Characterization of Nucleoside Adducts of

Characterization of Nucleoside Adducts of...
3 downloads 0 Views 98KB Size
Download PDF
Chem. Res. Toxicol. 2002, 15, 373-379

373

Characterization of Nucleoside Adducts of cis-2-Butene-1,4-dial, a Reactive Metabolite of Furan Michael C. Byrns, Daniel P. Predecki,† and Lisa A. Peterson* Cancer Center and Division of Environmental and Occupational Health, School of Public Health, University of Minnesota, Mayo Mail Code 806, 420 Delaware Street SE, Minneapolis, Minnesota 55455 Received August 28, 2001

Furan is a hepatic toxicant and carcinogen in rodents. Its microsomal metabolite, cis-2butene-1,4-dial, is mutagenic in the Ames assay. Consistent with this observation, cis-2-butene1,4-dial reacts with 2′-deoxycytidine, 2′-deoxyguanosine, and 2′-deoxyadenosine to form diastereomeric adducts. HPLC analysis indicated that the rate of reaction with deoxyribonucleosides was dependent on pH. At pH 6.5, the relative reactivity was 2′-deoxycytidine > 2′-deoxyguanosine > 2′-deoxyadenosine whereas it was 2′-deoxyguanosine > 2′-deoxycytidine > 2′-deoxyadenosine at pH 8.0. Thymidine did not react with cis-2-butene-1,4-dial. The primary 2′-deoxyguanosine and 2′-deoxyadenosine reaction products were unstable and decomposed to secondary products. NMR and mass spectral analysis indicated that the initial 2′-deoxyadenosine and 2′-deoxyguanosine reaction products were hemiacetal forms of 3-(2′-deoxy-β-Derthyropentafuranosyl)-3,5,6,7-tetrahydro-6-hydroxy-7-(ethane-2′′-al)-9H-imidazo[1,2-R]purine9-one (structure 2) and 3-(2′-deoxy-β-D-erythropentafuranosyl)-3,6,7,8-tetrahydro-7-(ethane2′′-al)-8-hydroxy-3H-imidazo[2,1-i]purine (structure 4), respectively. These adducts resulted from the addition of cis-2-butene-1,4-dial to the exo- and endocyclic nitrogens of 2′-deoxyadenosine and 2′-deoxyguanosine. The data provide support for the hypothesis that cis-2-butene1,4-dial is an important genotoxic intermediate in furan-induced carcinogenesis.

Introduction Furan is an industrial chemical that has also been detected as an environmental contaminant in smog, tobacco smoke, coffee, and canned foods (1). Furan is the parent compound for a group of chemicals that are widespread in the environment (2). It is a liver and kidney toxicant and a hepatocarcinogen in rodents (3-7), and has been classified as a possible human carcinogen (8). Metabolic activation plays an important role in the toxicity of furan (9-12). Furan is converted to a toxic metabolite in a reaction catalyzed by cytochrome P-450 2E1 (9, 10). We identified the putative reactive intermediate, cis-2-butene-1,4-dial, as a microsomal metabolite of furan (Scheme 1) (13, 14). The proposed mechanism of furan-induced carcinogenesis involves reaction of cis2-butene-1,4-dial with proteins to elicit a toxic response. This toxicity then stimulates liver cell proliferation, which is thought to lead to tumor formation (4). There is some evidence that furan may act through at least a partial genotoxic mechanism (15). Furan-induced rodent tumors contain altered ras genes with unique mutations, indicating the formation of specific DNA damage (15). While furan was inactive in the Ames assay (16), its metabolite, cis-2-butene-1,4-dial, was mutagenic in the Ames tester strain TA104 (17). Therefore, it is likely that this metabolite contributes to the carcinogenic activity of furan by reacting with DNA to form mutagenic * To whom correspondence should be addressed at the University of Minnesota Cancer Center, Box 806 Mayo, 420 Delaware St. SE, Minneapolis, MN 55455. Phone: 612-626-0164, Fax: 612-626-5135, E-mail: [email protected]. † Current address: Department of Chemistry, Emory University, Atlanta, GA 30322.

Scheme 1. Proposed Bioactivation Pathway for Furan

adducts. Consistent with this proposal is a recent report that cis-2-butene-1,4-dial reacts with dCyd to form stable oxadiazabicyclooctaimine adducts (18). In this report, we characterize the products formed upon reaction of cis-2butene-1,4-dial with dGuo and dAdo.

Experimental Procedures Materials. cis-2-Butene-1,4-dial was prepared as previously described (13, 17). Instrumental Analysis. HPLC analyses were performed with a Prodigy C18 column (5 µm, 150 × 4.6 mm; Phenomenex, Torrence, CA) on a Waters HPLC system linked to a Waters 996 photodiode array detector. HPLC chromatograms were recorded at 254 nm unless otherwise noted. HPLC Method 1. The column was eluted for 5 min with 100% 100 mM ammonium acetate, followed by a linear gradient to 100 mM ammonium acetate containing 25% acetonitrile over 25 min. Retention times (min): dCyd, 11.2; dCyd-adducts, 19.5, 19.7; dGuo, 19.5; dGuo-adducts, 20.2, 20.7; dAdo, 22.5; dAdoadducts, 18.5, 19.0. HPLC Method 2. This method used the same gradient as HPLC Method 1, but 10 mM ammonium acetate was substituted for 100 mM ammonium acetate. HPLC Method 3. The column was isocratically eluted with 3% acetonitrile in water. LC-MS and LC-MS/MS analyses were carried out in the positive electrospray ionization mode with a ThermoFinnigan LCQ Deca instrument (San Jose, CA) interfaced with a Waters Alliance HPLC system equipped with a Shimadzu SPD-10A

10.1021/tx0101402 CCC: $22.00 © 2002 American Chemical Society Published on Web 02/23/2002

374

Chem. Res. Toxicol., Vol. 15, No. 3, 2002

Byrns et al.

UV-visible detector, using HPLC Method 1. The API source of the mass spectrometer was set as follows: voltage, 1 kV; current, 80 µA; capillary temperature, 275 °C. NMR analyses were performed on either a Varian VI-300 or a VI-500 MHz NMR. Reaction of cis-2-Butene-1,4-dial with Deoxyribonucleosides. cis-2-Butene-1,4-dial (25 mM) was reacted with dAdo, dCyd, dGuo, or dThd (5 mM) in 50 mM sodium phosphate, pH 7.4, at 37 °C (total volume 1 or 10 mL). The effect of pH on the reaction was determined in 50 mM sodium phosphate, pH 6, 6.5, 7, 7.5, or 8. Reaction progress was monitored by HPLC analysis using HPLC Method 1. Preparative scale isolation of the reaction products was achieved using HPLC Methods 2 (dAdo) and 3 (dGuo). dGuo product peaks were collected separately. dAdo adduct peaks were collected as a mixture of isomers. All NMR assignments were supported with COSY1 results. 13C NMR chemical shift assignments were determined by HMQC experiments. Adduct stability was established by incubating the purified reaction products in 50 mM sodium phosphate, pH 7.4, at 37 °C. Aliquots were removed at various times and analyzed by HPLC using HPLC Method 1. (A) cis-2-Butene-1,4-dial-dGuo Reaction Products. ESIMS m/z (rel intensity) 410 (M+1+acetate, 100), 352 (M+1, 70), 236 (M+1-2′-deoxyribose); ESI-MS/MS on M+1 (relative intensity) m/z 236 (100, M-2′-deoxyribose); 13C NMR (DMSO-d6) major isomer in both peak A1 and peak A2: δ 135 (C2), 98 (C2′′), 88 (C6), 88 (C4′), 84 (C1′), 70 (C3′), 61 (C5′), 59 (C7), 40 (C2′), 38 (C1′′). (B) cis-2-Butene-1,4-dial-dAdo Reaction Products. ESIMS m/z (rel intensity) 336 (M+1, 100), 394 (M+1+acetate, 50), ESI-MS/MS on M+1 (rel intensity) m/z 220 (100, M-2′-deoxyribose); 13C NMR (D2O) δ 144 (C5), 144 (C2), 100 (C2′′), 91 (C8), 87 (C4′), 84 (C1′), 71 (C3′), 60 (C5′), 36 (C2′), 38 (C1′′).

Results General Reactions. Incubation of cis-2-butene-1,4dial with 2′-deoxyribonucleosides in sodium phosphate, pH 7.4, at 37 °C led to the formation of new peaks in the HPLC traces of the reactions with dCyd, dGuo, and dAdo but not dThd (Figure 1). In all cases, the reaction with the deoxyribonucleosides generated two new closely eluting peaks of equal intensity. The cis-2-butene-1,4dial-dC reaction products have been previously characterized as oxadiazabicyclo(3.3.0)octaimine adducts of dCyd (structure 1, Scheme 2A) (18). On the basis of product formation, we judged the relative reactivity to be dCyd > dGuo > dAdo at pH 7.4. This differs from the order reported by Gingipalli and Dedon (18). One possible explanation for this difference is the observation that the initial dAdo and dGuo adducts were not detected when the HPLC analyses were performed with a Phenomenex Bondclone C18 column. It is likely that the initial dAdo and dGuo reaction products coeluted with the unreacted deoxyribonucleosides under these circumstances. The effect of pH on the reactions was examined (Figure 2). The reaction between dGuo and cis-2-butene-1,4-dial was profoundly affected by small changes in pH (Figure 2B), but the reactions between cis-2-butene-1,4-dial and dCyd/dAdo were relatively unaffected by pH (Figure 2A,C). Therefore, the pH of the reaction solution dramatically influenced the relative reactivity of the deoxyribonucleosides (Table 1). In addition, the stability of the cis-2-butene-1,4-dialdGuo reaction products decreased as the pH increased. At pH 6.5, the disappearance of dGuo was accompanied 1 Abbreviations: COSY, 1H-1H two-dimensional correlated spectroscopy; HMQC NMR experiment, heteronuclear multiple quantum correlation NMR experiment.

Figure 1. HPLC traces of the reaction mixtures of 25 mM cis2-butene-1,4-dial and 25 mM nucleoside following 6 h at 37 °C. (A) dGuo; (B) dAdo. The reaction mixtures were separated on a Phenomenex Prodigy C18 column using HPLC method 1.

by a corresponding increase in cis-2-butene-1,4-dial-dGuo reaction products. However, the decrease in dGuo levels was greater than the increase in adduct levels when the pH of the reaction exceeded 7.0. Adduct formation at 6 h accounted for only 90%, 70%, and 40% of dGuo disappearance at pH 7.0, 7.5, and 8.0, respectively. In the dAdo and dCyd reactions, nucleoside disappearance directly coincided with adduct formation at all pHs investigated. Characterization of cis-2-Butene-1,4-dial-dGuo Reaction Products. The cis-2-butene-1,4-dial-dGuo initial reaction products were relatively unstable. When the reaction progress was monitored by HPLC for more than 24 h, the initial reaction products disappeared, and multiple reaction products with longer retention times appeared (data not shown). This instability was also observed in the pH studies described above. The stability of the adducts was further examined by incubating the isolated peaks in pH 7.4 buffer at 37 °C. Aliquots were removed at various times up to 24 h and analyzed by HPLC with diode array detection. No equilibration between the two peaks was observed. Both peaks decomposed to multiple unidentified compounds under these conditions as judged by HPLC analysis (data not shown). The nucleoside dGuo was not a significant decomposition product. Peak A1 had a half-life of 2.2 h, and peak A2 had a half-life of 3.4 h. The adducts were stable for at least 1 week if stored at temperatures at -20 °C. With these precautions, sufficient amounts of peaks A1 and A2 were isolated for structural characterization. The UV spectra of both product peaks were identical, indicating that these products were isomers. The spectra

cis-2-Butene-1,4-dial-Derived Adducts

Chem. Res. Toxicol., Vol. 15, No. 3, 2002 375

Scheme 2. Reaction of cis-2-Butene-1,4-dial with Deoxyribonucleosidesa

a

(A) dCyd; (B) dGuo; (C) dGuo; (D) dAdo; dR ) 2′-deoxyribose.

at pH 7 contained a λmax at 250.9 nm with a shoulder at 275 nm (Figure 3A). They are similar to UV spectra reported for N2-substituted dGuo adducts at pH 1 and 7 (Figure 3B) (19, 20). At pH 13, the UV spectra of the adducts changed irreversibly (λmax at 238 and 349 nm). Neutralization of the solution did not restore the UV spectrum initially observed at pH 7. Analysis of this solution by HPLC indicated that the adducts had partially decomposed to multiple products, including dGuo (data not shown). Positive electrospray mass spectral analysis also demonstrated that the reaction products were isomeric compounds. The mass spectra for both HPLC peaks contained a protonated molecular ion at m/z 352. The MS/MS of this ion yielded a single fragment at m/z 236, which results from loss of 2′-deoxyribose. These data are consistent with the addition of one molecule of cis-2-butene1,4-dial to the base region of dGuo with no loss of water. NMR analysis of each peak indicated that each one contained two coeluting isomers at a ratio of 1:2. Since there was no observed equilibration between the two peaks, cis-2-butene-1,4-dial reacted with dGuo to form four isomers. Consistently, the 1H NMR spectrum of the two peaks collected together displayed signals for all four isomers.

The 1H NMR chemical shifts for each isomer are listed in Table 2. All protons were assigned based on D2O exchange, COSY and HMQC experiments. The 1H NMR spectra indicated that each isomer contained 13 protons attached to carbon atoms and 4 exchangeable protons on heteroatoms. The chemical shifts of the protons derived from cis-2-butene-1,4-dial are similar to those reported for the cis-2-butene-1,4-dial-dCyd reaction products (18), supporting a bicyclic structure for the dGuo adducts. The cyclic nature of the adducts is further supported by the distinct chemical shifts observed for the diastereotopic C1′′ protons which are separated by 0.2-0.3 ppm. The involvement of the N1 and N2 nitrogen atoms of dGuo in adduct formation was indicated by the disappearance of the N1 proton at 10.6 ppm and the N2 protons at 6.4 ppm. The five-membered ring structure proposed in Scheme 2B is supported by the coupling relationships between the ring protons and by the integration of the signals. The adduct-related exchangeable proton resonating at either 6.7 ppm for the minor isomer or 6.3 ppm for the major isomer was assigned to the 2′′-hydroxyl group. This proton was coupled to the 2′′ proton that resonates at 5.5 ppm for the minor isomer and at 5.4 ppm for the major isomer. The 2′′ protons were coupled to the diastereotopic protons resonating at 2.1 and 2.3 ppm. These protons were assigned to the 1′′ position. These

376

Chem. Res. Toxicol., Vol. 15, No. 3, 2002

Byrns et al.

Figure 3. (A) UV spectra for cis-2-butene-1,4-dial-dGuo adducts at pH 6.8. (B) UV spectra of peak A2 at pH 1, 7, and 13.

Figure 2. pH dependence of cis-2-butene-1,4-dial-nucleoside reactions. (A) dCyd; (B) dGuo; and (C) dAdo. The reactions were monitored by HPLC analysis at 280 nm (dCyd) or 254 nm (dGuo and dAdo). The data points are averages from two experiments. pH 6.5 ([); pH 7.0 (9); pH 7.5 (2); pH 8.0 (b). The data point at 6 h for the dCyd reaction at pH 7.5 and the data points at 5.2 and 6 h for the dCyd reaction at pH 8.0 are from single experiments. Table 1. Effect of pH on Relative Reactivity of cis-2-Butene-1,4-dial with Deoxyribonucleosidesa pH

relative reactivity

6.5 7.0 7.5 8.0

dCyd . dAdo ≈ dGuo dCyd > dGuo > dAdo dCyd > dGuo . dAdo dGuo > dCyd . dAdo

a Based on disappearance of nucleoside as determined by HPLC analysis.

protons, in turn, were coupled to the methine proton assigned to H7 (4.9 ppm in all isomers). Finally, H7 was also coupled to the methine proton signal at 5.8 ppm, which was assigned to H6. The D2O exchangeable singlets occurring at approximately 9.0 ppm were assigned to H5. The HMQC experiment indicated that this proton was not attached to a carbon atom, eliminating the possibility that it is part of an exposed aldehydic group. In addition, the C2′′ carbon atom has a 13C chemical shift of 98 ppm, which is more characteristic of a carbon

attached to two oxygen atoms than a carbonyl carbon (180-220 ppm) (21). There are two possible structures that could result from reaction of cis-2-butene-1,4-dial at both the N1 and N2 positions of dGuo. One is structure 2 in Scheme 2B, and the second is structure 3 in Scheme 2C. These compounds differ in the attachment of the terminal carbon of cis-2-butene-1,4-dial to either the N2 (2) or the N1 (3) nitrogen of dGuo. Compound 3 was eliminated since H5 was coupled to H6 in all four isomers. Therefore, all four compounds are stereoisomers of 2, the hemiacetal form of 3-(2′-deoxy-β-D-erthyropentafuranosyl)-3,5,6,7tetrahydro-6-hydroxy-7-(ethane-2′′-al)-9H-imidazo[1,2-R]purine-9-one. Further support for this structure is the similarity in chemical shift for the N5 proton of 2 to that reported for a structurally related 4-oxo-2-nonenalderived dGuo adduct, the hemiacetal form of 3-(2′-deoxyβ-D-erthyropentafuranosyl)-3,5,6,7-tetrahydro-6-hydroxy7-(heptane-2′′-one)-9H-imidazo[1,2-R]purine-9-one (21). Characterization of cis-2-Butene-1,4-dial-dAdo Reaction Products. The initial cis-2-butene-1,4-dialdAdo reaction products were also isomeric. They appeared as two closely eluting HPLC peaks just prior to dAdo (Figure 1C). These compounds had identical UV spectra with λmax at 263 nm and a slight shoulder at 275 nm (Figure 4A). These spectra are similar to those reported for N1,N6-dAdo adducts of 4-oxo-2-nonenal (22). The UV spectrum of the adducts was not dramatically affected by pH (Figure 4B). LC-ESI MS analysis of each peak yielded a protonated molecular ion at m/z 336, consistent with 1:1 addition of cis-2-butene-1,4-dial to dAdo. MS/MS indicated the loss of the 2′-deoxyribose from these adducts, consistent with the alkylation occurring on the base portion of the nucleoside. The two peaks were in equilibrium with one another and, thereby, difficult to separate, so they were isolated

cis-2-Butene-1,4-dial-Derived Adducts

Chem. Res. Toxicol., Vol. 15, No. 3, 2002 377

Table 2. 500 MHz 1H NMR Data for cis-2-Butene-1,4-dial-dGuo Reaction Productsa

peak A1 major isomer peak A1 minor isomer peak A2 major isomer peak A2 minor isomer

peak A1 major isomer peak A1 minor isomer peak A2 major isomer peak A2 minor isomer

H5b

H2

2′′-OHb

H6

H2′′

H7

H1′′

9.02 (s) 8.98 (s) 9.00 (s) 8.97 (s)

7.94 (s) 7.95 (s) 7.94 (s) 7.95 (s)

6.32 (br s) 6.67 (br s) 6.31 (br s) 6.67 (br s)

5.78 (m) 5.78 (m) 5.77 (m) 5.77 (m)

5.50 (m) 5.39 (m) 5.51 (m) 5.40 (m)

4.92 (m) 4.92 (m) 4.91 (m) 4.91 (m)

2.42 (m), 2.15 (m) 2.32 (m), 2.12 (m) 2.42 (d), 2.19 (m) 2.33 (dm), 2.14 (m)

1′H

3′-OHb

5′-OHb

3′H

4′H

5′H

2′H

6.09 (dd)d 6.09 (dd)d 6.09 (dd)d 6.09 (dd)d

5.29 (br s) 5.29 (br s) 5.33 (br s) 5.33 (br s)

4.95 (br s) 4.95 (br s) 4.95 (br s) 4.95 (br s)

4.31 (m) 4.31 (m) 4.32 (m) 4.32 (m)

3.80 (m) 3.80 (m) 3.83 (m) 3.83 (m)

3.6-3.4 (m) 3.6-3.4 (m) 3.6-3.4 (m) 3.6-3.4 (m)

2.50c (m), 2.21 (m) 2.50c (m), 2.21 (m) 2.50c (m), 2.2 (m) 2.50c (m), 2.2 (m)

a Values are chemical shifts (δ) in DMSO-d . b Exchanged upon addition of deuterium oxide. c Obscured by DMSO peak; assignment 6 determined from 1H-1H two-dimensional COSY NMR spectrum. d J ) 6.4, 6.4 Hz.

Figure 4. (A) UV spectra for cis-2-butene-1,4-dial-dAdo adducts at pH 6.8. (B) UV spectra for cis-2-butene-1,4-dial-dAdo adducts at pH 1, 7, and 13.

together for NMR analysis. 1H NMR analysis in DMSOd6 indicated the presence of diastereomers present in approximately equal amounts. The chemical shift assignments are given in Table 3. As with the dGuo adducts, all protons were assigned based on COSY and HMQC experiments. These data support the conclusion that the adducts formed as a result of the addition of cis2-butene-1,4-dial to the N1 and N6 positions of dAdo (compound 4, Scheme 2D). As with the dGuo adducts, the chemical shifts of the protons associated with the carbons derived from cis-2-butene-1,4-dial are very similar to those reported for the dCyd adducts (18). The coupling relationships and signal integrations are con-

sistent with the five-membered ring structure proposed in Scheme 2D (structure 4). The exchangeable protons for each isomer that resonated at 6.06 and 6.49 ppm were assigned to the hydroxyl groups attached to the 2′′ position. These protons were coupled to the 2′′-methine protons resonating at 5.47 and 5.28 ppm, respectively. The 2′′ protons were also coupled to the 1′′-methylene protons corresponding to each isomer (2.27 and 2.17 ppm). The 1′′ protons were also coupled to the H7 methine proton (4.77 and 4.85 ppm) which was coupled to the H8 methine proton (6.01 and 6.00 ppm). On the basis of this NMR data, we concluded that the adducts were the hemiacetal form of 3-(2′-deoxy-β-D-erythropentafuranosyl)-3,6,7,8-tetrahydro-7-(ethane-2′′-al)-8-hydroxy3H-imidazo[2,1-i]purine. The 1H NMR spectrum of the adducts in D2O (Table 3) indicated significant exchange of the 1′′ protons. The 1′′ protons had exchanged by >50% in approximately 0.5 h. In addition, electrospray mass spectral analysis of the NMR sample stored in D2O for 2 weeks at -20 °C indicated the incorporation of 1-3 atoms of deuterium. These data support the involvement of a ring-opened aldehyde intermediate in the process (Scheme 2D). However, the predominant form in D2O was the hemiacetal form since the 13C chemical shift of the C2′′ atom was 100 ppm, which is distinctly different from that expected for a carbonyl carbon (180-220 ppm) (21). The NMR spectrum acquired in D2O also provided evidence that the adducts were unstable and decomposed to new compounds that are currently unidentified. The stability of the adducts under physiological conditions was determined by incubating the isolated peaks in pH 7.4 buffer at 37 °C. Aliquots were removed at various times up to 24 h and analyzed by HPLC with diode array detection. The half-life of the adducts was 6.1 h. They decomposed to unidentified compounds and a significant amount of unmodified dAdo, indicating that the adducts are significantly unstable under physiological conditions.

Discussion The results demonstrate that cis-2-butene-1,4-dial reacts with dGuo and dAdo to form bicyclic adducts

378

Chem. Res. Toxicol., Vol. 15, No. 3, 2002

Byrns et al.

Table 3. 500 MHz 1H NMR Data for cis-2-Butene-1,4-dial-dAdo Reaction Products in DMSO-d6a

DMSO-d6 isomer 1a isomer 2a D2O isomer 1a isomer 2a

DMSO-d6 isomers 1 and 2a D2O isomers 1 and 2b a

H5

H2

2′′ OH

H8

H2′′

H7

H1′′

8.24 (s) 8.19 (s)

8.17 (s) 8.17 (s)

6.06 (br s) 6.49 (br s)

6.01 (m) 6.00 (m)

5.47 (m) 5.28 (m)

4.77 (m) 4.85 (m)

2.27 (m) 2.17 (m)

8.69 (s) 8.69 (s)

8.42 (s) 8.41 (s)

6.24 (m) 6.24 (m)

5.66 (m) 5.66 (m)

5.50 (m) 5.50 (m)

2.53 (m) 2.53 (m)

1′H

3′ OH

5′ OH

3′H

4′H

5′H

2′H

6.23 (m)

5.3 (br s)

5.0 (br s)

4.35 (m)

3.83 (m)

3.6-3.4 (m)

2.55 (m), 2.1 (m)

4.4 (m)

3.96 (m)

3.6 (m)

2.7 (m), 2.4 (m)

6.38 (dd)c

Values are chemical shifts (δ) in DMSO-d6. Values are chemical shifts (δ) in D2O. J ) 6.4, 6.4 Hz. b

c

Scheme 3. Tautermeric Forms of cis-2-Butene-1,4-dial-dGuo Adductsa

adR

) 2′-deoxyribose.

involving the N1 and N2 positions of dGuo and the N1 and N6 positions of dAdo. These adducts are structurally related to the oxadiazabicyclo(3.3.0)octaimine adducts reported for dCyd (18). Based on these structures, we propose a general reaction mechanism involving initial reaction of the C1 atom of cis-2-butene-1,4-dial with the exocyclic nitrogen atom of the nucleoside (N4 of dCyd, N2 of dGuo, and N6 of dAdo; Scheme 2). This event is followed by 1,4-addition of the adjacent endocyclic nitrogen atom (N3 of dCyd, and N1 of dGuo, and dAdo) to the double bond of the remaining R,β-unsaturated aldehyde and subsequent attack of the alcohol on the second aldehyde group to form the final product. An alternative reaction mechanism, which cannot be ruled out, involves initial Michael addition to the double bond by the endocyclic nitrogen followed by addition of the exocyclic nitrogen to the terminal aldehyde. The relative reactivity of the deoxyribonucleosides with cis-2-butene-1,4-dial (Table 1) parallels that observed for formaldehyde (23). The relative reactivity of formaldehyde with the deoxyribonucleosides is the same as cis2-butene-1,4-dial at pH 6.5 (dCyd . dGuo ≈ dAdo). As with cis-2-butene-1,4-dial, the rate of reaction of formaldehyde with dAdo or dCyd was relatively unaffected by pH whereas the reaction with dGuo was profoundly increased when the pH exceeded 7.0. The increased reaction of formaldehyde with dGuo was attributed to a shift from reaction with the N2 position to reaction with the N1 position. In the case of cis-2-butene-1,4-dial, the deprotonation of the N1 of dGuo probably assists the cyclization reaction. Alternatively, it would also increase the extent of 1,4-addition of this position to cis-2-butene1,4-dial. Since there were only small effects of pH on the

reactivity of cis-2-butene-1,4-dial with dAdo and dCyd, pH effects on the reactivity of cis-2-butene-1,4-dial were minimal. The dGuo- and dAdo-cis-2-butene-1,4-dial adducts are substantially unstable under physiological conditions. In comparison, dCyd-cis-2-butene-1,4-dial adducts have a much longer half-life of 275 h under similar conditions (18). In addition, the dCyd (18) and dAdo isomeric adducts interconverted, but the dGuo adducts did not. It is possible that all three reaction products are in equilibrium with the ring-open aldehyde form of the adduct (Scheme 2). For the dCyd adducts, this aldehydic form of the adduct preferentially ring closes to form the cyclic hemiacetal, explaining the facile interconversion of the isomers. The dAdo adducts have a similar equilibrium between the ring-opened aldehyde and the cyclic hemiacetal, as indicated by the rapid exchange of the C1′′ protons in D2O. However, these isomeric adducts are not only in equilibrium with one another but also decompose to starting material as well as other uncharacterized decomposition products. It is likely that the dGuo adducts’ instability results from the presence of the nucleophilic lone pair of electrons on the N2 nitrogen atom. This allows for rearrangement to a variety of compounds (Scheme 3). It is possible that both the dAdo and the dGuo adducts are decomposing to form 1,N2-etheno adducts. A structurally related compound, cis-4-oxo-2-pentenal, reacts with 2′-deoxyguanosine to form a 7-(2-oxopropyl)-substituted 1,N2-ethenodeoxyguanosine adduct (24). Similarly, 4-oxo-2-nonenal reacts with dAdo and dGuo to form relatively stable intermediates that rearrange to substituted 1,N2-etheno adducts (21, 22). The structure of the

cis-2-Butene-1,4-dial-Derived Adducts

dAdo- and dGuo-cis-2-butene-1,4-dial decomposition products is currently under investigation. Once this chemistry is fully characterized, we will investigate the relative amounts of these adducts in cis-2-butene-1,4-dial-treated DNA. The ultimate goal of this project is to develop an analytical method to detect these adducts in DNA from furan-treated animals. The presence of the masked aldehyde in all of the cis2-butene-1,4-dial-derived adducts indicates that these adducts likely retain electrophilic reactivity similar to that reported for the exocyclic DNA adduct pyrimido[1,2R]purin-10-(3H)-one, which is formed by reaction of dGuo with malondialdehyde or base propenal (25, 26). cis-2Butene-1,4-dial causes DNA single-strand breaks and DNA cross-links in CHO cells (27). The mono adducts characterized in this report are potential precursors to these cross-links. It is likely that these adducts contribute to both the carcinogenic and toxic effects of furan. The ability of cis-2-butene-1,4-dial to react with deoxyribonucleosides to form reasonably stable adducts indicates that it can play a direct genotoxic role in furaninduced carcinogenesis. Consistently, cis-2-butene-1,4dial was a direct activity mutagen in TA104 (17). The observation that furan induces unique ras oncogene mutational profiles in rodent liver tumors is also consistent with a genotoxic mechanism for tumor induction (15, 28). Future studies will explore the role of these adducts in furan-induced tumorigenesis and toxicity.

Acknowledgment. We thank Ms. Rebecca Krenz and Ms. Selina Jaman for their technical assistance, Dr. Letitia Yao for her assistance with the NMR studies, and Dr. Peter Villalta for his assistance with mass spectral analysis. The Analytical Chemistry and Biomarkers Core Facility at the University of Minnesota Cancer Center is funded by National Cancer Institute Center Grant CA77598. This research was funded by ES-10577 from the National Institutes of Health. M.C.B. is the 2001-2002 Bond Scholar in the Division of Environmental and Occupational Health, University of Minnesota, and is funded by NIH Training Grant ES-10956. Supporting Information Available: 1H NMR and 1H-1H two-dimensional COSY NMR spectra of the dGuo- and dAdocis-2-butene-1,4-dial adducts. This material is available free of charge via the Internet at http://pubs.acs.org.

References (1) Capurro, P. U. (1973) Effects of exposure to solvents caused by air pollution with special reference to CCl4 and its distribution in air. Clin. Toxicol. 6, 109-124. (2) Maga, J. (1979) Furans in foods. Crit. Rev. Food Sci. Nutr. 11, 355-366. (3) Sirica, A. E. (1996) Biliary proliferation and adaptation in furaninduced rat liver injury and carcinogenesis. Toxicol. Pathol. 24, 90-99. (4) Wilson, D. M., Goldsworthy, T. L., Popp, J. A., and Butterworth, B. E. (1992) Evaluation of genotoxicity, pathological lesions, and cell proliferation in livers of rats and mice treated with furan. Environ. Mol. Mutagen. 19, 209-222. (5) National Toxicology Program (1993) Toxicology and carcinogenesis studies of furan in F344/N rats and B6C3F1 mice, U.S. Department of Health and Human Services, Public Health Service, National Institutes of Health, Research Triangle Park, NC. (6) Elmore, L. W., and Sirica, A. E. (1993) “Intestinal-type” of adenocarcinoma preferentially induced in right/caudate liver lobes of rats treated with furan. Cancer Res. 53, 254-259. (7) Maronpot, R. R., Giles, H. D., Dykes, D. J., and Irwin, R. D. (1991) Furan-induced hepatic cholangiocarcinomas in Fischer 344 rats. Toxicol. Pathol. 19, 561-570.

Chem. Res. Toxicol., Vol. 15, No. 3, 2002 379 (8) National Toxicology Program (2000) 9th Report on Carcinogens, U.S. Department of Health and Human Services, Washington, DC. (9) Carfagna, M. A., Held, S. D., and Kedderis, G. L. (1993) Furaninduced cytolethality in isolated rat hepatocytes: Correspondence with in vivo dosimetry. Toxicol. Appl. Pharmacol. 123, 265273. (10) Kedderis, G. L., Carfagna, M. A., Held, S. D., Batra, R., Murphy, J. E., and Gargas, M. L. (1993) Kinetic analysis of furan biotransformation by F-344 rats in vivo and in vitro. Toxicol. Appl. Pharmacol. 123, 274-282. (11) Parmar, D., and Burka, L. T. (1993) Studies on the interaction of furan with hepatic cytochrome P-450. J. Biochem. Toxicol. 8, 1-9. (12) Masuda, Y., Nakayama, N., Yamaguchi, A., and Murohashi, M. (1984) The effects of diethyldithiocarbamate and carbon disulfide on acute nephtrotoxicity induced by furan, bromobenzene and cephaloridine in mice. Jpn. J. Pharmacol. 34, 221-229. (13) Chen, L. J., Hecht, S. S., and Peterson, L. A. (1995) Identification of cis-2-butene-1,4-dial as a microsomal metabolite of furan. Chem. Res. Toxicol. 8, 903-906. (14) Chen, L. J., Hecht, S. S., and Peterson, L. A. (1997) Characterization of amino acid and glutathione adducts of cis-2-butene-1,4dial, a reactive metabolite of furan. Chem. Res. Toxicol. 10, 866874. (15) Johansson, E., Reynolds, S., Anderson, M., and Maronpot, R. (1997) Frequency of Ha-ras-1 gene mutations inversely correlated with furan dose in mouse liver tumors. Mol. Carcinog. 18, 199205. (16) Mortelmans, K., Haworth, S., Lawlor, T., Speck, W., Tainer, B., and Zeiger, E. (1986) Salmonella mutagenicity tests. II. Results from the testing of 270 chemicals. Environ. Mutagen. 7 (Suppl.), 1-119. (17) Peterson, L. A., Naruko, K. C., and Predecki, D. (2000) A reactive metabolite of furan, cis-2-butene-1,4-dial, is mutagenic in the Ames assay. Chem. Res. Toxicol. 13, 531-534. (18) Gingipalli, L., and Dedon, P. C. (2001) Reaction of cis- and trans2-butene-1,4-dial with 2′-deoxycytidine to form stable oxadiazabicyclooctaimine adducts. J. Am. Chem. Soc. 123, 2664-2665. (19) Young-Sciame, R., Wang, M., Chung, F. L., and Hecht, S. S. (1995) Reactions of R-acetoxy-N-nitrosopyrrolidine and R-acetoxy-Nnitrosopiperidine with deoxyguanosine: formation of N2-tetrahydrofuranyl and N2-tetrahydropyranyl adducts. Chem. Res. Toxicol. 8, 607-616. (20) 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-nitrosopiperdine and cis-4-oxo-2-pentenal. Chem. Res. Toxicol. 9, 774-780. (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) Lee, S. H., Rindgen, D., Bible, R. H., Jr., Hajdu, E., and Blair, I. A. (2000) Characterization of 2′-deoxyadenosine adducts derived from 4-oxo-2-nonenal, a novel product of lipid peroxidation. Chem. Res. Toxicol. 13, 565-574. (23) McGhee, J. D., and Von Hippel, P. H. (1975) Formaldehyde as a probe of DNA structure. I. Reaction with exocyclic amino groups of DNA bases. Biochemistry 14, 1281-1296. (24) 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. (25) Schnetz-Boutaud, N., Daniels, J. S., Hashim, M. F., Scholl, P., Burrus, T., and Marnett, L. J. (2000) Pyrimido[1,2-R]purin-10(3H)-one: a reactive electrophile in the genome. Chem. Res. Toxicol. 13, 967-970. (26) Niedernhofer, L. J., Riley, M., Schnetz-Boutaud, N., Sanduwaran, G., Chaudhary, A. K., Reddy, G. R., and Marnett, L. J. (1997) Temperature-dependent formation of a conjugate between tris(hydroxymethyl)aminomethane buffer and the malondialdehydeDNA adduct pyrimidopurinone. Chem. Res. Toxicol. 10, 556561. (27) Marinari, U. M., Ferro, M., Sciaba, L., Finollo, R., Bassi, A. M., and Brambilla, G. (1984) DNA-damaging activity of biotic and xenobiotic aldehydes in chinese hamster ovary cells. Cell Biochem. Funct. 2, 243-248. (28) Reynolds, S. H., Stowers, S. J., Patterson, R. M., Maronpot, R. R., Aaronson, S. A., and Anderson, M. W. (1987) Activated oncogenes in B6C3F1 mouse liver tumors: Implications for risk assessment. Science 237, 1309-1316.

TX0101402