Catalyzation of cocaine N-demethylation by cytochromes P4502B, P4503A, and P4502D in fish liver

Article abstract of DOI:10.1002/jbt.10075

Cocaine N-demethylation by microsomal cytochrome P450s is the principal pathway in cocaine bioactivation and hepatotoxicity. P450 isozymes involved in N-demethylation of cocaine have not been elucidated yet and they differ from species to species. In humans and mice, P4503A contributes to cocaine N-demethylase activity, whereas in rats, both P4503A and P4502B participate. In the present study, contribution of different P450 isozymes to cocaine N-demethylase activity was studied in vitro with fish liver microsomes. The specific cocaine N-demethylase activity was found to be 0.672 ± 0.22 nmol formaldehyde formed/min/mg protein (mean ± SD, n = 6). Cocaine N-demethylase exhibited biphasic kinetics, and from the Lineweaver-Burk plot, two K_m values were calculated as 0.085 and 0.205 mM for the high- and low-affinity enzyme. These results indicate that N-demethylation of cocaine in mullet liver microsomes is catalyzed by at least two cytochrome P450 isozymes. Inhibitory effects of cytochrome P450 isozyme-selective chemical inhibitors, ketoconazole, cimetidine, SKF-525A, and quinidine, on cocaine N-demethylase activity were studied at 50, 100, and 500 μM concentrations of these inhibitors. At 100 μM final concentrations, ketoconazole (P4503A inhibitor), SKF-525A (inhibitor of both P4502B and P4503A), and cimetidine (P4503A inhibitor) inhibited N-demethylation activity by 73, 69, and 63%, respectively. Quinidine, P4502D-specific inhibitor, at 100 μM final concentration, reduced N-demethylation activity down to 64%. Aniline, a model substrate for P4502E1, did not alter N-demethylase activity in the final concentration of 100 μM. IC₅₀ values were calculated to be 20 μM for ketoconazole, 48 μM for cimetidine (both specific P4503A inhibitors), 164 μM for quinidine (P4502D inhibitor), and 59 μM for SKF-525A (inhibitor of both P4503A and P4502B). The contribution of P4502B to cocaine N-demethylase activity in mullet liver microsomes was further explored by the use of purified mullet cytochrome P4502B in the reconstituted system containing purified mullet P450 reductase and lipid. The turnover number was calculated as 4.2 nmol HCOH/(min nmol P450). Overall, these results show that P4503A and P4502B are the major P450s responsible for N-demethylation of cocaine, whereas contribution of P4502D is a minor one, and P4502E1 is not involved in the N-demethylation of cocaine in mullet liver microsomes.

Full text of DOI:10.1002/jbt.10075

J BIOCHEM MOLECULAR TOXICOLOGY
Volume 17, Number 3, 2003
Catalyzation of Cocaine N-Demethylation by
Cytochromes P4502B, P4503A, and P4502D in Fish Liver
Emel Arinc¸ and Azra Bozcaarmutlu
Biochemistry Graduate Program, Department of Biological Sciences, Middle East Technical University, 06531 Ankara, Turkey;
E-mail: earinc@metu.edu.tr
Received 14 January 2003; revised 11 April 2003; accepted 8 May 2003
ABSTRACT: Cocaine N-demethylation by microsomal
cytochrome P450s is the principal pathway in cocaine
bioactivation and hepatotoxicity. P450 isozymes in-
volved in N-demethylation of cocaine have not been
elucidated yet and they differ from species to species.
In humans and mice, P4503A contributes to cocaine N-
demethylase activity, whereas in rats, both P4503A and
P4502Bparticipate. Inthepresentstudy, contributionof
different P450 isozymes to cocaine N-demethylase ac-
tivity was studied in vitro with fish liver microsomes.
The specific cocaine N-demethylase activity was found
to be 0.672 0.22 nmol formaldehyde formed/min/mg
protein (mean SD, n = 6). Cocaine N-demethylase
exhibited biphasic kinetics, and from the Lineweaver–
Burk plot, two Km values were calculated as 0.085 and
0.205 mM for the high- and low-affinity enzyme. These
results indicate that N-demethylation of cocaine in
mullet liver microsomes is catalyzed by at least two
cytochrome P450 isozymes. Inhibitory effects of cy-
tochrome P450 isozyme-selective chemical inhibitors,
ketoconazole, cimetidine, SKF-525A, and quinidine,
on cocaine N-demethylase activity were studied at 50,
100, and 500 ꢀM concentrations of these inhibitors.
At 100 ꢀM final concentrations, ketoconazole (P4503A
inhibitor), SKF-525A (inhibitor of both P4502B and
P4503A), and cimetidine (P4503A inhibitor) inhibited
N-demethylation activity by 73, 69, and 63%, respec-
tively. Quinidine, P4502D-specific inhibitor, at 100 ꢀM
final concentration, reduced N-demethylation activity
down to 64%. Aniline, a model substrate for P4502E1,
did not alter N-demethylase activity in the final con-
centration of 100 ꢀM. IC50 values were calculated to
be 20 ꢀM for ketoconazole, 48 ꢀM for cimetidine
(both specific P4503A inhibitors), 164 ꢀM for quini-
dine (P4502D inhibitor), and 59 ꢀM for SKF-525A (in-
hibitor of both P4503A and P4502B). The contribution
of P4502B to cocaine N-demethylase activity in mullet
liver microsomes was further explored by the use of
purified mullet cytochrome P4502B in the reconstituted
system containing purified mullet P450 reductase and
lipid. The turnover number was calculated as 4.2 nmol
HCOH/(min nmol P450). Overall, these results show
that P4503A and P4502B are the major P450s respon-
sible for N-demethylation of cocaine, whereas contri-
bution of P4502D is a minor one, and P4502E1 is not
involved in the N-demethylation of cocaine in mul-
ꢀ
C
let liver microsomes.
2003 Wiley Periodicals, Inc. J
Biochem Mol Toxicol 17:169–176, 2003; Published on-
line in Wiley InterScience (www.interscience.wiley.com).
DOI 10.1002/jbt.10075
KEYWORDS: Cocaine N-Demethylation; Kinetics; Leap-
ingMullet;Fish;Liver;Ketoconazole;Cimetidine;SKF-
525A (Proadifen); Quinidine; Aniline
INTRODUCTION
Cytochrome P450 (CYP) comprises a superfamily
of hemoproteins that play a pivotal role in the activation
and detoxification of xenobiotics such as drugs, pesti-
cides, antioxidants, organic solvents, anesthetic agents,
dyes, environmental pollutants, and food additives,
and endogenous compounds including steroids, bile
acids, vitamin D3, fatty acids, prostaglandins, biogenic
amines, and retinoids [1–8]. Among the cytochrome
P450 gene families, CYP1, CYP2, and CYP3 are cur-
rently thought to be responsible from the majority of
hepatic xenobiotic metabolism [9].
Cocaine is a potent and dangerous central nervous
system (CNS) stimulant obtained from the leaves of
Erythroxylon coca. Cocaine is a highly addicting drug
and can lead to many complications [10]. Cocaine
also causes hepatotoxicity in humans and laboratory
animals [11–15]. The cocaine-induced hepatotoxicity
has been shown to be associated with the metabolites
of cocaine, that is norcocaine and N-hydroxycocaine
[12,16–18]. The first step in the bioactivation of co-
caine is the N-demethylation of cocaine to norcocaine
Correspondence to: Emel Arinc¸.
Contract Grant Sponsor: Turkish Academy of Sciences (TUBA).
¨
Contract Grant Sponsor: Middle East Technical University,
Ankara, Turkey.
Contract Grant Number: AFP-2001-01-08-02.
2003 Wiley Periodicals, Inc.
c
ꢀ
169

170
ARINC¸ AND BOZCAARMUTLU
Volume 17, Number 3, 2003
by cytochrome P450 isozymes [16,17,19–21]. The P450
isozymes involved in N-demethylation of cocaine have
not been clarified yet, and they differ from species
to species. In humans and mice, cytochrome P4503A
catalyzes the N-demethylation of cocaine [17,20,22],
whereas in rats, both P4503A and P4502B catalyze
this reaction [14]. The toxicity of cocaine has been
shown to be associated with only the P4502B-mediated
metabolism in rats [14,19,20].
salt (EDTA), formaldehyde (HCOH), glycerol, mag-
nesium chloride, and methanol were purchased from
E. Merck, Darmstadt, Germany. Ammonium acetate,
ε-amino caproic acid (ε-ACA), bovine serum albumin,
dilauroyl phosphatidycholine (DLPC), D-glucose-6-
phosphate monosodium salt, D-glucose-6-phosphate
dehydrogenase type XI, N-2-hydroxyethylpiperazine-
Nꢁ-2-ethanesulfonic acid (HEPES), ketoconazole,
ꢀ-nicotinamide adenine dinucleotide phosphate
(NADP+), phenylmethane sulfonyl fluoride (PMSF),
proadifen (SKF-525A, N,N-diethylaminoethyl-2,2-
diphenylvalerate hydrochloride), quinidine, and
sodium potassium tartrate were purchased from
Sigma Chemical Co., St. Louis, MI. All the other
chemicals were of analytical grade and were obtained
from commercial sources at the highest grade of purity
available.
The balance between detoxification and bioactiva-
tion of a compound in a particular species or organ is
highly dependent on the relative amounts and activi-
ties of the different forms of P450s that are expressed.
Thus, identification of individual cytochrome P450 en-
zymes responsible for formation of specific metabolites
of xenobiotics, including drugs, environmental toxins,
and procarcinogens, is important from evolutionary,
toxicological, and pharmacological standpoints. As in
mammals, different cytochrome P450 isozymes have
been identified in fish species [23–27]. Knowledge on
the metabolism of xenobiotics by fish P450-dependent
monooxygenases is important for environmental toxi-
cology, chemical carcinogenesis, and comparative bio-
chemistry [6,27]. There is no information on cocaine
metabolism and the role of specific P450 isozymes cat-
alyzing cocaine N-demethylation reaction in fish liver.
Like in the case of smaller laboratory animals, cocaine
has not been used as a veterinary drug in the fish. How-
ever, there exists extensive research on the involvement
of P450 enzymes in cocaine metabolism, the objective of
which is that the information gained from the smaller
laboratory animal studies would be useful for the com-
parative drug metabolism studies including those for
humans.
Leaping mullet (Liza saliens) is the most common
fish in the Mediterranean. It has been used in biomon-
itoring studies in Izmir Bay, Turkey [28–30]. Biochemi-
cal and immunochemical properties of mullet liver cy-
tochrome P450 reductase and P4501A1 have been char-
acterized [26,31–33].
In the present study, we aimed to determine the
specific isozymes of P450 participating in cocaine N-
demethylation reaction in the liver of leaping mullet
byusingisoform-selectivechemicalinhibitors, isoform-
specific substrate, and reconstitution studies using the
purified components.
Fish Collection
The fish used in this study, leaping mullet (Liza
saliens), were collected by fish net in Izmir Bay, Aegean
coast of Turkey. The fish collected were of comparable
size and weight (300–400 g).
Preparation of Leaping Mullet
Liver Microsomes
The livers, weighing approximately 2–5 g, were
removed immediately. The gall bladder was removed
carefully with scissors so as to avoid the spillage of its
contents that are known to be inhibitory to monooxy-
genase activities. The livers were placed in liquid ni-
trogen and transported from Izmir to the University
laboratories in Ankara, about 650 km away. Liver mi-
crosomes were prepared with differential centrifuga-
tion as described by Arinc¸ and Sen [34] except that
the homogenization solution contained 10 mM EDTA,
0.25 mM ε-ACA, and 0.1 mM PMSF. In the last step,
the washed microsomal pellet was resuspended in 10%
glycerol containing 10 mM EDTA. For each gram of
liver, 0.5 mL of suspension solution was used. Ten liv-
ers were used for the preparation of the first and second
batch of microsomes, which were used for characteriza-
tion of other mixed-function oxidase activities of mul-
let [35]. Other four batches were prepared from a single
liver. The aliquots of microsomal suspensions contain-
ing approximately 25–35 mg protein/mL were gassed
with nitrogen and stored in liquid nitrogen until use.
MATERIAL AND METHODS
Chemicals
Protein Determination
Cocaine-HCl was purchased from Tarim ve Koy Is-
leri Bakanligi, Toprak Mahsulleri Ofisi, Ankara, Turkey.
Acetic acid glacial, cupper sulfate, dimethyl sulfoxide
(DMSO), ethylenediaminetetraacetic acid disodium
Protein concentration in microsomes was deter-
mined according to the method of Lowry et al. [36]
using bovine serum albumin as a standard.

Volume 17, Number 3, 2003
COCAINE METABOLISM BY P450s IN FISH
171
90 U of cytochrome P450 reductase, and 0.04 mg of
synthetic lipid, dilauroyl phosphatidycholine (DLPC).
The mixture was mixed thoroughly and preincubated
for 20 min at room temperature. The reaction proceeded
for 10 min at 25◦C after addition of 0.5 mM NADPH
generating system in a final volume of 0.5 mL.
Determination of Cytochrome P450 Content
Cytochrome P450 content was determined from
the carbon monoxide difference spectra of dithion-
ite reduced samples, using an extinction coefficient of
91 mM−1 cm−1 for the difference in the absorption be-
tween 450 and 490 nm, as described by Omura and Sato
[37].
Kinetic Analysis of Cocaine
N-Demethylation Reaction
Determination of NADPH–Cytochrome
Enzyme kinetics were carried out using substrate
concentrations ranging from 0.1 to 4.0 mM (10 cocaine
concentrations). Kinetic parameters (apparent Km and
Vmax values) for the formation of norcocaine were an-
alyzed by Lineweaver–Burk plot of reaction velocity
(1/V) against 1/substrate concentration.
P450 Reductase Activity
NADPH–cytochrome P450 reductase activity was
assayed by measuring the rate of cytochrome c reduc-
tion at 550 nm spectrophotometrically, according to
the method of Masters et al. [38], except that reaction
was carried out at 25◦C. The assay mixture contained
0.3 M potassium phosphate buffer (pH 7.7), containing
80 nmol cytochrome c, 130 nmol NADPH, and appro-
priate amounts of enzyme preparation in a final volume
of 0.8 mL. The enzyme activity was calculated using
the extinction coefficient of 19.6 mM−1 cm−1 for differ-
ence in absorbance between reduced and oxidized cy-
tochrome c at 550 nm [39]. One unit of reductase was
defined as the amount of enzyme catalyzing the reduc-
tion of 1 nmol cytochrome c/min under the described
conditions.
Inhibition of Cocaine N-Demethylation
by Organic Solvents
Since the selected specific cytochrome P450
isozyme inhibitors, ketoconazole, SKF-525A (proad-
ifen), cimetidine, and quinidine, used in this study are
poorly soluble in water, organic solvents are used to
aid solubilization of these chemical inhibitors. The ef-
fects of frequently used solvents, methanol, ethanol,
and DMSO, on the enzyme activity were determined
by adding these solvents in a final concentration of
1% of the reaction mixture containing cocaine in the
absence of NADPH generating system. Control assay
was also carried out. This mixture was preincubated for
15 min at 25◦C with moderate shaking in water bath.
The reaction was initiated by adding NADPH generat-
ing system and the reaction proceeded exactly 10 min at
25◦C.
Determination of Cocaine N-Demethylase
Activity in Microsomes and Reconstituted
Systems
The N-demethylation of cocaine was measured
spectrophotometrically by measuring the amount of
formaldehyde formed according to the method of
Nash [40] as modified by Cochin and Alexrod [41].
A typical assay mixture contained 100 mM HEPES
buffer (pH 7.8), 3.0 mM cocaine, 0.5 mM NADPH
generating system, consisting of 2.5 mM magne-
sium chloride, 2.5 mM glucose-6-phosphate, 0.5 U
glucose-6-phosphate dehydrogenase (EC 1.1.1.49; 1 U
glucose-6-phosphate dehydrogenase reduces 1 nmol
NADP+/min at 25◦C), 14.6 mM HEPES buffer (pH 7.8),
0.5 mM NADP+, and 3 mg microsomal protein in a
final volume of 0.5 mL. The reaction was initiated
with the addition of NADPH generating system, and
carried out for 10 min at 25◦C with moderate shaking
in a water bath.
Inhibition of Cocaine N-Demethylation
by Selective Inhibitors
Several chemicals known as selective inhibitors of
P450 isozymes were preincubated in various concen-
trations with fish liver microsomes in the presence of
cocaine to determine their potential inhibitory effect
on cocaine N-demethylation reaction. Ketoconazole
(P4503A inhibitor), SKF-525A (inhibitor of both P4502B
and P4503A), cimetidine (P4503A inhibitor), and quini-
dine (P4502D inhibitor) were all dissolved in ethanol.
The volume of ethanol added to the reaction mixture
was 1% of the reaction volume. All four inhibitors were
preincubated with 3 mg microsomal protein contain-
ing 3 mM cocaine and buffer for 15 min at 25◦C before
initiating the enzymatic reaction with NADPH generat-
ing system. In addition, the reaction mixture containing
Reconstitution of cocaine N-demethylase activity
was performed by using electrophoretically pure mul-
let cytochrome P4502B [35] and cytochrome P450 re-
ductase [26]. The reconstitution medium contained
0.2 nmol of purified mullet cytochrome P4502B,

172
ARINC¸ AND BOZCAARMUTLU
Volume 17, Number 3, 2003
SKF-525A was also preincubated with microsomes and
NADPH generating system for 15 min at 25◦C, and the
reaction was initiated adding the substrate, cocaine.
of cocaine concentration on cocaine N-demethylation
activity. Ten different cocaine concentrations, starting
from 0.1 mM to 4 mM, were used in the activity
measurements. The activity appeared to be saturated
at or above 1.5 mM cocaine. Lineweaver–Burk plot
demonstrates biphasic kinetics (Figure 1B). The ap-
parent Km values were calculated as 0.085 and 0.205
mM, respectively. The respective Vmax values for the
high- and low-affinity fish liver enzyme activities
were found to be 0.388 and 0.485 nmol formalde-
hyde formed/min/mg protein. These results indicate
that N-demethylation of cocaine in mullet liver micro-
somes is catalyzed by at least two cytochrome P450
isozymes having different affinities for the substrate,
cocaine. On the contrary, kinetics of human and rat liver
microsomal enzymes exhibited a linear monophasic
enzymatic reaction after transformation of Michaelis–
Menten data to either Eadie–Hofstee or Lineweaver–
Burk plot. The Km values for rat [14] and adult human
[21] liver microsomal enzymes were determined as
0.42 mM and 1.8 mM cocaine, respectively. Similar to
what was observed with mullet enzyme in this study,
Pellinen et al. [22] reported that murine liver cocaine N-
demethylase enzyme also exhibited two Km values as
0.040–0.060 mM and 2–3 mM. It is important to note
that in the case of mullet, even the low-affinity en-
zyme with a Km value of 0.205 mM exhibited about 10-
fold higher affinity toward cocaine than did the human
enzyme.
Inhibition of Cocaine N-Demethylation
by a Selective Substrate
Aniline, a P4502E1 substrate, was also preincu-
bated with fish liver microsomes at 100 ꢁM final
aniline concentrations in the presence of cocaine for
15 min at 25◦C prior to the addition of NADPH gen-
erating system to determine whether it could inhibit
N-demethylation of cocaine.
RESULTS
Kinetics of Cocaine N-Demethylase Activity
Cocaine N-demethylase activity was measured in
six different microsomal preparations. The average spe-
cific activities were found to be 0.672
0.22 nmol
formaldehyde formed/min/mg protein (mean SD,
n = 6). Batch 1 and batch 2 microsomes were pre-
pared from 10 fish livers, while in the preparation of
remaining batches of microsomes, a single fish liver
was used. Specific activity of mullet liver enzyme
was found to be similar to that of rabbit liver [42],
while it was less than that in either rat (3.8 nmol
formaldehyde formed/min/mg protein) [14] or mice
(1.59 formaldehyde formed/min/mg protein) [15] liver
microsomes.
Effect of Solvents on Cocaine
N-Demethylase Activity
The kinetic parameters, apparent Km and Vmax of
cocaine N-demethylation, were determined in mul-
let liver microsomes. Figure 1A illustrates the effect
To find out the most suitable organic solvent for
the preparation of P450 selective inhibitors, the effects
FIGURE 1. Kinetics of fish liver cocaine N-demethylase activity. (A) Substrate saturation curve of cocaine N-demethylase activity. Incubation
mixture contained varying concentrations of cocaine ranging from 0.1 to 4.0 mM. The points are the average of duplicate determinations. (B) The
correponding Lineweaver–Burk double-reciprocal plot. The apparent Km values were calculated as 0.085 and 0.205 mM.

Volume 17, Number 3, 2003
COCAINE METABOLISM BY P450s IN FISH
173
of three frequently used solvents, ethanol, methanol,
and DMSO, on cocaine N-demethylase activity were
determined by co-incubating these solvents in a fi-
nal concentration of 1% with microsomes in the pres-
ence of 3 mM cocaine. Figure 2 illustrates the effects of
thesethreedifferentsolventsoncocaineN-demethylase
activity. The enzyme activity was inhibited approxi-
mately by 15% and 44% by ethanol and DMSO, re-
spectively. Ontheotherhand, cocaineN-demethylation
rate was enhanced by about 230% with 1% methanol
in the incubation medium. Since the organic solvent
with the least effect on cocaine N-demethylation reac-
tion was found to be ethanol, this solvent was used
to dissolve ketoconazole, SKF-525A, quinidine, and
cimetidine.
Inhibition of P450-dependent monooxygenation
reactions by ethanol, methanol, acetone, and DMSO
at 1% final concentration has been observed in several
in vitro xenobiotic metabolism studies [14,20,21,43,44].
The reason for the enhancement of cocaine N-
demethylase activity by about 230% with methanol is
not clear. Hickman et al. [44] reported that acetone at
1% final concentration increased P4501A2-dependent
caffeine N-demethylase activity by about 215% while
methanol at the same concentration inhibited the same
reaction by more than 95%. They did not attempt to
explain the observed differences between the solvents
but they recommended that inhibitory profile of each
organic solvent used for each substrate in the presence
of various P450s must be ascertained prior to use [44].
Participation of Cytochrome P4503A,
P4502B, and P4502D in Cocaine
N-Demethylation
Effects of various cytochrome P450 isozyme-
specific inhibitors on the enzyme activity were
determined in order to assess the participation of
specific P450 isoforms that may contribute to N-
demethylation of cocaine in mullet liver. Figure 3 shows
the effects of various inhibitors, at a final concentration
of 100 ꢁM, on cocaine N-demethylation reaction.
The contribution of P4503A to cocaine N-
demethylase activity was investigated using cimeti-
dine and ketoconazole as P4503A-selective inhibitors
[21,45–47].
As seen in Figure 3, ketoconazole and cimetidine
at 100 ꢁM final concentration inhibited cocaine N-
demethylase activity markedly by 73 and 63%, re-
spectively. Because both inhibitors were considered to
be selective for P4503A, these data indicated that co-
caine N-demethylation reaction was largely mediated
by P4503A subfamily in mullet liver.
As seen in Figure 3, quinidine, a P4502D-specific in-
hibitor, at 100 ꢁM final concentration reduced cocaine
N-demethylase activity to 64%, suggesting the involve-
ment of P4502D subfamily in cocaine N-demethylation
in fish.
Aniline, a model substrate for P4502E1, at a fi-
nal concentration of 100 ꢁM, was incubated with
cocaine to determine if it could inhibit the forma-
tion of norcocaine. As it can be seen in Figure 3,
FIGURE 2. The effects of three different solvents (ethanol, methanol,
and DMSO) on the rate of cocaine N-demethylation in leaping mullet
liver microsomes. Solvents in a final concentration of 1% were prein-
cubated with 3 mg microsomal protein containing 3 mM cocaine and
buffer for 15 min at 25◦C. Then the reaction was initiated with the
addition of NADPH generating system. Data are means SD of four
(for ethanol) and of two (for methanol and DMSO) independent ex-
periments performed in duplicate.
FIGURE 3. The effects of selective P450 inhibitors and a substrate
probe on the rate of cocaine N-demethylation in leaping mullet liver
microsomes. All chemicals were used at 100 ꢁM final concentration.
Each bar represents the mean of duplicate determinations.

174
ARINC¸ AND BOZCAARMUTLU
Volume 17, Number 3, 2003
aniline did not alter cocaine N-demethylation in mullet
liver.
As illustrated in Figure 3, SKF-525A (proadifen),
at a final concentration of 100 ꢁM, substantially in-
hibited cocaine N-demethylase activity by 69%. Nei-
ther preincubation of SKF-525A with NADPH nor its
preincubation with cocaine caused detectable differ-
ences in the degree of the inhibition. Since SKF-525A
is an inhibitor of both P4502B and P4503A subfamily
isoforms [48], the inhibition data by itself was not
conclusive.
Participation of P4502B in fish liver cocaine N-
demethylation activity was further investigated in a
reconstituted system containing purified mullet liver
P4502B [35], purified mullet P450 reductase [26], and
artificial lipid DLPC. The results given in Table 1
demonstrated that purified fish P4502B substantially
supported cocaine N-demethylase activity in the recon-
stituted system. For comparison, the results of reconsti-
tution studies carried out by Poet and co-workers [14]
with other purified P4502Bs are also given in Table 1.
Purified mullet liver P4502B isozyme had similar abil-
ity in supporting N-demethylation rate to that of rat
P4502B1 (Table 1). Poet et al. [14] showed that puri-
fied rat liver P4502B1 was three times more active in
supporting cocaine N-demethylation reaction than rat
P4502B2, while PB-inducible rabbit P4502B4 exhibited
the lowest activity (Table 1).
Furthermore, the inhibitory effects of SKF-525A,
ketoconazole, cimetidine, and quinidine on cocaine N-
demethylase activity were studied at varying concen-
trations (50, 100, 500 ꢁM) of these inhibitors so as to
determine the IC50 values. As illustrated in Figure 4,
all these chemicals inhibited cocaine N-demethylation
reaction in a concentration-dependent manner. At
500 ꢁM concentration, ketoconazole (P4503A in-
hibitor), SKF-525A (P4502B and P4503A inhibitor),
cimetidine (P4503A inhibitor), and quinidine (P4502D
inhibitor) inhibited cocaine N-demethylase activity by
96, 90, 78, and 73%, respectively. IC50 values were de-
termined by linear regression (% control) against loga-
rithmic inhibitor concentration and found to be 20 ꢁM
FIGURE 4. The effects of varying concentrations of SKF-525A,
ketoconazole, quinidine, and cimetidine on cocaine N-demethylase
activity. Values are the means of duplicate determinations.
(for ketoconazole), 59 ꢁM (for SKF-525A), 48 ꢁM (for
cimetidine), and 164 ꢁM (for quinidine).
DISCUSSION
Generally, more than one cytochrome P450 metab-
olizes the same compound. The P450 isoforms involved
in the various steps of cocaine hepatic metabolism have
not yet been completely elucidated and may differ from
species to species. The N-demethylation of cocaine to
norcocaine is suggested as the first step in the bioacti-
vation of cocaine [14,16–21].
In the present investigation, we determined
specific P450 isoforms participating in cocaine N-
demethylation reaction in fish, leaping mullet, by us-
ing P450 isoform-selective inhibitors, isoform-specific
substrate, and reconstitution studies using the purified
components.
The present study demonstrated that P4503A sub-
family participates in cocaine N-demethylation reac-
tion in mullet liver microsomes. This conclusion was
reached from the fact that both cimetidine and keto-
conazole, diagnostic inhibitors of P4503A [21,45–47],
potentially inhibited N-demethylase activity (Figures 3
and 4) with low IC50 values of 48 and 20 ꢁM, respec-
tively. The presence of members of P4503A family in
liver was demonstrated in several fish species [6,27].
Results obtained with the inhibition studies
carried out with quinidine, a P4502D-specific in-
hibitor [45,46], suggested the possible involvement of
P4502D subfamily in cocaine N-demethylase activity
in mullet. The expression of P4502D-like protein
in fish was previously shown (http://drnelson.-
utmem.edu/cytochromeP450.html). Participation of
P4502D subfamily in cocaine N-demethylase activity
TABLE1. CocaineN-DemethylationRatesinaReconstituted
System Containing Purified P4502Bs
Enzyme Activity
Purified P4502B
(nmol HCOH/min/nmol P450)
Fish liver P4502Ba
Rat liver P4502B1b
Rat liver P4502B2b
Rabbit liver P4502B4b
4.20
5.05
1.70
0.70
a Reconstitution medium contained 0.2 nmol of purified fish liver P4502B,
90 U of purified fish liver P450 reductase, and 0.04 mg of DPLC in a final
volume of 0.5 mL.
b Taken from Ref. [14].

Volume 17, Number 3, 2003
COCAINE METABOLISM BY P450s IN FISH
175
Molecular Aspects of Oxidative Drug Metabolizing En-
zymes: Their Significance in Environmental Toxicology,
Chemical Carcinogenesis and Health. NATO-ASI Series.
New York: Springer; 1995. pp 1–19.
has not been demonstrated in any species yet.
An indirect evidence came from a recent work of
Ramamoorthy et al. [49]. Like cocaine, ecstasy [( )-
3,4-methyenedioxymethamphetamine or MDMA] is a
CNS stimulant. In vitro, MDMA is demethylated to
( )-3,4-dihydroxymethamphetamine by the polymor-
phic P4502D6. Cocaine was shown to be a potent in-
hibitor of MDMA metabolism in both human liver
(P4502D6s) and P4502D6-expressing microsomes, with
very low Ki values in micromolar (2.2 ꢁM) and sub-
micromolar (0.2 ꢁM) range, respectively [49]. This
indirect evidence suggests that cocaine, as in the case
of “ecstasy,” is most likely metabolized by P4502D6 in
human liver.
Boyer and Peterson [50] observed that hepatotoxic
effect of cocaine increased significantly following the
treatment of mice with ethanol, but the major ethanol-
inducible form of P450, P4502E1, was not involved
in the metabolism of cocaine [19]. In agreement with
the results of Boelsterli et al. [19], our results given in
Figure 3 indicated no involvement of P4502E1 in co-
caine N-demethylase activity in fish liver microsomes.
SKF-525A has been used in xenobiotic metabolism
studies as a general inhibitor of the P450 activity since
1958. However, recent findings suggested that SKF-
525A did not uniformly inhibit all P450 forms but
the PB-induced forms, P4502Bs and P4503As, in rats
[12,48]. In fish microsomes, SKF-525A inhibited cocaine
N-demethylase activity (Figures 3 and 4) strongly with
a low IC50 value of 59 ꢁM, indicating possible involve-
ment of P4502B in cocaine N-demethylation reaction.
Reconstitution of fish cocaine N-demethylase activity
in the presence of purified mullet P4502B and P450 re-
ductase unequivocally demonstrated participation of
P4502B in cocaine N-demethylation reaction in mullet.
In conclusion, the results of this investigation indi-
cate that P4502E1 is not involved in N-demethylation
of cocaine. P4503A and P4502B are the major P450s re-
sponsible for N-demethylation of cocaine whereas con-
tribution of P4502D is minor in mullet liver.
6. Stegeman JJ. Diversity and regulation of cytochrome P450
in aquatic species. In: Arinc¸ E, Hodgson E, Schenkman JB,
editors. Molecular Aspects of Oxidative Drug Metaboliz-
ing Enzymes: Their Significance in Environmental Toxi-
cology, Chemical Carcinogenesis and Health. NATO-ASI
Series. New York: Springer; 1995. pp 135–158.
7. Nelson DR, Koymans L, Kamataki T, Stegeman JJ,
Feyereisen R, Waxman DJ, Waterman MR, Gotoh O, Coon
MJ, Estabrook RW, Gunsalus IC, Nebert DW. P450 su-
perfamily: Update on new sequences, gene mapping,
accession numbers, and nomenclature. Pharmacokinet-
ics 1996;6:1–43.
8. Newcomb M, Hollenberg PF, Coon MJ. Multiple mecha-
nisms and multiple oxidants in P450-catalyzed hydroxy-
lations. Arch Biochem Biophys 2003;409:72–79.
9. Wringhton SA, Stevens JC. The human hepatic cy-
tochrome P450 involved in drug metabolism. Crit Rev
Toxicol 1992;22:1–22.
10. Cregler LL, Mark H. Medical complications of cocaine
abuse. N Engl J Med 1986;315:1495–1500.
11. EvansMA, HarbisonRD. Cocaine-inducedhepaticnecro-
sis in mice. Toxicol Appl Pharmacol 1978;45:739–754.
12. Thompson ML, Shuster L, Shaw K. Cocaine-induced hep-
atic necrosis in mice. Biochem Pharmacol 1979;28:2389–
2395.
13. Perino LE, Warren GH, Levine JS. Cocaine-induced hep-
atotoxicity in humans. Gastroenterology 1987;93:176–
180.
14. Poet TS, Mcqueen CA, Halpert JR. Participation of cy-
tochrome P4502B and P4503A in cocaine toxicity in rat
hepatocytes. Drug Metab Dispos 1996;24:74–80.
15. Pellinen P, Stenback F, Kojo A, Honkakoski P, Gelboin HV,
Pasanen M. Regenerative changes in hepatic morphology
and enhanced expression of CYP2B10 and CYP3A during
daily administration of cocaine. Hepatology 1996;23:515–
523.
16. Kloss MW, Rosen GM, Rauckman EJ. N-Demethylation
of cocaine to norcocaine. Evidence for participation by
cytochrome P450 and FAD-containing monooxygenase.
Mol Pharmacol 1983;23:482–485.
17. LeDuc BW, Sinclair PR, Walton HS, Sinclair JF, Greenblatt
DJ, Shuster L. Cocaine toxicity in cultured chicken hepa-
tocytes: Role of cytochrome P450. Toxicol Appl Pharma-
col 1994;125:322–332.
18. Pellinen P, Kulmala L, Konttila J, Auriola S, Pasanen
M, Juvonen R. Kinetic characteristics of norcocaine N-
hydroxylation in mouse and human liver microsomes:
involvement of CYP enzymes. Arch Toxicol 2000;74:511–
520.
19. Boelsterli UA, Lanzotti AC, Goldlin C, Oertle M. Identifi-
cation of cytochrome P450IIBI as a cocaine-bioactivating
isoform in rat hepatic microsomes and in cultured rat
hepatocytes. Drug Metab Dispos 1992;20:96–101.
20. Poet TS, Brendel K, Halpert JR. Inactivation of cy-
tochrome P450 2B protects against cocaine-mediated
toxicity in rat liver slices. Toxicol Appl Pharmacol
1994;126:26–32.
REFERENCES
1. LuAYH, LevinW. Theresolutionandreconstitutionofthe
livermicrosomalhydroxylationsystem. BiochimBiophys
Acta 1974;344:205–240.
2. Arinc¸ E, Philpot RM. Preparation and properties of par-
tially purified pulmonary cytochrome P-450 from rabbits.
J Biol Chem 1976;251:3213–3220.
3. Nebert DW, Gonzalez FJ. P-450 genes: Structure, evolu-
tion, and regulation. Ann Rev Biochem 1987;56:945–993.
4. Guengerich FP. Enzymatic oxidation of xenobiotic chem-
icals. Crit Rev Biochem Mol Biol 1990;25:97–153.
5. Schenkman JB, Jansson I. Introduction to cytochrome
P450. In: Arinc¸ E, Hodgson E, Schenkman JB, editors.
21. Ladona MG, Gonzalez ML, Rane A, Peter RM, Torre RDL.
Cocaine metabolism in human fetal and adult liver micro-
somes is related to cytochrome P4503A expression. Life
Sci 2000;68:431–443.

176
ARINC¸ AND BOZCAARMUTLU
Volume 17, Number 3, 2003
22. Pellinen P, Stenbach F, Raunio H, Pelkonen O, Pasanen
M. Modification of hepatic cytochrome P450 profile by
cocaine-induced hepatotoxicity in DBA/2 mouse. Eur J
Pharmacol Env Toxicol Pharmacol 1994;292:57–65.
23. Arinc¸ E, Adalı O. Solubilization and partial purification
of two forms of cytochrome P450 from trout liver micro-
somes. Comp Biochem Physiol 1983;76B:653–662.
24. Stegeman JJ, Kloepper-Sams PJ. Cytochrome P450
isozymes and monooxygenase activity in aquatic ani-
mals. Environ Health Perspect 1987;71:87–95.
25. Andersson T, Fo¨rlin L. Regulation of cytochrome P450
enzyme system in fish. Aquatic Toxicol 1992;24:1–20.
26. Sen A, Arinc¸ E. Preparation of highly purified cy-
tochrome P4501A1 from leaping mullet (Liza saliens) liver
microsomes and biocatalytic, molecular and immuno-
chemicalproperties. CompBiochemPhys1998;121C:249–
265.
37. Omura T, Sato R. The carbon monoxide-binding pige-
ment of liver microsomes, II: Solubilization, purification
and properties. J Biol Chem 1964;239:2379–2385.
38. Masters BSS, Williams CH, Kamin H. The preparation
andpropertiesofmicrosomalTPNH-cytochromecreduc-
tase from pig liver. In: Coon MJ, Conney AH, Estabook
RW, Gelboin HV, Gillette JR, O’Brien PJ, editors. Methods
in Enzymology. New York: Academic Press; 1967.
pp 709–720.
39. Yonetani T. Studies on cytochrome c peroxidase, II: Stoi-
chiometry between enzyme, H2O2 and ferrocytochrome
c and enzymic determination of extinction coefficients of
cytochrome c. J Biol Chem 1965;210:4509–4514.
40. Nash T. The colorimetric estimation of formaldehyde by
means of the Hantzsch reaction. Biochem J 1953;55:416–
421.
41. Cochin J, Alexrod J. Biochemical and pharmacological
changes in the rat following chronic administration of
morphine, nalmorphine and normorphine. J Pharmacol
Exp Ther 1959;125:105–110.
42. Ulusoy G, Boyunegˇmez T, Bozcaarmutlu A. Kinetic and
biochemical evidence suggest that at least two isoforms
of P450 participates in N-demethylation of cocaine in
rabbit. In: Abstract Book, XVII National Biochemistry
Congress, Ankara, Turkey, June 24–27, 2002. pp 533–
534.
43. Arinc¸ E. Characterization of sheep liver and lung mi-
crosomal ethylmorphine N-demethylase. Comp Biochem
Physiol 1985;80B:389–399.
44. Hickman D, Wang JP, Wang Y, Unadkat JD. Evalua-
tion of the selectivity of in vitro probes and suitability
of organic solvents for the measurement of human cy-
tochrome P452 monooxygenase activities. Drug Metab
Dispos 1998;26:207–215.
45. Halpert JR, Guengerich FP, Bend JR, Correia MA. Con-
temporary issues in toxicology: Selective inhibitors of
cytochrome P450. Toxicol Appl Pharmacol 1994;125:163–
175.
46. Bournie M, Meunier V, Berger Y, Fabre G. Cytochrome
P450 isoform inhibitors as a tool for the investigation
of metabolic reactions catalyzed by human liver micro-
somes. J Pharmacol Exp Ther 1996;277:321–332.
47. Kasahara M, Suzuki H, Komiya I. Studies on the cy-
tochrome P450 (CYP)-mediated metabolic properties of
miocamycin: Evalution of the possibility of a metabolic
intermediate complex formation with CYP, and identi-
fication of human CYP isoforms. Drug Metab Dispos
2000;28:409–417.
27. Buhler DR, Wang-Buhler J-L. Rainbow trout cytochrome
P450s: Purification, molecular aspects, metabolic activity,
induction and role in environmental monitoring. Comp
Biochem Physiol 1998;121C:107–137.
28. Arinc¸ E, Sen A. Hepatic cytochrome P4501A and 7-
ethoxyresorufin O-deethylase induction in mullet and
common sole as an indicator of toxic organic pollution
in Izmir Bay, Turkey. Mar Environ Res 1999;48:147–160.
29. Arinc¸ E, Sen A, Bozcaarmutlu A. Cytochrome P4501A
and associated mixed-function oxidase induction in fish
as a biomarker for toxic carcinogenic pollutants in the
aquatic environment. Pure Appl Chem 2000;72:985–994.
30. Arinc¸ E, Kocabiyik S, Su E. Induced CYP1A mRNA,
protein and catalytic activity in the liver of feral fish,
leaping mullet, Liza saliens. Comp Biochem Physiol
2001;128C:281–290.
31. Sen A, Arinc¸ E. Purification and characterization of
cytochrome P450 reductase from feral leaping mullet
(Liza saliens) liver microsomes. J Biochem Mol Toxicol
1998;12:103–113.
32. Sen A, Arinc¸ E. Further immunochemical and biocatalytic
characterization of CYP1A1 from feral leaping mullet
liver (Liza saliens) microsomes. Comp Biochem Physiol
2000;126C:235–244.
33. Sen A, Hu CH, Urbach E, Wang-Buhler JL, Yang YH,
Arinc¸ E, Buhler DR. Cloning, sequencing and charac-
terization of CYP1A1 cDNA from leaping mullet (Liza
saliens) liver and implications for the potential functions
of its conserved amino acids. J Biochem Mol Toxicol
2001;15:243–255.
34. Arinc¸ E, Sen A. Characterization of cytochrome P450
dependent mixed function oxidase system of gilthead
seabream (Sparus aurata; Sparidae) liver. Comp Biochem
Physiol 1993;104B:133–139.
48. Murray M. Complexation of cytochrome P-450 isozymes
in hepatic microsomes from SKF 525-A-induced rats.
Arch Biochem Biophys 1988;262:381–388.
35. Bozcaarmutlu A. Cytochrome P4502B-Type Protein from
Feral Leaping mullet (Liza saliens) Liver Microsomes: Its
Isolation and Biochemical and Immunological Charac-
terization. PhD Thesis. Turkey: Middle East Technical
University Ankara; 2002. pp 1–209.
36. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein
measurement with the folin phenol reagent. J Biol Chem
1951;193:265–275.
49. Ramamoorthy Y, Yu A, Suh N, Haining RL, Tyndale RF,
Sellers EM. Reduced ( )-3,4-methylenedioxymetham-
phetamine (“Ecstasy”) metabolism with cytochrome
P4502D6 inhibitors and pharmacogenetic variants in
vitro. Biochem Pharmacol 2002;63:2111–2119.
50. Boyer CS, Peterson DR. Potentiation of cocaine-mediated
hepatotoxicity by acute and chronic ethanol. Alcohol Clin
Exp Res 1990;14:28–31.