Effects

of imidazole

human

derivatives

hepatocytes

in primary

Words:

34033,

France;

tDepartement

Nutrition,
Institut du Cancer, Centre Paul Lamarque,
H#{244}pital
Saint Eloi, Montpellier
34059, France

34094,

France;

and

cell lysate

hepatocyte

culture

microsome

superfamily

of

hemo-

from plants to mammals. Although

expressed
in most tissues, these cytochromes
are concentrated
in the liver of humans
and higher
mammals, where they play a major part in the metabolism
of endogeneous
compounds
including
steroid
hormones,
biliary
salts, and fatty acids, as well as in the detoxication
of xenobiotic molecules
including
drugs and environmental
pollutants (1). Among
these, imidazole
and its derivatives
have
been recognized
for some
time as potent
ligands of the heme
iron atom of cytochromes
P450 (2). Cytochrome
P450 S1AI
(lanosterol
l4cs-demethylase)
from
Candidas
albicans
was
shown
recently
to be the target
of antifungal
imidazole
derivatives
such as ketoconazole
(KT),
miconazole
(MIC),
clotrimazole
(CLO),2
and fluconazole
(FLU)
(3, 4). These
compounds
are also able to interact
with cytochromes
P450
from
liver
microsomes,
thereby
inhibiting
some monoxygenase activities
(5-12). In addition,
N-substituted
imidazole

from

culture
DAUJAT
ISABELLE
MAUREL1

CYTOCHROME
P450s
cONsTtTUTE
a
proteins
present
in all living organisms

752

P450

MANUELLE
MAURICE,
LYDIANE
PICHARD,*
MARTINE
HENRI JOYEUX,t
JACQUES
DOMERGUE,1
AND PATRICK
*INSERM U-128, CNRS, Route de Meude, BP 5051, Montpellier

ABSTRACT
The
expression
of
several
forms
of
cytochrome
P450 including
P450 1A2, 2D6, 2E1, and 3A
was investigated
in human
hepatocytes
maintained
in
primary
culture for 96 h in the absence
or presence
of 50
sM of various
imidazole
derivatives.
These
included
ketoconazole,
clotrimazole,
miconazole,
fluconazole,
secnidazole
and metronidazole.
In addition,
the typical
inducers
rifampicin
and 3-naphthoflavone
were used for
comparison.
Western
and Northern
blot
analysis
of
micrososnes
and RNA prepared
from these cultures as well
as de novo synthesis
experiments
revealed
that, among
the imidazole
derivatives
tested, only clotrimazole
was a
strong rifampicin-like
inducer
of P450 3A. The expression of the other forms of P450 tested was not affected
by
the treatments.
Analysis
of the inhibition
of 13 monoxygenase activities,
including
ethoxyresorufin
and phenacetin 0-deethylases,
coumarin
7a-, lauric acid 11- and 12-,
mephenytoin
4-, debrisoquin
4-, and aniline
hydroxylases,
benzphetamine,
aminopyrine,
mephenytoin
and
erythromycin
demethylases,
and
cyclosporin
oxidase
(representative
of 10 different
forms of P450 in human
liver microsomes)
revealed
that ketoconazole
was a strong
and selective
in vitro inhibitor
of P450 3A (cyclosporin
oxidase)
with a K <1 tiM. Clotrimazole
and miconazole
were also strong inhibitors
of P450 3A-mediated
activities in contrast
to the other imidazole
derivatives.
Maurice,
M.; Pichard,
L.; Daujat,
M.; Fabre, I.; Joyeux,
H.; Domergue,
J.; Maurel, P. Effects of imidazole derivatives on cytochromes
P450
from human
hepatocytes
in
primary
culture.
FASEBJ
6: 752-758;
1992.
Key

on cytochromes

Montpellier

FABRE,

tService

et

de Chirurgie

de Chirurgie

C,

molecules
have been shown to be inducers
of several forms
of P450 in the rat (13-18). In comparison,
fewer data on the
inhibitory
effects
of these
derivatives
on human
liver
cytochromes
P450 have been published.
Moreover,
their inducing
capacity
has not been investigated
in humans.
With the recent development
of primary
cultures
of human hepatocytes
in which P450 genes retain their ability to
be induced,
systematic
studies of the inducibility
of human
cytochromes
P450 have recently
been undertaken
(19-21).
P450 1A1 and 1A2 were shown to respond
to -naphthoflavone
(BNF),
3-methylcholanthrene,
and
omeprazole,
whereas
P450
3A3 was inducible
by rifampicin
(RIF),
phenobarbital,
dexamethasone
(DEX),
troleandomycin,
and
several other compounds
(22-26).
In the present
study, we
investigated
the effects of several
N-substituted
imidazole
derivatives
currently
used as antifungal
or antibacterial
agents, including
KT, CLO, MIC, FLU, secnidazole
(SEC),
and metronidazole
(METR),
on human
cytochromes
P450
using adult human
hepatocytes
in primary
culture
and human liver microsomes.
MATERIALS

AND

METHODS

Chemicals
Chemicals
used in this study were obtained
from the following sources:
cyclosporin
A (CsA) and [mebmt-/3-3H]
CsAspecific
radioactivity
(11 Ci/mmol)
from Amersham,
England,
generously
supplied
by Sandoz
(Reuil-Malmaison,
France);
RIF from
Merrel
Dow-Lepetit
(Paris,
France);
CLO from Roger Bellon (Neuilly,
France);
FLU from Pfizer
(Massy,
France);
KT and MIC
from Jansen
(Boulogne,
France);
SEC and METR
from
Rh#{243}nePoulenc
(Paris,
France);
BNF and DEX
from Sigma
Chemical
Co. (St.
Louis, Mo.).
Tissue

sources,

hepatocyte

cultures,

and liver

microsomes

Lobes of liver, resected

for primary
or secondary
tumors
or
for angioma,
were obtained
at H#{244}pital
Saint Eloi and Centre

To whom correspondence
should be sent, at: INSERM
CNRS, Route de Mende, Montpellier
34033, France.
2Abbreviations:
KT, ketoconazole;
MIC, miconazole;
clotrimazole;
FLU, fluconazole;
BNF, fI-naphthoflavone;
rifampicin;
DEX,
metronidazole.

dexamethasone;

SEC,

secnidazole;

U-128,
CLO,
RIF,
METR,

3The human P450 3A subfamily comprises at least five genes.

These were shown to encode proteins of 83-98% similarity in terms
of primary sequence. The anti-P450 3A6 polyclonal antibody used
is expected to cross-react with all human P450 3A forms. Accordingly, we shall use the term P450 3A to refer to these proteins,
although P450 3A4 is most likely the P450 cyclosporin oxidase.

0892-6638/92/0006-752/$01 .50. EASEB

Paul Lamarque,
Montpellier,
France.
Hepatocytes
were isolated and cultured
as described
in previous
papers (22-26).
Culture
FT8 was prepared
from a 65-year-old
male who underwent
left lobectomy
for a hepatocellular
carcinoma.
Culture FT1O was prepared
from a 65-year-old
male who underwent left lobectomy
for a hepatocellular
carcinoma.
Culture
FT1I was prepared
from a 70-year-old
female who underwent left lobectomy
for a metastasis
from colonic
cancer.
Culture
FT12 was prepared
from a 55-year-old
female who
underwent
left lobectomy
for a voluminous
angioma.
Culture FT18 was prepared
from a 40-year-old
female who underwent
left lobectomy
for nodular
hyperplasia.
Culture
HTL34
was prepared
from a 76-year-old
male who underwent right lobectomy
for a metastasis
from colonic cancer.
Human
liver microsomes
were prepared
from lobe fragments
as described
(22). Clinical
characteristics
of liver
microsome
donor patients
FT1, FT3, FT4, and HTL27
are
given elsewhere
(23, 24). Patient FT6 was a 60-year-old
male
who underwent
right lobectomy
for a metastasis
from a colon
cancer. Patient 61289 was a 60-year-old
male who became
an
organ donor after cerebral
hemorrhage.
Patient
18190 was a
30-year-old
male who became
an organ donor after a severe
brain injury in a traffic accident.
Patient
FHI was a 30-yearold male who became an organ donor after a spontaneous
intracerebral
hemorrhage.

3A4
CLO

of cell lysates

and

microsomes

from

isC

:
STD

UT

RIF

DEX

STD

UT

KT

RIF

UT

FLU

MIC

CLO

CLO

1A2
STD

?&

Preparation
hepatocytes

MJ1K

METR

SEC

.*L-

cultured
STD

[3Hjlabeled
cell lysates and microsomes
were prepared
as
described
(22) from three and five 60 mm plates (3.5 x 106
cells per plate), respectively.
Cells used for cell lysates preparation
were previously
pulse-labeled
for 2-3 h with
10
tCi/plate
[3H]leucine
in a Leu-free
culture
medium
before
scraping.
Protein
concentration
was determined
using the
bicinconinic
acid protein
assay
reagent
from
Pierce
Chemical Co. (Rockford,
Ill.).

uT

TO

..-.

KT

..

BNF

...-..

2D6

UT

FLU

MIC

CLO

METR

SEC

UT

FLU

MIC

CLO

METR

SEC

2E1
Immunoquantitation

of cytochromes

P450

Cytochromes
P450 were quantified
by Western
blot analysis of microsomes
extracted
from cell cultures
as previously
described
(22-24),
using specific anti-P450
antibodies
prepared
in our laboratory
(22) or provided
by Drs. Dennis
Koop,
Cleveland
Ohio (anti 2El) and Urs Meyer,
Basel,
Switzerland
(anti 2D6). Relative
content
of the P450 forms
was evaluated
by densitometric
analysis
of the blots with a
Shimadzu
scanner
(Kyoto, Japan).
Dc novo

synthesis

and half-like

evaluation

of cytochromes

P450
De novo synthesis
and half-life of P450 1A2 and 3A were determined
by a radioimmunoprecipitation
assay on [3HJlabeled cell lysates, using specific anti-P450
1A2 and 3A antibodies as described
in detail elsewhere
(22).
Preparation
of RNA from
Northern
blot analysis

cultured

hepatocytes

and

Total RNA was extracted

from 15 plates and polyA RNA was
purified
as described
(22). Four micrograms
of polyA RNA
were analyzed
in Northern
blot under previously
indicated
conditions,
oligonucleotide
87-27 complementary
to nucleotides 132 to 161 of P450 3A4 mRNA
(27) being used as a
probe.

EFFECT OF IMIDAZOLE

DERIVATIVES ON HUMAN

P450

Figure 1. Western
blot analysis
of cytochromes
P450 lA2, 2D6,
2E1, and 3A in microsomes
from human hepatocytes
treated with
various compounds.
Human hepatocytes
were maintained
in culture for 96 h in the absence (UT) or presence of 50 iM of various
compounds

including

ketoconazole

(KT),

clotrimazole

(CLO),

miconazole
(MIC), fiuconazole
(FLU), metronidazole
(METR),
secnidazole (SEC), rifampicine
(RIF), dexamethasone
(DEX), or
/3-naphthoflavone
(BNF). Microsomes
were prepared
and an aliquot of 10 cg was analyzed in Western blot using anti-P450 antibodies directed against P450 1A2, 2E1, 2D6, and 3A6. STD refers to
purified human P450 3A(CsA oxidase) or rabbit P450 1A2; T0
refers to microsomes
prepared
at time zero of culture (plating).
These experiments
are representative
of cultures FT8 (blots 1A2
upper, 2D6 and 2E1), FT1O (blot 3A4 middle), FT11 (blot 1A2
lower), FT12 (blot 3A4 lower), and HTL34 (blot 3A4 upper).

Monoxygenase

activities

and

inhibition

studies

icrosomesp
reparedf
roml iverf ragmentso
rf romc ulturedh
epatocytesw
erer esuspendedi
n0 .1M p otassiump
hosphateb
ufferp H7 .4i nt hep resenceo
ft hes ubstratea
ndi nt hea bsenceo rp resenceo
f0 -300
iMo ft hei midazoled
erivatives.T
her eactionw asi nitiatedb yt hea dditiono
f1 m MN ADPHa
nda llowedt
op roceeda
t3 7#{176}Cu
nderl
ineark ineticsc
onditions.A lls ubstratesa
ndi midazoled
erivativesw
ered iluted

753

051
#{149}Oh
tel

96h

ft

U,
It)

I,,

0,3

0,2

0,1

RESULTS

UT

FLU

CLO

METR

SEC

#{174}

0,81

0,6

0,4 T

0,2

0,0-

UT

KT

CLO

RIF

UT

13.0

kb

12.2

kb

RIFDEXCLO

Figure 2. Cytochromes
P450 3A4 de novo synthesis and mRNA
level in human hepatocytes
treated with various compounds.
Human hepatocytes were maintained
in culture for 96 h in the absence
(UT) or presence
of 50 iM of various compounds,
including
ketoconazole
(KT),
clotrimazole
(CLO),
miconazole
(MIC),
fluconazole
(FLU), metronidazole
(METR),
secnidazole
(SEC),
rifampicine
(RIF), or dexamethasone
(DEX). At this time, either
cell lysate was prepared
from cultures
labeled
for 2 h (from 94 to
96 h) with tritiated leucine, for de novo synthesis analysis, or RNA
was extracted and polyA RNA was prepared. A, B) de novo synthesis of P450 3A determined
by a radioimmunoprecipitation
assay (in
% of total protein synthesis) from cultures HTL34 and FT11. For
untreated
cells, time 0 refers to 2 h of labeling after plating.
C)
Northern
blot analysis of 4 jsg polyA RNA from culture FT1O,
oligonucleotide
87-27 complementary
to P450 3A4 mRNA nucleotides 132-161 being used as a probe.

754

in dimethylsulfoxide.
The activities
studied included:
ethoxyresorufin
0-deethylase
(28), phenacetin
0-deethylase
(29),
coumarin
7a-hydroxylase
(30), benzphetamine
demethylase
(22), aminopyrine
demethylase
(22), lauric
acid
11 and
12-hydroxylases
(31),
mephenytoin
N-demethylase
and
4-hydroxylase
(32), debrisoquin
4-hydroxylase
(29), aniline
hydroxylase
(33),
erythromycin
demethylase
(22),
and
cyclosporin
A oxidase
(23). Cyclosporin
A oxidase
activity
was also determined
with hepatocytes
in culture under previously described
conditions
(23).

Vol. 6

January

1992

Human
hepatocytes
were maintained
in primary
culture for
96 h in absence
or presence
of 50 tM of the various
imidazole derivatives,
or of the typical
inducers
RIF, DEX,
or
BNF. Microsomes
prepared
from these cultures
were then
analyzed
by Western
immunoanalysis
developed
with
specific anti-P450
antibodies
against
P450 1A2, 2D6, 2E1,
and 3A. Results from different
cultures
are presented
in Fig.
1. In these cultures,
RIF and BNF specifically
induced
P450
3A and 1A1/2, respectively,
as previously
shown (22, 23).
Among the compounds
tested, only CLO was an inducer of
P450 3A, although
it did not affect accumulation
of the other
P450s investigated
here. This was confirmed
at the mRNA
level as shown in Fig. 2. Both P450 3A de novo synthesis
(Fig. 2A, Fig. 2B) and P450 3A mRNA
level (Fig. 2C) were
increased
in parallel
to protein
accumulation
in these cultures. These results show that CLO is as good an inducer
of
P450 3A at the level of protein
and specific mRNA
as RIF,
which in our hands is the best inducer
of human
P450 3A.
However,
no increase
of CsA oxidase
activity
(most likely
reflecting
P450 3A4 contribution)
was observed
in CLOtreated
cultures,
in contrast
to RIF-treated
cultures
(Fig.
3A). This discrepancy
probably
is related
to the fact that a
substantial
amount
of CLO, a strong inhibitor
of CsA oxidase activity
(K1 = 0.15 feM, see below),
remained
in the
cells at the time of the experiment.
Thus, CLO strongly induces P450 3A but concomitantly
binds tightly to its active
site and competes
with other substrates
(CsA). In these experiments,
accumulation
of P450 3A mRNA,
as determined
from Northern
blot, correlated
with P450 3A de novo synthesis, and these correlated
also with P450 3A accumulation
determined
by Western
blot as well as with CsA oxidase (excepted with CLO).
No evidence
for post-translational
effect
was observed.
In contrast,
the other
imidazole
derivatives
affected
neither
the accumulation
of the P450 forms
investigated
(Fig. 1) nor their specific messengers,
as detected
by de novo
synthesis
and Northern
blot analysis
(not shown).
In some
(but not all) cultures,
KT weakly induced
accumulation
of
P450 3A, as detected
by Western
blot. However,
neither
an
increase
in de novo synthesis
nor prolonged
half-life
of the
protein
(305
H, not shown)
was associated
with this
process.
Microscopic
examination
and strong
inhibition
of
de novo total protein
synthesis
indicated
that KT, and to a
lesser extent MIC, were highly toxic to the cells; such toxicity
was not observed
with the other derivatives.
This inhibition
of protein
synthesis
could have affected
the protein
pool of
microsomes
more than P450 3A, thus producing
an artifactually increased
accumulation
of this cytochrome.
In previous
experiments
(23) we observed
that KT was a
strong inhibitor
of CsA oxidase
in vitro, in agreement
with
many
clinical
reports
focusing
on this drug
interaction
(34-3 7). In this work we investigated
the inhibitory
effect of
KT on other monoxygenase
activities
using nine different

The FASEB Journal

MAURICE

ET AL.

KT and FLU have an IC50 of 50 nM for the 14 a-mevalonate

demethylase
by cell free extracts
from Candida albicans (3).
The other imidazole
derivatives
were also assayed as inhibitors
of both CsA oxidase
and erythromycin
demethylase.
Results are shown in Table 2. In addition
to KT, CLO and
MIC were strong inhibitors
of these activities
with K1 of 0.15
and 1.3 tiM, respectively.
In contrast,
FLU, METR,
and
SEC were weak inhibitors
or did not inhibit at all these activities. These results were clearly confirmed
when inhibitory
experiments
on CsA oxidase
were performed
with cultured
hepatocytes
(Fig. 3B).

4i
3.

1-

0
UT

RIF

KT

CLO

MIC

FLU

SEC

METR

DISCUSSION

3-

1-

us

in

in

In

ii.,

KT

CLO

In in
#{248}e

MIC

FLU

in
el

SEC

In

in

in

METR

Figure 3. Cyclosporin

A oxidase activity in primary

cultures
of human
hepatocytes
treated
with various
compounds.
A) Human
hepatocytes
(FT18) were maintained
in culture
for 96 h in the absence (UT)
ketoconazole
fluconazole
rifampicine
absence of

zCi

or presence of 50 ILM of various compounds

including
(KT),
clotrimazole
(CLO),
miconazole
(MIC),
(FLU), metronidazole
(METR),
secnidazole (SEC) or
(RIF). At this time, culture medium was renewed in the
the inducers but in the presence of 5 M CsA and 0.1

tritiatedCsA; then CsA

oxidase activitywas analyzed by radio

HPLC in the extra- and intracellular

medium for 24 h. Presented
here is the amount of oxidized metabolites of CsA (in ILM) released
in the extracellular
medium
after 4 h. B) Human
hepatocytes
(FT18) were maintained
in culture for 72 h in the absence (UT) or
presence of 50 iM RIF. Culture medium was renewed in the absence of RIF but in the presence of CsA, as indicated above. In addition, the RIF-treated
cultures received 0, 5, 25, or 100 ItM of the
various imidazole derivatives, and CsA metabolism was analyzed as
indicated above. The
ILM)

amount

of oxidized metabolites of CsA

released in the extracellularmedium

(in

after 4 h is presented.

Data were normalized

from culture plate to culture plate with
respect to the protein concentration
in cell lysate from each plate
after the CsA oxidase experiment
was completed (24 h).
preparations
of human
liver microsomes.
The data concerning 13 different
activities,
representative
of 10 different
forms
of P450 from families
I to 4, are reported
in Table 1. Both
CsA oxidase
and erythromycin
demethylase
were strongly
inhibited
with IC50 lower than 1 I4M. An intermediary
IC50
of 5-20 &M were observed
with mephenytoin
N-demethylase
and 4-hydroxylase
activities.
Other activities
were inhibited
to a much lesser extent,
IC50 values being higher
than 50
sM. These results indicate
that KT is a strong and selective
inhibitor
of P450 3A-related
activities
in human
liver microsomes when used in the micromolar
range of concentration.

EFFECT OF IMIDAZOLE

DERIVATIVES ON HUMAN

P450

The results
presented
in this paper
show that among
the
imidazole
derivatives
tested, only CLO was a potent inducer
of P450 3A in human
hepatocytes
in primary
culture.
The
finding
that increased
accumulation
of the protein
was
accompanied
by a parallel
increase
in the concentration
of
specific mRNA
suggests
that this compound
activates
the
transcription
of gene (or genes) 3A. CLO appears
therefore
as a RIF-like
inducer of human
P450 3A in the liver. Several
investigators
have reported
that
N-substituted
imidazole
derivatives
were strong inducers
of P450-mediated
monoxygenase activities
in rat liver microsomes
(13-17). The most
detailed
study of this is that of Hostetler
et al. (17). These
authors
observed
that CLO, KT, and MIC were potent inducers of P450 3A and 2B, and weak inducers
of P450 1A
and 2E. Notably,
CLO appeared
as the most efficacious
inducer of rat P450 3A ever reported;
our conclusion
that this
molecule
is as strong an inducer
as RIF in human
hepatocytes is in agreement
with this observation.
However,
in contrast to what was observed in the rat, CLO did not affect accumulation
of P450 1A1/2 and 2E1 (P450 2D6, a constitutive
form for which
no inducer
has yet been found,
was not
affected by this compound
either).
In addition,
neither
KT
nor MIC were inducers
of P450 in human
cultures
at 50
tM.
Owing
to the high
toxic effect of both compounds,
larger concentrations
could not be tested in induction
experiments. It is not clear whether
this absence of induction
in human hepatocyte
cultures
results from the in vitro conditions
used here or reflects an interspecies
difference
in response
to
these molecules,
because
to our knowledge
similar
experiments were not carried out with rat cultures.
This would not
be the first illustration
of interspecies
differences
in the
responsiveness

of genes

from

the

3A

family

to

various

in-

ducers:
previously
documented
examples
include
RIF, a
strong inducer
in humans
and rabbits but not in the rat, and
pregnenolone
16a-carbonitrile,
a typical
inducer
in the rat
but not in humans
(26).
The
most
thoroughly
documented
effect of imidazole
derivatives
on cytochromes
P450 concerns
their inhibitory
potency (3-13). In this respect, our results are in close agreement with previous
reports
on rat, mouse,
or human
liver
microsomes.
The literature
suggests that activities
known to
be related
to P450 3A i.e., CsA oxidase,
erythromycin
demethylase,
ethynylestradiol
hydroxylase,
ethylmorphine
demethylase,
and
63-hydroxylation
of steroids,
are more
strongly
inhibited
by the N-substituted
imidazole
derivatives
than activities
known to be related
to other forms of P450.
For example,
Tarbit and co-workers
(7, 10) measured
IC50
values for KT of 0.08 and 31 tM for 6f3- (P450 3A family)
and 16a-hydroxylation
(P450
2C family)
of testosterone,
respectively,
in mouse
liver microsomes
and KT values of
0.055 and 90 ,LM for the same activities
in rat liver micro-

755

TABLE

1. Inhibition

of monoxygenase

activities

by ketoconazole

Monooxygenase

P450 forms

liver microsomes
Residual

activities5

1A1

Ethoxyresorufin

1A2

Phenacetin

2A6
2B

Coumarin
7a-hydroxylase
Benzphetamine
demethylase
Lauric
acid 12-hydroxylase

2C
lA2-2C

in human

0-deethylase

activity,

IC50d

109-102-100

214-226-233
207-264-238

109-113-71
82-89-119
100-105-78

42-33-77
267-162-44
300-205-326

95-115-96

0-deethylase

67-72

170-105

2C8-l0
2C8-l0
2D6
2E1
3A4

Mephenytoin
4-hydroxylase
Mephenytoin
N-demethylase
Debrisoquin
4-hydroxylase
Aniline hydroxylase
Erythromycin
demethylase

125-85-90
87-75-75
62-98
67-89-84

12-23-17
18-5-14
52-235
455-429-155

27-47-36

3A4

Cyclosporin

oxidase

0.48-0.86-0.28
0.34 (K = 0.3)

4A

Lauric

1 1-hydroxylase

27
86-98-91

Aminopyrine

acid

demethylase

47-50-5 1

Forms of P450 most likely involved as the major enzyme catalyzing the corresponding
monoxygenase
reaction are listed in second column. When
the human form associated with a monoxygenase
activity has not been definitely identified, only the P450 familyisindicatedby reference to animal (rat
or rabbit) studies; for example: benzphetamine
catalyzed by P450 2B1 and 2B4 in rat and rabbit, respectively.
5Substrate concentration
during the
assay was: 3.3 tM cyclosporin
A; 5 M ethoxyresorufin
and coumarin;
50 M phenacetin,
debrisoquin,
lauric acid, and mephenytoin;
100 tM
benzphetamine,
aminopyrine
erythromycin;
5 mri aniline.
Residual activity for I tM KT, in % of uninhibited
activity. The different values were
obtained with liver microsomes
from different patients.
dIC50 is the concentration
of KT (in .tM) producing
50% inhibition
as determined
from the
graph of activity against KT concentration
(from 0 to 300 tM). The different values were obtained with liver microsomes
from different patients. These
experiments
are representative
of liver microsomes
from patients FF1, FT3, FT4, FT5, FT6, HTL27, 61289, 18190, and FHI.

somes. Similarly,
Back et al. (12) measured
IC50 values for
KT of 2.0 and 17.9 M for CsA oxidase
or ethynylestradiol
hydroxylase
(P450 3A family) and tolbutamide
hydroxylase
(P450 2C8-lO), respectively,
in human
liver microsomes.
Our
study of 13 different
P450-dependent
activities
representative of 10 different
forms of human
P450 (Table 1) allowed
us to demonstrate
that KT is indeed
a strong and selective
inhibitor
of P450 3A-mediated
monoxygenase
activities
in
human
liver microsomes
with a K, of 0.3 tM and IC50 <1
riM. It would appear
that substantial
inhibition
can be observed with other activities
but at much higher
concentrations. IC50 determined
with both mephenytoin
oxidase
activities (4-hydroxylase
and N-demethylase)
was in excellent
agreement
with that found by Back et al. (12). Several P450
form-specific
inhibitors
have already
been characterized
in
humans.
These
include
furafylline
(P450
1A2) (38), sulfaphenazole
(P450 2C8-10) (39), and quinidine
(P450 2D6)
(40). The use of such inhibitors
represents
an easy and inexpensive way to rapidly
characterize
both in vitro and in vivo
the form of P450 involved
in the metabolism
of pharmacologically
or toxicologically
important
molecules.
Among the compounds
tested here, only the N-substituted
imidazole
derivatives
KT, CLO, and MIC were strong inhi-

of CsA oxidase
and erythromycin
demethylase;
in contrast, FLU, a triazole
derivative,
and SEC and METR,
two
nitroimidazoles,
exhibited
much larger K (Table 2). Such a
structure
relationship
has been
observed
and discussed
previously (10) with the following conclusions: 1) imidazole
moiety seems to interact
with the P450 3A heme iron more
strongly
than does the triazole;
2) lipophilicity
of side chains
is also a determining
factor explaining
the low (if any) inhibition brought
about by SEC and METR.
Although
CLO
appears
as a strong
inducer
of human
P450 3A, this compound
is not likely to behave as such in
a clinical situation
because
its inhibitory
effect should overcome the inducing
one, as was observed
in hepatocyte
cultures (Fig. 3). Moreover,
this molecule
is generally
applied
to the skin, a way of administration
that is unlikely
to result
in in vivo concentration
in the range of that used here. KT
has been shown by some investigators
to strongly
inhibit CsA
oxidase in vivo (34-3 7), in agreement
with the low K1 values
reported
here (Table 2). Although
FLU has been reported
to
accelerate
renal dysfunction
in a patient
with polycystic
kidney disease receiving
CsA (41), a serious drug interaction
between CsA and FLU has been excluded
(42). This is to be
expected
owing to the large K1 values determined
in this

TABLE

by imidazole

2. Inhibition

of erythromycin

N-demethylase
Residual

Drugs

ER

Clotrimazole

42-28-30

Miconazole

49-33-42

Fluconazole
Metronidazole

Secnidazole

and cyclosporin

activities,

oxidase

bitors

derivatives

in human

liver microsomes

IC50

%
CsA ox

ERa

CsA ox

Ki (CsA ox

98-80-78
98-86-88

10.4
15.5
84.0
94.0

0.61-0.34-0.76
4.38-3.12-3.34
167-177-164
954-1050-411

0.20
1.74
93
302

0.15
1.30
63
223

90-83

83.0

750-810

181

165

Concentration
range: 0-10 M for CLO and MIC, and 0-300 M for FLU, METR,
and SEC.
6Residual activity for 10 aM of the inhibitors
in % of uninhibited
activity. Erythromycin
concentration
was 100 aM in this assay.
rResidual activity for 10 aM of the inhibitors in % of uninhibited
activity. Cyclosporin
concentration
was 3.3 zM in this assay.
dICSO is the concentration
of inhibitor (in SM), producing 50% inhibition as determined
from the graph of activity against KT concentration
(from 0 to 300 ELM). The different values were obtained
with liver microsomes
from different
patients.
These IC50 values were determined
from Lineweaver
Burk plots of cyclosporin
A oxidase activity.
These
K, values were determined
from Lineweaver Burk plots of cyclosporin
A oxidase activity for different concentrations
of inhibitors: 0, 0.4, 1, and 4 iM CLO; 0, 1, 4, and 10gM MIC;
0, 50, 100, and 200 gM for FLU, METR,
and SEC. These results are representative
of liver microsomes
from patients FT6, 61289, and 18190.

756

Vol. 6

January

1992

The FASEB Journal

MAURICE

ET AL.

study for this compound.

Although
no drug interaction
has
yet been reported
between
CsA and MIC, this compound
should
be used with care when administered
orally or intravenously
to patients
receiving
CsA. Recently,
a clinically
relevant
interaction
between
KT and mephenytoin
(but not
with debrisoquin)
was reported
(43). This is in agreement
with the results of Back et al. (12) and with this study showing that this compound
inhibits mephenytoin
oxidase activities (P450 2C8-10) with a potency
that is intermediary
between that for P450 3A and that for other forms of P450.
Thus,
although
KT is a selective
in vitro inhibitor
of P450
3A when used at concentrations
lower than 5 sM, this drug
can be expected
to interact
with both P450 3A and 2C8-10
in vivo, owing to the high dose generally
administered
and
the respective
K,.

This

in part by a grant from the Fondation

M#{233}dicale
Francaise. The authors would like to
thanks
to Drs. D. Koop (Cleveland,
Ohio) and U.

work

was supported

pour Ia Recherche
express
Meyer

and

their
(Biocenter,

2D6

Debr#{233},Paris,
mephenytoin

Hospital,

Basel,

antibodies,
France)
oxidase

Cambridge,

Switzerland)

to Dr.
for

for providing

anti-P450

E. Jackz-Aigrain

her

activities,

help
and

during

the

determination

to Dr. G. Park

U.K.) for critical

2E1

(H#{244}pital
Robert

reading

of

(Addenbrookes

11. Pasanen,
Kairaluoma,

hepatic

Taskinen,
T.,
M., and Pelkonen,

and placental

Iscan,
M., Sotaniemi,
0. (1988) Inhibition

E. A.,
of human

monooxygenase
by imidazole
37, 3861-3866
12. Back, D. J., Stevenson,P.,and Tjia,J. F. (1989) Comparative
effects of two antimycotic
agents ketoconazole
and terbinafine
on the metabolism
of tolbutamide,
ethinyloestradiol,
cyclospoantimycotics.

Biochem.

xenobiotic

Pharsnacol.

rin and ethoxycoumarin

by human
liver microsomes
in vitro.
Br. j Clin. Pharmacol. 28, 166-170
13. Niemegeers,
C.J. E., Levron,J.
Cl., Awouters, F., andJanssen,

P. A.

J.

(1981) Inhibition

and induction

of microsomal

enzymes

in the rat. A comparative

study of four antimycotics:

miconazole, econazole,
clotrimazole
and ketoconazole.
Arch. In!. Pharmacodyn. 251, 26-38
14. Lavrijsen,
K., Van Houdt, J., Thijs, D., Meuldermans,
W., and
Heykants,
J. (1986) Induction potential of antifungals
containing an imidazole
or tiazole moiety.
Miconazole
and ketoconazole, but not itraconazole
are able to induce
hepatic
drug
metabolizing
enzymes
of male rats at high doses. Biochem. Pharmacol. 35, 1867-1878
15. Ritter, J. K., and Franklin, M. R. (1987) Induction
and inhibition of rat hepatic
drug metabolism
by N-substituted
imidazole
drugs.
Drug Metab. Dispos. 15, 335-343
16. Ritter, J. K., and Franklin,
M. R. (1987) High magnitude
hepatic
cytochrome
P-450 induction
by an N-substituted
imidazole
antimycotic,clotrimazole.
Biochem. Phar,nacol. 36, 2783-2787

17.

of the manuscript.

M.,

Hostetler, K. A., Wrighton,

Levin,
hepatic
treated

S. A., Molowa,

D. T., Thomas,

P. E.,

W., and Guzelian,

P.S. (1989) Coinduction
of multiple
cytochrome
P-450
proteins
and their mRNAs
in rats
with imidazole
antimycotic
agents.
Mol. Pharmacol. 35,

279-285
18. Harmsworth,

REFERENC

ES

hepatic

1. Gonzalez,
F. J. (1990) Molecular
genetics of the P-450 superfamily. Phar,nacol. Ther 45, 1-38
2. Jefcoate, C. R. (1978) Measurement
of substrate
and inhibitor
binding to microsomal
cytochrome
P-450 by optical-difference
spectroscopy.
In Methods in Ezymology:
Biomembranes,
Part C
(Fleicher,
S., and Packler,
demic,
New York

3. Shaw,

J.

L., eds)

P-450

52, pp.

258-279,

Aca-

M.

mediated

H.,
sterol

and

Troke,

synthesis

P. F. (1987)

and

metabolism:

differences in sensitivity to fluconazole and other azoles. In Recent Trends in the Discovery, Development and Evaluation of An4fungal
Agents (Fromtling,
R. A., eds) pp. 125-139, JR. Prous Science
Publishers, S. A.
4. Kirsch,
D. R., Lai, M. H., and OSullivan,
J. (1988) Isolation
of the gene for cytochrome
P450 L1A1 (lanosterol
14 a-demethylase) from candida albicans. Gene 68, 229-237

5. Sheets,

J. J.,

and Mason,

J.

I. (1984) Ketoconazole:

hibitor
of cytochrome
P-450-dependent
liver. Drug Metab. Dispos. 12, 603-606
6. Meredith,
C. C., Maldonado,
A. L.,

7.

8.

9.

10.

a potent

drug

metabolism

and

Speeg,

K.

in-

in rat
V., Jr.

(1985) The effect of ketoconazole

on hepatic oxidative drug
metabolism
in the rat in vivo and in vitro. Drug Metab. Dispos.
13, 156-162
Houston, J. B., Humphrey,
M. J., Matthew, D. E., and Tarbit,
M. H. (1988) Comparison
of two azole antifungal
drugs,
ketoconazole
and
fluconazole,
as modifiers
of rat hepatic
monooxygenase
activity. Biochem. PharmacoL 37, 401-408
Thomson,
R. 0., Rawlins, M. D., James, 0. F. W., Wood, P.,
and Williams, F. M. (1988) The acute and subchronic effects of
ketoconazole on hepatic microsomal monooxygenase
in the rat.
Biochem. Pharmacol. 37, 3975-3980
Capdevila, J., Gil, L., Orellana, M., Marnett, L. J., Mason, J. I.,
Yadagiri, P., and Falck, J. R. (1988) Inhibitors of cytochrome
P-450-dependent
arachidonic
acid metabolism.
Arch. Biochem.
Biophys. 261, 257-263
Ballard, S. A., Lodola, A., and Tarbit, M. H. (1988) A comparative
study of 1-substituted
imidazole
and 1,2,4-triazole
antifungal
compounds
as inhibitors
of testosterone
hydroxylations
catalyzed
by mouse hepatic
microsomal
cytochromes
P-450. Biochem. Pharmacol.
37, 4643-4651

EFFECT OF IMIDAZOLE

DERIVATIVES ON HUMAN

P450

extrahepatic

Franklin,

M.

cytochrome

R. (1990)

Induction

of

and monooxygenase activity by N-substituted imidazoles. Xenobiotica 20,

1053-1063
19. Begue, J. M., Le Bigot, J. F., Guguen-Guillouzo,
C., Kiechel,
J. R., and Guillouzo, A. (1983) Cultured human adult hepato-

P-450

cytes: a new model for drug metabolism

studies.
Biochem. Pharmacol. 32, 1643-1646
20. Fabre, G., Rahmani,
R., Placidi, M., Combalbert,
J, Covo, J.,

Cano,

T. B., Tarbit,

Cytochrome

Vol.

W. L., and

and

J.

P., Coulange,

C., Ducros,

M., and Rampal,

M. (1988)

Characterization
of midazolam
metabolism
using
human
hepatic microsomal fractions and hepatocytes in suspension
obtained
by perfusing
whole human
livers. Biochem. Pharmacol.
37, 4389-4397
21. Morel,
F, Beaune,
P. H., Ratanasavanh,
D., Flinois,J.
P., Yang,

C. S., Guengerich,
F. P., and Guillouzo, A. (1990) Expression
of cytochrome
P-450 enzymes in cultured human hepatocytes.
Eur j Biochem. 191, 437-444
22. Diaz, D., Fabre, I., Daujat, M., Saint-Aubert,
B., Bones, P.,
Michel, H., and Naurel,
P. (1990) Omeprazole
is an aryl
hydrocarbon-like
inducer
of human
hepatic
Gastroenterology 99, 737-747
23. Pichard,
L., Fabre, I., Fabre, G., Domergue,

B., Mourad,

G., and Maurel,

cytochrome

J.,

P. (1990) Cyclosporin

P450.

Saint-Aubert,

A drug in-

teractions:
screening
for inducers
and inhibitors
of cytochrome
P-450
(cyclosporin
A oxidase)
in primary
cultures
of human
hepatocytes
and in liver microsomes.
Drug Metab. Dispos.
18,
595-606

24. Pichard,
Thenot,

L., Gillet, 0., Fabre, I., Dalet-Beluche,

I., Bonfils, C.,
P., and Maurel, P. (1990) Identification
of the rabbit

J.

and human
cytochromes
P-450111A
as the major enzymes
involved in the N-demethylation
of diltiazem.
Drug Metab. Dispos.

18, 711-719
25. Daujat, M., Fabre, I., Diaz, D., Pichard, L., Fabre, 0., Fabre,
J. M., Saint-Aubert, B., and Maurel, P. (1991) Inducibility and
expression of class IA and lilA cytochromes P450 in primary
cultures of adult human hepatocytes.
Lfe Sci. In press

26. Daujat,
M., Pichard,
L., Fabre, I., Pineau, T., Fabre, G.,
Bonfils, C., and Maurel,
P. (1991) Induction
protocols
for
cytochromes
P45OIIIA
in vivo and in primary
cultures
of
animal and human hepatocytes. Methods Enzymol. 206, 345-353
27. Bork, R. W., Muto, T., Beaune, P. H., Srivastava, P. K., Lloyd,
R. S., and Guengerich,
F. P. (1989) Characterization
of mRNA

757

species related to human

livercytochrome P450 nifedipin oxidase and the regulation of catalyticactivity.
j Biol. Chem. 264,

910-919
28. Burke, M. D., Prough, R. A., and Mayer, R. T. (1975) Characterization of a microsomal cytochrome P450 mediated reaction:
ethoxyresorufin
0-deethylation.
Drug Metab. Dispos. 3, 245-253
29. Distelnath,
L. M., Reilly, P. B. E., Martin,
M. V., Davis, C. C.,

Wilkinson,

C. R., and Guengenich,

F. P. (1985) Purification

and

characterization
of the human
liver cytochromes
P450 involved
in debnisoquine
4-hydroxylation
and phenacetin
0-deethylase,
two prototypes
for genetic
polymorphism
in oxidative
drug
metabolism. j Biol. Chem. 260, 9057-9067
30. Maurice, M., Emiliani, S., Dalet-Beluche, I.,Derancourt, J.,
and Lange, R. (1991) Isolation and characterization of a
cytochrome
P450 of the hA subfamily
from human
liver microsomes.
Eur. j Biochem. 200, 511-517
31. SalaOn, J. P., Simon,
A., and Durst,
F (1986) Specific induction
of launic
acid
w-hydroxylase
by clofibrate,
diethylhexyl-

phthalate and 2,4-dichlorophenoxyacetic

acid in higher plants.
Lipids 21, 776-779
32. Shimada, T., Misono, K. S., and Guengerich,
F. P. (1986) Human
liver
microsomal
4-hydroxylase,
a prototype
tive drug metabolism.
j

cytochrome
P450
mephenytoin
of genetic
polymorphism
in oxidaBid. Chem. 261, 909-921

33. Bonfils, C., Combalbert,

J., Pineau, T, Angevin, J., Larroque,
C., Derancourt,
J., Capony, J. P., and Maurel, P. (1990) Ontogenesis
of rabbit
liver
cytochrome
P450.
Evidence
for a
cytochrome
P450 lIE (3a)-related
form prevailing
during the
post-natal
period.
Eur. j Biochem. 188, 187-194
34. Morgenstern,
G. R., Powles, R., Robinson,
B., and McElwain,
T. J. (1982) Cyclosponin
interaction
with ketoconazole
and mel-

phalan. Lance! 2, 1342

35. Ferguson, R. M., Sutherland,
D. E. R., Simmons,
Najarian,
J. S. (1982) Ketoconazole,
cyclosponin
and renal transplantation.
Lance! 2, 882-883

758

Vol. 6

January

1992

36. White, D. J. G., Blatchford,

N. R., and Cauwenbergh,
G. (1984)
Cyclosponin
and ketoconazole.
Transplantation
37, 214-215
37. First, M. R., Weiskittel,
P., Alexander,
J. W., Schroeder, T J.,
Myre,
S. A., and Pesce, A. J. (1989) Concomitant
administration
of cyclosporin
and
ketoconazole
in renal
transplant
recipients.
Lancet 2, 1198-1201

38. Sesardic,

J,

D., Boobis,

De La Tome,

tent and selective

j Clin. Pharmacol.
39. Veronese, M. E.,
L. N., Miners, J.
tween the human
phenytoin.
Toxicology
Orrenius,

A. R., Murray,

R., and Davies,

inhibitor
29,

B. P., Murray,

D. S. (1990)

of cytochrome

S., Segura,

Furafylline

P4501A2

is a po-

in man. Br.

651-663

Doecke, C. J., McManus, M. E., Sansom,

0., and Birkett, D. J. (1990) Relationship
bemicrosomal
metabolism
of tolbutamide
and

In Drug Metabolizing
(Ingelman-Sundberg,
S., eds)
p. 203

Enzymes:
Genetics,
M.,
Gustafsson,
(abstr.)
Karolinska

Regulation

J.

and
and
Institutet,

A.,

Stockholm
40. Otton,
S. V., Crewe,
H. K., Lennard,
M. S., Tucker,
Woods,
H. F (1988) Use of quinidine inhibition to
role
of the
sparteine/debnisoquine
cytochrome
metoprolol
oxidation
by human
liver microsomes.
j
Exp. Ther. 247, 242-247

41. Collignon, P., Hurley, B., and Mitchell,

fluconazole with cyclosponin. Lancet 2,
42. Ehninger, G., Jaschonek, K., Schuler,
(1989) Interaction
of fluconazole
with

C. T., and
define the

P450
in
Pharmacol.

D. (1989) Interaction
1262

of

U., and Kruger, H. U.

cyclosporin.

Lancet 2,

104-105

43. Atiba,J.

0., Blaschke,

of ketoconazole
mephenytoin
161-165

on
and

T. F, and Wilkinson,

G. R. (1989) Effects

the polymorphic
4-hydroxylations
of 5debrisoquine.
Br. j
Glitz. Pharmacol.
28,

R. L., and
metabolism

The FASEB Journal

Received for publication September 30, 1991.

Accepted for publication October 25, 1991.

MAURICE

ET AL.