The epoxide-diol pathway in the metabolism of hexobarbital in rat and man

[Studies on the structure-activity relationship of allyl substituted oxopyrimidines searching for the novel antagonist or agonist of barbiturates to the sleep mechanism based on the uridine receptor theory--barbituric acid to uridine (part I)]

Thirty-six allyl substituted oxopyrimidine analogues such as barbituric acid (BA), barbiturates, uracil, thymine, and related derivatives including 13 new compounds were synthesized and their pharmacologic effects ([hypnotic activity, anticonvulsant activity against pentylentetrazol (PTZ)-induced seizures, and LD(50)]) and interactions with the barbiturates were evaluated in mice and rats. The results are briefly and parially summarized as follows. BA prolonged pentobarbital (PB)-induced sleep and had some central depressant effects. N,5,5-triallyl-BA exhibited some hypnotic and anticonvulsant activities, although the other 5,N-allyl-compounds did not show any activity except for allobarbital (AlloB). N-allyl-BA, 5-allyl-BA, N(1),N(3),5-triallyl-BA, N,5,5-triallyl-BA, and N(1),N(3),5,5-tetraallyl-BA also prolonged PB-induced sleep. Interestingly, N,5,5-triallyl-BA was the most potent in the interaction with AlloB, phenobarbital (PheB), amobarbital (AB), PB, and thiopental (TP) but not barbital (B). N(1),N(3),5,5-tetraallyl-BA prolonged AlloB-, PB-, and AB-induced sleep but not B-, PheB-, and TP-induced sleep. N(1),N(3),5-triallyl-B prolonged only PB- and TP-induced sleep. 5,5-diallyl-BA prolonged PheB- and TP-induced sleep. N,5-diallyl-BA prolonged only TP-induced sleep. In contrast, BA and N(1),N(3),5-triallyl-AB tended to antagonize AlloB, AB, and B. N(1),N(3),5,5-tetraallyl-BA also slightly antagonized B, PheB, and TP. 5,5-diallyl-BA antagonized only AB. The prolonging effects of BA, N,5,5-triallyl-BA, and N(1),N(3),5,5-tetraallyl-BA on PB-induced sleep were dose dependent. These results indicate that the position and number of allyl groups substituted on the structure of BA play an important role in their depressant activities. This review deals with the structure-activity relationship of allyl-substituted oxopyrimidines as part of our search for antagonists and agonists of barbiturates as well as their mechanisms of action.

New Aspects of Hexobarbital Metabolism: Stereoselective Metabolism, New Metabolic Pathway via GSH Conjugation, and 3-Hydroxyhexobarbital Dehydrogenases

Hexobarbital, a short-acting hypnotic, is metabolized to 3'-hydroxyhexobarbital by cytochrome P450, and then to 3'-oxohexobarbital by liver cytosolic dehydrogenase. New methods of separation for hexobarbital and its metabolites by TLC have been developed and applied to study the metabolism of hexobarbital enantiomers and stereoselective metabolism of hexobarbital. (+)-Hexobarbital preferentially was transformed into beta-3'-hydroxyhexobarbital and the (-)-enantiomer preferentially transformed into alpha-3'-hydroxyhexobarbital by rat liver microsomes. Glucuronidation and dehydrogenation of 3'-hydroxyhexobarbital were also stereoselective and the S-configuration at the 3'-position was preferred. Alpha-3'-hydroxyhexobarbital from (-)-hexobarbital and the beta-isomer from (+)-hexobarbital were shown to be preferentially conjugated with glucuronic acid in rabbit urine, and to be preferentially dehydrogenated to form 3'-oxohexobarbital by rabbit and guinea pig 3-hydroxyhexobarbital dehydrogenases. A new metabolic pathway of hexobarbital was found in which 3'-oxohexobarbital reacts with glutathione to form 1,5-dimethylbarbituric acid and a cyclohexenone-glutathione adduct, a novel metabolite. 1,5-dimethylbarbituric acid was excreted into the urine and the cyclohexenone-glutathione adduct into the bile of rats dosed with hexobarbital. 3-hydroxyhexobarbital dehydrogenases that dehydrogenate 3-hydroxyhexobarbital into 3'-oxohexobarbital were purified from the liver cytosol of rabbits, guinea pigs, goats, rats, mice, hamsters, and humans and characterized. These enzymes were monomeric proteins and had molecular weights of about 34500-42000, and used NAD(+) and NADP(+) as cofactors, except for the human enzyme that had a molecular weight of about 58000 and used NAD(+) alone. Each enzyme exhibited its own characteristics. Substrate specificity demonstrated that 3-hydroxyhexobarbital dehydrogenases dehydrogenate not only alpha,beta-unsaturated cyclic and acyclic secondary alcohols but also some 17 beta-, 3 alpha-hydroxysteroids or both, except for the human enzyme. The amino acid sequence of the hamster enzyme indicated that it belongs to the aldo-keto reductase superfamily and hydroxysteroid dehydrogenase subfamily.

Hexobarbital metabolism: a new metabolic pathway to produce 1,5-dimethylbarbituric acid and cyclohexenone-glutathione adduct via 3'-oxohexobarbital

1. In the presence of glutathione under physiological conditions, 3'-oxohexobarbital was non-enzymically converted to 1,5-dimethylbarbituric acid and a cyclohexenone-glutathione adduct. 2. The two reaction products were characterized by mass spectrometry, 1H- and 13C-n.m.r. spectrometry, and UV spectral analyses. 3. 1,5-Dimethylbarbituric acid was excreted in urine of rat given hexobarbital, 3'-oxohexobarbital, or 1',2'-epoxyhexobarbital, and accounted for 13.4, 14.5 and 4.7% of dose, respectively. 4. The cyclohexenone-glutathione adduct, a novel metabolite of hexobarbital, was excreted in the bile of rat given hexobarbital. 5. The route of 1,5-dimethylbarbituric acid formation via 3'-oxohexobarbital in the metabolism of hexobarbital was discussed in comparison with the epoxide-diol pathway.

Enantiospecific quantification of hexobarbital and its metabolites in biological fluids by gas chromatography/electron capture negative ion chemical ionization mass spectrometry

A highly sensitive and specific assay based on gas chromatography/electron capture negative ion chemical ionization mass spectrometry has been developed for the analysis of the enantiomers of hexobarbital and its major metabolites in human urine and plasma. S-(+)-(5-2H3)hexobarbital and R-(-)-(5-2H3)hexobarbital were synthesized for clinical studies along with (+/-)-(1,5-2H6)hexobarbital and the deuterated major metabolites for use as internal and reference standards. Hexobarbital enantiomers and their metabolites were analyzed after pentafluorobenzyl and trimethylsilyl derivatization, following solid-phase extraction from plasma and urine. Intense negative ion spectra were observed for all of the derivatives. The base peak in the spectra corresponded to the M-pentafluorobenzyl anion [M-PFB]- except for 1,5-dimethylbarbituric acid, where M-. was the most abundant ion. The applicability of the method was demonstrated by following the plasma concentration-time profiles and urinary excretion in a male extensive metabolizer of mephenytoin who was given a pseudoracemic oral dose of hexobarbital containing equal 50 mg amounts of S-(+)-2(H0)hexobarbital and R-(-)-(2H3)hexobarbital. Marked stereoselective disposition was observed, with the R-(-)-enantiomer being more efficiently metabolized, primarily by alicyclic oxidation and ring cleavage.

High-performance liquid chromatographic method for the simultaneous determination of R-(-)- and S-(+)-hexobarbital in rat plasma

The enantiospecific determination of R- and S-hexobarbital in rat plasma is described. The method involves liquid-liquid extraction of racemic hexobarbital from plasma, separation of the underivatized enantiomers by high-performance liquid chromatography on an alpha 1-acid glycoprotein column and ultraviolet detection. The mobile phase consists of a phosphate buffer (pH 5.4) containing 0.4% 2-propanol as organic modifier. An alpha 1-acid glycoprotein guard column is used to increase the lifetime of the analytical column. Heptabarbital is the achiral internal standard. With detection limits of ca. 0.05 microgram/ml for both R- and S-hexobarbital, the assay is suitable for pharmacokinetic studies of the enantiomers in rats.

Pharmacokinetics of hexobarbital in young and old rats

Hexobarbital is a drug widely used to study the capacity of the liver to metabolize drugs. The pharmacokinetics of hexobarbital in 3- and 30-month-old male BN/BiRij rats were studied. The half-life of hexobarbital in 30-month-old rats (39.9 +/- 4.1 min) was significantly higher than that of 3-month-old ones (21.3 +/- 3.8 min). The volume of distribution (ml.kg-1 body weight) did not change with age. The intrinsic clearance, expressed as ml.min-1.kg-1 body weight, of hexobarbital in 30-month-old rats (20.2 +/- 6.6) was half that of the 3-month-old ones (39.5 +/- 7.6). Further studies will be performed to investigate the effect of age on the intrinsic clearance of S(+)- and R(-)- hexobarbital.

N,N'-diallylpentobarbital (DAPB) metabolites and their effects on pentobarbital-induced sleep and hepatic drug-metabolizing enzymes

1. The biological half-life (t 1/2) of N,N'-diallylpentobarbital (DAPB) in brain after i.p. injection to mouse was 96 min (first phase) and 11 h (second phase). The t 1/2 values in plasma were 102 min and 9.4 h, respectively, after i.p. injection. After intracerebroventricular (i.c.v.) administration, the t 1/2 values in brain and plasma were 18 and 120 min, and 42 and 177 min, respectively. 2. Following i.p. administration of 2-14C-DAPB (80 mg/kg), 58% of the 14C was excreted in the urine in 72 h. Several urinary metabolites were identified by g.l.c.-mass spectrometry, DAPB was metabolized by three major pathways, i.e., omega-1 hydroxylation, epoxide-diol pathway and N-deallylation. 3. The effects of DAPB and its metabolites on pentobarbital (PB)-induced sleep were examined after i.p., i.v. and i.c.v. administration. Metabolite 1 [M-1; (omega-1)-hydroxy-DAPB], an active metabolite, exhibited the most potent prolonging effect. 4. M-1 and other metabolites, as well as unchanged DAPB, showed significant inhibitory effects on mouse hepatic drug-metabolizing enzymes.

Correlation between the metabolism of hexobarbital and aminopyrine in vivo in rats

Two model substrates for oxidative hepatic enzyme activity, namely hexobarbital and aminopyrine, were simultaneously orally administered to rats, and blood concentrations of the substrates measured by g.l.c. The apparent intrinsic clearances of hexobarbital (Cl*int.HB) and of aminopyrine (Cl*int,AM) were correlated in untreated rats, and in rats pretreated with phenobarbital, 3-methylcholanthrene, polychlorinated biphenyls or carbon tetrachloride. Cl*int,HB and Cl*int,AM were both increased by phenobarbital and polychlorinated biphenyl pretreatment. Pretreatment with 3-methylcholanthrene had hardly any effect, and carbon tetrachloride caused a strong diminution of Cl*int.HB and Cl*int.AM. When the dose of aminopyrine was decreased, both Cl*int,HB and Cl*int,AM increased. This indicated that the primary metabolite of aminopyrine, monomethylaminopyrine, inhibits cytochrome P-450. The correlation coefficient for all clearance data was 0.92 (N = 36). It was concluded that both hexobarbital and aminopyrine are metabolized in vivo by the same or closely related cytochrome P-450 isozymes, and both may be used as model substrates in vivo for metabolic conversions primarily mediated by the major phenobarbital-inducible cytochrome P-450 subspecies.

Disposition of allylic oxidation pathway metabolites of racemic hexobarbital in the rat

The pharmacokinetics in blood of the major metabolites of hexobarbital (HB), 3'-hydroxyhexobarbital (OH-HB) and 3'-ketohexobarbital (K-HB) were studied in rats. In addition urinary excretion of OH-HB and K-HB and 1,5-dimethylbarbituric acid (DMBA) was determined. Half-lives of OH-HB and K-HB were slightly longer than that of the parent drug. Urinary recovery of OH-HB, K-HB and DMBA following i.a. administration of OH-HB (75%) was more complete than the recovery following i.a. administration of K-HB (52%). Most probably further metabolism of K-HB takes place. Of K-HB, 41% was excreted renally, and 3.4% of K-HB reverted back to OH-HB. Of OH-HB, about 45% was excreted renally, following p.o. or i.a. administration. Since about 10% of both OH-HB and K-HB was converted to DMBA, it seems that the epoxide-diol pathway as proposed for HB also plays a minor role in the metabolism of OH-HB and K-HB. It is further concluded that measuring allylic pathway oxidation metabolites of HB does not improve the usefulness of HB as a model compound in the assessment of the activity of oxidative drug metabolizing activity.

Selected ion monitoring analysis of pseudoracemic hexobarbital and its major metabolites in blood and urine of rats

A stereospecific synthesis of N1-(2H3)-labelled R(-)-hexobarbital is described. A sensitive and rapid selected ion monitoring assay procedure for pseudoracemic hexobarbital, consisting of equal amounts of S(+)-hexobarbital (1a) and (2H3)-R(-)-hexobarbital (1b) in 100 microliters blood samples of rats was developed. Both the parent enantiomers in blood and three major metabolites excreted in urine were quantified. An application of the method in rats is described, and the results are compared to previously obtained data of separately administered enantiomers.

Route- and dose-dependent pharmacokinetics of hexobarbitone in the rat: a re-evaluation of the use of sleeping times in metabolic studies

The metabolic clearance (CL) and half-life of racemic hexobarbitone and sleeping time were studied in rats following intra-arterial (i.a.), intraperitoneal (i.p.) and oral (p.o.) administration, at dose levels of 25 and 100 mg kg-1 of its sodium salt. CLp.o. was higher than CLi.a. at both 25 and 100 mg kg-1. CLi.a. and CLi.p. values were much lower, but CLi.p. was higher than CLi.a. at 25 mg kg-1 and lower than CLi.a. at 100 mg kg-1. There was no distinct dependency of the half-lives on route of administration, but a slight increase upon increasing the dose was observed. Hexobarbitone blood concentrations at which the rats awoke were significantly higher after 100 mg kg-1 i.p. than after 100 mg kg-1 i.a., although there was only a small difference in sleeping time. It is postulated that the rate of uptake of the barbiturates into the portal system after i.p. administration is so high that transient saturation of hepatic first-pass metabolism occurs. Therefore neither CLi.p. nor sleeping times can be used as an accurate reflection of drug-metabolizing enzyme activity in the rat; instead CLp.o. should be used.

Biotransformation of N-methylcyclobarbital in vivo in rabbit and rat

Metabolism of N-methylcyclobarbital in the rabbit and rat has been studied in vivo for the purpose of comparison with the C5-methylated analogue, hexobarbital. In the rabbit, the main route of the metabolism of N-methylcyclobarbital is glucuronide formation after hydroxylation at the 3'-position of the parent compound. Dehydrogenation of the 3'-hydroxy product, a major pathway in the metabolism of hexobarbital, was a minor route in the case of N-methylcyclobarbital. In addition, a new type of metabolite, thought to be dihydroxylated products from spectral studies, was isolated. In the rat, there were almost no differences in the metabolic fates of N-methylcyclobarbital and hexobarbital. Profiles of metabolism of four analogous barbiturates (N-methylcyclobarbital, hexobarbital, cyclobarbital and norhexobarbital), which have a cyclohexene ring on the 5-carbon, reveal the contribution of alkyl substituents in the barbiturate ring on the bioavailability and metabolism of these compounds.

Endogenous hepatic growth-modulating factors and effects of a choline-devoid diet and of phenobarbital on hepatocarcinogenesis in the rat

The activities of an endogenous inhibitor and of a stimulator of cell proliferation were assayed in the livers of sham-operated (SO) or partially hepatectomized (PH) adult rats; rats fed a choline-supplemented (CS) or a choline-devoid (CD) diet; the same diets followed by acute CCl4 intoxication; the same diets supplemented with phenobarbital (PHB); or a CD diet containing DL-ethionine (ETH). The inhibitor and the stimulator were semipurified by fractional ethanol precipitation of a liver cytosolic fraction, and their activities were assessed by means of bioassays in vitro. The livers of SO rats and of rats fed the CS diet contained only inhibitor activity. Following PH, a CD diet, or CCl4 intoxication the inhibitor activity was suppressed, and there was a simultaneous appearance of a stimulator activity. Thus, PH, a CD diet, and CCl4 intoxication cause similar cellular (loss and regeneration) and humoral-homeostatic changes in adult rat livers. We propose that these changes constitute a basic attribute of the mechanism whereby the three conditions affect similarly hepatocarcinogenesis in the rat, especially in the case of a CD diet, because the changes it induces are chronic rather than acute. PHB, another promoter of chemical hepatocarcinogenesis, affected neither the inhibitor nor the stimulator activity. Thus, PHB seems to be acting by a different mechanism than that of the other three agents. ETH did not modify the shift in the balance of the growth-modulating factors induced by a plain CD diet. This shift may account for the marked stimulation of carcinogen-induced oval cell proliferation exerted by a CD diet. The significance of these results is discussed in the context of known effects of a CD diet and of PHB on hepatocarcinogenesis in rats.

Disposition of hexobarbitone in healthy man: kinetics of parent drug and metabolites following oral administration

1 Hexobarbitone plasma kinetics were determined in six healthy volunteers, who received 500 mg hexobarbitone orally. In addition urinary excretion rate and cumulative excretion were measured of its three major metabolites: 3'-hydroxyhexobarbitone, 3'-ketohexobarbitone and 1,5-dimethylbarbituric acid. 2 The mean plasma elimination half-life of hexobarbitone was 3.7 +/- 0.9 h (n = 6). Assuming complete absorption, the volume of distribution and the metabolic clearance were 81.3 +/- 20.5 1 and 16.4 +/- 2.9 1/h, respectively. The mean maximal plasma concentration was 7.1 +/- 2.1 micrograms/ml and was reached 1.2 +/- 0.4 h after drug administration. 3 3'-Hydroxyhexobarbitone and 3'-ketohexobarbitone, which are products of allylic side-chain oxidation of hexobarbitone, were excreted in 24 h to the extent of 4.7 +/- 1.3 and 32.1 +/- 11.9% of the dose, respectively. In the same period, 1,5-dimethylbarbituric acid, which is the end product of the epoxide-diol pathway, was excreted to 18.0 +/- 7.8% of the dose. The ratio of the sum of 3'-hydroxy- and 3'-ketohexobarbitone vs 1,5-dimethylbarbituric acid excreted varied with time and amounted ultimately in 24 h urine to 2.3 +/- 1.0. 4 The half-lives of 3'-hydroxyhexobarbitone and 1,5-dimethylbarbituric acid, calculated from their renal excretion rate curves, amounted 5.2 +/- 0.9 and 6.6 +/- 1.3 h and were significantly longer than the half-life of hexobarbitone in plasma. The half-life of 3'-ketohexobarbitone was 4.2 +/- 0.8 h. The maximum excretion rate of 1,5-dimethylbarbituric acid was reached at 7.7 +/- 1.0 h after administration of hexobarbitone. 3'-Hydroxy- and 3'-ketohexobarbitone were excreted with a maximal rate at 2.2 +/- 0.8 and 2.8 +/- 0.4 h respectively.

The metabolic fate of 1',2'-epoxyhexobarbital in the rat

1. In urine of rats treated with 1',2'-epoxyhexobarbital, unchanged compound and six metabolites were identified: 1,5-dimethylbarbituric acid, which is the end product of an epoxide-diol pathway, two stereochemically different 3'-hydroxy-1',2'-epoxyhexobarbitals, a hydroxyfuropyrimidine, 3'-hydroxyhexobarbital and 3'-ketohexobarbital. 2. The analytical methods used were based on capillary g.l.c. with nitrogen-selective or mass spectrometric detection. Identification was by electron impact and chemical ionization mass spectrometry. All the reference compounds needed for comparison were synthesized. 3. The mean plasma elimination half-life of 1',2'-epoxyhexobarbital after intra-arterial administration to the rat was 13.7 +/- 1.5 min (mean +/-S.D.; n = 3). A total body clearance of 35.2 +/- 9.6 ml/min (mean +/- S.D.) was calculated, which includes renal clearance of unchanged epoxide. 4. In rat liver microsomal preparations it was demonstrated that 1',2'-epoxyhexobarbital is hydrated by epoxide hydratase. With 1 mM 1,1,1,-trichloropropene-2,3-oxide (TCPO) this enzymic reaction could be inhibited completely. 5. On administration of the individual metabolites of the epoxide to rats, no evidence was found for their possible intermediacy in the formation of 3'-hydroxy- or 3'-ketohexobarbital, which are major metabolites of hexobarbital.

Mutagenicity of structurally related oxiranes: derivatives of benzene and its hydrogenated congeners

The mutagenicities of 17 closely related oxiranes were determined in 4 tester strains (Salmonella typhimurium TA98, TA100, TA1535, TA1537). The test compounds comprised all possible oxides of benzene and its partially hydrogenated congeners. In TA100 and TA1535, 12 of the tested oxiranes were weak to moderate mutagens. 4 of these were also active in TA98. No mutagenicity was observed with the remaining 5 compounds in any of the 4 strains. The presence of a double bond in formal conjugation with the epoxide ring increased the mutagenicity relative to that of the saturated oxirane. Interestingly, additional epoxide rings within the same molecule did not markedly increase the mutagenic activity, and for the oxiranes that are not activated by a double bond, the relationship between mutagenic activity and the number of epoxide rings in the molecule was even inverse. The influence of bromo and hydroxyl substitution on oxirane mutagenicity is discussed. Most notably, a compound having a 4-hydroxyl group in syn position to a 1,2-epoxide ring fused to the cyclohexane ring, a structure which has been suggested to increase the electrophilic reactivity of dihydrodiol epoxides through hydrogen bonding, was almost inactive.

Hexobarbital-binding, hydroxylation and hexobarbital-dependent hydrogen peroxide production in hepatic microsomes of guinea pig, rat and rabbit

Cytochrome P-450 dependent oxygenase (3'-hydroxy-hexobarbital) and oxidase activities (hydrogen peroxide) have been measured in hepatic microsomes from guinea pigs, rats and rabbits. A sensitive gas-chromatographic assay was developed to measure the hydroxylated product 3'-hydroxy-hexobarbital. The kinetics of its formation were determined and correlated to hexobarbital type I binding and compared with oxidase activity: in the rat, Vmax for 3'-hydroxyhexobarbital formation was 5.1 and 2.6 nmoles/mg/min, resp. This was increased by phenobarbital treated rabbits, Vmax was 15.0 nmoles/mg/min for hydroxylation and 40.8 for H2O2 formation. Spectral affinity constants (Ks) in control animals were 0.12 mM (rats) and 0.14 mM (rabbits). Phenobarbital treatment decreased these affinity constants, which were similar for each activity measured. In guinea pigs, however, hydroxylation of exobarbital was low (3.1 nmoles/mg/min) and hexobarbital-dependent formation of H2O2 was higher than hydroxylation (Vmax: 7.0 nmoles/mg/min). Phenobarbital treatment led here to two affinity constnts for each activity measured, which however, were alike. The existence of low in addition to high affinity constants observed here might explain the difficulties seen hitherto in correlating hexobarbital binding and metabolism in this species. Total oxidase activity was higher than oxygenase activity in all species tested. It is suggested that oxygenase activity of cytochrome P-450 is not limited by binding but by a competition with oxidase activity for a common intermediary species. This might be peroxy-P-450 (substrate-Fe3+O2(2-), rendering either substrate-Fe3+ O for hydroxylation reaction, or oxidized cytochrome P-450-substrate and hydrogen peroxide as product of oxidase function.

The epoxide-diol pathway in the metabolism of vinylbital in rat and man

1. In urine of rats given vinylbital (5-vinyl-5-(1'-methylbutyl)barbituric acid) i.p., unchanged vinylbital and its devinylated metabolite, 5-(1'-methylbutyl)barbituric acid, were identified. Rats synthetic 1',2'-epoxyvinylbital excreted the same compound as a major metabolite. No unchanged epoxide, nor 1',2'-dihydroxyvinylbital, could be identified in the urine of rats treated with vinylbital or its epoxide. 2. Attempts to synthesize 1',2'-dihydroxyvinylbital from 1',2'-epoxyvinylbital by acidic hydrolysis revealed that this possible metabolite decomposes readily to 5-(1'-methylbutyl)barbituric acid by a 'retro-aldol type' reaction. 3. In rat liver microsomal preparation 1',2'-epoxyvinylbital is readily hydrated by epoxide hydratase, but this reaction is almost completely inhibited by 0.8 mM 1,1,1-trichloropropene-2,3-oxide (TCPO). This finding and the identification of 5-(1'-methylbutyl)barbituric acid as end-product of this enzyme reaction provides further evidence for the existence of an epoxide-diol pathway in the metabolism of vinylbital. 4. Vinylbital and its devinylated metabolite are excreted in 36 h urine of rats treated with vinylbital, to an extent of 0.6 +/- 1.7% of the dose (n = 5), respectively. Upon administration of 1',2-epoxyvinylbital, 59.8 +/- 14.2% of the dose (n = 5) was excreted as 5-(1'-methylbutyl)barbituric acid. 5. In 60 h urine of three human volunteers who had taken 150 mg of vinylbital orally, 2.6 +/- 1.7% of the dose was excreted as vinylbitaland 11.0 +/- 4.1% as 5-(1'-methylbutyl) barbituric acid, illustrating that also in humans the epoxide-diol pathway plays a role in the metabolism of vinylbital.