Degradation of chemicals by reactive radicals produced by cellobiose dehydrogenase from Phanerochaete chrysosporium.

Phanerochaete chrysosporium, grown on cellulose, produced a cellobiose-dependent dehydrogenase which reduced both ferric iron and molecular oxygen, resulting in the generation of the hydroxyl radical. The hydroxyl radical was detected in reaction mixtures with and without the addition of exogenous H2O2. The purified reductase and the fungus grown under nonligninolytic conditions that promote the production of the reductase were able to depolymerize an insoluble polyacrylate polymer. When oxalate, a secondary metabolite of P. chrysosporium, was used as the iron chelator, it was oxidized by the hydroxyl radical to form the carboxylate anion radical, a strong reductant. Under these reductive conditions, the enzyme was shown to catalyze the reduction of bromotrichloromethane to the trichloromethyl radical. We propose that these oxidative and reductive mechanisms may contribute to the degradation of a wide range of environmental pollutants by fungi which produce this enzyme.

Heme oxygenase active-site residues identified by heme-protein cross-linking during reduction of CBrCl3.

The reduction of CBrCl3 by the heme-heme oxygenase complex forms dissociable and covalently bound heme products. No such products are formed with mesoheme in which the heme vinyl substituents are replaced by ethyl groups. The dissociable heme products are chromatographically similar but not identical to those obtained in the analogous reaction with myoglobin. Tryptic digestion of the heme-protein adduct and Edman sequencing and mass spectrometric analysis of the heme-linked peptide identify His-25, the proximal iron ligand, as the alkylated residue. Reaction of CBrCl3 with the heme complexes of the T135V mutant and a Delta221 C-terminal truncated protein yields heme-linked peptides in addition to that from the wild-type reaction. The sequence of the principal labeled peptide from the T135V reaction, 205TAFLLNIQLFEELQELLTHDTK226 , and the lability of the adduct suggest the heme is attached to one of the carboxylic acid residues. A carboxylic acid residue is probably also labeled in the modified peptide 49LVMASLYHIYVALEEEIER67 from the Delta221 truncated protein. Thus, addition of the reductively generated trichloromethyl radical to a heme vinyl group produces a species that alkylates active-site residues. The changes in the alkylated residue caused by the Thr-135 mutation or truncation of the protein places residues in the sequences 49-67 and 205-226 within the active site. Furthermore, this is the first demonstration that heme oxygenase, like cytochrome P450, may catalyze the reductive metabolism of halocarbons and thus contribute to the toxicity of these agents.

Cholesterol interaction with free radicals produced from carbon tetrachloride or bromotrichloromethane by either catalytic decomposition or via liver microsomal activation.

The reaction between cholesterol (Ch) and trichloromethyl or trichloromethyl peroxyl radicals was studied. The latter were generated from CCl4 either by benzoyl peroxide (BP) catalysis or via thermal activation or by liver microsomal NADPH-dependent biotransformation of CBrCl3. The structure of the products formed was elucidated by gas chromatography-mass spectrometry (GC/MS). Under aerobic conditions and using thermal activation of CCl4, the formation of 6 products was observed. Two (I and II) were dehydrated Ch derivatives (one also having a third double bond) (I). Another product was a delta(5)-3 ketone derivative of Ch (III). Two additional reaction products were determined as ketocholesterols (IV and V). One chloro Ch was also formed (VI). At low concentrations of BP, reaction was more extensive than under thermal activation, and the formation of peaks I to IV was also observed. When the reaction was conducted anaerobically and using thermal activation of CCl4 to generate radicals, only products I and II were formed in low yield. Under anaerobic conditions, but using catalyst, compounds I and III were produced plus two new isomeric ketocholesterol derivatives (VIII and IX) and also a compound having an extra hydroxyl group on the Ch structure (X). In order to check whether similar reactions are observable under biological experimental conditions, we used activation of CBrCl3 by liver microsomes. The incubation using only microsomes (without CBrCl3 or NADPH) showed two ketocholesterol peaks (A and B). In the presence of CBrCl3 we could detect peak B and hydroxycholesterol (C) and two others, ketocholesterols (D and E). D was the only peak showing close similarity (spectrum and retention time) to one of those observed in the chemical reaction system (V). The reaction of CBrCl3 in the presence of NADPH showed peaks B, C, D and E, in low abundance and a 7-ketocholesterol (F). If some of the reaction products reported here were formed during the intoxication with these haloalkanes, significant biological consequences might be expected.

Suicidal inactivation of haemoproteins by reductive metabolites of halomethanes: a structure-activity relationship study.

Human haemoglobin (Hb), methaemalbumin (MHA) or rat liver microsomal cytochrome P-450 (P-450) were incubated anaerobically at microM concentrations with 1 mM carbon tetrachloride (CCl4), trichlorobromomethane (CCl3Br), chloroform (CHCl3) or methylene chloride (CH2Cl2) in presence of 1 mM sodium dithionite as the reducing agent. At the end of a 5-min incubation, haem was measured by various methods, i.e. binding spectrum with CO, pyridine-haemochromogen haem assay and porphyrin fluorescence, and compared for the four analogues. Statistically significant losses were observed, with all three haemo-protein systems, for CCi3Br, CCl4 and CHCl3, but not CH2Cl2. For Hb, the loss was greater with CCl3Br (haem assay, 63%; porphyrin fluorescence, 48%; CO binding, 24%) than with CCl4 (haem assay, 31%) or CHCl3 (haem assay, 13%). On the other hand, with MHA, CCl4 gave a dramatic loss (haem assay, 88%; porphyrin fluorescence, 83%; CO binding, 67%), which was greater than that observed with CCl3Br (haem assay, 49%; porphyrin fluorescence, 38%; CO binding, 25%). No loss was found with CHCl3. Finally, with microsomes, the inactivation was larger with CCl4 (CO binding, 58%; haem assay, 50%; porphyrin fluorescence, 33%) than with CCl3Br (CO binding, 33%; haem assay, 10%) or CHCl3 (haem assay, 9%; CO binding, 6%). In a separate set of similar experiments, an ion-pairing reverse phase HPLC method showed the formation of substrate-dependent hae-derived products during incubation of CCl3Br with Hb or microsomes, and of CCl4 with Hb. A correlation between potential for free radical formation (CCl3Br > CCl4 > CHCl3 > CH2Cl2) and extent of haem inactivation was observed with all methods for Hb, but not for microsomal P-450 or MHA. The results indicate that these halomethanes may be activated differently by different haemoproteins and suggest that their potential ability to undergo reductive metabolism may not be the only critical factor involved in P-450 haem inactivation by these chemicals.

Structure of the novel heme adduct formed during the reaction of human hemoglobin with BrCCl3 in red cell lysates.

It was previously shown that the reductive debromination of BrCCl3 to trichloromethyl radical by human hemoglobin leads to formation of dissociable altered heme products, two of which are identical to those formed from myoglobin and one which is novel. In this study, we have elucidated the structure of this novel adduct with the use of mass spectrometry, as well as 1H and 13C NMR as a substitution product of a -C(Cl) = CCl2 moiety for a beta-hydrogen atom on the prosthetic heme's ring I vinyl group. From studies with the use of 13C-enriched BrCCl3, it was determined that the added carbon atoms were derived from 2 eq of BrCCl3. A mechanism that involves multiple reductive events and a radical cation heme intermediate is proposed. Consistent with this mechanism, cellular reductants were found to selectively enhance the amount of this novel dissociable heme adduct. These studies reveal fine differences between myoglobin and hemoglobin in the accessibility of reactive intermediates to the ring I vinyl group, as well as the potential importance of cellular reductants on the course of heme alteration.

Cytosine attack by free radicals arising from bromotrichloromethane in the presence of benzoyl peroxide catalyst: a mass spectrometric study.

We and others previously reported that CCl4 reactive metabolites are able to covalently bind to liver DNA either in vivo or in vitro. However, no demonstration of the structure of resulting adducts is available in literature. That information would be of relevance, for CCl4 exhibits null or contradictory mutagenic properties and is currently considered a non-genotoxic carcinogen. In the present study we report the nature of the reaction products formed when the putative CCl4 metabolites, .CCl3 and CCl3O2. attack cytosine in a purely chemical system where they were generated from CCl3Br in a benzoyl peroxide catalyzed reaction. Reaction products formed and identified were a) under nitrogen (.CCl3 present)--5-bromo cytosine and cytosine-5-carboxylic acid; b) under air (CCl3O2. present)--5-bromo cytosine, 5-chloro cytosine, 5-hydroxy cytosine, 6-hydroxy cytosine (tentative), chloro hydroxy uracil, 5,6-dihydroxy uracil, and chloro trichloromethyl cytosine. Results from present experiments suggest that if these reaction products were also produced in vivo during either CCl4 or CCl3Br poisoning and they were not repaired in due time prior to replication, they would lead to mutagenic events. Studies directed to obtain evidence for their in vivo formation are in course in our laboratory.

Heme peptide as a model substance for halogenomethane activation and heme modification.

Octa-heme peptide (CHP) obtained from Candida krusei cytochrome c was tested for suicidal activation of halogenomethanes. Under anaerobic conditions, CHP was kept in the reduced state in the presence of NADPH and NADPH-cytochrome P-450 reductase. Addition of CBrCl3 to the reduced CHP caused spectral changes such as rapid disappearance of alpha and beta bands and gradual decrease in the gamma-peak height, accompanied by oxidation of NADPH. Heme content of the reaction mixture, determined as pyridine hemochrome, also decreased NADPH dependently. CCl4 was less effective than CBrCl3, while CHCl3 had almost no effect. N-tert-butyl-alpha-phenylnitrone (PBN) suppressed the CBrCl3-induced heme damage, and resulted in the formation of radical adduct .PBN-CCl3 as evidenced by ESR spectroscopy. Radical formation was also observed with CCl4. The CHP damage induced by CBrCl3 was also accompanied by the release of Br- about 11-12-times molar excess of CHP, whereas the release of CHCl3 was about 20% that of Br-.FD-MS assay of the product of CHP reaction suggested that 10 trichloromethyl radicals bonded with CHP. Thus, CBrCl3 undergoes single-electron reduction in the presence of reduced CHP to trichloromethyl radicals, which covalently bind to CHP molecules. Heme peptide may be a useful tool in the study of mechanisms involved in the destruction of cytochrome P-450 by halogenomethanes.

PB-PK derived metabolic constants, hepatotoxicity, and lethality of BrCCl3 in rats pretreated with chlordecone, phenobarbital, or mirex.

Pharmacokinetic modeling has been very useful in examining the complex relationships between exposure concentration and target tissue dose. This study utilizes a physiologically based pharmacokinetic (PB-PK) modeling approach for assessing the metabolism of BrCCl3 and to investigate its relationship with hepatotoxicity and lethality. Male Sprague-Dawley rats maintained for 15 days on normal diet (control), or on diets containing either chlordecone (CD, 10 ppm), phenobarbital (PB, 225 ppm) or mirex (M, 10 ppm), were used in gas uptake studies to determine the kinetic constants of BrCCl3 metabolism. Four initial concentrations of BrCCl3 at approximately 30, 200, 700, and 3000 ppm were used for each group. The uptake data were analyzed by computer simulation using a PB-PK model containing relevant tissue solubilities and physiological parameters as well as an equation describing the behavior of BrCCl3 in the closed chamber atmosphere. Liver injury was assessed by serum enzyme elevations (alanine aminotransferase, aspartate aminotransferase, and sorbitol dehydrogenase) and histopathological examination, at 24 hr after the exposure to BrCCl3. Another group of similarly pretreated rats was exposed to BrCCl3 and observed over a 14-day period for mortality. Dietary exposures resulted in increased Vmaxc value for BrCCl3 metabolism as compared to control (3.55 +/- 0.14 mg/hr/kg) for PB (8.52 +/- 0.28 mg/hr/kg) and M (5.06 +/- 0.19 mg/hr/kg) but not for CD (3.92 +/- 0.19 mg/hr/kg). Kfc, the first-order rate constant for BrCCl3 metabolism, was decreased after PB (12.9 +/- 0.5 hr-1/kg) and increased after M (17.6 +/- 0.5 hr-1/kg), but unchanged after CD (15.5 +/- 0.6 hr-1/kg) exposure as compared to control (15.0 +/- 0.3 hr-1/kg). The total amount of BrCCl3 metabolism at any initial concentration employed remained unchanged in all the pretreated groups as compared to control. However, the amount of BrCCl3 metabolized through saturable pathway only, at higher initial concentrations, was increased in the PB and M pretreated groups, but not in the CD pretreated group. It is concluded that the rates of metabolism of BrCCl3 were unchanged after CD pretreatment as compared to control, while PB and M pretreatment alter both the saturable and first-order rates. Serum enzymes were significantly increased in all the groups after exposure to BrCCl3 at 200 and 700 ppm concentrations. The increase was more pronounced in PB and M pretreated groups as compared to control and CD pretreated groups. Similarly, histopathological examination of liver showed alterations in the lobular architecture, the extent of alterations being dependent on the dose of BrCCl3 and the pretreatment.(ABSTRACT TRUNCATED AT 400 WORDS)

Metabolism-based covalent bonding of the heme prosthetic group to its apoprotein during the reductive debromination of BrCCl3 by myoglobin.

The reductive metabolism of BrCCl3 by ferrous myoglobin leads to the alteration of the prosthetic heme to form products that can be dissociated from the protein and to those that are irreversibly bound to the protein. The major dissociable or soluble heme metabolites have recently been characterized. In this study, the irreversibly bound heme product was characterized by Edman degradation, amino acid analysis, and electronic absorption and mass spectrometry of peptides derived from the altered protein. It was found that the prosthetic heme was modified by a CCl2 moiety derived from BrCCl3 and was covalently bound to histidine residue 93, the normal proximal ligand to the heme-iron. The data are consistent with a mechanism by which the trichloromethyl radical reacts with the heme to form an intermediate that either can alkylate the proximal histidine residue or form soluble metabolites. The covalent bonding of the heme prosthetic moiety to the apoprotein likely leads to a change in the tertiary structure of the protein that may be responsible for its altered catalytic activity as well as its enhanced susceptibility to proteolysis. Similar processes may account, at least in part, for the covalent alteration of the heme prosthetic group of other hemoproteins caused by xenobiotics and endogenous substrates.

Vitamin E and glutathione are required for preservation of microsomal glutathione S-transferase from oxidative stress in microsomes.

Glutathione (GSH) inhibited lipid peroxidation induced by NADPH-BrCCl3 in vitamin E sufficient microsomes, but did not in phenobarbital (PB)-treated microsomes (containing about 60% of normal vitamin E) or in vitamin E-deficient microsomes (containing about 30% of normal vitamin E). There was a good correlation between the increased formation of CHCl3 from BrCCl3 in the presence of GSH under anaerobic conditions and the vitamin E level in the microsomes. A normal level of vitamin E in microsomes was thus very important for GSH-dependent inhibition of lipid peroxidation and for the efficient formation of CHCl3 from BrCCl3. Bromosulfophthalein (BSP) eliminated the effects of GSH on lipid peroxidation and CHCl3 formation. The apparent Km and Vmax of substrates for GSH S-transferase were changed by in vivo depletion of vitamin E in microsomes, and the Vmax/Km values were significantly reduced. The enzyme activity in microsomes was inactivated following the loss of vitamin E during in vitro lipid peroxidation, and GSH prevented the loss of vitamin E and protected the enzyme from attack by free radicals. GSH inhibited lipid peroxidation induced by NADPH-Fe2+ and the loss of GSH S-transferase activity during the peroxidation in PB-treated microsomes, but did not in the case of induction by NADPH-BrCCl3. A possible relation between the microsomal GSH S-transferase activity and defense by GSH against lipid peroxidation in microsomes is discussed.

Reaction of glutathione with a free radical metabolite of carbon tetrachloride.

Carbon tetrachloride and bromotrichloromethane are both metabolized by cytochrome P-450 in the presence of phenyl-N-t-butyl nitrone PBN) to the PBN/trichloromethyl (PBN/.CCl3) and the PBN carbon dioxide anion (PBN/.CO2-) radical adducts in the liver. The formation of the latter but not the former species in perfused liver was reduced markedly by prior depletion of hepatic glutathione with either diethyl maleate or buthionine sulfoximine treatments. In microsomal incubations, the PBN/.CO2- radical adduct was detected only upon the addition of cytosol. In microsomal incubations containing PBN, CCl4, and GSH, but no added cytosol, a novel radical adduct with distinctive coupling constants was detected. This radical adduct's ESR spectrum exhibited 13C isotope effects when it was formed in an incubation containing 13CCl4 or Br13CCl3. The presence of GSH in the radical adduct is postulated based on the radical adduct's hydrophilicity and slow rate of rotation in solution. The detection of this new radical adduct, PBN/[GSH-.CCl3], establishes the reaction of GSH with a CCl4-derived free radical as a significant event in the metabolism of CBrCl3 and CCl4. The cytosolic conversion of PBN/[GSH-.CCl3] into PBN/.CO2- has been demonstrated and characterizes the PBN/.CO2- radical adduct as the product of metabolism of PBN/[GSH-.CCl3], a primary radical adduct. Thus, it is concluded that GSH rather than oxygen is obligatory for the formation of PBN/.CO2- from .CCl3 in intact cells.

The carbon dioxide anion radical adduct in the perfused rat liver: relationship to halocarbon-induced toxicity.

CCl4 has been shown previously to be metabolized to the trichloromethyl radical (.CCl3) and to a novel oxygen-containing carbon dioxide anion radical (.CO2-) in the perfused rat liver and in vivo. Since the role of free radicals in CCl4-induced hepatotoxicity is unclear, these studies were designed to determine if a relationship between .CO2- formation and halocarbon-induced hepatotoxicity exists. CCl4 or bromotrichloromethane (CBrCl3) was infused into livers from control or phenobarbital-treated rats perfused with either nitrogen- or oxygen-saturated Krebs-Henseleit bicarbonate buffer. Samples of effluent perfusate and chloroform/methanol extracts of liver were analyzed by ESR spectroscopy for free radical adducts following infusion of halocarbon and the spin trap, phenyl-t-butylnitrone (PBN). Hyperfine coupling constants and 13C-isotope effects observed in the ESR spectra of organic extracts of liver demonstrated the presence of the PBN radical adduct of .CCl3 from both halocarbons. Radical adducts in aqueous extracts of liver and effluent perfusate had hyperfine coupling constants and 13C-isotope effects identical to those of PBN/.CO2- generated chemically from formate. The PBN/.CO2- radical adduct was also observed in urine following the intragastric administration of CBrCl3 and PBN. Detection of PBN/.CO2- adducts in the effluent perfusate was decreased 3- to 4-fold by DIDS (0.2 mM), an inhibitor of the plasma membrane anion transport system. The rate of formation of PBN/.CO2- was decreased 2- to 3-fold following inhibition of cytochrome P-450-dependent monooxygenases by metyrapone (0.5 mM) and was increased about 2-fold by induction of cytochrome P-450 by phenobarbital pretreatment. Toxicity of halocarbons in the perfused liver was assessed by measuring the release of lactate dehydrogenase (LDH) into the effluent perfusate in livers from phenobarbital-treated rats under conditions identical to those employed to detect radical adducts (i.e., during the infusion of CCl4 or CBrCl3 into livers perfused with either nitrogen- or oxygen-saturated perfusate). Under all conditions studied, PBN/.CO2- was detected in the effluent perfusate within 2-4 min. Metabolism of halocarbons to PBN/.CO2- was 6- to 8-fold faster during perfusion with nitrogen-saturated rather than with oxygen-saturated perfusate. Concomitantly, liver damage detected from LDH release occurred much sooner during halocarbon infusion in the presence of nitrogen-saturated rather than oxygen-saturated perfusate. A good correlation between the rate of formation of PBN/.CO2- and the time of onset of LDH release following halocarbon infusion was observed.(ABSTRACT TRUNCATED AT 400 WORDS)

Factors influencing the formation of the carbon dioxide radical anion (.CO2-) spin adduct of PBN in the rat liver metabolism of halocarbons.

Spin trapping techniques have been used to detect free radicals generated from the in vitro metabolism by rat liver microsomes of carbon tetrachloride (CCl4) and bromotrichloromethane (BrCCl3) under conditions of varying oxygen tension and pH. Dispersions of rat liver microsomes incubated with 12CCl4, 13CCl4 or Br12CCl3, alpha-phenyl-tert-butyl nitrone (PBN) and NADPH/NADH in a phosphate buffer varying in pH from 6.6 to 8.0 under varying oxygen tensions produced various amounts of four different PBN adducts: PBN-CCl3, PBN-L, PBN-OL and PBN-CO2- where L is a carbon-centered lipid type radical and LO is an oxygen-centered lipid type radical. The relative amount of PBN-CO2- increases with the absence of oxygen. With the use of 31P-NMR in vivo spectroscopy it was possible to detect a pH change from 7.4 to 6.8 in the livers of rats treated with CCl4 or BrCCl3. These results suggest that halocarbon metabolism in biological systems may depend on both oxygen tension as well as pH.

Bromotrichloromethane-induced damage to bronchiolar Clara cells.

Administration of bromotrichloromethane (BrCCl3) to rats results in selective damage to bronchiolar non-ciliated Clara cells; ciliated bronchiolar cells and pneumocytes were unaffected. The cellular alterations begin very early (10 min) after poisoning. Lipid peroxidation, as measured by the malonic dialdehyde (MDA) content in the lung, is greatly increased 10 min after BrCCl3 administration. A histochemical technique to detect, in vitro, lipid peroxidation in frozen sections was used to demonstrate whether sufficient activation of BrCCl3 occurs in lung tissue. The positivity for the histochemical reaction was observed in bronchiolar epithelium in which cytochrome P450-dependent monooxygenase activity is predominantly located. The data obtained strongly support that BrCCl3 is highly metabolized in bronchiolar Clara cells.

Potentiation of bromotrichloromethane hepatotoxicity and lethality by chlordecone preexposure in the rat.

These studies were designed to provide dose-response relationships for chlordecone (CD) potentiation of BrCCl3 hepatotoxicity in male rats using biochemical, functional and histopathological parameters. The influence of this interaction on BrCCl3 lethality was also examined. Male Sprague-Dawley rats (175-200 g) received a single ip dose of 1, 5, 10, 15 or 25 microL BrCCl3/kg following a 15 day dietary pretreatment of 0 or 10 ppm CD. Twenty four hrs after BrCCl3 challenge, biliary excretion of phenolphthalein glucuronide (PG), bile flow, serum transaminases (SGOT and SGPT), serum ICD and OCT were examined as functional and biochemical indices of hepatic injury. Effect of CD on 48 hr LD50 of BrCCl3 was also examined using the method of moving averages. With the exception of 1 microL BrCCl3/kg dose which had no effect, CD-BrCCl3 combination resulted in potentiation of hepatotoxicity by all parameters examined. Activity of all the serum enzymes was elevated in a dose related manner. A dose related decrease in the biliary excretion of PG and bile flow was observed. These effects were more pronounced at the higher doses of BrCCl3. Extensive centrilobular necrosis was observed in the animals given CD-BrCCl3 combination and the necrogenic effect was more severe at the doses of 15 microL and 25 microL BrCCl3/kg. BrCCl3-lethality was increased 5-fold by CD as indicated by the decreased LD50. The results suggest that CD-induced BrCCl3 toxicity is manifested both in the form of hepatotoxicity and lethality and since the hepatic functional status is greatly compromised, the CD potentiated hepatic failure is related to lethality.

Oxidation of carbon tetrachloride, bromotrichloromethane, and carbon tetrabromide by rat liver microsomes to electrophilic halogens.

In order to determine whether CCl4, CBrCl3, CBr4 or CHCl3 undergo oxidative metabolism to electrophilic halogens by liver microsomes, they were incubated with liver microsomes from phenobarbital pretreated rats in the presence of NADPH and 2,6-dimethylphenol. The analysis of the reaction mixtures by capillary gas chromatography mass spectrometry revealed that 4-chloro-2,6-dimethylphenol was a metabolite of CCl4 and CBrCl3 whereas 4-bromo-2,6-dimethylphenol was a metabolite of CBr4. The formation of the metabolites was significantly decreased when the reactions were conducted with heat denatured microsomes, in the absence of NADPH or under an atmosphere of N2. These results indicate that the chlorines of CBrCl3 and CCl4 and the bromines of CBr4 are oxidatively metabolized by rat liver microsomes to electrophilic and potentially toxic metabolites.

The formation of diglutathionyl dithiocarbonate as a metabolite of chloroform, bromotrichloromethane, and carbon tetrachloride.

One hour after the intraperitoneal administration of CHCl3, CBrCl3, or CCl4 to phenobarbital (PB)-treated rats, hepatic GSH levels decreased to 30, 59, and 88% of control levels, respectively; after 4 hr, the GSH levels had returned to 46, 65, 99%, respectively, of control levels. When incubated for 15 min in air with rat liver microsomes from PB-treated rats, a NADPH-generating system, and GSH (5 mM), all of the compounds were converted to diglutathionyl dithiocarbonate (GSCOSG). The rate of conversion of CHCl3, CBrCl3, and CCl4 to GSCOSG was 180, 58, and 8 nmol per mg of protein per 15 min, respectively. The GSCOSG was also identified in bile by 13C-NMR spectroscopy and HPLC as an in vivo metabolite of CHCl(3), CBrCl3, and CCl4. After the administration of CHCl3, CBrCl3, and CCl4, 2.89, 0.64, or 0.11 mumol of GSCOSG, respectively, was excreted in 6 hr. These results suggest that CHCl3, CBrCl3, and CCl4 are metabolized in vitro and in vivo to phosgene (COCl2), which reacts with GSH to produce GSCOSG. The reaction of GSH with COCl2 may be responsible at least in part for the GSH-depleting properties of CHCl3, CBrCl3, and CCl4, inasmuch as the relative amounts of formation of GSCOSG in vitro and in vivo paralleled their relative GSH-depleting activities.