Synthesis and Characterization of Aflatoxin B1 Mercapturic Acids and

For coelution studies, a Spherosorb (Phenomenex), 250 × 4.6 mm i.d., C18, 5 μm column was used. The mobile phase ...... BUSINESS CONCENTRATES ...
0 downloads 0 Views 198KB Size
1144

Chem. Res. Toxicol. 1997, 10, 1144-1151

Synthesis and Characterization of Aflatoxin B1 Mercapturic Acids and Their Identification in Rat Urine Peter F. Scholl,†,§ Steven M. Musser,‡ and John D. Groopman*,† Department of Environmental Health Sciences, School of Hygiene and Public Health, Johns Hopkins University, 615 North Wolfe Street, Baltimore, Maryland 21205, and Center for Food Safety and Applied Nutrition, United States Food and Drug Administration, 200 C Street, S.W., M/S HFS-717, Washington, D.C. 20204 Received September 9, 1996X

Biologic effects of the hepatocarcinogenic mycotoxin aflatoxin B1 are principally induced by one of its metabolites, the exo-aflatoxin B1 epoxide which produces both DNA and protein adducts in vivo. Detoxication of the exo-aflatoxin B1 epoxide can be mediated in part by glutathione S-transferases whose induction could be important in chemoprotection interventions. Thus, biomarkers of the enzymatic conjugation of exo-aflatoxin B1 epoxide with glutathione may be important indices of protection against the toxic effects of this agent. Since glutathione conjugates undergo further metabolic processing in vivo to yield mercapturic acids, increased urinary excretion of exo-aflatoxin B1 mercapturate could be expected during chemoprotection intervention. To determine if this mercapturic acid could be used as a biomarker, techniques for its specific measurement were developed using monoclonal antibody immunoaffinity chromatography and reverse phase high-performance liquid chromatography with ultraviolet absorbance and mass spectral detection. First, a synthetic exo-aflatoxin B1 mercapturate was characterized using mass spectrometry, ultraviolet absorbance, circular dichroism spectrometry, and chemical derivatization. In vivo metabolite characterization was then facilitated by comparison with the synthetically prepared exo-aflatoxin B1 mercapturate and both aflatoxin B1-glutathione conjugate diastereoisomers. In rats, 1% of the aflatoxin dose was excreted as exo-aflatoxin B1 mercapturate within 24 h. The finding that exo-aflatoxin B1 mercapturate was excreted in urine in a dose-dependent manner provides the basis for investigating its applicability as a biomarker of glutathione S-transferase status in aflatoxin chemoprotection studies.

Introduction Aflatoxin B1 (AFB1)1 is a naturally occurring mycotoxin that has been demonstrated to be one of the most potent liver carcinogens tested in experimental animal models (1, 2). Further, human epidemiologic studies indicate that exposure to AFB1 is a risk factor in the etiology of hepatocellular carcinoma (3-7). Prevention of the adverse health effects from the unavoidable, chronic, and high AFB1 exposure of Asian and sub-Saharan African populations is an important public health goal, and experimental models have provided insights into the molecular mechanisms of action of this carcinogen that can be used to design prevention measures (8-12). The field of molecular dosimetry has developed in response to the difficulties of traditional methods of analysis to assess the exposure status of an individual. Evolving biomarker methods are providing important * To whom correspondence should be addressed. Tel: 410-955-3720. Fax: 410-955-0617. E-mail: [email protected]. † Johns Hopkins University. ‡ Center for Food Safety and Applied Nutrition. § Present address: Department of Biochemistry, Vanderbilt University, Nashville, TN 37209. X Abstract published in Advance ACS Abstracts, September 15, 1997. 1 Abbreviations: aflatoxin B (AFB ), AFB -glutathione conjugates 1 1 1 (AFB1-GS), AFB1 mercapturic acids (AFB1-NAC), liquid chromatography/electrospray ionization mass spectrometry (LC/ESI-MS), aflatoxin M1 (AFM1), aflatoxin Q1 (AFQ1), aflatoxin G1 (AFG1), aflatoxin P1 (AFP1), 1,2-dithiole-3-thione (D3T), triethylammonium formate (TEAF), electrospray ionization mass spectrometry (ESI-MS), collisioninduced dissociation mass spectrometry (CID/MS), dimethyl sulfoxide (DMSO).

S0893-228x(96)00161-0 CCC: $14.00

tools for identifying individuals at high risk from AFB1, and past work has focused upon developing methods for analyzing the DNA and protein adducts derived from its ultimate carcinogenic metabolite, the exo-aflatoxin 8,9epoxide (5, 13-18). Other biomarkers reflecting the fate of the aflatoxin epoxide detoxication pathways are also needed, since a major factor for assessing the overall toxicological hazard of an individual following exposure to AFB1 is the integrated balance between the activation and detoxication processes (19-22). This work reports a method for assessing the glutathione S-transferase (GST)-mediated detoxication of exoAFB1 epoxide by measuring excretion of its heretofore unidentified mercapturic acid in urine (23). Metabolism of AFB1 by specific cytochrome P450 enzymes produces a mixture of two diastereoisomeric epoxides (24-26), and in vitro studies using GSTs have shown that both the exo- and endo-AFB1 epoxides are substrates that yield two diastereoisomeric AFB1-glutathione conjugates (AFB1GS) (25, 27). However, only the exo-epoxide isomer covalently reacts at the N7 position of guanine in DNA to form adducts, and this isomer is mutagenic in bacterial assays (17). Since the exo-AFB1-GS conjugate is degraded in vitro by the mercapturic acid pathway enzymes (28), it was predicted that AFB1 mercapturic acids (AFB1NAC) would be found in urine. In this study, both AFB1GS diastereoisomers and exo-AFB1-NAC were synthesized and the metabolite eluting at the same chromatographic retention time as the synthetic mercapturate was isolated from the urine of rats dosed with AFB1. This metabolite was initially characterized by comparison of © 1997 American Chemical Society

Synthesis and Characterization of AFB1-NAC

its UV absorbance spectra and acid hydrolysis products with those of aflatoxin standards. Since insufficient amounts of mercapturate can be isolated from urine for 1H NMR analysis, circular dichroism spectra of stereochemically defined aflatoxin standards and the in vivo metabolite were measured and compared to deduce the orientation of thiol addition at the C8 position of the urinary mercapturate. Synthetic standards and in vivo metabolites were analyzed by liquid chromatography/ electrospray ionization mass spectroscopy (LC/ESI-MS) to characterize the structures of urinary aflatoxin mercapturic acids. Finally, rats were dosed at three AFB1 levels, and 24 h urinary mercapturate excretion was measured.

Materials and Methods Caution: AFB1 is a human carcinogen and dimethyldioxirane is a volatile oxidant. Great care should be exercised to avoid personal exposure, and proper decontamination procedures should be used. Chemicals. AFB1, AFM1, AFQ1, AFG1, and AFP1 were purchased from Sigma Chemical Co. (St. Louis, MO). Potassium peroxysulfate (Oxone; DuPont trademark) was obtained from Aldrich Chemical Co. (Milwaukee, WI). Dimethyldioxirane was prepared by the alkaline oxidation of acetone with potassium peroxysulfate as previously described (29, 30). A 10:1 mixture of exo- and endo-AFB1 epoxides was prepared by oxidizing AFB1 with dimethyldioxirane (24, 31). exo- and endo-AFB1-GS conjugates were synthesized using established methods, and structures were confirmed by 1H NMR and mass spectrometry (32). 1,2-Dithiole-3-thione (D3T) was kindly provided by Dr. T. W. Kensler. All chromatographic solvents were of HPLC grade quality. All other chemicals were of the highest grade commercially available. A molar extinction coefficient of 20 000 at 365 nm was used for quantitation of all AFB-thiol adducts (28, 33). Synthesis and Purification of exo-AFB1-NAC. exo-AFB1NAC was prepared by reacting the disodium salt of N-acetylL-cysteine (90 µmol) dissolved in MeOH with the AFB1 epoxide mixture (3 mL, ∼9 µmol of epoxide). After 1 min, the reaction mixture was neutralized by addition of 1 M acetic acid (5 mL). Solvent was removed using a centrifugal vacuum concentrator. The reaction mixture was dissolved in 12% MeOH/H2O (3 mL), and the product was isolated by HPLC using a Microsorb (Rainin), 5 µm, C18, 250 × 21.4 mm i.d. preparative column eluted isocratically with acetonitrile/MeOH/H2O (1:1:8, v/v) at a flow rate of 12 mL/min at ambient temperature. The eluate absorbance (365 nm) was monitored using a Beckman Model 160 single-wavelength detector. The fraction eluting at 13 min was concentrated and HPLC purified by multiple injections on an Econosphere (Alltech), 5 µm, C18, 250 × 4.6 mm i.d. column. The mobile phase consisted of solvent A, 25 mM TEAF (pH 3), and solvent B, acetonitrile/MeOH (1:1, v/v). A linear gradient of 18-30% solvent B was generated over 32 min at a flow rate of 1 mL/min at 35 °C. The retention times of AFB1-GS (exoand endo-forms coelute), AFB1-diol, and exo-AFB1-NAC were 16.0, 21.6, and 22.6 min, respectively. Isolation of AFB1 Mercapturate from Rat Urine. Eight male Fischer rats F344 (Charles River Breeding Labs) (100110 g) were maintained under conditions of controlled humidity and lighting and fed ad libitum a menadione free AIN-76 diet supplemented with 0.03% D3T. During AFB1 dosing, the rats were individually housed in glass metabolic cages and urine was collected daily. Rats were dosed 5 days/week for 8 weeks by gavage with DMSO (100 µL) containing AFB1 (40 µg) (18). Urine was pooled (700 mL), MeOH (2800 mL) was added, and the mixture was stored at 4 °C overnight to precipitate proteins. This solution was decanted, and the supernatant was passed through glass frit filters (40-60 then 10-20 µm pore size) and then concentrated by rotary evaporation. The concentrate was

Chem. Res. Toxicol., Vol. 10, No. 10, 1997 1145 subjected to preliminary cleanup by two different methods. One fraction (50 mL) was diluted with acetone/MeOH (30 mL), acidified to pH 4.5 using glacial acetic acid, and spotted onto preparative thin layer chromatography (TLC) plates (C18 Si on glass, 20 × 20 cm, Sigma-Aldrich Z29295-8, 1000 µm thickness, 2-10 µm, 60 Å pore size). The mobile phase was acetone/ acetonitrile/MeOH/H2O (5:1:1:3, v/v). Under these conditions the exo-AFB1-NAC standard had an Rf ∼0.45. After development, a green fluorescent band extending from Rf 0.35 to 0.45 was visualized by long wavelength UV illumination, scraped from the plate, formed into a column, and eluted with 80% MeOH/H2O. Due to excessive band spreading during the TLC plate spotting, the remaining urine concentrate (80 mL) was diluted with H2O (40 mL) and the pH adjusted to 4.5 using glacial acetic acid. Three units consisting of three stacked Sep Pak cartridges (C18, 2 mL bed volume; Waters-Millipore Corp.) were successively prewashed with H2O, 80% MeOH/H2O, and H2O (20 mL each). Separate portions of diluted sample (40 mL) were loaded on each unit which was then progressively rinsed with 15%, 25%, 35%, and 50% MeOH/H2O (20 mL each). Analytical HPLC was performed to identify fractions containing AFB1-NAC. TLC and Sep Pak fractions (25% and 35% washes) containing AFB1NAC were combined and concentrated to dryness by rotary evaporation. exo-AFB1-NAC was isolated using the preparative HPLC column described above. The mobile phase consisted of solvent A, acetic acid (1 mM), and solvent B, acetonitrile/MeOH (1:1, v/v). The aflatoxins were chromatographed using a 1035% linear gradient generated over 30 min followed by a 3550% linear gradient generated over 30 min at a flow rate of 12 mL/min at ambient temperature. The retention time of exoAFB1-NAC was 25 min. The HPLC fraction containing the aflatoxin mercapturate was subsequently purified using analytical HPLC as described in the synthesis of AFB1-NAC. Chromatographic and Acid Hydrolysis Studies. The in vivo metabolite was initially characterized by comparison of its chromatographic properties and acid hydrolysis products with synthetic standards. For coelution studies, a Spherosorb (Phenomenex), 250 × 4.6 mm i.d., C18, 5 µm column was used. The mobile phase consisted of solvent A, 5 mM TEAF (pH 3.0), and solvent B, EtOH. A linear gradient of 18-35% B generated over 50 min at a flow rate of 1 mL/min at 40 °C was used. The retention time of exo-AFB1-NAC was 6.8 min. An Econosphere (Alltech), 250 × 4.6 mm i.d., C18, 5 µm column was also used. The mobile phase consisted of solvent A, H2O (pH 4.5), and solvent B, acetonitrile/MeOH (1:1, v/v). A linear gradient of 2035% B was generated over 30 min, 1 mL/min, at 40 °C. The retention times of exo-AFB1-NAC, AFB1-GS, AFB1-diol, and AFB1-N7-Gua were 13.0, 14.0, 17.8, and 19.0 min, respectively. Hydrolysis studies of all synthetic AFB1-thiol standards and the in vivo metabolite were performed by individually heating samples in 0.1 N HCl for 15 min. Samples were then adjusted to pH 4 and analyzed by HPLC. The HPLC column was Econosphere (Alltech), 250 × 4.6 mm i.d., C18, 5 µm. The mobile phase consisted of solvent A, H2O (pH 4.5), and solvent B, acetonitrile/MeOH (1:1, v/v). A linear gradient of 10-30% B was generated over 20 min at a flow rate of 1 mL/min at 40 °C. The retention times of the standards exo-AFB1-NAC, exoAFB1-GS, endo-AFB1-GS, and AFB1-diol were 19.9, 21.1, 21.4, and 24.8 min, respectively. Characterization of Synthetic and in Vivo Aflatoxinthiol Adducts. Buffer salts were removed from samples prior to analysis by loading onto a prewashed Sep Pak cartridge, washing with water (5 mL), eluting with 80% MeOH/H2O, and concentrating to dryness in a vacuum centrifuge without heating. Samples were then characterized by UV/visible spectroscopy, electrospray ionization spectrometry, collision-induced dissociation mass spectrometry (CID/MS), and 1H NMR spectroscopy. ESI mass spectra were obtained on a Finnigan Model TSQ-7000 triple-quadrupole mass spectrometer using a standard Finnigan electrospray ion source. For CID experiments xenon was used as the collision gas with a collision energy of -13 eV and cell pressure of 0.8 Torr. After mass spectral

1146 Chem. Res. Toxicol., Vol. 10, No. 10, 1997 analysis, samples were lyophilized from D2O twice prior to acquisition of 1H NMR in D2O using a 300 MHZ Bruker FT spectrometer. LC/ESI-MS Characterization of AFB1 Mercapturate Isolated in Vivo. Metabolites isolated for LC/ESI-MS analysis were obtained from a male Fischer rat dosed with AFB1 (1 mg/ kg, ip, 150 µL of DMSO) and housed in a glass metabolic cage. Urine was collected for 24 h. This animal was maintained on an AIN76A diet for 1 week prior to AFB1 dosing. Urine (pH 6) was diluted with water (10 mL), acidified to pH 4.5 using 5 M acetic acid, and loaded onto a prewashed Sep Pak cartridge. Cartridges were washed with 5% MeOH/H2O (5 mL) and eluted with 80% EtOH/H2O (5 mL). Eluant was concentrated to dryness in a vacuum centrifuge without heating. Solids were dissolved at room temperature in 1 mM formic acid (100 µL) followed by 10 mM ammonium acetate-saline buffer (10 mL, pH 5.5). Samples were gravity loaded onto an anti-aflatoxin immunoaffinity column (4 mL bed volume, resin capacity ) 800 ng of aflatoxin equiv/mL) and gravity rinsed with buffer (15 mL) (13). The affinity chromatographic method used here differs from that previously reported. In the previous method urinary aflatoxin metabolites were heated in 0.1 N HCl for 10 min at 50 °C; however, this process degrades exo-AFB1-NAC to AFB1diol and other polar aflatoxin metabolites to more lipophilic compounds. To prevent metabolite degradation, samples were briefly dissolved at room temperature in 1 mM formic acid and diluted with ammonium acetate buffer (pH 5.5). An ammonium acetate-saline buffer (pH 5.5) was used for affinity chromatography instead of phosphate-buffered saline (pH 7.4). Finally, solutions were acidified to pH 4.5 before Sep Pak removal of DMSO to improve the recovery of polar metabolites. Using AFG1 as an internal standard, the method recovery was 65%. Affinity columns were eluted with 70% DMSO/H2O (6 mL) followed by ammonium acetate-saline buffer (10 mL). Eluant was diluted to