Time Course of Hepatic 1-Methylpyrene DNA Adducts in Rats

Sep 13, 2008 - The method was validated with herring sperm DNA reacted with 1-SMP in ... Klaus Albert , Heinz Frank , Albrecht Seidel , and Hansruedi ...
0 downloads 0 Views 1MB Size
Download PDF
Chem. Res. Toxicol. 2008, 21, 2017–2025

2017

Time Course of Hepatic 1-Methylpyrene DNA Adducts in Rats Determined by Isotope Dilution LC-MS/MS and 32P-Postlabeling Bernhard H. Monien,*,† Carolin Mu¨ller,† Wolfram Engst,† Heinz Frank,‡ Albrecht Seidel,‡ and Hansruedi Glatt† Department of Toxicology, German Institute of Human Nutrition (DIfE), 14558 Nuthetal, Germany, and Biochemical Institute for EnVironmental Carcinogens, Prof. Dr. Gernot Grimmer-Foundation, 22927 Grosshansdorf, Germany ReceiVed June 15, 2008

The alkylated polycyclic aromatic hydrocarbon 1-methylpyrene is a carcinogen in rodents and has been detected in various environmental matrices and foodstuffs. It is activated metabolically by benzylic hydroxylation to 1-hydroxymethylpyrene followed by sulfoconjugation to yield electrophilic 1-sulfooxymethylpyrene (1SMP) that is prone to form DNA adducts. An LC-MS/MS method using multiple reaction monitoring (MRM) of fragment ions has been developed for specific detection and quantification of N2-(1-methylpyrenyl)-2′deoxyguanosine (MP-dGuo) and N6-(1-methylpyrenyl)-2′-deoxyadenosine (MP-dAdo) formed in DNA in the presence of 1-SMP. DNA samples were spiked with stable isotope internal standards, [15N5,13C10]MPdGuo and [15N5]MP-dAdo, followed by enzymatic digestion to 2′-deoxynucleosides and solid-phase extraction to remove unmodified 2′-deoxynucleosides prior to analysis by LC-MS/MS. The limits of detection were 10 fmol of MP-dGuo and 2 fmol of MP-dAdo or three molecules of MP-dGuo and 0.6 molecules of MP-dAdo per 108 2′-deoxynucleosides using 100 µg of herring sperm DNA as the sample matrix. The method was validated with herring sperm DNA reacted with 1-SMP in vitro. Hepatic DNA was analyzed from rats that were dosed intraperitoneally with 9.3 mg 1-SMP per kg body weight and killed after various time periods. Levels of MP-dGuo and MP-dAdo in rat liver were found to increase, reaching their maxima at ∼3 h, and then decrease over time. A good correlation was observed between the results obtained using LC-MS/MS and MRM and those from 32Ppostlabeling. MRM allowed the more precise quantification of specific 1-MP adducts, in addition to a time reduction of the analysis when compared with 32P-postlabeling. Introduction Polycyclic aromatic compounds (PAHs)1 are formed as a result of incomplete combustion of organic materials. More than 500 PAHs have been identified in complex mixtures, such as vehicle exhausts (1, 2), tobacco smoke (3), or coal tar (4), of which a substantial part have been shown to be carcinogenic in animals and humans. Heavy exposure to PAHs entails increased risks of lung, skin, and bladder cancer, especially relevant for workers of coal-processing and metal-producing industries (4). The tumorigenic effect is exerted by covalent binding of metabolically activated PAHs to nucleobases of DNA, inducing mutations by disruption of DNA replication. In contrast to the activation of a bay or a fjord region-containing PAHs via the diol epoxide pathway, alkyl-substituted PAHs were hypothesized to be activated by benzylic hydroxylation and subsequent sulfonation, resulting in reactive sulfuric acid esters (5). 1-Methylpyrene (1-MP), a prototype of alkylated PAHs and a common environmental hydrocarbon, exists in cigarette and marijuana smoke (6-9), * To whom correspondence should be addressed. Tel: +49-33200-88332. Fax: +49-33200-88426. E-mail: [email protected]. † German Institute of Human Nutrition (DIfE). ‡ Prof. Dr. Gernot Grimmer-Foundation. 1 Abbreviations: 1-HMP, 1-hydroxymethylpyrene; 1-MP, 1-methylpyrene; 1-SMP, 1-sulfooxymethylpyrene; B[a]P, benzo[a]pyrene; CID, collision-induced dissociation; ESI, electrospray ionization; LOD, limit of detection; MRM, multiple reaction monitoring; MP-dGuo, N2-(1-methylpyrenyl)-2′-deoxyguanosine; MP-dAdo, N6-(1-methylpyrenyl)-2′-deoxyadenosine; PAHs, polycyclic aromatic hydrocarbons; SIR, single ion recording; SPE, solid-phase extraction; SULT, sulfotransferase; UPLC, ultra performance liquid chromatography.

exhaust of diesel engines (10), cellulose pyrolysates (11), and as bioaccumulated pollutants in marine tissues (12). Levels of 1-MP at 4.1-36.2 µg/100 cigarettes were 3-10 times higher as compared to those of benzo[a]pyrene (B[a]P) in eight different brands of cigarettes (6, 7, 13). Approximately equal amounts of 1-MP and B[a]P were found in restaurants and other locations of environmental tobacco smoke (1-MP, 1.3-8 ng/m3; B[a]P, 2.2-13.3 ng/ m3) (8), in samples of smoked cheese (1-MP, 0.04-0.3 µg/kg; B[a]P, 0.08-0.52 µg/kg) (14), and in olive oil (1-MP, 1.3-35 µg/ L; B[a]P, 0.35-93 µg/L) (15). Evidence accumulated that initiation of hepatic tumors in newborn mice after administration of 1-MP (16) originates from metabolic activation by benzylic hydroxylation, followed by sulfotransferase (SULT)-catalyzed sulfonation (Scheme 1). Hydroxylation at the methyl group of 1-MP was observed in rat hepatic homogenates (17) and genetically engineered V79 cell lines expressing human or rodent P450 (18). The resulting 1-hydroxymethylpyrene (1-HMP) was converted into the mutagenic sulfuric acid ester 1-sulfooxymethylpyrene (1-SMP) in the presence of rat liver cytosol (19, 20) and human liver cytosol (21). 1-SMP was shown to initiate tumor growth at the site of subcutaneous injection in rats (22) and in a two-stage mouse model (19). Concurrently, 1-MP adducts of dGuo and dAdo appeared in hepatic DNA, which were chromatographically identical to those observed after incubation of 1-SMP with dAdo and dGuo (19, 23). Adducts of 1-MP were also found in rats that were treated with 1-HMP and 1-SMP by 32P-postlabeling (24, 25). The data support the hypothesis that the ultimate

10.1021/tx800217d CCC: $40.75  2008 American Chemical Society Published on Web 09/13/2008

2018

Chem. Res. Toxicol., Vol. 21, No. 10, 2008

Scheme 1. Metabolic Activation of 1-MP by P450 and SULT

carcinogen of 1-MP metabolism is 1-SMP, which tends to form adducts with nucleophilic residues of proteins and DNA nucleobases (Scheme 1). Detection and quantification of 1-MP adducts in tissue of animals dosed with 1-MP, 1-HMP, or 1-SMP were carried out using 32P-postlabeling (23, 24) and HPLC with fluorescence detection (19). 32P-Postlabeling of nucleotide adducts followed by TLC and quantification of DNA adducts has specific limitations and advantages. On the one hand, its sensitivity and ability to detect many different classes of DNA adducts is attractive. On the other hand, radioactive phosphorus has to be handled. Furthermore, the technique lacks the incorporation of internal standards to determine adduct recoveries and compensate for incomplete labeling efficiencies and provides no structural information about the adduct (26). HPLC coupled with fluorescence detection has been used to measure oxidative DNA lesions (27) or PAH adducts (19, 28), but these detection methods are not applicable to all types of DNA adducts. Internal standards are usually not incorporated into the assay, and structural information on the nature of the DNA adduct is again lacking. Following the advent of electrospray ionization (ESI) LC-MS/MS, recent years showed an increasing number of studies using the technology of multiple reaction monitoring (MRM) to identify and quantify DNA adducts (29). It has become the method of choice for biologically effective dose and for human carcinogenic risk assessments of exposure to chemical carcinogens. Typically, the detection of DNA adducts with LC-MS/MS is about 1 order of magnitude less sensitive as compared to 32P-postlabeling. However, molecule fragmentation allows for thorough characterization of DNA adducts, whereas the TLC pattern of radiolabeled samples often presents a confusing number of spots with questionable identity. In this investigation, we studied 1-MP nucleoside adducts formed in solutions of DNA with 1-SMP and after injection of 1-SMP in rats. The most abundant adducts identified were N2(1-methylpyrenyl)-2′-deoxyguanosine (MP-dGuo) and N6-(1methylpyrenyl)-2′-deoxyadenosine (MP-dAdo). To detect and quantify MP-dGuo and MP-dAdo in DNA samples, an isotope dilution technique was devised. A robust solid-phase extraction (SPE) procedure was established to enrich MP-dAdo and MPdGuo after DNA digestion, and ESI LC-MS/MS was used for both the corroboration and the quantification of these adducts. 32 P-Postlabeling was conducted in parallel to compare the sensitivity of both techniques.

Experimental Procedures Caution: 1-SMP is carcinogenic to rodents and should be handled in a well-Ventilated fume hood with the appropriate protectiVe clothing.

Monien et al. Chemicals. 1-SMP (30) and MP-dGuo (31) were synthesized as described previously. O6-(2,4,6-triisopropylphenylsulfonyl)-3′,5′O-bis(tert-butyldimethylsilyl)-2′-deoxyinosine was prepared according to the method of Gao and Jones (32). Deoxyribonuclease I from bovine pancreas, snake venom phosphodiesterase I (from Crotalus atrox), shrimp alkaline phosphatase (from Pandalus borealis), and micrococcal nuclease (from Staphylococcus aureus) were purchased from Sigma (Steinheim, Germany). Calf spleen phosphodiesterase was from Merck Biosciences (Darmstadt, Germany). Nuclease P1 (from Penicillium citrinum) was purchased from MP Biomedicals (Heidelberg, Germany), and T4 polynucleotide kinase was from MBI Fermentas (St. Leon-Rot, Germany). [15N5]dAdo was from Spectra Stable Isotopes (Columbia, MD), [15N5,13C10]dGuo was from Campro Scientific (Berlin, Germany), and [γ-32P]ATP was from Hartmann Analytik (Braunschweig, Germany). Polyethyleneimine-coated TLC plates Polygram CEL300 PEI (20 cm × 20 cm) were obtained from Macherey-Nagel (Du¨ren, Germany). 1-Pyrenylmethylamine hydrochloride was obtained from Anaspec (San Jose, CA). HPLC-grade methanol and 2-propanol were from Carl Roth GmbH (Karlsruhe, Germany), and HPLCgrade water was obtained from Merck. All other reagents (analytical grade) were from Sigma. Synthesis of MP-dAdo. An aliquot of 1.68 g (2.25 mmol) of O6-(2,4,6-triisopropylphenylsulfonyl)-3′,5′-O-bis(tert-butyldimethylsilyl)-2′-deoxyinosine was dissolved in 20 mL of dry dimethylformamide. After the addition of 1.4 g (6 mmol) of 1-pyrenylmethylamine, the solution was heated at 90 °C for 4 days. The solvent was evaporated, and the crude product was purified on silica gel column eluting with a mixture of acetone and dichloromethane (1: 19, v/v). After evaporation of the solvent, the residue was dissolved in 90 mL of dry THF, and 18 mL of tetrabutylammonium fluoride (1.1 M solution in THF) was added dropwise. After 10 min, the solvent was evaporated, and the residue was purified by flash chromatography on silica gel using a mixture of methanol and trichloromethane (1:10, v/v) as an eluent to provide 0.82 g of MPdAdo in 78% yield. NMR spectra were recorded on a Bruker AMX400 spectrometer. Chemical shifts are reported in ppm downfield from tetramethylsilane. 1H NMR (400 MHz, dimethyl sulfoxide-d6): δ (ppm) 8.81 (d, 1, -NH-CH2-Ar), 8.63 (s, 1, H2purine), 8.55 (d, 1, H10, J9,10 ) 7.6 Hz), 8.45-8.52 (m, 5, H2,3,6,8, H8-purine), 8.38 (s, 2, H4,5), 8.30-8.34 (m, 2, H7,9), 6.60-6.63 (m 1, H1′), 5.71 [s(b), 2, Ar-CH2], 5.57 (d, 1, -OH5′), 5.44-5.47 (m, 1, -OH3′), 4.63-4.68 (m, 1, H3′), 4.11-4.15 (m, 1, H4′), 3.83-3.90 (m, 1, H5′), 3.73-3.79 (m, 1, H5′*), 2.94-3.03 (m, 1, H2′), 2.49-2.54 (m, 1, H2′*). Synthesis of Internal Standards [15N5]MP-dAdo and 15 [ N5,13C10]MP-dGuo. Nine hundred microliters of 9 mM [15N5]dAdo in water was mixed with 100 µL of 50 mM 1-SMP in dimethyl sulfoxide. The reaction mixture was stirred, incubated at 37 °C in the dark for 16 h, and subjected to SPE using an Oasis HLB column (1 cm3, 30 mg, Waters). The column was initially conditioned with 1 mL of methanol followed by 1 mL of water. The reaction mixture was loaded onto the column, and the column bed was washed with 1 mL of water/methanol (95:5). [15N5]MPdAdo was eluted from the column with 1 mL of methanol followed by 1 mL of ethyl acetate. The latter fractions were combined and dried using a centrifugal vacuum evaporator. The [15N5]MP-dAdo was purified with an HPLC system consisting of an Alliance 2695 separations module (Waters) coupled to a 996 PDA detector (Waters) and a tandem quadrupole mass spectrometer VG Quattro II (Micromass, u.k.) using an Ultrasphere column (5 µm, 4.6 mm × 250 mm, Beckman Coulter). The column was eluted with a 15 min gradient starting from 50:50 water/methanol to 0:100 water/ methanol at a flow rate of 1 mL/min. The UV spectra of the eluting peaks were monitored between 210 and 400 nm, and single ion recording (SIR) in the ESI+ mode was used to identify the correct peak with the molecule mass of 471.3 m/z for [15N5]MP-dAdo. The corresponding peaks of multiple runs were collected, pooled, dried, and redissolved in methanol. The identity of [15N5]MP-dAdo was confirmed via MS. The concentration of the solution was determined with analytical HPLC and fluorescence detection (λEXC

Analysis of 1-Methylpyrene Adducts in Hepatic DNA ) 334 nm, λEM ) 390 nm) using an Alliance 2695 (Waters) and a 474 scanning fluorescence detector (Waters). The column NovaPak Silica (5 µm, 3.9 mm × 150 mm, Waters) was eluted with a 15 min gradient starting from 70:30 water/acetonitrile to 0:100 water/acetonitrile at a flow rate of 0.6 mL/min. Both solvents were acidified with 0.25% formic acid and 0.25% acetic acid. An external calibration line of the MP-dAdo standard was used for quantification. [15N5,13C10]MP-dGuo was synthesized using the same procedure and purified via HPLC, and the concentration of the solution was determined by analytical HPLC as described. Reaction of Herring Sperm DNA with 1-SMP. DNA (10 mg) was dissolved in deionized water at a concentration of 1 mg/mL. After the addition of 1 mL of 5 mM 1-SMP in dimethyl sulfoxide, the solution was incubated for 6 h at 37 °C in the dark. The reaction mixture was cooled to 4 °C and washed consecutively with 5 mL each of n-butanol and heptane to remove excess 1-SMP and its decomposition products, such as 1-HMP. The modified DNA was precipitated by addition of 1 mL of cold 3 M sodium acetate (pH 5.2) and 1.5 mL of cold 2-ethoxyethanol. Precipitation was completed for 45 min at -80 °C. The suspension was centrifuged for 30 min at 20, 000g and 4 °C. The supernatant was discarded, and the modified DNA was air-dried. The product was redissolved in 7 mL of H2O. The concentration of the DNA solution was calculated by determining the absorbance at 260 nm using a nanodrop ND-1000 spectrophotometer (peqlab Biotechnologie, Erlangen, Germany). The DNA solution was stored at -80 °C. Animal Dosing with 1-SMP and DNA Isolation. Thirty-two 7 week old male rats (Wistar Han) from Charles River Laboratories (Sulzfeld, Germany) were allowed to habituate for 1 week. Thirty animals received 27.8 µmol/kg 1-SMP dissolved in dimethyl sulfoxide and injected ip. As a control, two rats were injected with dimethyl sulfoxide alone. Rats dosed with 1-SMP were killed after different time periods (0.75, 1.5, 3, 6, 24, and 168 h), five rats at one time. Livers were harvested, and the DNA was isolated by phenol-chloroform extraction (33). Enzymatic Digestion of DNA and SPE of Nucleoside Adducts for LC-MS/MS Analysis. Prior to enzymatic digestion, a sample containing 100 µg of DNA and 250 fmol of [15N5]MPdAdo (or 2.5 pmol of [15N5,13C10]MP-dGuo) was dried. The residue was taken up in 27 µL of 50 mM Tris-HCl (pH 8.5) and 5 mM MgCl2 and incubated with 2 µL of deoxyribonuclease I (4 U/µL) for 3 h at 37 °C. Forty microliters of snake venom phosphodiesterase I (1.25 × 10-3 U/µL) and 2 µL of shrimp alkaline phosphatase (1 U/µL) were added, resuming the incubation at 37 °C for 14 h. The digestion mixture was diluted by the addition of 25 µL of methanol and 700 µL of water and centrifuged at 20000g for 10 min. The supernatant was purified using an Oasis HLB column (1 cm3, 30 mg, Waters), preconditioned with 1 mL of methanol and 1 mL of water. The DNA digest was loaded onto the column and washed with 1 mL of water/methanol (95:5). DNA adducts were eluted with 1 mL of methanol and 1 mL of ethyl acetate. After evaporation of the solvents, the residuals were taken up in 50 µL of methanol. Samples were centrifuged at 20000g for 10 min, and the supernatant was used for mass spectrometric analysis. LC-MS/MS Analysis. Samples of DNA adducts were analyzed by LC followed by mass spectrometric MRM. An Acquity ultra performance liquid chromatography (UPLC) System (Waters) with a UPLC BEH Phenyl column (1.7 µM, 2.1 mm × 100 mm, Waters) was used for sample separation. Samples of 4 µL were injected into a starting gradient of 90:10 water/acetonitrile (both solvents containing 0.25% acetic acid and 0.25% formic acid), eluting DNA adducts with a 4.5 min gradient to 20:80 water/acetonitrile at 0.35 mL/min flow rate. The UPLC was connected to a Quattro Premier XE tandem quadrupole MS (Waters) with an electrospray interface operated in the positive ion mode. The neutral loss of the furanose unit (MP-dAdo 466.3 f 350.1 and MP-dGuo 482.3 f 366.1) was used for quantification (quantifier) with [15N5]MP-dAdo (471.3 f 355.1) and [15N5,13C10]MP-dGuo (497.3 f 376.1) as internal standards, while the fragmentation into nucleoside and the 1-MP cation (215.1 m/z) was used to confirm the identity of the substance

Chem. Res. Toxicol., Vol. 21, No. 10, 2008 2019 (qualifier). The tune parameters for MP-dAdo were as follows (deviating parameters for MP-dGuo are given in parentheses): temperature of the electrospray source, 110 °C; desolvation temperature, 485 °C (MP-dGuo 495 °C); desolvation gas, nitrogen (850 L/h); cone gas, nitrogen (50 L/h); collision gas, argon (indicated cell pressure ∼ 4.6 × 10-3 mbar). For the fragmentation of MP-dAdo, collision energies were 35 and 18 eV for the transitions 466.3 f 215.1 and 466.3 f 350.3, respectively. Optimal fragmentation of MP-dGuo required collision energies of 30 and 10 eV for the transitions 482.3 f 215.1 and 482.3 f 366.3, respectively. The dwell time was set to 100 ms, and the capillary voltage was set to 4 kV (MP-dGuo 3.8 kV). The cone and RF1 lens voltages were 30 (MP-dGuo 25 V) and 0.1 V, respectively. Data acquisition and handling were performed with MassLynx software. 32 P-Postlabeling Analysis. DNA samples (5 µg) were digested with micrococcal nuclease (28.8 mU) and calf spleen phosphodiesterase (0.5 mU) in digestion buffer [16.6 mM sodium succinate, 8.3 mM CaCl2 (pH 6)] for 2 h at 37 °C in a total volume of 4.8 µL. The incubation was extended for 2 h following addition of 28.8 mU micrococcal nuclease and 0.5 mU calf spleen phosphodiesterase. For enrichment of adducts of 3′-phosphomononucleotides, 22.2 mU nuclease P1 was added in 4.8 µL of 125 mM sodium acetate and 40 mM ZnCl2 (pH 5), and the digestion continued for 60 min at 37 °C. The reaction was stopped with 1.9 µL of 0.5 M Tris (pH 11). DNA adducts were 32P-labeled by adding [γ-32P]ATP (50 µCi, ∼7000 Ci/mmol) and 6 U T4 polynucleotide kinase in 2 µL of 100 mM bicine, 50 mM MgCl2, 50 mM DTT, and 5 mM spermidine (pH 9), followed by incubation for 30 min at 37 °C. 32 P-Labeled adducts were resolved by chromatography on polyethyleneimine-cellulose plates using the following solvents: D1, 1 M sodium phosphate buffer (pH 6); D3, 3.5 M lithium formate and 8.5 M urea (pH 3.5); and D4, 0.8 M lithium chloride, 0.5 M Tris, and 8.5 M urea (pH 8). TLC sheets were scanned using an Instant Imager (Canberra Packard, Dreieich, Germany), and adducts were quantified via their specific radioactivity as described by Gupta (34).

Results Nucleoside Adducts of 1-MP in Herring Sperm DNA. The formation of 1-MP adducts in vitro was studied by incubating the reactive metabolite 1-SMP with herring sperm DNA. For analysis of nucleoside adducts, samples were subjected to DNA digestion, SPE, and LC-MS/MS but without addition of internal standards. Using the parent scan mode, all of those nucleoside adducts were identified that lost a fragment of 215.1 m/z, characteristic for the cleavage of the 1-MP cation (Figure 1A). The parent scan suggested that all nucleosides formed 1-MP adducts. Importantly, the nucleoside-specific transitions were also found when single 3′-phosphomononucleotides were incubated with 1-SMP alone but not after digestion and SPE of untreated DNA. For further characterization, the DNA digest was analyzed for the neutral loss of 116, related to the fragmentation of the deoxyribosyl moiety, a typical fragmentation of nucleoside adducts. Only two transitions among the putative adducts in Figure 1A, 482.3 f 215.1 and 466.3 f 215.1, were accompanied by a neutral loss of 116. This indicated that 1-MP adducts of dGuo and dAdo existed (data not shown), whereas dissociation of deoxyribose was not observed for adducts of dCyd and dThd. To verify the incidence of 1-MP adducts of dAdo and dGuo, two standard compounds with 1-MP linked to the exocyclic nitrogens of the pyrimidine ring, MP-dAdo and MP-dGuo, were synthesized (compare Scheme 1). They show the characteristic fragmentation patterns in the MS/MS scan: loss of the cation of 1-MP (466.3/482.3 f 215.1) and the neutral loss of the deoxyribosyl unit (MP-dAdo: 466.3 f 350.1, and MP-dGuo: 482.3 f 366.1) and coelute with the peaks assigned to MP-dAdo (5.12 min) and MP-dGuo (4.64 min).

2020

Chem. Res. Toxicol., Vol. 21, No. 10, 2008

Monien et al.

Figure 1. 1-MP adducts formed upon incubation of herring sperm DNA and 1-SMP. The LC-MS/MS parent scan allowed determination of all of those molecules with daughter ions of 215.1 m/z, which is characteristic for the loss of the 1-MP cation (A). From the parent scan, those with the parent mass of 482.3, 466.3, 442.4, and 457.3 are indicative for 1-MP adducts of dGuo, dAdo, dCyd, and dThd, respectively. The sample was analyzed on a UPLC-MS/MS as described in the Experimental Procedures. Representative 32P-postlabeling TLC profile (B). On the basis of chromatography of incubations of individual 2′-deoxynucleoside-3′-phosphates with 1-SMP, spots have been assigned to 1-MP adducts of dGuo (spots 1 and 2), dAdo (spot 3), and dCyd (spot 4) (23, 24).

The modified DNA was further analyzed by 32P-postlabeling. Chromatographic separation of the radiolabeled adducts allowed distinguishing four main spots (Figure 1B). Spot 1 comigrated with the incubation product of dGuo-3′-phosphate and 1-SMP and was assigned to the 3′,5′-bisphosphate of MP-dGuo (23, 24). A fainter spot closer to the application site (spot 2) was assigned to a second 1-MP adduct of the 3′,5′-bisphosphate of dGuo. Synthesis and Characterization of the Stable Isotope Standards [15N5]MP-dAdo and [15N5,13C10]MP-dGuo. Incubation mixtures containing isotope-labeled 2′-deoxynucleosides and 1-SMP were subjected to SPE to remove unreacted nucleosides, followed by purification of the reaction products via reversed-phase HPLC. Figure 2A shows the chromatogram of the reaction mixture of MP-dAdo at 343 nm. The UV spectra of the peaks were indistinguishable and did not allow identification of the correct product. Monitoring the eluate via SIR at 471.3 m/z ([15N5]MP-dAdo) and 497.3 m/z ([15N5,13C10]MPdGuo) allowed identification of the peaks containing the desired products (Figure 2B). The isotope-labeled nucleoside adducts of 1-MP were further characterized by collision-induced dissociation (CID). Figure 3A shows the characteristic fragmentation scheme of [15N5]MP-dAdo, depicting the bonds cleaved under conditions as described in the Experimental Procedures. The resulting fragmentation mass spectrum recorded in the positive electrospray mode is specific for the nucleoside adduct (Figure 3B). The concentrations of solutions of [15N5]MP-dAdo and [15N5,13C10]MP-dGuo were determined by analytical HPLC and fluorescence detection using standard curves from unlabeled MP-dAdo and MP-dGuo, respectively. Performance of the Isotope Dilution Quantification Using [15N5]MP-dAdo and [15N5,13C10]MP-dGuo. High specificity of our MS/MS-based method was achieved by monitoring

Figure 2. Analysis of products formed upon incubation of [15N5]dAdo with 1-SMP. HPLC chromatogram recorded at 340 nm for the separation of the reaction mixture using a C18 Ultrasphere column (5 µm, 4.6 mm × 250 mm) and a gradient from water/methanol (50:50 at 5 min) to water/methanol (10:90 at 20 min) at 1 mL/min (A). SIR analysis with m/z ) 471.3 was used to identify the correct peak at 18.4 min for collection of [15N5]MP-dAdo (B). The product showed a pyrene characteristic UV spectrum (inset).

of two characteristic transitions of DNA adducts (MP-dAdo: 466.3 f 215.1 and 466.3 f 350.1; MP-dGuo: 482.3 f 215.1 and 482.3 f 366.1) and two transitions for the internal standards ([15N5]MP-dAdo:471.3f215.1and471.3f355.1;[15N5,13C10]MPdGuo 497.3 f 215.1 and 497.3 f 376.1). The transition resulting from the neutral loss of 2′-deoxyfuranose was used as a quantifier signal, and the cleavage of the MP group (215.1

Analysis of 1-Methylpyrene Adducts in Hepatic DNA

Figure 3. Fragmentation pathways of [15N5]MP-dAdo observed by positive ESI-MS/MS CID. Principal fragmentation ions were m/z ) 355.1 (the aglycone of [15N5]MP-dAdo), m/z ) 215.1 (the cation of 1-MP), m/z ) 152.8 (N-methyladenine), and 117.1 (the furanose). The asterisks indicate istope-labeled atoms 15N. CID of [15N5,13C10]MPdGuo and unlabeled DNA adducts yielded corresponding ions (the fragmentation pattern of [15N5,13C10]MP-dGuo is shown in Figure S1 of the Supporting Information).

m/z) was used as a qualifier signal (compare Figure 3). Considering a signal-to-noise ratio of 4, the limits of detection (LODs) for direct injection of standard solutions of MP-dAdo and MP-dGuo were determined to be 0.4 and 4 fmol, respectively. To determine the LODs under conditions of sample preparation, herring sperm DNA (100 µg) was spiked with different amounts of MP-dAdo (MP-dGuo) and a constant 250 fmol of the stable isotope standard [15N5]MP-dAdo (2.5 pmol [15N5,13C10]MP-dGuo) and subjected to the entire analysis procedure. Levels of MP-dAdo (MP-dGuo) were measured. Figure 4 shows the transitional chromatograms of the injection that contained 20 fmol each of MP-dAdo and [15N5]MP-dAdo. Peak areas for the secessions of the 1-MP cation (Figure 4A,C) and for the neutral loss of the sugar moiety (Figure 4B,D) match with great accuracy, confirming that ionization yield and fragmentation performance of both molecules are the same. The LOD for MP-dAdo under conditions of sample preparation was 2 fmol, and the resulting calibration was linear in the range of 2-200 fmol (r2 ) 0.9995). Interestingly, the detection of MPdGuo was less sensitive than that of MP-dAdo, with a LOD of 10 fmol and a linear calibration between 10 and 2000 fmol (r2 ) 0.9996). Using 100 µg of DNA for the analyses, these LODs correspond to 3 and 0.6 molecules of MP-dGuo and MP-dAdo, respectively, per 108 2′-deoxynucleosides. Analysis of MP-dAdo and MP-dGuo in Herring Sperm DNA Reacted with 1-SMP. To validate the MRM technique, DNA was incubated with 1-SMP in vitro and analyzed in the presence of [15N5]MP-dAdo and [15N5,13C10]MP-dGuo as internal standards. In these samples, peaks with retention times as previously observed for synthetic MP-dAdo (5.12 min) and MP-dGuo (4.64 min) were observed that coeluted with the internal standards [15N5]MP-dAdo and [15N5,13C10]MP-dGuo,

Chem. Res. Toxicol., Vol. 21, No. 10, 2008 2021

Figure 4. Chromatograms obtained after coinjection of 20 fmol of MPdAdo and 20 fmol of [15N5]MP-dAdo under conditions of sample preparation. The formation of the 1-MP cation (A and C) and the neutral loss of the furanose unit (B and D) resulted in similar peak areas (in parentheses) for analytes and internal standards. Corresponding results were obtained for MP-dGuo and [15N5,13C10]MP-dGuo (Figure S2 of the Supporting Information).

Table 1. Interday Precision for the Detection of MP-dAdo and MP-dGuo in Samples of Modified DNA Containing Various Adduct Levelsa MP-dAdo dilution factor

adducts/108 nucleosides

1 2 4 10 20 40 100 200 400

1870 ( 140 990 ( 40 460 ( 30 190 ( 20 103 ( 9 56 ( 4 18 ( 2 12 ( 1 6(1

MP-dGuo

variation factorb

adducts/108 nucleosides

bariation factorb

0.90 0.95 0.89 0.92 0.99 1.07 0.87 1.19 1.24

7310 ( 680 4390 ( 360 1950 ( 160 740 ( 30 350 ( 50 180 ( 12 79 ( 9 38 ( 1 24 ( 4

0.94 1.12 1.00 0.94 0.89 0.91 1.01 0.97 1.22

a Interday precision was determined using a solution of 1 mg/mL modified DNA, which was diluted with 1 mg/mL unmodified DNA to yield nine DNA solutions with different adduct levels. Means ( SD of the adduct levels were determined from four independent measurements on different days. b The variation factor of a particular sample vf(i) ) A(i)/At(i) describes the deviation of the measured adduct level A(i) from the calculated value At(i),with the theoretical adduct level At(i) ) 1/n [Σ(A(i)·di)]/di, the dilution factor di, and the overall number of measurements n ) 9.

respectively. Levels of MP-dAdo and MP-dGuo in the modified herring sperm DNA were determined to be 1870 ( 150 and 7310 ( 680 adducts per 108 2′-deoxynucleosides, respectively. Thus, MP-dGuo occurred in a ∼4-fold excess as compared to MP-dAdo. To further validate the MRM of MP-dAdo and MPdGuo quantification, interday precision was determined. The original incubation solution with the highest level of adducts and eight dilutions were subjected to adduct analysis on four separate days. The results are summarized in Table 1. The interday sample variation of adduct levels resulted in an average coefficient of variation (CV) of 10% for MP-dAdo and 9.2% for MP-dGuo. The variation factors listed in Table 1 illustrate

2022

Chem. Res. Toxicol., Vol. 21, No. 10, 2008

Figure 5. LC-MS/MS chromatograms of digested DNA isolated from hepatic tissue of a 1-SMP-treated rat. Fragmentations of MP-dAdo, 466.3 f 215.1 (A) and 466.3 f 350.1 (B), were monitored together with the transitions 471.3 f 215.1 (C) and 471.3 f 355.1 (D) of the isotope-labeled standard [15N5]MP-dAdo (20 fmol/injection). The ratio of peak areas for the transition 466.3 f 350.1 (indicated in parentheses) was used to calculate the MP-dAdo content of the DNA. A minor peak coeluting with the expected signal of MP-dAdo in the chromatogram of the fragmentation 466.3 f 215.1 was observed for DNA from untreated rats, which precluded its use for quantification (Figure S5 of the Supporting Information).

the deviation of measured adduct levels from average adduct levels that were calculated from the mean of all samples. Determination of MP-dAdo and MP-dGuo in Rat Liver. Thirty rats were treated with 27.8 µmol/kg 1-SMP and killed after different intervals to observe the time dependence of DNA adduct formation and repair. Figure 5 shows representative LC-MS/MS chromatograms of a hepatic DNA digest containing a defined amount of 250 fmol [15N5]MP-dAdo in 50 µL of methanol. Peaks corresponding to fragmentations 466.3 f 215.1 and 466.3 f 350.1 were observed at 5.10 min, corresponding to MP-dAdo formed in vivo, that coeluted with the internal standard. The amount of MP-dAdo adducts formed was calculated using the peak area of the stable isotope internal standard. The results for the formation of MP-dAdo and MPdGuo in rat liver over time following dosing with 1-SMP are summarized in Figure 6. The increase in adduct levels from 45 min lead to the maximum level between 3 and 6 h. The highest mean level (133 ( 15 molecules MP-dGuo and 7.9 ( 1.3 molecules MP-dAdo per 108 2′-deoxynucleosides) was observed at 3 h following the 1-SMP dosing. Later, levels of MP-dGuo and MP-dAdo declined, such that the adduct levels were below LODs of three molecules of MP-dGuo and 0.6 molecules of MP-dAdo per 108 2′-deoxynucleosides at 168 h. A similar time course was observed by 32P-postlabeling analysis of hepatic DNA samples followed by two-dimensional TLC. The audioradiographs showed a main spot comigrating with the radiolabeled product of 1-SMP and dGuo-3′-phosphate (spot 1 in Figure 1B), which was chosen for quantification. This spot contained 82 ( 2.7% of the overall radioactivity on the TLC plate (means ( SD determined from six measurements) in samples of herring sperm DNA modified with 1-SMP. Analyses of hepatic DNA revealed a fainter spot closer to the

Monien et al.

Figure 6. Comparison of DNA adducts detected by LC-MS/MS MRM and 32P-postlabeling. Following administration of 1-SMP, animals were killed at 0.75, 1.5, 3, 6, 24, and 168 h. Levels of MP-dGuo (black bars) and MP-dAdo (gray bars) were determined by MRM (A) and compared with levels of MP-dGuo-3′,5′-bisphosphate measured by 32Ppostlabeling (B). Bars represent means ( SD of five rats. At 168 h, adduct levels were below the LODs of both techniques.

Figure 7. Correlation between adduct levels in hepatic DNA of 1-SMPtreated rats determined by MRM and 32P-postlabeling. Linear regression revealed that levels of MP-dGuo determined by LC-MS/MS were about 3.4 times higher as compared to the results of densitometric analyses of spot 1, which was assigned to the 3′,5′-bisphosphate of MP-dGuo (regression line y ) 3.4x + 19.4). Results of seven animals with adduct levels below the LODs in one or both analyses were excluded from the regression analysis.

origin (spot 2 in Figure 1B, presumably a minor 1-MP adduct of deoxyguanosine-3′,5′-bisphosphate), which was detected only between 3 and 6 h after 1-SMP administration. This spot was not included in the densitometric analysis of the TLC plate. The DNA adduct level increased from 45 min to the maximum between 3 and 6 h and declined thereafter, such that DNA adducts could not be detected in rat liver after 168 h (Figure 6). Comparing the results from MRM and 32P-postlabeling showed that DNA adduct levels determined by the MS/MS based technique were ∼3.4 times higher than those from postlabeling (Figure 7).

Discussion As a representative molecule for the class of alkylated PAHs, 1-MP is activated metabolically by hydroxylation and subsequent sulfo conjugation (18, 19, 21). The carcinogenicity of the

Analysis of 1-Methylpyrene Adducts in Hepatic DNA

resulting 1-SMP originates probably from DNA alkylation, primarily at purine residues of dGuo and dAdo (19). In this study, we describe the development of a MRM technique for the detection and quantification of DNA adducts of 1-MP. The technique is based upon enzymatic hydrolysis of isolated DNA to nucleosides, followed by SPE, chromatographic separation, and quantification by UPLC-MS/MS in the presence of stable isotopes of MP-dAdo and MP-dGuo. Nucleotide adducts of 1-MP have been studied previously by HPLC with fluorescence detection (19) and 32P-postlabeling (24), indicating that 3′,5′-bisphosphates of MP-dAdo and MPdGuo were the most abundant adducts (23, 24). Here, we employed LC-MS/MS to identify putative nucleoside adducts formed by incubation of herring sperm DNA and 1-SMP. The parent scan mode allowed distinguishing molecules that lost a fragment of 215.1 m/z, corresponding to the characteristic secession of the cationic 1-MP moiety ([M-H]+ f [1-MP H-]+). Only two among six putative adducts (Figure 1A) also showed the anticipated neutral loss of the deoxyribose ([M-H]+ f [M-H - 116]+), id sunt two molecules with masses 466.3 m/z (MP-dAdo, peak at 5.12 min) and 482.3 m/z (MP-dGuo, peak at 4.64 min). 32P-Postlabeling analyses support this result (Figure 1B). Spot 4 formed upon reaction of 1-SMP with DNA in vitro comigrated with a product from incubation of 1-SMP with dCyd-3′-phosphate, suggesting that a dCyd adduct may exist (24), whose occurrence was not corroborated by LC-MS/ MS experiments. On the basis of the results from digestion and parent scan analysis of modified herring sperm DNA, we synthesized 1-MP adducts of dAdo and dGuo for further studies. Considering the preference of bulky PAHs for reactions with exocyclic nitrogens, MP-dAdo and MP-dGuo were synthesized with 1-MP attached to the exocyclic amino groups N6 of dAdo and N2 of dGuo (25, 31, 35-37). Both adducts showed identical elution profiles and fragmentation patterns to those formed by incubation of 1-SMP with single 3′-phosphomononucleotides or with DNA. For the quantification of MP-dAdo and MP-dGuo, an isotope dilution technique was devised that allows for convenient and fast screening of DNA samples by LC-MS/MS. The internal standards were prepared by incubating [15N5]MP-dAdo and [15N5,13C10]MP-dGuo with 1-SMP and applied as follows: DNA samples were spiked with defined amounts of [15N5]MP-dAdo or [15N5,13C10]MP-dGuo and subjected to enzymatic digestion. Adducts of 1-MP were separated from unmodified 2′-deoxynucleosides by SPE and analyzed by LC-MS/MS MRM. Two specific transitions for each of the protonated adducts were monitored; the neutral loss of the deoxyribose ([M-H]+ f [M-H - 116]+) and the fragmentation into the nucleoside and the cation of 1-MP ([M-H]+ f [1-MP - H-]+). The internal standards adjust the described technique from all daily deviations that may occur. Determination of sample recoveries at the SPE step is redundant, but also variances in chromatography (shifts in retention time, progressive loss of column performance) and MS/MS quantification (matrix effects that affect ionization of analyte molecules) are compensated. Thus, the use of the internal standards facilitated greatly the workup and analysis of DNA samples and, in addition, enhanced the accuracy of the MS/MS quantification as compared to the use of an external standard calibration line. Interestingly, MP-dAdo was detected with a sensitivity about 5-10 times higher as compared to the analysis of MP-dGuo. This was essential for quantification of MP-dAdo, because respective adduct levels were about four times lower in vitro and about 15 times lower in vivo as compared to MP-

Chem. Res. Toxicol., Vol. 21, No. 10, 2008 2023

dGuo. The sensitivity of MP-dAdo detection may originate from increased ionization yield when compared to MP-dGuo. Adduct levels were determined in hepatic DNA of 1-SMPtreated rats to assess the persistence of adducts of 1-MP, a prerequisite for their potential use as biomarkers of exposure (38). MRM and radiolabeling analyses showed similar time courses of adduct levels comprising an increase of adducts to a maximum level at about 3 h after administration of 1-SMP, reflecting a delay in the distribution and reaction of 1-SMP. It appears that following the uptake phase, DNA repair mechanisms removed efficiently 1-MP adducts, whose levels dropped below the LODs after 7 days. A good correlation was observed between adduct levels detected using LC-MS/MS MRM and 32 P-postlabeling (Figure 7). However, the average yield of MPdGuo was 3.4-fold higher than the amount of adductsassigned to MP-dGuo-3′,5′-bisphosphatesobtained by DNA digest and enrichment prior to 32P-postlabeling. Similar effects have been observed comparing LC-MS/MS and radiolabeling techniques for the detection of adducts of B[a]P (39) and 2-amino-1-methyl6-phenylimidazo[4,5-b]pyridine (PhIP) (40). The disparities likely result from varying efficiencies of sample preparation. Our technique employs a different strategy of enzymatic hydrolysis, an effective SPE for the enrichment of nucleoside adducts, and incorporates internal standards to account for the losses of analyte molecules. In contrast, the total adduct yield of 32P-postlabeling may be cut by incomplete digestion, partial labeling, and loss of adduct bisphosphates during TLC (41, 42). We discovered that “overdigestion” of adduct nucleotides contribute to the underestimation by 32P-postlabeling. The level of nucleoside adducts in DNA digests prepared for 32Ppostlabeling was analyzed after different steps of hydrolysis. We found that MP-dGuo-3′-phosphate is not entirely resistant to dephosphorylation in the presence of micrococcal nuclease, spleen phosphodiesterase, and nuclease P1. About 20% of the nucleotide adduct MP-dGuo-3′-phosphate is lost by hydrolysis to yield MP-dGuo. In the future, we will evaluate the influence of incomplete 32P-postlabeling by polynucleotide kinase on the overall efficiency of the assay. Despite the disadvantages of the postlabeling technique, the LOD surpasses those of the MRM method. Using 5 µg of DNA for postlabeling, the LOD is ∼7 adducts/108 nucleosides. This may be lowered by changing chromatographic conditions, which were devised for an optimal separation of adduct spots and not to yield the lowest LOD possible for the detection of 1-MP adducts of dGuo-3′,5′-bisphosphate. MRM analyses of MPdGuo and MP-dAdo have LODs of 10 and 2 fmol per injection, respectively, corresponding to three molecules of MP-dGuo and 0.6 molecules of MP-dAdo per 108 2′-deoxynucleosides, when 100 µg of DNA was used. If the sample amount is limited to 5 µg of DNA, the LODs increase to 60 molecules of MP-dGuo and 12 molecules of MP-dAdo per 108 2′-deoxynucleosides. Hence, radiolabeling analysis of the 1-MP adduct of dGuo-3′phosphate is about nine times more sensitive as compared to quantification of MP-dGuo by MRM. The 1-MP adduct of dAdo-3′-phosphate could not be detected by 32P-postlabeling, hindering the direct comparison with results from MRM. Comparing the sensitive detection of MP-dAdo with quantification of MP-dGuo-3′,5′-bisphosphate, the latter analysis was merely 1.7 times more sensitive than MP-dAdo analysis. Taken together, a highly sensitive MRM method was developed for specific detection and quantification of MP-dGuo and MP-dAdo in DNA samples from in vitro and in vivo sources. The LOD of 32P-postlabeling analysis of dGuo-3′phosphate was ∼9 times and ∼1.7 times lower when compared

2024

Chem. Res. Toxicol., Vol. 21, No. 10, 2008

to the LC-MS/MS quantification of MP-dGuo and MP-dAdo, respectively, presenting a sensitive alternative technique to the MRM method, particularly when sample amounts are limited. When sufficient DNA is available, the advantages of the MRM technique prevail over the limits of sensitivity. The use of the stable isotope standards ensures a quantitative yield from sample preparation and a highly specific detection of adducts. Furthermore, the total time for sample preparation through chromatographic analysis allows for the daily workup of 100 samples, making the method suitable for routine analytical applications. In our workgroup, quantification of 1-MP adducts is currently used as an end point analytical tool for metabolic characterization of mice and rats, for example, to follow pathways of 1-SMP within organisms dosed with 1-MP or 1-HMP. The potential of 1-MP adducts for future use as possible biomarkers of exposure is challenged to some extend by the observation of effective repair, which was faster than observed for other bulky adducts from B[a]P in mouse liver (t1/2 ∼ 21 days) (39), tamoxifen in hepatic tissue of rats (t1/2 ∼ 10-15 days) (43), and 7,12-dimethylbenz[a]anthracene in mouse epidermis (t1/2 ∼ 6 days) (44). However, tissue-specific differences in DNA repair can greatly affect the adduct level, and 1-MP adducts may remain stable and accumulate in other, less repair-proficient tissues. Further studies will be conducted to study the half-life of MP adducts in other tissues and other species and to test whether 1-MP adducts are detectable in human tissue, for example, in leukocytes of smokers, where PAH adducts have been detected (45, 46). Acknowledgment. We thank Christine Gumz, Brigitte Knuth, Martina Scholtyssek, and Elke Thom for their excellent technical assistance. Research described in this article was supported by Phillip Morris Inc. Supporting Information Available: Analyses of MP-dGuo containing samples: Fragmentation pathways of [15N5,13C10]MPdGuo and prominent product ions observed by positive ESIMS/MS CID (Figure S1), validation of MP-dGuo analysis under conditions of sample preparation (Figure S2), LC-MS/MS chromatograms of a hepatic DNA digest from a 1-SMP treated rat (Figure S3), LC-MS/MS chromatograms of a hepatic DNA digest from a rat treated with the vehicle dimethyl sulfoxide, recorded under conditions of MP-dGuo analysis (Figure S4), and LC-MS/MS chromatograms monitored with parameters of MP-dAdo analysis in DNA digests from a rat treated with the vehicle dimethyl sulfoxide (Figure S5). This material is available free of charge via the Internet at http://pubs.acs.org.

References (1) Bostro¨m, C. E., Gerde, P., Hanberg, A., Jernstrom, B., Johansson, C., Kyrklund, T., Rannug, A., Tornqvist, M., Victorin, K., and Westerholm, R. (2002) Cancer risk assessment, indicators, and guidelines for polycyclic aromatic hydrocarbons in the ambient air. EnViron. Health Perspect. 110, 451–488. (2) Han, X., and Naeher, L. P. (2006) A review of traffic-related air pollution exposure assessment studies in the developing world. EnViron. Int. 32, 106–120. (3) Rodgman, A., and Perfetti, T. A. (2006) The composition of cigarette smoke: A catalogue of polycyclic aromatic hydrocarbons. Beitr. Tabakforsch. Int.sContr. Tob. Res. 22, 13–69. (4) Boffetta, P., Jourenkova, N., and Gustavsson, P. (1997) Cancer risk from occupational and environmental exposure to polycyclic aromatic hydrocarbons. Cancer Causes Control 8, 444–472. (5) Watabe, T., Ishizuka, T., Isobe, M., and Ozawa, N. (1982) A 7-hydroxymethyl sulfate ester as an active metabolite of 7,12dimethylbenz[a]anthracene. Science 215, 403–405.

Monien et al. (6) Lee, M. L., Novotny, M., and Bartle, K. D. (1976) Gas chromatography/mass spectrometric and nuclear magnetic resonance spectrometric studies of carcinogenic polynuclear aromatic hydrocarbons in tobacco and marijuana smoke condensates. Anal. Chem. 48, 405–416. (7) Severson, R. F., Snook, M. E., Arrendale, R. F., and Chortyk, O. T. (1976) Gas chromatographic quantitation of polynuclear aromatic hydrocarbons in tobacco smoke. Anal. Chem. 48, 1866–1872. (8) Husgafvel-Pursiainen, K., Sorsa, M., Moller, M., and Benestad, C. (1986) Genotoxicity and polynuclear aromatic hydrocarbon analysis of environmental tobacco smoke samples from restaurants. Mutagenesis 1, 287–292. (9) Bi, X., Sheng, G., Feng, Y., Fu, J., and Xie, J. (2005) Gas- and particulate-phase specific tracer and toxic organic compounds in environmental tobacco smoke. Chemosphere 61, 1512–1522. (10) Jensen, T. E., and Hites, R. A. (1983) Aromatic diesel emissions as a function of engine conditions. Anal. Chem. 55, 594–599. (11) Okamoto, H., and Yoshida, D. R. (1981) Metabolic formation of pyrenequinones as enhancing agents of mutagenicity in Salmonella. Cancer Lett. 11, 215–220. (12) Pancirov, R. J., and Brown, R. A. (1977) Polynuclear aromatic hydrocarbons in marine tissues. EnViron. Sci. Technol. 11, 989–992. (13) Grimmer, G. (1979) Prozesse, bei denen PAH entstehen. Luftqualita¨tskriterien fu¨r Ausgewa¨hlte Polyzyklische Aromatische Kohlenwasserstoffe, pp 54-76, Erich Schmidt-Verlag, Berlin. (14) Guillen, M. D., and Sopelana, P. (2004) Occurrence of polycyclic aromatic hydrocarbons in smoked cheese. J. Dairy Sci. 87, 556–564. (15) Guillen, M. D., Sopelana, P., and Palencia, G. (2004) Polycyclic aromatic hydrocarbons and olive pomace oil. J. Agric. Food Chem. 52, 2123–2132. (16) Rice, J. E., Rivenson, A., Braley, J., and LaVoie, E. J. R. (1987) Methylated derivatives of pyrene and fluorene: evaluation of genotoxicity in the hepatocyte/DNA repair test and tumorigenic activity in newborn mice. J. Toxicol. EnViron. Health 21, 525–532. (17) Rice, J. E., Geddie, N. G., DeFloria, M. C., and LaVoie, E. J. (1988) Structural requirements favoring mutagenic activity among methylated pyrenes in S. typhimurium. In Polynuclear Aromatic Hydrocarbons: A Decade of Progress (Cooke, M., and Dennis, A., Eds.) pp773785, Battelle Press, Columbus, OH. (18) Engst, W., Landsiedel, R., Hermersdo¨rfer, H., Doehmer, J., and Glatt, H. R. (1999) Benzylic hydroxylation of 1-methylpyrene and 1-ethylpyrene by human and rat cytochromes P450 individually expressed in V79 Chinese hamster cells. Carcinogenesis 20, 1777–1785. (19) Surh, Y. J., Blomquist, J. C., Liem, A., and Miller, J. A. (1990) Metabolic activation of 9-hydroxymethyl-10-methylanthracene and 1-hydroxymethylpyrene to electrophilic, mutagenic and tumorigenic sulfuric acid esters by rat hepatic sulfotransferase activity. Carcinogenesis 11, 1451–1460. (20) Glatt, H. R., Henschler, R., Phillips, D. H., Blake, J. W., Steinberg, P., Seidel, A., and Oesch, F. (1990) Sulfotransferase-mediated chlorination of 1-hydroxymethylpyrene to a mutagen capable of penetrating indicator cells. EnViron. Health Perspect. 88, 43–48. (21) Glatt, H., Seidel, A., Harvey, R. G., and Coughtrie, M. W. (1994) Activation of benzylic alcohols to mutagens by human hepatic sulphotransferases. Mutagenesis 9, 553–557. (22) Horn, J., Flesher, J. W., and Lehner, A. F. (1996) 1-Sulfooxymethylpyrene is an electrophilic mutagen and ultimate carcinogen of 1-methyl- and 1-hydroxymethylpyrene. Biochem. Biophys. Res. Commun. 228, 105–109. (23) Monnerjahn, S., Seidel, A., Steinberg, P., Oesch, F., Hinz, M., Stezowsky, J. J., Hewer, A., Phillips, D. H., and Glatt, H. R. (1993) Formation of DNA adducts from 1-hydroxymethylpyrene in liver cells in vivo and in vitro. In Postlabelling Methods for Detection of DNA Adducts (Phillips, D. H., Castegnaro, M., and Bartsch, H., Eds.) pp 189-193, IARC, Lyon, France. (24) Ma, L., Kuhlow, A., and Glatt, H. (2002) Ethanol enhances the activation of 1-hydroxymethylpyrene to DNA adduct-forming species in the rat. Polycyclic Aromat. Compds. 22, 933–946. (25) Ma, L., Landsiedel, R., Seidel, A., and Glatt, H. R. (2000) Detection of mercapturic acids and nucleoside adducts in blood, urine and faeces of rats treated with 1-hydroxymethylpyrene. Polycyclic Aromat. Compds. 21, 135–149. (26) Phillips, D. H., Hewer, A., and Arlt, V. M. (2005) 32P-Postlabeling analysis of DNA adducts. Methods Mol. Biol. 291, 3–12. (27) Nair, J. (1999) Lipid peroxidation-induced etheno-DNA adducts in humans. IARC Sci. Publ. 55–61. (28) Weston, A., and Bowman, E. D. (1991) Fluorescence detection of benzo[a]pyrene-DNA adducts in human lung. Carcinogenesis 12, 1445–1449. (29) Singh, R., and Farmer, P. B. (2006) Liquid chromatography-electrospray ionization-mass spectrometry: The future of DNA adduct detection. Carcinogenesis 27, 178–196. (30) Enders, N., Seidel, A., Monnerjahn, S., and Glatt, H. R. (1993) Synthesis of 11 benzylic sulfate esters, their bacterial mutagenicity

Analysis of 1-Methylpyrene Adducts in Hepatic DNA

(31)

(32) (33) (34) (35) (36)

(37)

(38)

(39)

and its modulation by chloride, bromide and acetate anions. Polycyclic Aromat. Compds. 3, 887s–894s. Steinbrecher, T., Wameling, C., Oesch, F., and Seidel, A. (1993) Activation of the C-2 position of purine by the trifluoromethanesulfonate group: Synthesis of N2-alkylated deoxyguanosines. Angew. Chem. Engl. Ed. 32, 404–406. Gao, X., and Jones, R. A. (1987) Nitrogen-15-labeled deoxynucleosides. Synthesis of [6-15N]- and [1-15N]deoxyadenosines from deoxyadenosine. J. Am. Chem. Soc. 109, 1275–1278. Gupta, R. C. (1984) Nonrandom binding of the carcinogen N-hydroxy2-acetylaminofluorene to repetitive sequences of rat liver DNA in ViVo. Proc. Natl. Acad. Sci. U.S.A. 81, 6943–6947. Gupta, R. C. (1993) 32P-postlabelling analysis of bulky aromatic adducts. IARC Sci. Publ. 11–23. La, D. K., and Swenberg, J. A. (1996) DNA adducts: Biological markers of exposure and potential applications to risk assessment. Mutat. Res. 365, 129–146. Kokontis, J. M., Tsung, S. S., Vaughan-Johnson, J., Lee, H., Harvey, R. G., and Weiss, S. B. R. (1993) Mutation in Escherichia coli and mammalian cells induced by closely spaced 1-methylpyrene-deoxyadenosine adducts in opposite DNA strands. Carcinogenesis 14, 645– 651. Lee, H., Hinz, M., Stezowski, J. J., and Harvey, R. G. (1990) Syntheses of polycyclic aromatic hydrocarbon-nucleoside and oligonucleotide adducts specifically alkylated on the amino functions of deoxyguanosine and deoxyadenosine. Tetrahedron Lett. 31, 6773–6776. Swenberg, J. A., Fryar-Tita, E., Jeong, Y. C., Boysen, G., Starr, T., Walker, V. E., and Albertini, R. J. (2008) Biomarkers in toxicology and risk assessment: Informing critical dose-response relationships. Chem. Res. Toxicol. 21, 253–265. Singh, R., Gaskell, M., Le Pla, R. C., Kaur, B., Azim-Araghi, A., Roach, J., Koukouves, G., Souliotis, V. L., Kyrtopoulos, S. A., and Farmer, P. B. (2006) Detection and quantitation of benzo[a]pyrene-

Chem. Res. Toxicol., Vol. 21, No. 10, 2008 2025

(40)

(41)

(42) (43) (44) (45)

(46)

derived DNA adducts in mouse liver by liquid chromatography-tandem mass spectrometry: Comparison with 32P-postlabeling. Chem. Res. Toxicol. 19, 868–878. Goodenough, A. K., Schut, H. A., and Turesky, R. J. (2007) Novel LC-ESI/MS/MS(n) method for the characterization and quantification of 2′-deoxyguanosine adducts of the dietary carcinogen 2-amino-1methyl-6-phenylimidazo[4,5-b]pyridine by 2 D linear quadrupole ion trap mass spectrometry. Chem. Res. Toxicol. 20, 263–276. Phillips, D. H., and Castegnaro, M. (1999) Standardization and validation of DNA adduct postlabelling methods: Report of interlaboratory trials and production of recommended protocols. Mutagenesis 14, 301–315. Reddy, M. V. (2000) Methods for testing compounds for DNA adduct formation. Regul. Toxicol. Pharmacol. 32, 256–263. Divi, R. L., Osborne, M. R., Hewer, A., Phillips, D. H., and Poirier, M. C. (1999) Tamoxifen-DNA adduct formation in rat liver determined by immunoassay and 32P-postlabeling. Cancer Res. 59, 4829–4833. DiGiovanni, J., Fisher, E. P., and Sawyer, T. W. (1986) Kinetics of formation and disappearance of 7,12-dimethylbenz(a)anthracene:DNA adducts in mouse epidermis. Cancer Res. 46, 4400–4405. Mooney, L. A., Santella, R. M., Covey, L., Jeffrey, A. M., Bigbee, W., Randall, M. C., Cooper, T. B., Ottman, R., Tsai, W. Y., Wazneh, L., et al. (1995) Decline of DNA damage and other biomarkers in peripheral blood following smoking cessation. Cancer Epidemiol. Biomarkers PreV. 4, 627–634. Rothman, N., Poirier, M. C., Baser, M. E., Hansen, J. A., Gentile, C., Bowman, E. D., and Strickland, P. T. (1990) Formation of polycyclic aromatic hydrocarbon-DNA adducts in peripheral white blood cells during consumption of charcoal-broiled beef. Carcinogenesis 11, 1241–1243.

TX800217D