poly(dG-dT)-l poly(dG-dT)-2 poly(dG-dT)-3 poly(dG-dC)-l poly(dG-dC)-P p0ly(dG-dC)-3~ 300
400
500
Wavelength (nm)
Figure 1. Corrected fluorescence spectraof lJV-etheno-2'-deoxyguanosine (2) at pH 12. Right side: Emission spectrum with a maxima at 407 nm (excitation at 300 nm); the small peak at 340 nm is a Raman water peak. Left side: Excitation spectrum with a maxima at 308 nm (observed at 450 nm).
of Dowex 1X 2 resin for purification of the reaction product assures the separation of 2 from unreacted 1, which is less retarded by the resin than the more hydrophobic etheno adduct 2. Another advantage of Dowex 1 X 2 is that the resin almost completely retains the brown tarry side products which were formed during the reaction. Decolorization by Dowex is much more efficient in this case than classicalcharcoal treatment. The analogous reaction of guanosine and subsequent workup of the reaction mixture yield 1,W-ethenoguanosine. However, the yield of the modified ribonucleoside was repeatedly lower and did not exceed 20 % ,compared to 2. The acid hydrolysis of 2 yields base 3. Properties of 1,W-Ethenoguanine Nucleosides. Both the ribonucleoside and deoxynucleoside of 1,W-tG exhibit virtually identical UV spectra, which are in agreement with the spectral data published for 1,W-tGuo ( 4 9 )and 1,W-cdGuo(3). TheUVspectraof lJP-ethenoguanine nucleosides are distinctively different from the spectra of W,3-ethenoguanine nucleosides (1I , 12). The pKa values of 1JP-tGuo obtained by spectrophotometric titration in Britton-Robinson buffers are 2.2 and 9.2. These are similar to these of isomericW,3-tGuo [pK,: -2.5 and 8.85 (11)l and unmodified guanosine [pKa: 1.6 and 9.2 (13)l. 1,WEthenoguanine nucleosides exhibit weak fluorescence in alkaline solution. The emission and excitation spectra of 2 at pH 1 2 are presented in Figure 1. The approximatefluorescencequantum yield (a)of 2 compared to the fluorescenceof 1,Wethenoadenosine [@ = 0.56 (14)l gives the value of about 0.01 at pH 12. The spectrofluorometric titration of 2 from pH 12 to pH 7 results in about a 10-fold decrease of fluorescence intensity. The pK, of about 9, determined by this method, is close to the value determined by absorption spectrophotometry (see above). This indicates that the anionic species of 1,WtdGuo is responsible for its fluorescence. The fluorescence efficiency of the anionic form of 1JP-tdGuo is about the same as the fluorescence efficiency of the neutral form of W,3-tdGuo. However, the fluorescence maxima of the 1,N2-isomer (407 nm) is different from the fluorescence maxima of the N2,3-isomer (430 nm) (8, 11). The hydrolysis of the glycosyl bond of 1,W-tdGuo in 0.1 N HCl at 37 OC was measured by following UV absorption changes. The t1/2 value for this reaction was found to be 2.3 h, which indicates that the glycosyl bond
("C) 5 5 50 22 37 37
reaction time
% of Darent base transfbrmed into* 1P-cG W.3-cG
9 days 15 days lh 27 h 40 h 96 h
2.9 7.2 13.2 W,3-eG >>> 1,W-eG (see Table I11 and footnote e in this table). There are numerous reports on using chloroacetaldehyde (or bromoacetaldehyde) as a probe for DNA structural features such as cruciforms and B-Z junctions (18, 19), Z-DNA (20),triple-helical structures (21), and DNA bending by bulky adducts (22). Generally, the method exploits the fact that in such structures the reactive sites of bases are more exposed to the action of the agent than the sites of bases involved in regular B-DNA pairing. Under CAA limiting conditions, there is apparent formation of etheno adducts in such regions but not in the regions where DNA is in the B-form. The extensive treatment of bulk salmon sperm DNA, which exist mainly in B-form under the reaction conditions employed in our experiments, results in the formation of all possible etheno adducts. This is not in contradiction with the results where the action of CAA was limited, since the long exposure of DNA to a high concentration of CAA will allow bases involved in B-structure to react via fluctuational opening of the helix. An etheno base formed would be likely to result in a bulge which could further distort DNA, allowing additional reaction. On the basis of our findings some speculation can be made about possible formation of 1 P - e G in vivo and its possible role in mutagenesis. It is obvious that, due to the fact that DNA in the cell is mainly in double-stranded B-form, the overall formation of this adduct will be very low in comparison to the formation of other etheno adducts. However, the other forms which exist in cells, especially single-stranded regions, which are present in replication forks will react differently. The efficiency of formation of 1,W-tG is, in the single-stranded regions, expected to be comparable to the formation of 3,N4-eC and lJV-eA, and higher thanW,3-eG (Table 11). The 1,W-ethenoring blocks the essential sites in guanine for hydrogen bonding (see Scheme I, structure 2). This would imply that I F eG will be a noninstructive lesion, in contrast to its isomer, W,3-tdGuo, which has been shown to produce a high level of transitions (23,241. Studies on the role of 1,W-edGuo in replication are in progress. Acknowledgment. This work was supported by Grant CA-47723 from the National Cancer Institute, NIH, Bethesda, MD (B.S.) and was administered by Lawrence Berkeley Laboratory under DOE Contract DEOAC0376SF00098. Additional support was from a grant from the Polish Academy of Sciences and from Grant BST62/91 from theuniversityof Warsaw (J.T.K.). Theauthors are indebted to Mr. Krzysztof Krawiec of the Institute of
Kuimierek and Singer
638 Chem. Res. Toxicol., Vol. 5, No. 5, 1992
Biochemistryand Biophysics,Polish Academy of Sciences for fluorescence spectra.
References (1) Swenberg, J. A., Fedtke, N., Fennell, T. R, and Walker, V. E. (1990) Relationship between carcinogen exposure, DNA adducts and carcinogenesis. In Progress i n Predictive Toxicology (Clayson, D. B., et al., Eds.) pp 161-184, Elsevier Science Publishers B.V. (Biomedical Division), Amsterdam. (2) Mroczkowska, M. M., and KuBmierek, J. T. (1991) Miscoding potentialof N2,3-ethenoguanine studied in anEscherichia coli DNAdependent RNA polymerase in uitro system and possible role of this adducts invinyl chloride-induced mutagenesis. Mutagenesis 6,385395. (3) Sodum, R. S., and Chung, F.-L. (1988) 1JP-Ethenodeoxyguanosine as a potential marker for DNA adduct formation by trans-Chydroxy2-nonenal. Cancer Res. 48,320-323. (4) Chung, F.-L., Young, R., and Hecht, S. S. (1984) Formation of cyclic, 1JP-propanodeoxyguanosine adducts in DNA upon reactions with acrolein or crotonaldehyde. Cancer Res. 44, 990-995. (5) Foiles, P. G., Akerkar, S. A., and Chung, F.-L. (1988) Application of an immunoassay for cyclic acrolein deoxyguanosine adducts to assess their formation in DNA of Salmonella typhimurium under conditions of mutation induction by acrolein. Carcinogenesis 10, 87-90. (6) Chung, F.-L., Young, R., and Hecht, S. S. (1989) Detection of cyclic 1,Wpropanodeoxyguanosine adducts in DNA of rats treated with N-nitrosopyrrolidine and mice treated with crotonaldehyde. Carcinogenesis 10, 1291-1297. (7) Basu, A. K., O’Hara, S. M., Valladier, P., Stone, K., Mols, O., and Marnett, L. J. (1988) Identification of adducts formed by reaction of guanine nucleosides with malondialdehyde and structurally related aldehydes. Chem. Res. Toxicol. 1, 53-59. (8) Sattsangi, P. D., Leonard, N. J., and Frihart, C. R. (1977) 1 P ethenoguanine and N2,3-ethenoguanine. Synthesis and comparison of the electronic spectral properties of these linear and angular triheterocycles related to Y bases. J . Org. Chem. 42, 3292-3296. (9) Boryski,J. (1990) 1,Wethenoguanosine: three methodsofsynthesis. Nucleosides Nucleotides 9, 803-813. (10) Barrio, J. R., Secrist, J. A., and Leonard, N. J. (1972) Fluorescent adenosine and cytidine derivatives. Biochem. Biophys. Res. Commun. 46, 597-604. (11) KuBmierek, J. T., Jensen, D. E., Spengler, S. J., Stolarski, R., and Singer, B. (1987) Synthesis and properties of N2,3-ethenoguanosine and N2,3-ethenoguanosine 5’-diphosphate. J . Org. Chem. 52,23742378.
(12) KuBmierek, J. T., Folkman, W., and Singer, B. (1989) Synthesis of N2,3-ethenodeoxyguanosine,N2,3-ethenodeoxyguanosine5’-phosphate, and N2,3-ethenodeoxyguanosine5’-triphosphate. Stability of the glycosyl bond in the monomer and inpoly(dG,fdG-dC). Chem. Res. Toxicol. 2, 23C-233. (13) Fasman, G. D. Ed. (1975) Handbook ofBiochemistry andMolecular Biology. Nucleic Acids, Volume I, CRC Press, Boca Raton, FL. (14) Secrist, J. A. Barrio, J. R., Leonard, N. J., and Weber, G . (1972) Fluorescent modification of adenosine-containing coenzymes. Biological activities and spectroscopic properties. Biochemistry l l , 3499-3506. (15) Singer, B., and Grunberger, D. (1983)Molecular Biology ofMutagem and Carcinogens, Plenum Press, New York. (16) KuBmierek, J. T.,and Singer, B. (1982) Chloroacetaldehyde-treated ribo- and deoxyribopolynucleotides. 1. Reaction products. Biochemistry 21, 5717-5722. (17) Early, T. A., Olmsted, J., Kearns, D. R., and Lezius, A. G. (1978) Base pairing structure in poly d(G-T) double-helix: wobble base pairs. Nucleic Acids Res. 5, 1955-1970. (18) McLean, M. J., Larson, J. E., Wohlrab, F., and Wells, R. D. (1987) Reaction conditions affect the specificity of bromoacetaldehyde as a probe for DNA cruciforms and B-Z junctions. Nucleic Acids Res. 15, 6917-6935. (19) Dayn, A., Malkhosyan, S., Duzhy, D., Lyamichev, V., Panchenko, Y., and Mirkin, S. (1991) Formation of (dA-dT), cruciforms in Escherichia coli cells under different environmental conditions. J. Bacteriol. 173, 2658-2664. (20) Vogt, N., Marrot, L., Rousseau, N., Malfoy, B., and Leng, M. (1988) Chloroacetaldehyde reacts with Z-DNA. J . Mol. Biol. 201,773-776. (21) Kohwi, Y., and Kohwi-Shigematsu, T. (1988) Magnesium ion dependent, novel triple-helix structure formed by homopurinehomopyrimidine sequences in supercoiled plasmid DNA. Proc. Natl. Acad. Sci. U.S.A. 85, 3781-3788. (22) Schwartz, A., Marrot, L., and Leng, M. (1989) The DNA bending by acetylaminofluorene residues and by apurinic sites. J . Mol. Biol. 207, 445-450. (23) Singer, B., KuBmierek, J. T., Folkman, W., Chavez, F., and Dosanjh, M. K. (1991) Evidence for the mutagenic potential of the vinyl chloride induced adduct, N2,3-ethenodeoxyguanosine,using a sitedirected kinetic assay. Carcinogenesis 12, 745-747. (24) Cheng, K. C., Preston, B. D., Cahill, D. S., Dosanjh, M. K., Singer, B., and Loeb, L. A. (1991) The vinyl chloride derivative, Nz,3ethenoguanine, produces G A transitions in E. coli. Proc. Natl. Acad. Sci. U.S.A. 88,9974-9978.
-
