Characterization of 2 '-Deoxyadenosine Adducts Derived from 4-Oxo-2

Jun 9, 2000 - Seon Hwa Lee,† Diane Rindgen,†,‡ Roy H. Bible, Jr.,§ Elisabeth Hajdu,§ and. Ian A. Blair*,†. Center for Cancer Pharmacology, U...
0 downloads 0 Views 250KB Size
Download PDF
Chem. Res. Toxicol. 2000, 13, 565-574

565

Characterization of 2′-Deoxyadenosine Adducts Derived from 4-Oxo-2-nonenal, a Novel Product of Lipid Peroxidation Seon Hwa Lee,† Diane Rindgen,†,‡ Roy H. Bible, Jr.,§ Elisabeth Hajdu,§ and Ian A. Blair*,† Center for Cancer Pharmacology, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19104-6160, and Physical Methodology Department, Searle, 4901 Searle Parkway, Skokie, Illinois 60077 Received March 9, 2000

Analysis of the reaction between 2′-deoxyadenosine and 4-oxo-2-nonenal by liquid chromatography/mass spectrometry revealed the presence of three major products (adducts A1, A2, and B). Adducts A1 and A2 were isomeric; they interconverted at room temperature, and they each readily dehydrated to form adduct B. The mass spectral characteristics of adduct B obtained by collision-induced dissociation coupled with multiple tandem mass spectrometry were consistent with those expected for a substituted etheno adduct. The structure of adduct B was shown by NMR spectroscopy to be consistent with the substituted etheno-2′deoxyadenosine adduct 1′′-[3-(2′-deoxy-β-D-erythropentafuranosyl)-3H-imidazo[2,1-i]purin-7yl]heptane-2′′-one. Unequivocal proof of structure came from the reaction of adducts A1 and A2 (precursors of adduct B) with sodium borohydride. Adducts A1 and A2 each formed the same reduction product, which contained eight additional hydrogen atoms. The mass spectral characteristics of this reduction product established that the exocyclic amino group (N6) of 2′-deoxyadenosine was attached to C-1 of the 4-oxo-2-nonenal. The reaction of 4-oxo-2-nonenal with calf thymus DNA was also shown to result in the formation of substituted ethano adducts A1 and A2 and substituted etheno adduct B. Adduct B was formed in amounts almost 2 orders of magnitude greater than those of adducts A1 and A2. This was in keeping with the observed stability of the adducts. The study presented here has provided additional evidence which shows that 4-oxo-2-nonenal reacts efficiently with DNA to form substituted etheno adducts.

Introduction Promutagenic exocyclic DNA adducts have been detected in both human and animal tissues that have not been treated with carcinogens (1). There is compelling evidence that these lesions arise from the reaction of DNA bases with bifunctional electrophiles generated during lipid peroxidation (2). Malondialdehyde (MDA, β-hydroxyacrolein) is a well-characterized bifunctional electrophile, which arises from both free radical (3-5) and enzymatic (6, 7) peroxidation of polyunsaturated fatty acids. MDA modifies purine bases to generate a tricyclic adduct with guanine (8) and an acyclic adduct with adenine (9, 10). Lipid hydroperoxides are also thought to be the precursors in the generation of 4-hydroxy-2-nonenal, another well-studied bifunctional electrophile (11, 12). It reacts directly with 2′-deoxyguanosine (dGuo) to produce a tricyclic substituted propano adduct (13). The reaction of 4-hydroxy-2-nonenal with 2′-deoxyadenosine (dAdo) has not been well characterized. However, 4-hydroxy-2-nonenal is readily oxidized to the * To whom correspondence should be addressed: Center for Cancer Pharmacology, University of Pennsylvania School of Medicine, 1254 BRB II/III, 421 Curie Blvd., Philadelphia, PA 19104-6160. Fax: (215) 573-9889. E-mail: [email protected]. † University of Pennsylvania School of Medicine. ‡ Present address: Drug Metabolism and Pharmacokinetics, Schering-Plough Research Institute, K-15-3700, 2015 Galloping Hill Rd., Kenilworth, NJ 07033-0539. § Searle.

epoxide derivative, which can covalently modify both dGuo and dAdo to form etheno adducts (14, 15). Other bifunctional electrophiles derived from lipid peroxides that may have relevance to the formation of DNA adducts include acrolein (2), crotonaldehyde (2), trans-2-hexenal (16), trans,trans-2,4-decadienal (17), and 4,5-epoxy-2decenal (18). Previous studies with lipid hydroperoxides have identified a series of novel fluorescent adducts of adenine and adenosine (19, 20). These adducts were only tentatively identified but did not appear to correspond to any of the etheno or propano adducts derived from known bifunctional electrophiles. Therefore, there may be unidentified covalent modifications to dAdo that arise as a consequence of lipid peroxidation. We have recently demonstrated that 4-oxo-2-nonenal is a novel product of lipid peroxides and that it covalently modifies dGuo to form a substituted etheno adduct (21). This raised the possibility that 4-oxo-2-nonenal could also covalently modify dAdo to produce a similar novel structural modification. We report the characterization of the stable etheno adduct that arises in the interaction between 4-oxo-2-nonenal and dAdo. We also show that 4-oxo-2-nonenal can modify calf thymus DNA to produce identical covalent modifications that are observed with the free 2′-deoxynucleoside.

10.1021/tx000057z CCC: $19.00 © 2000 American Chemical Society Published on Web 06/09/2000

566

Chem. Res. Toxicol., Vol. 13, No. 7, 2000

Materials and Methods Materials. Nuclease P1, dAdo, 3-(N-morpholino)propanesulfonic acid (MOPS),1 alkaline phosphatase, trifluoroacetic acid, sodium borohydride, etheno-2′-deoxyguanosine, ethenoguanine, and calf thymus DNA were purchased from Sigma Chemical Co. (St. Louis, MO). Ammonium acetate was obtained from J. T. Baker (Phillipsburg, NJ), and DNAse I was from Calbiochem (La Jolla, CA). [15N5]-2′-Deoxyadenosine was obtained from Cambridge Isotope Laboratories (Andover, MA). HPLC grade water was obtained from Fisher Scientific Co. (Fair Lawn, NJ). HPLC grade methanol and acetonitrile were purchased from Burdick and Jackson (Muskegon, MI). Gases were supplied by BOC Gases (Lebanon, NJ). NMR. The NMR spectra were determined at 25 °C using a 600 MHz Bruker DRX-600 instrument equipped with a 2.5 mm indirect detection triple-axis gradient probe. The sample of adduct B was dissolved in 180 µL of DMSO-d6. The proton spectrum was determined using a 30° flip angle with a 1 s equilibrium delay. The unsymmetrized COSY spectrum shown here was determined using a 100 ms delay before the second 90° pulse. Both the 1H-13C two-dimensional heteronuclear multiple-quantum correlation (HMQC) and 1H-13C two-dimensional heteronuclear multiple-bond correlation (HMBC) spectra were determined using gradient pulses for coherence selection. The HMQC spectrum was determined with decoupling during acquisition. The HMBC spectrum, which was optimized for 5 Hz 1H-13C couplings, was determined with no decoupling during acquisition. Transverse rotating-frame nuclear Overhauser effect (ROESY) experiments were carried out with both 500 ms and 1 s mixing times, and the correlations were made by taking slices at the chemical shifts of the selected protons. Mass Spectrometry. The data were acquired on a Finnigan LCQ ion trap mass spectrometer (ThermoQuest, San Jose, CA) equipped with a Finnigan electrospray source. The mass spectrometer was operated in the positive ion mode with a potential of 4.25 kV applied to the electrospray needle. Nitrogen was used as the sheath (60 units) and auxiliary (5 units) gas to assist with nebulization. The capillary temperature was held at 200 °C. Full scanning analyses were performed in the range of m/z 100-800. Collision-induced dissociation (CID) experiments coupled with multiple tandem mass spectrometry (MSn) employed helium as the collision gas. The relative collision energy was set at 20% of the maximum (1 V). Liquid Chromatography. Chromatography was performed using a Waters Alliance 2690 HPLC system (Waters Corp., Milford, MA). Gradient elutions were all performed in the linear mode. Gradient systems 1-3 employed a YMC C18 ODS-AQ column (250 mm × 4.6 mm i.d., 5 µm; YMC, Inc., Wilmington, NC) at a flow rate of 1 mL/min. The LC/UV/MS experiments were performed using a Hitachi L-4200 UV detector at 231 nm. A postcolumn split was employed such that a flow of 950 µL/ min was delivered to the UV detector and 50 µL/min to the mass spectrometer. For preparative HPLC, all of the mobile phase was diverted through the UV detector. For systems 1 and 2, solvent A was 5 mM ammonium acetate in water containing 0.01% trifluoroacetic acid, and solvent B was 5 mM ammonium acetate in methanol containing 0.01% trifluoroacetic acid. For system 1, the gradient conditions were as follows: 30% B at 0 min, 30% B at 5 min, 100% B at 16 min, 100% B at 24 min, and 1 Abbreviations: bs, broad singlet; CID, collision-induced dissociation; CNL, constant neutral loss; COSY, 1H-1H two-dimensional correlation spectroscopy; dd, doublet of doublets; ddd, doublet of doublets of doublets; ESI, electrospray ionization; HMBC, 1H-13C twodimensional heteronuclear multiple-bond correlation; HMQC, 1H-13C two-dimensional heteronuclear multiple-quantum correlation; 13HPODE, 13-hydroperoxy-[S-(Z,E)]-9,11-octadecadienoic acid; HRMS, high-resolution mass spectrometry; LC/MS, liquid chromatography/ mass spectrometry; LOX, lipoxygenase; MOPS, 3-(N-morpholino)propanesulfonic acid; MSn, multiple tandem mass spectrometry; NOE, nuclear Overhauser effect; ROESY, rotating-frame nuclear Overhauser effect; SPE, solid-phase extraction; SRM, selected reaction monitoring; TIC, total ion current.

Lee et al. 30% B at 26 min, followed by a 4 min equilibration period. For system 2, the gradient conditions were as follows: 30% B at 0 min, 30% B at 3 min, 33% B at 5 min, 33% B at 20 min, 100% B at 23 min, 100% B at 33 min, 30% B at 35 min, and 30% B at 40 min, followed by a 4 min equilibration period. For system 3, solvent A was 5 mM ammonium acetate in water and solvent B was 5 mM ammonium acetate in acetonitrile. The gradient for system 3 was as follows: 6% B at 0 min, 6% B at 3 min, 12% B at 9 min, 12% B at 10 min, 70% B at 13 min, 70% B at 18 min, and 6% B at 20 min, followed by a 4 min equilibration period. Synthesis of 4-Oxo-2-nonenal. 4-Hydroxy-2-nonenal diethyl acetal was oxidized with activated MnO2 as described by Esterbauer and Weger (22). The resulting 4-oxo-2-nonenal diethyl acetal was then hydrolyzed by citric acid/HCl as described previously (21). Reaction of 4-Oxo-2-nonenal with dAdo. A solution of 4-oxo-2-nonenal (1300 µg, 8.4 µmol) in ethanol (15 µL) was added to dAdo (1200 µg, 4.8 µmol) in water (250 µL). The reaction mixture was sonicated for 15 min at room temperature and incubated at 37 or 60 °C for 24 h, after which it was placed in ice. An aliquot of the sample (20 µL) was diluted to 100 µL with a water/methanol mixture (7:3 v/v) and filtered through a 0.2 µm CoStar cartridge prior to analysis of a portion of the sample (10 µL) by LC/MS using gradient system 1. Preparation of Adducts A1 and A2 for UV and LC/MS Analysis. A solution of 4-oxo-2-nonenal (7.4 mg, 0.05 mmol) in 15 µL of ethanol was added to dAdo (12.5 mg, 0.05 mmol) in 250 µL of water. The reaction mixture was incubated at 37 °C for 16 h after sonication for 20 min. At the end of the incubation, the reaction mixture was placed in ice for 30 min and the reaction product was isolated by preparative HPLC. Chromatography was conducted using gradient system 2. Reaction products were fraction-collected between 19.0 and 24.0 min every 10 s by monitoring the UV response at 263 nm. Fractions containing primarily A1 or A2 were neutralized with aqueous 1 M ammonium hydroxide and concentrated under nitrogen at room temperature. The individual diastereomers were then further purified using gradient system 2. Adduct A1 had a retention time of 20.9 min under these conditions, and adduct A2 had a retention time of 23.3 min. The purified samples were extremely unstable, and so the amounts were determined by UV spectroscopy using the molecular extinction coefficient for dAdo (λmax ) 260.5 nm,  ) 14 347 M-1 cm-1). The adducts were stored at - 80 °C. Preparation of [15N5]A1 and [15N5]A2. A solution of 4-oxo2-nonenal (7.4 mg, 0.05 mmol) in 15 µL of ethanol was added to [15N5]dAdo (12.5 mg, 0.05 mmol) in 250 µL of water. [15N5]A1 and [15N5]A2 were purified as described above for the unlabeled adducts and stored at -80 °C. Decomposition of Adducts A1 and A2. Adduct A1 or A2 (0.10 mg, 0.24 µmol) was dissolved in water (pH 7.0, 250 µL) and the mixture allowed to stand at room temperature for 72 h after vortex mixing for 2 min. Samples (10 µL) from the aqueous solution were diluted to 100 µL with a water/methanol mixture (7:3 v/v) and filtered through a 0.2 µm CoStar cartridge prior to analysis of a portion of the sample (10 µL) by LC/MS using gradient system 1. Reconstructed total ion current (TIC) chromatograms were obtained after 0, 24, and 72 h. Reduction of Adducts A1 and A2. Adduct A1, adduct A2, [15N5]A1, or [15N5]A2 (0.48 mg, 1.2 µmol of each) was dissolved in water (pH 7.0, 250 µL) and the mixture cooled to 4 °C in an ice bath. Sodium borohydride (10 mg, 270 µmol) was added, and the sample was vortex-mixed and kept at 4 °C for 1 h in an ice bath. It was then warmed to room temperature and allowed to stand for a further 23 h. The reaction mixture was analyzed directly by LC/MSn. Reduction of Ethenoadenine and Etheno-2′-deoxyadenosine. Ethenoadenine (0.25 mg, 1.6 µmol) or etheno-2′deoxyadenosine (0.50 mg, 1.8 µmol) was dissolved in water (pH 7.0, 250 µL) and cooled to 4 °C in an ice bath. Sodium borohydride (10 mg, 270 µmol) was added, and the sample was

dAdo Adducts of 4-Oxo-2-nonenal vortex-mixed and kept at 4 °C for 1 h in an ice bath. It was then warmed to room temperature and allowed to stand for a further 23 h. The reaction mixture was analyzed directly by LC/ MS. Preparation of Adduct B for UV and NMR Analysis. A solution of 4-oxo-2-nonenal (64.8 mg, 0.4 mmol) in ethanol (15 µL) was added to dAdo (10.5 mg, 0.04 mmol) in water (250 µL). After sonication for 20 min at room temperature, the mixture was incubated at 60 °C for 24 h. It was placed on ice and adduct B isolated using gradient system 3, by monitoring the UV absorbance at 231 nm. The fraction, which eluted between 15.5 and 15.8 min, was concentrated under nitrogen at room temperature. The residue was dissolved in a methanol/water mixture (1 mL, 3:7 v/v) and rechromatographed on the same system. Fractions were collected between 14.0 and 17.0 min. The fractions containing pure adduct B (as determined by LC/ MS) were combined, concentrated under reduced pressure, and dried under vacuum. The retention time of pure adduct B on system 3 was 16.5 min. Hydrolysis of Calf Thymus DNA. Calf thymus DNA (200 µg, 0.67 µmol) was dissolved in 1 mL of 10 mM MOPS containing 100 mM NaCl (pH 7.0) by sonication for 5 min. DNAse I (556 units) dissolved in 200 µL of 10 mM MOPS containing 120 mM MgCl2 (pH 7.0) was added and the sample incubated for 90 min at 37 °C. At the end of this incubation, 50 µL of 10 mM MOPS containing 100 mM NaCl (pH 7.0) and 15.5 units of nuclease P1 was added followed by 25 mM ZnCl2 (50 µL). The incubation was then continued for 2 h at 37 °C. Finally, 30 units of alkaline phosphatase in 0.5 mL of 0.4 M MOPS (pH 7.8) was added, followed by incubation for an additional 1 h. After the mixture had been cooled to room temperature, a portion of the hydrolysate (10 µL) was subjected to LC/selected reaction monitoring (SRM)/MS analysis using gradient system 3. The elution times for the 2′-deoxyribonucleosides were 4.1 (dCyd), 6.7 (dGuo), 8.8 (dThd), and 10.81 min (dAdo). SRM analyses were performed on [MH]+ f [BH2+] of the individual base: m/z 228 f 112 (dCyd), m/z 268 f 152 (dGuo), m/z 243 f 127 (dThd), and m/z 252 f 136 (dAdo). Quantitation of normal DNA bases was carried out from standard curves constructed by injection of solutions (10 µL) containing known amounts (0.25, 0.5, 1, 2, and 4 µg) of each 2′-deoxyribonucleoside. Typical regression lines were as follows: y ) 1.215x - 0.114 and r2 ) 0.997 (dCyd), y ) 14.609x + 1.392 and r2 ) 0.994 (dGuo), y ) 0.341x - 0.019 and r2 ) 0.977 (dThd), and y ) 63.90x + 7.896 and r2 ) 0.999 (dAdo). Recovery of individual normal DNA bases was essentially quantitative. Preparation of 4-Oxo-2-nonenal-Modified Calf Thymus DNA. Calf thymus DNA (200 µg, 0.67 µmol) was dissolved in water (250 µL) and treated with 4-oxo-2-nonenal (1.0 mg, 6.5 µmol) in ethanol (15 µL). After sonication for 15 min, the reaction mixture and appropriate control reaction mixtures were incubated at 37 °C for 8 h. Samples were placed in ice for 30 min, and the DNA was precipitated by adding ice-cold ethanol (1050 µL) followed by ice-cold 2.9 M sodium acetate (35 µL). The samples were centrifuged at 14000g for 15 min in an Eppendorf centrifuge, and the supernatant was removed. The DNA pellet was washed with an ethanol/water mixture (1 mL, 7:3 v/v) and residual solvent removed by evaporation under nitrogen. The final pellet was dissolved in 1 mL of 10 mM MOPS containing 100 mM NaCl (pH 7.0) by sonication for 5 min. Enzymatic hydrolysis was conducted as described above for normal DNA. Isolation of DNA Adducts from Modified Calf Thymus DNA. The hydrolysate was applied directly to a solid-phase extraction (SPE) cartridge (500 mg, 3 mL, Supelclean LC-18, Supelco, Bellefonte, PA) that had been prewashed with methanol (10 mL) and water (10 mL). The cartridge was washed with water (5 mL) and a methanol/water mixture (5 mL, 1:4 v/v). Adducts A1, A2, and B were eluted in methanol (4 mL). The eluates was evaporated to dryness under nitrogen and dissolved in methanol (100 µL). LC/MS analysis was conducted on a 10 µL aliquot of this solution using gradient system 1 at a flow rate of 1.0 mL/min.

Chem. Res. Toxicol., Vol. 13, No. 7, 2000 567

Figure 1. Analysis of the reaction between 4-oxo-2-nonenal and dAdo for 24 h at 37 °C by concurrent LC/MS and UV detection using gradient system 1. The three upper panels are the reconstructed selected ion chromatograms for the MH+ of dAdo (m/z 252), adducts A1 and A2 (m/z 406), and adduct B (m/z 388). The lower panel is the UV absorbance at 231 nm. Recovery of Adduct B through the Hydrolysis Procedure. Triplicate samples of adduct B (5 µg) were separately added to calf thymus DNA (200 µg, 0.67 µmol) in water (250 µL). The DNA was hydrolyzed and adduct B extracted using the SPE procedure described above. The LC/MS response from an injection of 10 µL of hydrolysate was compared with that from a 10 µL injection of a blank DNA hydrolysate (100 µL) containing adduct B (5 µg). This took care of any suppression of the electrospray signal by residual constituents from the biological matrix.

Results Reaction of 4-Oxo-2-nonenal with dAdo. LC/MS analysis of the products from the reaction between dAdo and 4-oxo-2-nonenal at 37 °C for 24 h revealed the presence of three major compounds with MH+ ions at m/z 406 (adducts A1 and A2) and m/z 388 (adduct B) (Figure 1). Residual dAdo (MH+ m/z ) 252) was also observed. The LC effluent was also allowed to pass concurrently through a UV spectrometer (231 nm). The resulting chromatogram confirmed the presence of three major products (adducts A1, A2, and B), together with residual dAdo. When the reaction mixture was heated at 60 °C for 24 h, the magnitudes of the signals for adducts A1 and A2 were reduced considerably with a concomitant increase in the magnitudes of the MS and UV signals for adduct B (Figure 2). NMR Analysis of Adduct B. The assignments of the proton signals were made on the basis of the proton chemical shifts and proton-proton couplings (Table 1) and COSY, HMQC, and HMBC (Table 2) correlations. All assignments refer to structure I in Scheme 1. No signal for H-6 was detected in the 1H NMR spectrum of adduct B (Table 1). However, a signal for H-8 appeared at 7.37 ppm, which had a COSY correlation to H-1′′. These data were consistent with the presence of an

568

Chem. Res. Toxicol., Vol. 13, No. 7, 2000

Figure 2. Analysis of the reaction between 4-oxo-2-nonenal and dAdo for 24 h at 60 °C by concurrent LC/MS and UV detection using gradient system 1. The three upper panels are the reconstructed selected ion chromatograms for the MH+ of dAdo (m/z 252), adducts A1 and A2 (m/z 406), and adduct B (m/z 388). The lower panel is the UV absorbance at 231 nm.

olefinic bond between C-7 and C-8 (Scheme 1). The triplet at 2.64 ppm was assigned to the H-3′′ protons because of their chemical shift and their triplet character (Figure 3). The COSY spectrum showed that the H-3′′ protons at 2.64 ppm were coupled mostly to the protons with the 1.53 ppm shift, which were therefore assigned to the H-4′′ protons (Figure 4). This spectrum also showed that the C-4′′ protons were coupled to the high-field side of the four-proton signal at 1.29 ppm and that the H-7′′ protons were coupled to the low-field side of this same band. This then meant that the signals from the H-6′′ protons were at a slightly lower field than the H-5′′ protons. Examination of the HMQC spectrum placed the 13C shift for C-5′′ (31.6 ppm) downfield of C-4′′ (23.6 ppm). The HMQC spectrum also showed that H-5 at 8.98 ppm and H-8 at 7.37 ppm both correlated with C-9a at 141.6 ppm (Table 2). Although the transverse ROESY spectra were very noisy and contained artifacts, there were clear nuclear Overhauser effect (NOE) cross-peaks between the H-1′′ protons and both H-5 and H-8. While the NOE data were not decisive, additional HMQC and HMBC data were consistent with the proposed structure of 1′′-[3-(2′-deoxyβ-D-erythropentafuranosyl)-3H-imidazo[2,1-i]purin-7-yl]heptane-2′′-one (I, Scheme 1). However, on the basis of NMR data alone, it was not possible to completely rule out the alternative structure 1′′-[3-(2′-deoxy-β-D-erythropentafuranosyl)-3H-imidazo[2,1-i]purin-8-yl]heptane2′′-one (II, Scheme 1). LC/MSn Analysis of Adducts A1 and A2. The mass spectra for adducts A1 and A2 were identical (Figure 5). Each spectrum exhibited an intense MH+ ion at m/z 406 together with a fragment ion at m/z 290 corresponding to the protonated modified base (B) to which an additional proton had been transferred (BH2+) (Figure 5A). MS2 analysis of m/z 406 resulted in the almost exclusive formation of the BH2+ product ion at m/z 290 (Figure 5B).

Lee et al.

CID of m/z 290 (MS3) gave rise to product ions at m/z 272 (-H2O) and m/z 136 (Ade + H) (Figure 5C). Finally, CID of the ion at m/z 272 (MS4) gave rise to an ion at m/z 174 (Figure 5D). These mass spectral characteristics were consistent with diastereomeric substituted ethano dAdo adduct structures. In particular, the MS3 analysis revealed the presence of an adenine moiety (Scheme 2), and MS4 analysis was consistent with the presence of a heptanone moiety (Scheme 2). Furthermore, a product ion corresponding to protonated Ade was not observed in the MS/MS analysis of the product ion at m/z 272 (dehydrated BH2+), suggesting that this product ion had an etheno dAdo structure (Figure 5D). LC/MSn Analysis of Adduct B. The mass spectrum of adduct B exhibited an intense MH+ ion at m/z 388 together with a BH2+ ion at m/z 272 (Figure 6A). MS2 analysis of MH+ again gave rise to an intense BH2+ ion 18 Da below that for adducts A1 and A2 (Figure 6B). Further CID of the BH2+ ion at m/z 272 (MS3) gave rise to an ion at m/z 174 identical with that observed for adducts A1 and A2 (Figure 6C and Scheme 2). These data were consistent with the proposal that adduct B was a dehydration product of adducts A1 and A2. Decomposition of Adducts A1 and A2. Adducts A1 and A2 were shown to be >95% pure by LC/MS analysis (Figures 7A and 8A). Incubation of adduct A1 at room temperature for 24 h resulted in the formation of adduct A2 and adduct B (Figure 7B). After 72 h, almost all of adduct A1 and adduct A2 had been converted to adduct B (Figure 7C). At this time point, residual adducts A1 and A2 were present in almost equimolar amounts. The decomposition of adduct A2 followed a time course almost identical to that of adduct A1 (panels B and C of Figure 8). These data showed that adducts A1 and A2 were able to interconvert with each other and confirmed that they both could be dehydrated to give adduct B. UV Analysis of Adducts A1, A2, and B. Both adduct A1 and A2 exhibited a λmax of 263 nm at pH 7.0 (Figure 9A). Adduct B exhibited a different UV spectrum with a λmax at 231 nm ( ) 15 800 M-1 cm-1) at pH 7.0 (Figure 9B). The λmax was shifted to 237 nm ( ) 15 232 M-1 cm-1) at pH 13 and to 224 nm ( ) 15 960 M-1 cm-1) at pH 1. The effect of pH on the UV spectra of adduct B was consistent with that reported previously for substituted etheno dAdo adducts (14). LC/MS2 Analysis of Adducts A1 and A2 after Borohydride Reduction. LC/MS analysis of adduct A1 and adduct A2 after reaction with sodium borohydride for 5 min at 4 °C revealed that no starting material was left. Two major products were formed, each with intense MH+ ions at m/z 408. MS4 analysis revealed the presence of two hydroxyl groups in each product [m/z 408 (MH+) f m/z 292 (BH2+) f m/z 274 (BH2+ - H2O) f m/z 256 (BH2+ - 2H2O)]. When the reaction with NaBH4 was allowed to proceed at 4 °C for 2 h, products were observed that corresponded to the addition of 4 Hs, 6 Hs, and 8 Hs. The reaction mixture was then allowed to warm to room temperature and to stand for a further 22 h. One major peak appeared upon reversed-phase LC/MS analysis of the reduction products (Figure 10A). Although this peak seemed to be homogeneous, it could have contained a mixture of diastereomers. An intense MH+ ion was observed at m/z 414 (Figure 10B). LC/MS2 analysis of m/z 414 (Figure 10C) revealed major product ions at m/z 397 corresponding to MH+ - NH3 and m/z 298 corresponding to BH2+ (Figure 10D). LC/MS3 analysis of m/z 397

dAdo Adducts of 4-Oxo-2-nonenal

Chem. Res. Toxicol., Vol. 13, No. 7, 2000 569 Table 1. 1H NMR Assignments for Adduct Ba

assigned H H-5 H-2 H-8 H-1′ 3′-OH 5′-OH H-3′ H-1′′a,b H-4′ H-5′b H-5′a H-2′b H-3′′a,b H-2′a H-4′′a,b H-5′′a,b H-6′′a,b H-7′′a,b,c a

δ (ppm)

multiplet

8.98 8.53 7.37 6.49 5.39 5.00 4.45 4.35 3.91 3.64 3.55 2.76 2.64 2.38 1.53 1.24 1.29 0.87

singlet singlet slightly bs triplet bs bs ddd singlet triplet dd dd ddd triplet ddd pentet multiplet multiplet triplet

H-coupled J (Hz)

small coupling to H-1′′ H-2′a (6.1), H-2′b (6.0) (exchange-broadened) (exchange-broadened) H-2′a (3.3), H-2′b (7.3), H-4′ (4.1) H-3′ (4.1), H-5′a (4.6), H-5′b (4.5) H-4′ (4.6), H-5′a (11.5) H-4′ (4.5), H-5′b (11.5) H-1′ (6.0), H-2a′ (13.4), H-3′ (7.3) H-4′′a,b (2 × 7.4) H-1′ (6.1), H-2′b (13.4), H-3′ (3.3) H-3′′a,b (2 × 7.4), H-5′′a,b (2 × 7.4) H-4′′a,b (2 × 7.4), H-6′′a,b (2 × 7.4) H-5′′a,b (2 × 7.4), H-7′′a,b,c (3 × 7 0.1) H-6′′a,b (2 × 7.1)

type NdCH NdCH CdCH NCHO CHOH CH2OH OCH CH2C OCH CH2OH CH2OH CH2C CH2C CH2C CH2C CH2C CH2C CH3

Spectra were obtained in DMSO-d6. Table 2. 13C NMR Assignments for Adduct Ba

assigned carbon δ (ppm) C-7′′ C-6′′ C-4′′ C-5′′ C-1′′ C-2′ C-3′′ C-5′ C-3′ C-1′ C-4′ C-7 C-1a C-8 C-5 C-3a C-2 C-9a C-2′′ a

14.7 22.8 23.6 31.6 37.9 40.6 42.1 62.6 71.6 84.9 88.9 119.4 123.9 132.8 136.6 138.7 140.7 141.6 207.3

coupling

type

CH3 H-7′′ CH2 H-3′′ CH2 H-3′′, H-4′′, H-6′′, H-7′′ CH2 CH2 C3′-OH CH2 H-4′′ CH2 C5′-OH CH2OH H-2′a, 3′-OH, H-4′, H-5′a, H-5′b, 5′-OH CHOH H-2′b NCHO C3′-OH, H-5′a, H-5′b, C5′-OH CHO H-1′′, H-8 CdCN H-2, H-5 CdCN H-1′′ CdCN NdCN H-2, H-5, H-1′ NCdC H-1′ NCHdN H-5, H-8 NCdN H-1′′, H-3′′ Cd

Figure 3. 1H NMR spectrum of adduct B: (A) δ 0.5-9.5 ppm, (B) expanded range of δ 3.4-6.7 ppm, and (C) expanded range of δ 0.8-3.0 ppm.

Spectra were obtained in DMSO-d6.

Scheme 1. Possible Structures for Adduct B Based on NMR Data

Figure 4. Two-dimensional COSY analysis of adduct B.

resulted in formation of the expected product ion at m/z 281 (loss of 2′-deoxyribose + H). LC/MS4 analysis revealed a diagnostic product ion at m/z 183 (loss of C7H14) that was derived from cleavage of the hemiacetal moiety (Scheme 3). LC/MS5 analysis resulted in the formation of a product ion at m/z 156 (loss of CHN, Scheme 3). Product ions derived from [15N5]A1 and [15N5]A2 were in accord with these assignments. Thus, ions were observed at m/z 401 (MH+ - 15NH3) and m/z 303 (BH2+) in the

MS2 analysis. The MSn analyses were also in accord with the assignments listed above. LC/MS Analysis of Ethenoadenine and Etheno2′-deoxyadenosine after Borohydride Reduction. Ethenoadenine was recovered intact after reaction with sodium borohydride. However, etheno-2′-deoxyadenosine was converted to a more polar product that eluted as a single peak on LC/MS analysis. An intense MH+ ion was observed at m/z 280. MS2 analysis revealed major ions

570

Chem. Res. Toxicol., Vol. 13, No. 7, 2000

Lee et al.

Figure 6. LC/MSn analysis of adduct B: (A) full-scan mass spectrum, (B) MS2 spectrum, and (C) MS3 spectrum. Figure 5. LC/MSn analysis of adduct A1. The mass spectra of adduct A2 were identical: (A) full-scan mass spectrum, (B) MS2 spectrum, (C) MS3 spectrum, and (D) MS4 spectrum.

Scheme 2. LC/MSn Product Ions from Adducts A1, A2, and B

Figure 7. Analysis of the decomposition of adduct A1 for 24 h at 25 °C by LC/MS using gradient system 1. Total ion current (TIC) chromatograms at (A) time zero, (B) 24 h, and (C) 72 h.

at m/z 262 (MH+ - H2O) and m/z 164 (MH+ - 2′deoxyribose + H). These data show that the purine ring of etheno-2′-deoxyadenosine can be reduced by sodium borohydride and that the 2′-doxyribose is required for this reduction to occur. Formation of Adducts A1, A2, and B in Calf Thymus DNA. Calf thymus DNA was treated with a 10fold excess of 4-oxo-2-nonenal. The DNA was then hydrolyzed with a mixture of DNAse I, nuclease P1, and alkaline phosphatase in MOPS buffer. Previous studies have shown that aldehyde-derived DNA adducts can react with Tris-containing buffers, and so MOPS was used instead (23). These hydrolysis conditions were shown to give essentially quantitative recovery of normal DNA bases. Modified DNA bases were separated from normal bases using the SPE procedure described in Materials and Methods. Adducts A1 and A2 are unstable,

and so it was difficult to obtain a reliable estimate of their recovery through the extraction and hydrolysis procedure. However, the recovery of adduct B (5 µg) was determined to be 71 ( 3% (n ) 3). Using both LC/MS and LC/SRM/MS, it was possible to detect adducts A1 and A2 (panels A and B of Figure 11) in the DNA hydrolysate. Relatively strong signals were observed for adduct B (panels C and D of Figure 11). On the basis of a 71% recovery through the extraction and hydrolysis procedure, the signal for adduct B corresponded to 4.4 adducts/ 103 normal bases.

Discussion We have proposed recently that 4-oxo-2-nonenal is a previously unrecognized product of lipid peroxidation

dAdo Adducts of 4-Oxo-2-nonenal

Chem. Res. Toxicol., Vol. 13, No. 7, 2000 571

Figure 8. Analysis of the decomposition of adduct A2 for 24 h at 25 °C by LC/MS using gradient system 1. Reconstructed TIC chromatograms at (A) time zero, (B) 24 h, and (C) 72 h.

Figure 10. LC/MSn analysis of a mixture of adduct A1 and adduct A2 after reduction with sodium borohydride for 24 h. The chromatography was conducted using system 2: (A) reconstructed TIC chromatogram, (B) full-scan mass spectrum of the peak at 13.3 min, (C) reconstructed TIC chromatogram from LC/MS2 analysis of MH+ (m/z 414), and (D) MS2 spectrum of the peak at 13.3 min.

Figure 9. (A) UV spectra of adducts A1 and A2 at pH 7.0. (B) UV spectra of adduct B at pH 1, 7, and 13.

(21). In our previous study, we characterized a substituted etheno dGuo adduct that arose through sequential nucleophilic additions to 4-oxo-2-nonenal. An identical adduct was observed when dGuo was allowed to react with synthetic 13-hydroperoxylinoleic acid (13-HPODE) or with 13-HPODE generated from linoleic acid by soybean lipoxygenase. We reasoned that a similar reaction may occur between dAdo and 4-oxo-2-nonenal and that a novel etheno dAdo adduct would result. Analysis of the reaction between 4-oxo-2-nonenal and dAdo revealed the presence of three major products (adducts A1, A2, and B). Adduct B was characterized as a 7-heptanone-etheno dAdo adduct by NMR spectroscopy (I, Scheme 1), although it was not possible to completely rule out the isomeric 8-heptanone-etheno dAdo adduct structure (II) shown in Scheme 1. Adducts A1 and A2 were shown to interconvert with each other and to each

dehydrate to form adduct B (Figures 7 and 8). The structures of adducts A1 and A2 were consistent with diastereomeric substituted ethano-dAdo adducts because of their mass spectral characteristics (Scheme 2). Reactions between R,β-unsaturated aldehydes and dGuo are thought to take place through a two-step mechanism involving an initial Michael addition (24). However, it has been suggested that R,β-unsaturated aldehydes undergo epoxidation before reaction with dAdo (14, 25). Nucleophilic addition of N6 from dAdo occurs at the carbonyl carbon of the epoxy aldehyde with subsequent reaction of N-1 at C-2 to form an ethano intermediate. Dehydration of this intermediate then results in the formation of an etheno adduct (14, 25). We propose that 4-oxo-2-nonenal reacted with dAdo in a similar manner (Scheme 3). Thus, nucleophilic addition of N6 occurred initially at aldehyde C-1. This was followed by reaction of N-1 at C-2 of the resulting R,β-unsaturated ketone to generate adducts A1 and A2 as a mixture of diastereomers that were able to interconvert. Subsequent dehydration of adducts A1 and A2 resulted in the formation of adduct B. Our data could not distinguish between initial Michael addition of N-1 at C-2 of the R,β-unsaturated aldehyde followed by nucleophilic addition of N-6 at aldehyde C-1. We favor initial reaction at aldehyde C-1 because this appeared to be the most electrophilic site in reactions with dGuo (21). The interconversion of adducts A1 and A2 was thought to occur through the common acyclic aldehyde intermediate as shown in Scheme 3. We reasoned that sodium borohydride would initially reduce the heptanone moiety

572

Chem. Res. Toxicol., Vol. 13, No. 7, 2000

Lee et al.

Scheme 3. Proposed Mechanism of Formation of Adduct B and Reduction Products of Adducts A1 and A2

156

Figure 11. LC/MS analysis of the dAdo adducts released by enzymatic hydrolysis of DNA that had been treated with 4-oxo2-nonenal for 8 h at 37 °C. (A) Reconstructed selected ion chromatograms for MH+ of adducts A1 and A2 (m/z 406) from a full-scan LC/MS analysis. (B) Reconstructed SRM chromatograms for adducts A1 and A2 (m/z 406 f 290) from a full-scan LC/MS2 analysis. (C) Reconstructed selected ion chromatogram for MH+ of adduct B (m/z 388) from a full-scan LC/MS analysis. (D) Reconstructed SRM chromatogram for adduct B (m/z 388 f 272) from a full-scan LC/MS2 analysis.

of adducts A1 and A2. The resulting C-2′′ hydroxyl group from adduct A1 or adduct A2 would then react with the

ring-opened aldehyde form to give a hemiacetal derivative. Further reduction of the hemiacetal derivatives would result in the addition of six more hydrogen atoms to give a compound with an MH+ ion at m/z 414 (Scheme 3). LC/MS2 analysis of this compound should result in the loss of N-9 as NH3 for generation of a product ion at m/z 397, in addition to the expected BH2+ product ion at m/z 298. A fragmentation pathway involving the loss of NH3 is impossible for the isomeric structure II (Scheme 1) because N-9 would still be attached to the nine-carbon fragment derived from 4-oxo-2-nonenal. The addition of eight hydrogen atoms would require the reduction of the purine ring. Although there does not seem to be a literature precedence for this reaction, we have shown that the structurally related molecule etheno-2′-deoxyadenosine readily undergoes sodium borohydride reduction with the addition of four hydrogen atoms. As this reaction does not proceed in the case of ethenoadenine, we hypothesize that the 2′-deoxyribose moiety activates the purine ring to facilitate reduction of the imidazole moiety. Identical reduction products were in fact observed when adducts A1 and A2 were treated with sodium borohydride. LC/MS analysis showed that eight hydrogen atoms had been added (panels A and B of Figure 10). CID and LC/MS2 analysis (Figure 10C) of the reduction product’s MH+ ion (m/z 414) revealed a major product ion derived from the loss of NH3 (Figure 10D). Therefore, adduct B must have been formed by initial nucleophilic attack from N6 of dAdo at C-1 of the 4-oxo-2-nonenal (Scheme 3). This provided unequivocal evidence that adduct B had structure I rather than the alternative structure II (Scheme 1), which would have resulted from initial nucleophilic attack from N-1 of dAdo at C-1 of the

dAdo Adducts of 4-Oxo-2-nonenal

4-oxo-2-nonenal. The presence of an intact hemiacetal ring was confirmed by LC/MS4 analysis of the product ion at m/z 281 derived from the loss of the exocyclic amine originally at N6 and the 2′-deoxyribose moiety. A major product ion observed at m/z 183 corresponded to the loss of C7H14 from the hemiacetal (Scheme 3). Evidence for reduction of the imidazole ring came from LC/MS5 analysis of m/z 183 where the loss of the CHN moiety gave an ion at m/z 156. If the imidazole ring had been intact, this fragmentation reaction would not have been possible. There is a possibility that the initial nucleophilic attack by dAdo on 4-oxo-2-nonenal had actually taken place from N-1 and that a subsequent Dimroth rearrangement to N6 similar to that observed with butadiene and styrene oxide had occurred (26, 27). However, we were unable to detect any polar N-1 products with M+ ions at m/z 406 or any deaminated (inosine) adducts with MH+ ions at m/z 407 that are typical of Dimroth rearrangement products. Therefore, we consider that initial nucleophilic attack occurred from the N6 exocyclic amino group (Scheme 3). Adducts A1, A2, and B were also characterized after enzymatic hydrolysis of 4-oxo-2-nonenal-treated calf thymus DNA (Figure 11). The detection of substituted ethano and etheno adducts in double-stranded DNA implies that they may be formed as a consequence of lipid peroxidation-mediated DNA damage. In summary, we have provided evidence that 4-oxo-2nonenal can generate a substituted etheno dAdo adduct. Furthermore, 4-oxo-2-nonenal-induced structural alterations to dAdo in double-stranded DNA in vitro were identical to those observed for the free nucleoside. There are a number of reports that etheno adducts can be formed from products of lipid peroxidation (2, 14, 17, 21, 25, 28, 29), and there is accumulating evidence which suggests that they are mediators of mutagenesis and carcinogenesis (1). Interestingly, a substituted etheno adduct was observed in the reaction between the carcinogen R-acetoxy-N-nitrosopiperidine and dGuo that was shown to arise through the intermediacy of 4-oxo-2pentenal (30, 31). An examination of the biological relevance of 4-oxo-2-nonenal-mediated formation of substituted etheno DNA adducts will be necessary in understanding its potential contribution to the etiology of cancer. It has been observed previously that 4-oxo-2nonenal can also form adducts with model amino acids. This suggests that it may also play a role in lipid peroxidation-mediated cross-linking of proteins as suggested by Xu and Sayre (32).

Acknowledgment. We gratefully acknowledge financial support from the National Institutes of Health in the form of an RO1 grant to I.A.B. (CA65878) and an NRSA fellowship to D.R. (GM19388-02).

References (1) Burcham, P. C. (1998) Genotoxic lipid peroxidation products: Their DNA damaging properties and role in formation of endogenous DNA adducts. Mutagenesis 13, 287-305. (2) Chung, F.-L., Chen, H.-J. C., and Nath, R. G. (1996) Lipid peroxidation as a potential endogenous source for the formation of exocyclic DNA adducts. Carcinogenesis 17, 2105-2111. (3) Basu, A. K., and Marnett, L. J. (1983) Unequivocal demonstration that malondialdehyde is a mutagen. Carcinogenesis 4, 331-333. (4) Esterbauer, H. (1982) Aldehydic products of lipid peroxidation. In Free Radicals, Lipid Peroxidation and Cancer (McBrien, D. C. H., and Slater, T. F., Eds.) pp 101-128, Academic Press, London.

Chem. Res. Toxicol., Vol. 13, No. 7, 2000 573 (5) Dedon, P. C., Plastaras, J. P., Rouzer, C. A., and Marnett, L. J. (1998) Indirect mutagenesis by oxidative DNA damage: formation of the pyrimidopurinone adduct of deoxyguanosine by base propenal. Proc. Natl. Acad. Sci. U.S.A. 95, 11113-11116. (6) Hamberg, M., and Samuelsson, B. (1967) On the mechanism of the biosynthesis of prostaglandins E1 and F1R. J. Biol. Chem. 242, 5336-5343. (7) Hecker, M., and Ullrich, V. (1989) On the mechanism of prostacyclin and thromboxane A2 biosynthesis. J. Biol. Chem. 264, 141150. (8) Marnett, L. J., Basu, A. K., O’Hara, S. M., Weller, P. E., Rahman, A. F. M. M., and Oliver, J. P. (1986) Reaction of malondialdehyde with guanine nucleosides: formation of adducts containing oxadiazabicyclononene residues in the base-pairing region. J. Am. Chem. Soc. 108, 1348-1350. (9) Stone, K., Ksebati, M. B., and Marnett, L. J. (1990) Investigation of the adducts formed by reaction of malondialdehyde with adenosine. Chem. Res. Toxicol. 3, 33-38. (10) Chaudhary, A. K., Reddy, R. G., Blair, I. A., and Marnett, L. J. (1996) Characterization of an N6-oxopropenyl-2′-deoxyadenosine adduct in malondialdehyde-modified DNA using liquid chromatography/electrospray ionization tandem mass spectrometry. Carcinogenesis 17, 1167-1170. (11) Esterbauer, H., Schaur, R. J., and Zollner, H. (1991) Chemistry and biochemistry of 4-hydroxynonenal, malonaldehyde and related aldehydes. Free Radical Biol. Med. 11, 81-128. (12) Pryor, W. A., and Porter, N. A. (1990) Suggested mechanisms for the production of 4-hydroxy-2-nonenal from the autoxidation of polyunsaturated fatty acids. Free Radical Biol. Med. 8, 541-543. (13) Yi, P., Zhan, D. J., Samokyszyn, V. M., Doerge, D. R., and Fu, P. P. (1997) Synthesis and 32P-postlabeling/high-performance liquid chromatography separation of diastereomeric 1,N2-(1,3-propano)2′-deoxyguanosine 3′-phosphate adducts formed from 4-hydroxy2-nonenal. Chem. Res. Toxicol. 10, 1259-1265. (14) Sodum, R. S., and Chung, F.-L. (1991) Stereoselective formation of in vitro nucleic acid adducts by 2,3-epoxy-4-hydroxynonanal. Cancer Res. 51, 137-143. (15) Chen, H.-J. C., and Chung, F.-Y. (1996) Epoxidation of trans-4hydroxy-2-nonenal by fatty acid hydroperoxides and hydrogen peroxide. Chem. Res. Toxicol. 9, 306-312. (16) Winter, C. K., Segall, H. J., and Haddon, W. F. (1986) Formation of cyclic adducts of deoxyguanosine with the aldehydes trans-4hydroxy-2-hexenal and trans-4-hydroxy-2-nonenal in vitro. Cancer Res. 46, 5682-5686. (17) Carvalho, V. M., Di Mascio, P., de Arruda Campos, I. P., Douki, T., Cadet, J., and Medeiros, M. H. G. (1998) Formation of 1,N6etheno-2′-deoxyadenosine adducts by trans,trans-2,4-decadienal. Chem. Res. Toxicol. 11, 1042-1047. (18) Zamora, R., Alaiz, M., and Hidalgo, F. J. (1999) Modification of histidine residues by 4,5-epoxy-2-alkenals. Chem. Res. Toxicol. 12, 654-660. (19) Hasegawa, K., Fujimoto, K., Kaneda, T., and Frankel, E. N. (1988) Characterization of fluorescent products from reaction of methyl linoleate hydroperoxides with adenine in the presence of Fe2+ and ascorbic acid. Biochim. Biophys. Acta 962, 371-376. (20) Hasegawa, K., Fujimoto, K., Kaneda, T., Neff, W. E., and Frankel, E. N. (1989) Formation of fluorescent products in the reaction of adenine and other bases with secondary oxidation products of methyl linoleate hydroperoxides. Agric. Biol. Chem. 53, 15751581. (21) Rindgen, D., Nakajima, M., Wehrli, S., Xu, K., and Blair, I. A. (1999) Covalent modifications to 2′-deoxyguanosine by 4-oxo-2nonenal a novel product of lipid peroxidation. Chem. Res. Toxicol. 12, 1195-1204. (22) Esterbauer, H., and Weger, W. (1967) U ¨ ber die wirkungen von aldehyden auf gesunde und maligne zellen, 3. mitt.: synthese von homologen 4-hydroxy-2-alkenalen, II. Monatsh. Chem. 98, 19942000. (23) Rouzer, C. A., Chaudhary, A. K., Nokubo, M., Ferguson, D. M., Reddy, G. R., Blair, I. A., and Marnett, L. J. (1997) Analysis of the malondialdehyde-2’-deoxyguanosine adduct pyrimidopurinone in human leukocyte DNA by gas chromatography/electron capture/ negative chemical ionization/mass spectrometry. Chem. Res. Toxicol. 10, 181-188. (24) Boerth, D. W., Eder, E., Hussain, S., and Hoffman, C. (1998) Structures of acrolein-guanine adducts: a semi-empirical selfconsistent field and nuclear magnetic resonance spectral study. Chem. Res. Toxicol. 11, 284-294. (25) Chen, H.-J. C., and Chung, F.-L. (1994) Formation of etheno adducts in reactions of enals via autoxidation. Chem. Res. Toxicol. 7, 857-860.

574

Chem. Res. Toxicol., Vol. 13, No. 7, 2000

(26) Selzer, R. R., and Elfarra, A. A. (1996) Characterization of N1and N6-adenosine adducts and N1-inosine adducts formed by the reaction of butadiene monoxide with adenosine: evidence for the N1-adenosine adducts as major initial products. Chem. Res. Toxicol. 9, 875-881. (27) Barlow, T., Takeshita, J., and Dipple, A. (1998) Deamination and Dimroth rearrangement of deoxyadenosine-styrene oxide adducts in DNA. Chem. Res. Toxicol. 11, 838-845. (28) Sodum, R. S., and Chung, F.-L. (1989) Structural characterization of adducts formed in the reaction of 2,3-epoxy-4-hydroxynonanal with deoxyguanosine. Chem. Res. Toxicol. 2, 23-28. (29) Ghissassi, F. E., Barbin, A., Nair, J., and Bartsch, H. (1995) Formation of 1,N6-ethenoadenine and 3,N4-ethenocytosine by lipid peroxidation products and nucleic acid bases. Chem. Res. Toxicol. 8, 278-283.

Lee et al. (30) Hecht, S. S., Young-Sciame, R., and Chung, F.-L. (1992) Reaction of R-acetoxy-N-nitrosopiperidine with deoxyguanosine: Oxygendependent formation of 4-oxo-2-pentenal and a 1,N2-ethenodeoxyguanosine adduct. Chem. Res. Toxicol. 5, 706-712. (31) Liu, Z., Young-Sciame, R., and Hecht, S. S. (1996) Liquid chromatography-electrospray ionization mass spectrometric detection of an ethenodeoxyguanosine adduct and its hemiaminal precursors in DNA reacted with R-acetoxy-N-nitrosopiperidine and cis-4-oxo-2-pentenal. Chem. Res. Toxicol. 9, 774-780. (32) Xu, G., and Sayre, L. M. (1998) Structural characterization of a 4-hydroxy-2-alkenal-derived fluorophore that contributes to lipoperoxidation-dependent protein cross-linking in aging and degenerative disease. Chem. Res. Toxicol. 11, 247-251.

TX000057Z