Synthesis and Cleavage of Oligodeoxynucleotides Containing a 5

Nov 17, 1997 - To facilitate biochemical and biophysical studies aimed at elucidating the effects of this modification in DNA, we have developed a sol...
0 downloads 10 Views 134KB Size
Download PDF
1254

Chem. Res. Toxicol. 1997, 10, 1254-1258

Synthesis and Cleavage of Oligodeoxynucleotides Containing a 5-Hydroxyuracil Residue at a Defined Site June Fujimoto, Linh Tran, and Lawrence C. Sowers* Division of Pediatrics, City of Hope National Medical Center, Duarte, California 91010 Received June 11, 1997X

Oxidation and hydrolysis of a cytosine residue can lead to the formation of 5-hydroxyuracil in DNA. The biological consequences of this modification are not fully understood. To facilitate biochemical and biophysical studies aimed at elucidating the effects of this modification in DNA, we have developed a solid-phase synthetic method for the placement of 5-hydroxyuracil residues at defined sites in oligodeoxynucleotides. This method is based upon the enhanced acidity of the 5-hydroxyl proton which allows selective aqueous acetylation. Under standard aqueous ammonia deprotection conditions, however, we observed that 5-hydroxyuracil residues are lost substantially from synthetic oligonucleotides. Substitution of aqueous ammonia with methanolic potassium carbonate and the use of phosphoramidite derivatives with alternatively protected amino groups allow synthesis of oligonucleotides containing 5-hydroxyuracil and all normal bases in high yield. The composition of the oligodeoxynucleotides prepared by this method has been verified by enzymatic digestion followed by high-performance liquid chromatography (HPLC) analysis as well as acid hydrolysis followed by GC/MS analysis. The location of the 5-hydroxyuracil residue is demonstrated by selective permanganate oxidation of the 5-hydroxyuracil residue followed by β-elimination. We have also probed a synthetic oligonucleotide containing a unique 5-hydroxyuracil residue with uracil DNA N-glycosylase, previously reported to remove this lesion from DNA.

Introduction Oxidation of cytosine residues can lead to the formation of 5-hydroxyuracil (HOU) residues in DNA. Dizdaroglu (1) originally identified HOU in the acid hydrosylate of γ-irradiated DNA. It was proposed that the initial adduct was cytosine glycol which would deaminate, forming uracil glycol. The HOU identified by the GC/MS method was then attributed to the dehydration of uracil glycol, after release from DNA, under hydrolysis or derivatization conditions (2). Subsequently, Ames and co-workers identified 5-hydroxy-2′-deoxyuridine in enzymatic hydrolysates of γ-irradiated DNA by HPLC and electrochemical detection, establishing that HOU could indeed be generated in DNA (3). Previously, HOU residues have been incorporated into oligonucleotides biosynthetically (4-6). In 1970, RoyBurman and co-workers demonstrated that HO-dUTP could serve as an alternative substrate for DNA polymerase, substituting specifically for dTTP but with reduced efficiency (4). More recently, Wallace and coworkers incorporated HOU residues into oligonucleotides at selected sites using a combination of terminal deoxynucleotidyltransferase and T4 DNA ligase (6). It was demonstrated in one sequence context that HOU could pair with adenine and, in another, with cytosine during DNA polymerase-catalyzed DNA replication (7). The mechanism for such mispair formation is not yet known. Using oligonucleotides with a unique HOU residue, this group also presented evidence that HOU is removed by uracil glycosylase, endonuclease III, and FPG protein (8). In contrast, others have presented evidence that HOU is not removed by endonuclease III (9), FPG protein (10), or Escherichia coli uracil DNA N-glycosylase (11). * Author to whom correspondence should be addressed. Phone: (818) 359-8111, ext. 3845. Fax: (818) 301-8458. E-mail: Isowers@ smtplink.coh.org. X Abstract published in Advance ACS Abstracts, October 15, 1997.

S0893-228x(97)00102-1 CCC: $14.00

In order to provide substrates which will allow structural insights into the mispairing of HOU in DNA as well as potentially to reconcile some of the conflicting glycosylase results, we have developed a method for the specific placement of HOU residues in oligonucleotides by chemical synthesis. The method reported here is clean and efficient and provides oligonucleotides in sufficient quantity and purity for subsequent biophysical and biochemical studies.

Experimental Section Materials. 2′-Deoxyuridine was obtained from Chem-Impex, Inc. Silica gel H was obtained from Fluka. Phosphoramidites of the normal DNA bases were obtained from Glen Research, including phenoxyacetyl dA (PAc-dA), 4-isopropylphenoxyacetyl dG (iPr-PAc-dG), and acetyl dC (Ac-dC), monomers which can be deprotected under very mild conditions. Fluorescein was added to the 5′-end of the oligonucleotides using a fluorescein phosphoramidite, also from Glen Research. Uracil DNA Nglycosylase (E. coli, cloned) was obtained from Amersham. Synthesis. 5-Hydroxy-2′-deoxyuridine. 2′-Deoxyuridine was converted to 5-hydroxy-2′-deoxyuridine by the method of Podrebarac and Cheng (12). A 100-mL round-bottom flask was charged with 2′-deoxyuridine (4 g, 17.5 mmol) and distilled water (25 mL). Bromine (∼4 mL) was added dropwise with stirring until a permanent yellow color persisted. Air was bubbled through in order to remove excess bromine until the color of the solution cleared. The flask was cooled to 0 °C, and pyridine (12 mL) was added dropwise. The solution was stirred at 0 °C for 30 min and then at room temperature overnight. Silica gel TLC (10% methanol/90% dichloromethane) indicated that the reaction was complete. Solvents were removed under reduced pressure, and the product was obtained by silica gel chromatography in 12% methanol/88% dichloromethane in 87% yield. The HO-dU prepared by this method was chromatographically pure by TLC. The purity of the HO-dU was verified further by HPLC and GC/MS. In particular, we wished to verify that the HO-dU was free of the parent 2′-deoxyuridine which could interfere with subsequent glycosylase assays. We deter-

© 1997 American Chemical Society

Synthesis of 5-Hydroxyuracil Oligodeoxynucleotides mined by GC/MS that our HO-dU preparation contained less than 0.1% 2′-deoxyuridine. The UV spectral characteristics of HO-dU were consistent with those previously reported (13). Exact mass determination (FAB, negative ion): calcd, 243.0618; obsd, 243.0617. 1H NMR (300 MHz, DMSO-d6): δ 11.41 (s, NH), 8.64 (s, 5-OH), 7.33 (s, H6), 6.18 (t, H1′, J ) 6.0 Hz), 5.22 (d, 3′OH), 5.01 (m, 5′OH), 4.22 (m, H3′), 3.75 (m, H4′), 3.54 (m, H5′,H5′′), 2.06-2.00 (m, H2′, H2′′). 5-Acetoxy-2′-deoxyuridine. The 5-hydroxyl group of 5-hydroxy-2′-deoxyuridine was selectively protected by the method of Rabi and Fox (14). A round-bottom flask was charged with 5-hydroxy-2′-deoxyuridine (1 g, 4.1 mmol). To this flask was added an aqueous sodium carbonate solution (16.67 M, 20 mL) followed by dropwise addition of acetic anhydride (0.5 mL, 5.33 mmol, 1.3 equiv). Bubbling began, and the reaction was complete after 5 min as indicated by TLC. Solvents were removed under reduced pressure, and the product was obtained as a white solid by silica gel chromatography in 8% methanol/ 92% dichloromethane in 81% yield. 1H NMR (300 MHz, DMSOd6): δ 11.74 (s, NH), 8.00 (s, H6), 6.14 (t, H1′, J ) 6.0 Hz), 5.26 (d, 3′OH), 5.08 (m, 5′OH), 4.21 (m, H3′), 3.78 (m, H4′), 3.55 (m, H5′, H5′′), 2.22 (s, CH3), 2.10 (m, H2′, H2′′). 5-Acetoxy-5′-O-(4,4′-dimethoxytrityl)-2′-deoxyuridine. The protected 5-hydroxy-2′-deoxyuridine was converted to the fully protected phosphoramidite derivative by standard methods (15). A 100-mL round-bottom flask was charged with 5-acetyl2′-deoxyuridine, and the material was dried by coevaporation with distilled pyridine two times. Pyridine was then added, and the solution was purged with argon for 10 min. (Dimethylamino)pyridine (75 mg, 0.61 mmol, 0.2 equiv) was added followed by dimethoxytrityl chloride (1.35 g, 4.0 mmol, 1.3 equiv). The solution was stirred at room temperature for 15 min. Solvents were removed under reduced pressure, and then the crude material was dissolved in dichloromethane. The organic solution was washed with aqueous sodium bicarbonate, brine, and distilled water, one time each. The organic phase was dried over sodium sulfate and filtered, and solvents were removed under reduced pressure. The product was obtained by silica gel chromatography in 1% triethylamine/99% dichloromethane followed by 3% methanol/97% dichloromethane to produce a white foam in 73% yield. 1H NMR (300 MHz, DMSO-d6): δ 7.79 (s, H6), 7.38-6.86 (m, trityl), 6.17 (s, H1′), 5.37 (m, 3′OH), 4.33 (m, H3′), 3.89 (m, H4′), 3.74 (s, OCH3), 3.35 (m, H5′, H5′′), 2.22 (m, H2′, H2′′), 1.95 (s, CH3). 5-Acetoxy-5′-O-(4,4′-dimethoxytrityl)-2′-deoxyuridine 3′O-(2-Cyanoethyl)-N,N-diisopropylphosphoramidite. A 100mL round-bottom flask was charged with 5-acetyl-5′-O-(4,4′dimethoxytrityl)-2′-deoxyuridine (1.30 g, 2.21 mmol), and the material was dried by coevaporation with toluene two times. Dichloromethane (22 mL) was added, and the solution was purged with argon for 10 min. Diisopropylethylamine (1.54 mL, 8.84 mmol, 4 equiv) was added followed by dropwise addition of 2-cyanoethyl N,N-diisopropylchlorophosphoramidite (0.74 mL, 3.31 mmol, 1.5 equiv). The yellow solution was stirred at room temperature for 1 h. The reaction was determined to be complete by TLC. Solvents were removed under reduced pressure, and then the crude material was dissolved in dichloromethane. The solution was washed with saturated sodium bicarbonate, brine, and distilled water, one time each. The organic phase was dried over sodium sulfate and filtered, and solvents were removed under reduced pressure. The product was purified by gel chromatography using a 3.5-cm × 6.5-cm column of silica gel H in 10% triethylamine/90% toluene to yield 1.13 g of a cream-colored foam in 65% yield. 1H NMR (300 MHz, DMSO-d6): δ 11.09 (s, NH), 7.82 (s, H6), 7.39-6.86 (m, trityl), 6.16 (t, H1′), 4.56 (m, H3′), 4.03 (m, H4′), 3.73 (s, OCH3), 3.633.53 (m, H5′, H5′′, NC-CH2, N-CH), 2.77-2.66 (CH2-O-P, JHP ) 34.77 Hz), 1.96 (s, CH3), 2.40 (m, H2′, H2′′), 1.25-0.84 (m, isopropyl). Oligonucleotide Synthesis and Purification. Oligonucleotides were synthesized using a Pharmacia gene assembler. The 5-hydroxyuracil phosphoramidite was placed in one of the extra ports. Synthesis was conducted using the

Chem. Res. Toxicol., Vol. 10, No. 11, 1997 1255 standard coupling cycles without removal of the final dimethoxytrityl group. Monomers which were deprotected under mild conditions were used for 5-hydroxyuracil-containing oligonucleotides, including PAc-dA, iPr-PAc-dG, and Ac-dC. Deprotection of oligonucleotides containing 5-hydroxyuracil was conducted in methanolic potassium carbonate at room temperature for 4 h (16). Following deprotection, the methanol was removed under reduced pressure and the oligonucleotides were dissolved in 1 M triethylammonium acetate buffer, pH 7.0. Tritylcontaining oligonucleotides were isolated using HPLC on a Hamilton PRP column and a gradient of acetonitrile in aqueous triethylammonium acetate as the mobile phase. Tritylated oligonucleotides were then detritylated using 80% aqueous acetic acid (room temperature, 30 min) and repurified using a Pharmacia HR5/5 anion-exchange column eluted with a gradient of aqueous potassium chloride in potassium phosphate buffer. Oligonucleotides with a 5′-fluorescein tag were prepared on the synthesizer using a fluorescein phosphoramidite containing a 5-dimethoxytrityl group and purified as described previously (17). Enzymatic Digestion and HPLC Analysis. A portion of each purified oligonucleotide was digested with nuclease P1 (200 µL, 20 mM sodium acetate buffer, pH 4.8, 37 °C, 1 h) followed by bacterial alkaline phosphatase (0.1 M Tris-HCl, pH 7.5, 37 °C, 1 h). The liberated deoxynucleosides were separated by HPLC on a Supelco Supelcosil-LS-C18 reverse-phase column eluted with a methanol gradient in aqueous sodium phosphate buffer, pH 5.0. Deoxynucleosides were detected with an LKB diode array detector. Acid Hydrolysis and GC/MS Analysis. A portion of each purified oligonucleotide was hydrolyzed in 88% formic acid for 30 min at 150 °C. Following removal of formic acid, the liberated free bases were silylated with bis(trimethylsilyl)trifluoroacetamide in acetonitrile. Gas chromatography was performed with a Hewlett-Packard 5890 gas chromatograph using conditions previously reported (1). The gas chromatograph was interfaced directly to a Hewlett-Packard 5970 mass spectral detector. Gel Electrophoresis and Chemical and Enzymatic Cleavage. The purity of the isolated oligonucleotides was verified by gel electrophoresis on 20% polyacrylamide denaturing gels. In order to probe for chemical cleavage of the HOU residue, portions of the HOU-containing oligonucleotide (2 nmol) were treated and then examined by gel electrophoresis. Piperidine cleavage was performed in 100 µL of 1 M piperidine, at 90 °C for 30 min or at 37 °C for 15 min. To examine cleavage in ammonia, oligonucleotides were treated with 100 µL of 30% concentrated ammonia with or without 0.25 M mercaptoethanol at 60 °C, overnight. Permanganate cleavage was performed by resuspending the labeled oligonucleotide in 32 µL of distilled water followed by 8 µL of a 0.3 M aqueous potassium permanganate solution. Following incubation at room temperature for approximately 12 min, 10 µL of allyl alcohol was added. In all cases, the samples were dried under reduced pressure, resuspended in 10 µL of loading buffer containing urea-saturated formamide, and extracted with 5 µL of ethyl acetate. To probe for uracil DNA N-glycosylase activity, 2-nmol portions of the oligonucleotides were resuspended in 14 µL of distilled water. Buffer solution (10 mM Tris-HCl, pH 8.0, 1 mM EDTA, 1 mM DTT) was added followed by 1 unit of uracil DNA N-glycosylase. Samples were incubated at 37 °C for 60 min. Samples were dried under reduced pressure and treated with piperidine (37 °C, 15 min) as described above. The resulting oligonucleotides were examined by gel electrophoresis.

Results and Discussion Previously, 5-hydroxyuracil residues have been generated in DNA by either in situ oxidation of cytosine residues or incorporation of 5-hydroxy-2′-deoxyuridine5′-triphosphate biosynthetically. These studies have served well to stimulate interest in such DNA damage products. However, in order to carry out biophysical

1256 Chem. Res. Toxicol., Vol. 10, No. 11, 1997

Fujimoto et al.

Scheme 1. Preparation of the 5-Hydroxyuracil Phosphoramidite

Figure 1. HPLC analysis of the enzymatic digest of a model oligodeoxynucleotide containing HOU, G, T, and A: (A) HPLC chromatogram at 275 nm, (B) UV spectrum of the 5-hydroxy2′-deoxyuridine peak.

studies aimed at elucidating the structural role of specific damages, it is necessary to have efficient methods for the chemical synthesis of defined sequence oligonucleotides containing such lesions. Preparation of a suitable phosphoramidite derivative of 5-hydroxyuracil first requires selective protection of the 5-hydroxyl group (Scheme 1). Here, we used the method of Rabbi and Fox (14), which exploits the enhanced acidity of the 5-hydroxyl proton to acetylate selectively the 5-position in high yield. The acetylated derivative quantitatively regenerates the starting material under mild conditions and is sufficiently stable during subsequent modification of the deoxynucleoside including 5′-tritylation and formation of the 3′-phosphoramidite. Under normal conditions for the deprotection of synthetic oligodeoxynucleotides, the resin is treated with concentrated aqueous ammonia at 60 °C for several hours. These conditions, however, destroy 5-hydroxy-2′deoxyuridine and 5-hydroxyuracil residues in synthetic oligonucleotides. We found that more mild conditions, specifically potassium carbonate in methanol at room temperature (16), successfully deblock 5-acetoxy-2′-deoxyuridine and quantitatively regenerate 5-hydroxy-2′deoxyuridine within minutes. These mild deprotection conditions are incompatible with the standard benzoyl and isobutyryl protecting groups for dC, dA, and dG. However, normal bases with alternative protecting groups have been reported and are now commercially available (see ref 18). To establish the chemistry for the incorporation of HOU into synthetic oligonucleotides, we first prepared a 12-base sequence containing A, T, G, and HOU using the HOU phosphoramidite described above, along with PAc-dA and iBuPAc-dG corresponding to the following sequence, where HOU is 5-hydroxyuracil: 5′-A(HOU)GTA(HOU)GTA(HOU)GT. Deoxycytidine was excluded from this test sequence because 2′-deoxycytidine and 5-hydroxy-2′-deoxyuridine are not well-resolved under the HPLC conditions reported

here. The HOU phosphoramidite coupled with high yield, similar to the coupling efficiencies of the other bases. Following deprotection in methanolic potassium carbonate, the tritylated oligonucleotide was purified by HPLC using a PRP column and then by ion-exchange chromatography following detritylation. The composition of the oligonucleotide was verified by HPLC following enzymatic hydrolysis. As shown in Figure 1A, the HO-dU peak is eluted prior to the normal deoxynucleosides. The chromatographic retention as well as UV spectrum of the HO-dU peak obtained with the diode array detector (Figure 1B) are indistinguishable from those of an authentic standard. No other peaks were observed in the chromatogram, and the integrated areas of the four peaks are consistent with the expected areas based upon composition of the oligonucleotide. The composition of the oligonucleotide was also examined by GC/MS following acid hydrolysis and silylation of the free bases. The silylated HOU peak is observed between thymine and adenine at 11.7 min (Figure 2A). The mass spectrum of the HOU peak (Figure 2B) displays a parent ion at 344 amu as well as a characteristic M methyl peak (329 amu) as previously described (1). Having established a viable synthetic strategy, we then prepared a longer oligonucleotide containing a single HOU residue at the indicated site, where X is fluorescein: 5′-XCGACGGATCCGGTACTCGAGAAGCTTGA(HOU)ATCAGAATTCGCACCT. To prepare this oligonucleotide, we added the remaining acetylated dC phosphoramidite. This oligonucleotide was deprotected and isolated as described above. The composition was similarly confirmed by both HPLC and GC/MS methods. The purity of the full-length oligonucleotide is shown in the gel electropherogram (Figure 3, lanes 1 and 2). We prepared this sequence in order to examine both the chemical reactivity of HOU as well as the enzymatic removal of the HOU residue by uracil DNA N-glycosylase. To examine the reactivity of the HOU residue under alkaline conditions, the HOU-containing oligonucleotide was heated in 1 M piperidine at 90 °C for 30 min. These are the standard conditions for cleavage of oligonucleotides by the chemical sequencing method (19). As shown in Figure 3, lane 3, substantial cleavage is observed at the position of the HOU residue. Similarly, heating the

Synthesis of 5-Hydroxyuracil Oligodeoxynucleotides

Figure 2. GC/MS analysis of the acid hydrolysate of the model oligonucleotide containing HOU, G, T, and A: (A) gas chromatogram of the silylated bases obtained in the scan mode, (B) mass spectrum of the silylated 5-hydroxyuracil peak.

Figure 3. Gel electrophoresis analysis of chemical cleavage of an oligonucleotide containing HOU: lane 1, DMT-on full-length oligo directly after deprotection and prior to HPLC purification; lane 2, control, full-length oligonucleotide; lane 3, oligonucleotide treated with 1 M piperidine at 90 °C, 30 min; lane 4, oligonucleotide heated in concentrated aqueous ammonia at 60 °C, overnight; lane 5, same as lane 3, except containing 0.25 M mercaptoethanol; lane 6, sequential treatment with 0.3 M potassium permanganate followed by 1 M piperidine, 90 °C, 30 min.

oligo in concentrated aqueous ammonia at 60 °C overnight (standard oligonucleotide deprotection conditions) results in substantial cleavage at the HOU residue (Figure 3, lane 4). We therefore used the strategy proposed by Johnson and co-workers (18) for the deprotection of oligonucleotides containing the oxidationsensitive modified base 8-oxo-2′-deoxyguanosine. When the HOU oligonucleotide was heated at 60 °C overnight in concentrated aqueous ammonia containing 0.25 M mercaptoethanol, slight protection of the HOU residue was observed (Figure 3, lane 5), suggesting that oxidation may precede cleavage. Although mercaptoethanol affords some protection from cleavage under standard ammonia deprotection conditions, the use of the milder potassium carbonate-methanol strategy is clearly superior. The electrophoretic mobilities of the oligonucleotides cleaved in piperidine and ammonia are slightly different, which may suggest that the chemical structures of the 3′-oligonucleotide ends differ between the two cleavage

Chem. Res. Toxicol., Vol. 10, No. 11, 1997 1257

conditions. These results confirm that standard oligonucleotide deprotection conditions are incompatible with HOU-containing oligonucleotides. Previously, it has been shown that 5-hydroxyuracil ribonucleosides are readily oxidized (20, 21). Such oxidation would be expected for this semiquinone-type structure. In contrast to the ribonucleoside, however, the glycosidic bond of the deoxynucleoside would be unstable following oxidation. The potential for oxidation of the HOU residue may be of both biological and chemical interest. Verdine and co-workers (22) have exploited the selective cleavage of oligonucleotides containing HOU as part of a strategy to define DNA-protein contacts. Therefore, we treated our oligonucleotide with potassium permanganate (0.3 M) followed by piperidine treatment. As shown in Figure 3, lane 6, the HOU-containing oligonucleotide is cleaved with high selectivity at the location of the HOU residue. We note that the oligonucleotide displayed in Figure 3 is single-stranded. Enhanced selectivity would be expected in duplex structures. These results confirm that the 5-hydroxyuracil lesion is not chemically stable in the presence of some metals (20-22) and suggest that oxidation followed by hydrolysis of the glycosidic bond may be biologically important. Further, the selective cleavage of 5hydroxyuracil residues may prove useful in ligationmediated PCR strategies to map the location and repair of such lesions in vivo (23). The HOU residue in DNA, which is formed from a cytosine residue, would pair predominantly with adenine during subsequent DNA replication and thus generate a transition mutation if unrepaired. Previously, it has been reported that HOU is removed by uracil glycosylase, although with significantly reduced efficiency (8). In contrast, however, another group has reported that HOU is not a substrate for E. coli uracil DNA N-glycosylase (11). Our primary interest was to determine if uracil DNA N-glycosylase could aid in detecting the presence and indicating the location of HOU residues in DNA. The removal of base residues by uracil DNA Nglycosylase generates an abasic site. Subsequent cleavage of the abasic site has been accomplished previously with endonuclease IV which has an associated apurinic/ apyrimidinic phosphodiesterase activity. Although uracil DNA N-glycosylase is very active on single-stranded DNA, the phosphodiesterase activity of endonuclease IV requires a duplex substrate. Piperidine cleavage can also be used to cleave DNA containing abasic sites; however, we demonstrated above (Figure 3, lane 3) that the standard piperidine cleavage conditions (90 °C, 30 min) result in considerable cleavage of the HOU-containing oligonucleotide. We therefore modified the piperidine cleavage conditions and determined that more mild piperidine treatment at 37 °C for 15 min did not cause observable cleavage of the HOU-containing oligonucleotide (Figure 4, lane 2). As a positive control, we prepared an oligonucleotide containing uracil in place of HOU. When treated with uracil DNA N-glycosylase followed by mild piperidine treatment, the uracil-containing oligonucleotide was completely cleaved (Figure 4, lane 6). Under identical conditions, no cleavage of the HOU-containing oligonucleotide was observed (Figure 4, lane 3). Our results with E. coli uracil DNA N-glycosylase are therefore consistent with the results of Zastawny et al. (11) and contrast with those of Wallace and co-workers (8). However, we note that a significant difference

1258 Chem. Res. Toxicol., Vol. 10, No. 11, 1997

Fujimoto et al.

(6)

(7)

(8)

(9)

Figure 4. Gel electrophoresis analysis of oligonucleotides containing either uracil or HOU following treatment with uracil DNA N-glycosylase and piperidine (1 M, 37 °C, 15 min): lane 1, control HOU-containing oligonucleotide; lane 2, control HOUcontaining oligonucleotide treated with piperidine alone; lane 3, HOU-containing oligonucleotide treated with glycosylase followed by piperidine; lane 4, control uracil-containing oligonucleotide; lane 5, uracil-containing oligonucleotide treated with piperidine only; lane 6, uracil-containing oligonucleotide treated with uracil DNA N-glycosylase followed by piperidine.

(10)

(11)

(12)

between the results reported here and those of Wallace and co-workers is that the present study utilized singlestranded DNA whereas the previous study utilized duplex DNA. Studies are currently in progress to examine more closely the enzymatic removal of HOU from DNA. In conclusion, the method reported here provides defined sequence oligonucleotides with HOU residues at selected sites conveniently and in high yield. Such substrates will allow structural characterization of HOUcontaining mispairs, studies of which are currently in progress. The oxidation-induced cleavage of oligonucleotides at HOU sites, catalyzed by endogenous as well as toxic metals, may prove biologically important.

Acknowledgment. This work was supported in part by the National Institutes of Health (GM 41336, GM 50351, and CA 33572).

(13) (14) (15) (16)

(17) (18) (19)

References (20) (1) Dizdaroglu, M. (1985) Application of capillary gas chromatography-mass spectrometry to chemical characterization of radiationinduced base damage of DNA: Implications for assessing DNA repair processes. Anal. Chem. 144, 593-603. (2) Dizdaroglu, M., Holwitt, E., Hagan, M. P., and Blakley, W. F. (1986) Formation of cytosine glycol and 5,6-dihydrocytosine in deoxyribonucleic acid on treatment with osmium tetroxide. Biochem. J. 235, 531-536. (3) Wagner, J. R., Hu, C.-C., and Ames, B. (1992) Endogenous oxidative damage of deoxycytidine in DNA. Proc. Natl. Acad. Sci. U.S.A. 89, 3380-3384. (4) Roy-Burman, S., Roy-Burman, P., and Visser, D. W. (1966) Inhibition of ribonucleic acid polymerase by 5-hydroxyuridine 5′triphosphate. J. Biol. Chem. 241, 781-786. (5) Roy-Burman, S., Roy-Burman, P., and Visser, D. W. (1970) Studies on the effect of triphosphates of 5-aminouridine and

(21)

(22) (23)

5-hydroxydeoxyuridine on ribonucleic acid and deoxyribonucleic acid polymerase. Biochem. Pharmacol. 19, 2745-2756. Hatahet, Z., Purmal, A. A., and Wallace, S. S. (1993) A novel method for the site specific introduction of single model oxidative DNA lesions into oligodeoxynucleotides. Nucleic Acids Res. 21, 1563-1568. Purmal, A. A., Kow, Y. W., and Wallace, S. S. (1994) Major oxidative products of cytosine, 5-hydroxycytosine and 5-hydroxyuracil, exhibit sequence and context-dependent mispairing in vitro. Nucleic Acids Res. 22, 72-78. Hatahet, Z., Kow, Y. W., Purmal, A. A., Cunningham, R. P., and Wallace, S. S. (1994) New substrates for old enzymes. 5-Hydroxy2′-deoxycytidine and 5-hydroxy-2′-deoxyuridine are substrates for Escherichia coli endonuclease III and formamidopyrimidine DNA N-glycosylase, while 5-hydroxy-2′-deoxyuridine is a substrate for uracil DNA N-glycosylase. J. Biol. Chem. 269, 18814-18820. Dizdaroglu, M., Laval, J., and Boiteux, S. (1993) Substrate specificity of the Escherichia coli endonuclease III: Excision of thymine- and cytosine-derived lesions in DNA produced by radiation-generated free radicals. Biochemistry 32, 12105-12111. Karakaya, A., Jaruga, P., Bohr, V. A., Grollman, A. P., and Dizdaroglu, M. (1997) Kinetics of excision of purine lesions from DNA by Escherichia coli Fpg protein. Nucleic Acids Res. 25, 474479. Zastawny, T. H., Doetsch, P. W., and Dizdaroglu, M. (1995) A novel activity of E. coli uracil DNA N-glycosylase. Excision of isodialuric acid (5,6-dihydroxyuracil), a major product of oxidative DNA damage, from DNA. FEBS Lett. 364, 255-258. Podrebarac, E. G., and Cheng, C. C. (1968) 2′-Deoxy-5-hydroxyuridine. Hydroxylation via addition of aqueous bromine. In Synthetic Procedures in Nucleic Acid Chemistry (Zorbach, W. W., Tipson, R. S., Eds.) Vol. 1, pp 412-413, Interscience Publishers, New York. Beltz, R. E., and Visser, D. W. (1955) Growth inhibition of Escherichia coli by new thymidine analogs. J. Am. Chem. Soc. 77, 736-738. Rabi, J. A., and Fox, J. J. (1972) Nucleosides. LXXVI. A synthesis of a carbon-carbon bridged pyrimidine cyclonucleoside. J. Org. Chem. 37, 3898-3901. Gait, M. J. (1984) Oligonucleotide Synthesis. A Practical Approach, IRL Press, Oxford, U.K. Kuijpers, W. H., Kuyl-Yeheskiely, E., van Boom, J. H., and Boeckel, C. A. A. (1993) The application of the AMB protecting group in the solid-phase synthesis of methylphosphonate DNA analogues. Nucleic Acids Res. 21, 3493-3500. Mazurek, M., and Sowers, L. C. (1996) The paradoxical influence of thymine analogues on restriction endonuclease cleavage of oligodeoxynucleotides. Biochemistry 35, 11522-11528. Beaucage, S., and Iyer, R. (1993) The synthesis of modified oligonucleotides by the phosphoramidite approach and their applications. Tetrahedron 49, 6123-6194. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning, a Laboratory Manual, 2nd ed., Cold Spring Harbor Laboratory Press, Plainview, New York. Yanagawa, H., Ogawa, Y., and Ueno, M. (1992) Redox ribonucleosides. Isolation and characterization of 5-hydroxyuridine, 9-hydroxyguanosine, and 8-hydroxyadenosine from Torula yeast RNA. J. Biol. Chem. 267, 13320-13326. Bodepudi, V., Iden, C., and Johnson, F. (1991) An improved method for the preparation of the phosphoramidites of modified 2′-deoxynucleotides: Incorporation of 8-oxo-2′-deoxy-7H-guanosine into synthetic oligomers. Nucleosides Nucleotides 10, 755761. Mascarenas, J. L., Hayabashi, K. C., and Verdine, G. L. (1993) Template-directed interference footprinting of protein-thymine contacts. J. Am. Chem. Soc. 115, 373-374. Tornaletti, S., and Pfeifer, G. P. (1994) Slow repair of pyrimidine dimers at p53 mutation hotspots in skin cancer. Science 263, 1436-1438.

TX970102B