DNA Adducts of Acrolein: Site-Specific Synthesis of an

3-(2-Deoxy-β-d-erythro-pentofuranosyl)-6-hydroxy-5,6,7,8-tetrahydropyrimido[1,2-a]purin-10(3H)-one is formed in low yield by the reaction of acrolein...
0 downloads 0 Views 96KB Size
Download PDF
MAY 2002 VOLUME 15, NUMBER 5 © Copyright 2002 by the American Chemical Society

Articles DNA Adducts of Acrolein: Site-Specific Synthesis of an Oligodeoxynucleotide Containing 6-Hydroxy-5,6,7,8-tetrahydropyrimido[1,2-a]purin-10(3H)-one, an Acrolein Adduct of Guanine Lubomir V. Nechev, Ivan D. Kozekov, Angela K. Brock, Carmelo J. Rizzo, and Thomas M. Harris* Department of Chemistry and Center in Molecular Toxicology, Vanderbilt University, Nashville, Tennessee 37235 Received December 12, 2001

3-(2-Deoxy-β-D-erythro-pentofuranosyl)-6-hydroxy-5,6,7,8-tetrahydropyrimido[1,2-a]purin-10(3H)-one is formed in low yield by the reaction of acrolein with 2′-deoxyguanosine. The nucleoside and an oligodeoxynucleotide containing it have been synthesized. For preparation of the nucleoside 2′-deoxyguanosine was alkylated at the N1 position using 1-bromo-3-butene to give 1-(3-butenyl)-2′-deoxyguanosine. Oxidation with OsO4 and N-methylmorpholine-N-oxide to give the 3,4-dihydroxybutyl adduct followed by oxidation with NaIO4 gave the 1-(3-oxopropyl) adduct which cyclized spontaneously to yield the title compound as a rapidly epimerizing mixture of two diastereomers. Reduction of the nucleoside with NaBH4 gave the unfunctionalized compound plus 1-(3-hydroxypropyl)-2′-deoxyguanosine showing that epimerization was occurring via both the imine and the 1-(3-oxopropyl) adduct. Reduction with NaCNBH3 gave exclusively unfunctionalized 3-(2-deoxy-β-D-erythro-pentofuranosyl)-5,6,7,8-tetrahydropyrimido[1,2-a]purin-10(3H)-one. The phosphoramidite reagent needed for preparation of oligonucleotides was prepared from 1-(3-butenyl)-2′-deoxyguanosine by glycolation after protection of the 3′ and 5′ hydroxyl groups as silyl derivatives. Acetylation of the vicinal hydroxyl groups and the exocyclic amino group followed by removal of silyl protection gave the protected nucleoside. Protection of the 5′ hydroxyl group as the 4,4′-dimethoxytrityl ether followed by phosphitylation with 2-cyanoethyl-N,N,N′,N′-tetraisopropylphosphorodiamidite gave the prosphoramidite reagent which was used to prepare a 12-mer oligodeoxynucleotide.

Introduction Acrolein is ubiquitous in the environment and reported to be a product of endogenous oxidation of lipids (1-4). It is a mutagen; in mammalian cells it produces a * To whom correspondence should be addressed. Telephone: (615) 322-2649. Fax: (615) 322-7591. E-mail: harristm@ toxicology.mc.vanderbilt.edu.

complex mutational spectrum including point mutations at both G:C and A:T sites (5). Acrolein reacts with deoxyguanosine to form two regioisomeric 1,N2 cyclic adducts, the deoxyribonucleosides of 8- and 6-hydroxy5,6,7,8-tetrahydropyrimido[1,2-a]purin-10(3H)-ones 2 and 4 (6, 7). Pyrimidopurinone 2, in which the hydroxyl group is proximal to the purine ring system, probably arises by conjugate addition of N2 of guanine to C3 of acrolein

10.1021/tx010181y CCC: $22.00 © 2002 American Chemical Society Published on Web 04/26/2002

608

Chem. Res. Toxicol., Vol. 15, No. 5, 2002 Scheme 1

followed by cyclization of the resulting N2-(3-oxopropyl) adduct 1 (Scheme 1).2 The 6-hydroxy isomer (4), in which the hydroxyl group is distal to the ring, arises by conjugate addition of N1 to C3 of acrolein followed by cyclization of the resulting 1-(3-oxopropyl) adduct 3.3 Nucleosides 2 and 4 are formed in a ∼2:3 ratio when the reaction with dG is carried out at pH 7.0; 2 becomes the primary adduct when the reaction is carried out in base (8). Evaluation of the mutagenicity of 2, 4, and other adducts that may be formed in DNA requires the availability of oligonucleotides containing site-specific, structurally defined adducts. Oligonucleotides containing 2 and 4 represent synthetic challenges because these adducts are not compatible with the reaction conditions normally utilized for solid-phase oligonucleotide synthesis. Two strategies for the preparation of DNA containing 2 have been developed, one by Johnson and the other by ourselves (9, 10). Both involve periodate cleavage of 1,2diols after assembly and deprotection of the oligonucleotide. Johnson’s method utilizes deoxyguanosine having an appropriately protected 3,4-dihydroxybutyl adduct at the N2 position in the assembly of the oligonucleotide, ours a post-oligomerization strategy in which 3,4-dihydroxybutylamine displaces the halogen atom from 2-fluorodeoxyinosine in the oligonucleotide. No syntheses of oligonucleotides containing adduct 4 have been reported. Studies of the mutagenicity of adduct 2 carried out in two laboratories showed it to have remarkably low mutagenicity in bacteria (11, 12). In mammalian cells, divergent results have been obtained, with one study showing the adduct to be essentially nonmutagenic while a second one found a significant level of mutations (13, 14). The analogous acrolein adduct of cytosine, if formed in DNA, is apparently not stable (15). Consequently, we have turned our attention to adduct 4 as a possible source 1 Abbreviations: DMA, dimethylacetamide; DMF, dimethylformamide, DMAP, 4-(dimethylamino)pyridine; FAB-HRMS, fast atom bombardment-high-resolution mass spectrometry; MALDI-TOF MS, matrix-assisted laser desorption-time-of-flight mass spectrometry; CZE, capillary zone electrophoresis; PAC, phenoxyacetyl. 2 Adducts 2 and 4 are numbered by the IUPAC convention as pyrimido[1,2-a]purin-10(3H)-ones rather than as guanine derivatives. Consequently, the hydroxyl groups on 2 and 4 are at C8 and C6, respectively. 3 The alternative sequence for formation of 4 cannot be excluded which involves imine formation at N2 followed by conjugate addition at N1 and then hydration of the imine. However, the sequence of events shown in Scheme 1 is supported by the fact that crotonaldehyde gives only adducts analogous to 2, whereas methylvinyl ketone gives adducts analogous to both 2 and 4 and 2-cyclohexenone gives an adduct analogous to 4 plus a simple Michael adduct at N2 analogous to 1 (6, 7).

Nechev et al.

of the observed mutations. Herein we report a sitespecific synthesis of an oligonucleotide containing 4. Adducts involving the N1 position of deoxyguanosine are not amenable to the post-oligomerization approach that we employed for preparation of oligonucleotides containing 2. Consequently, in the present case we have employed an adducted nucleoside strategy for the synthesis of oligonucleotides containing 4. Although the primary site of nucleophilicity of deoxyguanosine with simple electrophiles is generally N7, under basic conditions the reactivity is diverted to N1 by formation of the anion (16). We recently reported application of this strategy to the synthesis of the 1-hydroxy-3-buten-2-yl adduct at the N1 position of deoxyinosine and extension of the strategy to an oligonucleotide containing the nucleoside (17). It seemed reasonable that the method could be extended to synthesis of oligonucleotides containing 4.

Experimental Procedures General Methods. 1H NMR spectra were recorded at 400.13 MHz on a Bruker AM400 NMR spectrometer in D2O or DMSOd6. DMSO was distilled under vacuum over CaH2. Other chemicals were used as purchased without further purification. Thin-layer chromatography was performed on silica gel glass plates (Merck, Silica Gel 60 F254, layer thickness 250 µm). The chromatograms were visualized under UV light (254 nm) or by staining with an anisaldehyde/H2SO4 solution, followed by heating. Column chromatography was performed using silica gel (Merck, 70-230 mesh). Oligodeoxynucleotides were synthesized on a Perseptive Biosystems Model 8909 DNA synthesizer using their Expedite reagents with the standard synthetic protocol on a 1-µmol scale. The modified oligodeoxynucleotides were cleaved from the solid support and the exocyclic amino groups were deprotected in a single step using concentrated NH4OH at 60 °C. Enzymatic hydrolysis was carried out in one step as follows: oligonucleotide (0.5 A260 units) was dissolved in 20 µL of buffer (pH 7, 0.01 M Tris‚HCl, 0.01 M MgCl2). DNase I (5 units), alkaline phosphatase (1.7 units), and snake venom phosphodiesterase I, type II (0.02 units) were added and the solution was incubated at 37 °C for 1.5 h. The purification of nucleosides and oligonucleotides and the analysis of reaction mixtures and mixtures of nucleosides obtained from enzymatic degradations were performed on a Beckman HPLC system with a diode array UV detector monitoring at 260 nm using YMC ODS-AQ columns (250 × 4.6 mm i.d., 1.5 mL/min for analysis and 250 × 10 mm i.d., 5 mL/min for purification) with H2O (A) and CH3CN (B) for nucleosides and 0.1 M aq ammonium formate (A) and CH3CN (B) for oligonucleotides. HPLC gradient: 99% A initial mixture, 15 min linear gradient to 90% A, 5 min linear gradient to 80% A, 5 min at 80% A, 3 min linear gradient to 0% A, 2 min at 0% A followed by 3 min linear gradient to the initial conditions. 1-(3-Butenyl)-2′-deoxyguanosine (5). A mixture of deoxyguanosine monohydrate (100 mg, 0.35 mmol) and DMA (3 mL) was heated at 70 °C until all the nucleoside was dissolved (∼10 min). The solution was cooled to room temperature and powdered NaOH (21 mg) was added. The mixture was heated at 70 °C until most of the NaOH had dissolved (∼1 h) and then cooled to room temperature. 1-Bromo-3-butene (57 mg, 0.42 mmol) was added in one portion. The mixture was stirred at room temperature for 1 h and at 60 °C overnight. The solvent was evaporated, and the product was purified by silica gel column (CH3CN:H2O:concentrated NH4OH 90:5:5) to yield 73.5 mg (65%) of 5. 1H NMR (DMSO-d6) δ 7.81 (s, 1H, H-8), 6.99 (bs, 2H, NH2), 6.04 (dd, 1H, H-1′, J1 ) J2 ) 6.1 Hz), 5.75 (m, 1H, CHdC), 5.18 (d, 1H, 3′-OH, J ) 4.04 Hz), 4.93 (m, 2H, CH2d C), 4.85 (t, 1H, 5′-OH, J ) 5.5 Hz), 4.26 (m, 1H, H-3′), 3.94 (dd, 2H, N-CH2, J1 ) J2 ) 7.3 Hz), 3.73 (m, 1H, H-4′), 3.45 (m, 2H,

Acrolein Adduct of Deoxyguanosine H-5′, H-5′′), 2.50 (m, 1H, H-2′′), 2.25 (m, 2H, CH2-CdC), 2.12 (m, 1H, H-2′). FAB-HRMS [M + H]+: calcd 322.1515, found 322.1508. 1-(3,4-Dihydroxybutyl)-2′-deoxyguanosine (6). A solution of nucleoside 5 (23 mg, 0.072 mmol) in 0.20 mL of H2O was added to a mixture of water (0.20 mL), acetone (0.40 mL), N-methylmorpholine-N-oxide (9.4 mg, 0.08 mmol) and ∼1 mg of OsO4. The mixture was stirred overnight at room temperature. The solvents were evaporated, and the crude product was purified by silica gel column chromatography (CH3CN:H2O: concentrated NH4OH 85:10:5) to give 23.4 mg (92%) of 6. 1H NMR (DMSO-d6) δ 7.93 (s,1H, H-8), 7.00 (bs, 2H, NH2), 6.12 (dd, 1H, H-1′, J1 ) 7.8 Hz, J2 ) 6.1 Hz), 5.26 (bs, 1H, 3′-OH), 4.94 (m, 1H, 5′-OH), 4.84 (m, 1H, CH-OH), 4.62 (m, 1H, CH2OH), 4.34 (m, 1H, H-3′), 4.06 (m, 1H, CH2N), 3.98 (m, 1H, CH2N), 3.80 (m, 1H, H-4′), 3.47 (m, 5H, H-5′, H-5′′, 2 × CH2OH, CH-OH), 2.51 (m, 1H, H-2′), 2.21 (m, 1H, H-2′′), 1.75 (m, 1H, C-CH2-C), 1.50 (m, 1H, C-CH2-C). FAB-HRMS [M + H]+: calcd 356.1570, found 356.1578. 3-(2-Deoxy-β-D-erythro-pentofuranosyl)-6-hydroxy-5,6,7,8tetrahydropyrimido[1,2-a]purin-10(3H)-one (6-hydroxy1,N2-propano-2′-deoxyguanosine, 4). Nucleoside 6 (15 mg, 0.042 mmol) was dissolved in a solution of NaIO4 (3 mL, 20 mM). The mixture was stirred for 10-15 min at room temperature; the product was purified by HPLC. It was a ∼1:1 mixture of diastereomers of 4, which equilibrated too rapidly for them to be individually isolated. Yield 12.1 mg (89%). 1H NMR (DMSOd6) δ 8.38 (bs, 1H, NH), 7.93 (s, 1H, H-2), 6.11 (m, 1H, H-1′), 5.93 (m, 1H, 6-OH), 5.26 (bs, 1H, 3′-OH), 4.96 (m, 1H, H-6), 4.92 (m, 1H, 5′-OH), 4.42 (m, 1H, H-8), 4.33 (m, 1H, H-3′), 3.80 (m, 1H, H-4′), 3.50 (m, 3H, H-5′, H-5′′, H-8), 2.55 (m, 1H, H-2′), 2.18 (m, 1H, H-2′′), 1.94 (m, 1H, H-7), 1.75 (m, 1H, H-7). FABHRMS [M + H]+: calcd 324.1308, found 324.1317. Reduction of 4 with NaBH4. A solution of 4 (3.0 mg) in 2.5 mL of phosphate buffer (50 mM, pH 7.0) was treated with NaBH4 (17.5 mg, 50 equiv) at room temperature. The reaction was followed by HPLC. After days 1, 2, and 3, additional NaBH4 (17.5 mg, 50 equiv) was added for a total of 200 equiv. By the fifth day, the starting material had been completely consumed, giving two products (3:1 ratio), which were separated by HPLC. The major component (16.2 min) was assigned on the basis of NMR and mass spectra as 1-(3-hydroxypropyl)-2′-deoxyguanosine (7). 1H NMR (DMSO-d6) δ 7.95 (s, 1H, H-8), 7.17 (bs, 2H, NH2), 6.12 (dd, 1H, H-1′, J ) 6.1 Hz), 4.95-5.33 (bs, 2H, 5′-OH, 3′-OH), 4.35 (m, 1H, H-3′), 3.98 (m, 2H, CH2N in the side chain), 3.81 (m, 1H, H-4′), 3.55 (m, 2H, H-5′, H-5′′), 3.45 (m, 2H, CH2O in the side chain), 2.52 (m, 1H, H-2′), 2.21 (m, 1H, H-2′′), 1.73 (m, 2H, C-CH2-C in the side chain. FABHRMS [M + H]+: calcd 326.1464, found 326.1475. The minor one (19.0 min) was shown to be 1,N2-propano-2′-deoxyguanosine (9) by HPLC comparison with an authentic sample.4 Reduction of Nucleoside 4 with NaCNBH3. A solution of 4 (0.5 mg) and NaCNBH3 (1.5 mg, 5 equiv) in 1 mL of phosphate buffer (50 mM, pH 7.0) was stirred for 3 days at room temperature. After 3 days the temperature was raised to 40 °C, an additional 1.5 mg of NaCNBH3 was added, and stirring continued for 5 more days, at which time HPLC analysis showed the reaction to be complete. Only a single product, 1,N2propanodeoxyguanosine (9), was observed, the identity of which was confirmed by HPLC comparison with an authentic sample. 1-(3-Hydroxypropano)-2′-deoxyguanosine (7) was not detected. 1-(3-Butenyl)-2′-deoxy-3′,5′-O-(1,1,3,3-tetraisopropyldisiloxane-1,3-diyl)guanosine (10). A mixture of 5 (400 mg, 1.24 mmol) and 1H-imidazole (420 mg, 6.2 mmol) was coevaporated with anhydrous DMF (2 × 20 mL). The resulting residue was suspended in anhydrous DMF (5 mL), cooled to 0 °C, and then 1,3-dichloro-1,1,3,3-tetraisopropyl-1,3-disiloxane (480 mg, 1.5 mmol) was added dropwise with stirring. The suspension was stirred overnight to give a turbid solution, which was poured onto ice. The resulting mixture was extracted with CH2Cl2. The 4

The authentic sample of 9 was a gift from L. J. Marnett.

Chem. Res. Toxicol., Vol. 15, No. 5, 2002 609 combined organic extracts were washed with water and dried over Na2SO4, and the solvents were evaporated. The residue was heated in MeOH for a few minutes; MeOH was evaporated, and the mixture was purified by flash chromatography on silica gel (ethyl acetate:hexane 2:1) to give 629.5 mg (90%) of protected nucleoside 10. 1H NMR (DMSO-d6) δ 7.88 (s, 1H, H-8), 7.13 (bs, 2H, NH2), 6.11 (dd, 1H, H-1′, J1 ) 4.1 Hz, J2 ) 7.3 Hz), 5.86 (m, 1H, CdCH), 5.06 (m, 2H, CH2dC), 4.72 (m, 1H, H-3′), 4.06 (m, 2H, CH2N), 3.97 (m, 2H, H-5′, H-5′′), 3.84 (m, 1H, H-4′), 2.75 (m, 1H, H-2′), 2.56 (m, 1H, H-2′′), 2.36 (m, 2H, C-CH2-C), 1.09 (m, 28H, 4 × i-Pr-Si). FAB-HRMS [M + H]+: calcd 564.3038, found 564.3043. 1-(3,4-Dihydroxybut-1-yl)-2′-deoxy-3′,5′-O-(1,1,3,3-tetraisopropyldisiloxane-1,3-diyl)guanosine (11). A solution of 10 (271 mg, 0.48 mmol) in 200 µL of THF was added to a mixture of water (2 mL), THF (2 mL), N-methylmorpholine-Noxide (62.7 mg) and ∼1 mg of OsO4. The mixture was stirred overnight at room temperature. The solvents were evaporated, and the crude product was subjected to a silica gel column (CH3CN:H2O:concentrated NH4OH 90:5:5) to give 249.4 mg (87%) of 11. 1H NMR (DMSO-d6) δ 7.94 (s, 1H, H-8), 7.00 (bs, 2H, NH2), 6.04 (m, 1H, H-1′), 4.81 (d, 1H, CH-OH, J ) 5.2 Hz), 4.66 (m, 1H, H-3′), 4.59 (m, 1H, CH2-OH), 4.05 (m, 1H, CH2-N-CdO), 3.95 (m, 1H, CH2-N-CdO), 3.88 (m, 2H, H-5′, H-5′′), 3.78 (m, 1H, H-4′), 3.45 (m, 1H, CH-OH), 3.35 (m, 2H, CH2-OH), 2.63 (m, 1H, H-2′), 2.44 (m, 1H, H-2′′), 1.73 (m, 1H, C-CH2-C), 1.49 (m, 1H, C-CH2-C), 1.00 (m, 28H, 4 × i-Pr). FAB-HRMS [M + H]+: calcd 598.3092, found 598.3120. 1-(3,4-Diacetoxybut-1-yl)-2′-deoxy-3′,5′-O-(1,1,3,3-tetraisopropyldisiloxane-1,3-diyl)guanosine (12). Nucleoside 11 (550 mg, 0.92 mmol) was coevaporated with anhydrous pyridine (2 × 10 mL) and then mixed with pyridine (40 mL), acetic anhydride (0.86 mL, 10 equiv), triethylamine (1.5 mL), and DMAP (11.1 mg). The mixture was stirred at 55 °C for ∼80 h; the course of the reaction was followed by TLC (CH2Cl2:MeOH 9:1). When TLC analysis indicated the completion of the reaction, the mixture was cooled (ice bath) and methanol (20 mL) was added. After 15 min at room temperature, the solvents were evaporated under reduced pressure. Flash chromatography of the residue on silica gel (EtOAc:hexanes 4:1) gave 559.6 mg (84%) of triacetylated product 12. 1H NMR (DMSO-d6) δ 10.58 (bs, 1H, NH), 8.20 (s, 1H, H-8), 6.20 (dd, 1H, H-1′, J1 ) J2 ) 7.6 Hz), 4.96 (m, 1H, CH-OAc), 4.80 (m, 1H, H-3′), 4.18 (m, 1H, 1 × CH2-OAc), 4.08 (m, 3H, 2 × CH2-N, 1 × CH2-OAc), 3.90 (m, 2H, H-5′, H-5′′), 3.79 (m, 1H, H-4′), 2.78 (m, 1H, H-2′), 2.57 (m, 1H, H-2′′), 2.11 (s, 3H, CH3CO), 2.00 (s, 3H, CH3CO), 1.97 (s, 3H, CH3CO), 1.88 (m, 2H, C-CH2-C), 1.03 (m, 28H, isopropyl). FAB-HRMS for [M + H]+: calcd 724.3409, found 724.3373. 1-(3,4-Diacetoxybutyl)-N2-acetyl-2′-deoxyguanosine (13). Compound 12 was treated with tetrabutylammonium fluoride (3.8 mL, 1.0 M in THF) in THF (5 mL) for 2 h at ambient temperature. The solvent was removed under reduced pressure. Purification by flash chromatography (CH3CN:H2O:concentrated NH4OH, 90:5:5) gave 333.6 mg (96%) of 13 as a white solid. 1H NMR (DMSO-d6) δ 10.56 (bs, 1H, NH), 8.26 (s, 1H, H-8), 6.16 (dd, 1H, H-1′, J1 ) J2 ) 6.7 Hz), 5.26 (d, 1H, 3′-OH, J ) 4.2 Hz), 4.88 (m, 2H, 5′-OH, CH-OAc), 4.29 (m, 1H, H-3′), 4.09 (m, 1H, CH2-OAc), 4.01 (m, 3H, 2 × CH2-N, 1 × CH2-OAc), 3.77 (m, 1H, H-4′), 3.46 (m, 2H, H-5′, H-5′′), 2.43 (m, 1H, H-2′), 2.20 (m, 1H, H-2′′), 2.06 (s, 3H, CH3CO), 1.94 (s, 3H, CH3CO), 1.91 (s, 3H, CH3CO), 1.82 (m, 2H, C-CH2-C). FAB-HRMS for [M + H]+: calcd 482.1887, found 482.1896. 5′-O-(4,4′-Dimethoxytrityl)-1-(3,4-diacetoxybutyl)-N2acetyl-2′-deoxyguanosine (14). Compound 13 (110 mg, 0.23 mmol) was coevaporated with pyridine (2 × 3 mL) and then redissolved in pyridine (5 mL). 4,4′-Dimethoxytrityl chloride (94 mg, 0.28 mmol) was added, and the mixture was stirred at room temperature for 5 h. The solvent was evaporated under vacuum, and the residue was purified by silica gel column chromatography (CH2Cl2:MeOH:pyridine 95:0:5 to 90:5:5) to give 14 (132 mg, 73%). 1H NMR (DMSO-d6) δ 10.56 (bs, 1H, NHCO), 8.16

610

Chem. Res. Toxicol., Vol. 15, No. 5, 2002

Nechev et al. Scheme 2

Scheme 3

Figure 1. (s, 1H, H-8), 7.30 (m, 2H, aromatic), 7.21 (m, 7H, aromatic), 6.80 (m, 4H, aromatic), 6.24 (dd, 1H, H-1′, J1 ) J2 ) 6.3 Hz), 5.36 (d, 1H, 3′-OH, J ) 4.6 Hz), 4.95 (m, 1H, CH-OAc), 4.35 (m, 1H, H-3′), 4.15 (m, 1H, 1 × CH2-OAc), 4.06 (m, 3H, 1 × CH2OAc, 2 × CH2N), 3.89 (m, 1H, H-4′), 3.70 (s, 6H, 2 × CH3O), 3.14 (m, 2H, H-5′, H-5′′), 2.68 (m, 1H, H-2′), 2.32 (m, 1H, H-2′′), 2.10 (s, 3H, CH3CO), 1.99 (s, 3H, CH3CO), 1.95 (s, 3H, CH3CO), 1.84 (m, 2H, C-CH2-C). FAB-HRMS for [M + H]+: calcd 784.3194, found 784.3211. 3′-O-[(N,N-Diisopropylamino)-2-cyanoethoxyphosphinyl]5′-O-(4,4′-dimethoxytrityl)-1-(3,4-diacetoxybutyl)-N2-acetyl2′-deoxyguanosine (15). Compound 14 (80 mg, 0.10 mmol) was dried by coevaporation with anhydrous pyridine (2 × 3 mL) and placed under vacuum overnight. The oily residue was treated with a solution of anhydrous 1H-tetrazole (8.4 mg, 0.12 mmol) in freshly distilled CH2Cl2 (5 mL) followed by addition of 2-cyanoethyl-N,N,N′,N′-tetraisopropylphosphorodiamidite (39.2 mg, 0.13 mmol). After 3 h, the solvents were removed under reduced pressure, and the crude product was purified by flash column chromatography on silica gel (EtOAc:hexanes:pyridine 90:5:5) to yield 95.3 mg (95% yield) of phosphoramidite 15. 1H NMR (DMSO-d6) δ 10.58 (bs, 1H, NHCO), 8.22, 8.20 (s, 1H, diastereomeric H-8), 7.33 (m, 2H, aromatic), 7.20 (m, 7H, aromatic), 6.82 (m, 4H, aromatic), 6.28 (dd, 1H, H-1′, J1 ) J2 ) 6.24 Hz), 4.96 (m, 1H, CH-OAc), 4.63 (m, 1H, H-3′), 4.18 (m, 1H, 1 × CH2-OAc), 4.06 (m, 4H, 1 × CH2-OAc, 2 × CH2N, H-4′), 3.72 (s, 7H, 2 × CH3O, 1 × POCH2), 3.63 (m, 1H, 1 × POCH2), 3.52 (m, 2H, isopropyl CH), 3.19 (m, 2H, H-5′, H-5′′), 2.84 (m, 1H, H-2′), 2.76 (t, 1H, 1 × CH2-CN, J ) 5.89), 2.64 (t, 1H, 1 × CH2-CN, J ) 5.89), 2.57 (m, 1H, H-2′′), 2.21 (s, 3H, CH3CO), 2.00 (s, 3H, CH3CO), 1.97 (s, 3H, CH3CO), 1.88 (m, 2H, C-CH2-C), 1.10, 1.00 (m, 12H, isopropyl CH3). 31P NMR (DMSO-d6, 121 MHz) δ 148.6, 147.8. FAB-HRMS for C50H63N7O12P [M + H]+: calcd 984.4272, found 984.4279. Preparation of an Oligonucleotide Containing 6. The 12-mer 5′-d(GCT-AGC-6-AG-TCC) was prepared on a 1-µmol

scale using tert-butylphenoxyacetyl-protected cyanoethyl phosphoramidites and adducted phosphoramidite 15. Deprotection of the oligonucleotide, including the three acetyl groups and cleavage of the oligonucleotide from the CPG beads was achieved by treatment with concentrated aqueous ammonia for 10 h at 55 °C. Purification by reversed-phase HPLC gave 25 A260 units of the 12-mer containing 6. The constitution was confirmed by MALDI-TOF MS: m/z calcd [M - H]- 3732.7, found 3732.7. Nucleolytic digestion gave 6 plus the four normal deoxynucleosides in the correct stoichiometric ratios. The identity of 6 was confirmed by HPLC comparison with an authentic sample (Figure 1A). A solution of 5′-d(GCT-AGC-6-AG-TCC) (10 A260 units) in phosphate buffer (0.80 mL, 0.05 M, pH 7.0) was treated with aqueous NaIO4 (20 mmol, 0.050 mL) for 30 min at room temperature. Purification by reversed-phase HPLC gave 8.5 A260 units (∼85%) of 5′-d(GCT-AGC-4-AG-TCC). The oligonucleotide was characterized by MALDI-TOF MS: m/z calcd [M - H]3700.7, found 3701.4. Nucleolytic digestion gave 4 plus the four normal deoxynucleosides in the correct stoichiometric ratios. The identity of 4 was confirmed by HPLC comparison with the authentic sample (Figure 1B).

Results and Discussion The goal of this project was to synthesize not just nucleoside 4 but also oligonucleotides containing it. Therefore, we focused our attention on a synthetic approach that would be suitable for both. For the synthesis of nucleoside 4, 1-bromo-3-butene reacted smoothly with deoxyguanosine under alkaline conditions

Acrolein Adduct of Deoxyguanosine

Chem. Res. Toxicol., Vol. 15, No. 5, 2002 611 Scheme 4

to give butenyl adduct 5 in 65% yield (Scheme 2). Treatment of 5 with OsO4/N-methylmorpholine-N-oxide gave 92% of diol 6. Oxidation with NaIO4 gave aldehyde 3, which cyclized spontaneously to hydroxypropano-dG (4) in 89% yield. The structures of compounds 4, 5, and 6 were established by NMR and mass spectroscopy. Nucleoside 4 exists as two diastereomers on account of C6 being a stereogenic center. The nucleoside appears as two peaks on HPLC (Figure 1B). As reported by previous workers, the two diastereomers are in a facile equilibrium (6). The diastereomers equilibrate too rapidly for it to be feasible to isolate them individually. There are two pathways by which the epimerization could be occurring: reversion to 3-oxopropyl adduct 3 or dehydration to imine 8 followed by rehydration. Neither 3 nor 8 can be detected in the NMR spectrum of 4, but it is possible that they are present at concentrations too small to be detected. As a probe of the mechanism, reductions

were carried out with NaBH4 and NaCNBH3 (Scheme 3). Reduction by NaBH4 at ambient temperature gave a 3:1 mixture of 1-(3-hydroxypropyl) adduct 75 and 1,N2propanodeoxyguanosine 9, whereas reduction with NaCNBH3 gave exclusively 9. The reduction with NaCNBH3 was too slow at ambient temperature but an acceptable reaction rate was observed at 40 °C. The two reducing agents differ in their selectivity for reduction of imines and aldehydes. These results show that both epimerization pathways are operative with 4. Chung et al. obtained exclusively 9 when they carried out the reduction of 4 using a mixture of NaBH4 and NaOH at 100 °C (6). 5 Compound 7 was used to confirm that the site of alkylation of dG in the initial step of the synthesis had, in fact, been N1. An HMBC spectrum showed three-bond connectivity between C6 of the purine (156 ppm) and the protons of the R methylene group of the side chain (3.98 ppm). The R carbon (39 ppm) showed three-bond connectivity with the protons of the γ methylene group (3.45 ppm).

612

Chem. Res. Toxicol., Vol. 15, No. 5, 2002 Scheme 5

For the extension of this strategy to the synthesis of oligonucleotides, the phosphoramidite reagent, i.e., 15, needed to have the hydroxyl groups on the butyl side chain in place but protected. For synthesis of 15, the deoxyribose hydroxyl groups of butenyl-dG (5) were protected with a bis-silyl agent, after which vicinal hydroxylation of the olefin in 10 was achieved with OsO4/ N-methylmorpholine-N-oxide giving diol 11 in 87% yield (Scheme 4). The hydroxyl groups of the diol and the exocyclic amino group were then protected by treatment with acetic anhydride to give 12 in 84% yield. The bissilyl group was removed with tetrabutylammonium fluoride (96%), followed by tritylation with dimethoxytrityl chloride (73%) and phosphitylation with 2-cyanoethylN,N,N′,N′-tetraisopropylphosphorodiamidite to give phosphoramidite 15 in 95% yield. A 12-mer oligodeoxynucleotide containing hydroxypropano-dG (4) was prepared using 15 by the standard protocol for automatic solid-phase DNA synthesis with the exception that the exocyclic amino groups of dG, dC, and dA were protected with labile PAC groups (Scheme 5). The deprotection step was carried out with hot, concentrated ammonia, which cleaved the tether to the beads and removed all protective groups including the three acetyl groups of the adducted nucleoside to give an oligonucleotide containing dihydroxybutyl-dG (6). Cleavage of the vicinal diol with NaIO4 gave the oligonucleotide containing hydroxypropano-dG (4). Homogeneity of the oligonucleotides containing 6 and 4 was demonstrated by CZE. The molecular weights of the oligonucleotides were established by MALDI-TOF MS. Enzymatic degradation of the oligonucleotides using a mixture of DNase I, snake venom phosphodiesterase I, and alkaline phosphatase gave the appropriate nucleosides in the correct molar ratios (Figure 1). Nucleosides 6 and 4 were identified by HPLC comparison with authentic samples. The structurally similar 8- and 6-hydroxypropanodeoxyguanosines 2 and 4 can be expected to have quite different effects on the structures and biological properties of DNAs that contain them. Both 2 and 4 disrupt Watson-Crick base-pairing. However, 2 is in a mobile equilibrium with the acyclic N2-oxopropyl nucleoside 1 and can thereby achieve base-pairing. The equilibrium between 1 and 2 lies so far on the side of 2 that 1 cannot be detected by NMR. However, the driving force for forming the base pair is sufficient that 1 becomes a major species in a DNA duplex (18). Nucleoside 4 is also in a

Nechev et al.

mobile equilibrium with acyclic N1-oxopropyl nucleoside and in this case the amount of 3 present is below the limits of spectroscopic detection. However, in duplexed DNA there will be no driving force to favor formation of the acyclic species since the oxopropyl substituent at the N1 position of 3 still blocks Watson-Crick base-pairing. With 4 there is close structural similarity to 1,N2propano-dG (9) and it is likely that the three-dimensional structures of oligonucleotides containing 4 will closely resemble those containing 9. The solution structure of an oligonucleotide duplex containing 9 obtained by NMR showed the presence of two conformations; in both of them 9 had rotated into a syn conformation so that the propano fragment lay in the major groove providing an opportunity for Hoogsteen base-pairing (19). It seems likely that 4 will resemble 9 rather than 8-hydroxypropano adduct 2 in its effects on replication. Whereas 2 was essentially nonmutagenic in bacteria (11, 12), 9 substantially degraded replication fidelity (20-23). These findings are consistent with the fact that 9 perturbs duplex structure far more than 2. We have recently reported that DNA duplexes containing 2 in a CpG context form imine cross-links between the strands (24). The mechanism of formation of the cross-link involves trapping of the N2-oxopropyl adduct 1 by the exocyclic amino group of the guanine in the complementary strand. The situation with respect to the 6-hydroxypropano adduct 4 is more complex because both N1-oxopropyl species 3 and imine 8 can potentially crosslink.

Acknowledgment. We thank the National Institute for Environmental Health Sciences for generous support of this project (ES00267 and ES05355) and Pamela J. Tamura and Amanda S. Wilkinson for assistance with oligonucleotide syntheses. A.K.B. acknowledges a GAANN fellowship (P200A980237-03) and an NIEHS traineeship (ES007028). Supporting Information Available: 1H NMR spectra of compounds 4-6 and 10-15; 31P NMR spectrum of 15, COSY spectra of compounds 4, 10, 11, and 14, and CZE analyses of oligonucleotides containing 6 and 4. This information is available free of charge via the Internet at http://pubs.acs.org.

References (1) Nath, R. G., Ocando, J. E., Guttenplan, J. B., and Chung, F.-L. (1998) 1,N2-propanodeoxyguanosine adducts: potential new biomarkers of smoking-induced DNA damage in human oral tissue. Cancer Res. 58, 581-584. (2) Yang, K., Fang, J.-L., Li, D., Chung, F.-L., and Hemminki, K. (1999) 32P-postlabelling with high-performance liquid chromatography for analysis of abundant DNA adducts in human tissues. IARC Sci. Publ. 150, 205-217. (3) Chung, F.-L., Nath, R. G., Nagao, M., Nishikawa, A., Zhou, G. D., and Randerath, K. (1999) Endogenous formation and significance of 1,N2-propanodeoxyguanosine adducts. Mutat. Res. 424, 71-81. (4) Pan, J., and Chung, F.-L. (2002) Formation of cyclic deoxyguanosine adducts from ω-3 and ω-6 polyunsaturated fatty acids under oxidative conditions. Chem. Res. Toxicol. 15, 367-372. (5) Kawanishi, M., Matsuda, T., Nakayama, A., Takebe, H., Matsui, S., and Yagi, T. (1998) Molecular analysis of mutations induced by acrolein in human fibroblast cells using supF shuttle vector plasmids. Mutat. Res. 417, 65-73. (6) Chung, F.-L., Young, R., and Hecht, S. S. (1984) Formation of cyclic 1,N2-propanodeoxyguanosine adducts in DNA upon reaction with acrolein or crotonaldehyde. Cancer Res. 44, 990-995.

Acrolein Adduct of Deoxyguanosine (7) Chung, F.-L., Roy, K., and Hecht, S. S. (1988) A study of reactions of R,β-unsaturated carbonyl compounds with deoxyguanosine. J. Org. Chem. 53, 14-17. (8) Huang, Y., and Johnson, F. (2002) Regioisomeric synthesis and characteristics of the R-hydroxy-1,N2-propanodeoxyguanosine. Chem. Res. Toxicol. 15, 236-239. (9) Khullar, S., Varaprasad, C. V., and Johnson F. (1999) Postsynthetic generation of a major acrolein adduct of 2′-deoxyguanosine in oligomeric DNA. J. Med. Chem. 42, 947-950. (10) Nechev, L. V., Harris, C. M., and Harris, T. M. (2000) Synthesis of nucleosides and oligonucleotides containing adducts of acrolein and vinyl chloride. Chem. Res. Toxicol. 13, 421-429. (11) VanderVeen, A. A., Hashim, M. F., Nechev, L. V., Harris, T. M., Harris, C. M., and Marnett, L. J. (2001) Evaluation of the mutagenic potential of the principal DNA adduct of acrolein. J. Biol. Chem. 276, 9066-9070. (12) Yang, I.-Y., Hossian, M., Miller, H., Khullar, S., Johnson, F., Grollman, A., and Moriya, M. (2001) Responses to the major acrolein-derived deoxyguanosine adduct in Escherichia coli. J. Biol. Chem. 276, 9071-9076. (13) Yang, I.-Y., Johnson, F., Grollman, A. P., and Moriya, M. (2002) Genotoxic mechanism for the major acrolein-derived deoxyguanosine adduct in human cells. Chem. Res. Toxicol. 15, 160-164. (14) Kanuri, M., Minko, I. G., Nechev, L. V., Harris, T. M., Harris, C. M., and Lloyd, R. S. (2002) Error prone translesion synthesis past γ-hydroxypropano deoxyguanosine, the primary acrolein-derived adduct in mammalian cells. J. Biol. Chem. (in press). (15) Smith, R. A., Sysel, I. A., Tibbles, T. S., and Cohen, S. M. (1988) Implications for the formation of abasic sites following modification of polydeoxycytidylic acid by acrolein in vitro. Cancer Lett. 40, 103-109. (16) Moon, K.-Y., and Moschel, R. C. (1998) Effect of ionic state of 2′deoxyguanosine and solvent on its aralkylation by benzyl bromide. Chem. Res. Toxicol. 11, 696-702.

Chem. Res. Toxicol., Vol. 15, No. 5, 2002 613 (17) Kowalczyk, A., Harris, C. M., and Harris, T. M. (2001) Synthesis and characterization of oligodeoxynucleotides containing an N1 β-hydroxyalkyl adduct of 2′-deoxyinosine. Chem. Res. Toxicol. 14, 746-753. (18) de los Santos, C., Zaliznyak, T., and Johnson, F. (2001) NMR characterization of a DNA duplex containing the major acroleinderived deoxyguanosine adduct γ-hydroxy-1,N2-propano-2′-deoxyguanosine. J. Biol. Chem. 276, 9077-9082. (19) Singh, U. S., Moe, J. G., Reddy, G. R., Weisenseel, J. P., Marnett, L. J., and Stone, M. P. (1993) 1H NMR of an oligodeoxynucleotide containing a propanodeoxyguanosine adduct positioned in a (CG)3 frameshift hotspot of Salmonella typhimurium hisD3052: Hoogsteen base-pairing at pH 5.8. Chem. Res. Toxicol. 6, 825-836. (20) Hashim, M. F., Schnetz-Boutaud, N., and Marnett, L. J. (1997) Replication of template-primers containing propanodeoxyguanosine by DNA polymerase β. Induction of base pair substitution and frameshift mutations by template slippage and deoxynucleoside triphosphate stabilization. J. Biol. Chem. 272, 20205-20212. (21) Burcham, P. C., and Marnett, L. J. (1994) Site-specific mutagenesis by a propanodeoxyguanosine adduct carried on an M13 genome. J. Biol. Chem. 269, 28844-28850. (22) Shibutani, S., and Grollman, A. P. (1993) On the mechanism of frameshift (deletion) mutagenesis in vitro. J. Biol. Chem. 268, 11703-11710. (23) Moriya, M., Zhang, W., Johnson, F., and Grollman, A. P. (1994) Mutagenic potency of exocyclic DNA adducts: marked differences between Escherichia coli and simian kidney cells. Proc. Natl. Acad. Sci. U.S.A. 91, 11899-11903. (24) Kozekov, I., Nechev, L. V., Sanchez, A., Harris, C. M., Lloyd, R. S., and Harris, T. M. (2001) Interchain cross-linking of DNA mediated by the principal adduct of acrolein. Chem. Res. Toxicol. 14, 1482-1485.

TX010181Y