Chem. Res. Toxicol. 1998, 11, 1137-1144
1137
1,3-Dichloropropene Epoxides: Intermediates in Bioactivation of the Promutagen 1,3-Dichloropropene Manfred Schneider,† Gary B. Quistad, and John E. Casida* Environmental Chemistry and Toxicology Laboratory, Department of Environmental Science, Policy and Management, University of California, Berkeley, California 94720-3112 Received June 1, 1998
1,3-Dichloropropene (1,3-D), a major soil fumigant nematicide, is genotoxic in many types of assays, leading to its classification as possibly carcinogenic in humans. This study tests in three steps the hypothesis that 1,3-D is a promutagen activated by epoxidation and further reaction of the 1,3-D-epoxides. Stereospecific epoxidation of 1,3-D (examined as the cis/trans mixture and as individual isomers) to the corresponding cis- and trans-1,3-D-epoxides is demonstrated here for the first time, both in vitro in a mouse liver microsome-NADPH system and in vivo in the liver of ip-treated mice, using GC/MS for product identification and quantitation. The cis epoxide is observed in higher yield than the trans epoxide, both in vitro and in vivo, and the cis isomer also reacts slower than the trans isomer with GSH alone or catalyzed by GSH S-transferase. cis- and trans-1,3-D-Epoxides are stable in acetone or chloroform but degrade completely in Me2SO exclusively to 2-chloroacrolein (30 min at 40 °C). Epoxide decomposition is slower in pH 7.4 phosphate buffer (t1/2 ) 116 and 64 min for cis and trans, respectively, at 41 °C) with a >99% yield of 3-chloro-2-hydroxypropanal (and its dimer) and 95% cis) is used in The Netherlands as an improved formulation (5). 1,3-D is classified as possibly carcinogenic in humans on the basis of extensive genotoxicity studies by the National Toxicology Program (2). It induces sister chromatid exchanges with and without metabolic activation * To whom correspondence should be addressed. Telephone: (510) 642-5424. Fax: (510) 642-6497. E-mail:
[email protected]. † Present address: Zeneca Ag Products, Western Research Center, 1200 S. 47th St., Richmond, CA 94804. 1 Abbreviations: CI, chemical ionization; 1,3-D, 1,3-dichloropropene; 1,3-D-epoxides, 1,3-dichloropropene epoxides; GST, GSH S-transferase; imp, solvent impurity; is, internal standard which is chlorobenzene; MG, methylglyoxal; mEH, rat liver microsomal epoxide hydrolase; RRT, relative retention time or kcompound/kis, where the capacity factor k is (tR - tO)/tO and tO is the gas holdup time; S9 fraction, postmitochondrial supernatant fraction; sEH, rat liver soluble epoxide hydrolase; SIM, selected ion monitoring.
(6), induces DNA fragmentation in rat liver, gastric mucosa, and kidney (7), and is genotoxic and cytotoxic in cultured rodent and human cells, including metabolically deficient V79 Chinese hamster lung cells (8). There is also a possible relationship between human exposure to 1,3-D and the development of hematologic malignancies (9). Three primary pathways are proposed for 1,3-D metabolism in mammals. The first is cytochrome P450dependent epoxidation to cis- and trans-1,3-dichloropropene epoxides (cis- and trans-1,3-D-epoxides) (Figure 1), consistent with a common pathway for chloroalkenes (10, 11) but only recently demonstrated with 1,3-D (12). The second is P450 oxidation to 3-chloroacrolein (13). The third involves GSH conjugation at the chloromethyl moiety leading to excretion of mercapturate metabolites of 1,3-D in human urine (11, 14). The GSH/GSH Stransferase (GST) conjugation pathway is the major elimination route in mammals (2, and references therein). 1,3-D-containing pesticides appeared to be direct-acting mutagens in the TA100 and TA1535 strains of Salmonella typhimurium (15), but this was due in TA100 to polar impurities (16) subsequently identified as cis- and trans-1,3-D-epoxides formed by autoxidation in equal amounts (17). The 1,3-D-epoxide impurities must be removed prior to studies of intrinsic genotoxicity and
10.1021/tx980126y CCC: $15.00 © 1998 American Chemical Society Published on Web 09/02/1998
1138 Chem. Res. Toxicol., Vol. 11, No. 10, 1998
Schneider et al.
Figure 1. Metabolic epoxidation of 1,3-D isomers and hydrolysis of the epoxides to 3-chloro-2-hydroxypropanal proposed to be the principal bioactivation sequence. The other dehydrochlorination, rearrangement, and dimerization reactions illustrated are less important or not relevant under biological conditions. Major conjugation reactions of 1,3-D with GSH, mostly leading to detoxification, are not shown. Reaction conditions were as follows: (a) cytochrome P450, (b) hydrolysis possibly by epoxide hydrolase, (c) reversible dimerization (at high concentrations), (d) dehydrochlorination but not in pH 7.4 buffer, and (e) rearrangement and dehydrochlorination in Me2SO but not in pH 7.4 buffer.
metabolic activation and other toxicological investigations. Purified 1,3-D is not active in TA100 except on P450 activation by rat liver postmitochondrial supernatant (S9 fraction) or microsomes with the putative mutagenic metabolite(s) proposed to be 1,3-D-epoxide (17) or 3-chloroacrolein (13). cis- and trans-1,3-D-Epoxides do not require metabolic activation for mutagenic activity in S. typhimurium TA100 (17) and TA1535 where the cis isomer is twice as potent as the trans isomer (10). The genotoxicity of cis- and trans-1,3-D-epoxides is also evident from neoplastic transformations and increased cell lethality in Syrian hamster cells (18) and the induction of squamous carcinomas in mouse skin (19). This study focuses on the formation and toxicological relevance of cis- and trans-1,3-D-epoxides from three standpoints: direct evidence for stereoselective epoxidation of 1,3-D in vitro and in vivo using the mouse as a mammalian model, the decomposition of 1,3-D-epoxides with special emphasis on possible formation of 2-chloroacrolein (a potent mutagen; 20) on rearrangement in Me2SO and 3-chloro-2-hydroxypropanal in aqueous medium, and the mutagenic activity in S. typhimurium TA100.
Materials and Methods Caution: 1,3-D, 1,3-D-epoxides, 2-chloroacrolein, 3-chloroacrolein, and methylglyoxal are hazardous chemicals and must be used under containment conditions. Spectroscopy and Chromatography. NMR spectra were determined with a Bruker AM300 spectrometer at 300 and 75 MHz for 1H and 13C, respectively. GC/MS measurements were
performed as follows: Hewlett-Packard 5890 gas chromatograph coupled with a 5971A mass spectrometer in the chemical ionization (CI) mode using methane; HP-1701 capillary column (30 m × 0.32 mm i.d., 0.25 µm film thickness); helium carrier gas at a constant rate of 33 cm/s; inlet at 250 °C and oven program from 40 to 250 °C at 20 °C/min. CI/MS selected ion monitoring (SIM) was based on the most abundant masses, including the chlorine isotope ratios, i.e., [MH]+ and [M - Cl]+. All quantitative analyses were in the linear range (r >0.99) of the MS detector versus chlorobenzene as an internal standard (is). These GC conditions gave tR values in minutes (and relative retention times or RRT) of 3.71 (0.684) for cis-1,3-D, 4.08 (0.806) for trans-1,3-D, 4.72 (1.000) for the internal standard, 5.23 (1.166) for cis-1,3-D-epoxide, and 5.63 (1.284) for trans-1,3-Depoxide. Chemicals. (A) 1,3-D and 1,3-D-Epoxides. 1,3-D from Pfaltz & Bauer (Waterbury, CT) was 92% pure (1:1 cis:trans determined by GC/MS). MS: m/z 111, 113, 115 [MH]+, 75, 77 [M - Cl]+. This sample contained 0.2% cis- and 0.2% trans1,3-D-epoxides (GC/MS). cis-1,3-D (99% pure) was from Supelco (Bellefonte, PA) and trans-1,3-D (>96% pure) from TCI America (Portland, OR). Oxidation of 1,3-D preparations with 3-chloroperoxybenzoic acid (10) was used to prepare cis/trans-1,3-Depoxides (>96% pure by 1H NMR, 1:1 cis:trans by GC/MS and 1H NMR), cis-1,3-D-epoxide (80% cis, 6% trans; prepared from 80-90% cis-1,3-D), and trans-1,3-D-epoxide (97% pure by 1H NMR). MS data were identical for cis- and trans-1,3-Depoxides: m/z (relative intensity) 127, 129, 131 (53, 36, 6 [MH]+), 91, 93 (100, 33 [M - Cl]+). 1H NMR (CDCl3): cis-1,3-D-epoxide, δ 3.79 (2H, m), 3.33 (1H, m), 5.20 (1H, d); trans-1,3-D-epoxide, δ 3.60 (2H, m), 3.42 (1H, m), 5.01 (1H, d). 13C NMR (acetoned6): cis-1,3-D-epoxide, δ 69.97 (ClCO), 55.83 (CCH2Cl), 42.21 (CH2Cl); trans-1,3-D-epoxide, δ 68.08 (ClCO), 59.48 (CCH2Cl),
Bioactivation of 1,3-Dichloropropene 42.59 (CH2Cl). The epoxides are stable without solvent for months at 4 °C and in acetone or chloroform for several days at 25 °C. (B) 2-Chloroacrolein. 2-Chloroacrolein can be synthesized by chlorination of acrolein and then dehydrochlorination of 2,3dichloropropanal with diethylaniline (21, 22). We find that 2-chloroacrolein is conveniently prepared by stirring a solution of the 1,3-D-epoxides in Me2SO (5:1, v/v) for 60 min at 40 °C (the reaction is complete within 30 min) and then distillation (60 mm/50 °C) (60% yield, 92% purity by 1H NMR). MS: m/z (relative intensity) 91, 93 (100, 33 [MH]+). 1H NMR (Me2SOd6): δ 9.42 (1H, s), 6.58 (1H, d), 6.40 (1H, d) (see also ref 23). 13C NMR (Me SO): δ 186.78 (CHO), 140.10 (CCl), 134.81 (dCH ). 2 2 Me2SO and traces of dimethyl sulfide are the only impurities. (C) 3-Chloroacrolein. The procedure of Basu and Marnett (24) was used to prepare an ether solution of 3-chloroacrolein (20% cis, 80% trans as determined by GC/MS; same as the reported ratio). MS data were identical for cis- and trans-3chloroacrolein (and 2-chloroacrolein): m/z (relative intensity) 91, 93 (100, 33 [MH]+). The RRTs compared with the internal standard on the HP-1701 column are 0.592, 0.635, and 0.654 for 2-chloroacrolein and cis- and trans-3-chloroacrolein, respectively; i.e., they can be easily separated for analysis. (D) Methylglyoxal. The 40% (w/w) aqueous solution of methylglyoxal (MG) (Aldrich, Milwaukee, WI) was 92% pure with ∼1:1 monohydrate and dihydrate (25) and 8% pyruvic acid (1H NMR). 13C NMR (pH 7.4, 100 mM phosphate, trace of the Me2SO standard): MG monohydrate, δ 210.61 (C-2), 91.55 (C-1), 23.59 (C-3); MG dihydrate, δ 93.98 (C-1), 83.64 (C-2), 26.22 (C-3). Metabolic Epoxidation of 1,3-D. Male albino SwissWebster mice (22-27 g) were obtained from Charles River Laboratories (Wilmington, MA). (A) In Vitro. Mouse liver microsomes (2.6 mg of protein) (26) were incubated with 1,3-D [0.1 mM cis or trans or a 0.2 mM cis/trans mixture; added in Me2SO (5 µL)], NADPH (0 or 1 mM), and GSH (0 or 5 mM) in 100 mM phosphate buffer (pH 7.4, 505 µL) for 15 min at 37 °C. The incubated mixtures were spiked with 2 µg of internal standard (in 20 µL of ethyl acetate) and immediately extracted with ethyl acetate (500 µL). Two extractions led to the recovery of >95% of the internal standard, and a third extraction did not produce additional 1,3-D-epoxides. The extracts were combined and dried over sodium sulfate. Aliquots (2 µL) were analyzed (GC/MS) for cis/trans-1,3-D and cis/trans-1,3-D-epoxides. (B) In Vivo. The 1,3-D-epoxides were analyzed in liver at various times up to 150 min after ip treatment with 1,3-D (350 mg/kg for individual isomers or 700 mg/kg for cis/trans-1,3-D). The carrier solvent was Me2SO, and 1,3-D solutions were held for at least 4 h at 25 °C before the experiment to degrade any epoxide impurities that might be present (see above). Some mice were pretreated with metribuzin (300 mg/kg), a potent GSH depletor (27), 90 min before 1,3-D administration. The liver was homogenized in 100 mM sodium phosphate buffer (pH 7.4, 2 mL) containing ∼0.2 g of NaCl, and then ethyl acetate (2 mL) containing 2 µg of internal standard was added and the mixture homogenized and centrifuged at 10000g for 30 min. The organic layer was removed, and the pellet was extracted again with ethyl acetate but without internal standard. Recoveries for cis/trans-1,3-D and cis/trans-1,3-D-epoxides were 81-95%, and a third extraction with ethyl acetate did not increase the recoveries; hexane or dichloromethane for extraction gave poor recoveries. The ethyl acetate extracts were analyzed as above. Reactions of 1,3-D-Epoxides. (A) Products in Me2SO. The conversion of the 1,3-D-epoxides in Me2SO-d6 to 2,3dichloropropanal and 2-chloroacrolein was monitored by 1H NMR at 22 °C by integration of the epoxide proton at the terminal carbon, which gives a well-isolated doublet for both epoxide isomers. The rate constants were determined for degradation of the epoxides to 2,3-dichloropropanal. 13C NMR (Me2SO): δ 193.55 (C-1), 63.01 (C-2), 43.75 (C-3).
Chem. Res. Toxicol., Vol. 11, No. 10, 1998 1139 (B) Products in pH 7.4 Buffer. A mixture of the 1,3-Depoxides (200 mg) and 100 mM phosphate buffer (pH 7.4, D2O, 500 µL) was kept for up to 7 days at 25 °C. At 3 days, the epoxide had completely reacted and dissolved and no further 13C NMR spectral changes were observed. (C) Reaction Rates with GSH and GST. cis/trans-1,3-DEpoxides or 2-chloroacrolein (0.1 mM final concentration) added in acetone (50 µL) to 100 mM phosphate buffer (pH 7.4, 1 mL) with or without GSH (5 mM) and rat liver GST (200 µg of protein; Sigma, St. Louis, MO) were incubated for 1, 30, 60, 90, 200, 325, and 425 min at 41 °C. They were analyzed by GC/ MS using 25 µL aliquots added to 5 mL of acetone containing 1 µg of internal standard. Rate constants were calculated assuming first-order reaction kinetics (r > 0.97). (D) Interaction with Epoxide Hydrolase. Rat liver microsomal epoxide hydrolase (mEH) and soluble epoxide hydrolase (sEH) were expressed in insect cells using recombinant baculovirus (28) and assayed with selective substrates (50 µM): [3H]-cis-stilbene oxide for mEH in 100 mM Tris-HCl buffer (pH 9, 100 µL) and [3H]-trans-stilbene oxide for sEH in 100 mM phosphate buffer (pH 7.4, 100 µL) (29). Bovine serum albumin (10 µg) was present in the buffer to control nonspecific reactions, and enzyme levels were adjusted (by dilution of the crude extract) to 2.1 (mEH) and 0.9 (sEH) nmol of diol formed min-1 mL-1, providing a linear relationship between the enzyme level and diol yield in the standard 5 min assay. The potential inhibitory effect of cis/trans-1,3-D-epoxides (1 mM final concentration) was then examined as a function of preincubation time (1-20 min) before [3H]stilbene oxide addition. Mutagenicity of 1,3-D-Epoxides. S. typhimurium TA100, which detects mutagens that cause base pair substitutions, was used by the standard plate incorporation method, including all required controls (30). cis- and trans-1,3-D-Epoxides were added in acetone (100 µL) and 2-chloroacrolein in Me2SO (100 µL) using a 2-fold dilution series. In some tests, washed microsomes like those described above (0.64 mg of protein) and NADPH (1.5 mg) were added, each in 100 µL of 100 mM phosphate buffer (pH 7.4). cis/trans-Diallate (40 nmol) (from Chem Service, West Chester, PA) was used as a positive control in this activation assay, giving 0 or 1000 revertants assayed directly or with microsomes and NADPH, equivalent to about ∼15% conversion of diallate to 2-chloroacrolein (20). The effect of GSH and/or GST on the mutagenic activity of cis- and trans-1,3-D-epoxides and 2-chloroacrolein was tested by addition of rat liver GST as described above (10 µg in 100 µL of buffer) and varying levels of GSH in buffer (100 µL). The spontaneous revertants (averaging 180 per plate) were subtracted. Each assay dose involved duplicate plates, and each experiment was carried out three times. The mutagenicity values were determined from the slopes of the linear dose-response range and are given in revertants per nanomole of compound with standard deviations.
Results Metabolic Epoxidation of 1,3-D. (A) Identification of Epoxide Metabolites. cis- and trans-1,3-Depoxides are identified as metabolites of the respective 1,3-D isomers in mice both in vitro (microsomes and NADPH) and in vivo (liver) by GC/MS (Figure 2). The structural assignments are based on (1) identical tR and RRT values compared with standards, (2) detection of both [MH]+ and [M - Cl]+, and (3) the expected isotope ratios for two chlorines (m/z 127, 129, 131 in a 9:6:1 ratio for [MH]+ and one chlorine m/z 91, 93 in a 3:1 ratio for [M - Cl]+). 2-Chloroacrolein and cis- and trans-3chloroacrolein (easily separated from the other compounds) were not detected as in vitro or in vivo metabolites of cis- and trans-1,3-D. (B) Liver Microsomes. Microsomal metabolism of cis- and trans-1,3-D examined as individual isomers is enhanced by addition of NADPH and further enhanced
1140 Chem. Res. Toxicol., Vol. 11, No. 10, 1998
Figure 2. Detection of cis- and trans-1,3-D-epoxides as metabolites of cis/trans-1,3-D (1:1 ratio) in vitro in the mouse liver microsome-NADPH system and in vivo in mouse liver (150 min after treatment). Analyses were by GC/MS.
Schneider et al.
Figure 3. Stereospecific epoxidation of cis-1,3-D in vitro in the mouse microsome-NADPH system and trans-1,3-D in vivo in mouse liver (4 min after treatment). The same stereospecificity is observed for trans-1,3-D in vitro and cis-1,3-D in vivo. Analyses were as described in the legend of Figure 2.
Table 1. Mouse Liver Microsomal Metabolism of cis/ trans-1,3-D and the Effect of NADPH and GSH recovery (%) with the indicated addition compound cis-1,3-D parent epoxide acroleinsc trans-1,3-D parent epoxide acroleinsc
none
NADPH
NADPH and GSH
100a
