Document 3J3Lk3Grr048BBmXgnQbZ4xvO

V ' * k I TOXICOLOGY AND APPLIED PHARMACOLOGY 36, 339-352 (1976) Fate of [HC]Vinyl Chloride after Single Oral Administration in Rats1 P. G. Watanabe, G. R. McGowan, and P. J. Gehring Toxicology Research Laboratory, and Analytical Laboratory, The Dow Chemical Company, Midland, Michigan 48640 Received September 9, 1975; accepted January 9, 1976 Fate of [uC]Vinyl Chloride after Single Oral Administration in Rats. Watanabe, P. G., McGowan, G. R., and Gehring, P. J. (1976). Toxicol. Appl. Pharmacol. 36,339-352. Male rats were given single oral doses of0.05, 1, and 100 mg/kg of ['`Cjvinyl chloride (VC), and the routes and rates of elimination of MC activity followed for 72 hr. Following 0.05 and 1 mg/kg, excretion in the urine as nonvolatile metabolites and as MCOj in expired air accounted for 59-68% and 9-13%, respectively of the administered dose. Only 1-2% of the dose was expired by the lungs as VC. Conversely, after 100 mg/kg, 67% of the dose was eliminated by the lungs as VC, while urinary nonvolatile metabolites and UC02 comprised 11 and 3%, respectively. Pulmonary elimination after 100 mg/kg showed an apparent biphasic clearance with half-times (tul) of 14.4 and 40.8 min for the respec tive fast and slow phases. Following 0.05 and 1 mg/kg the pulmonary clearance of VC was monophasic with tm of 53.3 and 57.8 min. The percentage of the dose remaining in the carcass after 72 hr was 10,11, and 2% for the 0.05-, 1- and 100-mg/kg doses, respectively. The urinary radio activity was separated by high pressure liquid chromatography into three major metabolites. Two of the three major urinary metabolites have been identified as A'-acetyl-5'-(2-hydroxyethyl)-cysteine and thiodiglycolic acid by gas chromatography-mass spectrometry. The proportions of the urinary metabolites were not influenced by the dose. The fate of VC following an oral dose between 1 and 100 mg/kg was clearly dose-dependent. Con sistent with our previous studies on the fate of VC following inhalation ex posure in rats, the metabolism of VC appears to be a saturable process. The hazard of industrial exposure to vinyl chloride (VC) during the production of polyvinyl chloride and other plastics has received considerable attention. Maitoni and Lefcmine (1974) demonstrated the carcinogenic potential of VC in rats exposed daily by inhalation to concentrations ranging from 50-10,000 ppm. Subsequent epidemio logic data obtained from industrial workers with long-term exposure to high conccntrations of VC demonstrated an association between exposure and various hepatic abnormalities including induction of angiosarcoma (Creech and Johnson, 1974). The fate of inhaled VC has been studied previously in rats (Hefner et ai, 1975). Although some aspects of the fate of inhaled VC remain to be elucidated, the data indicated that in rats exposed to 50 ppm for 1 hr the inhaled VC was metabolized to polar products which were excreted predominantly in the urine. Using kinetic para meters and inhibitors of drug metabolism, additional evidence indicated that the fate 1 This study was funded by the companies supporting the vinyl chloride projects being administered by the Manufacturing Chemists Association, Washington, D.C. Copyright O 1976 by Academic Press, Inc. All rights of reproduction in any form reserved. Primed in Great Britain 339 R&S155631 340 WATANABE, MCGOWAN AND GEHRING of inhaled VC was concentration-dependent and that at least two metabolic pathways were involved in its metabolism. Consideration of these data led to the speculation that the carcinogenic activity of VC may be mediated through formation of alkylating meta bolites such as chloroacetaldehyde and chloroethylene oxide. Even more important is the fact that the production ofcarcinogenic metabolites may increase disproportionately with the extent of exposure; or conversely, the capacity to detoxify alkylating metabol ites may decrease disproportionately with the extent of exposure. Resolution of these possibilities is critical in assessing the hazard of exposure to low levels of VC. The objective of this study was to determine the fate of orally administered VC in rats at various dose levels in order to provide data useful in evaluating the hazard of ingested VC. METHODS Compound. [1,2-uC] Vinyl chloride ([l4C] VC) was synthesized directly from [ 1,2-14CJ1,2-dichloroethane (lot no. 819-021, 3.4 mCi/mmol, New England Nuclear, Corp.) by the method of Wagner and Muelder (1975). The radiochemical purity of a representative sample of [I4C] VC, from this synthesis, has been reported to be 95-96% pure (Wagner et al,, 1975). The primary l4C-containing contaminant (4-5%) in the [UC]VC prepara tion was [14C]acetylene. Nonlabeled VC (Matheson Gas Products) of 99.9% minimum purity was mixed with the 14C material to obtain the desired specific activity. Preparation ofdose. The [14C] VC was synthesized immediately prior to use. Typically 20 ml of the [I4C]VC, helium mixture (approx. 2.6 mCi/mmo!) was bubbled directly into 15 g of USP corn oil in a sealed septum vial. An appropriate quantity of nonlabeled VC was then bubbled into the corn oil to obtain the desired concentration. One microliter of the corn oil-dosing solution was subjected to gas chromatography and the final concentration of VC in the corn oil was determined by comparison to standard gas samples. The VC was analyzed by gas chromatography (Hewlett-Packard Model 5750) on a Porapak Q column (80-100 mesh, 6 ft x i in. stainless steel) with a carrier gas (He) flow rate of 15 ml/min. The flame ionization detector, injection and column tempera tures were 280, 250, and I80C, respectively. The radioactivity of the dosing solution was determined by placing aliquots (0.0500. lOOg) into prefilled scintillation vials containing 20 ml scintillant, Concifluor (Mallinckrodt Chemical Works), 2-methoxyethanol;toluene, 6:11:83. The radio activity was determined in a Nuclear Chicago Mark II liquid scintillation spectrometer (Searle, Inc.). External standard channel ratios were used to determine counting efficien cy and the counts per minute (cpm) were converted to disintegrations per minute (dpm) with a standard quench curve. The specific activities for the 0.05-, I - and 100-mg/ kg dose solutions were 187.36, 20.50, and 0.18 /tCi/'mg VC, respectively. The VC-corn oil solution was administered by gavage with a glass syringe and stainless steel dosing needle in a volume not exceeding 5 ml/kg. Animals and apparatus. Male Sprague-Dawley rats weighing from 180-224 g pur chased from Spartan Research were used throughout the studies. All animals were fasted overnight, and the VC was administered between 8 and 10 am the following morning. R&S 155632 FATE OF [UC]VINYL CHLORIDE IN RATS 341 The rats were housed in glass Roth-type metabolism chambers designed for the separate collection of urine, feces, and expired air. Room air was drawn by vacuum through the chambers at 400-500 ml/min. The air leaving the chamber was passed through a series of traps to collect the expired [14C]VC and I4C02. The air leaving the chamber was first passed through a glass tube containing about 40 g of Drierite (W. A. Hammond Drierite Co.) to remove moisture. Subsequent transit through a series of two cold finger traps containing 50 ml of toluene, 2-methoxyethanol (80:20, immersed in Dry Ice baths) and a single trap containing 120 ml of 5m ethanolamine in 2-methoxyethanol (room temperature) enabled the collection of [14C]VC and 14C02 respectively. Experimental procedure. The animals were placed in the modified metabolism cages immediately after dosing. The VC traps were changed at 30-min intervals for the first 4 hr. Two subsequent changes at 4-hr intervals (8 and 12 hr) completed the collection of expired VC. The C02 trap, urine (immersed in a Dry Ice bath), and feces receptacles were changed at 12-hr intervals for 72 hr. At termination of thestudytheanimalsweredecapitated.exsanguinatedandsamplesof tissue (liver, lung, perirenal fat, muscle, plasma) were collected foranalysis of I4C activity. The remaining carcass was skinned and homogenized (50%, w/v) in distilled water. Sample preparation and assay. Aliquots of the VC traps (5 ml) were prepared for counting by addition of an equal volume of scintillating agent containing Concifluor (Mallinckrodt Chemical Works) 2-methoxyethanol, and toluene (6:11:33). Five-milliliter samples of the solution from the C02 trap were added to 5 ml of 5 m ethanolamine in 2-methoxyethanol and 10 ml of the scintillating agent described above. The urine samples, 250 p\, were prepared by adding 1 ml of distilled water and 12 ml of Aquasol (New England Nuclear). Aqueous homogenates (33 or 50%, w/v) of feces, tissues, and the remaining carcass were oxidized to C02 and H20 in a Biological Material Oxidizer (Beckman Instru ments), The 14C02 from the combustion was trapped in 8 ml of 5 m ethanolamine in 2-methoxyethanol and added to the Concifluor as described previously. Combustion of the samples of skin and fat was performed without homogenization. 14C activity in all samples was determined by scintillation counting. Isolation of urinary metabolites of VC by high pressure liquid chromatography (hplc). Urine (5 to 10 ml) containing 0.05 to 1.0 /iCi of 14C activity (ranging from 9-2000 pg VC equivalents) was lyophylized and the solid residue extracted once with 5 ml of methanol, once with 0.5 ml of water, and twice more with 5 ml of methanol. The solu tion was centrifuged after each extraction and the clear supernatants removed and combined. The extraction of 14C activity into the combined supernatants was 100 2 (SD)%. The combined supernatant fractions were evaporated to dryness under a stream of nitrogen at room temperature and the residue reconstituted in 1 ml of methanol. A Corasil 11 (37-50 /<m. Waters Associates, 2-mm i.d. x 50-cm glass) liquid chroma tography column was used to separate the 14C-containing urinary metabolites for both identification and routine quantitation. The pumping system used was a Waters Model 660 Solvent Programmer with two Waters Model 6000 Pumps. The flow rate was maintained at 2.0 ml/min while the solvent was programmed to form nonlinear gradient No. 7 from hexane:dioxane (7:1) to 2-propanol:methanol (3:2) over 20 min. The R&S155633 I 342 WATANABE, MCGOWAN AND GEHRING column temperature was ambient temperature, about 23C. Typically, from 10-75 pi of methanol solution containing from 1000-150,000 dpm of 14C activity were injected onto the column followed by a hexane:dioxane (7:1) wash for about 10 sec. The eluant was monitored at 254 or 280 nm using a Chromatronix 220 uv monitor and collected in 2-ml fractions. Aliquots (25-250 pi) of the fractions from the HPLC were combined with 10 ml of Aquasol (New England Nuclear) and the 14C activity determined by liquid scintillation counting. The fractions making up individual peaks of I4C activity were combined, evaporated to dryness under N2, and dissolved in a small volume of methanol. A 1-m Porasil B (250) column (Waters Associates) was also used for preparative scale work-up of urine samples for structure identification. The Porasil column was eluted with a linear gradient from hexane:chloroform (1:1) to methanol. The results were similar to those obtained from the Corasil II column except for a higher column capacity and poorer peak resolution. With both columns the recovery of 14C activity was quantitative and the columns could be reused numerous times before peak resolu tion decreased significantly. For further purification of Metabolite A additional HPLC was done using alumina. The column was prepared by packing a 2-mm i.d. x 1-m glass column with acidic alumina AG-4 (40 pm, Bio-Rad Laboratories). The column was repacked for each run. The packed column was washed with methanol until a stable baseline was indicated by the uv monitor. About 50-pg VC equivalents of Metabolite A from the initial Porasil separation were injected on the column and eluted with 6% concentrated aqueous NH*OH in methanol at a flow of 1 ml/min. The eluant was collected in fractions of 1- 2.5 ml and an aliquot counted as before. The fractions making up the peak of I4C activity were combined and evaporated to dryness under N2. Gas chromatography (gc). The collected eluant fractions from the initial hplc separa tion on Corasil II or Porasil B(250) were combined to give three fractions containing the three peaks of l4C activity. The first two major peaks, designated Metabolites A and B, were evaporated to dryness under N2, dissolved in methanol:diethylether (1:1), and methylated using diazomethane. They were then evaporated to dryness and dissolved in methanol to give a concentration of 0.5- to 1-pg VC equivalents and 500-1200 dpm of ,4C activity/ml. The derivatized fractions were chromatographed on one of two columns: (A) 6-ft. x 2- mm i.d. glass packed with 10% UCW-98 on 80/100 Gas Chrom Q, or(B) 6-ft. x 2-mm i.d. glass packed with 3% OV-210 on 80/100 Chromsorb 750. Column A was program med from 100 to 250C at 10C/min and column B was run at 125C isothermally. The outlet of the column was routed into an effluent splitter using a 5:1 split ratio. One part was fed into the flame ionization detector while five parts exited through a 1/8-in. o.d. stainless steel heated exit line to a fraction-trapping apparatus. A HewlettPackard 5750B gas chromatograph was used with the injection port and flame ionization detector maintained at 250 and 275C, respectively. The helium carrier gas flow rate was 35 ml/min. When fractions were trapped from the gc for counting of 14C-labeled metabolites, the glass capillary containing the condensed metabolite was washed into a scintillation vial using 2 ml of Aquasol, Eight additional milliliters of Aquasol were added to the vial for counting. R&S 155634 FATE OF (UC]VINYL CHLORIDE IN RATS 343 Mass spectroscopy (ms). Low resolution mass spectra were run on a Finnigan Model 3000D gc-ms operating at 70 keV using both direct insertion probe and gc inlets inter faced with the Model 6000 ms data system. When using the gc inlet the columns and gc conditioned were identical to those previously given. When using the direct insertion probe, samples (about 1 p% of metabolite) were placed in a quartz cup and the tempera ture of the probe was slowly raised from ambient to 250C. High resolution gc-ms were run on an AEI MS-30/DS-50 double beam mass spectro meter at Dow Corning Analytical Services. The resolution was 4300 and the mass measuring accuracy was generally within +0.005 mass units over the range of interest. The gc conditions were as previously described. Synthetic metabolite standards. Samples of W-acetyI-5'-(2-hydroxyethyl)-cysteine and thiodiglycolic acid were synthesized by N. Peet of Dow Lepetit. Pharmaceutical R & D. Samples of 14C-labeled jV-acetyI-S-(2-hydroxyethyl)-cysteine and thiodiglycolic acid were synthesized by D. Gransden of Dow Environmental Sciences Research. RESULTS Disposition of [14C] VC in Mats. Excretion of 14C activity within 72 hr following a single oral dose of 0.05, 1, and 100 mg/kg [,4C]VC is shown in Table 1. The percentage TABLE 1 Percentage of Administered i4C Activity Recovered Following a Single Oral Dose of Vinyl Chloride (VC)* 0.05 Dose (mg/kg) 1.0 100 Expired: As VC As COj Urine Feces Carcass and tissues Cage wash' Total recovery 1.4310.13" 8.96 0.59 68.34 0.54 2.39 0.52 10.13 1.93 0 91.25 + 2.47 2.13 0.22 13.26 0.47 59.30 + 2.75 2.20 + 0.39 11.10 + 0.47 0.84 + 0.45 88.83 + 1.98 66.64 + 0.67 2.52 + 0.13 10.84 + 0.95 0.47 + 0.06 1.83 + 0.14 0 82.30 0.43 * Percentage of dose excreted over 72 hr. Only the UC activity associated with the expired VC can be attributed to VC per se. * Mean SE five rats per dose. * Distilled water wash of metabolism cage at termination of the study. of the dose expired as VC perse was 1, 2, and 67%, respectively. Due to the dispropor tionate pulmonary elimination of VC, a greater percentage of the dose was metabo lized and eliminated in the urine, feces, and as expired 14C0, by rats eiven 0.05 and 1 mg/kg than by rats given 100 mg/kg. No special precautions were taken to insure detection of volatile compounds when collecting and processing the urine, feces, and carcass. Thus, the 14C activity in the excreta and carcass represent nonvolatile meta bolites of [14C]VC. The overall recovery of ,4C activity was 91.3, 88.8. and 82.3% at the 0.05-, 1-, and 100-mg/kg dose levels, respectively. The primary radiochemical contaminant in the [14C]VC preparation was 14C-labeled acetylene (4-5 %). Due to the 344 WATANABE, MCGOWAN AND GEHR1NG physicochemical properties of acetylene, the solvent cold traps used for the collection of expired VC would not trap the highly volatile [14C]acetylene. Based on this assump tion the total recovery of 14C activity due solely to VC would be slightly higher than expressed in Table 1. The characteristic pattern of pulmonary elimination of VC differed greatly between rats given 0.05 or I mg/kg than those given 100 mg/kg (Fig. 1). During the first 4 hr after administration of 100 mg/kg, the pulmonary elimination of VC was biphasic. The two linear portions of the curves were determined by regression analysis of the logarithmically transformed data. The data were feathered to obtain an approximation of the rate constant for the rapid phase of elimination. The apparent first-order rate Fir.. 1. Expired vinyl chloride expressed as percentage of the dose administered (0.05, 1, and 100 mg/kg) versus time (hr). Each point represents the mean SE of the mean of five rats. The linear phases of the curves were fit by linear regression analysis. constants for the rapid and slow phases were 0.048 0.005 and 0.017 + 0.008 min"1 (SD). These rate constants correspond to half-lives of 14.4 and 40.8 min. Following the two low doses of 0.05 and 1 mg/kg, pulmonary elimination of VC was monophasic with apparent first-order rate constants of 0.013 0.001 and 0.012 0.001 min"1 (SD) corresponding to half-lives of 53.3 and 57.8 min, respectively. To assure that the 14C activity collected in the cold traps was VC. the trapping solutions were analyzed by gas chromatography. Other than the components of the trapping solution the only compound detected had an identical retention time as a standard gas sample of VC. The elimination of 14C activity in the urine as a function of time after 1 and 100 mg/kg is shown in Fig. 2. The initial linear portions of the excretion curves from 12-36 hr were fit by regression analysis of the logarithmically transformed data. Similar biphasic elimination was evident at all dose levels. The curve for the 0.05-mg/kg dose was essentially identical to the 1-mg/kg dose and therefore was not graphically represented. R& S 155636 FATE OF [14C] VINYL CHLORIDE IN RATS 345 Fig. 2. MC activity excreted in the urine expressed as percentage of the dose administered (1 and 100 mg/kg) versus time (hr). Each point represents the mean SE of the mean for five rats. The initial linear segments of the curves (12-36 hr) were fit by linear regression analysis. For rats given 0.05, 1, and 100 mg/kg, respectively, estimates of the apparent first-order rate constants for the initial phase of elimination were 0.155 0.006, 0.150 + 0.020, and 0.152 0.011 (SD) hr-1. These correspond to half-lives of 4.5, 4.6, and 4.6 hr. The data for the secondary phase of the urinary excretion curves were extremely variable and since this phase accounted for less than 3 % of the total urinary radioactivity, no attempt was made to estimate the rate constants. The liver contained the highest concentration of 14C activity after 72 hr at all dose levels (Table 2). The concentration in the liver expressed on a percentage dose per gram TABLE 2 Percentage of the Administered mC Activity per Gram of Tissue after Administration of [14C]Vinyl Chloride' Dose (mg/kg) Tissue 0.05 1.0 100 Liver Skin Carcass Plasma Muscle Lung Fat 0.172 0.025` 0.070 0.023 0.027 0.007 0.041 +0.004 0.028 0.003 0.050 0.003 0.030 0.004 " Remaining in the body after 72 hr. * Mean SE, five rats per dose. c Not detectable above background. 0.182 0.005 0.076 0.010 0.046 0.002 0.053 0.007 0.031 0.003 0.061 0.003 0.045 0.008 0.029 0.002 0.010 0.002 0.007 + 0.001 NDf 0.006 0.001 0.011 0.001 0.006 + 0.001 R&S155637 346 WATANABE, MCGOWAN AND GEHRING tissue basis was three- to fivefold greater than muscle, lung, or fat. Consistent with the proportionally greater metabolism at the 0.05- and 1-mg/kg level, the proportion of the dose remaining in the tissues after 72 hr was considerably higher in rats given the low doses than those given 100 mg/kg. Isolation and identification of urinary metabolites of VC. Using hplc on a Corasil II column, methanol extracts of urine from rats given 0,05 to 100 mg/kg [14C]VC orally 100 so 60 -to 20 o Fig. 3. Separation of urinary metabolites by high pressure liquid chromatography (hplc) on a Corasil II column. The profile formed by the solid line shows the separation of radioactivity representing metabolites A, B, and C, by hplc versus time (min, flow rate = 2 ml/min). The dashed line depicts the nonlinear gradient profile used to effect the separation expressed as percentage solvent b. Solvent a, hexanetdioxane (7:1); solvent b, 2-propanol:methanol (3:2), were separated into three major peaks containing about 95 % of the applied 14C activity and several minor peaks. Figure 3 shows a chromatogram of a typical sample. The metabolites were designated by their elution order from the column as A, B, and C. The proportions of radioactivity determined by the three metabolites were not influ enced by the dose (Table 3). TABLE 3 Separation of 14C-Containing Urinary Metabolites from Rats Given Vinyl Chloride" Compound Dose (mg/kg) 0.05(4)* 1.0(5) 100(5) (A) jV-acetyI-S-(2-hydroxyethyl)-cysteine (B) Thiodiglycolic acid (C) Unidentified Total 30.4 + 2.0C 25.6+ 1.9 38.6+2.9 94.6 36.2 3.9 23.7+ 1.1 34.5 4.6 94.4 29.1 + 2.0 25.4 + 0.9 36.6 + 2.0 91.1 " Metabolites were separated and quantitated by high pressure liquid chromatography. Values arc expressed as percentage of total urinary radioactivity. * () = Number of animals per dose. c Mean + SE. R&S155638 FATE OF [14C] VINYL CHLORIDE IN RATS 347 Additional purification of the metabolite A fraction from the initial Corasil II or Porasil separation was carried out on an acidic alumina hplc column with a methanol: aqueous NH4OH eluant. The 14C activity eluted as a single peak containing over 90% of the applied radioactivity. A mass spectrum of the combined fraction was determined using a direct sample introduction probe. The sample was shown to have an apparent weak ion of mje = 189 and a prominent peak at mje= 130. Earlier work with 5-(2hydroxyethyl)-cysteine had shown that the highest mass peak found corresponded to the dehydrated molecular ion (M-18). If a similar dehydration had occurred in the metabolite, the mass spectrum obtained corresponded to that expected for 7V-acetyl-5-(2-hydroxyethyl)-cysteine (1). OO NH--C--CHj NH--C--CHj HO--CH2--CH2--S--CH2--CH--C02H M.W.= 207 (I) ------ CH3=CH--S -CH2--CH--COjH + H20 M. W, = 189 (2) The mass spectrum of a synthesized sample of A,'-acetyl-5'-(2-hydroxyethyl)-cysteine was found to be virtually identical to that of the material found in the Metabolite A fraction (Fig. 4). Fio, 4. Mass spectra of A^acetyl-S-^-hydroxyethyO-cysteine (top) and urinary Metabolite A (bot tom). All peaks mje greater than 140 were expanded by a factor of 10. Using a 10% UCW-98 column programmed from 120 to 250C, the methylated Metabolite A fraction was shown to contain a peak which had the identical retention time and mass spectrum as that of the methyl ester of the previously synthesized .Vacetyl-5-(2-hydroxyethyl)-cysteine standard. Additional confirmation of the identity of Metabolite A was obtained by coinjection of a synthetic IJC-labeled //-acetyl-5-(2-hydroxyethyl)-cysteine with the urinary' R&S 155639 348 WATANABE, MCGOWAN AND GEHRING metabolites using hplc. It was found that the added l*C activity coeluted quantitatively with Metabolite A on both the Corasil II and acidic alumina columns described previously. When the methylated Metabolite B fraction from the initial Porasil separation was run on the gas chromatograph it was possible to associate a single peak in the chromato gram with the UC activity in the sample by trapping the material at the column outlet. The best separation was obtained using an OV-210 column at 125C (isothermal). Initial gc-ms analysis of methylated Metabolite B indicated that it was a sulfur contain ing carboxylic acid. By high resolution gc-ms Metabolite B was shown to be a carboxylic acid (methyl ester) with a nominal molecular weight 178 and molecular formula C6H1004S. A listing of the important peaks is shown in Table 4. TABLE 4 High Resolution gc-ms of Metabolite B mje Molecular formula Structure" 178.0302 146.0064 119.0165 118.0127 91.0247 74.0393 61.0142 59.0118 45.9905 45.0379 * Molecular ion. QH.oO.S CjHsOjS c*h7o2s ch6o2s CjH7OS CjH602 C2HsS c2h3o2 ch2s C2H50 OO I! II CH30--C--CHj--s--ch2c--och3 [M minus -CH3OH]+ 0 1+ II .M" minus --C--OCHj. --CH2--S--CH2--O--CHj]1- 0 1+ II CHjC--OCHi. 'O j+ II --c--och3. [CHjCHjOH]* When a sample of thiodiglycolic acid (dimethyl ester) (3) was run on the Finnigan gc-ms, the mass spectrum was found to be identical to that of Metabolite B (Fig. 5). oO CH30--C--CHj--S--CHj--C--OCH, (3) DISCUSSION The results of the current study establish that the fate of VC following ingestion by rats is dose-dependent. Following a dose of 0.05 or 1 mg kg [1JC]VC, most of the UC activity was excreted in the urine as nonvolatile metabolites and as uCO, in expired air. After 100 mg/kg, the predominant mode of excretion was by expiration of VC. There- R&S 155640 FATE OF [14C]V1NYL CHLORIDE IN RATS 100 45 80 - OII OII h3C-0-C-CH,-S-CHj-C-0-CHj MW 178 349 Fig. 5. Mass spectra of the methyl ester of ihiodiglycolic acid (top) and the methyl ester of metabolite B. fore, it appears that the metabolism of VC is a dose-dependent, saturable process. Similar results on the excretion of [14C]VC following oral ingestion in rats have been found recently by T. Green and D. E. Hathway (personal communication). The fate of VC following oral administration is consistent with its fate after inhalation (Hefner et al., 1975). In those studies, the rate of VC metabolism was more rapid in rats exposed to 100 ppm or less than in rats exposed to 220 ppm or greater. Because of the experimental procedure used by Hefner et al. (1975), evidence was obtained indi cating that at least two pathways may be involved in the metabolism of VC and that the degree of their involvement was concentration-dependent. Recently, R. J. Withey (personal communication) has found a biphasic clearance of VC from the plasma of rats after cessation of inhalation exposure to concentrations of 500 to 7000 ppm or after iv injection of 50 to 75 mg/kg VC. The half-lives of the biphasic process were 4 to 9 min and approx. 40 min. These results deviate to some degree from those reported here. In this study the biphasic pulmonary excretion of VC following a dose of 100 mg/kg had half-lives of 14,4 and 40.8 min for the two phases. The half-lives for the slow component are similar. With regard to the differences for the half-lives of the first phase, 4 to 9 min vs 14.4 min, our value may be somewhat slower because of delayed absorption from the gastrointestinal tract. In any case, the results of this study confirm and complement those of R. J. Withey (personal communication). Of paramount importance to assessing the hazard of exposure to VC is the fate of that portion metabolized. In the previous study (Hefner et al., 1975) it was speculated that potential alkylating metabolites such as chloroacetaldehyde and chloroethylene oxide may be formed in vivo from VC. Two of the three major urinary metabolites of VC have been identified as JV-acetyl-S-(2-hydroxyeihyl)-cysteine and thiodiglycolic acid. The identification of these metabolites is consistent with the proposed pathways for meta bolism of VC. Chloroacetaldehyde and chloroethylene oxide will conjugate with gluta- R&S1 SI 350 WATANABE, MCGOWAN AND GEHRtNG thione and cysteine leading ultimately to the types of metabolites identified in the urine. Gothe (1974) reported recently the trapping of acetaldehyde formed from VC by an in vitro microsomal preparation. This provides additional support for the formation of chloroacetaldehyde. Recently, T. Green and D. E. Hathway (personal communication) have shown that after multiple dosing of [,4C] VC in rats (50 mg/kg orally, three times at 3-hr intervals), thiodiglycolic acid was the major metabolite of vinyl chloride (about 47% of the total urinary I4C activity). The present work confirms that thiodiglycolic acid is one of the major metabolites of vinyl chloride. Two other major urinary metabolites identified by T. Green and D. E. Hathway (personal communication) were S-(2-chloroethyl)cysteine and its acetylated analog, A:-acetyl-5-(2-chloroethyl)cystrine. Since we have identified a major metabolite as A-acetyl-S-U-hydroxyethyOcysteine the question is raised whether the vinyl chloride metabolites exist in the urine as hydroxyethyl or chloroethyl conjugates. S-(2-chloroethyl)-Cysteine is a monofunctional sulfur mustard and has been shown to be mutagenic (Fahmy and Fahmv, 1970). It seems doubtful, however, that it could be detected intact in the urine even if formed because of its susceptibility to hydrolysis to S-(2-hydroxyethyI)-cysteine. Jones (1973) reviewing the metabolism of 1,2-dibromoethane reported that the initially formed 5-(2-bromoethyl)-glutathione is unstable and spontaneously hydrolyzes to S-(2-hydroxyethyl)-glutathione producing S-(2-hydroxyethyl)-cysteine as the primary metabolite, A similar hydrolysis would be expected for the chloroethyl conjugate. More direct evidence for the instability of S-(2-chloroethyl)cysteine is its reported half-life of 7 min in aqueous solution at 373C and pH 7 (Ross, 1962). T. Green and D. E. Hathway (personal communication) identified the vinyl chloride metabolites by preparing the V-trifluoroacetyl n-butyl esters by the method of Gehrke and Stalling (1967). This involves the formation of the methyl ester using 1,25 m HC1 (gas) in methanol at room temperature followed by the transesterification to the n-butyl ester using 1,25 m HC1 (gas) in n-butanol at I00C. It has been reported by Connors and Ross (1958) and Carson and Wong (1964) that 5-(2-chIoroethyl)-cysteine can be prepared by heating S'-(2-hydroxyethyl)-cysteine with concentrated HCI. It seems likely, therefore, that if the urinary vinyl chloride metabolites were present as S-(2hydroxyethyl)-cysteine and its acetylated analog they may be converted into the corre sponding chloroethyl compounds by the derivaiization procedure used by T. Green and D. E. Hathway (personal communication). Therefore, it appears that the vinyl chloride metabolites identified by T, Green and D. E, Hathway (personal communication) correspond to those identified in this study but that the chloroethyl conjugates were in fact artifacts of the derivitization procedure. This does not, however, rule out the initial formation of the chloroethyl conjugate in the animal followed by hydrolysis to the corresponding hydroxyethyl compounds before excretion. As a final point for discussion, it is important to consider how the data gathered by this laboratory and others may relate to assessing the hazard of exposure to VC. Such a discussion has been made previously (Hefner et al.. 1975). Data gathered since that report continue to support the hypothesis that the carcinogenicity of VC is related to the metabolic formation of alkylating metabolites. Numerous studies have reported R&S 155642 FATE OF [UC] VINYL CHLORIDE IN RATS 351 the enhancement of the positive mutagenic response in Salmonella typhimurium exposed to VC if microsomal enzymes or fortified liver homogenates are present (Rannug et al., 1974; Bartsch et al., 1975; Malavielle et al,, 1975). The metabolites of VC identified in the urine indicate that the primary deactivating mechanism is by conjugation with the nonprotein free sulfhydryl compounds, gluta thione and cysteine. Studies in this laboratory have shown that the nonprotein free sulfhydryl groups of the liver are depleted in rats exposed to VC both as a function of concentration and exposure duration (Watanabe et al., 1976). As the nonprotein free sulfhydryl concentrations are depleted, the alkylating metabolites are more likely to react with protein, DNA, and RNA, eliciting proportionally greater toxicity. This phenomena has been demonstrated to markedly influence the toxicity of compounds such as bromobenzene, furosemide, and acetaminophen (Gillette, 1974a,b; Mitchell et at., 1973; Jotlow et al., 1974). The threshold for the toxicity of these materials coin cides with their reactions with tissue macromolecules (protein, DNA, and RNA). Reactions with these cellular components and discernible toxicity occurs only after the glutathione content is sufficiently depleted to preclude deactivation of the reactive metabolites of these agents. ACKNOWLEDGMENTS The authors wish to express appreciation to M. Hiser, R. Francisco, and J. Zempel for technical assistance throughout the study. Appreciation is also gratefully acknowledged to N. Peet and D. Gransden for synthesizing the metabolite standards, and to the Michigan Division Analytical Laboratory of the Dow Chemical Company for their continued support. REFERENCES Bartsch, H., Malavielle. C. and Montesano, R. (1975). Human, rat, and mouse liver mediated mutagenicity of vinyl chloride in Salmonella typhimurium strains. Int. J. Cancer 15, 429-437, Carson, J. F. and Wong, F. F. (1964). The synthesis of L-l,4-thiazane-3-carboxy!ic acid 1-oxide, J. Organ. Chem. 29, 2203. Connors, T. A. and Ross. W. C. J. (1958). a-Methylcysteine and 5-2-chloroethyl-cysteine. Chemistry and Industry, 366. Creech, J. L. and Johnson. M. N. (1974). Angiosarcoma of liver in the manufacture of polyvinyl chloride. J. Oceup. Med. 16, 150-151. Fahmy, O. G. and Fahmy. M. J, (1970). Gene elimination in carcinogenesis: Reinterpretation of the somatic mutation. Cancer Acs. 30, 195-205, Gehrke, C. W, and Stalling, D. L. (1967). Quantitative analyis of twenty natural protein amino acid by gas-liquid chromatography. Separation Science 2(1) 101--130. Gillette, J. R. (1974a). A perspective on the role of chemically reactive metabolites of foreign compounds in toxicity.--I. Biochem. Pharmacol. 23, 2785-2794. Gillette, J. R. (1974b). A perspective on the role of chemically reactive metabolites of foreign compounds in toxicity.--II. Biochem. Pharmacol. 23, 2927-2938. Gothe, R.. Calleman, C. J.. Ehrenberg. L. and Wachtmeister. C. A. (1974). Trapping with 3,4-dichlorobenzenethiol of reactive metabolites formed in vitro from the carcinogen vinyl chloride. Amhio 3, 224--226. Hefner, R. E., Jr., Watanabe. P. G. and Gehring, P. J. (1975). Preliminary studies of the fate of inhaled vinyl chloride monomer (VCM) in rats. Ann. N. Y. Acad. Sci. 246, 135-148. Jollow, D. J,, Thorgeirsson. S. S., Potter, W. Z., Hashimoto, M. and Mitchell, J. R. (1974). Acetaminophen-induced hepatic necrosis VI. Pharmacology 12, 251-271. R&S 156643 352 WATANABH, MCGOWAN AND GEHRING Jones, A. R. (1973). The metabolism of biological alkylating agents. Drug Metabolism Reviews 2(1), 71-100. Malavielle, C., Bartsch, H., Barbin. A., Camus, A. M. and Montesano, R. (1975). Muta genicity of vinyl chloride, chloroethyleneoxide, chloroacetaldehyde, and chloroethanol. Biochem. Biophys. Res. Common. 63, 363-370. Maltoni, C. and Lefemine, G. (1974). La potenzialita die saggi spermentali nella predizione dei rischi oncogeni chi ambientali. Un esempio: 11 chloruro di vinile. Accaci. National Dei Lincei (Roma) (Ser. VIII) 56, 1-11. Mitchell, J. R., Jollow, D. J., Potter, W. Z,, Gillette, J. R. and Brodie, B. B. (1973). Acetaminophen-induced hepatic necrosis IV. J. Pharmacol. Exp. Ther. 187, 211-217. Rannug, U., Johansson, A,, Ramel, C. and Wachtmeister, C. A. (1974). The mutagenicity of vinyl chloride after metabolic activation. Ambio 3. 194-197. Ross, W. C, J. (1962). Biological Alkylating Agents, p. 173, Butterworth, Washington, D.C. Wagner, E. R. and Muelder, W, W. (1975). A procedure for preparing 14C-Iabeled vinyl chloride. Ann. N. Y. Acad. Sci. 246, 152-153. Wagner, E. R., Muelder, W. W., Watanabe, P. G., Hefner, R. E., Jr., Braun, W. H. and Gehring, P. J. (1976). Gas chromatographic method for the preparation of 1JClabeled vinyl chloride. J. Labeled Compounds (in press). Watanabe, P. G., Hefner, R. E., Jr. and Gehring. P. J. (1976). Vinyl chloride induced depression of hepatic nonprotcin sulfhydrvl content and effects on bromsulphalein (BSP) clearance in rats. Toxicology (in press). R&S155644