Inhibition of CYP2C9 by selective serotonin reuptake inhibitors: In vitro studies with tolbutamide and (S)-warfarin using human liver microsomes

Article abstract of DOI:10.1007/s002280050580

Objective: To investigate the in vitro potential of selective serotonin reuptake inhibitors (SSRIs) to inhibit two CYP2C9-catalysed reactions, tolbutamide 4-methylhydroxylation and (S)-warfarin 7-hydroxylation. Methods: The formation of 4-hydroxytolbutami

Full text of DOI:10.1007/s002280050580

Eur J Clin Pharmacol (1999) 54: 947±951
Ó Springer-Verlag 1999
PHARMACOKINETICS AND DISPOSITION
A. Hemeryck á C. De Vriendt á F. M. Belpaire
Inhibition of CYP2C9 by selective serotonin reuptake inhibitors:
in vitro studies with tolbutamide and (S)-warfarin using human liver
microsomes
Received: 20 July 1998 / Accepted in revised form: 6 October 1998
Abstract Objective:Toinvestigatetheinvitropotentialof Key words CYP2C9 á SSRI á (S)-warfarin
selective serotonin reuptake inhibitors (SSRIs) to inhibit
two CYP2C9-catalysed reactions, tolbutamide 4-methyl-
hydroxylation and (S)-warfarin 7-hydroxylation.
Methods: The formation of 4-hydroxytolbutamide from
Introduction
tolbutamide and that of 7-hydroxywarfarin from (S)- Selective serotonin reuptake inhibitors (SSRIs) interfere,
warfarinasafunctionofdi€erentconcentrationsofSSRIs both in vitro and in vivo, with cytochrome P450-medi-
and some of their metabolites was studied in microsomes ated drug metabolism. Most studies on this subject
from three human livers.
Results: Both tolbutamide 4-methylhydroxylation and by CYP1A2, CYP2C19, CYP2D6 and CYP3A4 [1]. Few
S)-warfarin 7-hydroxylation followed one enzyme data are available on the e€ect of SSRIs on CYP2C9-
concern interactions with SSRIs for pathways mediated
(
Michaelis-Menten kinetics. Kinetic analysis of 4-hy- mediated drug biotransformation. Only one in vitro
droxytolbutamide formation yielded a mean apparent study investigated systematically the e€ect of SSRIs on
Michaelis-Menten constant (K ) of 133 lM and a CYP2C9 activity, using phenytoin as substrate [2].
m
mean apparent maximal velocity (Vmax) of 248 pmol á p-Hydroxylation of phenytoin is mainly catalysed by
)1
)1
min á mg ; formation of 7-hydroxywarfarin yielded a CYP2C9 and to a lesser extent by CYP2C19, and
mean K of 3.7 lM and a mean V of 10.5 pmol á therefore phenytoin cannot be considered as a model
m
max
)1
)1
min á mg . Amongst the SSRIs and some of their substrate for CYP2C9 [3]. The aim of this study was to
metabolites tested, only ¯uvoxamine markedly inhibited evaluate the inhibitory potential of SSRIs and some of
both reactions. The average computed inhibition constant their metabolites on the in vitro metabolism of tolbut-
(
K ) values and ranges of ¯uvoxamine when tolbutamide amide and (S)-warfarin using human liver microsomes.
i
and (S)-warfarin were used as substrate, were 13.3 (6.4± Tolbutamide was chosen as substrate because tolbut-
7.3) lM and 13.0 (8.4±18.7) lM, respectively. The av- amide 4-methylhydroxylation is generally regarded as a
erage K value of ¯uoxetine for (S)-warfarin 7-hydroxyl- marker of CYP2C9 activity [4]. (S)-warfarin, the phar-
1
i
ation was 87.0 (57.0±125) lM.
macologically active enantiomer of warfarin, was also
Conclusion: Amongst the SSRIs tested, ¯uvoxamine was studied as substrate because it is mainly metabolized by
shown to be the most potent inhibitor of both tolbutamide CYP2C9 to form 7-hydroxywarfarin [5]. We also ex-
4-methylhydroxylation and (S)-warfarin 7-hydroxyl- amined in vitro, whether the potentiation of the anti-
ation. Fluoxetine, nor¯uoxetine, paroxetine, sertraline, coagulant e€ect of warfarin by some SSRIs, reported
desmethylsertraline, citalopram, desmethylcitalopram in vivo, can be explained by inhibition of (S)-warfarin's
had little or no e€ect on CYP2C9 activity in vitro. This is main metabolic pathway [6±8].
consistent with in vivo data indicating that amongst the
SSRIs, ¯uvoxamine has the greatest potential for inhibi-
ting CYP2C9-mediated drug metabolism.
Materials and methods
Chemicals and reagents
A. Hemeryck á C. De Vriendt á F.M. Belpaire (&)
Heymans Institute of Pharmacology,
University of Gent Medical School,
b-NADP,
D-glucose-6-phosphate, glucose-6-phosphate dehydro-
genase, tolbutamide, 7-ethoxycoumarin, chlorpropamide and
sulphaphenazole were purchased from Sigma Chemicals (St. Louis,
Mo., USA). 4-Hydroxytolbutamide, (S)-warfarin and 7-hydroxy-
warfarin were obtained from Ultra®ne Chemicals (Manchester,
9000 Gent, Belgium
e-mail: frans.belpaire@rug.ac.be
Tel.: +32-9-2403355; Fax: +32-9-2404988

948
UK). Sertraline hydrochloride and desmethylsertraline maleate
(
For the preparation of the (S)-warfarin stock solution [1 mM
P®zer Pharmaceuticals, New York, N.Y., USA), ¯uoxetine hy- (S)-warfarin in 5 mM KOH], 0.0145 mmol (S)-warfarin was dis-
drochloride, nor¯uoxetine hydrochloride (Eli Lilly, Indianapolis, solved in 70 ll 1 M KOH and further diluted with water to a ®nal
Ind., USA), ¯uvoxamine maleate (Solvay Duphar, Weesp, The volume of 14.5 ml. Each incubation mixture of (S)-warfarin con-
Netherlands), citalopram hydrobromide, desmethylcitalopram hy- tained a ®nal concentration of 0.5 mM KOH; the ®nal pH of the
drochloride (Lundbeck, Copenhagen, Denmark) and paroxetine incubation mixtures was 7.4.
hydrochloride (Smithkline Beecham Pharmaceuticals, Welwyn,
UK) were obtained from the manufacturers.
Inhibition experiments
The following compounds were tested in the same three livers for
their inhibitory e€ect at substrate concentrations of 200 lM tol-
Microsomal preparations
butamide and 4 lM (S)-warfarin: sulphaphenazole (0.1±100 lM),
citalopram (0.1±100 lM), desmethylcitalopram (0.1±100 lM), ¯u-
oxetine (0.1±100 lM), nor¯uoxetine (0.1±100 lM), ¯uvoxamine
Liver samples from three healthy organ donors (aged 7, 32 and 62
years) were used, with the approval of the local medical ethics
committee. The samples were sliced into portions and placed in
vials, frozen in liquid nitrogen and stored at )80 °C until prepa-
ration of microsomes by di€erential ultracentrifugation [9]. The
microsomal pellets were suspended in a 0.05 M potassium phos-
phate bu€er (pH 7.4). The protein content of the microsomal
preparations was determined by the method of Bradford [10].
Microsomal preparations were aliquoted, frozen and stored at
(0.1±100 lM), sertraline (0.1±100 lM), desmethylsertraline (0.1±
30 lM) and paroxetine (0.1±100 lM). The experiments were re-
peated for those substances displaying marked inhibition, at dif-
ferent tolbutamide (25±1000 lM) and (S)-warfarin (0.75±50 lM)
concentrations in order to determine the apparent inhibitory con-
stant (K
i
).
)
80 °C until use.
HPLC assay of 4-hydroxytolbutamide
-Hydroxytolbutamide was quanti®ed with an HPLC method ac-
cording to Miners et al. with slight modi®cations [11]. The HPLC
system consisted of a Gilson 307 pump and a Gilson 234 automatic
injector (Gilson, Middelton, Wisc., USA) with a 100±ll loop
4
Incubation conditions for 4-methylhydroxylation of tolbutamide
)
1
Tolbutamide was incubated with microsomal protein (1 mg á ml
)
and glucose-6-phosphate dehydrogenase (2 U á ml) ) in 100 mM
1
(
Rheodyne, Cotati, Calif., USA), a Waters 2487 tunable absorbance
detector (Waters, Milford, Mass., USA) and an HP 3395 integrator
Hewlett Packard, Palo Alto, Calif., USA). A Symmetry C18 col-
potassium phosphate bu€er (pH 7.4). The ®nal incubation volume
was 500 ll. The 4-hydroxytolbutamide formation rate from tol-
butamide was determined in a concentration range of 25 to
(
umn (150 mm ´ 3.0 mm ID, 5 lm; Waters, Milford, Mass., USA)
was used to separate 4-hydroxytolbutamide and the internal stan-
dard chlorpropamide at ambient temperature. The mobile phase
consisted of a 10-mM sodium acetate bu€er with 0.05 N triethyl-
1000 lM, in three human livers (HL 7, HL 9 and HL 10). After
pre-incubation (5 min at 37 °C) the reaction was initiated by the
addition of 50 ll of a NADPH-generating system [NADP (1 mM),
glucose-6-phosphate (10 mM) and MgCl
trations]. After incubation at 37 °C for 30 min in a shaking water
bath, the reaction was stopped by the addition of 50 ll HClO
0%; 2.17 nmol chlorpropamide was added as internal standard.
The samples were vortexed, followed by centrifugation at
6000 ´ g for 4 min. The pH of the supernatant was adjusted with 5
2
(5 mM), ®nal concen-
amine (adjusted to pH 4.3 with H
3 4
PO ), acetonitrile and methanol
(
71:25:4 v/v) mixture and was delivered at a ¯ow rate of 0.6 ml á
4
,
)
1
min . Detection was performed by UV absorbance at 230 nm. A
calibration curve was constructed by spiking microsomes not con-
taining the NADPH-generating system, with 4-hydroxytolbutamide
from 0.49 to 13.90 lM. Height ratios were determined utilizing the
internal standard method. The intra-day coecients of variation
6
1
N KOH to a value between 1.5 and 2. Samples were again vortexed
and centrifugated at 16000 ´ g for 4 min prior to injection of 25 ll
of the supernatant onto the HPLC column. Because
sulphaphenazole interfered in the HPLC assay of 4-hydroxytol-
butamide, an extraction step was required (see below) after the
(
4
n ˆ 6) at low (0.60 lM), medium (6.98 lM) and high (12.57 lM)
-hydroxytolbutamide concentrations were all below 3%; the inter-
day coecients of variation (n ˆ 3) at the low, medium and high
concentrations were all below 7%.
4
addition of HClO .
The samples incubated with tolbutamide and sulphaphenazole
were extracted after incubation and termination of the reaction by
HClO . After centrifugation, tolbutamide, 4-hydroxytolbutamide
In preliminary experiments it was ascertained that the forma-
tion of 4-hydroxytolbutamide from tolbutamide (1 mM) was linear
up to 2 mg protein ml)1 incubation mixture during 30 min incu-
bation time. The reaction was also linear throughout 40 min in-
4
and the internal standard were extracted from 400 ll incubation
cubation for a microsomal protein content of 1 mg á ml)
1
mixture in 2 ml ether by 3-min vortexing followed by centrifuga-
tion. The aqueous phase was discarded and the ether phase was
evaporated to dryness; the residue was reconstituted in 300 ll
mobile phase and 25 ll was injected onto the column. A calibration
curve was also extracted.
.
Incubation conditions for 7-hydroxylation of (S)-warfarin
The incubation method for (S)-warfarin 7-hydroxylation was the
same as that for tolbutamide 4-methylhydroxylation. (S)-warfarin HPLC assay of 7-hydroxywarfarin
was incubated in a ®nal volume of 250 ll. The 7-hydroxywarfarin
formation rate from (S)-warfarin was determined in a concentra- 7-Hydroxywarfarin was quanti®ed with the HPLC method ac-
tion range of 0.75±50 lM, in the same three human livers. After cording to Lang and Boker with slight modi®cations [12]. The
È
pre-incubation the reaction was initiated by the addition of 25 ll HPLC system consisted of a Gilson 307 pump and a Gilson 234
NADPH-generating system and terminated by the addition of automatic injector (Gilson, Middelton, Wisc., USA) with a 100-ll
4
25 ll HClO , 60%. 7-Ethoxycoumarin (0.033 nmol) was added as loop (Rheodyne, Cotati, Calif., USA), a Waters 470 ¯uorescence
internal standard. After centrifugation and pH adjustment as de- detector (Waters, Milford, Mass., USA) and an HP 3395 integrator
scribed for tolbutamide, 50 ll of the supernatant was injected onto (Hewlett Packard, Palo Alto, Calif., USA). A Symmetry C18 col-
the HPLC column. umn (150 mm ´ 3.0 mm ID, 5 lm; Waters, Milford, mass., USA)
In preliminary experiments it was ascertained that the forma- was used to separate 7-hydroxywarfarin and the internal standard
tion of 7-hydroxywarfarin from (S)-warfarin (50 lM) was linear up 7-ethoxycoumarin at ambient temperature. The eluent consisted of
)
1
to 2 mg protein ml incubation mixture during 30 min incubation a 0.05 N triethylamine (pH adjusted to 2.45 with H
3 4
PO ), aceto-
time. The reaction was also linear throughout 40 min incubation nitrile and methanol (59:31:10 v/v) mixture; the ¯ow rate was
for a microsomal protein content of 1 mg á ml
)
1
)1
.
0.8 ml á min . Detection was performed by ¯uorescence detection

9
49
at an excitation wavelength of 320 nm and an emission wavelength 10.5 pmol á min) á mg . Eadie-Hofstee plots for both 4-
1
)1
of 415 nm. A calibration curve was constructed by spiking micro-
somes not containing the NADPH-generating system, with 0.012 to
hydroxytolbutamide and 7-hydroxywarfarin formation
displayed a monophasic pattern (data not shown).
0.617 lM 7-hydroxywarfarin. Height ratios were determined using
the internal standard method. The intra-day coecients of varia-
tion (n ˆ 6) at low (0.024 lM), medium (0.247 lM) and high
(
0.493 lM) 7-hydroxywarfarin concentrations were all below 2%;
Inhibition experiments
the inter-day coecients of variation (n ˆ 3) at the low, medium
and high concentrations were all below 5%.
In a ®rst series of experiments, the formation rates of
4
-hydroxytolbutamide and 7-hydroxywarfarin were
measured in microsomal preparations of three di€erent
livers as function of the concentration of
sulphaphenazole and the various SSRIs, at a ®xed tol-
Data analysis
a
Formation rates were expressed as pmoles per minute per milligram
of protein. The data points consisting of the 4-hydroxytolbutamide
and 7-hydroxywarfarin formation velocities (v) at varying tolbut- butamide (200 lM) and (S)-warfarin (4 lM) concen-
amide and (S)-warfarin concentrations, respectively, in the absence
of inhibitor, were ®tted by non-linear regression analysis to a one
Fig. 1. Amongst the SSRIs tested, only ¯uvoxamine
enzyme Michaelis-Menten equation:
tration respectively. A representative plot is shown in
markedly inhibited both reactions with average IC50
values of 23 lM for tolbutamide 4-methylhydroxylation
and 27 lM for (S)-warfarin 7-hydroxylation. The other
Vmax  S
Km ‡ S
v ˆ
(Eq. 1)
m
Vmax and K are the apparent maximal velocity and the apparent substances tested, except desmethylsertraline, had IC50
Michaelis-Menten constants, respectively. The drug concentration
producing 50% decrease in metabolite formation rate (IC50) was
determined for the various inhibitors by visual analysis of plots of
the percentage control activity vs inhibitor concentration.
values over 100 lM for both reactions studied. De-
smethylsertraline was only tested up to 30 lM due to its
limited water solubility. No apparent inhibition by des-
For K
determinations, data for inhibitors were also ®tted to methylsertraline was observed up to 30 lM.
i
0
0
Eq. 1, yielding K and V , apparent kinetic parameters in the
presence of inhibitor. K values were determined by least-squares
linear regression analysis of secondary plots (``K =V '' ratio as a
function of inhibitor concentration) [13].
m
max
Sulphaphenazole, a potent prototypical CYP2C9 in-
hibitor, strongly inhibited both reactions with average
IC50 values of 0.63 lM for tolbutamide hydroxylation
i
0
0
m
max
The data sets were ®tted using an iterative program (Statistica, and 0.28 lM for (S)-warfarin hydroxylation.
version 5.0, Statsoft, Tulsa, Okla., USA).
In a second series of experiments, ¯uvoxamine was
retested at di€erent tolbutamide and (S)-warfarin
concentrations in order to determine the K value. A
i
Results
representative plot of the 4-hydroxytolbutamide and
7
-hydroxywarfarin formation rate as a function of the
respective substrate concentration in the absence and in
the presence of di€erent concentrations of ¯uvoxamine
Kinetic analysis of 4-hydroxytolbutamide
and 7-hydroxywarfarin formation
(0, 10, 20 lM and 40 lM), is shown in Fig. 2. The av-
As shown in Table 1, kinetic analysis of 4-hydroxytol- erage computed K values of ¯uvoxamine were 13.3 lM
i
butamide formation from tolbutamide yielded a mean for tolbutamide hydroxylation and 13.0 lM for (S)-
)
1
Km of 133 lM and a mean Vmax of 248 pmol á min
á
warfarin hydroxylation (Table 1). For both substrates,
)1
mg ; formation of 7-hydroxywarfarin from (S)-warfa- inhibition by ¯uvoxamine was characterized by a con-
rin yielded a mean Km of 3.7 lM and a mean Vmax of centration-dependent increase in Km values, whereas
Table 1 Kinetic parameters of tolbutamide 4-methylhydroxylation
and (S)-warfarin 7-hydroxylation; inhibition constants of
¯
uvoxamine for inhibition of tolbutamide 4-hydroxylation and
S)-warfarin 7-hydroxylation and inhibition constant of ¯uoxetine
(
for inhibition of (S)-warfarin 7-hydroxylation in three human liver
microsomal preparations (HL 7, HL 9, HL 10). Vmax apparent
m i
maximal velocity, K apparent Michaelis-Menten constant, K
apparent inhibition constant
HL 7 HL 9 HL 10 Mean
Tolbutamide 4-methylhydroxylation
)
)
1
)1
V
max (pmol á min á mg
160
205
369
84
217
110
248
133
13.3
K
m
(lM)
¯uvoxamine (lM)
K
i
17.3 16.1 6.4
Fig. 1 E€ect of citalopram (s), desmethylcitalopram (d),
desmethylsertraline (h), ¯uoxetine (n), ¯uvoxamine (n), nor¯uoxe-
tine (m), paroxetine (e), sertraline (r) and sulfaphenazole (+ ) on 4-
hydroxytolbutamide (left) and 7-hydroxywarfarin formation (right)
at a tolbutamide concentration of 200 lM and an (S)-warfarin
concentration of 4 lM. Activities are expressed as a percentage of
control activity. A representative plot for liver HL 7 is shown
(
V
K
K
K
S)-warfarin 7-hydroxylation
)
)
1
)1
max (pmol á min á mg
8.93 13.7 8.8
10.5
3.7
13.0
87
m
(lM)
4.5
18.7 8.4
125 80
2.9
3.8
11.9
56
i
i
¯uvoxamine (lM)
¯uoxetine (lM)

950
7-hydroxylation by ¯uoxetine and nor¯uoxetine, as as-
sessed by IC50 determinations. Furthermore, the average
K value of ¯uoxetine for the 7-hydroxylation of (S)-
i
warfarin was quite high. An explanation for this ap-
parent discrepancy between the e€ect of ¯uoxetine on
phenytoin p-hydroxylation and its much smaller inhib-
itory e€ect when (S)-warfarin or tolbutamide are used as
substrates, could be inhibition of the CYP2C19 com-
ponent of phenytoin p-hydroxylation [3]. Fluoxetine and
its metabolite nor¯uoxetine, have been shown to have a
moderate inhibitory e€ect on CYP2C19 activity in vitro
Fig. 2 Formation rates of 4-hydroxytolbutamide (left) and 7-
hydroxywarfarin (right) in function of concentrations of tolbutamide [18]; ¯uoxetine also moderately inhibited CYP2C19 ac-
or warfarin in a representative human liver microsomal preparation
(HL 7). Incubations were performed without inhibitor (s) and in the
presence of 10 lM (d), 20 lM (h) and 40 lM (n) ¯uvoxamine. All
data were ®tted to Eq. 1
tivity in vivo [17].
No formal in vivo studies are available about the
e€ect of ¯uvoxamine on warfarin metabolism but indi-
rect clinical data supports our in vitro ®ndings: ¯uvox-
amine co-administration produced a substantial increase
Vmax remained una€ected. Although no marked inhibi- in racemic warfarin plasma concentrations (+65%) and
tion by ¯uoxetine was observed at a ®xed (S)-warfarin an increase in prothrombin time [6].
concentration of 4 lM, the K value of ¯uoxetine for the
The in vivo situation is, however, more complex than
i
7-hydroxylation of (S)-warfarin was determined because might be expected from the in vitro results. Warfarin is
some case reports suggested a possible interaction be- used clinically as the racemate, and the anticoagulant
tween warfarin and ¯uoxetine [7, 8]. The mean K value e€ect resides in the (S)-enantiomer. Whereas (S)-warfa-
i
of ¯uoxetine (0, 50 lM and 100 lM) for the 7-hydrox- rin metabolism is mainly dependent on CYP2C9, the
ylation of (S)-warfarin was 87 lM.
major contributor to (R)-warfarin metabolism is
CYP1A2; CYP3A4 and CYP2C19 mediate minor
pathways [5]. (R)-warfarin has been reported to be an
inhibitor of (S)-warfarin 7-hydroylation [19]. Fluvox-
amine is known to be a potent inhibitor of CYP1A2
Discussion
As previously reported, tolbutamide 4-methylhydroxyl- activity in vitro and in vivo [20, 21]; there is also evi-
ation and (S)-warfarin 7-hydroxylation followed one dence that ¯uvoxamine can inhibit CYP3A4 and
enzyme Michaelis-Menten kinetics and the apparent CYP2C19 to some extent [17, 22]. Some authors have
kinetic parameters found in this study are similar to suggested that ¯uvoxamine could produce an increase of
those reported by others [11, 14].
(R)-warfarin plasma concentrations due to CYP1A2
Amongst the SSRIs and some of their metabolites (and CYP3A4 and CYP2C19) inhibition; this could in
tested, only ¯uvoxamine was able to markedly inhibit turn lead to inhibition of CYP2C9, leading to an in-
both tolbutamide 4-methylhydroxylation and (S)-war- crease in (S)-warfarin plasma concentrations and a po-
farin 7-hydroxylation. This is compatible with the results tentiation of the anticoagulant e€ect [1]. However, clear
of Yamazaki and Shimada, who demonstrated inhibi- evidence in favour of this hypothesis is lacking [5].
tion of both reactions by 50 lM ¯uvoxamine [15]. The
The fact that ¯uoxetine, paroxetine, sertraline and
K values of ¯uvoxamine found for both reactions citalopram produced no or only minor changes in the
i
studied were comparable. For both reactions, inhibition pharmacokinetics of warfarin [23±26], is compatible with
by ¯uvoxamine was characterized by a concentration- our in vitro results: these SSRIs had no or little e€ect on
dependent increase in Km value with no e€ect on Vmax CYP2C9 activity. However, some case reports describe
value, compatible with a competitive inhibition model. prolonged bleeding when warfarin and ¯uoxetine were co-
The K values of ¯uvoxamine were also comparable to administered; no interaction mechanism was suggested by
i
those reported when phenytoin was used as substrate [2]. the authors [7, 8]. On the other hand ¯uoxetine and
Fluvoxamine has, however, besides a moderate inhibi- sertraline were also reported to produce no clinically sig-
tory e€ect on CYP2C9 activity, the ability to inhibit ni®cant decrease in the clearance of tolbutamide [27, 28].
CYP2C19 activity, which could explain the comparable
In summary, clinicians should carefully monitor
K values of ¯uvoxamine for phenytoin p-hydroxylation, therapy when SSRIs are added to a regimen involving
i
tolbutamide 4-methylhydroxylation and (S)-warfarin 7- drugs mainly metabolized by CYP2C9, e.g. tolbutamide
hydroxylation [15±17]. Schmider et al. reported rather and (S)-warfarin. This is of particular importance when
low K values of ¯uoxetine (19 lM) and nor¯uoxetine ¯uvoxamine and warfarin are co-administered because
i
(
17 lM) when phenytoin was used as a substrate [2]. A of the narrow therapeutic index of warfarin.
substantial number of case reports support the hypoth-
esis that ¯uoxetine inhibits phenytoin kinetics in vivo [1].
Our in vitro results showed no marked inhibition of ei-
Acknowledgements This study was supported by grant no.
G.0011.97 from the National Fund for Scienti®c Research, Belgium
and by grant no 01104197 from the University of Gent Research
ther tolbutamide 4-methylhydroxylation or (S)-warfarin Foundation. We thank Prof. Dr. B. de Hemptinne and Mr. M. Van

9
51
Der Vennet of the Gent University Hospital Liver Transplantation 15. Yamazaki H, Shimada T (1997) Human liver cytochrome P450
Team. Critical review of the manuscript by Prof. Dr. M.G. Bogaert
is acknowledged.
enzymes involved in the 7-hydroxylation of R- and S-warfarin
enantiomers. Biochem Pharmacol 54: 1195±1203
1
1
6. Perucca E, Gatti G, Cipolla G, Spina E, Barel S, Soback S,
Gips M, Bialer M (1994) Inhibition of diazepam metabolism by
¯uvoxamine: a pharmacokinetic study in normal volunteers.
Clin Pharmacol Ther 56: 471±476
7. Jeppesen U, Gram LF, Vistisen K, Loft S, Poulsen HE, Brùsen
K (1996) Dose-dependent inhibition of CYP1A2, CYP2C19
and CYP2D6 by citalopram, ¯uoxetine, ¯uvoxamine and
paroxetine. Eur J Clin Pharmacol 51: 73±78
References
1
. Harvey AT, Preskorn SH (1996) Cytochrome P450 enzymes:
interpretation of their interactions with selective serotonin re-
uptake inhibitors. Part II. J Clin Psychopharmacol 16: 345±355
2. Schmider J, Greenblatt DJ, von Moltke LL, Karsov D, Shader 18. Kobayashi K, Yamamote T, Chiba K, Tani M, Ishizaki T,
RI (1997) Inhibition of CYP2C9 by selective serotonin reup-
take inhibitors in vitro: studies of phenytoin p-hydroxylation.
Br J Clin Pharmacol 44: 495±498
. Levy RH (1995) Cytochrome P450 isozymes and antiepileptic
drug interactions. Epilepsia 36 [Suppl 5]: S8±S13
. Miners JO, Birkett DJ (1996) The use of tolbutamide as a
substrate probe for human hepatic cytochrome P4502C9.
Methods Enzymol 272: 139±145
Kuroiwa Y (1995) The e€ects of selective serotonin reuptake
inhibitors and their metabolites on S-mephenytoin 4¢-hydrox-
ylase activity in human liver microsomes. Br J Clin Pharmacol
40: 481±485
3
4
19. Kunze KL, Eddy AC, Gibaldi M, Trager WF (1991) Metabolic
enantiomeric interactions: the inhibition of human (S)-warfa-
rin-7-hydroxylase by (R)-warfarin. Chirality 3: 24-29
20. Brùsen K, Skjelbo E, Rasmussen BB, Poulsen HE, Loft S
(1993) Fluvoxamine is a potent inhibitor of cytochrome
P4501A2. Biochem Pharmacol 45: 1211±1214
5
6
. Kaminsky LS, Zhang Z-Y (1997) Human P450 metabolism of
warfarin. Pharmacol Ther 73: 67±74
. Ben®eld P, Ward A (1986) Fluvoxamine. A review of its 21. Van Harten J (1993) Clinical pharmacokinetics of selective
pharmacodynamic and pharmacokinetic properties, and ther-
apeutic ecacy in depressive illness. Drugs 32: 313±334
. Hanger HC, Thomas F (1995) Fluoxetine and warfarin inter- 22. Fleishaker JC, Hulst LK (1994) A pharmacokinetic and
serotonin reuptake inhibitors. Clin Pharmacokinet 24: 203±
220
7
8
9
actions. N Z Med J 108: 157
pharmacodynamic evaluation of the combined administration
. Woolfrey S, Gammack NS, Dewar MS, Brown PJ (1993)
of alprazolam and ¯uvoxamine. Eur J Clin Pharmacol 46: 35±
Fluoxetine-warfarin interaction. BMJ 307: 241
39
. Amar-Costesec A, Beaufay H, Wibo M, Thines-Sempoux D, 23. Rowe H, Carmichael R, Lemberger L (1978) The e€ect of
Feytmans E, Robbi M, Berthet J (1974) Analytical study of
microsomes and isolated subcellular membranes from rat liver.
J Cell Biol 61: 201±212
0. Bradford MM (1976) A rapid and sensitive method for the
quanti®cation of microgram quantities of protein using the
principle of protein-dye binding. Ann Clin Biochem 72: 248±252
¯uoxetine on warfarin metabolism in the rat and man. Life Sci
23: 807±812
24. Bannister SJ, Houser VP, Hulse JD, Kisicki JC, Rasmussen
JGC (1989) Evaluation of the potential for interactions of
paroxetine with diazepam, cimetidine, warfarin, and digoxin.
Acta Psychiatr Scand 80: 102±106
1
1
1
1. Miners JO, Smith KJ, Robson RA, McManus ME, Veronese 25. Apselo€ G, Wilner KD, Gerber N, Tremaine LM (1997) E€ect
ME, Birkett DJ (1988) Tolbutamide hydroxylation by human
liver microsomes. Biochem Pharmacol 37: 1137±1144
of sertraline on protein binding of warfarin. Clin Pharmaco-
kinet 32 [Suppl 1]: 37±42
È
2. Lang D, Boker R (1995) Highly sensitive and speci®c high- 26. Priskorn M, Sidhu JS, Larsen F, Davis JD, Khan AZ, Rolan
performance liquid chromatographic analysis of 7-hydroxy-
warfarin, a marker for human cytochrome P-4502C9 activity.
J Chromatogr Biomed Sci Appl 672: 305±309
PE (1997) Investigation of multiple dose citalopram on the
pharmacokinetics and pharmacodynamics of racemic warfarin.
Br J Clin Pharmacol 44: 199±202
13. Bourrie M, Meunier V, Berger Y, Fabre G (1996) Cytochrome 27. Lemberger L, Bergstrom RF, Wolen RL, Farid NA, Enas GG,
Â
P450 isoform inhibitors as a tool for the investigation of met-
abolic reactions catalyzed by human liver microsomes. J Phar-
macol Exp Ther 277: 321±332
4. Rettie AE, Eddy AC, Heimark LD, Gibaldi M, Trager WF
(1989) Characteristics of warfarin hydroxylation catalyzed by
human liver microsomes. Drug Metab Dispos 17: 265±270
Arono€ GR (1985) Fluoxetine: clinical pharmacology and
physiologic disposition. J Clin Psychiatry 46: 14±19
28. Tremaine LM, Wilner KD, Perskorn SH (1997) A study of the
potential e€ect of sertraline on the pharmaokinetics and pro-
tein binding of tolbutamide. Clin Pharmacokinet 32 [Suppl 1]:
31±36
1