Mass Spectrometric Analysis of Relative Levels of ... - ACS Publications

Metabolic activation of NNK leads to the formation of DNA adducts, which play a critical role in NNK carcinogenesis. Adducts specific to NNK result fr...
0 downloads 0 Views 160KB Size
Download PDF
1048

Chem. Res. Toxicol. 2005, 18, 1048-1055

Mass Spectrometric Analysis of Relative Levels of Pyridyloxobutylation Adducts Formed in the Reaction of DNA with a Chemically Activated Form of the Tobacco-Specific Carcinogen 4-(Methylnitrosamino)-1-(3-pyridyl)-1-butanone Shana J. Sturla,† Jana Scott, Yanbin Lao, Stephen S. Hecht, and Peter W. Villalta* The Cancer Center, University of Minnesota, Mayo Mail Code 806, 420 Delaware Street SE, Minneapolis, Minnesota 55455 Received February 4, 2005

Exposure to the tobacco-related nitrosamines 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) and N′-nitrosonornicotine is carcinogenic to humans. Metabolic activation of NNK leads to the formation of DNA adducts, which play a critical role in NNK carcinogenesis. Adducts specific to NNK result from covalent linkage of a pyridyloxobutyl (POB-1-yl) group to DNA. Furthermore, some such adducts are unstable, releasing the degradation product 4-hydroxy1-(3-pyridyl)-1-butanone (4-HPB). Previous qualitative reports from our laboratory have established the chemical structures of the major POB-1-yl-DNA adducts. In this study, we have quantitated the levels of each of these adducts in vitro, as well as their contribution to the biomarker of DNA pyridyloxobutylation, 4-HPB. Standards for the POB-DNA adducts O6-(POB-1-yl)dGuo, 7-(POB-1-yl)Gua, O2-(POB-1-yl)dThd, and O2-(POB-1-yl)Cyt were synthesized and used to determine standard responses by reverse phase HPLC-electrospray ionizationtandem mass spectrometry (ESI-MS/MS). DNA was incubated with varying amounts of 4-(acetoxymethylnitrosamino)-1-(3-pyridyl)-1-butanone in the presence of an esterase, conditions favorable to the formation of an active pyridyloxobutylating agent. After sequential enzymatic and neutral thermal hydrolysis, isolation, and purification, the pyridyloxobutylated mixture was analyzed by HPLC-ESI-MS/MS to quantify the relative level of each of these four adducts as well as the released 4-HPB. The most abundant product was 4-HPB, which accounted for two-thirds of the analyzed mixture. The highest adduct levels measured were those of bases that result from loss of deoxyribose upon neutral thermal hydrolysis. These adducts, 7-(POB1-yl)Gua and O2-(POB-1-yl)Cyt, comprised an average of 23 and 6% of the analyzed mixture, respectively. O2-(POB-1-yl)dThd and the mutagenic adduct O6-(POB-1-yl)dGuo were detected at the lowest levels, 4 and 2%, respectively. The relative levels of adducts determined in this study provide further insight regarding the chemical reactivity of the activated form of NNK with respect to DNA bases. Furthermore, the analytical standards and mass spectrometric methods used lay the groundwork for establishing a representative array of pyridyloxobutylation adducts as biomarkers of tobacco exposure in further biochemical and in vivo studies.

Introduction Tobacco alkaloids, such as nicotine, can be converted to nitrosamines during the curing and processing of tobacco (1). Two of these nitrosamines, N′-nitrosonornicotine (NNN)1 and 4-(methylnitrosamino)-1-(3-pyridyl)1-butanone (NNK), are present in unburned tobacco and its smoke. The chemistry and biology of NNN and NNK have been comprehensively reviewed (2). These compounds are among the most potent carcinogens in tobacco, and extensive animal studies have demonstrated that NNN and NNK cause a number of cancers (2). Significantly, NNK is a strong pulmonary carcinogen in rats (2). An International Agency for Research on Cancer * To whom correspondence should be addressed. Tel: 612-626-0408. Fax: 612-626-5135. E-mail: [email protected]. † Current address: Department of Medicinal Chemistry, University of Minnesota, Minneapolis, Minnesota, 55455.

(IARC) working group has recently concluded that exposure to NNN and NNK is carcinogenic to humans (3). The conclusions of the IARC working group were supported in part by mechanistic information provided by studies of DNA and hemoglobin adducts resulting from exposure to NNN and NNK (1). DNA adducts can play a significant role in chemical carcinogenesis; persistent 1 Abbreviations: DBU, 1,8-diazabicyclo[5.4.0]undec-7-ene; FT-NMR, Fourier transform nuclear magnetic resonance; 4-HPB, 4-hydroxy-1(3-pyridyl)-1-butanone; ESI, electrospray ionization; EtOAc, ethyl acetate; IARC, International Agency for Research on Cancer; MS/MS, tandem mass spectrometry; NNAL, 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol; NNN, N′-nitrosonornicotine; NNK, 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone; NNKOAc, 4-(acetoxymethylnitrosamino)1-(3-pyridyl)-1-butanone; POM, (3-pyridyl)oxomethyl; POB-1-yl, 4-(3pyridyl)-4-oxobut-1-yl; O2-(POB-1-yl)dThd, O2-[4-(3-pyridyl)-4-oxobut1-yl]dThd; O6-(POB-1-yl)dGuo, O6-[4-(3-pyridyl)-4-oxobut-1-yl]dGuo; 7-(POB-1-yl)Gua, 7-[4-(3-pyridyl)-4-oxobut-1-yl]Gua; O2-(POB-1-yl)Cyt, O2-[4-(3-pyridyl)-4-oxobut-1-yl]Cyt; PDA, photodiode array; SRM, selected reaction monitoring; SPE, solid phase extraction.

10.1021/tx050028u CCC: $30.25 © 2005 American Chemical Society Published on Web 05/17/2005

Pyridyloxobutylation Adduct Levels

Chem. Res. Toxicol., Vol. 18, No. 6, 2005 1049

Scheme 1. DNA Adduct Formation by r-Hydroxylation of NNK and 2′-Hydroxylation of NNN

adducts can result in genetic mutations that may alter normal cellular growth control processes (4-6). NNN and NNK require metabolic activation to alkylate DNA and form adducts. An overview of this relationship is presented in Scheme 1 (2). Cytochrome P450-mediated hydroxylation of NNK at the R-methylene carbon yields intermediate 3, which eventually gives rise to methyl DNA adducts. On the other hand, hydroxylation at the R-methyl carbon produces intermediate 2. Intermediate 2 spontaneously loses formaldehyde, to generate unstable species 4. The intermediate 4 can be hydrolyzed to 4-hydroxy-1-(3-pyridyl)-1-butanone (4-HPB) or act as an alkylating reagent that will deliver a pyridyloxobutyl (POB-1-yl) group to DNA-based nucleophiles. It was established in early experiments that some of these adducts are unstable to neutral thermal hydrolysis and release 4-HPB (7). Under such conditions, these HPBreleasing adducts partition between the generation of 4-HPB and the depurination or depyrimidation to the corresponding POB-1-yl-base adducts. The release of 4-HPB was thus established as a biomarker for tobaccorelated nitrosamine exposure and has been quantified in tissues of experimental animals and humans (2). Other POB-1-yl adducts are stable deoxyribonucleoside adducts. An additional major metabolic pathway for NNK is reduction of the ketone, giving rise to 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL), which in turn can be metabolized to methyl and pyridylhydroxybutyl DNA adducts (8). POB-1-yl adducts, including the stable and 4-HPBreleasing adducts, are specific to tobacco-related carcinogens. Therefore, their presence in human DNA results exclusively from exposure to tobacco, making them

important as carcinogen specific markers (2). To gain mechanistic insight relevant to the source of 4-HPB, we have characterized the stable and unstable POB-1-ylDNA adducts and the corresponding depurination/depyrimidation products. These structures are illustrated in Chart 1 (2, 9-11). Additional NNN- and NNK-derived POB-1-yl-DNA adducts resulting from rearrangement or cyclization of the oxobutyl moiety, as well as reaction with the phosphate backbone of DNA, have also been reported (10, 12). A chemically activated form of NNK, 4-(acetoxymethylnitrosamino)-1-(3-pyridyl)-1-butanone (NNKOAc, 1), has facilitated the study of these adducts by acting as a source of the reactive POB-1-yl group (13). In the presence of an esterase, the acetate group of 1 is hydrolyzed to generate the reactive intermediate 2. In previous studies, analytical standards were not available to quantitate the relative levels of the adducts illustrated in Chart 1. Peterson and co-workers have recently reported a quantitative method for the analysis of O6-[4-(3-pyridyl)-4-oxobut-1-yl]dGuo [O6-(POB-1-yl)dGuo, 6] (14). The technology for quantitation of adducts is important in providing mechanistic insight for the chemistry and biology of NNN and NNK exposure. Furthermore, quantitative analysis capabilities are needed for the implementation of new biomarkers. In this study, we determine the relative levels of four of the major POB1-yl adducts formed in vitro. Measurements were carried out for reactions of varying concentrations of 1 with DNA using HPLC-electrospray ionization (ESI)-tandem mass spectrometry (MS/MS) and synthesized standards of each adduct. Furthermore, we related the levels of these adducts to the levels of the previously established biomarker of tobacco nitrosamine exposure, 4-HPB.

1050

Chem. Res. Toxicol., Vol. 18, No. 6, 2005

Sturla et al.

Chart 1. Structures of POB-1-yl Adducts Formed in DNA upon Exposure to 1 in Vitro (6, 7, 9, and 11) and Corresponding Decomposition Products (8 and 10)

Experimental Procedures Caution: The work described involves the handling of hazardous agents and was therefore conducted in accordance with NIH guidelines for the Laboratory use of Chemical Carcinogens (15). Compound 1 is a carcinogen that should be handled with extreme caution. Apparatus and Assay Conditions. Five HPLC systems were used during the course of the experiment as follows. All solvent gradients were linear. 1. System 1. A Waters Associates (Milford, MA) HPLC equipped with a model 991 photodiode array (PDA) detector was used. A 250 mm × 4.6 mm LUNA 5 µm C18(2) column (Phenomenex, Torrance, CA) was used. The solvent elution system consisted of isocratic elution with 5% CH3OH in 40 mM NH4OAc buffer, pH 6.6, for 10 min, then a gradient from 5 to 35% CH3OH in 60 min, and finally a gradient from 35 to 50% CH3OH in 10 min. The flow rate was 0.5 mL/min, and detection was by UV (254 nm). 2. System 2. This system was the same as system 1, except a SPD-10 A UV detector (Shimadzu Scientific Instruments, Columbia, MD) was used instead of the PDA detector. 3. System 3. The same components as system 1 were used. The solvent elution method consisted of initial conditions of 100% H2O, a gradient to 60% CH3OH over the course of 30 min, and a gradient to 100% CH3OH over the course of the subsequent 5 min. The flow rate was 1.0 mL/min. 4. System 4. A Waters Associates HPLC with a SPD-10 A UV detector operating at 254 nm was used. A 250 mm × 21.2 mm LUNA 5 µm C18(2) column was used. The method involved isocratic elution with 5% CH3OH in 40 mM NH4OAc buffer, pH 6.6, for 10 min, then a gradient from 5 to 35% CH3OH over the course of 20 min, and finally a gradient from 35 to 50% CH3OH in 10 min. The flow rate was 7.0 mL/min. 5. System 5. An Agilent 1100 capillary flow HPLC (Agilent Technologies, Palo Alto, CA) with a 150 mm × 0.5 mm C18 column (Agilent Zorbax SB-C18) with a 5 µm particle size in line with either a Finnigan Quantum Ultra AM or Discovery Max (Thermoelectron, San Jose, CA) triple quadrupole mass spectrometer was used. The solvent elution system consisted of a 15 µL/min gradient from 0% acetonitrile in 40 mM NH4OAc buffer to 25% acetonitrile in 29 min, then a gradient from 25 to 75% acetonitrile in 1 min, followed by isocratic elution of 75% acetonitrile for 5 min. The ESI source was set in the positive ion mode as follows: voltage, 5 kV; current, 50 µA; and heated ion transfer tube, 330 °C. The adducts were measured by MS/MS using the selected reaction monitoring (SRM) mode, and the ion transitions are listed in Table 1. The collision energy for all of the transitions was 13 eV, and the argon collision gas pressure was 1.3 mTorr. The retention times of the adducts

Table 1. Parent and Product Ion Masses and Structures of DNA Adducts (6, 8, 10, and 11) Formed from Reaction of DNA with 1a parent ion M+ or (M + H)+

product ion M+ or (M + H)+

adduct

m/z

m/z

fragment

6 8 d4-8 10 11 4-HPB

415 299 303 259 390 166

299 148 152 148 274 106

O6-(POB-1-yl)Gua POB-1-yl d4-(POB-1-yl) POB-1-yl O2-(POB-1-yl)Thy POM

a The m/z values shown represent the fragments monitored by SRM in the LC/MS/MS method described here. 4-HPB is a hydrolytic byproduct of the decomposition of the unstable adducts. The deuterium-labeled Gua adduct d4-8 was used as an injection standard.

using this system were as follows: 7-[4-(3-pyridyl)-4-oxobut-1yl]Gua [7-(POB-1-yl)Gua, 8], 20.7 min; O2-[4-(3-pyridyl)-4oxobut-1-yl]dThd [O2-(POB-1-yl)dThd, 11], 24.0 min; 6, 27.6 min; and O2-(POB-1-yl)Cyd (10), 27.7 min. LC-ESI-MS/MS for qualitative identification of standards was carried out with a Thermo Finnigan LCQ Deca (Thermo Finnigan LC/MS Division) ion trap mass spectrometer interfaced with a Waters Alliance 2690 HPLC multisolvent delivery system and equipped with an SPD-10 A UV detector (Shimadzu Scientific Instruments). The HPLC column and elution conditions were identical to system 1. The ESI source was set in positive ion mode as follows: voltage, 2 kV; current, 10 µA; and capillary temperature, 270 °C. MS/MS data were acquired with the following parameters: isolation width, 1.5 amu; activation amplitude, 30%; activation Q, 0.25; and activation time, 30 ms. NMR spectra were recorded in either DMSO-d6 or D2O using a Varian Inova (Varian, Inc., Palo Alto, CA) spectrometer operating at 600 or 800 MHz at 25 °C using a triple resonance 3 mm HCN probe. Chemicals and Enzymes. Calf thymus DNA, dCyd, dThd, and all enzymes were obtained from Sigma Aldrich Chemical Co. (Milwaukee, WI) and used as supplied. Previously published methods were used for the synthesis of 4-HPB (13), 1 (13), 8 (10), and 6 (9). d4-Ethyl nicotinate was purchased from Cambridge Isotope Labs (Andover, MA). O2-(POB-1-yl)Cyt. A mixture of 1 (100 mg, 0.33 mmol), porcine liver esterase (170 µL, 50 units), and dCyd (18 mg, 0.067 mmol) was allowed to react in 5.0 mL of 0.1 M phosphate buffer, pH 7, at 37 °C for 1 h. The reaction mixture was washed three times with 5 mL of CHCl3. Neutral thermal hydrolysis of the dCyd adduct was carried out on the aqueous phase by heating

Pyridyloxobutylation Adduct Levels at 100 °C for 1 h. The reaction was repeated four times. The material was subjected to crude purification by system 2, collecting the eluent between 90 and 95 min. This solution was concentrated and further purified by HPLC system 1. The desired compound had a retention time of 82.9 min on this system. The HPLC solvent was removed to yield 170 µg, 0.2% based on 1, of the title compound. 1H NMR (800 MHz, DMSOd6): δ 9.1 (s, 1H), 8.8 (s, 1H), 8.3 (d, J ) 8.0 Hz, 1H), 7.8 (s, 1H), 7.6 (s, 1H), 6.0 (s, 1H), 4.2 (m, 2H), 3.2 (t, J ) 8.0 Hz, 2H), 2.0 (t, J ) 8.0 Hz, 2H). Thymidine-5′-toluenesulfonate Ester (14). dThd (0.30 g, 1.2 mmol) was added to a dry flask, placed under N2, dissolved in pyridine (1.5 mL, anhydrous), and cooled to 0 °C. In a separate dry flask, p-toluenesulfonyl chloride (0.30 g, 1.5 mmol) was dissolved in pyridine (1.5 mL, anhydrous) and the resulting solution was slowly cannula-transferred to the dThd solution. The resulting mixture was stirred for 30 min at 0 °C. After storage in a desiccator at 4 °C for 12 h, a small amount of ice was added. The solution was extracted with CHCl3 (0.5 mL × 3) and rinsed with saturated aqueous sodium bicarbonate. The combined organic fractions were dried with MgSO4, filtered, and concentrated to yield 0.26 g (55%) of the title compound as a white solid; mp 170 °C dec; Rf (20% hexanes/EtOAc) 0.45. IR (KBr pellet, cm-1): 3381, 1716, 1660, 1476, 1360, 1273, 1176, 1190, 1095, 1076, 925, 817. 1H NMR (300 MHz, DMSO-d6): δ 11.35 (s, 1H), 7.81 (d, J ) 8.1 Hz, 2H), 7.49 (d, J ) 7.8 Hz, 2H), 7.40 (s, 1H), 6.16 (dd, J ) 6.6, 6.6 Hz, 1H), 5.46 (s, 1H), 4.27 (dd, J ) 10.8, 3.0 Hz, 1H), 4.23-4.13 (m, 2H), 3.87 (m, 1H), 2.42 (s, 3H), 2.23-1.98 (m, 2H), 1.77 (s, 3H). 13C NMR (75 MHz, DMSO): δ 164.0, 150.7, 145.5, 136.3, 132.4, 130.6, 128.0, 125.3, 110.1, 84.3, 83.5, 70.2, 21.4, 12.3. ESI HRMS calcd for [C17H20N2O7S + Na]+, 419.0889; found, 419.0880. O2-(POB-1-yl)dThd. Tosylate 14 (61 mg, 0.15 mmol) was added to a dry flask and placed under N2. 4-HPB (0.5 g, 3.0 mmol), followed by 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) (150 µL), was added by syringe. The reaction mixture was stirred for 17 h at room temperature. The crude reaction mixture was passed through a 35 cm3 (10 g) C18 Sep-Pak solid phase extraction (SPE) cartridge (Waters Corp.) eluting with H2O (40 mL), 5% MeOH (20 mL), 10% MeOH (20 mL), 50% MeOH (40 mL), and MeOH (40 mL). Fractions (20 mL) were collected; fraction 5 was concentrated and further purified by HPLC system 4, collecting the eluent between 45 and 50 min. The solution was concentrated and further purified by HPLC system 1; retention time, 87.0 min. The solvent was removed to yield 234 µg, 0.4% of the title compound. MS and 1H NMR analysis matched previously reported data (11). d4-4-HPB. γ-Butyrolactone (0.5 mL, 6.6 mmol), NaH (0.27 g, 6.6 mmol), and d4-ethyl nicotinate (1.0 g, 6.5 mmol) were allowed to react and purified as previously reported for the preparation of 4-HPB to yield 478 mg (44%) of the title compound. 1H NMR analysis matched previously reported data for 4-HPB 4-toluenesulfonate ester but lacked the pyridyl proton signals (13). 4-Hydroxy-1-(3-d4-pyridyl)-1-butanone 4-Toluenesulfonate Ester (13). d4-4-HPB (176 mg, 1.1 mmol), p-toluenesulfonyl chloride (305 mg, 1.6 mmol), and anhydrous pyridine (1.5 mL) were allowed to react and subjected to workup, as previously reported for the preparation of 4-hydroxy-1-(3pyridyl)-1-butanone 4-toluenesulfonate ester (16). Crude material was chromatographed on silica gel using a stepwise gradient of CH2Cl2, 50% CH2Cl2 in EtOAc, and 30% CH2Cl2 in EtOAc to yield 741 mg (95%) of the title compound. 1H NMR analysis matched previously reported data for 4-HPB 4-toluenesulfonate ester but lacked the pyridyl proton signals (16). 7-[4-(3-d4-Pyridyl)-4-oxobut-1-yl]Gua (d4-8). Compound 13 (380 mg, 1.2 mmol) and 2′,3′,5′-triacetylguanosine (1.1 g, 2.7 mmol) (12) were allowed to react according to the previously published procedure for the synthesis of 8. The product was purified by HPLC system 3; retention time, 24.2 min. HPLC solvent was removed under vacuum to yield 40 µg, 0.01% of the title compound. LC-ESI-MS m/z (rel intensity): 303 (M + H,

Chem. Res. Toxicol., Vol. 18, No. 6, 2005 1051 100), 304 (M + 2, 15). ESI-MS/MS of m/z 303 gave m/z 152 (100). NMR analysis matched previously reported data for 8 but lacked the pyridyl proton signals (10). Standardization of Analytical Solutions. A standard solution of 8 was prepared as previously described (10). The concentration of 11 was determined by 1H NMR analysis in DMSO-d6 with toluene as an internal standard. The areas for the 4-pyr, 5-pyr, and H6 signals (δ ) 8.29, 7.55, and 7.81 ppm) and aromatic toluene signals (δ ) 7.13, 7.24 ppm) were integrated four times, and the values were averaged. The range of these values was used in the regression analysis for standard curves. The concentration was determined after dilution with H2O to make a standard solution suitable for analysis. The concentration of 10 was determined in the same manner, using an average of integrated areas for the adduct signals corresponding to 4-pyr and H6 (δ ) 7.81 and 8.28 ppm) and toluene aromatic signals (δ ) 7.13, 7.24 ppm). Standard solutions of 4-HPB and 6 were prepared gravimetrically. Standard Curves. Adduct solutions with concentrations ranging from 2 to 250 fmol/injection of 6, 8, 10, and 11 were used to generate calibration curves. A constant concentration of d4-8 (58 fmol/µL) was used as an injection standard. LC/MS/ MS analysis was carried out using system 5. Reaction of 1 with DNA. Reactions of 1 with DNA (reactions a-d) were carried out as previously reported (10). Varying amounts of 1 (a, 10 mg, 40 µmol; b, 20 mg, 80 µmol; c, 264 mg, 1000 µmol; and d, 150 mg, 570 µmol) were allowed to react with varying amounts of calf thymus DNA, measured by weight (a, 20 mg, 66 µmol; b, 20 mg, 66 µmol; c, 40 mg, 132 µmol; and d, 15 mg, 49.5 µmol, respectively) in 0.1 M phosphate buffer (0.5 mL per mg of DNA), pH 7.0, in the presence of porcine liver esterase (0.5-10 µL of esterase, activity 3.57 units/µl, per mg of 1) at 37 °C for 1.5 h. Relative ratios of DNA:1 expressed as ratios of moles of DNA bases to moles of 1 are approximately 2:1, 1:1, 1:8, and 1:12, respectively. Each mixture was diluted with a volume of H2O equal to the amount of buffer used in the reaction and extracted twice with an equal volume of CHCl3/ isoamyl alcohol (24:1) and ethyl acetate (EtOAc). DNA was precipitated by the addition of cold ethanol and washed with 70% aqueous ethanol and ethanol sequentially. For enzyme hydrolysis, treated DNA (2.5 mg) was dissolved in 10 mM TrisHCl/5 mM MgCl2 buffer, pH 7.0 (1 mL). The mixture was incubated at 37 °C for 70 min with DNAse I, phosphodiesterase I, and alkaline phosphatase as described previously (10). For neutral thermal hydrolysis, the enzyme hydrolysate was heated at 100 °C for 1 h. Hydrolysates were further purified by SPE as described (17). The resulting CH3OH solutions were concentrated under vacuum, and H2O (approximately 0.5-1.5 mL) was added. A constant concentration of injection standard d4-8 was added, and the samples were analyzed in triplicate using system 5. 1H

Results Exposure of calf thymus DNA to 1 in the presence of an esterase, followed by enzyme and neutral thermal hydrolysis, gives rise to nucleoside and base adducts including 6, 8, 10, and 11 (Chart 1). We report here the preparation of standards d4-8, 10, and 11, as well as the results of a quantitative mass spectrometric method for analyzing relative levels of adducts present in mixtures of the POB-1-yl adducts. Synthesis. On the basis of the published procedure for the preparation of 8, the injection standard d4-8 was prepared by alkylation of triacetylguanosine 12 with 13 (Scheme 2) (10). The adduct 11 was synthesized as illustrated in Scheme 3. The proposed intermediate for this transformation is the seven-membered ring that results from intramolecular displacement of the tosyl group (18). The hydroxy group of 4-HPB serves as a nucleophile to open this ring resulting in the desired

1052

Chem. Res. Toxicol., Vol. 18, No. 6, 2005

Sturla et al.

Scheme 2. Synthesis of Deuterium-Labeled 8, Used as an Injection Standarda

a

Alkylation by the d-labeled tosylate is followed by depurination in situ.

Scheme 3. Synthesis of an Analytical Standard for the Hydrolytically Stable dThd Adduct 11

adduct. A drawback of this route is its requirement for a large excess of 4-HPB. Efforts aimed to develop conditions that allow for the reaction between thymidine-5′-toluenesulfonate ester and 4-HPB in an organic solvent have been unsuccessful thus far. The product of this reaction provides further confirmation of the structure of 11, as the HPLC retention times and NMR data are consistent with the material obtained by incubation of 1 with dThd. The concentrations of 11 solutions were measured by adding a known amount of toluene as an internal NMR standard. The 1H NMR spectrum was acquired with careful attention to minimize contributions to integration errors inherent in Fourier transform nuclear magnetic resonance (FT-NMR) (19). The estimated error for determining the concentration of standard solutions in this manner was 2-10%, based on comparison to gravimetrically prepared solutions of test nucleosides (data not shown). Efforts aimed at the independent synthesis of 10 met with little success, and for the current study, we chose to isolate the adduct produced using 1 as the alkylating reagent. In the presence of an esterase, 1 was incubated with dCyd and the adduct was isolated by reverse phase HPLC and characterized by 1H NMR. The concentration of a standard solution of 10 was determined in a manner analogous to that described above for the dThd adduct. Reaction of 1 with DNA. Calf thymus DNA was exposed to 1 in the presence of an esterase at 37 °C for 1.5 h, as described previously (10). The molar ratios of DNA bases to 1 were 2:1, 1:1, 1:8, and 1:12. After the incubation period, treated DNA was washed thoroughly with ethanol to remove residual 4-HPB. Samples were subjected to enzymatic and neutral thermal hydrolysis, sequentially. Samples, spiked with a constant concentration of the injection standard d4-8, were analyzed using a triple quadrupole mass spectrometer with an ESI source operating in the positive ion mode with SRM. Transitions for the ions of interest were monitored as listed in Table 1. To calibrate the response of the instrument and demonstrate linearity of response within our concentration range of interest, calibration curves were constructed using standard solutions (Figure 1). Each of the analytes was quantitated using a linear

Figure 1. Plot of ion signal peak area as a function of the amount of standard injected. The data were generated by LC/ MS/MS analysis of standard solutions of the DNA-POB-1-yl adducts. Error bars reflect uncertainty in determining analytical solution concentrations using 1H NMR with an internal standard. Lines are the linear regression fits (with the y-intercept set to 0) to the data. The slope of each line is the MS response to a given adduct for the mass transition monitored (Table 1). For the lines shown below, the slopes and R2 values are as follows: 6, slope ) 5920, R2 ) 0.998; 8, slope ) 870, R2 ) 0.99; 10, slope ) 6900, R2 ) 0.994; and 11, slope ) 8620, R2 ) 0.992.

regression from the corresponding calibration curve. Detection limits for each adduct using this method, with a signal-to-noise ratio of two, were 0.4 fmol for 11, 1 fmol for 6 and 10, and 3 fmol for 8. Five adduct masses were simultaneously monitored; a typical chromatogram is shown in Figure 2, and m/z values are listed in Table 1. We observed small variations in the absolute instrument response from day to day, and calibration curves were performed immediately prior to each session of sample analyses. The slopes of the adduct-calibration curves (area units/fmol) were 870, 5920, 6900, and 8620, for 8, 6, 10, and 11, respectively, and were linear in the concentration ranges of interest for this study. The same procedure was carried out to determine the concentration of 4-HPB in each incubation. The limit of detection was

Pyridyloxobutylation Adduct Levels

Chem. Res. Toxicol., Vol. 18, No. 6, 2005 1053

Figure 3. Relative abundance of DNA-POB-1-yl adducts measured by LC-ESI-MS/MS and expressed as percentages of the total DNA adducts monitored. The MS response was determined from standard curves using analytical standards for each adduct. Levels were measured for DNA:1 ratios of 2:1, 1:2, 1:8, and 1:12. The order of the relative abundances of the adducts was the same for the four incubations with the highest levels detected for 8. Error bars represent 95% confidence intervals from the linear regression fits of the standard curves.

Discussion

Figure 2. Typical LC-ESI-MS/MS analysis of enzymatic hydrolysate of POB-1-yl-DNA. d4-8 was the injection standard. Each chromatogram represents the total ion current for the SRM of the mass transitions listed in Table 1 for each adduct shown below. Table 2. Relative Abundance of Each Analyte Determined by LC/MS/MS Expressed as a Percentage of All Quantified POB Biomarkersa DNA:1 ratio

6

8

10

11

4-HPB

2:1 1:2 1:8 1:12

2.2 ( 0.7 2.3 ( 0.6 1.0 ( 0.4 1.2 ( 0.4

19.0 ( 5.5 24.5 ( 7.0 23.3 ( 6.7 23.3 ( 6.7

5.8 ( 3.5 6.6 ( 3.0 6.9 ( 1.9 6.1 ( 2.1

2.6 ( 2.5 4.1 ( 2.1 3.6 ( 1.1 3.8 ( 1.3

70.4 ( 6.0 62.4 ( 2.4 65.1 ( 0.8 65.5 ( 1.5

a Values represent concentrations determined from mixtures produced after sequential enzymatic and neutral thermal hydrolysis and purification by SPE. Concentrations were determined using the linear regression fits of the standard curves and the peak area integrations from the incubation chromatograms. The error bars were determined from the 95% prediction intervals of the standard curve linear regression fits where the error represents uncertainty in the standard concentrations.

11 fmol, and the slope of the standard curve between 40 and 10500 fmol was 870 area units/fmol (data not shown, R2 ) 0.9998). Using regression analysis (20), the concentration of each of the analytes was determined and the relative abundances as percentages in the resulting mixture are illustrated in Table 2. The levels of adducts remained similar regardless of DNA:1 ratio, and on average, HPB accounted for 66% of the mixture. The modified bases 8 and 10 comprised an average of 23 and 6% of the mixture, respectively. Deoxynucleoside adducts were detected at generally lower levels: 4% 11 and 2% 6. Figure 3 illustrates the relative distribution of nucleoside or base adducts as a percentage of the total adducts analyzed excluding the amount of 4-HPB detected.

The reaction of DNA with 1, followed by enzymatic and neutral thermal hydrolysis, results in the formation of POB-1-yl adducts (Chart 1) that can be categorized based on their hydrolytic stability. Adducts 6 and 11 are stable to neutral thermal hydrolysis. In contrast, 7 and 9 release either deoxyribose or 4-HPB under such conditions. Loss of the sugar results in the formation of base adducts 8 and 10, respectively. The structures of these adducts, and their corresponding stabilities, have been characterized previously by reacting 1 with DNA, but the relative levels could only be approximated based on areas of chromatographic peaks. In the present study, the relative levels of the adducts 6, 8, 10, and 11 were measured by HPLCESI-MS/MS using analytical standards for each compound. For practical reasons, the concentrations of 1 used in these experiments were higher than the corresponding levels of activated NNK exposure in vivo. Nonetheless, the levels measured provide valuable information about the relative propensity for formation of adducts and demonstrate their relative chemical reactivity toward the POB-1-yl electrophile. It is important that this method be applied to the analysis of in vivo samples to determine absolute levels, chemical recoveries, and whether the distribution of adducts in vivo mirror those presented from these in vitro experiments. Two important considerations are the relative recoveries of each adduct during workup and the efficiency of DNA hydrolysis. On the basis of the solubility and reverse phase chromatography behavior of the various analytes, we considered the adduct recoveries to be adequately similar for the purposes of comparing relative abundances in this study. While beyond the scope of this report, future efforts are focused on the preparation of stable isotope-labeled analogues of each of the analytes to be used as internal standards prior to sample cleanup. Use of such internal standards will give both accurate recoveries for each adduct and allow for the determination of absolute levels of adducts in DNA. The hydrolysis protocol used in this study reflects the general method used for the analysis of 4-HPB release, an established

1054

Chem. Res. Toxicol., Vol. 18, No. 6, 2005

biomarker for pyridyloxobutylation. However, it has been shown that 6 can inhibit the enzymatic digestion of oligonucleotides (21) and that unstable POB-1-yl adducts may interfere with complete enzymatic digestion of alkylated DNA (14). To address this issue, Peterson and co-workers have developed a digestion method, the efficiency of which was supported by experiments using [3H]-labeled modified DNA, that can be used to accurately assess the level of 6 in DNA. Using their modified hydrolysis procedure and analyzing with the method described here, we found the relative abundance of modified bases to change from an average of 5 to 11% for 6, from 66 to 60% for 8, from 19 to 8% for 10, and from 10 to 21% for 11. When considering the overall distribution of these adducts in NNKOAc-treated DNA, losses resulting from potentially incomplete digestion do not appear to dramatically influence relative abundances. However, as evidenced by changes such as the 2-fold increase in 6 levels detected using the Peterson method, we conclude that optimization of hydrolytic conditions for each adduct may be required to determine their absolute levels with a high degree of accuracy. The pattern of nucleoside and base adduct distribution, as illustrated by the values shown in Figure 3, varied only slightly with changes in the concentration of 1 with the relative propensities for formation of each of the adducts remaining similar. The most abundant adduct, 8, was formed in relative levels ranging from 64 to 68% of the total analyzed adduct mixture. Compound 10, also a modified base that results from decomposition of an unstable nucleoside adduct, comprised 18-20% of the adducts. Compound 11 is formed with relative abundances ranging from 9 to 11%. The lowest adduct levels detected were those of the stable mutagenic adduct 6. Because 8 and 10 are the decomposition products of their DNA adduct precursors 7 and O2-(POB-1-yl)dCyd (9), the measured levels of base adducts represent lower limits of their corresponding levels in DNA. These 4-HPBreleasing adducts are therefore present at levels higher than the stable adducts, as is further supported by the levels of 4-HPB released from these DNA samples (vide infra). The high levels of 4-HPB measured in this study indicate that a significant portion of deoxynucleoside adducts decompose to 4-HPB using the hydrolytic method described here. On the basis of the levels of corresponding base adducts, we have characterized 7 and 9 as major sources of 4-HPB. However, the possibility remains that other 4-HPB-releasing adducts, for which the corresponding base adducts have not been identified, contribute to the high levels of 4-HPB. The partition coefficient for the formation of 4-HPB and 8 or 10 from 7 and 9, respectively, is difficult to determine because of the instability of the deoxynucleoside adducts. The rate of depyrimidation of unmodified deoxyribonucleosides has been estimated to be about 20 times lower than the rate of depurination in DNA under conditions of neutral thermal hydrolysis, both occurring at low but biologically relevant rates. When these bases are modified at certain positions, however, rates for both processes are greatly accelerated. O2-EthyldCyd has a half-life of less than 0.1 h at neutral pH at 95 °C and 26 h at 37 °C. For comparison, 7-methyldGuo has a half-life of 6.5 h at neutral pH at 37 °C (4, 22). By analogy, we can estimate that 7 may lose deoxyribose about four times as fast as the corresponding loss from 9. While we can assume that

Sturla et al.

the ratio of base adducts (8:10) is proportional to the ratios of their corresponding rates of formation, the relative rates of loss of 4-HPB from 7 and 9 is not known. In considering the relative levels of the POB-1-yl adducts, as is illustrated in Figure 3, the consistently most abundant adduct is 7. These data are consistent with the high nucleophilicity of the 7-position of dGuo relative to other positions on the DNA bases (23). While studies of the solvolysis of 1 support alkylation of the 7-position of dGuo by the proposed electrophilic intermediate, they also illustrate the potentially complex and unusual pattern of modification of strong nucleophiles including evidence for an oxonium ion intermediate (13). It is often generalized that less ionic alkylating agents give rise to adducts involving ring nitrogen alkylation and alkylating agents with higher ionic character modify oxygen atoms (6). The pattern of DNA alkylation determined in this study, including high levels of 8, is similar to simple R-X alkylating agents such as methyl methanesulfonate and N-methyl-N-nitrosourea, with the caveat of relatively high levels of pyrimidine adducts 9 and 11 (4, 6). The latter result may be due to noncovalent preassociation of the pyridyl ring with the pyrimidine sites, as discussed previously (11). The prevalence of the 7 adduct is notable because it has been shown that relatively large 7-Gua adducts, for which the pyridyloxobutyl group qualifies, can have significant biological influence including genetic mutations (24-28). While stable in DNA, the initially formed 7 is highly unstable on the deoxynucleoside level, consistent with the high rate of depurination of 7-alkyldGuo (22). We can detect the presence of this adduct by MS (10), but its lability precludes the synthesis of a suitable analytical standard. The corresponding depurination adduct 8 has been a useful tool for the identification of its precursor 7. The same is true for the unstable dCyd adduct 9, which upon depyrimidation gives rise to the measured base adduct 10. Compound 6 is present in low levels relative to other adducts but is of biological significance and has been quantified in liver DNA isolated from mice treated with NNK or 1 (14). It has been demonstrated to be highly mutagenic in Escherichia coli and human cells and to interfere with the activity of the repair enzyme O6alkylguanine-DNA alkyltransferase (1, 9, 29, 30). Recently, the solution structure of 6 opposite dC in doublestranded DNA has been reported (31). While 6 is present at low levels relative to other adducts presented in this study, this minor DNA adduct has profound biological influence. The biological consequences of the more prevalent adducts, however, are not known. Nonetheless, by understanding the distribution of various adducts, more abundant lesions can serve as dosimeters of the low level lesions. In summary, we have determined the relative levels of 4-HPB and four major POB-1-yl adducts from the reaction of DNA with 1. The highest levels were those of the adducts that result from loss of deoxyribose upon neutral thermal hydrolysis, 8 and 10. Taken together with the 4-HPB levels, these data support that unstable adducts are present at levels up to 20 times the amount of the stable adducts. Furthermore, these data support that levels of 4-HPB released from DNA reflect a significant portion of the DNA damage resulting from exposure to the tobacco specific carcinogen NNK. Although the levels of 11 were one of the lowest of the adducts

Pyridyloxobutylation Adduct Levels

analyzed, it is notable because its high MS response factor allows for its detection at levels below those of the more abundant 8. The relative adduct levels determined in this study in situ reflect the chemical reactivity of the pyridyloxobutylating agent derived from NNK toward DNA-based nucleophiles; however, biological considerations such as DNA sequence context, DNA repair, and metabolic variability can impact in vivo levels (32, 33). Therefore, we currently are carrying out studies aimed at using MS methods to measure adduct levels in laboratory animals exposed to NNK.

Acknowledgment. This study was supported by Grant CA-81301 from the National Cancer Institute. S.S.H. is an American Cancer Society (ACS) Research Professor supported by ACS Grant 00-138. S.J.S. was supported by a Bowman Cancer Research Fellowship from the ACS. We thank Guang Cheng for technical assistance in DNA isolation and Pramod Upadhyaya for generously providing an authentic sample of 6 for this study, Steven Carmella for helpful suggestions, and Tracy Bergemann for assistance in error analysis.

Chem. Res. Toxicol., Vol. 18, No. 6, 2005 1055

(15)

(16)

(17)

(18)

(19) (20) (21)

References (1) Hecht, S. S., and Hoffmann, D. (1988) Tobacco-specific nitrosamines, an important group of carcinogens in tobacco and tobacco smoke. Carcinogenesis 9, 875-884. (2) Hecht, S. S. (1998) Biochemistry, biology, and carcinogenicity of tobacco-specific N-nitrosamines. Chem. Res. Toxicol. 11, 559-603. (3) IARC (2004) International Agency for Research on Cancer Press Release No. 154. (4) Singer, B., and Grunberger, D. (1983) Molecular Biology of Mutagens & Carcinogens, Plenum Press, New York. (5) Gooderham, N. J., and Carmichael, P. L. (2002) Mechanisms of chemical carcinogenesis. In The Cancer Handbook (Alison, M. R., Ed.) pp 259-272, Nature Publishing Group, London. (6) Dipple, A. (1995) DNA adducts of chemical carcinogens. Carcinogenesis 16, 437-441. (7) Hecht, S. S., Spratt, T. E., and Trushin, N. (1988) Evidence for 4-(3-pyridyl)-4-oxobutylation of DNA in F344 rats treated with the tobacco specific nitrosamines 4-(methylnitrosamino)-1-(3pyridyl)-1-butanone and N′-nitrosonornicotine. Carcinogenesis 9, 161-165. (8) Upadhyaya, P., Sturla, S. J., Tretyakova, N., Ziegel, R., Villalta, P. W., Wang, M., and Hecht, S. S. (2003) Identification of adducts produced by the reaction of 4-(acetoxymethylnitrosamino)-1-(3pyridyl)-1-butanol with deoxyguanosine and DNA. Chem. Res. Toxicol. 16, 180-190. (9) Wang, L. J., Spratt, T. E., Liu, X. K., Hecht, S. S., Pegg, A. E., and Peterson, L. A. (1997) Pyridyloxobutyl adduct O6-[4-oxo-4(3-pyridyl)butyl] guanine is present in 4-(acetoxymethylnitrosamino)-1-(3-pyridyl)-1-butanone-treated DNA and is a substrate for O6-alkylguanine-DNA alkyltransferase. Chem. Res. Toxicol. 10, 562-567. (10) Wang, M., Cheng, G., Sturla, S. J., Shi, Y., McIntee, E. J., Villalta, P. W., Upadhyaya, P., and Hecht, S. S. (2003) Identification of adducts formed by pyridyloxobutylation of deoxyguanosine and DNA by 4-(acetoxymethylnitrosamino)-1-(3-pyridyl)-1-butanone, a chemically activated form of tobacco specific carcinogens. Chem. Res. Toxicol. 16, 616-626. (11) Hecht, S. S., Villalta, P. W., Sturla, S. J., Cheng, G., Yu, N. X., Upadhyaya, P., and Wang, M. Y. (2004) Identification of O2substituted pyrimidine adducts formed in reactions of 4-(acetoxymethylnitrosamino)-1-(3-pyridyl)-1-butanone and 4-(acetoxymethylnitrosamino)-1-(3-pyridyl)-1-butanol with DNA. Chem. Res. Toxicol. 17, 588-597. (12) Haglund, J., Henderson, A. P., Golding, B. T., and Tornqvist, M. (2002) Evidence for phosphate adducts in DNA from mice treated with 4-(N-methyl-N-nitrosamino)-1-(3-pyridyl)-1-butanone (NNK). Chem. Res. Toxicol. 15, 773-779. (13) Spratt, T. E., Peterson, L. A., Confer, W. L., and Hecht, S. S. (1990) Solvolysis of model compounds for R-hydroxylation of N′-nitrosonornicotine and 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone: Evidence for a cyclic oxonium ion intermediate in the alkylation of nucleophiles. Chem. Res. Toxicol. 3, 350-356. (14) Thomson, N. M., Mijal, R. S., Ziegel, R., Fleischer, N. L., Pegg, A. E., Tretyakova, N. Y., and Peterson, L. A. (2004) Development

(22)

(23) (24)

(25)

(26)

(27)

(28)

(29)

(30)

(31)

(32)

(33)

of a quantitative liquid chromatography/electrospray mass spectrometric assay for a mutagenic tobacco specific nitrosaminederived DNA adduct, O6-[4-oxo-4-(3-pyridyl)butyl]-2′-deoxyguanosine. Chem. Res. Toxicol. 17, 1600-1606. NIH (1981) NIH Guidelines for the Laboratory Use of Chemical Carcinogens Vol. NIH Publication No. 81-2385, U.S. Department of Health and Human Services, Public Health Service, National Institutes of Health, Washington, DC. Carmella, S. G., Kagan, S. S., Spratt, T. E., and Hecht, S. S. (1990) Evaluation of cysteine adduct formation in rat hemoglobin by 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone and related compounds. Cancer Res. 50, 5453-5459. Cheng, G., Shi, Y., Sturla, S. J., Jalas, J. R., McIntee, E. J., Villalta, P. W., Wang, M., and Hecht, S. S. (2003) Reactions of formaldehyde plus acetaldehyde with deoxyguanosine and DNA: Formation of cyclic deoxyguanosine adducts and formaldehyde cross-links. Chem. Res. Toxicol. 16, 145-152. Reist, E. J., Benitez, A., and Goodman, L. (1964) The synthesis of some 5′-thiopentofuranosylpyrimidines. J. Org. Chem. 29, 554558. Derome, A. E. (1987) Modern NMR Techniques for Chemistry Research, Vol. 6, Pergamon Press, New York. Myers, R. H. (1990) Classical and Modern Regression With Applications, Duxbury/Thompson Learning, Pacific Grove, CA. Park, S., Seetharaman, M., Ogdie, A., Ferguson, D., and Tretyakova, N. (2003) 3′-Exonuclease resistance of DNA oligodeoxynucleotides containing O6-[4-oxo-4-(3-pyridyl)butyl]guanine Nucleic Acids Res. 31, 1984-1994. Gates, K. S., Nooner, T., and Dutta, S. (2004) Biologically relevant chemical reactions of N7-alkylguanine residues in DNA. Chem. Res. Toxicol. 17, 839-856. Pullman, A., and Pullman, B. (1981) Molecular electrostatic potential of the nucleic acids. Q. Rev. Biophys. 14, 289-380. Smela, M. E., Currier, S. S., Bailey, E. A., and Essigmann, J. M. (2001) The chemistry and biology of aflatoxin B-1: From mutational spectrometry to carcinogenesis. Carcinogenesis 22, 535545. Chakravarti, D., Mailander, P. C., Li, K. M., Higginbotham, S., Zhang, H. L., Gross, M. L., Meza, J. L., Cavalieri, E. L., and Rogan, E. G. (2001) Evidence that a burst of DNA depurination in SENCAR mouse skin induces error-prone repair and forms mutations in the H-ras gene. Oncogene 20, 7945-7953. Kim, M. S., and Guengerich, F. P. (1998) Polymerase blockage and misincorporation of dNTPs opposite the ethylene dibromidederived DNA adducts S-[2-(N7-guanyl)ethyl]glutathione, S-[2-(N2guanyl)ethyl]glutathione, and S-[2-(O6-guanyl)ethyl]glutathione. Chem. Res. Toxicol. 11, 311-316. Sahasrabudhe, S. R., Luo, X., and Humayun, M. Z. (1990) Induction of G-C to A-T transitions by the acridine half-mustard ICR-191 supports a mispairing mechanism for mutagenesis by some bulky mutagens. Biochemistry 29, 10899-10905. Nooner, T., Dutta, S., and Gates, K. S. (2004) Chemical properties of the leinamycin-guanine adduct in DNA. Chem. Res. Toxicol. 17, 942-949. Liu, X. K., Spratt, T. E., Murphy, S. E., and Peterson, L. A. (1996) Pyridyloxobutylation of guanine residues by 4-(acetoxymethylnitrosamino)-1-(3-pyridyl)-1-butanone generates substrates of O6allylguanine-DNA allyltransferase. Chem. Res. Toxicol. 9, 949953. Pauly, G. T., Peterson, L. A., and Moschel, R. C. (2002) Mutagenesis by O6-[4-oxo-4-(3-pyridyl)butyl]guanine in Escherichia coli and human cells. Chem. Res. Toxicol 15, 165-169. Peterson, L. A., Vu, C., Hingerty, B. E., Broyde, S., and Cosman, M. (2003) Solution structure of an O6-[4-oxo-4-(3-pyridyl)butyl]guanine adduct in an 11mer DNA duplex: Evidence for formation of a base triplex. Biochemistry 42, 13134-13144. Rajesh, M., Wang, G., Jones, R., and Tretyakova, N. (2005) Stable isotope labeling-mass spectrometry analysis of methyl- and pyridyloxobutyl-guanine adducts of 4-(methylnitrosamino)-1-(3pyridyl)-1-butanone in p53-derived DNA sequences. Biochemistry 44, 2197-2207. Mijal, R. S., Thomson, N. M., Fleisher, N. L., Pauly, G. T., Moschel, R. C., Kanugula, S., Fang, Q., Pegg, A. E., and Peterson, L. A. (2004) The repair of the tobacco-specific nitrosamine derived adduct O6-[4-oxo-4-(3-pyridyl)butyl]guanine by O6-alkylguanineDNA alkyltransferase variants. Chem. Res. Toxicol. 17, 424-434.

TX050028U