![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 27, Issue 6, 717-723, June 1999
Departments of Anesthesiology, Medicinal Chemistry, and Pharmaceutics, University of Washington, Seattle, Washington and the Anesthesiology Service, Puget Sound Veterans Affairs Health Care System, Seattle, Washington
 |
Abstract |
|---|
 Top
 Abstract
 Introduction
Materials and Methods
Results
Discussion
References
|
|---|
Disulfiram and its primary metabolite diethyldithiocarbamate are effective mechanism-based inhibitors of cytochrome P-450 2E1 (CYP2E1)1 in vitro. Single-dose disulfiram diminishes CYP2E1 activity in vivo and has been used to identify CYP2E1 participation in human drug metabolism and prevent CYP2E1-mediated toxification. Specificity of single-dose disulfiram toward CYP2E1 in vivo, however, remains unknown. This investigation determined single-dose disulfiram effects on human CYP 2C9, 2C19, 2D6, and 3A4 activities in vivo. In four randomized crossover experiments, volunteers received isoform-selective probes (oral tolbutamide, mephenytoin, dextromethorphan, or i.v. midazolam) on two occasions, 10 h after oral disulfiram or after no pretreatment (controls). Plasma and/or urine parent and/or metabolite concentrations were measured by HPLC or gas chromatography-mass spectrometry. CYP2C9, 2C19, 2D6, and 3A4 activities were determined from the tolbutamide metabolic ratio, 4'-hydroxymephenytoin excretion, and dextromethorphan/dextrorphan ratios in urine and midazolam systemic clearance, respectively. Midazolam clearance (670 ± 190 versus 700 ± 240 ml/min, disulfiram versus controls), dextromethorphan/dextrorphan metabolic ratio (0.013 ± 0.033 versus 0.015 ± 0.035), 4'-hydroxymephenytoin excretion (122 ± 22 versus 128 ± 25 µmol), and tolbutamide metabolite excretion (577 ± 157 versus 610 ± 208 µmol) were not significantly altered by disulfiram pretreatment, although the tolbutamide metabolic ratio was slightly diminished after disulfiram (60 ± 17 versus 81 ± 40, p < .05). Results show that single-dose disulfiram does not cause clinically significant inhibition of human CYP2C9, 2C19, 2D6, and 3A4 activities in vivo. When single-dose disulfiram is used as an in vivo probe for P-450, inhibition of drug metabolism suggests selective involvement of CYP2E1. Single-dose disulfiram should not cause untoward drug interactions from inhibition of other P-450 isoforms.
| |
Introduction |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
|
|---|
The metabolism of numerous
halogenated xenobiotics used in industrial applications, endogenous
compounds, ethanol, and a few drugs such as chlorzoxazone, isoniazid,
acetaminophen, and volatile anesthetics is catalyzed by human liver
cytochrome P-450 (CYP) 2E1 (Lieber, 1997
). Potential consequences of
such metabolism include bioactivation resulting in toxification and/or
carcinogenesis. One approach for identifying CYP2E1 participation in
human xenobiotic biotransformation in vivo is to manipulate CYP2E1
activity, by induction or inhibition, and assess the effect on
candidate drug metabolism. When bioactivation portends toxification,
the latter approach is preferable. A corollary application of this
technique is that CYP2E1 inhibition may be used prophylactically, or
possibly postexposure, to prevent CYP2E1-mediated toxification.
Single-dose disulfiram inhibition of CYP2E1 has been explored as an in
vivo probe for CYP2E1 participation in biotransformation and as a
potential preventative agent against CYP2E1-mediated toxification.
Disulfiram and its reduced metabolite diethyldithiocarbamate inhibit
human liver microsomal CYP2E1 in vitro (Guengerich et al., 1991) and
rat CYP2E1 in vivo (Brady et al., 1991). Using chlorzoxazone
6-hydroxylation as a measure of CYP2E1 activity (Peter et al., 1990),
disulfiram was found to reduce chlorzoxazone elimination clearance and
6-hydroxychlorzoxazone formation clearance to 15% and 7% of control
values, respectively (Kharasch et al., 1993). This established
single-dose disulfiram as an effective inhibitor of human CYP2E1 in
vivo, a useful probe for delineating CYP2E1 participation in drug
disposition and for potential prevention of CYP2E1-mediated
toxification. Single-dose disulfiram was subsequently used to
establish CYP2E1 participation in human in vivo metabolism of
enflurane, sevoflurane, and halothane (Kharasch et al., 1994, 1995,
1996) and to diminish CYP2E1-mediated metabolism of halothane to
potentially toxic reactive intermediates (Kharasch et al., 1996).
Although initial in vitro investigations indicated that disulfiram and
diethyldithiocarbamate were selective, mechanism-based inhibitors of
CYP2E1 (Guengerich et al., 1991), their selectivity was subsequently
questioned. For example, diethyldithiocarbamate was also reported to
inhibit human liver microsomal P-450s 2A6 (Yamazaki et al., 1992; Chang
et al., 1994; Ono et al., 1996) and 2C19 (Ono et al., 1996) and, at
higher concentrations, both 2C8 and 3A3/4 (Chang et al., 1994).
Furthermore, although disulfiram efficacy toward CYP2E1 is
unquestioned, the in vivo specificity of single-dose disulfiram for
only CYP2E1 remains unknown. Accurate interpretation of clinical
studies using disulfiram as a (presumably) selective inhibitor of
CYP2E1 rests on this identification. Furthermore, before disulfiram can
be recommended to prevent CYP2E1-mediated drug or xenobiotic
bioactivation and toxicity, its safety, vis-a-vis the absence of
unwanted and potentially hazardous non-CYP2E1 drug interactions, must
be demonstrated. We recently showed that single-dose disulfiram does
not inhibit human CYP2A6 activity (coumarin hydroxylation) in vivo
(Kharasch et al., 1998). The present investigation determined the
effect of single-dose disulfiram on the other human P-450 isoforms
responsible for metabolizing the majority of therapeutically used
drugs, CYPs 2C9, 2C19, 2D6, and 3A4 (Guengerich, 1995; Wrighton et al.,
1996).
| |
Materials and Methods |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
|
|---|
Patient Selection and Clinical Protocol. Male and nonpregnant female nonsmoking volunteers participated in this investigation after written informed consent was obtained. The investigational protocols were approved by the Institutional Human Subjects Committee. Not all subjects participated in each study pair; group size was determined by prospective power analysis. Subjects were in good health, within 20% of ideal body weight, had no history of hepatic or renal disease, and were taking no prescription medications (excluding a few subjects on oral contraceptives) during the investigation. Subjects abstained from caffeine, grapefruit, grapefruit juice, and ethanol beginning the day before each study day and continuing throughout the period of urine collection. Subjects also abstained from ethanol for 5 days after disulfiram administration. Each substrate probe was studied in a crossover design, with subjects randomly assigned first to the control or disulfiram phase and a washout period of 1 to 2 weeks between phases. Pretreated subjects received 500 mg of disulfiram orally at bedtime, 9 to 10 h before substrate probes, whereas controls received nothing.
CYP3A4 activity was assessed by the clearance of midazolam (Thummel et al., 1994a,b). Subjects received 1 mg i.v. midazolam, and venous blood
samples were obtained through an indwelling catheter in the
contralateral arm for 12 h after dosing. Plasma was frozen at
20°C for later analysis. Based on the population variability of
midazolam disposition (Kassai et al., 1988), 15 subjects (30 ± 5 years, 72 ± 13 kg, 8:7 male/female) were studied to detect a 30%
difference in midazolam clearance (
= 0.05, 
= 80%). CYP3A4 activity was also estimated by the N-demethylation of
dextromethorphan to 3-methoxymorphinan (Jacqz-Aigrain et al., 1993;
Ducharme et al., 1996; Jones et al., 1996).
CYP2D6 activity was determined from the metabolism of dextromethorphan
(Schmid et al., 1985). Subjects received 30 mg oral dextromethorphan
hydrobromide (85.5 µmol of dextromethorphan) with 200 ml of water
after an overnight fast, followed 2 h later by a standard
breakfast. Urine was collected 0 to 8 h after dextromethorphan, the volume was measured, and an aliquot was frozen at 20°C for metabolite analysis. Forty subjects (28 ± 5 years, 69 ± 12 kg, 20:20 male/female) were studied to detect a 30% difference in dextromethorphan O-demethylation ( = 0.05, = 80%),
based on the population variability of dextromethorphan
O-demethylation and the expected incidence of poor
metabolizers (Evans et al., 1993).
CYP2C19 activity was determined by the 4'-hydroxylation of
S-mephenytoin (Wrighton et al., 1993), as described
previously (Kupfer and Preisig, 1984; Wedlund et al., 1984). Subjects
received racemic mephenytoin (100 mg orally with 200 ml of water; 229 µmol of S-mephenytoin), followed 2 h later by a
standard breakfast. Urine was collected from 0 to 8 h after
mephenytoin dosing, the volume was measured, and an aliquot was frozen
at 20°C for metabolite analysis. Fifteen subjects (30 ± 5 years, 72 ± 13 kg, 8:7 male/female) were studied, based on the
calculation that 13 subjects would be needed to detect a 30%
difference in mephenytoin metabolism ( = 0.05, = 80%) using the
published population variability of mephenytoin hydroxylation and the
expected incidence of poor metabolizers (Wilkinson et al., 1989;
Goldstein et al., 1997).
CYP2C9 activity was assessed by the metabolism of tolbutamide to
hydroxytolbutamide and its secondary metabolite carboxytolbutamide (Brian et al., 1989; Relling et al., 1990), as described previously (Peart et al., 1987; Veronese et al., 1990). Subjects received tolbutamide (500 mg orally, 1849 µmol) with 200 ml of water after breakfast. Urine was collected from 6 to 12 h after tolbutamide, the volume was measured, and an aliquot was frozen at 20°C for metabolite analysis. Subjects self-administered glucose tablets at
regular intervals until the midafternoon urine collection to prevent
hypoglycemia. Twenty subjects (29 ± 4 years, 70 ± 12 kg, 9:11 male/female) were studied to detect a 30% difference in
tolbutamide metabolic ratio ( = 0.05, = 80%), based on the
population variability of tolbutamide metabolism (Veronese et al.,
1993).
Analytical Methods. Dextromethorphan, dextrorphan, 3-methoxymorphinan, 3-hydroxymorphinan, levallorphan, tolbutamide, hydroxytolbutamide, and 4'-hydroxymephenytoin were obtained from Research Biochemicals, Inc. (Natick, MA), carboxytolbutamide was purchased from UltraFine Chemicals (Manchester, England), and all other chemicals were from Sigma (St. Louis, MO).
Plasma midazolam concentrations were determined by gas chromatography-mass spectrometry as described previously (Thummel, 1994b), with minor modification. Briefly, 1.5 ml of plasma containing the internal standard diazepam (25 ng) and 0.5 ml of 1 M sodium hydroxide was extracted with 2 × 3 ml of ethyl acetate-heptane (1:1, v/v). The combined organic layers were evaporated under nitrogen
at 40°C, reconstituted in 75 µl of acetate-heptane, and analyzed on
a Hewlett-Packard 5890 series II GC/5972 mass selective detector using
a DB-5 capillary column (30 m × 0.32 mm × 0.25 µm film
thickness) (J&W, Folsom, CA). The column head pressure (helium carrier
gas) was 25 psi for 1 min and then decreased to 8 psi. The oven
temperature was 200°C for 1 min, increased 10°C/min to 280°C, and
then to 320°C at 5°C/min. Injector and transfer line temperatures
were 290°C and 300°C, respectively. Ions monitored were
m/z 283.1 and 310.1 for diazepam and midazolam,
respectively. Standard curves were prepared daily (0.1-25 ng/ml) and
were linear (r2 > 0.99 over 0.1-5
and 0.1-25 ng/ml). The interday variability was 12, 7, and 1% at 0.1, 5, and 25 ng/ml, respectively.
Urine dextromethorphan and metabolite concentrations were determined by
HPLC with fluorescence detection as described (Chen et al., 1990), with
the following modifications. Urine (1 ml) was incubated overnight at
37°C with 1 ml of 0.1 M potassium acetate buffer (pH 5) containing
-glucuronidase (5000 U/ml), spiked with 5 µg of levallorphan and
diluted with 1 ml of saturated sodium carbonate. Samples were twice
extracted with 3 ml of diethyl ether/chloroform/isopropanol (20:9:1) by
vortexing for 10 min and centrifuged at 2000g for 10 min;
organic layers were combined into a 15-ml tapered polypropylene screw
cap tube containing 200 µl of 0.1 N HCl. Analytes were
back-extracted by vortexing for 10 min and centrifuging
(2000g, 10 min), and the organic layer was aspirated and the
aqueous phase evaporated to dryness under nitrogen at 55°C. Samples
were reconstituted in 100 µl of 0.1 N HCl, vortexed, and transferred
to an autosampler vial. Chromatography was performed with a Hewlett
Packard 1050 HPLC system coupled to a Kratos Spectroflow 980 fluorescence detector (excitation 228 nm, no emission cut-off filter)
using a Rainin Microsorb phenyl column (250 × 4.6 mm, 5 microns).
The mobile phase was methanol/acetonitrile/10 mM potassium phosphate
(pH 3.5) (20:25:55) at 1.2 ml/min. Analyses were carried out at room temperature. Each sample was injected twice (1 and 10 µl) to ensure that peaks were within the linear range of the detector. Calibration standards were prepared daily using blank urine containing 80 to 10,000 ng/ml dextrorphan and 3-hydroxymorphinan and 8 to 1000 ng/ml
dextromethorphan and 3-methoxymorphinan. Peak area ratios were used for
analysis of dextromethorphan and dextrorphan, and 3-methoxymorphinan
and 3-hydroxymorphinan were quantified from peak height ratios because
complete baseline resolution was unattainable in some patients.
Standard curves were linear over the concentration ranges used
(r2 > 0.99, 0.99, 0.98, and 0.96 for
dextromethorphan, dextrorphan, 3-methoxymorphinan, and
3-hydroxymorphinan, respectively). Interday coefficients of variation
were 25 and 2% at 16 and 1000 ng/ml dextromethorphan, 22 and 15% at
0.3 and 2.5 µg/ml dextrorphan, 19 and 2% at 16 and 1000 ng/ml
3-methoxymorphinan, and 18 and 8% at 0.3 and 2.5 µg/ml
3-hydroxymorphinan.
The concentration of 4'-hydroxymephenytoin in urine was measured by
HPLC with UV detection as described, with minor modification (Xie et
al., 1995). Urine (100 µl), water (900 µl), and 50 µl of 0.5 M
acetate buffer (pH 5.0, containing 5000 U/ml -glucuronidase) were
incubated in polyethylene tubes at 37°C for 18 h. The
hydrolysate was transferred to a glass tube containing 5 µg of
phenobarbital and twice extracted by vortexing with 3 ml of diethyl
ether. Following centrifugation (2000g, 10 min), the
combined organic layers were evaporated to dryness at 45°C under
nitrogen, reconstituted in 50 µl of mobile phase, and transferred to
an autosampler vial. Chromatography was performed at room temperature
on an HP 1050 HPLC system with variable wavelength UV detector at 204 nm, using a Rainin Microsorb-MV C-18 analytical column (5 microns,
250 × 4.6 mm) (Varian, Walnut Creek, CA) and Opti-Guard C-18 (1 mm) guard column (Optimize Technologies, Oregon City, OR). The mobile phase gradient began at 69:31 water/acetonitrile, increased linearly to
38:62 over 10 min, held for 2 min, returned to its initial composition
over 3 min, and then re-equilibrated for 5 min. The flow rate was 1.75 ml/min. Injection volume was 20 µl. Analyte stock solutions were
prepared in acetonitrile and used to formulate 4'-hydroxymephenytoin
standards in urine for calibration curves (0.5-25 µg/ml,
r2 = 0.999). The intraday coefficient
of variation for 4'-hydroxymephenytoin was 2% at 0.5 and 25 µg/ml.
All urine samples were analyzed on the same day.
Hydroxytolbutamide, carboxytolbutamide, and unchanged tolbutamide in
urine were determined by HPLC after glucuronidase treatment using a
modification of a previous method (Csillag et al., 1989). Briefly,
urine (1 ml) containing 20 µg of chlorpropamide (internal standard)
and 100 µl of 2 M HCl was twice extracted with 2 ml of diethyl ether
by vortexing and was then centrifuged for 10 min at 2000g.
The combined organic layers were evaporated to dryness under nitrogen
at 40°C, reconstituted in 100 µl of
methanol/acetonitrile/isopropanol/0.1% phosphoric acid buffer (pH 2.0)
(50:8.5:8.5:33) and transferred to autosampler vials. Analyses were
performed at room temperature on an HP 1050 HPLC system with variable
wavelength detector (235 nm) and a Rainin Microsorb-MV C-18 column (5 microns, 4.6 × 250 mm) at room temperature. The isocratic mobile
phase was acetonitrile/isopropanol/0.1% phosphoric acid (pH 2.0)
(17:17:66) at 1.0 ml/min, and the injection volume was 10 µl.
Tolbutamide in urine was also analyzed by HPLC-electrospray mass
spectrometry using an HP 1100 HPLC-MSD system and Rainin Microsorb C-18
analytical column (5 microns, 2.1 × 150 mm) coupled to a HAIPEEK
C-18 guard column. Injection volume was 5 µl, and the isocratic
mobile phase was acetonitrile/isopropanol/aqueous 0.5% trifluoroacetic
acid (20:20:80) at 200 µl/min. The mass spectrometer acquisition
parameters included positive ionization, quadrupole temperature 99°C,
gas temperature 350°C, 55 nA cap current, and drying gas at 10 liters/min. Tolbutamide and chlorpropamide were monitored at
m/z 271.2 and 277.2 (MH+), respectively. Analyte stock
solutions were prepared in methanol, and standard curves of peak area
ratios were created daily using drug-free urine (5-115 µg/ml
hydroxytolbutamide and carboxytolbutamide, 0.5-5.0 µg/ml tolbutamide) and used to quantify concentrations in patient samples. Standard curves were linear over the concentration ranges used (r2 > 0.995 for all analytes), and
the interday coefficient of variation was 8 and 10% for
hydroxytolbutamide at 5 and 115 µg/ml, 3 and 0.3% for
carboxytolbutamide at 5 and 115 µg/ml, and 7 and 2% for tolbutamide
at 0.5 and 5 µg/ml (intraday, liquid chromatography-mass spectrometry).
To verify that disulfiram had been ingested, subjects kept medication
diaries, and urine was analyzed for 2-thiothiazolidine-4-carboxylic acid (TTCA), an established human biomarker for the major disulfiram metabolite carbon disulfide and also for disulfiram itself (van Doorn
et al., 1981, 1982). TTCA was synthesized, and TTCA in urine was
analyzed by gas chromatography-mass spectrometry, as described previously (Johnson et al., 1996). Briefly, 1 ml of filtered urine and
0.8 ml of 1 N HCl were extracted with 2 ml of ethyl acetate, dried over
magnesium sulfate, evaporated to dryness, dissolved in methanol,
derivatized with diazomethane, evaporated to dryness, reconstituted in
acetonitrile, and analyzed by selected ion mode gas chromatography-mass
spectrometry (m/z 132;
M+-COOCH3). A linear standard curve
was prepared using blank urine and TTCA (10-1000 ng/ml,
r2 > 0.99) and used for
quantification. Urine samples were collected as described above and
also for 0 to 24 h after midazolam.
Data Analysis.
CYP2C9 activity was assessed by the
tolbutamide metabolic ratio [(hydroxytolbutamide + carboxytolbutamide)/tolbutamide] in the 6 to 12-h urine (Veronese et
al., 1990) and tolbutamide metabolite recovery (Peart, 1987). CYP2C19
activity was determined by the amount of 4'-hydroxymephenytoin excreted
in the 0 to 8-h urine, expressed also as the hydroxylation index
(µmol S-mephenytoin/µmol 4'-hydroxymephenytoin) (Kupfer
and Preisig, 1984; Wedlund et al., 1984; Xie et al., 1997). Individuals
excreting <2% of the dose as 4'-hydroxymephenytoin were considered
phenotypically poor metabolizers (Wedlund et al., 1984). The molar
dextromethorphan/dextrorphan metabolic ratio in 0 to 8-h urine was used
to determine CYP2D6 activity (Schmid et al., 1985; Jacqz-Aigrain et
al., 1993; Jones et al., 1996). A metabolic ratio 
0.3 was considered
evidence of CYP2D6 poor metabolizer phenotype (Schmid et al., 1985).
Midazolam systemic clearance (dose/area under the curve) was
determined by noncompartmental analysis (WinNonlin 1.5; Scientific
Consulting, Inc. Cary, NC) and used as a measure of hepatic CYP3A4
activity (Thummel et al., 1994a, 1994b).
| |
Results |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
|
|---|
Mean midazolam plasma concentration versus time profiles in control and disulfiram-treated subjects were superimposable (Fig. 1A). Systemic clearance (670 ± 190 versus 700 ± 240 ml/min, p = .39) was similarly unchanged by disulfiram (Fig. 1B).
|
Dextromethorphan disposition in untreated subjects resembled that in
previous reports (Jacqz-Aigrain et al., 1993; Ducharme et al., 1996;
Jones et al., 1996). Urine excretion (0-8 h) of dextrorphan,
3-methoxymorphinan, 3-hydroxymorphinan, and dextromethorphan was
27 ± 12, 0.05 ± 0.06, 14 ± 8, and 0.3 ± 0.5 µmol, respectively, representing 32 ± 15, 0.06 ± 0.07, 16 ± 10, and 0.3 ± 0.5% of the dose, with 48 ± 20%
metabolite recovery. Dextromethorphan O-demethylation was
unaffected by disulfiram pretreatment (Fig. 2A). The mean
dextromethorphan/dextrorphan metabolic ratio was unchanged by
disulfiram (0.013 ± 0.033 versus 0.015 ± 0.035, p = .26), and no subject was converted to a
phenotypically poor metabolizer. There was an excellent correlation
(r = 0.94, not shown) between the two ratio measures of
CYP2D6 activity, dextromethorphan/dextrorphan and
dextromethorphan/3-hydroxymorphinan, as described previously (Jones et
al., 1996). The mean dextromethorphan/3-hydroxymorphinan metabolic
ratio was also unchanged by disulfiram (0.028 ± 0.061 versus
0.032 ± 0.064, p = .22, data not shown).
Dextromethorphan N-demethylation, another putative marker of
CYP3A4 activity, was also unaffected by disulfiram pretreatment, with
no change observed in the dextromethorphan/3-methoxymorphinan metabolic
ratio (Fig. 2B). Neither molar excretion of dextromethorphan and
metabolites nor metabolite recovery was significantly different after
disulfiram pretreatment.
|
Mephenytoin metabolism in controls was similar to that described
previously (Kupfer and Preisig, 1984; Xie et al., 1997), averaging
128 ± 25 µmol of 4'-hydroxymephenytoin (56 ± 11% of the
dose) excreted in 0 to 8-h urine, corresponding to a hydroxylation index of 1.9 ± 0.5. One subject could not void in the control phase, and two subjects were found to be poor metabolizers; these were
excluded from the statistical analysis. Single-dose disulfiram effects
on mephenytoin metabolism are shown in Fig.
3. No subject was converted to a
phenytypic poor metabolizer. Average 4'-hydroxymephenytoin excretion
(122 ± 22 µmol; hydroxylation index 1.9 ± 0.4) was
unchanged compared with controls (p = .72). In
one subject, 4'-hydroxymephenytoin control excretion appeared low,
although s/he was not a poor metabolizer. Reanalysis without this
subject showed that 4'-hydroxymephenytoin excretion was somewhat lower
after disulfiram (120 ± 22 versus 134 ± 18 µmol;
hydroxylation index 2.0 ± 0.4 versus 1.7 ± 0.2, p = .045).
|
Tolbutamide metabolism in controls was similar to that described
previously (Veronese et al., 1990, 1993), with one difference. Urine
excretion (6-12 h) of hydroxytolbutamide (104 ± 37 µmol; 6 ± 2% of the dose), carboxytolbutamide (506 ± 179 µmol;
27 ± 10% of the dose), and percent of the dose recovered
(33 ± 11%) were similar to those reported by Veronese et al.
(4 ± 2, 22 ± 8, and 26 ± 9%) (1993). In contrast,
urinary tolbutamide excretion (8.5 ± 4.6 µmol; 0.46 ± 0.25% of the dose) was higher than that reported previously (0.49 ± 0.23 µmol; 0.027 ± 0.013% of the dose) (Veronese et al.,
1993). Consequently, the 6 to 12-h tolbutamide metabolic ratio we
observed (84 ± 39) was lower than those reported previously
(794 ± 87 and 1144 ± 529) (Veronese et al., 1990, 1993). This difference may be related in part to assay conditions. Urine tolbutamide concentrations were close to the limit of quantification for the HPLC-UV assay, as reported previously (Veronese et al., 1990).
Thus, urine tolbutamide was re-assayed using liquid chromatography-mass spectrometry, which afforded greater sensitivity and specificity. Concentrations measured by liquid chromatography-mass spectrometry were
approximately 2-fold greater than when measured by HPLC-UV and were
used to calculate the tolbutamide metabolic ratio reported.
The effect of single-dose disulfiram on tolbutamide metabolism is shown in Fig. 4. Tolbutamide metabolite excretion was unchanged (577 ± 157 versus 610 ± 208 µmol, p = .54) after disulfiram pretreatment (Fig. 4A). The tolbutamide metabolic ratio, however, was slightly, although significantly, diminished after disulfiram (60 ± 17 versus 81 ± 40, p = .032), and this difference persisted after reanalysis without the subject with unusually low urine tolbutamide recovery (60 ± 18 versus 74 ± 26, p = .033) (Fig. 4B).
|
TTCA excretion after control and disulfiram pretreatment is shown in
Fig. 5 for each isoform probe. TTCA was
present in urine of untreated subjects, as described previously, due
presumably to dietary sources (Simon et al., 1994). All subjects
instructed to ingest disulfiram the night before probe drug
administration, with the exception of one subject in the
dextromethorphan study, showed increased TTCA excretion.
|
| |
Discussion |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
|
|---|
Single-dose disulfiram effects on CYP3A4 activity, assessed by
systemic midazolam clearance and dextromethorphan
N-demethylation, can be compared against the influence of
known CYP3A4 inhibitors. Neither midazolam clearance nor the
dextromethorphan/3-methoxymorphinan metabolic ratio was affected by
disulfiram pretreatment. In contrast, known CYP3A4 inhibitors such as
macrolide antibiotics and azole antifungals markedly impair i.v.
midazolam clearance (Olkkola et al., 1993, 1994; Kharasch et al.,
1997). Furthermore, erythromycin pretreatment caused a significant 35 to 50% increase in the dextromethorphan/3-methoxymorphinan metabolic
ratio (Ducharme et al., 1996; Jones et al., 1996). These results
suggest that single-dose disulfiram does not significantly inhibit
CYP3A4 activity in vivo. Lack of CYP3A4 inhibition, based on unaltered
carbamazepine and methadone disposition, has also been seen with
chronic disulfiram administration (Enghusen Poulsen et al., 1992).
Single-dose disulfiram effects on dextromethorphan
O-demethylation can be compared against CYP2D6 activity in
genetically deficient poor metabolizers and the effects of known CYP2D6
inhibitors. Phenotypic poor metabolizers have a
dextrorphan/dextromethorphan metabolic ratio >0.3 (Schmid et al.,
1985). In extensive metabolizers, the CYP2D6 inhibitor quinidine (1 week pretreatment) increased the dextrorphan/dextromethorphan metabolic
ratio from 0.015 ± 0.061 to 1.9 ± 1.6, and all subjects but
one were converted from an extensive to a phenotypically poor
metabolizer (Zhang et al., 1992). Similarly, a single quinidine dose
administered to extensive metabolizers 12 h before
dextromethorphan increased the average metabolic ratio from 0.005 to
0.4 (Schadel et al., 1995). In contrast, single-dose disulfiram did not
significantly change the dextrorphan/dextromethorphan metabolic ratio,
and no subject was converted from an extensive to a poor metabolizer.
Thus, these data suggest that single-dose disulfiram does not
significantly inhibit CYP2D6 activity in vivo.
Single-dose disulfiram effects on S-mephenytoin
hydroxylation can be compared against CYP2C19 activity in genetically
deficient poor metabolizers and the effects of known CYP2C19 alternate
substrates. Poor metabolizers excrete less than 1 to 5% of
S-mephenytoin as 4'-hydroxymephenytoin (<2-11 µmol)
(Wedlund et al., 1984; Sohn et al., 1992; Kubota et al., 1996; Xie et
al., 1997). A single dose of chloroguanide increased the mephenytoin
hydroxylation index from 1.4 ± 0.2 to 2.5 ± 1.8, corresponding to a reduction in 4'-hydroxymephenytoin excretion from
164 to 92 µmol (Partovian et al., 1995). A single dose of
mephobarbital, administered shortly before mephenytoin, decreased
S-mephenytoin clearance by approximately one-third (Jacqz et
al., 1986). Chronic omeprazole treatment decreased urine
4'-hydroxymephenytoin excretion from 123 ± 16 to 74 ± 19 µmol (Caraco et al., 1996). In contrast to these 35 to 45% decreases in apparent CYP2C19 activity, single-dose disulfiram did not
significantly change urine 4'-hydroxymephenytoin excretion, and, if the
subject with low control 4'-hydroxymephenytoin excretion was omitted, it caused only an 11% decrease. Thus, these data suggest that single-dose disulfiram does not cause clinically significant inhibition of CYP2C19 activity in vivo.
Single-dose disulfiram effects on tolbutamide metabolism can be
compared against the effects of known CYP2C9 inhibitors. Sulfaphenazole decreased plasma tolbutamide clearance by 80%, caused a 2-fold decrease in urine hydroxytolbutamide and carboxytolbutamide excretion, a 3-fold increase in tolbutamide excretion, and reduced the 6 to 12-h
tolbutamide metabolic ratio to one-sixth that of controls (Veronese et
al., 1990). In contrast, disulfiram had no significant effect on urine
tolbutamide or metabolite excretion and caused only a 20% overall
decrease in the metabolic ratio. These results suggest that single-dose
disulfiram does not cause clinically significant inhibition of CYP2C9
activity in vivo. Chronic disulfiram administration also had no effect
on tolbutamide clearance (Svendsen et al., 1976).
The observed lack of disulfiram effect on CYP 2C9, 2C19, 2D6, and 3A4 activities cannot be attributed to a failure of subjects to ingest the drug as instructed. Both the medication diaries and the urine monitoring clearly demonstrated disulfiram ingestion. Thus, single-dose disulfiram was administered but had minimal or no inhibitory effect on the above isoforms.
Effects of single-dose disulfiram differ from those of chronic
disulfiram administration. Single-dose disulfiram diminishes CYP2E1
activity by more than 90% (Kharasch et al., 1993), and inhibition is
relatively selective for this isoform. Daily (>4 days) disulfiram use
did not alter CYP2C9 or CYP3A4 activities but did decrease the
clearances of antipyrine, theophylline, caffeine, phenytoin, diazepam,
and chlordiazepoxide (Enghusen Poulsen et al., 1992), suggesting
inhibition of P-450s 1A2 and 2C19. However, even daily use decreased
CYP1A2 (caffeine and theophylline clearance) and 2C19 (phenytoin
clearance) activities by only 30 to 35% (Svendsen et al., 1976; Beach
et al., 1986; Loi et al., 1989). Differences between single-dose and
chronic disulfiram effects on extent and selectivity of inhibition
likely result from the relative selectivity of disulfiram and
diethyldithiocarbamate toward CYP2E1 (Guengerich et al., 1991),
combined with rapid disulfiram elimination, which reduces inhibitor
concentrations at the time of study (10 h after disulfiram dosing)
(Petersen, 1992), thereby maximizing mechanism-based and minimizing
competitive components of P-450 inhibition. Consistent with this
hypothesis is the observation that antipyrine clearance was diminished
10, 16, and 32% after 1, 3, and 5 days of disulfiram (Loft et al.,
1986). Thus, single-dose disulfiram has unique, isoform-selective
inhibitory characteristics. These conclusions pertain only to
mechanism-based inhibition, as competitive components of disulfiram
interactions were not evaluated.
Results of the current investigation, together with previous
observations that single-dose disulfiram did not significantly diminish
CYP2A6 activity (coumarin hydroxylation) (Kharasch et al., 1998), while
profoundly inhibiting chlorzoxazone 6-hydroxylation (Kharasch et al.,
1993), demonstrate that single-dose disulfiram is a highly effective
and selective inhibitor of human CYP2E1 activity in vivo. These results
also support the selectivity of chlorzoxazone as an in vivo human
CYP2E1 probe. There are two implications of the single-dose disulfiram
attribute of CYP2E1 selectivity with respect to clinical P-450 isoform
typing and therapeutic interventions. First, single-dose disulfiram
inhibition of candidate drug disposition suggests involvement of CYP2E1
but not CYPs 2A6, 2C9, 2C19, 2D6, and 3A4. Second, single-dose
disulfiram use to prevent CYP2E1-mediated drug or xenobiotic
bioactivation is unlikely to result in untoward drug interactions
resulting from inhibition of other P-450 isoforms and potentially
hazardous alteration in therapeutic drug disposition.
| |
Acknowledgment |
|---|
We thank Tauri Senn for determination of midazolam pharmacokinetic parameters.
| |
Footnotes |
|---|
Received December 18, 1998; accepted February 23, 1999.
Supported by National Institutes of Health Grants R01 GM48712, P01 GM32165, and M01 RR00037 to the University of Washington Clinical Research Center.
Send reprint requests to: Evan D. Kharasch, Department of Anesthesiology, Box 356540, University of Washington, Seattle, WA 98195. E-mail: kharasch{at}u.washington.edu
| |
Abbreviations |
|---|
Abbreviation used is: TTCA, 2-thiothiazolidine-4-carboxylic acid.
| |
References |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
|
|---|
-naphthoflavone and diethyldithiocarbamate as selective chemical probes for inhibition of human cytochromes P450.
Arch Biochem Biophys
311:
437-442[Medline].This article has been cited by other articles:
|
|
 |
|
  M. G. Emery, C. Jubert, K. E. Thummel, and E. D. Kharasch Duration of Cytochrome P-450 2E1 (CYP2E1) Inhibition and Estimation of Functional CYP2E1 Enzyme Half-Life after Single-Dose Disulfiram Administration in Humans J. Pharmacol. Exp. Ther., October 1, 1999; 291(1): 213 - 219. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
