Trimethadione metabolism by human liver cytochrome P450: Evidence for the involvement of CYP2E1

Article abstract of DOI:10.1080/004982598238930

1. Caucasian liver samples were used in this study. N-demethylation of trimethadione (TMO) to dimethadione (DMO) was monitored in the presence of chemical inhibitors of CYPs, such as fluconazole, quinidine, dimethyl-nitrosamine, acetaminophen, phenacetin, chlorzoxazone and mephenytoin. Trimethadione N-demethylation was selectively inhibited by dimethylnitrosamine and chlorzoxazone (> 50%) and weakly inhibited by tolbutamide (12%) and fluconazole (22%), whereas other inhibitors showed no effect. This result suggested that TMO metabolism to DMO is mainly mediated by CYP2E1 and marginally by CYP2C and CYP3A4. 2. Fifteen human livers were screened and interindividual variability of TMO N-demethylation activity was 3-fold. Chlorzoxazone 6-hydroxylation activity was also measured and both activities were significantly correlated (r = 0.735, p < 0.01). 3. DMO production by human cDNA expressed CYP enzymes was observed mainly for CYP2E1 (10.8 nmol/tube), marginally for CYP2C8 (0.22 nmol/tube) and not detectable for other CYP enzymes. 4. These results indicate that TMO metabolism is primarily catalysed by CYP2E1 and that trimethadione would be a suitable selective probe drug for the estimation of human CYP2E1 activity in vivo.

Full text of DOI:10.1080/004982598238930

xenobiotica , 1998, vol. 28, no. 11, 1041±1047
Trimethadione m etabolism by hum an liver
cytochrom e P450 : evidence for the involvem ent of
CYP2E1
N. KURATA,‹* Y. NISHIMURA,‹, M. IWASE,‹
N. E. FISCHER,Œ B. K. TANG,Œ T. INABAŒ
and H. YASUHARA‹
‹
Department of Pharmacology, School of Medicine, Showa University, 1-5-8
Hatanodai, Shinagawa-ku, Tokyo 142-8555, Japan
ΠDepartment of Pharmacology, Faculty of Medicine, University of Toronto, Toronto
M5S 1A8, Canada
Received 24 March 1998
1
. Caucasian liver samples were used in this study. N-demethylation of trimethadione
TMO) to dimethadione (DMO) was monitored in the presence of chemical inhibitors of
(
CYPs, such as ¯uconazole, quinidine, dimethyl-nitrosamine, acetaminophen, phenacetin,
chlorzoxazone and mephenytoin. Trimethadione N-demethylation was selectively in-
"
hibited by dimethylnitrosamine and chlorzoxazone ( 50 %) and weakly inhibited by
tolbutamide (12 %) and ¯uconazole (22 %), whereas other inhibitors showed no eåect. This
result suggested that TMO metabolism to DMO is mainly mediated by CYP2E1 and
marginally by CYP2C and CYP3A4.
2. Fifteen human livers were screened and interindividual variability of TMO N-
demethylation activity was 3-fold. Chlorzoxazone 6-hydroxylation activity was also
!
measured and both activities were signi®cantly correlated (r = 0.735, p 0.01).
3
. DMO production by human cDNA expressed CYP enzymes was observed mainly
for CYP2E1 (10.8 nmol}tube), marginally for CYP2C8 (0.22 nmol}tube) and not de-
tectable for other CYP enzymes.
4. These results indicate that TMO metabolism is primarily catalysed by CYP2E1 and
that trimethadione would be a suitable selective probe drug for the estimation of human
CYP2E1 activity in vivo.
Introduction
Trimethadione (TMO) has been employed as a probe drug for the estimation of
hepatic drug metabolising enzyme capacity in vivo (Tanaka et al. 1985a, 1989,
Kobayashi et al. 1984, Tanaka and Misawa 1985, Sakai et al. 1989). This compound
has several bene®cial properties as an indicator drug (Butler and Poole 1965) since
it is well absorbed from the gastrointestinal tract after oral administration, not
bound to plasma protein, only metabolized by hepatic drug-metabolizing enzymes,
and produces one metabolite, dimethadione (DMO) (®gure 1). It has also been well
characterized as an indicator drug in rat by Nishimura et al. (1992, 1995) and Tanaka
et al. 1985a±c). Nishimura et al. (1996) recently concluded that cytochrome P450
2
E1 (CYP2E1) is the main enzyme catalysing TMO N-demethylation in rat both in
vivo and in vitro.
In contrast, Nakamura et al. (1994) reported that TMO was metabolized by 14
CYP enzymes puri®ed from rat liver. The enzymes with higher than median N-
demethylation activity were CYPs 2A2, 2B1, 2B2, 2C7, 2C11, 2C12 and 2E1.
Nakamura’s group also tested three puri®ed CYP enzymes from human liver
*
Author for correspondence.
049±8254}98 $12.00 ’ 1998 Taylor & Francis Ltd
0

1
042
N. Kurata et al.
Figure 1. Chemical structure and metabolic pathway of trimethadione.
microsomes and reported the N-demethylation activity in the following order:
"
"
3
A4 2C
The present study was designed to investigate whether our previous rat data
Nishimura et al. 1992, 1995), indicating the major involvement of CYP2E1, is
1A2 (2E1 was not tested).
(
also relevant for human liver. We used commercially available human cDNA
expressed CYP enzymes as well as human liver microsomes with known selective
chemical inhibitors of various CYP enzymes.
M aterials and m ethods
Chemicals
Trimethadione, chlorzoxazone, 7-hydroxypropyltheophylline, tolubutamide, N,N-dimethylnitro-
samine, quinidine, acetaminophen, phenacetin and NADPH were purchased from Sigma Chemical Co.
(
St Louis, MO, USA). Dimethadione was purchased from Nachalai Tesque (Kyoto, Japan). 2,4-
Thiazolidinedione was purchased from Aldrich Chemical Company, Inc. (Milwaukee, WI, USA). Other
chemicals used were reagent or higher grade. Human B-lymphoblastoid cell- and -yeast expressed CYP
enzymes were obtained from Daiichi Pure Chemical Co. (Tokyo, Japan) and Sumika Chemical Analysis
Service Ltd (Tokyo, Japan), respectively.
Preparation of human liver microsomes
Fifteen human liver samples (all of Caucasian origin) stored at –70 C in the tissue bank of the
Ê
Department of Pharmacology, University of Toronto, were used in this study. The preparation of human
liver microsomes was based on a standard centrifugation method (Jurima-Romet et al. 1994). The frozen
liver tissue was weighed, thawed, and placed in 5 vol ice-cold 1.15 % (w}v) KCl. After brief
homogenization with a Brinkman Polytron for 15 s, the liver tissue was further homogenized with a
Potter±Elvehjem glass±Te¯on tissue grinder. The resulting homogenate was then centrifuged at 9000 g
for 20 min at 4 C. The supernatant was further centrifuged at 105000 g at 4 C for 60 min. The
Ê
Ê
remaining pellet was resuspended in 0.05 m Tris-HCl buåer, pH 7.6, gently homogenized in an ice bath
and centrifuged again at 105000 g for 60 min. The washed microsome pellet was ®nally resuspended in
a solution containing 0.3 m sucrose and 0.05 m Tris-HCl buåer (pH 7.6) and the volume was adjusted to
1
in prepared microsomal suspension was determined by the Pierce Chemical BCA assay method.
ml}g liver. This preparation was stored at –70 C and used within 2 months. The protein concentration
Ê
TMO N-demethylation assay development
Stock solutions of DMO (200 nmol}ml in water and ethyl acetate) were prepared from solid material
on the day of analysis. Working standard solution of DMO (20, 10, 5, 2, 1, 0.5, 0.2 and 0.1 nmol}ml in
water) were prepared by serial dilution with incubation mixture containing TMO and microsomal
protein. A working solution of internal standard (IS) was prepared by dilution of the stock solution
(
1 mg}ml) to a concentration of 100 lg}ml with water. This solution was used for addition of the internal
standard to each extraction.
Extraction eæ ciency
The eæciency of extraction of DMO and internal standard, 2,4-thiazolidinedione, were calculated
from the area ratios (DMO}IS) obtained from these samples (R´) and the area ratio obtained from ethyl

Trimethadione metabolism in human liver microsomes
1043
Table 1. Trimethadione metabolism in four human livers: Enzyme kinetic parameters for
N-demethylation.
Liver no.
Km (mm)
Vmax (pmol}mg protein}min)
Vmax }Km (ml}mg}hr)
K14
K19
K24
K26
2.21
3.51
5.41
4.14
622
531
378
413
4.69
2.52
1.16
1.66
mean‰ SD
3.82‰ 1.15
486‰ 97
2.49‰ 1.35
acetate solution containing the same amount of DMO and internal standard (R) with the following
equation :
Percentage eæciency = (R´}R)3 100%.
The percentage eæciency of this extraction procedure for DMO and internal standard were 94.2 and
89.3, respectively, and the coeæcient of variation for DMO and internal standard were 3.55 and 2.98 %,
respectively. By using this procedure, the relationship between the ratio of chromatographic areas
(
DMO}IS) and the DMO concentration was linear from 0 to 20 nmol}ml of reaction mixture. The
minimum detectable concentration of DMO was 0.05 nmol}ml. The retention time of DMO, TMO and
IS were 3.7, 5.2 and 9.4 min, respectively.
Trimethadione N-demethylase activity in human liver microsomes
Enzyme reactions were performed as follows. The reaction mixture consisted of 0.35 ml 0.1 m Na-K
phosphate buåer, pH 7.4, 0.05 ml 10 mm NADPH, 0.05 ml of microsomal suspension (protein content
was 20 mg}ml) and 0.05 ml 20 mm TMO. The reaction mixture (total volume of 0.5 ml) was combined
in a glass test tube and incubated for 20 min at 37 C. The reaction was terminated by placing in an ice
Ê
bath. Extraction of the metabolite of TMO, dimethadione, was carried out using glass tubes with 0.4 ml
5
m monosodium phosphate, 200 mg magnesium sulphate, 200 mg sodium sulphate, 0.2 ml aliquot of
reaction mixture, 1 ml ethyl acetate and the internal standard (5 lg 2,4-thazolidinedione). The extraction
mixture was mixed by vortex mixer for 30 s and then centrifuged at 1600 g for 20 min. The solvent layer
was transferred to a new glass tube and concentrated carefully to 0.1 ml under a nitrogen stream at room
temperature. Trimethadione N-demethylase activity was analysed by a gas-chromatograph (GC)
equipped with a ¯ame ionization detector (FID). Resulting extracts (2-ll aliquots) were injected onto the
GC. A Shimadzu GC-9A, equipped with DB-1 widebore capillary column (0.52 mm3 15 m), was
employed for the analysis of DMO. Injection port and column oven temperature were 200 C and
7
Ê
0±200 C, respectively. Nitrogen was used as the carrier gas. Gas ¯ows of air, hydrogen and nitrogen
Ê
were 500, 50 and 25 ml}min, respectively. GC data were integrated with a Shimadzu C-R4A integrater.
Trimethadione N-demethylase activity in cDNA-expressed CYP isoforms
The reaction mixture consisted of 0.7 ml 0.1 m Na-K phosphate buåer, pH 7.4, 0.1 ml NADPH
generation system containing 1 unit of glucose 6-phosphate (G6P) dehydrogenase, 1 lmol NADP 1and
8
0
lmol G6P, 0.1 ml expressed CYP suspension (protein content was approximately 10 mg}ml) and
.1 ml 20 mm TMO. The reaction mixture (®nal volume of 1 ml) was incubated for 120 min at 37 C. The
Ê
procedure after this step was the same as described above. Trimethadione N-demethylase activity was
analysed by a GC equipped with a ¯ame thermionic detector (FTD). A Shimadzu GC-14A equipped
with DB-1 widebore capillary column (0.52 mm3 15 m) was employed. Injection port and column oven
temperatures were 200 and 70±200 C, respectively. Helium was used as the carrier gas. Gas ¯ows of air,
hydrogen and helium were 150, 50 and 25 ml}min, respectively. GC data were integrated with a
Ê
Shimadzu C-R7A integrater.
HPLC assay method for chlorzoxazone 6-hydroxylase activity
The analytical procedure for human microsomal chlorzoxazone 6-hydroxylase activity was performed
as follows. Chlorzoxazone 6-hydroxylation was carried out in test tubes and the reaction mixture
consisted of 0.32 ml 0.1 m Na-K phosphate buåer, pH 7.4, containing of 1 lmol NADPH and 10 ll
human liver microsome suspension containing 200 lg protein. The mixture (®nal volume 0.45 ml) was
preincubated at 37 C for 3 min before adding chlorzoxazone (250 lm) and the enzyme reaction was
Ê
carried out for 20 min at 37 C. The reaction was terminated with 50 ll 8 m sulphuric acid. Internal
Ê
standard, 7-hydroxypropyl-theophylline (5 lg) was added and extracted twice with 2 ml dichloromethane
containing 10 % isopropanol. Organic solvent layers were transferred to a new glass test tube and
evaporated under a nitrogen stream at room temperature. The residue was dissolved in 150 ll 0.1 %
acetic acid in distilled water containing 10 % ethanol and then a 30-ll aliquot was injected onto the
HPLC.

1
044
N. Kurata et al.
The HPLC conditions for the analysis of 6-hydroxy-chlorzoxazone were as follows. Acetic acid in
distilled water (0.1 %) containing 10 % ethanol was used as a mobile phase and the ODS column was
obtained from Beckman (5 lm, 4.6 mm3 25 cm). The ¯ow rate of the mobile phase was 1 ml}min and the
UV detector was set at 300 nm ; the retention time for 6-hydroxy-chlorzoxazone and internal standard are
2
0.3 and 17.4 min, respectively. HPLC data were integrated with a Shimadzu C-R4A integrater.
Data analysis
All results are represented as mean‰ SD. Statistical analysis was done by employing the ANOVA and
Student’s t-test. The enzyme kinetic parameters K and Vmax were obtained by a PC program, Enz®tter
m
(
Elsevier Biosoft).
Results
Metabolism of trimethadione by human liver microsomes:
Trimethadione N-demethylase activity and Km were determined in microsomes
from four human livers (table 1). The Km for the N-demethylation determined by
the substrate concentration range 1±10 mm ranged from 2.2 to 5.4 mm. Based on
this, subsequent following experiments were carried out using a substrate con-
centration of 2 mm TMO.
Chemical inhibition of TMO N-demethylation
To clarify the CYP isoform responsible for TMO metabolism, chemical
inhibition studies were performed by employing various CYP substrates and
inhibitors. The inhibitor concentration was 200 lm, one-tenth of the substrate
(
TMO) concentrate was selectively inhibited by dimethylnitrosamine and chlorz-
"
oxazone ( 50 % inhibition) and marginally inhibited by tolbutamide (12 %) and
uconazole (22 %) (Figure 2). Other chemical inhibitors for CYP1A2, CYP2C19
¯
and CYP2D6 did not inhibit TMO metabolism. This result indicated that the TMO
N-demethylation is mediated mainly by CYP2E1, and only marginally by CYP2C
and CYP3A4 (Figure 2).
Correlation between TMO metabolism and chlorzoxazone metabolism
Trimethadione N-demethylation and chlorzoxazone 6-hydroxylation in 15
human livers were compared (®gure 3). In these human liver microsomes TMO N-
demethylation activity ranged over 3-fold, whereas chlorzoxazone 6-hydroxylation
activity ranged over 5-fold. Both activities were signi®cantly correlated with a
!
correlation coeæcient, r, of 0.735 (n = 15, p 0.01). The intercept at the y-axis
suggested that some TMO metabolism occurred when chlorzoxazone hydroxylation
was zero, indicating some contribution of other CYP enzymes in TMO N-
demethylation.
Trimethadione metabolism by 11 human cDNA expressed CYP isoforms
Trimethadione metabolism by human B-lymphoblastoid-expressed CYP1A1,
1
2
A2, 2A6, 2B6, 2D6, 2E1 and 3A4, and yeast-expressed CYP2C8, 2C9, 2C18 and
C19 was determined (®gure 4). Dimethadione production from TMO in the
incubation medium was only catalysed by CYP2E1 (10.8 nmol}tube) and marginally
in CYP2C8 (0.22 nmol}tube). No formation or lower than detection limit
(
0.05 nmol}tube) of dimethadione was observed for the other nine expressed CYP
isoforms.

Trimethadione metabolism in human liver microsomes
1045
Figure 2. Chemical inhibition of trimethadione N-demethylation. Six CYP inhibitors}substrates were
tested using two human livers (K14, K24, n = 2, control activity was 204.21‰ 72.19 pmol}mg
protein}min (mean‰ SD)). DMNA, dimethylnitrosamine ; MP, mephenytoin; TB, tolu-
butamide ; AAP, acetaminophen ; Ph, phenacetin; QN, quinidine. Two additional CYP inhibitors
were tested using four more human livers (n = 6, control activity was 140.86‰ 66.07 (mean‰ SD)).
FL, ¯uconazole; CZ, chlorzoxazone. Inhibitor concentration was 0.2 mm and TMO was 2 mm.
Values are presented as mean‰ SD (n = 2±6).
Figure 3. Correlation of trimethadione metabolism and chlorzoxazone metabolism in 15 human livers.
Each data point is the average of two determinations.

1
046
N. Kurata et al.
Figure 4. Trimethadione metabolism in human B-lymphoblastoid cell- and yeasts-expressed
cytochrome P450s. Data are derived from two determinations.
Discussion
In the current study, the human liver CYP enzymes responsible for tri-
methadione N-demethylation have been examined. In a recent study, Nakamura et
al. (1994) reported that TMO metabolism was catalysed by three CYP enzymes,
namely 3A4, 2C and 1A2, puri®ed from human liver microsomes. Our present
chemical inhibition data (®gure 2) indicated only minor contributions from CYP3A4
and CYP2C, as indicated by weak inhibition by ¯uconazole and tolbutamide,
respectively, whereas our previous rat data showed no detectable contributions from
CYP3A and CYP2C in TMO N-demethylation. In our current study, no inhibitory
eåect of phenacetin was observed, indicating the lack of involvement of CYP1A2.
The most extensive inhibition of DMO production was observed in the presence of
dimethylnitrosamine and chlorzoxazone, typical CYP2E1 substrates}inhibitors.
These results support the proposal that TMO metabolism is mainly mediated by
CYP2E1.
Further evidence for CYP2E1 mediation is the signi®cant correlation in
1
5 human livers between chlorzoxazone 6-hydroxylation activity and TMO
N-demethylation activity (®gure 2). It is well established that chlorzoxazone
metabolism is mainly mediated by CYP2E1 and minimally by CYP1A2 (Peter et al.
1
990, Kharasch et al. 1993). Furthermore, TMO N-demethylation by expressed
CYPs was only observed for CYP2E1 and marginally for CYP2C8 (®gure 4).
Unfortunately, comparison of our current data with the report of Nakamura et al.
(
1994) was not possible, since they did not use puri®ed CYP2E1 from human liver.
In summary, our results indicate that TMO metabolism is mainly mediated by
CYP2E1 and TMO would be a suitable selective in vivo probe drug for the
estimation of CYP2E1 activity in man.
Acknow ledgem ents
This project was partially supported by Grant-in-Aid for Scienti®c Research
from the Ministry of Education, Science and Culture in Japan.

Trimethadione metabolism in human liver microsomes
1047
References
Butler , T. C. and Poole , D. T., 1965, Determination of trimethadione and metharbital by rat liver
microsomal enzymes: substrate concentration-yield relationships and competition between
substrates. Biochemical Pharmacology, 14, 937±942.
Jurima -Romet , M., Crawford, K., Cyr, T. and Inab, T., 1994, Terfenadine metabolism in human
liver: in vitro inhibition by macrolide antibiotics and azole antifungals. Drug Metabolism and
Disposition, 22, 849±857.
Kharasch, E. D., Thummel , K. E., Mhyre , J. and Lillibridge , J. H., 1993, Single dose disul®ram
inhibition of chlorzoxazone metabolism : a clinical probe for P4502E1. Clinical Pharmacology and
Therapeutics, 53, 643±650.
Kobayashi , S., Tanaka, E., Oguchi , K., Yoshida , T., Kuroiwa , Y. and Yasuhara, H., 1984, A method
for estimation of hepatic drug-metabolizing capacity; determination of concentration of
trimethadione and its metabolite in human serum. Journal of Pharmaco-Bio Dynamics, 7, 329±335.
Nakamura , M., Tanaka, E., Misawa , S., Shimada , T., Imaoka , S. and Funae, Y., 1994, Trimethadione
metabolism, a useful indicator for assessing hepatic drug-oxidizing capacity. Biochemical
Pharmacology, 47, 247±251.
Nishimura , Y., Kobayashi , S., Kurata, N., Uchida , E. and Yasuhara, H., 1992, The eåects of
inducers of diåerent types of cytochrome P450 isozymes on serum dimethadione}trimethadione
ratio in rats. Showa University Journal of Medical Sciences, 4, 7±14.
Nishimura , Y., Kurata, N., Kobayashi , S., Uchida , E. and Yasuhara, H., 1995, The eåects of
inhibitors and substrates of diåerent types of cytochrome P450 isozymes on serum dimethadione}
trimethadione ratio in rats in vivo. Research Communications in Chemical Pathology and
Pharmacology, 87, 145±154.
Nishimura , Y., Kurata, N., Watanabe , M., Uchida , E. and Yasuhara, H., 1996, Trimethadione N-
demethylation by rat liver CYP2E1 in vitro. Research Communications in Molecular Pathology and
Pharmacology, 93, 43±56.
Peter , R., Bocker, R., Beaune , P. H., Iwasaki, M., Guengerich , F. P. and Yang, C. S., 1990,
Hydroxylation of chlorzoxazone as a speci®c probe for human liver cytochrome P-450 IIE1.
Chemical Research in Toxicology, 3, 566±573.
Sakai, H., Kobayashi , S., Uchida , E., Kai, K., Tanaka, E., Kurata, N. and Yasuhara, H., 1989, The
mechanism of inhibition of hepatic drug metabolizing enzymes by quinidine in wistar rats. Asia
Paci®c Journal of Pharmacology, 4, 197±202.
Tanaka, E., Kobayashi , S., Aramphongphan , A., Kurata, N., Oguchi , K., Kuroiwa , Y. and
Yasuhara, H., 1985a, The eåect of indomethacin on hepatic drug-oxidizing capacity in the rat:
trimethadione and antipyrine metabolism as an indicator. Journal of Pharmaco-Bio Dynamics, 8,
7
73±779.
Tanaka, E., Kobayashi , S., Nakamura , K., Uchida , E. and Yasuhara, H., 1989, Simpli®ed approach
for determination of hepatic drug-oxidizing capacity using trimethadione metabolism as an
indicator. Biopharmaceutics and Drug Disposition, 10, 617±620.
Tanaka, E., Kobayashi , S., Oguchi , K., Kuroiwa , Y. and Yasuhara, H., 1985b, The eåects of short-
term administration of ibuprofen on trimethadione metabolism and antipyrine metabolite
formation in the rat. Research Communications in Chemical Pathology and Pharmacology, 48,
3
17±320.
Tanaka, E. and Misawa , S., 1985, Eåects of cimetidine and ranitidine on trimethadione metabolism in
the rat. Research Communications in Chemical Pathology and Pharmacology, 47, 461±464.
Tanaka, E., Misawa , S. and Kuroiwa , Y., 1985c, Eåects of cimetidine and diethylaminoethyl 2,2-
diphenylvalerate HCL(SKF525-A) on trimethadione metabolism in the rat. Journal of Pharmaco-
Bio Dynamics, 8, 767±772.