Inhibition of glutathione synthesis with propargylglycine enhances N-acetylmethionine protection and methylation in bromobenzene-treated Syrian hamsters.

The finding that liver necrosis caused by the environmental glutathione (GSH)-depleting chemical, bromobenzene (BB) is associated with marked impairment in O- and S-methylation of BB metabolites in Syrian hamsters raises questions concerning the role of methyl deficiency in BB toxicity. N-Acetylmethionine (NAM) has proven to be an effective antidote against BB toxicity when given after liver GSH has been depleted extensively. The mechanism of protection by NAM may occur via a replacement of methyl donor and/or via an increase of GSH synthesis. If replacement of the methyl donor is an important process, then blocking the resynthesis of GSH in the methyl-repleted hamsters should not decrease NAM protection. This hypothesis was examined in this study. Propargylglycine (PPG), an irreversible inhibitor of cystathionase, was used to inhibit the utilization of NAM for GSH resynthesis. Two groups of hamsters were pretreated with an intraperitoneal (ip) dose of PPG (30 mg/kg) or saline 24 h before BB administration (800 mg/kg, ip). At 5 h after BB treatment, an ip dose of NAM (1200 mg/kg) was given. Light microscopic examinations of liver sections obtained 24 h after BB treatment indicated that NAM provided better protection (P < 0.05) in the PPG + BB + NAM group than in the BB + NAM group. Liver GSH content, however, was lower in the PPG + BB + NAM group than in the BB + NAM group. The Syrian hamster has a limited capability to N-deacetylated NAM. The substitution of NAM with methionine (Met; 450 mg/kg) resulted in a higher level of GSH in the BB + Met group than in the BB + NAM group (P < 0.05). The enhanced protection by PPG in the PPG + BB + NAM group was accompanied by higher (P < 0.05) urinary excretions of specificO- and S-methylated bromothiocatechols than in the BB + NAM group. The results suggest that NAM protection occurs primarily via a replacement of the methyl donor and that methyl deficiency occurring in response to GSH repletion plays a potential role in BB toxicity.

Glutathione depletion-induced thymidylate insufficiency for DNA repair synthesis.

Dietary methionine (Met) deficiency is known to divert folate away from de novo biosyntheses of purines and the pyrimidine, thymidylate, to the resynthesis of Met resulting in deoxynucleoside triphosphate imbalance. We have recently shown that Met can easily be depleted and methylation can be impaired by exposure to a model glutathione (GSH)-depleting agent, bromobenzene (BB). GSH depletion-induced Met depletion, therefore, could cause thymidylate insufficiency for DNA repair synthesis. The administration of thymidine (Thy) should repair this impairment. When this hypothesis was examined in the present study, several interesting results were found. The administration of Thy labeled with [2-14C]Thy to BB-treated Syrian hamsters at either 1, 5, 7 or 9 h after BB resulted in an attenuation of liver toxicity. Intrahepatic hemorrhage, which is a typical characteristic of BB toxicity in the Syrian hamster, was decreased in the BB + Thy groups. The attenuation of liver toxicity was accompanied by a progressive increase of Thy incorporation into liver genomic DNA at 24 h after BB. With respect to the time points chosen for Thy administration, Thy incorporation found in the BB + Thy groups were 2-, 2-, 3- and 4-fold of the controls that received only Thy. The results provide evidence that BB causes a progressive increase of thymidylate insufficiency in liver cells. Thymidylate insufficiency is due to Met depletion, a depletion that occurs as a result of GSH depletion.

Alterations of DNA methylation by glutathione depletion.

One of the most consistent findings in cancer cells is an overall decrease of 5-methylcytosine content in DNA. The causes that lead to this alteration are not known. We have shown in a recent study that the methyl-donor, methionine (Met), can easily be depleted and that O- and S-methylation can be impaired in response to glutathione (GSH) depletion. This is because mammalian cells are capable of resynthesizing GSH after GSH is depleted, and GSH turnover occurs at the expense of Met. An extensive utilization of Met for the resynthesis of GSH causes Met depletion and impairment in methylation. In the present study we now demonstrate that GSH depletion has a significant impact on DNA methylation. An i.p. dose of a model GSH-depleting hepatotoxin, bromobenzene (BB), caused a progressive impairment in genomic DNA methylation in the Syrian hamster. The administration of a single i.p. dose of Met labeled with [14CH3]Met to BB-treated hamsters at either 1, 3, 5.5 or 9 h after BB resulted in an increase of methyl-group incorporation into liver genomic DNA at 24 h after BB. With respect to the time points chosen for Met administration, methyl-group incorporation found in the BB + Met groups were 1-, 2-, 4- and 12-fold of the controls that received only Met. We further employed an in vitro methylation assay using specific bacterial SssI CpG methylase as the catalyzing enzyme to demonstrate that BB caused a progressive increase of unmethylated CpG sites in genomic DNA. Interestingly, the time response curve of global DNA methylation in vitro showed an identical pattern to that observed in the in vivo experiment. The results provide strong evidence that GSH-depleting agents significantly impair cytosine methylation. Thus, alterations in gene expression could result from a high dose and/or prolonged exposure to GSH-depleting agents, e.g. medications, chemotherapeutic agents and environmental toxins.

Methyl-donor deficiency due to chemically induced glutathione depletion.

Dietary deficiency of methionine (Met) is known to deplete cellular Met and cause DNA hypomethylation, but depletion of Met and impairment in methylation due to chemically induced glutathione (GSH) depletion has escaped recognition. In this study, the effect of GSH depletion on the Met pool and methylation capability was examined after bromobenzene (BB), a model GSH-depleting hepatotoxin, was administered to the Syrian hamster. An i.p. dose of BB (800 mg/kg) caused a rapid and extensive depletion of liver GSH; approximately 68% of the initial concentration was depleted during the first hour. The lowest level of GSH, only 4% of the control, was detected at 5 h. GSH depletion was accompanied by a prompt increase in liver Met during the first hour. This initial increase was followed by an extensive depletion during the next 4 h. At 5 h after BB, liver Met was 12% below the control value, and it remained around this concentration throughout the 24-h experiment. To further confirm these results, the endogenous Met pool was labeled with deuterated Met. The administration of (L)- Met-methyl-d(3) to the Syrian hamster after GSH had been depleted by BB resulted in a significant protection of the liver against necrosis. The protection was accompanied by a marked incorporation of deuterated Met into the liver Met pool. The incorporation, which was determined by gas chromatography-mass spectrometry, shows BB dose dependence. Approximately 53% of the liver Met was labeled when a toxic BB dose (800 mg/kg) was used, while only 25% incorporation was found for the nontoxic dose (100 mg/kg). These results were different from the controls, where only 15% incorporation was found. The differences in the incorporation indicate that there are differences in the degree of utilization and/or depletion of Met in these hamsters, and these differences apparently are dependent upon the degree of toxicity and GSH depletion. The marked incorporation of deuterated Met in the high-dose group was accompanied with a striking increase in the methylation capability. Urinary excretion of the O- and S-methylated 4- and 5-bromo-2-hydroxythiophenols and S-methylated 4- and 5-bromo-2-hydroxy-1,2-dihydrobenzenethiols was significantly increased when compared with the BB treated alone. Approximately 40-45% of the methyl groups in these methylated BB metabolites were methyl-d(3). These results provide direct evidence that depletion of GSH leads to Met depletion and also injures the methylation processes.

Prevention of bromobenzene toxicity by N-acetylmethionine: correlation between toxicity and the impairment in O- and S-methylation of bromothiocatechols.

Bromobenzene (800 mg/kg, ip) caused severe liver necrosis with massive hemorrhage in the golden Syrian hamster within the first 24 hr. Kidney injury was also observed. Treatment with N-acetylmethionine (NAM) at an ip dose of 1200 mg/kg at 5 hr after bromobenzene administration significantly protected the liver and kidney against injuries. Plasma glutamate pyruvate transaminase and blood urea nitrogen levels were substantially decreased in the NAM-treated animals. Histological evaluations confirmed these results. When the urinary neutral and phenolic metabolites of bromobenzene from NAM-treated and untreated hamsters were isolated and compared by GC and GC/MS, a striking result was observed in terms of O- and S-methylated thiol-containing metabolite formation. The NAM-treated animals showed approximately a 8- to 14-fold increase in the excretion of the four isomeric O- and S-methylated bromothiocatechols. These thiocatechols, which are now known to be the 3,4-series metabolites of bromobenzene, can undergo methylation at either the thiol or the hydroxyl functional group. The excretion of 3-S- and 4-S-methylated bromodihydrobenzene thiolols was also increased significantly in the NAM-treated hamster, but other neutral and phenolic metabolites were relatively unchanged. These results suggest that bromobenzene toxicity in the Syrian hamster may be associated with impaired methylation capabilities, an impairment that could be due to methionine and glutathione depletion.

O- and S-methylated bromothiocatechols.

Bromobenzene is metabolized by the Hartley guinea pig to two different bromothiocatechols, 4-bromo-2-hydroxythiophenol and 5-bromo-2-hydroxythiophenol. Both the thiol and phenol functional groups of thiocatechol undergo biological methylation. Methylation at the thiol group leads to the formation of (methylthio)bromophenol (S-methylated bromothiocatechol), while methylation of the phenol group leads to methoxybromothiophenol (O-methylated bromothiocatechol). This resulted in the urinary excretion of four O- and S-methylated bromothiocatechols. Bromothiocatechols could be formed by dehydrogenation of their corresponding bromodihydrobenzene thiolols. Both the 3-S- and 4-S-bromodihydrobenzene thiolols, as S-methylated products, were found as urinary metabolites of bromobenzene in the Hartley guinea pig. All four O- and S-methylated bromothiocatechols and two S-methylated bromodihydrobenzene thiolols were also found as urinary metabolites of bromobenzene in the golden Syrian hamster.

Pathways of formation of 2-, 3- and 4-bromophenol from bromobenzene. Proposed mechanism for C-S lyase reactions of cysteine conjugates.

Bromobenzene is metabolized by the rat and guinea pig to 2-, 3- and 4-bromophenol. 3-Bromophenol is formed through the sulfur-series pathway to phenols. This route involves the enterohepatic circulation; the key intermediate is the S-(2-hydroxy-4-bromocyclohexa-3,5-dienyl)-L-cysteine derived from the 4-S-glutathione conjugate of the 3,4-oxide. A sulfonium ion C-S lyase reaction is proposed in order to account for the pyridoxal phosphate-dependent cleavage/aromatization step, and a C-S beta-lyase reaction sequence is also proposed for the formation of bromodihydrobenzene thiolols. This route of phenol formation may prove to be a general one for aromatic hydrocarbons and closely related compounds that show arene oxide conjugation with glutathione. 2-Bromophenol is formed predominately by spontaneous isomerization of the 2,3-oxide. 4-Bromophenol is formed by the sulfur-series route from the S-(2-hydroxy-5-bromocyclohexa-3,5-dienyl)-L-cysteine. Additional in vivo routes to 3- and 4-bromophenol involve dehydration/aromatization of the 3,4-dihydro-3,4-diol, possibly by way of conjugates; these routes have transient ketonic intermediates. The pathways from bromobenzene to phenols and to sulfur-containing metabolites derived from premercapturic acids show species and dosage variation.

Formation of phenol and thiocatechol metabolites from bromobenzene premercapturic acids through pyridoxal phosphate-dependent C-S lyase activity.

When N-acetyl-S-(2-hydroxy-4-bromocyclohexa-3,5-dienyl)-L-cystein e (4-S-premercapturic acid) and N-acetyl-S-(2-hydroxy-5-bromocyclohexa-3,5-dienyl)-L-cystein e (3-S-premercapturic acid) were used as substrates in incubations with Hartley guinea pig kidney 9000 g supernatant preparations, the major products were the corresponding S-(2-hydroxy-4-bromocyclohexa-3,5-dienyl)-L-cysteine and S-(2-hydroxy-5-bromocyclohexa-3,5-dienyl)-L-cysteine. At the end of the incubation period, the percentage recovery of these N-deacetylate cysteine conjugates accounted for 77 +/- 2% of the substrates, 3-S- and 4-S-premercapturic acids. Removal of the N-acetyl group from premercapturic acids to form the corresponding cysteine conjugates by kidney N-deacetylase(s) showed no preference with respect to the 3-S- and 4-S-positional isomeric conjugates. Other metabolites which included the known sulfur-containing acids, mercaptolactate and mercaptoacetate, were also detected. 3- and 4-Bromophenol and 3- and 4-bromothioanisole were also formed. The addition of pyridoxal-5'-phosphate to the kidney incubation mixture resulted in a 5-fold increase in the formation of phenols and thioanisoles, along with four different isomeric O- and S-methylated 3-S-and 4-S-bromothiocatechols and two S-methylated 3-S- and 4-S-bromodihydrobenzene thiolols. This result indicated that a pyridoxal phosphate-dependent C-S lyase(s) is involved in the formation of both phenol and thiophenolic metabolites from S-(2-hydroxy-4-bromocyclohexa-3,5-dienyl)-L-cysteine and S-(2-hydroxy-5-bromocyclohexa-3,5-dienyl)-L-cysteine. Guinea pig liver 9000 g supernatant preparations did not N-deacetylate the 3-S- and 4-S-premercapturic acids to the same extent as kidney preparations, and this may account for decreased conversion of 3-S- and 4-S-premercapturic acids to 3- and 4-bromophenol and to thiophenolic products by liver preparations.

Conversion of bromobenzene to 3-bromophenol. A route to 3- and 4-bromophenol through sulfur-series intermediates derived from the 3,4-oxide.

Premercapturic acids derived from bromobenzene 3,4-oxide were found to act as precursors of 3- and 4-bromophenol in the rat and guinea pig. The 4-S- and 3-S- positional isomers used in this study were rat urinary metabolites and were prepared in unlabeled, radioactive, and 2,4,6-d3-labeled forms. These are not guinea pig urinary metabolites; the guinea pig does not completely acetylate cysteine conjugates, and this effect leads to urinary products arising from deamination of the cysteine moiety rather than to urinary premercapturic acids. Conversion to phenols was found to be much greater in the guinea pig than in the rat. We interpret our results as indicating that cysteine adducts, rather than the N-acetylcysteine adducts which were administered, are required intermediates in this metabolic route to 3- and 4-bromophenol. This route to phenols may be the major mode of phenol formation for many aromatic compounds. Sulfur-series metabolic products from bromobenzene also include thiocatechols, and these metabolites may be responsible for the hepatotoxicity of bromobenzene in high dosage.

Analytical methods for the study of urinary thioether metabolites in the rat and guinea pig.

Methods are described for the isolation and identification of three classes of bivalent sulfur metabolites characterized as neutral methylthio ethers, ethyl acetate-soluble acidic thioethers and ethyl acetate-insoluble acidic thioethers from rat and guinea pig urine. After extraction of the metabolites by the ammonium carbonate-ethyl acetate procedure, the individual metabolites are separated by capillary gas chromatography and/or by high-performance liquid chromatography with both mu Bondapak C18 and Porasil columns. Identification of the metabolites is based on gas chromatography-mass spectrometry (electron impact) and on fast atom bombardment mass spectrometry. Interesting species differences in metabolism were observed. The major ethyl acetate-soluble acidic thioethers in rat urine are mercapturic acids. In contrast, in the guinea pig a new pathway involving mercaptopyruvic, mercaptolactic and mercaptoacetic acids is operative. The thioether metabolites of styrene oxide and phenanthrene are described, but the procedures have been applied in studies of several drugs and environmental chemicals in our laboratory.

Metabolism of bromobenzene. Analytical chemical and structural problems associated with studies of the metabolism of a model aromatic compound.

Our studies of bromobenzene metabolism have shown that the 3,4-oxide is metabolized to 3- and 4-bromophenol through an extended glutathione pathway. The mechanism of sulfur elimination from a dihydrobromobenzene metabolite is not known, although it is known that the aromatization reaction will occur in a 9000-g supernatant fraction of rat liver. The hepatotoxic and nephrotoxic metabolites of bromobenzene are most likely bromthiocatechols and a bromothiopyrogallol, respectively.

Bromobenzene metabolism in the rat and guinea pig.

The metabolism of bromobenzene was studied in the rat and guinea pig with respect to three considerations: the dose and species dependence of 3-bromophenol excretion; the formation of methylthio analogs of dihydrodiols and catechols; and the identification of acidic bivalent sulfur metabolites. In the guinea pig, 3-bromophenol was the major monohydric phenolic metabolite under conditions of both relatively low and relatively high dosage. In the rat, 3-bromophenol and 4-bromophenol were formed in approximately equal amounts. 2-Bromophenol was a minor metabolite in both species. Methylthio analogs of dihydrodiols were found as guinea pig, but not rat, metabolites. Two di(methylthio)dihydroxytetrahydrobromobenzene metabolites were excreted by the rat but not by the guinea pig. These methylthio compounds have not been reported in earlier studies of bromobenzene metabolism. In the guinea pig, the acidic urinary metabolites were a mercaptoacetate, a mercaptolactate, and a mercapturate. In the rat, the acidic metabolites were a mercapturic acid and premercapturic acids. This species difference in urinary acids indicates a difference in acetylation/deacetylation processes for cysteine conjugates.

Urinary bivalent sulfur metabolites of phenanthrene excreted by the rat and guinea pig.

After the administration of phenanthrene (50 mg/kg, ip) to young adult male rats and guinea pigs, a series of bivalent sulfur urinary metabolites were isolated and characterized by gas chromatography and gas chromatography-mass spectrometry. Seven methylthio metabolites were isolated from the neutral fraction of hydrolyzed rat urine, whereas only two were detected in guinea pig urine. The major methylthio metabolite excreted by each species was 9-hydroxy-10-methylthio-9, 10-dihydrophenanthrene. This was observed as a second-day metabolite in the rat, and its appearance was accompanied by 9-phenanthrol. In addition to the methylthio compounds, which were excreted predominantly as glucuronides, six acidic bivalent sulfur metabolites were isolated from hydrolyzed rat urine and identified by GC/MS; five were present in hydrolyzed guinea pig urine. The major acidic metabolite in hydrolyzed rat urine was the hydroxydihydromercapturic acid N-acetyl-S-(9-hydroxy-9, 10-dihydro-10-phenanthryl)-L-cysteine. The major acidic metabolite in guinea pig urine was the mercaptoacetic acid S-(9-hydroxy-9, 10-dihydro-10-phenanthryl)mercaptoacetic acid, but the hydroxydihydromercapturic acid was also present.

Metabolism of carbamazepine.

Thirty-two metabolites of carbamazepine, in addition to the 10,11-epoxide, have been isolated from enzymatically hydrolyzed urine by HPLC and identified by gas-chromatographic and mass-spectrometric techniques. Eight were sulfur-containing methylthio, methylsulfinyl, or methylsulfonyl derivatives. Eighteen new metabolites, including five iminostilbene derivatives that have not been described earlier, were found. All of the metabolites were formed by processes involving epoxidation and a peroxidation, the structures of the metabolites support the hypothesis that multiple epoxides and a cyclic peroxide are involved in the in vivo metabolism of carbamazepine by the rat and man.

High-performance liquid chromatographic separation of carbamazepine metabolites excreted in rat urine.

A procedure for the separation and isolation of the urinary metabolites of carbamazepine by reversed-phase high-performance liquid chromatography is described. After extraction from urine, the metabolites were separated on either an analytical or semi-preparative C18 mu Bondapak column by gradient elution with methanol-water-acetic acid. Following derivatization the metabolites isolated by the use of the semi-preparative column were analyzed by gas chromatography and gas chromatography-mass spectrometry.

Use of saliva in therapeutic drug monitoring.

We measured the concentrations of phenobarbital, phenytoin, primidone, ethosuximide, antipyrine, and caffeine in paired samples of saliva and plasma by gas chromatograph-mass spectrometer-computer (GC/MS/COM) and enzyme immunoassay. Mixed saliva was collected for the antipyrine and caffeine studies, parotid saliva for the phenobarbital, primidone, ethosuximide and phenytoin studies. The saliva/plasma (S/P) ratios (by weight) obtained by GC/MS/COM were: phenobarbital, 0.31-0.37; phenytoin, 0.11; ethosuximide, 1.04; antipyrine, 0.83-0.95; caffeine, 0.55. The S/P ratio obtained by enzyme immunoassay were: phenobarbital, 0.32; phenytoin, 0.12; primidone, 0.85. The concentrations of phenytoin, primidone, ethosuximide and antipyrine in saliva correspond to the free fraction of the drug in plasma. When we analyzed samples containing phenobarbital or phenytoin (plasma or saliva) by both techniques, we found that the enzyme immunoassay values were generally higher than GC/MS/COM values, suggesting that the metabolites as well as the parent drug were measured in the immunoassay.

The use of gas chromatographic-mass spectrometric-computer systems in pharmacokinetic studies.

Pharmacokinetic studies involving plasma, urine, breast milk, saliva and liver homogenates have been carried out by selective ion detection with a gas chromatographic-mass spectrometric-computer system operated in the chemical ionization mode. Stable isotope labeled drugs were used as internal standards for quantification. The half-lives, the concentration at zero time, the slope (regression coefficient), the maximum velocity of the reaction and the apparent Michaelis constant of the reaction were determined by regression analysis, and also by graphic means.

Clinical applications of gas chromatograph/mass spectrometer/computer systems.

Gas chromatograph/mass spectrometer/computer systems can be used to quantify a wide variety of compounds of clinical interest. A quadrupole instrument operated in the chemical ionization (Cl) mode was used in these studies. Because of the sensitivity and specificity of selective ion detection, it is possible to make measurements routinely in the nanogram to picogram range, with 0.1-1.0 ml samples of plasma and 1-5 ml samples or urine. Internal standards, preferably stable-isotope-labeled compounds, were added to the biological samples before isolation was begun. We describe clinical applications of these procedures to problems in toxicology, pharmacokinetics, and perinatal pharmacology.