Structure of DNA Adduct Formed with Aminophenylnorharman, Being

Sep 25, 2002 - Structure of DNA Adduct Formed with Aminophenylnorharman, Being Responsible for the Comutagenic Action of Norharman with Aniline...
0 downloads 0 Views 128KB Size
Download PDF
1288

Chem. Res. Toxicol. 2002, 15, 1288-1294

Structure of DNA Adduct Formed with Aminophenylnorharman, Being Responsible for the Comutagenic Action of Norharman with Aniline Yukari Totsuka,*,† Takeji Takamura-Enya,† Nobuo Kawahara,‡ Rena Nishigaki,† Takashi Sugimura,† and Keiji Wakabayashi† Cancer Prevention Division, National Cancer Center Research Institute, 1-1 Tsukiji 5-chome, Chuo-ku, Tokyo 104-0045, and Division of Pharmacognosy and Phytochemistry, National Institute of Health Sciences, 1-18-1 Kamiyoga, Setagaya-ku, Tokyo,158-0098, Japan Received February 1, 2002

A mutagenic heterocyclic amine (HCA), 9-(4′-aminophenyl)-9H-pyrido[3,4-b]indole (aminophenylnorharman, APNH), is produced in the presence of S9 mix by the reaction of norharman and aniline, both of which are nonmutagenic and abundantly present in our environment. It has been previously reported that APNH-DNA adducts were detected in DNA of Salmonella typhimurium strain incubated with APNH and S9 mix. In the present study, we examined the structures of APNH-DNA adducts using the 32P-postlabeling method and various spectrometry techniques. When the reaction mixture of N-acetoxy-APNH and 2′deoxyguanosine 3′-monophosphate (3′-dGp) was analyzed, three adduct spots (two major and one minor) were observed by 32P-postlabeling under modified-standard conditions. No adduct formation was observed for reaction mixtures of N-acetoxy-APNH with 3′-dAp, 3′-dTp, or 3′dCp. The two major adduct spots (spots 1 and 2) detected by TLC were extracted and subjected to HPLC along with the standards 3′,5′-pdGp-C8-APNH and 5′-pdG-C8-APNH, which were independently chemically synthesized. On the basis of the results of co-chromatography, spots 1 and 2 were identified to be 5′-monophosphate and 3′,5′-diphosphate forms of dG-C8-APNH. When the extract of spot 2 (3′,5′-pdGp-C8-APNH) was further digested with nuclease P1 and phosphodiesterase I, a spot corresponding to spot 1 (5′-pdG-C8-APNH) was newly observed on TLC. From these observations, both of the two major spots were concluded to be dG-C8-APNH. A similar DNA adduct pattern to that apparent in vitro was observed in various organs of F344 rats fed 40 ppm of APNH for 4 weeks. The levels of APNH-DNA adducts were highest in the liver and colon, with RAL values of 1.31 ( 0.26 and 1.32 ( 0.11 adducts/107nucleotides, respectively. Thus, APNH was demonstrated to form DNA adducts primarily at the C-8 position of guanine residues in vitro and in vivo, like other mutagenic and carcinogenic HCAs.

Introduction There are many compounds in our environment showing mutagenicity and/or carcinogenicity, including components of cigarette smoke and cooked foods. Although most of these compounds are present at low levels, they constitute a major risk factor in combination. Since it is impossible for humans to completely avoid exposure to such mutagenic/carcinogenic compounds during daily life, it is important to evaluate potential harmful effects on human cancer development and for this the characteristics of mutagens/carcinogens require elucidation. A class of environmental mutagenic/carcinogenic compounds, heterocyclic amines (HCAs),1 has been identified in a variety of heated protein foods (1-3). Among these, 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) and 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline (MeIQx) are present in various cooked foods at relatively abundant levels, in the range of 0.9-480 ng/g (3-5). It has also been reported that the exocyclic amino * To whom correspondence should be addressed. † Cancer Prevention Division, National Cancer Center Research Institute. ‡ National Institute of Health Sciences.

groups of these compounds are metabolically activated, being converted first into N-hydroxyamino derivatives and then, through esterification, including O-acetylation, to ultimate forms which react with the guanine base of DNA to produce HCAs-deoxyguanosine adducts (1-3, 6-10). The first step is mainly catalyzed by P4501A2 and the second by acetyltransferase-2 or sulfotransferase. The major PhIP- and MeIQx-DNA adducts have been reported to be N2-(2′-deoxyguanosin-8-yl)-2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (dG-C8-PhIP) and N2(2′-deoxyguanosin-8-yl)-2-amino-3,8-dimethylimidazo[4,51 Abbreviations: HCAs, heterocyclic amines; PhIP, 2-amino-1methyl-6-phenylimidazo[4,5-b]pyridine; MeIQx, 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline; dG-C8-PhIP, N2-(2′-deoxyguanosin-8-yl)2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine; dG-C8-MeIQx, N2-(2′-deoxyguanosin-8-yl)-2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline; APNH, 9-(4′-aminophenyl)-9H-pyrido[3,4-b]indole; norharman, 9H-pyrido[3,4-b]indole; PEI, polyethyleneimine; RAL, relative adduct labeling; dG, 2′-deoxyguanosine; 3′-dGp, 2′-deoxyguanosine 3′-monophosphate; 3′,5′-pdGp, 2′-deoxyguanosine 3′,5′-diphosphate; 5′-dGp, 2′deoxyguanosine 5′-monophosphate; dG-C8-APNH, N2-(2′-deoxyguanosin-8-yl)-9-(4′-aminophenyl)-9H-pyrido[3,4-b]indole; amino-3′-methylphenylnorharman, 9-(4′-amino-3′-methylphenyl)-9H-pyrido[3,4-b]indole; nitrophenylnorharman, 9-(4′-nitrophenyl)-9H-pyrido[3,4-b]indole; N-acetoxy-APNH, 9-(N-acetoxy-4′-aminophenyl)-9H-pyrido[3,4b]indole; N-OH-APNH, 9-(4′-hydroxyaminophenyl)-9H-pyrido[3,4-b]indole.

10.1021/tx020007p CCC: $22.00 © 2002 American Chemical Society Published on Web 09/25/2002

Structure of Aminophenylnorharman-DNA Adduct

f]quinoxaline (dG-C8-MeIQx), respectively (8.10), and data on alteration of cancer-related genes such as ras, Apc, β-catenin and p53 in tumors induced by PhIP and MeIQx suggested that formation of dG adducts by HCAs is involved in their carcinogenesis (11-13). Recently, we identified a novel mutagenic HCA, 9-(4′aminophenyl)-9H-pyrido[3,4-b]indole (aminophenylnorharman, APNH), on reaction of 9H-pyrido[3,4-b]indole (norharman) with aniline, neither of which are mutagenic, in the presence of S9 mix (14, 15). In the presence of S9 mix the APNH-DNA adducts yielded in Salmonella typhimurium YG1024 are the same as those with the reaction mixture of norharman and aniline. Therefore, APNH is considered to be responsible for the observed co-mutagenic action. APNH exerts higher mutagenicity in S. typhimurium TA98 and YG1024, detectors of frame shift mutations, than in S. typhimurium TA100 and YG1029, detectors of base-pair-change mutations, in the presence of S9 mix, with mutagenic activities being 187 000 revertants of TA98 and 1 783 000 revertants of YG1024/µg, respectively. The mutagenic activity of this compound is comparable to those of MeIQx and 2-amino6-methyldipyrido[1,2-a:3′,2′-d]imidazole (Glu-P-1) (3, 14). Norharman is produced in the pyrolysis of tryptophan (16) and is reported to be present at levels of 900-8990 ng/cigarette and 2.39-795 ng/g of cooked foods (17). These levels are much higher than mutagenic/carcinogenic HCAs, including PhIP and MeIQx. Similarly, aniline has also been reported to be present in cigarette smoke and some kinds of vegetables (18, 19). Moreover, both of these compounds have been detected in human urine and breast milk samples (20-23). Thus, humans can be assumed to be exposed to norharman and aniline simultaneously in everyday life. Therefore, clarification of genotoxic effects of APNH is important in order to understand how this might impact on human health. In the present study, as part of this aim we elucidate the structure of a major APNH-DNA adduct using the 32Ppostlabeling method and various spectrometry techniques. We also report here the formation of APNH-DNA adducts and their adduct levels in various tissues of rats administered APNH.

Materials and Methods Chemicals. APNH and 9-(4′-nitrophenyl)-9H-pyrido[3,4-b]indole (nitrophenylnorharman) were obtained from the Nard Institute, Osaka, and were more than 99% pure based on HPLC analysis. 2′-Deoxyguanosine 3′,5′-diphosphate (3′,5′-pdGp) was from Pharmacia LKB Biotechnology(Uppsala, Sweden). 2′Deoxyribonucleotide 3′-monophosphate (3′-dAp, 3′-dGp, 3′-dCp, and 3′-dTp) and 2′-deoxyguanosine 5′-monophosphate (5′-pdG) were from Sigma Chemical Co. (St Louis, MO). Micrococcal nuclease and phosphodiesterase II were purchased from Worthington Biochemical Co. (Freehold, NJ). [γ-32P]ATP, T4 polynucleotidekinase, nuclease P1, and phosphodiesterase I were obtained from ICN Biochemicals (Irvine, CA), Takara Shuzo Co. (Kyoto, Japan), Yamasa Shoyu Co. (Choshi, Japan) and Worthington Biochemical Co. (Freehold, NJ), respectively. All other chemicals used were of analytical grade. Spectral Measurements. 1H NMR spectra were recorded with a JEOLGX-R FT-NMR spectrometer using a microprobe. Electrospray ionization mass spectrometry (ESI-MS) was performed using a JEOL JMS-7000Q equipped with HP 1100 HPLC system, and UV absorbance spectra were measured with a PD8020 photodiode array detector (Tosoh, Tokyo, Japan). In Vitro Formation of Deoxyribonucleotide-APNH Adducts. 9-(N-Acetoxy-4′-aminophenyl)-9H-pyrido[3,4-b]indole (N-

Chem. Res. Toxicol., Vol. 15, No. 10, 2002 1289 acetoxy-APNH) was synthesized by the same procedure as for N-acetoxy-2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (Nacetoxy-PhIP) with minor modifications (24). A mixture of nitrophenylnorharman (1 mg, 3.5 µmol) with 5% Pd-C (5 mg) in 2 mL of tetrahydrofuran was stirred under a hydrogen atmosphere for 30 min at 25 °C to yield 9-(4′-hydroxyaminophenyl)-9H-pyrido[3,4-b]indole (N-OH-APNH). After removing Pd-C by filtration, the filtrate was mixed with acetic anhydride (4 µmol), and kept for 15 min at 0 °C to produce N-acetoxy-APNH. Two hundred microliters of the reaction mixture was added to 600 µL of 20 mM Tris-HCl buffer (pH 7.4) containing 0.4 µmol of each 2′-deoxyribonucleotide 3′-monophosphate (3′-dAp, 3′dGp, 3′-dCp, and 3′-dTp), and the mixture was incubated for 30 min at 25 °C. Aliquots were evaporated in vacuo and the residues were dissolved in water and subjected to 32P-postlabeling analysis under modified-standard conditions as described below. Structural Analysis of dG-APNH. N-Acetoxy-APNH was synthesized from nitrophenylnorharman, as described above, and then incubated with 2′-deoxyguanosine (dG, 0.4 µmol). The resulting solution was separated by HPLC as follows. An aliquot of the solution was applied to an analytical-grade TSKgel ODS80Ts column (5 µm particle size, 4.6 × 250 mm; Tosoh, Tokyo, Japan) and material was eluted at a flow rate of 1 mL/min, with a linear gradient of acetonitrile (from 10 to 80%) in 25 mM phosphate buffer (pH 2.0) over 60 min. Although only a few peaks were detected in the solution without N-acetoxy-APNH, several peaks were newly observed in the reaction mixture in its presence. To identify the peak that corresponded to dGAPNH, an aliquot of each peak fraction was analyzed by LCMS. A peak eluted at a retention time of 18.0 min exhibited a molecular ion at m/z 525 and fragment at m/z 409, consistent with the loss of deoxyribose. Therefore, this compound was suggested to be dG-APNH. The peak fraction was therefore collected, and injected into the same TSKgel ODS-80Ts column for further purification. The applied material was eluted with a linear gradient of acetonitrile (from 10 to 50%) in water with 0.25% diethylamine adjusted to pH 6.5 with acetic acid over 60 min at a flow rate of 1 mL/min. All of the above HPLC procedures were performed more than 10 times at ambient temperature with monitoring of the eluate at 254 nm, and the compound was collected for the 1H NMR and MS spectral analyses. 1H NMR spectrum assigmment of an adduct was done with DQF-COSY, NOESY, NOEDIF, and a D2O exchange experiments. The following data were obtained. 1H NMR (DMSO-d ) 10.53 (s, 1H, H ) N1 of dG), 8.96 (s, 1H, H 6 ) NH of APNH) 8.75 (brs, 1H, H ) Ar-H1), 8.46 (brs, 1H, H ) Ar-H3), 8.36 (d, J ) 7.8, 1H, H ) Ar-H5), 8.25 (brs, 1H, H ) Ar-H4), 8.04 (d, J ) 9.0, 2H, H ) Ar-H3′, 5′), 7.59 (t, J ) 7.2, 1H, H ) Ar-H7), 7.55 (d, J ) 9.0, 2H, H ) Ar-H2′, 6′), 7.44 (d, J ) 8.4, 1H, H ) Ar-H8) 7.35 (t, J ) 7.2, 1H, H ) Ar-H6), 6.38, (dd, J ) 9.6, 6.0, 1H, H ) 1′ (dG)), 6.36 (s, 2H, H ) NH2 (dG)), 5.97 (s, 1H, H ) 5′-OH (dG)), 5.32 (s, 1H, H ) 3′-OH (dG)), 4.48 (s,1H, H dH3′ (dG)), 3.96 (s, 1H, H ) H4′ (dG)), 3.80 (s, 2H, H ) H5′ (dG)), 2.60 (m, 1H, H ) H2′ (dG)) 2.08 (m, 1H, H ) H2′′ (dG)); ESI-MS m/z 525.2, 409.0. Preparation of 2′-Deoxyguanosine 3′,5′-Diphosphate (3′,5′-pdGp)-APNH and 2′-Deoxyguanosine 5′-Monophosphate (5′-dGp)-APNH as Standard Compounds. N-Acetoxy-APNH was synthesized as described above and incubated with 2′-deoxyguanosine 3′,5′-diphosphate (3′,5′-pdGp, 0.4 µmol) or 2′-deoxyguanosine 5′-monophosphate (5′-dGp, 0.4 µmol) at 25 °C for 30 min. The resulting solution was evaporated in vacuo, and the residue was dissolved in an appropriate volume of 10% acetonitrile, then purified by HPLC on a TSKgel ODS80Ts column with a gradient of acetonitrile in 25 mM phosphate buffer (pH 2.0) as a mobile phase, these being the same conditions as for dG-APNH separation. When an aliquot of the reaction mixture of N-acetoxy-APNH with either 3′, 5′-pdGp or 5′-dGp was examined, the peak with the same UV absorption pattern as dG-APNH was eluted at a retention time of 13.0 and 15.5 min, respectively. These peak fractions were collected and

1290

Chem. Res. Toxicol., Vol. 15, No. 10, 2002

concentrated. To confirm the adduct structures, the peak fractions containing 3′, 5′-pdGp-APNH or 5′-dGp-APNH were dephosphorylated with 0.2 units of alkaline phosphatase in 10 mM Tris-HCl buffer (pH 7.4) containing 1.0 mM MgCl2, as described previously (24). dG-C8-APNH was clearly detected in both peak fractions, therefore, these compounds were concluded to be dG-C8-APNH 5′-mono- or 3′,5′-diphosphate. In Vivo Formation of APNH-DNA Adducts. Three male F344 rats, purchased from Charles River Japan, Inc. (Atsugi, Japan), were provided with food (CE-2 pellet diet, CLEA Japan) and tap water ad libitum. From 7 weeks of age, they were given a diet containing 40 ppm of APNH for 4 weeks, and then sacrificed under ether anesthesia. The major organs, such as the liver, lung, colon, kidney, pancreas, and testis, were excised and stored at -80 °C until DNA extraction by the standard procedure involving enzymatic digestion of protein and RNA followed by extraction with phenol and chloroform/isoamyl alcohol (24:1, v/v). 32P-Postlabeling Method. DNA obtained from the in vivo experiments was digested with micrococcal nuclease and phosphodiesterase II, and the digest was 32P-postlabeled under modified-standard and adduct-intensification conditions as reported previously (14, 25). Briefly, 2 µL aliquots were removed from 15 µL reaction mixtures after 32P-labeling under standard or adduct intensification conditions, and used for measurement of total nucleotide levels after treatment with apyrase. The remaining 13 µL were brought to pH 6.5 by addition of 1.2 µL of 0.3 M HCl, and then treated with 3.0 µL of 0.13 M sodiumcitrate buffer (pH 6.5) containing 4 µg of nuclease P1 and 0.3 mM ZnCl2 at 37 °C for 10 min. The resulting mixtures were subsequently mixed with 3 µL of 0.5 M Trizma base to adjust the pH to about 9.0 and incubated with 3 µL of an aqueous solution containing 30 milliunits of phosphodiesterase I at 37 °C for 30 min. In the case of PhIP-, MeIQx-, and MeIQ-DNA adducts (26-28), under this condition, most of the undigested di- or oligonucleotides including C8-guanine adducts, derived from micrococcal nuclease and phosphodiesterase II treatment are digested, and adducts can then be detected as single spots of 5′-monophosphate form. For the in vitro experiments, deoxyribonucleotide-APNH adduct formation was analyzed under modified-standard conditions without the DNA digestion. The 32P-postlabeled samples were next applied to a polyethyleneimine (PEI) cellulose TLC sheet (POLYGRAM CEL 300 PEI; Macherey-Nagel, Duren, Germany) and developed with 2.3 M sodium phosphate buffer (pH 6.0). The modified nucleotides were then subjected to two-dimensional TLC using the following solvent systems: 1.38 M lithium formate, 5.1 M urea, pH 3.5 (from bottom to top), 1.0 M lithium chloride, 0.5 M Tris-HCl, 8.5 M urea, pH 8.0 (from left to right) and 1.7 M sodium phosphate buffer, pH 6.0 (from left to right, with a 3.5 cm wick) as previously reported (14, 25). Adducts were detected with a Bio-Image Analyzer (BAS 2000; Fuji Photo Film Co., Tokyo, Japan) after exposing the TLC sheets to Fuji imaging plates. Relative adduct labeling (RAL) was determined by the method of Randerath et al. (29), and the values were calculated as averages of three assays. When necessary, adduct spots on the TLC sheets were cut out and extracted twice with 750 µL of 4 M pyridinium formate (pH 4.5) solution with shaking for 45 min as reported previously (26). The extract was centrifuged at 12 000 rpm for 20 min, the supernatant was passed through a 0.2 µm filter (Nihon Millipore Kogyo, Yonezawa, Japan), and the filtrate was evaporated and dissolved in an appropriate volume of water and analyzed by HPLC as described above.

Results In Vitro Formation of DeoxyribonucleotideAPNH Adducts. An N-acetoxy derivative of APNH was synthesized by reacting N-OH-APNH with acetic anhydride. Aliquots of the reaction mixtures were incubated with each 2′-deoxyribonucleotide 3′-monophosphate (3′-

Totsuka et al.

Figure 1. In vitro formation of deoxyribonucleotide-APNH adducts. Chemically synthesized N-acetoxy-APNH was incubated with 3′-dAp (A), 3′-dGp (B), 3′-dCp (C), or 3′-dTp (D) in 20 mM Tris-HCl buffer (pH 7.4) for 30 min at 25 °C. dG-Adduct formation was analyzed by the 32P-postlabeling method under modified-standard conditions. dG-APNH adducts are indicated by the arrowheads. The imaging plates were exposed for 2 h.

dAp, 3′-dGp, 3′-dCp, and3′-dTp), and their adduct formation was analyzed by the 32P-postlabeling method, under modified-standard conditions. As shown in Figure 1B, three adduct spots, two major (spots 1 and 2) and one minor (spot 3), were detected in the reaction mixture of N-acetoxy-APNH and 3′-dGp, and this deoxyribonucleotide-APNH adduct pattern was similar to that of APNHtreated S. typhimurium YG1024, described previously (14). The levels of three adduct spots were estimated to be 3.10 ( 0.23 (spot 1), 2.20 ( 0.16 (spot 2), and 0.17 ( 0.02 (spot 3) per 104 nucleotides, respectively. In contrast, when N-acetoxy-APNH was incubated with 3′-dAp, 3′dCp or 3′-dTp, no adduct spots were observed (Figure 1A, C, and D). Thus, N-acetoxy-APNH could react with the guanine base to form dG-APNH adducts. Structure of dG-APNH Formed in Vitro. N-Acetoxy-APNH was incubated with dG and the dG-APNH adduct was isolated by HPLC on an ODS column using two different solvent systems. An aliquot of each peak fraction was analyzed by LC-MS to identify the peak corresponding to dG-APNH (Figure 2A). A peak eluting at a retention time of 18.0 min exhibited a molecular ion at m/z 525 and fragment at m/z 409, consistent with the loss of deoxyribose, suggesting it to be dG-APNH. The UV-vis absorption spectrum of the responsible compound showed absorption maxima at 250, 300, and 384 nm (Figure 2B). By repeating the HPLC procedures, around 600 µg of the dG-APNH was able to be collected for measurement of the 1H NMR spectrum. The yield of dG-APNH adduct from nitrophenylnorharman was 0.1%. Figure 3 shows the 1H NMR spectrum for dG-APNH adduct measured in DMSO-d6, revealing the presence of 24 protons, which were assigned by DQF-COSY, NOESY, and D2O exchange experiments. Eleven downfield peaks from 7.3 to 8.8 ppm were assigned to the APNH (H-1, 3-8, 2′, 3′, 5′, 6′) moiety. While ortho-type coupling

Structure of Aminophenylnorharman-DNA Adduct

Figure 2. HPLC profile and UV-vis absorption spectrum of dG-APNH adduct. (A) N-Acetoxy-APNH was synthesized from nitrophenylnorharman and then incubated with dG. An aliquot of the reaction mixture was applied to HPLC on an analytical ODS-80Ts column with a linear gradient of acetonitrile (from 10 to 80%) in 25 mM phosphate buffer (pH 2.0). The UV absorbance of the eluate was monitored at 254 nm. The peak, corresponding to dG-APNH by LC-MS analysis, is indicated by the arrow. (B) The UV absorption spectrum of the compound in the peak fraction on an ODS column was measured with a photodiode array detector. Table 1. Levels of APNH-DNA Adducts in Various Tissues of F344 Rats Fed APNH

a

tissue

adduct levelsa (adducts/107 nucleotides)

colon liver pancreas testis kidney lung

1.32 ( 0.11 1.31 ( 0.26 0.44 ( 0.07 0.22 ( 0.01 0.10 ( 0.07 0.05 ( 0.02

Values are the mean ( SD of three analyses.

between H3 and H4 was not observed, the COSY interaction clearly supported the assignment (see Supporting Information). Proton signals corresponding to the deoxyribose (H-1′, 2′, 2", 3′, 4′, 5′, 5") moiety were also observed from 2.0 to 6.4 ppm. Furthermore, D2O exchangeable protons observed at 5.3 and 5.9 ppm were deduced to be protons of 3′-OH and 5′-OH of deoxyribose each of which showed apparent COSY correations with 3′H and 5′H of deoxyribose, respectively (see Supporting Information). The two protons assigned to the amino group at the N2 position of deoxyguanosine were observed at 6.39 ppm which was also D2O exchangeable. The exocyclic amino group of APNH was found at 8.9 ppm which was confirmed by NOESY correlation to H2′, 6′ protons and a D2O exchange expreriment. N1H of a

Chem. Res. Toxicol., Vol. 15, No. 10, 2002 1291

guanine moiety was also found at 10.5 ppm. However, no signal corresponding to the C8 position of guanine was observed. Moreover, the shift values of the H-2′ and H-2′′ protons of 2′-deoxyribose moiety proved similar to those of unmodified 2′-deoxyguanosine, suggesting that the preferred conformation of this adduct is the anti form (8, 30-35). On the basis of these observations, the structure of the APNH-dG adduct was concluded to be N2-(2′deoxyguanosin-8-yl)-9-(4′-aminophenyl)-9H-pyrido[3,4-b]indole (dG-C8-APNH) (Figure 3). Identification of Structures of Adducts in Major Spots on TLC in in Vitro Experiments. As mentioned above, when the reaction mixture of N-acetoxy-APNH and 3′-dGp was analyzed by the 32P-postlabeling method under modified-standard conditions, three adduct spots were detected. To elucidate the structures of the adducts, we first synthesized standard compounds, 3′,5′-pdGp-C8APNH and 5′-pdG-C8-APNH, by the same procedure as for dG-C8-APNH. Their structures were confirmed by UV absorption patterns, being similar to that of dG-C8APNH. Furthermore, when these synthetic 3′,5′-pdGp or 5′-pdG derivatives were treated with alkalinephosphatase (24), dG-C8-APNH was obtained. Under the HPLC conditions, using an analytical ODS column with a linear gradient of acetonitrile (10-80%) in 25 mM phosphate buffer (pH 2.0) for 60 min at a flow late of 1 mL/min, 3′,5′-pdGp-C8-APNH and 5′-pdG-C8-APNH were eluted at retention times of 13.0 and 15.5 min, respectively. Next, two major adduct spots (spots 1 and 2 in Figure 1B) were cut out and extracted from the TLC plates and then subjected to HPLC. As shown in Figure 4, the radioactivity of spot 1was mainly observed in a peak fraction at retention times of 15.0-16.0 min, which corresponded to that of the standard 5′-pdG-C8-APNH. Similarly, when an extract of adduct spot 2 was analyzed by HPLC, the radioactivity was mainly eluted at the same position as the standard 3′,5′-pdGp-C8-APNH at retention times of 12.0-14.0 min. Furthermore, an aliquot of the extracts derived from spot 2 was evaporated and dissolved in 25 mM sodium citrate buffer, pH 6.5, containing 0.06 mM ZnCl2, and then treated again with nuclease P1 and phosphodiesterase I, as described previously (26). The reaction mixture was subjected to TLC and a spot which corresponded to 5′-pdG-C8-APNH was newly observed (Figure 5). From these findings, the structures of these two major adduct spots observed by the 32P-postlabeling method under modified-standard conditions were identified to be the 5′-monophosphate or 3′,5′-diphosphate form of dG-C8-APNH. In Vivo Formation of dG-C8-APNH. APNH-DNA adduct formation was analyzed in various organs of F344 rats fed APNH at a dose of 40 ppm for 4 weeks. When the DNA samples from livers of APNH-treated rats were analyzed under modified adduct-intensification conditions, two major (spots I and II) and two minor (spots III and IV) adduct spots were observed (Figure 6). This TLC pattern was similar to that in the in vitro reaction except a minor spot (spot IV) was newly observed. Furthermore, the TLC patterns of APNH-DNA adducts in DNA samples from other organs, such as lung, colon, kidney, pancreas, and testis, were the same those of liver DNA samples. Their adduct levels are shown in Table 1. RAL values for liver and colon DNA were higher than for other organs, at 1.31 ( 0.26 and 1.32 ( 0.11 adducts/107

1292

Chem. Res. Toxicol., Vol. 15, No. 10, 2002

Totsuka et al.

Figure 3. 1H NMR spectrum and chemical structure of the dG-APNH adduct. 1H NMR was measured in DMSO-d6. D2O exchangeable protons are marked with an asterisk.

Figure 5. Autoradiograms of 3′,5′-pdGp-C8-APNH with or without nuclease P1 and phosphodiesterase I treatment. Spot 2 corresponding to 3′,5′-pdGp-C8-APNH described in Figure 4 was extracted from TLC sheet and then applied to another TLC sheet without (A) and with (B) nuclease P1 and phosphodiesterase I treatment. The imaging plates were exposed for 12 h. Figure 4. HPLC elution patterns of a mixture containing an extract of the dG-APNH adducts, spot 1 or spot 2, on a TLC plate as described for Figure 1 with spiked two standard compounds: 3′,5′-pdGp-C8-APNH and 5′-pdG-C8-APNH. The radioactivity of each fraction collected at 1-min intervals was measured with a liquid scintillation counter.

nucleotides, respectively. To clarify the structures of adducts produced in vivo, two major APNH-DNA adduct spots (spots I and II), which were detected on the same positions as standard 5′-pdG-C8-APNH and 3′,5′-pdGpC8-APNH on TLC, from liver DNA samples were cut out, extracted and subjected to HPLC. As a result, the radioactivities of spots I and II were detected at the same positions as standard 3′,5′-pdGp-C8-APNH and 5′-pdG-

C8-APNH, respectively (data not shown). Accordingly, the major DNA adduct produced in vivo was concluded to be dG-C8-APNH.

Discussion The present study using the 32P-postlabeling method and various spectrometry techniques revealed that the structures of two major dG-APNH adducts formed in vitro corresponded to standard 5′-pdG-C8-APNH and 3′,5′-pdGp-C8-APNH. Thus, in in vitro experiments, dGC8-APNH proved to be the major APNH-DNA adduct, analogous to the situation with other HCAs such as PhIP,

Structure of Aminophenylnorharman-DNA Adduct

Figure 6. Autoradiogram of APNH-DNA adducts in the liver of F344 rat fed a diet containing 40 ppm APNH for 4 weeks. The imaging plate was exposed for 24 h.

MeIQx, and 2-amino-3,4-dimethylimidazo[4,5-f]quinoline (MeIQ) (8, 10, 24). Under modified adduct intensification conditions in the 32P-postlabeling method, which includes treatment with nuclease P1 and phosphodiesterase I, most of the undigested di- or oligonucleotides including C8-guanine adducts are digested, and adducts can then be detected as single spots of 5′-monophosphate form, as is the case of PhIP-, MeIQx-, and MeIQ-DNA adducts (26-28). However, in the present study, a large amount of 3′,5′-pdGp-C8-APNH remained under the conditions used, suggesting that 3′,5′-pdGp-C8-APNH may be more resistant to treatment with nuclease P1 and phosphodiesterase I than the corresponding forms of other HCAs including PhIP and MeIQx. Besides dG-C8-APNH as the major adduct, a minor adduct spot was also detected with the reaction of N-acetoxy-APNH and 3′-dGp. However, the structure of the adduct involved has yet to be clarified. It has been previously reported that a minor HCA-dG adduct is 5-(deoxyguanosin-N2-yl)-HCA (dG-N2HCA), as demonstrated with MeIQx and 2-amino-3methylimidazo[4,5-f]quinoline (IQ) (8). Therefore, it is very likely that the minor APNH-dG adduct is dG-N2APNH. When APNH was given to F344 rats at a dose of 40 ppm for 4 weeks, the DNA adduct pattern, which resembles that for the in vitro reaction except for the appearance of one minor spot, was detected in all organs, including the liver, colon and lung. The major APNHDNA adduct was confirmed to be dG-C8-APNH. As described above, 3′,5′-pdGp-C8-APNH may be a little resistant to treatment with nuclease P1 and phosphodiesterase I, so it is suggested that the minor spot which was observed with only in the in vivo sample could be derived from the undigested di- or oligonucleotides including dG-C8-APNH. The adduct levels of DNA samples from various tissues were 0.05-1.32 adducts/ 107 nucleotides, with particularly high RAL values being found for liver and colon DNA. Recently, we reported that glutathione S-transferase placental form (GST-P)-positive liver preneoplastic lesions were clearly induced by APNH treatment at a dose of 50 ppm in F344 rats (36). Thus APNH-DNA adducts might be involved in the induction of preneoplastic lesions in the liver of F344 rats. Moreover, APNH has been demonstrated to show testicular toxicity in rats, probably through damage of the Sertoli cell (37), so that investigation of the significance of APNH-DNA adduct formation in this respect is clearly warranted.

Chem. Res. Toxicol., Vol. 15, No. 10, 2002 1293

It has been reported that norharman is mutagenic in the presence of o-toluidine and S9 mix (15, 25, 38, 39). Recently, we described the mutagenic compound produced by norharman and o-toluidine to be 9-(4′-amino3′-methylphenyl)-9H-pyrido[3,4-b]indole (amino-3′-methylphenylnorharman) (40). Amino-3′-methylphenylnorharman gave DNA adducts and induced mutations in S. typhimurium YG1024, like APNH. Although no data are available regarding the structure of the DNA adduct, it is possibly dG-C8-amino-3′-methylphenylnorharman, by analogy with the APNH case. As noted above, both norharman and aniline are widely distributed in our environment, the two compounds also being detected in human urine and breast milk samples. Recently, we have reported that APNH was detected in 24 h urine at a level of 19.6 ( 16.9 ng, from F344 rats administered norharman and aniline (41). In addition, APNH was also detected in the reaction mixture of norharman and aniline in the presence of human microsome fractions. Thus, it is likely that humans are simultaneously exposed to norharman and aniline in daily life, and APNH may be produced in our bodies. To understand the effect of APNH on human health, it is important to evaluate the long-term carcinogenicity of APNH in rodents and its carcinogenesis mechanism. The data obtained in the present study contribute a firm basis for further research in this area.

Acknowledgment. This study was supported by a Grant-in-Aid for Cancer Research from the Ministry of Health, Labor and Welfare, Japan. Supporting Information Available: 1H-1H-DQFCOSY spectrum of dG-APNH adduct. This material is available free of charge via the Internet at http://pubs.acs.org.

References (1) Sugimura, T. (1986) Studies on environmental chemical carcinogenesis in Japan. Science 233, 312-318. (2) Sugimura, T. (1992) Multistep carcinogenesis: a 1992 perspective. Science 258, 603-607. (3) Wakabayashi, K., Nagao, M., Esumi, H., and Sugimura, T. (1992) Food-derived mutagens and carcinogens. Cancer Res. 52, 2092s2098s. (4) Murray, S., Lynch, A. M., Knize, M. G., and Gooderham, N. J. (1993) Quantification of the carcinogens 2-amino-3, 8-dimethylimidazo[4,5-f]quinoxaline, 2-amino-3, 4, 8-trimethylimidazo[4,5f]quinoxaline, and 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine in food using a combined assay based on capillary column gas chromatography negative ion mass spectrometry. J. Chromatogr., Biomed. Appl. 616, 211-219. (5) Sinha, R., Rothman, N., Salmon, C. P., Knize, M. G., Brown, E. D., Swanson, C. A., Rhodes, D., Rossi, S., Felton, J. S., and Levander, O. A. (1998) Heterocyclic aromatic amine content of beef cooked by different methods to varying degrees of doneness and beef gravy made from meat drippings. Food Chem. Toxicol. 36, 279-287. (6) Kato, R., and Yamazoe, Y. (1987) Metabolic activation and covalent binding to nucleic acids of carcinogenic heterocyclic amines from cooked foods and amino acid pyrolysates. Jpn. J. Cancer Res. (Gann) 78, 297-311. (7) Hashimoto, Y., Shudo, K., and Okamoto, T. (1980) Activation of a mutagen, 3-amino-1-methyl-5H-pyrido[4,3-b]indole. Identification of 3-hydroxyamino-1-methyl-5H-pyrido[4,3-b]indole and its reaction with DNA. Biochem. Biophys. Res. Commun. 96, 355362. (8) Turesky, R. J., Rossi, S. C., Welti, D. H., Lay, J. O., Jr., and Kadlubar, F. F. (1992) Characterization of DNA adducts formed in vitro by reaction of N-hydroxy-2-amino-3-methylimidazo[4,5f]quinoline and N-hydroxy-2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline at the C-8 and N2 atoms of guanine. Chem. Res. Toxicol. 5, 479-490.

1294

Chem. Res. Toxicol., Vol. 15, No. 10, 2002

(9) Snyderwine, E. G., Roller, P. P., Adamson, R. H., Sato, S., and Thorgeirsson, S. S. (1988) Reaction of N-hydroxyamine and N-acetoxy derivatives of 2-amino-3-methylimidazo[4,5-f]quinoline with DNA. Synthesis and identification of N-(deoxyguanosin-8yl)-IQ. Carcinogenesis 9, 1061-1065. (10) Frandsen, H., Grivas, S., Anderson, R., Dragsted, L., and Larsen, J. C. (1992) Reaction of the N2-acetoxy derivative of 2-amino-1methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) with 2′-deoxyguanosine and DNA. Synthesis and identification of N2-(2′-deoxyguanosine-8-yl)-PhIP. Carcinogenesis 13, 629-635. (11) Kakiuchi, H., Watanabe, M., Ushijima, T., Toyota, M., Imai, K., Weisburger, J. H., Sugimura, T. and Nagao, M. (1995) Specific 5′-GGGA-3′f 5′-GGA-3′ mutation of the Apc gene in rat colon tumors induced by 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine. Proc. Natl. Acad. Sci. U.S.A. 92, 910-914. (12) Dashwood, R. H., Suzui, M., Sugimura, T., and Nagao, M. (1998) High frequency of β-catenin (Ctnnb1) mutations in the colon tumors induced by two heterocyclic amines in the F344 rat. Cancer Res. 58, 1127-1129. (13) Ushijima, T., Makino, H., Kakiuchi, H., Inoue, R., Sugimura, T., and Nagao, M. (1995) Genetic aiterations in HCA-induced tumors. In Heterocyclic Amines in Cooked Foods: Possible Human Carcinogens (Adamson, R. H., Gustafsson, J. A., Ito, N., Nagao, M., Sugimura, T., Wakabayashi, K., and Yamazoe, Y., Eds.) pp 281291, Princeton Scientific Publishing Co., Princeton, NJ. (14) Totsuka, Y., Hada, N., Matsumoto, K., Kawahara, N., Murakami, Y., Yokoyama, Y., Sugimura, T., and Wakabayashi, K. (1998) Structural determination of a mutagenic aminophenylnorharman produced by the co-mutagen norharman with aniline. Carcinogenesis 19, 1995-2000. (15) Sugimura, T. (1998) A new concept of co-mutagenicity from a phenomenon forgotten for the past two decades: Is it more important than previously expected? Environ. Health Perspect. 106, A522-A523. (16) Sugimura, T., Kawachi, T., Nagao, M., Yahagi, T., Seino, Y., Okamoto, T., Shudo, K., Kosuge, T., Tsuji, K., and Wakabayashi, K. (1977) Mutagenic principle(s) in tryptophan and phenylalanine pyrolysis products. Proc. Jpn. Acad. 53, 58-61. (17) Totsuka, Y., Ushiyama, H., Ishihara, J., Sinha, R., Goto, S., Sugimura, T., and Wakabayashi, K. (1999) Quantification of the co-mutagenic β-carbolines, norharman and harman, in cigarette smoke condensates and cooked foods. Cancer Lett. 143, 139-143. (18) International Agency for Research on Cancer (IARC) (1982). Aniline and aniline hydrochloride. IARC Monogr. Eval. Carcinog. Risk Chem. Hum. 27, 39-61. (19) Luceri, F., Giuseppe, P., Moneti, G., and Dolara, P. (1993) Primary aromatic amines from side-stream cigarette smoke are common contaminants of indoor air. Toxicol. Ind. Health 9, 405-413. (20) Ushiyama H., Oguri A., Totsuka Y., Itoh H., Sugimura T., and Wakabayashi K. (1995). Norharman and harman in human urine. Proc. Japan Acad. 71B, 57-60. (21) Riffelmann M., Muller G., Schmieding W., Popp W., and Norpoth K. (1995). Biomonitoring of urinary aromatic amines and arylamine hemoglobin adducts in exposed workers and nonexposed control persons. Int. Arch. Occup. Environ. Health 68, 36-43. (22) DeBruin, L. S., Pawliszyn, J. B., and Josephy, P. D. (1999). Detection of monocyclic aromatic amines, possible mammary carcinogenesis, in human milk. Chem. Res. Toxicol. 12, 78-82. (23) Bayoumy, K. E., Donahue, J. M., Hecht, S. S., and Hoffmann, D. (1986). Identification and quantitative determination of aniline and toluidines in human urine. Cancer Res. 46, 6064-6067. (24) Nagaoka, H., Wakabayashi, K., Kim, S. B., Kim, I. S., Tanaka, Y., Ochiai, M., Tada, A., Nukaya, H., Sugimura, T., and Nagao, M. (1992) Adduct formation at C-8 of guanine on in vitro reaction of the ultimate form of 2-amino-1-methyl-6-phenylimidazo[4,5b]pyridine with 2′-deoxyguanosine and its phosphate esters. Jpn. J. Cancer Res. 83, 1025-1029. (25) Mori, M., Totsuka, Y., Fukutome, K., Yoshida, T., Sugimura, T., and Wakabayashi, K. (1996) Formation of DNA adducts by the co-mutagen norharman with aromatic amines. Carcinogenesis 17, 1499-1503.

Totsuka et al. (26) Fukutome, K., Ochiai, M., Wakabayashi, K., Watanabe, S., Sugimura, T., and Nagao, M. (1994) Detection of guanine-C82-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine adduct as a single spot on thin-layer chromatography by modification of the32P-postlabeling method. Jpn. J. Cancer. Res. 85, 113-117. (27) Tada, A., Ochiai, M., Wakabayashi, K., Nukaya, H., Sugimura, T., and Nagao, M. (1994) Identification of N-(deoxyguanosin-8yl)-2-amino-3,4-dimethylimidazo[4,5-f]quinoline (dG-C8-MeIQ) as a major adduct formed by MeIQ with nucleotides in vitro with DNA in vivo. Carcinogenesis 15, 1275-1278. (28) Totsuka, Y., Fukutome, K., Takahashi, M., Takahashi, S., Tada, A., Sugimura, T., and Wakabayashi, K. (1996) Presence of N2(deoxyguanisin-8-yl)-2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline (dG-C8-MeIQx) in human urine. Carcinigenesis 17, 10291034. (29) Randerath, E., Agrawal, H. P., Weaver, J. A., Bordelon, C. B., and Randerath, K. (1985) 32P-Postlabeling analysis of DNA adducts persisting for up to 42 weeks in the skin, epidermis and dermis of mice treated topically with 7,12-dimethylbenz[a]anthracene. Carcinogenesis 6, 1117-1126. (30) Stolarski, R., Dudycz, L., and Shugar, D. (1980) NMR studies on the syn-anti dynamic equilibrium in purine nucleosides and nucleotides. Eur. J. Biochem. 108, 111-121. (31) Evans, F. E., and Kaplan, N. O. (1976) 8-alkylaminoadenyl nucleotides as probes of dehydrogenase interactions with nucleotide of different glycosyl conformation. J. Biol. Chem. 251, 67916797. (32) Evans, F. E., and Wright, J. W. (1980) Proton and phosphorous31 nuclear magnetic resonance study on the stabilization of the anti conformation about the glycosyl bond of 8-alkylaminoadenyl nucleotides. Biochemistry 19, 2113-2117. (33) Stolarski, R., Hagberg, C. E., and Shugar, D. (1984) Studies on the dynamic syn-anti equilibrium in purine nucleosides and nucleotides with the aid of 1H and 13C NMR spectroscopy. Eur. J. Biochem. 138, 187-192. (34) Marques, M. M., Mourato, L. L. G., Santos, M. A., and Beland, F. A. (1996) Synthesis, characterization, and conformational analysis of DNA adducts from methylated anilines present in tobacco smoke. Chem. Res. Toxicol. 9, 99-108. (35) Marques, M. M., Mourato, L. L. G., Amorim, M. T., Santos, M. A., Melchior, W. B., and Beland, F. A. (1997) Effect of substitution site upon the oxidation potentials of alkylanilines, the mutagenicities of N-hydroxyalkylanilines, and the conformations of alkylaniline-DNA adducts. Chem. Res. Toxicol. 10, 1266-1274. (36) Kawamori, T., Totsuka, Y., Ishihara, J., Uchiya, N., Sugimura, T., and Wakabayashi, K. (2000) Induction of liver preneoplastic lesions by aminophenylnorharman, formed from norharman and aniline, in F344 rats. Cancer Lett. 163, 157-161. (37) Totsuka, Y., Kawamori, T., Hisada, S., Mitsumori, K., Ishihara, J., Sugimura, T., and Wakabayashi, K. (2001) Testicular toxicity in F344 rats by aminophenylnorharman, formed from norharman and aniline. Toxicol. Appl. Pharmacol. 175, 169-175. (38) Nagao, M., Yahagi, T., Honda, M., Seino, Y., Matsushima, T., and Sugimura, T. (1977) Demonstration of mutagenicity of aniline and o-toluidine by norharman. Proc. Japan Acad. 53B, 34-37. (39) Nagao, M., Yahagi, T., and Sugimura, T. (1978) Differences in effects of norharman with various classes of chemical mutagens and amounts of S-9. Biochem. Biophys. Res. Commun. 83, 373378. (40) Hada, N., Totsuka, Y., Enya, T., Tsurumaki, K., Nakazawa, M., Kawahara, N., Murakami, Y., Yokoyama, Y., Sugimura, T., and Wakabayashi, K. (2001) Structures of mutagens produced by the co-mutagen norharman with o- and m-toluidine isomers. Mutat. Res. 493, 115-126. (41) Totsuka, Y., Kataoka, T., Enya-Takamura, T., Sugimura, T., and Wakabayashi, K. (2002) In vitro and in vivo formation of aminophenylnorharman from norharman and aniline. Mutat. Res. (in press).

TX020007P