Correlation between Sequence-dependent Glycosylase Repair and the Thermal Stability of Oligonucleotide Duplexes Containing 1,N 6-Ethenoadenine
1998; Elsevier BV; Volume: 273; Issue: 50 Linguagem: Inglês
10.1074/jbc.273.50.33406
ISSN1083-351X
AutoresBo Hang, János Sági, Brett C. Singer,
Tópico(s)HIV/AIDS drug development and treatment
ResumoPrevious experiments on DNA sequence context reported that base modification, replication, and repair are affected by the nature of neighbor bases. We now report that repair by mammalian alkylpurine-DNA-N-glycosylases (APNG) of 15-mer oligonucleotides with a central 1,N 6-ethenoadenine (εA), flanked by 5′ and 3′ tandem bases, is also highly sequence dependent. Oligonucleotides with the central sequences -GGεAGG- or -CCεACC- are repaired 3–5-fold more efficiently than those containing -AAεAAA- or -TTεATT- when using human or mouse APNG. Melting curves of the same duplexes showed that oligomers with G·C/C·G neighbors were less denatured than those with A·T/T·A neighbors at 37 °C. This sequence-dependent difference in denaturation correlates with the relative thermodynamic stability of oligomers with G·C/C·G or A·T/T·A neighbors. The dependence of repair on thermal stability was confirmed by enzyme reactions performed over 0–45 °C. Under these conditions, repair of εA flanked by G·C/C·G was dramatically increased at 37 °C with continuous increase up to 45 °C, in contrast to that with flanking A·T/T·A pairs, which was in agreement with the degree of denaturation of these duplexes. These results indicate that the thermodynamic stability conferred by base pairs flanking εA plays an essential role in maintaining the integrity of the duplex structure which is necessary for repair. Previous experiments on DNA sequence context reported that base modification, replication, and repair are affected by the nature of neighbor bases. We now report that repair by mammalian alkylpurine-DNA-N-glycosylases (APNG) of 15-mer oligonucleotides with a central 1,N 6-ethenoadenine (εA), flanked by 5′ and 3′ tandem bases, is also highly sequence dependent. Oligonucleotides with the central sequences -GGεAGG- or -CCεACC- are repaired 3–5-fold more efficiently than those containing -AAεAAA- or -TTεATT- when using human or mouse APNG. Melting curves of the same duplexes showed that oligomers with G·C/C·G neighbors were less denatured than those with A·T/T·A neighbors at 37 °C. This sequence-dependent difference in denaturation correlates with the relative thermodynamic stability of oligomers with G·C/C·G or A·T/T·A neighbors. The dependence of repair on thermal stability was confirmed by enzyme reactions performed over 0–45 °C. Under these conditions, repair of εA flanked by G·C/C·G was dramatically increased at 37 °C with continuous increase up to 45 °C, in contrast to that with flanking A·T/T·A pairs, which was in agreement with the degree of denaturation of these duplexes. These results indicate that the thermodynamic stability conferred by base pairs flanking εA plays an essential role in maintaining the integrity of the duplex structure which is necessary for repair. Many laboratories have published data on the effect on efficiency of modification, replication, or repair of nucleic acids conferred by the immediate neighbor bases. In almost all cases, there was some measurable difference depending on sequence context. Most of the experiments on chemical modification which used a gene or a naturally occurring DNA led to a statistical analysis of the bases neighboring the lesion studied. These important data contributed to the hypothesis that localized DNA structure was an important determinant in the non-random distribution of adducts (e.g. Refs. 1Haseltine W.A. Gordon L.K. Lindan C.P. Grafstrom R.H. Shaper N.L. Grossman L. Nature. 1980; 285: 634-641Crossref PubMed Scopus (190) Google Scholar, 2Burns P.A. Gordon A.J.E. Glickman B.W. J. Mol. Biol. 1987; 194: 385-390Crossref PubMed Scopus (112) Google Scholar, 3Dolan M.E. Oplinger M. Pegg A.E. Carcinogenesis. 1988; 9: 2139-2143Crossref PubMed Scopus (98) Google Scholar, 4Richardson F.C. Boucheron J.A. Skopek T.R. Swenberg J.A. J. Biol. Chem. 1989; 264: 838-841Abstract Full Text PDF PubMed Google Scholar, 5Carothers A. Urlaub G. Mucha J. Harvey R. Chasin L.A. Grunberger D. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 5464-5468Crossref PubMed Scopus (31) Google Scholar, 6Skopek T.R. Walker V.E. Cochrane J.E. Craft T.R. Cariello N.F. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 7866-7870Crossref PubMed Scopus (128) Google Scholar, 7Cmarik J.L. Humphreys W.G. Bruner K.L. Lloyd R.S. Tibbetts C. Guengerich F.P. J. Biol. Chem. 1992; 267: 6672-6679Abstract Full Text PDF PubMed Google Scholar, 8Ross H. Bigger C.A.H. Yagi H. Jerina D.M. Dipple A. Cancer Res. 1993; 53: 1273-1277PubMed Google Scholar, 9Yarema K.J. Wilson J.M. Lippard S.J. Essigmann J.M. J. Mol. Biol. 1994; 236: 1034-1048Crossref PubMed Scopus (44) Google Scholar, 10Wyatt M.D. Lee M. Garbiras B.J. Souhami R.L. Hartley J.A. Biochemistry. 1995; 34: 13034-13041Crossref PubMed Scopus (49) Google Scholar). Replication efficiency has been found in many cases to be a function of nearest neighbors (e.g. Refs. 11Petruska J. Goodman M.F. J. Biol. Chem. 1985; 260: 7533-7539Abstract Full Text PDF PubMed Google Scholar, 12Mendelman L.V. Boosalis M.S. Petruska J. Goodman M.F. J. Biol. Chem. 1989; 264: 14415-14423Abstract Full Text PDF PubMed Google Scholar, 13Singer B. Chavez F. Goodman M.F. Essigmann J.M. Dosanjh M.K. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 8271-8274Crossref PubMed Scopus (132) Google Scholar, 14Dosanjh M.K. Galeros G. Goodman M.F. Singer B. Biochemistry. 1991; 30: 11595-11599Crossref PubMed Scopus (69) Google Scholar, 15Cai H. Bloom L.B. Eritja R. Goodman M.F. J. Biol. Chem. 1993; 268: 23567-23572Abstract Full Text PDF PubMed Google Scholar, 16Bloom L.B. Otto M.R. Beechem J.M. Goodman M.F. Biochemistry. 1993; 32: 11247-11258Crossref PubMed Scopus (98) Google Scholar, 17Bloom L.B. Otto M.R. Eritja R. Reha-Krantz L.J. Goodman M.F. Beechem J.M. Biochemistry. 1994; 33: 7576-7586Crossref PubMed Scopus (104) Google Scholar, 18Goodman M.F. Cai H. Bloom L.B. Eritja R. Ann. N. Y. Acad. Sci. 1994; 726: 132-142Crossref PubMed Scopus (36) Google Scholar, 19Rao S. Chenna A. Slupska M. Singer B. Mutat. Res. 1996; 356: 179-185Crossref PubMed Scopus (9) Google Scholar, 20Hashim M.F. Marnett L.J. J. Biol. Chem. 1996; 271: 9160-9165Abstract Full Text PDF PubMed Scopus (29) Google Scholar, 21Litinski V. Chenna A. Sági J. Singer B. Carcinogenesis. 1997; 18: 1609-1615Crossref PubMed Scopus (23) Google Scholar). In these experiments, specific sequences containing a modified base were primed and insertion and extension examined using a variety of polymerases. Usually the 5′ neighbor base effect was found to be an important factor, although the 3′ neighbor can also play a role. In addition, it has also been reported that the proofreading efficiency of polymerases is sequence-dependent (11Petruska J. Goodman M.F. J. Biol. Chem. 1985; 260: 7533-7539Abstract Full Text PDF PubMed Google Scholar, 17Bloom L.B. Otto M.R. Eritja R. Reha-Krantz L.J. Goodman M.F. Beechem J.M. Biochemistry. 1994; 33: 7576-7586Crossref PubMed Scopus (104) Google Scholar). There are also numerous studies on the sequence dependence of enzymatic repair of various types of damage in defined DNA sequences (e.g. Refs. 22Jones M. Wagner R. Radman M. Genetics. 1987; 115: 605-610Crossref PubMed Google Scholar, 23Seeberg E. Fuchs R.P. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 191-194Crossref PubMed Scopus (67) Google Scholar, 24Foley C.K. Pedersen L.G. Darden T.A. Glickman B.W. Anderson M.W. Mutat. Res. 1991; 255: 89-93Crossref PubMed Scopus (2) Google Scholar, 25Mu D. Bertrand-Burggraf E. Huang J.-C. Fuchs R.P.P. Sancar A. Nucleic Acids Res. 1994; 22: 4869-4871Crossref PubMed Scopus (37) Google Scholar, 26Eftedal I. Volden G. Krokan H.E. Ann. N. Y. Acad. Sci. 1994; 726: 312-314Crossref PubMed Scopus (10) Google Scholar, 27Bender K. Federwisch M. Loggen U. Nehls P. Rajewsky M.F. Nucleic Acids Res. 1996; 24: 2087-2094Crossref PubMed Scopus (27) Google Scholar, 28Sander M. Benhaim D. Nucleic Acids Res. 1996; 24: 3926-3933Crossref PubMed Scopus (15) Google Scholar, 29Sibghat-Ullah Gallinari P. Xu Y.-Z. Goodman M.F. Bloom L.B. Jiricny J. Day III, R.S. Biochemistry. 1996; 35: 12926-12932Crossref PubMed Scopus (83) Google Scholar, 30Delagoutte E. Bertrand-Burggraf E. Dunand J. Fuchs R.P.P. J. Mol. Biol. 1997; 266: 703-710Crossref PubMed Scopus (23) Google Scholar, 31Mekhovich O. Tang M-s. Romano L.J. Biochemistry. 1998; 37: 571-579Crossref PubMed Scopus (27) Google Scholar, 32Hang B. Chenna A. Sági J. Singer B. Carcinogenesis. 1998; 19: 1339-1343Crossref PubMed Scopus (31) Google Scholar). Here again, specific neighbor sequences led to significant variations in the rate and extent of excision of a modified base by repair mechanisms including base excision repair, nucleotide excision repair, mismatch repair, andO 6-methylguanine-DNA methyltransferase. When a double-stranded structure is required, either as a partial duplex (template/primer) or as a full-length duplex, the stability of base pairs adjacent to the adduct can be an important factor in replication and repair. Thermodynamics has been a general method used to probe nucleic acid structure and strandedness (e.g. Refs.33Nelson J.W. Martin F.H. Tinoco Jr., I. Biopolymers. 1981; 20: 2509-2531Crossref PubMed Scopus (66) Google Scholar, 34Breslauer K.J. Frank R. Blocker H. Marky L.A. Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 3746-3750Crossref PubMed Scopus (1562) Google Scholar, 35Petruska J. Goodman M.F. Boosalis M.S. Sowers L.C. Cheong C. Tinoco Jr., I. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 6252-6256Crossref PubMed Scopus (289) Google Scholar, 36Ke S.-H. Wartell R.M. Biochemistry. 1995; 34: 4593-4600Crossref PubMed Scopus (45) Google Scholar, 37SantaLucia Jr., J. Allawi H.T. Seneviratne P.A. Biochemistry. 1996; 35: 3555-3562Crossref PubMed Scopus (708) Google Scholar, 38Allawi H.T. SantaLucia Jr., J. Biochemistry. 1998; 37: 2170-2179Crossref PubMed Scopus (206) Google Scholar). In some replication studies of templates containing modified bases, a correlation between the rate of insertion of a dNTP and the relative base pairing energies of the nearest neighbor base pairs was reported (16Bloom L.B. Otto M.R. Beechem J.M. Goodman M.F. Biochemistry. 1993; 32: 11247-11258Crossref PubMed Scopus (98) Google Scholar, 39Shibutani S. Margulis L.A. Geacintov N.E. Grollman A.P. Biochemistry. 1993; 32: 7531-7541Crossref PubMed Scopus (119) Google Scholar, 40Arghavani M.B. SantaLucia J. Romano L.J. Biochemistry. 1998; 37: 8575-8583Crossref PubMed Scopus (23) Google Scholar). In repair, it was assumed that G·C-rich neighbor regions would play a role in stabilizing the necessary double strand containing a mismatched base (22Jones M. Wagner R. Radman M. Genetics. 1987; 115: 605-610Crossref PubMed Google Scholar). This is likely to be true for all DNA modifying and repair enzymes which require a DNA duplex structure for activity except uracil-DNA glycosylase (41Lindahl T. Annu. Rev. Biochem. 1982; 51: 61-87Crossref PubMed Scopus (696) Google Scholar) andO 6-methylguanine-DNA methyltransferase both of which can act on single-stranded substrates, althoughO 6-methylguanine-DNA methyltransferase is more active on double-stranded DNA substrates (42Demple B. Jacobsson A. Olsson M. Robins P. Lindahl T. J. Biol. Chem. 1982; 257: 13776-13780Abstract Full Text PDF PubMed Google Scholar, 43Scicchitano D. Pegg A.E. Biochem. Biophys. Res. Commun. 1982; 109: 995-1001Crossref PubMed Scopus (24) Google Scholar). However, under the circumstances where repair is carried out by protein complexes such as the Escherichia coli UvrABC nuclease, the mechanism(s) underlying the sequence-dependent repair becomes more complicated as the efficiency of repair is also correlated with the stability of the preincision UvrB·DNA and UvrBC·DNA complexes (30Delagoutte E. Bertrand-Burggraf E. Dunand J. Fuchs R.P.P. J. Mol. Biol. 1997; 266: 703-710Crossref PubMed Scopus (23) Google Scholar, 31Mekhovich O. Tang M-s. Romano L.J. Biochemistry. 1998; 37: 571-579Crossref PubMed Scopus (27) Google Scholar). In the present work, we have carried out parallel studies of both repair by alkylpurine-DNA-N-glycosylase (APNG) 1The abbreviations used are: APNG, alkylpurine-DNA-N-glycosylase; HAP1, human AP endonuclease 1; εA, 1,N 6-ethenoadenine; AP site, abasic site. (also termed 3-methyladenine-DNA glycosylase) and thermal stability using a set of specifically designed 15-mer oligonucleotides with purine or pyrimidine 5′ and 3′ tandem flanking bases to a central 1,N 6-ethenoadenine (εA). This adduct is efficiently removed by mammalian APNGs (44Singer B. Antoccia A. Basu A.K. Dosanjh M.K. Fraenkel-Conrat H. Gallagher P.E. Kusmierek J.T. Qiu Z.-H. Rydberg B. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 9386-9390Crossref PubMed Scopus (103) Google Scholar, 45Saparbaev M. Kleibl K. Laval J. Nucleic Acids Res. 1995; 23: 3750-3755Crossref PubMed Scopus (214) Google Scholar, 46Hang B. Singer B. Margison G.P. Elder R.H. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 12869-12874Crossref PubMed Scopus (127) Google Scholar), which, in common with most repair enzymes, require a double-stranded duplex for activity (47Friedberg E. Walker G.C. Seide W. DNA Repair and Mutagenesis. ASM Press, Washington, D. C.1995Google Scholar). We have now found a connection between the rate and extent of the APNG-mediated cleavage of εA-containing 15-mers and the thermal stability of these duplexes. Our work appears to be the first correlation between sequence-dependent glycosylase repair and degree of denaturation of the oligomer substrates. This may provide insights into sequence specificity of in vivo repair of εA or other adducts requiring a stable duplex structure. The present work uses εA as a representative mutagenic adduct (48Singer B. Abbott L.G. Spengler S.J. Carcinogenesis. 1984; 5: 1165-1171Crossref PubMed Scopus (56) Google Scholar, 49Basu A.K. Wood M.L. Niedernhofer L.J. Ramos L.A. Essigmann J.M. Biochemistry. 1993; 32: 12793-12801Crossref PubMed Scopus (204) Google Scholar, 50Pandya G.A. Moriya M. Biochemistry. 1996; 35: 11487-11492Crossref PubMed Scopus (182) Google Scholar) which has been shown to be produced both exogenously and endogenously (51Marnett L.J. Burcham P.C. Chem. Res. Toxicol. 1993; 6: 771-785Crossref PubMed Scopus (276) Google Scholar). HeLa cells were obtained from Cell Culture Center, Endotronics Inc., Minneapolis (MN), a National Institutes of Health sponsored facility. The [γ-32P]ATP (specific activity 6000 Ci/mmol) was purchased from Amersham. T4 polynucleotide kinase was purchased from U. S. Biochemical (Cleveland, OH). 1,N 6-Ethenodeoxyadenine phosphoramidite was obtained from Glen Research (Sterling, VA). Formamide, spermidine, bovine serum albumin, 40% acrylamide/bis solution (19:1), and urea were obtained from Sigma. Poly(dI-dC) was purchased from Pharmacia. OPC cartridges were from Applied Biosystems. The purified 26-kDa truncated human APNG (0.58 mg/ml) was a gift from Dr. T. R. O'Connor. It was shown to be more heat stable than the full-length human protein (52O'Connor T.R. Nucleic Acids Res. 1993; 21: 5561-5569Crossref PubMed Scopus (135) Google Scholar). In addition, O'Connor reported that there is no significant differences in the kinetic parameters of the full-length human APNG and the truncated form (52O'Connor T.R. Nucleic Acids Res. 1993; 21: 5561-5569Crossref PubMed Scopus (135) Google Scholar). The purified 31-kDa mouse APNG protein (0.8 mg/ml) which lacks 48 residues from the amino terminus of the wild-type (wt) protein without loss of activity (53Roy R. Brooks C. Mitra S. Biochemistry. 1994; 33: 15131-15140Crossref PubMed Scopus (51) Google Scholar) was a gift from Drs. R. Roy and S. Mitra. As with the truncated human APNG, there was no difference in kinetic properties of releasing m3A and m7G due to size (53Roy R. Brooks C. Mitra S. Biochemistry. 1994; 33: 15131-15140Crossref PubMed Scopus (51) Google Scholar). The purity of both proteins was found to be homogeneous by SDS-polyacrylamide gel electrophoresis analysis. The human 5′ AP endonuclease (HAP1) (0.25 mg/ml) was a gift from Dr. I. D. Hickson. The crude HeLa cell-free preparation was used as a source for wild-type APNG protein activity and also an overall measurement of repair in the cell. The preparation of crude extracts from HeLa cells was carried out essentially as described by Singer et al.(44Singer B. Antoccia A. Basu A.K. Dosanjh M.K. Fraenkel-Conrat H. Gallagher P.E. Kusmierek J.T. Qiu Z.-H. Rydberg B. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 9386-9390Crossref PubMed Scopus (103) Google Scholar). The ammonium sulfate precipitates, dissolved in a buffer containing 25 mm Hepes-KOH, pH 7.8, 0.5 mmEDTA, 0.125 mm phenylmethylsulfonyl fluoride, 3 mm β-mercaptoethanol, and 10% glycerol, were used in enzyme assays. The 15-nucleotide long oligodeoxynucleotides with a central A or εA were synthesized on a 1 μm scale and purified as described previously (21Litinski V. Chenna A. Sági J. Singer B. Carcinogenesis. 1997; 18: 1609-1615Crossref PubMed Scopus (23) Google Scholar). The following 15-mer oligomers were used in this study: 5′-AGCGGNNXNNGAGCT-3′, where -NNXNN- are: -GGεAGG-, -CCεACC-, -AAεAAA-, and -TTεATT-. All control sequences were the same 15-mers in which A replaced εA in -NNXNN-. The complementary oligodeoxynucleotides synthesized contained a thymine opposite εA or A. Two complete sequences are shown in the form of duplexes (Fig.1). The sequences in common are outside the box and are shown as hydrogen bonded (•). The boxed central sequence is presented with unknown strength of hydrogen bonds (○) since it is possible that the region at the εA·T mismatch can be destabilized as reported for normal base mismatches and their neighbor bases (54Arnold F.H. Wolk S. Cruz P. Tinoco Jr., I. Biochemistry. 1987; 26: 4068-4075Crossref PubMed Scopus (62) Google Scholar, 55Borden K.L. Jenkins T.C. Skelly J.V. Brown T. Lane A.N. Biochemistry. 1992; 23: 5411-5422Crossref Scopus (49) Google Scholar, 56Lane A. Martin S.R. Ebel S. Brown T. Biochemistry. 1992; 31: 12087-12095Crossref PubMed Scopus (47) Google Scholar). The 15-mer oligonucleotides were 5′-end radiolabeled and annealed to their complementary strands as described by Rydberg et al. (57Rydberg B. Dosanjh M.K. Singer B. Proc. Natl. Acad. Sci. U. S. A. 1991; 86: 6839-6842Crossref Scopus (34) Google Scholar). Briefly, the εA- or A- containing oligonucleotides were 5′-end labeled with [γ-32P]ATP and T4 polynucleotide kinase in a kinase buffer containing 50 mm Hepes-KOH, pH 7.5, 10 mmβ-mercaptoethanol, 10 mm MgCl2 at 37 °C for 35 min. The labeled oligomers were then annealed to complementary 15-mers (1.5-fold molar excess) in a buffer containing 10 mm Hepes-KOH, pH 7.5, 100 mm NaCl by slowly cooling down from 90 °C to room temperature (1 h). The enzymatic assay used to examine APNG-mediated cleavage of oligonucleotide substrates was carried out essentially as described by Rydberg et al. (57Rydberg B. Dosanjh M.K. Singer B. Proc. Natl. Acad. Sci. U. S. A. 1991; 86: 6839-6842Crossref Scopus (34) Google Scholar, 58Rydberg B. Qiu Z.-H. Dosanjh M.K. Singer B. Cancer Res. 1992; 52: 1377-1379PubMed Google Scholar). The standard reaction was performed in a total volume of 10 μl in 35 mm Hepes-KOH, pH 7.8, 0.5 mm EDTA, 0.5 mm dithiothreitol, 0.5 mm spermidine, 50 mm KCl, 0.6 mmMgCl2, 400 μg of bovine serum albumin, 10% glycerol with 20 fmol of the 15-mer duplex. In the reactions when crude HeLa extracts were used, 0.5 μg of poly(dI-dC) was added and bovine serum albumin was omitted. The reactions were stopped by addition of equal volume of gel loading buffer containing 90% formamide, 50 mm EDTA, and 0.05% bromphenol blue, heated at 95–100 °C for 2 min. This treatment yielded virtually identical amounts of cleavage of AP sites generated by the action of APNG on εA as that produced by the human 5′ AP endonuclease (HAP1) cleavage of the AP sites (Fig.2). Note that the protein concentration of HAP1 used was greatly in excess (25 ng) over that normally needed to cleave the same amount of AP sites (59Hang B. Rothwell D.G. Sági J. Hickson I.D. Singer B. Biochemistry. 1997; 36: 15411-15418Crossref PubMed Scopus (26) Google Scholar). The nonenzymatic method was used for all the cleavage reactions in this work. Electrophoresis of the reaction mixtures was carried out using a 12% polyacrylamide gel containing 8 m urea. The bands corresponding to the cleavage products and the remaining uncut substrates were scanned and quantitated using a Molecular Dynamics PhosphorImager (Molecular Dynamics, Sunnyvale, CA). The curve fitting used the MicroCal Origin program (version 3.0). For the temperature-dependent experiments the standard enzyme reaction assay described above was used with differing reaction temperatures, 0–45 °C. In one type of experiment, a single reaction mixture was divided into six aliquots and each was incubated at a different temperature for 20 min with 2.5 ng of human APNG. 0 °C indicates on ice, 10 and 20 °C were controlled by adding ice to the water bath to maintain the required temperatures. Samples at 30, 40, and 45 °C were kept in water baths. All incubations were for 20 min with 2.5 ng of human APNG. In the second experiment, samples under standard conditions with 2.5 ng of human APNG were incubated at 20 °C for the indicated times (20 or 100 min) and then half of each sample continued an additional 20 min at 37 °C. Quantitation of APNG-mediated cleavage was by the standard gel electrophoresis which was scanned by a PhosphorImager as described under "Enzymatic Reaction." Molar absorbance of the single strands containing only natural nucleotides was calculated by using the DNA/Oligo Quantitation Software of a Beckman DU 7400 diode-array spectrophotometer. In the case of the εA-containing oligonucleotides, the nearest neighbor interactions (60Borer P.N. Fasman G.D. 3rd Ed. Handbook of Biochemistry and Molecular Biology: Nucleic Acids. 1. CRC Press, Cleveland, OH1975: 589Google Scholar) were recalculated using the molar absorption coefficient of 5000 for εA, which was calculated from published spectral data (61Secrist III, J.A. Barrio J.R. Leonard N.J. Weber G. Biochemistry. 1972; 11: 3499-3506Crossref PubMed Scopus (496) Google Scholar). Molar absorbance of the εA-containing 15-mers was found to be lower by an average of 6%, as compared with the A-containing corresponding sequences. The double strands were prepared by mixing equimolar amounts of the non-self-complementary single-strands in the buffer containing 0.1m NaCl, 0.01 m sodium phosphate, and 0.1 mm EDTA, pH 7.0. These conditions differ from those used in enzymatic assays primarily in that the oligomer concentration is higher. Thermal transition profiles were measured using a Beckman DU 7400 diode-array spectrophotometer at 260 or 280 nm. A buffer containing 0.1m NaCl, 0.01 m sodium phosphate, and 0.1 mm EDTA, pH 7.0, was used in all thermal denaturation experiments. For the determination of the melting curve at a single duplex concentration or for the analysis of the premelting region of the melting profiles a Beckman special Tm-6-cell holder was used. This has an internal thermometer, a Peltier temperature controller and is equipped with gas inlet accessory. The latter was used to flush the cell holder with nitrogen when measurements were started below 16 °C. The path length was 1 cm and the volume of samples was 0.32 ml. For the concentration dependence measurements, the Beckman normal 6-cell holder was used as described before (62Sági J. Chenna A. Hang B. Singer B. Chem. Res. Toxicol. 1998; 11: 329-334Crossref PubMed Scopus (9) Google Scholar). In both cases linear heating was used from 10 or 20 to 90 °C. The ramp rate was 0.2 °C/min in the range of ± 20 °C of theT m of the samples and 0.5 °C/min at other temperatures. Absorption values were measured at 0.5 °C intervals.T m values were obtained from the melting curves with the use of the MeltWin program (63McDowell J.A. Turner D.H. Biochemistry. 1996; 35: 14077-14089Crossref PubMed Scopus (326) Google Scholar). Enthalpy (−ΔH O) and entropy (−ΔS O) data for duplex formation of the 15-mers were obtained in two ways: from the 1/T m − ln Ct plots (Fig. 3), where Ct is the total strand or duplex concentration, and from the melting profiles by using MeltWin, version 3.0. This program fits the shape of each curve to the two-state model with sloping base lines using a nonlinear least-square program (63McDowell J.A. Turner D.H. Biochemistry. 1996; 35: 14077-14089Crossref PubMed Scopus (326) Google Scholar). For the former method six duplex concentrations were used which ranged from 1 to 125 μM and measurements were carried out as described before (62Sági J. Chenna A. Hang B. Singer B. Chem. Res. Toxicol. 1998; 11: 329-334Crossref PubMed Scopus (9) Google Scholar). From the parameters obtained from the 1/T m − ln Ctplots, only the −ΔH O values are shown in Table I. These were the averages of three to six determinations of separate samples. For the shape analysis with the melt curve processing program MeltWin the melting profiles were truncated at 20 and 90 °C.Table IThermodynamic parameters for the melting of the 15-mer duplexesCentral sequenceData from melting curve analysis with MeltWinData from 1/Tm − ln CtT m(°C)ΔTm(°C)−ΔH° (kcal/mol)−ΔS° (cal/mol · K)−ΔG°37(kcal/mol)ΔΔG° (kcal/mol)−ΔH° (kcal/mol)AAAAA55.7103.8289.714.0121.5AAɛAAA44.9−10.878.3220.310.0−4.093.9TTATT56.8100.1276.414.3115.8TTɛATT47.6−9.277.3214.910.5−3.8102.8GGAGG67.586.2227.015.8101.6GGɛAGG59.4−8.178.8211.413.3−2.596.2CCACC69.696.7255.217.4117.6CCɛACC62.2−7.480.4214.713.8−3.6102.0The measurements were carried out in 0.1 m NaCl, 0.01m sodium phosphate, and 0.1 mm EDTA (pH 7.0). The values shown are the averages of three to six separate determinations. The T m values shown were determined at a duplex concentration of 8 μm. Standard deviation of the T m values averaged ± 0.5 °C. The average standard deviation for thermodynamic parameters of the MeltWin analysis was ± 3.6 kcal/mol for ΔH°, ± 10.6 cal/mol · K for ΔS°, ± 0.21 kcal/mol for ΔG°37, and ± 3.1 kcal/mol for ΔH° obtained from the 1/T m − lnC t plots. Open table in a new tab The measurements were carried out in 0.1 m NaCl, 0.01m sodium phosphate, and 0.1 mm EDTA (pH 7.0). The values shown are the averages of three to six separate determinations. The T m values shown were determined at a duplex concentration of 8 μm. Standard deviation of the T m values averaged ± 0.5 °C. The average standard deviation for thermodynamic parameters of the MeltWin analysis was ± 3.6 kcal/mol for ΔH°, ± 10.6 cal/mol · K for ΔS°, ± 0.21 kcal/mol for ΔG°37, and ± 3.1 kcal/mol for ΔH° obtained from the 1/T m − lnC t plots. Preliminary data on extent of cleavage of four 25-mer εA-containing oligomers with the same central sequences exhibited a definitive sequence dependence of cleavage of εA when using human APNG (64Singer B. Hang B. Chem. Res. Toxicol. 1997; 7: 713-732Crossref Scopus (126) Google Scholar). The order of cleavage efficiency was -GGεAGG- or -CCεACC- > -AAεAAA- or -TTεATT-. The sequence -GGεATT- in the same 25-mer gave an intermediate level of cleavage. The length of these duplexes was considered to be too long for precise determination of the role of thermal stability in repair. Therefore, in this study the same flanking doublets were kept and both ends of the 15-mers truncated by 5 nucleotides in the new substrates, yielding the 15-mer sequences shown under "Experimental Procedures" (Fig.1). In order to show both the extent and rate of repair of the 15-mer duplexes with εA flanked by differing neighbor bases, cleavage with human APNG was determined as a function of enzyme concentration for 20 min (Fig. 4 A) and of time using 2.5 ng of human APNG (Fig. 4 B). The 20-min time and 2.5 ng of APNG, which was within the linear portion of these cleavage curves, was used for all subsequent experiments involving human APNG. A representative gel is shown in Fig. 5which presents the cleavage of εA-containing oligomers after reaction with APNG or APNG + HAP1. Note that lanes 5–8 have a double-band on cleavage presumably resulting from the β-elimination mechanism due to high pH and high temperature in the treatment of samples (65Doetsch P.W. Cunningham R.P. Mutat. Res. 1990; 236: 173-201Crossref PubMed Scopus (328) Google Scholar, 66Neddermann P. Jiricny J. J. Biol. Chem. 1993; 268: 21218-21224Abstract Full Text PDF PubMed Google Scholar). When HAP1 is added to the APNG reaction (lanes 9–12) there is only a single band seen which is the result of cleavage of the phosphodiester bond 5′ to the AP site (65Doetsch P.W. Cunningham R.P. Mutat. Res. 1990; 236: 173-201Crossref PubMed Scopus (328) Google Scholar). The 15-mers were in two classes in terms of repair efficiency: G·C or C·G neighbor pairing and A·T or T·A neighbor pairing. The A-containing control 15-mers were not cleaved in the presence of human APNG (data not shown). The εA-containing duplexes with G·C or T·A flanking bases were cleaved by wild-type crude HeLa extracts which contain full-length human APNG and the same preference for oligomers with G·C pairs was found (Fig. 6). Similar results of differential cleavage were also obtained with a purified cloned mouse APNG (Fig. 7). APNG, as expected, did not cleave the four single-stranded εA-containing 15-mers (data not shown).Figure 5Autoradiogram of cleavage of the four εA-containing oligomers using buffer only (lanes 1–4), 2.5 ng of human APNG (lanes 5–8), and 2.5 ng of human APNG plus 25 ng of HAP1 (lanes 9–12). The positions of the cleavage products are indicated on the right. Note that both intact and cleavage product show different mobilities as a result of different neighbor tandem bases. All the reactions were performed at 37 °C for 20 min.View Large Image Figure ViewerDownload (PPT)Figure 6Cleavage efficiency of εA-containing oligom
Referência(s)