Liquid Chromatography−Electrospray Ionization Mass Spectrometric

Zhi Liu, Ruth Young-Sciame, and Stephen S. Hecht*. American Health Foundation, One Dana Road, Valhalla, New York 10595. Received December 11, 1995X...
0 downloads 0 Views 345KB Size
774

Chem. Res. Toxicol. 1996, 9, 774-780

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 Zhi Liu, Ruth Young-Sciame, and Stephen S. Hecht* American Health Foundation, One Dana Road, Valhalla, New York 10595 Received December 11, 1995X

N-Nitrosopiperidine, a carcinogenic cyclic nitrosamine that occurs in the diet and may be formed endogenously, is believed to be metabolically activated by R-hydroxylation. The DNA reactive compounds that could be formed in this process have been studied using R-acetoxyN-nitrosopiperidine as a model. Previous studies have shown that 4-oxo-2-pentenal is one product of the hydrolysis of R-acetoxyNPIP and that it reacts with deoxyguanosine to produce 7-(2-oxopropyl)-5,9-dihydro-9-oxo-3-β-D-deoxyribofuranosylimidazo[1,2-a]purine (7-(2-oxopropyl)1,N2-ethenodG). Several other products were formed in that reaction, and these have now been identified as diastereomers of 7-(2-oxopropyl)-5-hydroxy-5,6,7,9-tetrahydro-9-oxo-3-β-Ddeoxyribofuranosylimidazo[1,2-a]purine, the hemiaminal precursors to 7-(2-oxopropyl)-1,N2ethenodG. Their structures were characterized by electrospray ionization mass spectrometry (ESI-MS), and by reduction with NaBH4 followed by hydrolysis to 7-(2-hydroxypropyl)-5,6,7,9tetrahydro-9-oxoimidazo[1,2-a]purine, which was characterized by 1H-NMR, MS, and UV. The presence of 7-(2-oxopropyl)-1,N2-ethenodG and its hemiaminal precursors in DNA reacted with either R-acetoxy-N-nitrosopiperidine or cis-4-oxo-2-pentenal was confirmed by LC-ESI-MS and LC-ESI-MS/MS. The results of this study demonstrate that ethenodG adducts and their precursors are present in DNA reacted with R-acetoxy-N-nitrosopiperidine, which suggests a possible basis for the unique carcinogenic properties of this nitrosamine.

Introduction N-Nitrosopiperidine (NPIP)1 induces tumors of the esophagus and liver in rats (1-3). This cyclic nitrosamine carcinogen has been identified in the diet, and there is substantial evidence supporting the hypothesis that it could form endogenously in humans (4, 5). Little is known about the DNA adducts which would be produced upon metabolic activation of NPIP. By analogy to other nitrosamines, R-hydroxylation of NPIP is expected to be a major pathway of metabolic activation (6). The relevant reactive intermediates resulting from R-hydroxylation can be generated in vitro by hydrolysis of R-acetoxyNPIP (5, 7). In previous work, we have demonstrated that one of the major adducts produced upon solvolysis of R-acetoxyNPIP in the presence of dG or DNA is N2-(3,4,5,6tetrahydro-2H-pyran-2-yl)dG (THP-dG) (5, 8). This was formed via a cyclic oxonium ion or related intermediates. Another electrophile produced in the solvolysis of R-acet* To whom correspondence and requests for reprints should be addressed. X Abstract published in Advance ACS Abstracts, May 1, 1996. 1 Abbreviations: NPIP, N-nitrosopiperidine; R-acetoxyNPIP, R-acetoxy-N-nitrosopiperidine; THP-dG, N2-(3,4,5,6-tetrahydro-2H-pyran-2yl)dG; dG, deoxyguanosine; 7-(2-oxopropyl)-1,N2-ethenodG, 7-(2oxopropyl)-5,9-dihydro-9-oxo-3-β-D-deoxyribofuranosylimidazo[1,2-a]purine; LC-ESI-MS, liquid chromatography-electrospray ionization mass spectrometry; LC-ESI-MS/MS, liquid chromatography-electrospray ionization tandem mass spectrometry; NPYR, N-nitrosopyrrolidine.

S0893-228x(95)00206-2 CCC: $12.00

oxyNPIP is 4-oxo-2-pentenal (7) (also referred to in the literature as acetylacrolein, 1, Figure 1). This reactive compound is also a metabolite of 2-methylfuran and an atmospheric degradation product of toluene (8-11). cis4-Oxo-2-pentenal reacts with dG to produce 7-(2-oxopropyl)-5,9-dihydro-9-oxo-3-β-D-deoxyribofuranosylimidazo[1,2-a]purine (7-(2-oxopropyl)-1,N2-ethenodG) (3, Figure 1), along with several major unidentified products (see Figure 2) (7). Adduct 3 was also detected upon solvolysis of R-acetoxyNPIP in the presence of dG (7). The main goal of the present study was to characterize these unidentified products of the reaction of 1 with dG and to determine whether these adducts as well as adduct 3 were present in DNA that had been reacted with either © 1996 American Chemical Society

LC-ESI-MS Detection of DNA Adducts

Figure 1. Adduct formation in the reaction of cis-4-oxo-2pentenal (1) with dG.

Figure 2. Chromatogram obtained upon HPLC analysis (system 1) of the reaction between cis-4-oxo-2-pentenal (1) and dG. Numbers above the peaks refer to structures in Figure 1.

R-acetoxyNPIP or cis-4-oxo-2-pentenal. Since UV detection did not appear to be sensitive enough to detect these adducts in hydrolysates of DNA, we employed liquid chromatography-electrospray ionization mass spectrometry (LC-ESI-MS) and liquid chromatograph-electrospray ionization tandem mass spectrometry (LC-ESI-MS/ MS).

Experimental Section Caution: R-AcetoxyNPIP is a mutagen and chemically activated form of the carcinogen NPIP. It should be handled with extreme care, using appropriate safetywear and ventilation at all times. Apparatus. 1H-NMR spectra were recorded on a Bruker AM 360 WB spectrometer. LC-ESI-MS and LC-ESI-MS/MS were carried out on a Finnigan MAT Model TSQ-700 instrument. Argon was used as the collision gas in the MS/MS experiments. UV spectra were run on a Hewlett Packard Model 8452A diode array spectrophotometer. HPLC was performed with a Waters Associates system (Millipore, Waters Division, Milford, MA) equipped with a Model 991 photodiode array detector and the following columns and solvent elution systems: (1) a 9.4 mm i.d. × 50 cm Partisil 10 ODS-3 Magnum 9 column (Whatman, Clifton NJ) eluted isocratically with 5% CH3CN in H2O for 5 min, then with a gradient from 5% to 25% CH3CN in H2O in 30 min at 4 mL/min; (2) two 4.6 mm × 25 cm B and J OD5 octadecyl columns (Baxter, Edison, NJ) in series eluted isocratically with 5% CH3CN in H2O for 5 min, then with a gradient from 5% to 25% CH3CN in H2O in 30 min at 0.6 or 1 mL/min; (3) a 3.9 mm

Chem. Res. Toxicol., Vol. 9, No. 4, 1996 775 × 25 cm Supelco ODS column (Whatman, Bellefonte, PA) eluted isocratically with 10% CH3CN in H2O for 5 min, then from 10% to 12% CH3CN in H2O in 20 min at 1 mL/min. Chemicals and Enzymes. cis-4-Oxo-2-pentenal (1) and R-acetoxyNPIP were synthesized (7, 9). dG, DNA, DNase I (Type II from bovine pancreas), phosphodiesterase I (Type VII from Crotalus atrox venom), and alkaline phosphatase (Type VII-NTA from bovine intestinal mucosa) were obtained from Sigma Chemical Co. (St. Louis, MO). Reactions. (A) cis-4-Oxo-2-pentenal and dG. dG (40 mg, 0.14 mmol) was dissolved in 5 mL of 0.1 M sodium phosphate buffer (pH 6.8). To this solution was added cis-4-oxo-2-pentenal (24 mg, 0.24 mmol), and the mixture was vortexed and then incubated in a shaking water bath at 37 °C for 7 h. The aqueous layer was extracted twice with 4 mL of CHCl3 and analyzed with HPLC system 1. The material eluting between 22 and 26 min was collected and reanalyzed with HPLC system 3, giving four peaks (2a-d) at retention times 11.5, 12.0, 14.0, and 17.0 min. Spectral data for 2a-d: LC-ESI-MS, m/z (rel intensity) 2a, 388 (M + Na+ , 22), 366 (M + 1, 48), 250 (100), 232 (20); 2b, 388 (M + Na+, 28), 366 (M + 1, 50), 250 (100), 232 (20); 2c, 388 (M + Na+, 20), 366 (M + 1, 43), 250 (100), 232 (17); 2d, 388 (M + Na+, 23), 366 (M + 1, 50), 250 (100). UV (H2O), λmax 248, 275 (sh) nm at pH 7 for all four peaks. The above mixture was concentrated to 2 mL and placed into a 7 mL scintillation vial. It was cooled to 0 °C, NaBH4 (0.5 g) was added, and the mixture was stirred overnight while the temperature gradually rose to room temperature. After neutralization with HCl, the solution was concentrated to 2 mL and analyzed with HPLC system 1. This gave four major peaks at retention times 39.8, 40.4, 40.7, and 41.5 min. Spectral data for 5a-d: LC-ESI-MS, m/z (rel intensity) 5a, 415 (M + 2Na+ + H2O, 100), 352 (M + 1, 77), 142 (72), 128 (54); 5b, 415 (M + 2Na+ + H2O, 100), 352 (M + 1, 95), 164 (48), 142 (53), 114 (62), 100 (78); 5c, 415 (M + 2Na+ + H2O, 100), 352 (M + 1, 93), 142 (18) 128 (67); 5d, 415 (M + 2Na+ + H2O, 63), 352 (M + 1, 48), 146 (100), 128 (50). Each peak corresponding to 5a-d was collected and concentrated to 1 mL. Each solution was added to 1 mL of 0.1 N HCl, heated at 75 °C for 60 min, and neutralized with NaHCO3. The resulting four products, two pairs of enantiomers (6a,b and 6c,d) of 7-(2-hydroxypropyl)-5,6,7,9-tetrahydro-9-oxoimidazo[1,2-a]purine, were purified with HPLC system 2 and dried under vacuum. Spectral data were as follows: 6a (retention time 16 min, 0.6 mg of white solid); UV (H2O), λmax 250, 281 (sh) nm at pH 1.0, 248, 280 (sh) nm at pH 7.0, and 257, 282 nm at pH 13; 1HNMR (DMSO-d6) δ 7.65 (s, 1H, H-2), 7.44 (s, 1H, N5-H), 4.69 (s, 1H, H-7), 4.58 (m, 1H, OH), 1.98 (m, 1H, H-1′), 1.63 (m, 1H, H-1′), 1.10 (d, 3H, J ) 7.2 Hz, H-3′); 1H-NMR (D2O) δ 7.75 (s, 1H, H-2), 3.97 (m, 1H, H-2′), 3.88 (m, 1H, H-6), 3.57 (m, 1H, H-6), 1.92 (m, 2H, H-1′), 1.18 (d, 3H, J ) 7.0 Hz, H-3′); EI-MS m/z (rel intensity) 258 (M + Na+, 100) 236 (M + 1, 78), 146 (55); high resolution FAB-MS, calcd for C10H14N5O2 236.11473, found 236.11520; 6b (retention time 16 min, 0.4 mg of white solid); UV (H2O), λmax 250, 281 (sh) nm at pH 1.0, 248, 280 (sh) nm at pH 7.0, and 257, 282 nm at pH 13; 1H-NMR (DMSO-d6) δ 7.65 (s, 1H, H-2), 7.44 (s, 1H, N5-H), 4.69 (s, 1H, H-7), 4.58 (m, 1H, OH), 1.98 (m, 1H, H-1′), 1.63 (m, 1H, H-1′), 1.10 (d, 3H, J ) 7.2 Hz, H-3′); 1H-NMR (D2O) δ 7.75 (s, 1H, H-2), 3.97 (m, 1H, H-2′), 3.88 (m, 1H, H-6), 3.57 (m, 1H, H-6), 1.92 (m, 2H, H-1′), 1.18 (d, 3H, J ) 7.0 Hz, H-3′); EI-MS m/z (rel intensity) 258 (M + Na+, 100) 236 (M + 1, 78), 146 (55); 6c (retention time 19 min, 0.5 mg of white solid); UV (H2O), λmax 250, 281 (sh) nm at pH 1.0, 248, 280 (sh) nm at pH 7.0, and 257, 282 nm at pH 13; 1H-NMR (DMSO-d6) δ 7.67 (s, 1H, H-2), 7.44 (s, 1H, N5-H), 4.69 (s, 1H, H-7), 4.57 (m, 1H, OH), 1.97 (m, 1H, H-1′), 1.61 (m, 1H, H-1′), 1.12 (d, 3H, J ) 7.0 Hz, H-3′); 1H-NMR (D2O) δ 7.75 (s, 1H, H-2), 3.98 (m, 1H, H-2′), 3.88 (m, 1H, H-6), 3.57 (m, 1H, H-6), 2.05 (m, 1H, H-1′), 1.85 (m, 1H, H-1′), 1.16 (d, 3H, J ) 7.0 Hz, H-3′); EI-MS m/z (rel intensity) 258 (M + Na+, 87), 236 (M + 1, 100), 146 (65); high resolution FAB-MS, calcd for C10H14N5O2 236.11473, found 236.11370; 6d (retention time

776 Chem. Res. Toxicol., Vol. 9, No. 4, 1996 19 min, 0.5 mg of white solid); UV (H2O), λmax 250, 281 (sh) nm at pH 1.0, 248, 280 (sh) nm at pH 7.0, and 257, 282 nm at pH 13; 1H-NMR (DMSO-d6) δ 7.67 (s, 1H, H-2), 7.44 (s, 1H, N5-H), 4.69 (s, 1H, H-7), 4.57 (m, 1H, OH), 1.97 (m, 1H, H-1′), 1.61 (m, 1H, H-1′), 1.12 (d, 3H, J ) 7.0 Hz, H-3′); 1H-NMR (D2O) δ 7.75 (s, 1H, H-2), 3.98 (m, 1H, H-2′), 3.88 (m, 1H, H-6), 3.57 (m, 1H, H-6), 2.05 (m, 1H, H-1′), 1.85 (m, 1H, H-1′), 1.16 (d, 3H, J ) 7.0 Hz, H-3′); EI-MS m/z (rel intensity) 258 (M + Na+, 87), 236 (M + 1, 100), 146 (65). (B) cis-4-Oxo-2-pentenal and DNA. cis-4-Oxo-2-pentenal (33.5 mg, 0.23 mmol) was added to a solution of 10 mg of calf thymus DNA in 2 mL of 0.1 M phosphate buffer (pH 7.0). The mixture was incubated in a shaking water bath at 37 °C for 24 h. After addition of 2 mL of H2O, the mixtures were extracted with 4 mL of CHCl3/isoamyl alcohol (20/1 v/v). The DNA was precipitated by adding 0.4 mL of 5 M NaCl followed by an equal volume of cold ethanol. The DNA was suspended in 3.0 mL of 10 mM Tris-HCl/5 mM MgCl2 (pH 7.0) and digested enzymatically at 37 °C as follows: 10 min with DNase I (3 mg, 8000 U), and 60 min with phosphodiesterase I (0.2 U) and alkaline phosphatase (1000 U). Enzymes were removed by centrifugation using an Amicon Centrifree Micropartition System, 30 000 molecular weight cutoff (W. R. Grace & Co., Amicon Div., Beverly, MA). For analysis of adduct 3 (Figure 1), the hydrolysates were further purified by HPLC system 2. Appropriate fractions were collected and concentrated to dryness. The samples were redissolved in 0.5 mL of H2O for LC-ESI-MS and LC-ESI-MS/ MS analysis using HPLC system 2, with one column. The eluant was split at the end of the column so flow to the MS was 150 µL/min. For the LC-ESI-MS/MS experiment, material eluting after 27 min was allowed to enter the source and was monitored for m/z 348-116. At retention time 31-35 min, m/z 352-116 was monitored. For analysis of adducts 2a-d, the appropriate fractions were collected using HPLC system 2. The collected fractions were concentrated to dryness, redissolved in H2O, and treated with NaBH4 (200 mg). After standing overnight, the mixture was neutralized and then hydrolyzed with 0.1 N HCl for 1 h at 75 °C. The products, 6a-d, were analyzed by LC-ESI-MS using HPLC system 2. (C) r-AcetoxyNPIP and DNA. The reaction was carried out with R-acetoxyNPIP (39.6 mg, 0.23 mmol) and 10 mg of calf thymus DNA as described in section B. Analyses for adducts 2a-d and 3 were also performed as above.

Liu et al.

Figure 3. Chromatogram obtained upon HPLC analysis (system 3) of the material eluting from 22 to 26 min in the chromatogram illustrated in Figure 2. Numbers above the peaks refer to structures in Figure 1.

Figure 4. UV spectra of adducts 2a (- -), 2b (‚‚‚), 2c (s), and 2d (- - -).

Results Initially, we investigated the reaction of cis-4-oxo-2pentenal (1) with dG. This produced the chromatogram illustrated in Figure 2. The peak eluting at 28 min was previously identified as 7-(2-oxopropyl)-1,N2-ethenodG (3). In this study, our goal was to obtain structural information on the peaks eluting from 22 to 26 min. This region was collected and reanalyzed by HPLC, giving the chromatogram illustrated in Figure 3. LC-ESI-MS analysis of this mixture indicated that the four compounds eluting at 11.5-17 min all had molecular weights of 365, since all gave m/z 366 (M + 1) and 388 (M + Na+). This corresponds to addition of one molecule of cis-4-oxo-2pentenal to one molecule of dG, consistent with four diastereomers 2a-d (Figure 1). Their UV spectra were identical, as illustrated in Figure 4. These spectra are similar to those of 1,N2-substituted dG derivatives (1214). Attempts to obtain 1H-NMR data on these peaks were unsuccessful, apparently due to instability and/or interconversion. Therefore, the material eluting from 22 to 26 min in Figure 2 was collected and treated with NaBH4, producing the chromatogram shown in Figure 5. Each of the 4 major peaks eluting at 39.8-41.5 min had similar ESI-

Figure 5. Chromatogram obtained upon HPLC analysis (system 1) of the products of NaBH4 treatment of 2a-d. Numbers above the peaks refer to structures in Figure 1.

MS, with m/z 352 (M + 1) and 415 (M + 2Na+ + H2O). This corresponds to reduction of the Schiff’s base 4 to four diastereomers 5a-d (Figure 1). Each of the 4 peaks was collected, treated with HCl, analyzed, and purified by HPLC. This gave two compounds eluting at 16 min and two compounds eluting at 19 min. The two compounds eluting at 16 min were enantiomers, as were the two eluting at 19 min. All four compounds had m/z 236 (M + 1) and 258 (M + Na+), consistent with hydrolysis of the glycosidic bond of 5a-d yielding 6a-d. The UV spectra of 6a-d were identical; representative spectra

LC-ESI-MS Detection of DNA Adducts

Chem. Res. Toxicol., Vol. 9, No. 4, 1996 777

Figure 6. UV spectra of 6a (see structure in Figure 1) at pH 7.0 (s), 1.0 (- - -), and 13.0 (- -). Figure 8. COSY spectrum of 6a in D2O. Table 1.

1H-NMR

Data for Adducts 6a-d and 8

chemical shifts, ppm (multiplicity)

Figure 7. 1H-NMR spectrum of 6a in D2O; inset, partial spectrum in DMSO.

of 6a are illustrated in Figure 6. These spectra are similar to those reported for 7 (15).

position

6a,ba

6c,da

8b

2 N3H N5H 6 7 8 1′ 2′ 3′ OH

7.75 (s) ND 7.44b (s) 3.57, 3.88 (m) 4.69 (m)b

7.75 (s) ND 7.44b (s) 3.57, 3.88 (m) 4.69 (m)b

7.62 (s) 12.3 (bs) 7.48 (s) 3.30 (t)c 1.90 (qui) 3.90 (t)d

1.92 (m) 3.97 (m) 1.18 (d) 4.58 (m)b

1.85, 2.05 (m) 3.98 (m) 1.16 (d) 4.57 (m)b

a In D O. b In DMSO-d . From ref 13. c Assignment not con2 6 firmed; could be C8-H2. d Assignment not confirmed; could be C6H2.

10 would be 3.57 and 3.88 ppm, which are inconsistent with literature data on the related compound 11, in which the protons at position 7 resonate at 4.05 and 4.23 ppm (15). Further evidence against structure 10 was obtained The 1H-NMR spectra of 6a-d and the related adduct 8 are summarized in Table 1; the spectra of 6a in D2O and DMSO are illustrated in Figure 7. The methine proton at position 7 cannot be seen in the spectrum run in D2O. However, when the spectrum was obtained in DMSO, this proton was evident as a multiplet at 4.69 ppm. The peaks at 4.58 and 7.44 ppm in the spectrum run in DMSO disappeared upon addition of D2O and were assigned to the OH and N5H protons, respectively. Two other structuress9 and 10scould be consistent with the 1H-NMR data. Structure 9 was eliminated by the COSY spectrum of 6a illustrated in Figure 8. This spectrum showed that the multiplet at 4.69 ppm, assigned to H-7, was strongly coupled to the multiplets at 3.57 and 3.88 ppm, corresponding to the two H-6 protons, and to the multiplet at 1.92 ppm, corresponding to the two H-1′ protons. This coupling pattern is not consistent with 9. The 1H-NMR data favor 6a-d over 10 because the chemical shifts of the protons at position 7 in structure

by examining the conversion of 2a-d (Figure 1) to the known adduct 3. Incubation of purified 2a-d at 37 °C for 7 h produced the HPLC trace illustrated in Figure 9; the new peak eluting at 30 min was identified as adduct 3.

778 Chem. Res. Toxicol., Vol. 9, No. 4, 1996

Liu et al.

Figure 9. Chromatogram obtained upon HPLC analysis (system 1) of products formed when adducts 2a-d were heated at 37 °C for 7 h. Adduct 3 of Figure 1 is indicated.

Figure 11. Chromatograms obtained upon LC-ESI-MS analysis (system 2) of an enzymatic hydrolysate of DNA that had been reacted with R-acetoxyNPIP. Selected ion monitoring was carried out at (A) m/z 348, (B) m/z 352, (C) m/z 370, and (D) m/z 374. Peaks corresponding to adduct 3 and THP-dG are indicated.

Figure 10. Chromatogram obtained upon HPLC analysis (system 2) of an enzymatic hydrolysate of DNA that had been reacted with R-acetoxyNPIP. The peak corresponding to adduct 3 of Figure 1 is indicated.

We used LC-ESI-MS to investigate the presence of 7-(2oxopropyl)-1,N2-ethenodG (3) and its precursor adducts 2a-d in DNA reacted with either cis-4-oxo-2-pentenal or R-acetoxyNPIP. DNA was enzymatically hydrolyzed, and the hydrolysates were analyzed initially by HPLC. A chromatogram from the reaction of R-acetoxyNPIP with DNA is illustrated in Figure 10. UV detection at 220 nm (the maximum for adduct 3) was employed. A small peak corresponding in retention time to 3 was observed at 25.9 min. Peaks corresponding to 2a-d (retention time 2225 min, Figure 10) could not be definitively observed by HPLC with UV detection at 220 or 254 nm (data not shown). For analysis of 3 by LC-ESI-MS, material eluting from 23 to 29 min was collected to separate it from the huge excess of unmodified deoxyribonucleosides. Selected ion monitoring at m/z 348 (M + 1) and m/z 370 (M + Na+) produced the chromatograms illustrated in Figure 11A,C. The peak eluting at 32.3 min corresponded in retention time to 3. Selected ion monitoring at m/z 352 (M + 1) and m/z 374 (M + Na+) also confirmed the presence of THP-dG in this DNA (Figure 11B,D). Further confirmation for the presence of 3 was obtained by LC-ESI-MS/ MS of m/z 348, with selected reaction monitoring of m/z 348 f m/z 232 (loss of deoxyribose with proton transfer). The resulting chromatogram is illustrated in Figure 12: the indicated peak had identical retention time to that of a standard, as confirmed by coinjection. These data establish the presence of 3 in DNA reacted with R-acetoxyNPIP. Similar data were obtained upon analysis of DNA that had been reacted with cis-4-oxo-2-pentenal. Levels of adduct 3 in these DNA samples were approximately 50 µmol/mol of dG from R-acetoxyNPIP and 20 µmol/mol of dG from cis-4-oxo-2-pentenal. For analysis of 2a-d, the DNA hydrolysates from reaction with R-acetoxyNPIP or cis-4-oxo-2-pentenal were

Figure 12. Chromatogram obtained upon LC-ESI-MS/MS analysis (system 2) of an enzymatic hydrolysate of DNA that had been reacted with R-acetoxyNPIP. The indicated peaks correspond to adduct 3 (m/z 348-116) and THP-dG (m/z 352116).

analyzed by HPLC and the region corresponding in retention time to 2a-d was collected. These fractions were treated with NaBH4 followed by hydrolysis with HCl, and then analyzed for 6a-d by LC-ESI-MS with selected ion monitoring at m/z 236 (M + 1) and m/z 258 (M + Na+). The results of this analysis for DNA reacted with cis-4-oxo-2-pentenal are illustrated in Figure 13. The retention times of the peaks eluting at approximately 33 min were established by coinjection with standard 6a,c. Similar results were obtained upon analysis of DNA reacted with R-acetoxyNPIP. Approximately 60 µmol/ mol of dG of 6a-d were present in these reaction mixtures.

LC-ESI-MS Detection of DNA Adducts

Chem. Res. Toxicol., Vol. 9, No. 4, 1996 779

Figure 13. Chromatogram obtained upon LC-ESI-MS analysis (system 2) of a hydrolysate of DNA that had been reacted with cis-4-oxo-2-pentenal. The DNA was enzyme hydrolyzed, and the appropriate fraction was purified by HPLC, treated with NaBH4, then hydrolyzed with HCl. Selected ion monitoring was carried out at (A) m/z 236 and (B) m/z 258. Peaks corresponding to adducts 6a,c of Figure 1 are indicated.

Discussion The results of this study demonstrate that the ethenodG adduct 3 as well as its hemiaminal precursors 2a-d are present in DNA reacted with either R-acetoxyNPIP or cis-4-oxo-2-pentenal (1). These results are of potential interest with respect to the carcinogenicity of NPIP. Although the adducts identified in this study form at levels substantially lower than THP-dG, they have potential biological significance since some ethenodG adducts are highly mutagenic and persistent in DNA (16-19). Moreover, the identification of ethenodG adducts in DNA reacted with R-acetoxyNPIP contrasts with the results of studies on the related cyclic nitrosamine R-acetoxyNPYR in which ethenodG adducts have not been detected (5, 8). This may be of interest with respect to the esophageal carcinogenicity of NPIP. Esophageal tumors have not been observed in animals treated with NPYR (1-3). Etheno type adducts are believed to be important in carcinogenesis by vinyl chloride, vinyl carbamate, and related compounds (15, 20-22). The hemiaminal adducts 2a-d are structurally related to the major products 12a,b and 14a,b formed upon the reaction of acrolein or methyl vinyl ketone with dG (13, 14). These products interconvert at room temperature via intermediate 13a,b (see Scheme 1) and are consequently reduced to the saturated analogue such as 8 (see structure in Table 1) upon treatment with NaBH4. This behavior is identical to that observed upon treatment of 2a-d with NaBH4. Adducts 12a,b and 14a,b are formed by Michael addition of N-1 of dG to acrolein or methyl vinyl ketone. In these reactions, N-1 was more reactive than the

exocyclic amino group as a nucleophile in the Michael addition reaction (13, 14). Such Michael additions with dG are sensitive to steric effects; crotonaldehyde, as one example, reacts with dG mainly by Michael addition of the exocyclic amino group (12). The reaction of cis-4-oxo2-pentenal with dG could be initiated by Michael addition at either carbon 2 or 3. All products identified to date result from attack of N-1 of dG at carbon 2, which is the less hindered position. Thus, the reaction of cis-4-oxo2-pentenal with dG follows the general principles observed in previous reactionssfavored addition by N-1 to a sterically accessible carbon of the Michael acceptor double bond (12-14). A stable hemiaminal 7 has been characterized as a product of the reaction of 2-chlorooxirane with dG (15). This compound is a hydrate of 1,N2-ethenodG, but is not readily dehydrated. This behavior contrasts to that of 2a-d which are at least partially converted to adduct 3 under physiological conditions. The more facile conversion of 2a-d than 7 to etheno adducts probably results from the intermediacy of Schiff’s base 4 (Figure 1), which can yield 3 by a proton shift. Schiff’s base formation of this type would not be favored in the case of 7. The sensitivity and specificity of LC-ESI-MS and LCESI-MS/MS were important for the characterization of the adducts in hydrolysates of DNA. The levels of the etheno adducts and their hemiaminal precursors formed in DNA were lowsapproximately 100-fold less than those of THP-dG produced under the same conditions. Conventional methods of adduct detection employed for in vitro reactions with DNA, such as HPLC with UV detection, were not useful in this study. The detection limit for adduct 3 by LC-ESI-MS/MS was approximately 0.3 pmol. Thus, these MS techniques appear to have great promise for detection of low levels of DNA adducts but have not been frequently employed to date for this purpose (23, 24). It will be important to determine whether the adducts characterized in this study are formed in vivo and are persistent in animals treated with NPIP. Formation of these adducts could also potentially occur in vivo upon exposure to 2-methylfuran, which is metabolically activated to cis-4-oxo-2-pentenal (9, 10). Although human exposure to cis-4-oxo-2-pentenal per se has not been documented, it could potentially form in the environment upon photooxidation of toluene (11, 12). Moreover, endogenous formation of 4-oxo-2-pentenal by reaction of glyoxal and acetone or pyruvaldehyde and acetaldehyde seems plausible. The development of technology to assess levels of the relevant DNA adducts in animal tissues would be one way to determine whether uptake or endogenous formation of NPIP or 4-oxo-2-pentenal occurred. MS techniques are likely to have the requisite sensitivity and specificity to address these questions.

Scheme 1

780 Chem. Res. Toxicol., Vol. 9, No. 4, 1996

Acknowledgment. This study was supported by Grant CA-44377 from the National Cancer Institute. This is paper 160 in the series “A Study of Chemical Carcinogenesis”. MS were determined in the American Health Foundation Instrument Facility, supported in part by National Cancer Institute Cancer Center Support Grant CA-17613. We thank Peifeng Hu for his contributions to the MS analyses. High resolution MS were obtained at the Washington University Resource for Biomedical and Bio-organic Mass Spectrometry, St. Louis, MO.

Liu et al.

(12)

(13)

(14)

(15)

References (1) Druckrey, H., Preussmann, R., Ivankovic, S., and Schma¨hl, D. (1967) Organotrope carcinogene Wirkungen bei 65 verschiedenen N-Nitroso-Verbindungen an BD-Ratten (Organspecific carcinogenic effects of 65 various N-nitroso compounds in BD rats). Z. Krebsforsch. 69, 103-201. (2) Lijinsky, W., and Reuber, M. D. (1981) Carcinogenic effect of nitrosopyrrolidine, nitrosopiperidine and nitrosohexamethyleneimine in Fischer rats. Cancer Lett. 12, 99-103. (3) Gray, R., Peto, R., Branton, P., and Grasso, P. (1991) Chronic nitrosamine ingestion in 1040 rodents: the effect of the choice of nitrosamine, the species studied, and the age of starting exposure. Cancer Res. 51, 6470-6491. (4) Hotchkiss, J. H. (1989) Preformed N-nitroso compounds in foods and beverages. Cancer Surv. 8, 295-321. (5) 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 or N2-tetrahydropyranyl adducts. Chem. Res. Toxicol. 8, 607-616. (6) Preussmann, R., and Stewart, B. W. (1984) N-Nitroso carcinogens. In Chemical Carcinogenesis, pp 643-828, American Chemical Society, Washington, DC. (7) Hecht, S. S., Young-Sciame, R., and Chung, F.-L. (1992) Reaction of R-acetoxy-N-nitrosopiperidine with deoxyguanosine: oxygen dependent formation of 4-oxo-2-pentenal and a 1,N2-ethenodeoxyguanosine adduct. Chem. Res. Toxicol. 5, 706-712. (8) Ravindranath, V., Burka, L. T., and Boyd, M. R. (1984) Reactive metabolites from the bioactivation of toxic methylfurans. Science 224, 884-886. (9) Ravindranath, V., and Boyd, M. R. (1985) Metabolic activation of 2-methylfuran by rat microsomal systems. Toxicol. Appl. Pharmacol. 78, 370-376. (10) Shepson, P. B., Edney, E. O., and Corse, E. W. (1984) Ring fragmentation reactions on the photooxidations of toluene and o-xylene. J. Phys. Chem. 88, 4122-4126. (11) O’Brien, R. J., Dumdei, B. E., Hummel, S. V., and Yost, R. A. (1984) Determination of atmospheric degradation products of

(16)

(17)

(18)

(19)

(20)

(21)

(22)

(23) (24)

toluene by tandem mass spectrometry. Anal. Chem. 56, 13291335. Chung, F.-L., and Hecht, S. S. (1983) Formation of cyclic 1,N2adducts by reaction of deoxyguanosine with R-acetoxy-N-nitrosopyrrolidine, 4-(carbethoxynitrosamino)-butanal, or crotonaldehyde. Cancer Res. 43, 1230-1235. 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. Chung, F.-L., Roy, K. R., and Hecht, S. S. (1988) A study of reactions of R,β-unsaturated carbonyl compounds with deoxyguanosine. J. Org. Chem. 53, 14-17. Guengerich, F. P., Persmark, M., and Humphreys, W. G. (1993) Formation of 1,N2- and N2,3-ethenoguanine from 2-halooxiranes: isotopic labeling studies and isolation of a hemiaminal derivative of N2-(2-oxoethyl)guanine. Chem. Res. Toxicol. 6, 635648. Leonard, N. J. (1993) Etheno-bridged nucleotides in structural diagnosis and carcinogenesis. Chemtracts: Biochem. Mol. Biol. 3, 273-297. Swenberg, J. A., Fedtke, N., Ciroussel, F., Barbin, A., and Bartsch, H. (1992) Etheno adducts formed in DNA of vinyl chloride-exposed rats are highly persistent in liver. Carcinogenesis 13, 727-729. Singer, B., Spengler, S. J., Chavez, F., and Kusmierek, J. T. (1987) The vinyl chloride-derived nucleoside, N2,3-ethenoguanine, is a highly efficient mutagen in transcription. Carcinogenesis 8, 745747. Singer, B., Kusmierek, J. T., Folkman, W., Chavez, F., and Dosanjh, M. K. (1991) Evidence for the mutagenic potential of the vinyl chloride induced adduct N2,3-etheno-deoxyguanosine, using a site-directed kinetic assay. Carcinogenesis 12, 745-757. Eberle, G., Barbin, A., Laib, R. J., Ciroussel, F., Thomale, J., Bartsch, H., and Rajewsky, M. F. (1989) 1,N6-Etheno-2′-deoxyadenosine and 3,N4-etheno-2′-deoxycytidine detected by monoclonal antibodies in lung and liver DNA of rats exposed to vinyl chloride. Carcinogenesis 10, 209-212. Guengerich, F. P., and Kim, D. H. (1991) Enzymatic oxidation of ethyl carbamate to vinyl carbamate and its role as an intermediate in the formation of 1,N6-ethenoadenosine. Chem. Res. Toxicol. 4, 413-421. Ribovich, M. L., Miller, J. A., Miller, E. C., and Timmins, L. G. (1982) Labeled 1,N6-ethenoadenosine and 3,N4-ethenocytidine in hepatic RNA of mice given [ethyl-1,2-3H or ethyl-1-14C] ethyl carbamate (urethan). Carcinogenesis 3, 539-546. Blair, I. A. (1993) Applications of mass spectrometry to toxicology. Chem. Res. Toxicol. 6, 741-747. Chaudhary, A. K., Nokubo, M., Oglesby, T. D., Marnett, L. J., and Blair, I. A. (1995) Characterization of endogenous DNA adducts by liquid chromatography/electrospray ionization tandem mass spectrometry. J. Mass Spectrom. 30, 1157-1166.

TX950206R