1110
Chem. Res. Toxicol. 1999, 12, 1110-1116
Role of Cytochrome P450 2E1 in the Metabolism of Acrylamide and Acrylonitrile in Mice Susan C. J. Sumner,*,† Timothy R. Fennell,† Timothy A. Moore,† Brian Chanas,‡ Frank Gonzalez,§ and Burhan I. Ghanayem‡ Chemical Industry Institute of Toxicology, Research Triangle Park, North Carolina 27709-2137, National Institute of Environmental Health Sciences, Research Triangle Park, North Carolina 27709, and National Cancer Institute, Bethesda, Maryland 20892 Received March 5, 1999
Acrylonitrile (AN) and acrylamide (AM) are commonly used in the synthesis of plastics and polymers. In rodents, AM and AN are metabolized to the epoxides glycidamide and cyanoethylene oxide, respectively. The aim of this study was to determine the role of cytochrome P450 in the metabolism of AM and AN in vivo. Wild-type (WT) mice, WT mice pretreated with aminobenzotriazole (ABT, 50 mg/kg ip, 2 h pre-exposure), and mice devoid of cytochrome P450 2E1 (P450 2E1-null) were treated with 50 mg/kg [13C]AM po. WT mice and P450 2E1-null mice were treated with 2.5 or 10 mg/kg [13C]AN po. Urine was collected for 24 h, and metabolites were characterized using 13C NMR. WT mice excreted metabolites derived from the epoxides and from direct GSH conjugation with AM or AN. Only metabolites derived from direct GSH conjugation with AM or AN were observed in the urine from ABT-pretreated WT mice and P450 2E1-null mice. On the basis of evaluation of urinary metabolites at these doses, these data suggest that P450 2E1 is possibly the only cytochrome P450 enzyme involved in the metabolism of AM and AN in mice, that inhibiting total P450 activity does not result in new pathways of non-P450 metabolism of AM, and that mice devoid of P450 2E1 do not excrete metabolites of AM or AN that would be produced by oxidation by other cytochrome P450s. P450 2E1-null mice may be an appropriate model for the investigation of the role of oxidative metabolism in the toxicity or carcinogenicity of these compounds.
Introduction (AM)1
Acrylamide and acrylonitrile (AN) are used in the manufacture of plastics and polymers. The potential for human exposure to AN via air or water can occur during monomer and polymer production, transportation, and usage (1). Human exposure to AM can occur during its manufacture and via its use in the production of polyacrylamide gels, as a grouting agent, and in soil, tunnel, and dam stabilization (2). Humans exposed to AM exhibit neurotoxic effects (3), and in exposed laboratory animals, AM produces neurotoxic, genotoxic, reproductive, and carcinogenic effects (3, 4). Rats exposed to AN develop tumors in the brain, stomach, and Zymbal’s gland (5). Rats and mice treated with AM (6-8) or AN (8-12) metabolize the parent compound by direct conjugation with glutathione (GSH), resulting in the urinary excretion of N-acetylcysteine and thioacetic acid derivatives (Scheme 1). Cyanoethylene oxide (CEO) and glycidamide (GA), formed presumably via cytochrome P450-mediated * To whom correspondence should be addressed: Chemical Industry Institute of Toxicology, 6 Davis Dr., P.O. Box 12137, Research Triangle Park, NC 27709-2137. Phone: (919) 558-1343. Fax: (919) 558-1300. E-mail:
[email protected]. † Chemical Industry Institute of Toxicology. ‡ National Institute of Environmental Health Sciences. § National Cancer Institute. 1 Abbreviations: AM, acrylamide; AN, acrylonitrile; CEO, cyanoethylene oxide; GA, glycidamide; GSH, glutathione; ABT, 1-aminobenzotriazole; P450 2E1, cytochrome P450 2E1; WT, wild-type mice; P450 2E1-null, mice devoid of P450 2E1; ABT WT, WT mice pretreated with ABT.
oxidation of the parent compounds, have been detected in blood and brain of rats treated with AN (13) and in blood of rats treated with AM.2 CEO and GA are believed to be involved in the carcinogenic activity of the parent compounds since both epoxides react with DNA (14-18) and are mutagenic (19-21). The mechanisms behind the neurotoxic effects in AM-exposed rodents are not well understood but are believed to be derived from both AM and GA (22). Rodents exposed to AN and AM excrete metabolites (Scheme 1) in urine that are derived from GSH conjugation at the 2- and 3-carbons of CEO and GA (7-9, 11, 23-25). Rodents treated with AM also excrete GA and the hydrolysis product of GA in the urine (7). In vitro studies with human and rat hepatic microsomes have indicated that cytochrome P450 2E1 (P450 2E1) is the major catalyst of AN epoxidation and that other cytochrome P450 isozymes can also metabolize AN (23, 26). Studies have not been conducted to evaluate the role of cytochrome P450 in the conversion of AM to GA (22). The purpose of the study presented here was to evaluate the role of P450 2E1 in the in vivo metabolism of AM and AN. 13C NMR was used to detect and quantitate metabolites in urine of WT mice and mice devoid of P450 2E1 (P450 2E1-null) following administration (po) of [1,2,3-13C]AN or [1,2,3-13C]AM. The me2 S.-Y. Cheng, T. R. Fennell, C. B. Brown, and S. C. J. Sumner (1996) unpublished results.
10.1021/tx990040k CCC: $18.00 © 1999 American Chemical Society Published on Web 10/28/1999
Cytochrome P450 and Metabolism of AM and AN
Chem. Res. Toxicol., Vol. 12, No. 11, 1999 1111
Scheme 1. Proposed Metabolism of AM and AN in Rats and Micea
a Processes that should involve several transformations are represented by broken arrows. N-AcCys, N-acetylcysteine; Cys, cysteine; GS, glutathione.
tabolism of [1,2,3-13C]AM was also evaluated in WT mice pretreated with 1-aminobenzotriazole (ABT). The metabolism either through direct GSH conjugation with the parent compounds or via oxidation to epoxides (Scheme 1) was compared to evaluate differences in the flux through these pathways in mice treated with AM or AN. Understanding the role of cytochrome P450 enzymes and other pathways of metabolism can provide valuable data for evaluating species differences in sensitivity to toxicity and carcinogenicity.
Materials and Methods Chemicals. [1,2,3-13C]Acrylamide was purchased from Isotec Inc. (Miamisburg, OH). [1,2,3-13C]Acrylonitrile was obtained from Cambridge Isotope Ltd. (Cambridge, MA). The 1H and 13C NMR spectra of [13C]AN (CDCl3) and [13C]AM (D2O) were consistent with those previously assigned (7-9). The chemical
purity of [13C]AM was 97.6%, and the isotopic enrichment was 99.5%. The chemical purity of [13C]AN was >98%, and the isotopic enrichment was 99%. AN was stabilized with 0.1% hydroquinone monomethyl ether. This small percentage of stabilizer is not expected to affect the metabolism or distribution of AN or cause toxicity. Acrylamide (CAS registry no. 79-06-1), acrylonitrile (CAS registry no. 107-13-1), and 1,4-dioxane (HPLC grade, CAS registry no. 123-91-1) were purchased from Aldrich Chemical Co. (Milwaukee, WI) and had >99% chemical purities. 1-Aminobenzotriazole (99%) was purchased from Sigma Chemical Co. (St. Louis, MO). Animals. Male and female wild-type (WT) mice and mice deficient of cytochrome P450 2E1 (P450 2E1-null; 26-29 g) of mixed background (C57BL/6N X Sv129) were obtained from a colony developed at the National Cancer Institute (27). The mice were quarantined at the National Institute of Environmental Health Sciences Animal Facility for a minimum of 2 weeks before being used in this study. They were maintained on a 12 h light-dark cycle with constant temperature (21-23 °C) and
1112
Chem. Res. Toxicol., Vol. 12, No. 11, 1999
humidity (40-60%). National Institutes of Health 31 rodent chow and tap water were provided ad libitum. All animal care and procedures were according to the National Institutes of Health guidelines (28). Dosing and Sample Collection. For the AM study, three WT male mice and three P450 2E1-null male mice were treated with ∼50 mg of [1,2,3-13C]AM/kg of body weight (0.67 mmol/ kg) by gavage (po). Three additional WT male mice were treated with ∼50 mg of [1,2,3-13C]AM/kg of body weight po approximately 2 h after an ip injection of ABT (50 mg/kg in 2.5 mL of saline/kg). 1-ABT has been used as a general inhibitor of cytochrome P450 enzymes (29). For the AN study, three male WT mice and four male P450 2E1-null mice were treated with ∼10 mg of [1,2,3-13C]AN/kg of body weight po (0.18 mmol/kg). In addition, three male WT mice and three male P450 2E1-null mice were treated with ∼2.5 mg of [1,2,3-13C]AN/kg of body weight po (0.045 mmol/kg). For all dosing solutions, the AM and AN were prepared in distilled water and administered at a volume of 10.0 mL/kg of body weight (0.27 mL/mouse). All solutions were administered within 1 h of preparation, and an aliquot of each dosing solution was analyzed by 1H and 13C NMR spectroscopy to confirm stability. The animals were placed in plastic metabolism cages following administration of the labeled material, and urine was collected for 24 h. Within 24 h of administration of [14C]AM (6, 30), [13C]AM (7), [14C]AN (12), or [13C]AN (9) to rats and mice, ∼50-70% of the dose was excreted in urine. The majority of the dose of [2,3-14C]AN administered (po) to rats (73-100%) and mice (83-94%) was recovered in urine within 72 h of administration (12). Rats treated with [2,3-14C]AN po exhale ∼2% of the dose as 14CO2 (31). In rats treated with [2,3-14C]AM (po or iv), approximately 90% of the radioacitivty recovered in excreta for up to 7 days was found in urine and no 14CO2 (iv study only) was recovered (6). The volume of urine recovered over a 24 h period ranged from ∼140 to 1400 µL. Overall, the urine volumes were lower than those previously collected (∼2 mL) from mice treated with [13C]AM or [13C]AN and placed in glass metabolism cages (7-9). The low recovery of urine and the variability in urine volume in the current study are attributed to the use of plastic metabolism cages. NMR Spectroscopy. Most urine samples were prepared for NMR analysis by adding ∼20% D2O to an aliquot of the urine. For small urine volumes, up to 70% D2O was added to provide a sufficient volume (∼400 µL total) for NMR analysis. NMR spectra were acquired with a 5 mm dual proton-multinuclear probe on a Varian VXR-300 spectrometer (Palo Alto, CA). 1Hdecoupled 13C NMR spectra were acquired in the doubleprecision mode with an acquisition time of 0.9 s, 30K data points, and a relaxation delay of 10 s. Samples were prepared for quantitation of metabolites by adding a known volume of dioxane to the urine aliquot. In previous studies (7-9) with AM and AN, relaxation times (T1) for dioxane and all AM and AN (except AM and SCN-)derived urinary products were determined to be e5 s. A quantitative method was established (7-9) that enabled the calculation of metabolite concentration based on the concentration of the internal standard (dioxane) and relative signal intensities (taking into account the percentage of 13C enrichment and the number of carbons that give rise to each signal). AM and SCN- were not quantitated due to the long carbon T1 relaxation times. Quantitative values for metabolites of AM and AN are presented as a percentage of total AM or AN excreted urinary metabolites and as a percentage of the AM or AN administered dose. The lowest values were quantitated to ∼0.1 mM, which is near the quantiation limit for a 13C-labeled compound run under the conditions used in this study.
Results Structural Assignments for [13C]AM Metabolites. The 1H-decoupled 13C NMR spectra of urine collected
Sumner et al.
Figure 1. (A) 1H-decoupled 13C NMR spectrum of urine collected from a WT mouse treated with 50 mg/kg [1,2,3-13C]AM and expanded regions of the 1H-decoupled 13C NMR spectra of urine from (B) a WT mouse, (C) an ABT-pretreated WT mouse, and (D) a P450 2E1-null mouse, which were all treated with 50 mg/kg [1,2,3-13C]AM. Signals for metabolites of [1,2,313C]AM are labeled in panel A. Each signal is labeled with a metabolite number (Scheme 1) and a letter designating the carbon derived from acrylamide (aCH2dbCHcCO2H).
from a WT mouse, a P450 2E1-null mouse, and an ABTpretreated WT mouse treated po with 50 mg of [1,2,313C]AM/kg of body weight are shown in Figure 1. Signals for metabolites of [1,2,3-13C]AM are recognized by coupling patterns that are produced by spin-spin interactions between the labeled carbons. Compounds endogenous in urine give rise to singlet patterns due to the low occurrence (1.1%) of adjacent 13C nuclei. The chemical shifts and coupling constants of signals (Table 1 and Figure 1) in the 1H-decoupled 13C NMR spectra of urine from WT mice are consistent with those previously assigned (7, 8) following administration of [1,2,3-13C]AM to F344 rats or B6C3F1 mice. The assignments include (Scheme 1 and Figure 1) N-acetyl-S-(3-amino-3-oxopropyl)cysteine (metabolite 1), S-(3-amino-3-oxopropyl)cysteine (metabolite 1′), diastereomers of N-acetyl-S-(3amino-2-hydroxy-3-oxopropyl)cysteine (metabolites 2 and 2′) and N-acetyl-S-(1-carbamoyl-2-hydroxyethyl)cysteine (metabolites 3 and 3′), GA (metabolite 4), and 2,3dihydroxypropionamide (metabolite 5). Additional signals (5′, ABT-pretreated P450 2E1-null mice, while the percentage recovered in expired air (volatiles and CO2) was ABT-pretreated P450 2E1-null mice > P450 2E1-null mice ) WT mice. These data may also indicate the possibility of ABT pretreatment altering GSH content or GS-transferase activity, since metabolites derived from GSH conjugation would be excreted in the urine or feces. In the study presented here, the largest percentage of metabolites from WT mice treated with the high or low doses of [13C]AN were derived from conjugation of GSH with the 3-carbon of CEO. Metabolites derived from direct GSH conjugation with AN and from GSH conjugation with the 2-carbon of CEO were also present. P450 2E1-null mice treated with the high or low dose of [13C]AN excreted only metabolites derived from direct GSH conjugation with AN. No metabolites were detected that could be produced via oxidative metabolism of AN to CEO. For the high- and low-dose regimens, the WT and P450 2E1-null mice excreted an approximately similar percentage of the administered dose. Analogous to the observations for the [13C]AM study, these data indicate that the P450 2E1-null mice compensated for the P450 2E1 deficiency by producing more metabolites from direct conjugation of AN with GSH. These results differ from previous research (37) with WT and P450 2E1-null mice treated with [14C]MAN (12 mg/kg po). That study (37) showed that P450 2E1-null mice had a significantly lower percentage of the dose, compared with WT mice, excreted as an epoxide-derived metabolite of [14C]MAN in P450 2E1-null mice. A further decrease was noted for P450 2E1-null mice pretreated with ABT. The decrease in the percentage of epoxide-derived metabolites was compensated only in part by an increase in the extent of direct
GSH-MAN-derived metabolism. These data suggest that cytochrome P450 enzymes other than P450 2E1 may be involved in the oxidative metabolism of MAN. Previous investigations in which 50 mg/kg [1,2,3-13C]AM (7) or 10 mg/kg [1,2,3-13C]AN (9) was administered to B6C3F1 mice resulted in the urinary excretion of ∼5055% of the administered dose within 24 h after administration. The lower percentage of dose recovered in the current study may be related to differences in the distribution of AM and AN in the C57BL/6N mice (WT mice used in this study) or may be related to the low urine volumes recovered from the plastic metabolism cages (
