Regio- and stereoselective oxidation of propranolol enantiomers by human CYP2D6, cynomolgus monkey CYP2D17 and marmoset CYP2D19

Article abstract of DOI:10.1016/j.cbi.2010.12.014

Toxic and pharmacokinetic profiles of drug candidates are evaluated in vivo often using monkeys as experimental animals, and the data obtained are extrapolated to humans. Well understanding physiological properties, including drug-metabolizing enzymes, of monkeys should increase the accuracy of the extrapolation. The present study was performed to compare regio- and stereoselectivity in the oxidation of propranolol (PL), a chiral substrate, by cytochrome P450 2D (CYP2D) enzymes among humans, cynomolgus monkeys and marmosets. Complimentary DNAs encoding human CYP2D6, cynomolgus monkey CYP2D17 and marmoset CYP2D19 were cloned, and their proteins expressed in a yeast cell expression system. The regio- and stereoselective oxidation of PL enantiomers by yeast cell microsomal fractions were compared. In terms of efficiency of expression in the system, the holo-proteins ranked CYP2D6 ≒ CYP2D17 CYP2D19. This may be caused by the bulky side chain of the amino acid residue at position 119 (leucine for CYP2D19 vs. valine for CYP2D6 and CYP2D17), which can disturb the incorporation of the heme moiety into the active-site cavity. PL enantiomers were oxidized by all of the enzymes mainly into 4-hydroxyproranolol (4-OH-PL), followed by 5-OH-PL and N-desisopropylpropranolol (NDP). In the kinetic analysis, apparent K_m values were commonly in the μM range and substrate enantioselectivity of R-PL < S-PL was observed in both K _m and V_max values for the formation of the three metabolites from PL enantiomers. The activity to produce NDP tended to be higher for the monkey enzymes, particularly CYP2D17, than for the human enzyme. These results indicate that in the oxidation of PL enantiomers by CYP2D enzymes, stereoselectivity is similar but regioselectivity is different between humans and monkeys.

Full text of DOI:10.1016/j.cbi.2010.12.014

Chemico-Biological Interactions 189 (2011) 146–152
Contents lists available at ScienceDirect
Chemico-Biological Interactions
journal homepage: www.elsevier.com/locate/chembioint
Regio- and stereoselective oxidation of propranolol enantiomers by human
CYP2D6, cynomolgus monkey CYP2D17 and marmoset CYP2D19
Shizuo Narimatsua,∗ , Toshiyuki Nakataa , Takeshi Shimizudania , Kenjiro Nagaokaa , Hironori Nakuraa ,
Kazufumi Masudab, Takashi Katsua, Akiko Koedac, Shinsaku Naitod, Shigeru Yamanoe, Atsuro Miyataf,
Nobumitsu Haniokaa
a Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama University, 1-1-1 Tsushima-naka, Kita-ku, Okayama 700-8530, Japan
b School of Pharmacy, Shujitsu University, 1-6-1 Nishigawara, Naka-ku, Okayama 703-8516, Japan
c Ina Research Inc., 1248-188 Nishiminowa, Ina, Nagano 399-4501, Japan
d Otsuka Pharmaceutical Factory Inc., Naruto, Tokushima 772-8601, Japan
e Faculty of Pharmaceutical Sciences, Fukuoka University, 8-19-1 Nanakuma, Jonan-ku, Fukuoka 814-0180, Japan
f
Kagoshima University, Graduate School of Medicine and Dentistry, 8-35-1 Sakuragaoka, Kagoshima 890-8520, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Toxic and pharmacokinetic profiles of drug candidates are evaluated in vivo often using monkeys as
experimental animals, and the data obtained are extrapolated to humans. Well understanding physio-
logical properties, including drug-metabolizing enzymes, of monkeys should increase the accuracy of the
extrapolation. The present study was performed to compare regio- and stereoselectivity in the oxida-
tion of propranolol (PL), a chiral substrate, by cytochrome P450 2D (CYP2D) enzymes among humans,
cynomolgus monkeys and marmosets. Complimentary DNAs encoding human CYP2D6, cynomolgus
monkey CYP2D17 and marmoset CYP2D19 were cloned, and their proteins expressed in a yeast cell
expression system. The regio- and stereoselective oxidation of PL enantiomers by yeast cell microsomal
fractions were compared. In terms of efficiency of expression in the system, the holo-proteins ranked
CYP2D6 ꢀ CYP2D17 ꢀ CYP2D19. This may be caused by the bulky side chain of the amino acid residue at
position 119 (leucine for CYP2D19 vs. valine for CYP2D6 and CYP2D17), which can disturb the incorpora-
tion of the heme moiety into the active-site cavity. PL enantiomers were oxidized by all of the enzymes
mainly into 4-hydroxyproranolol (4-OH-PL), followed by 5-OH-PL and N-desisopropylpropranolol (NDP).
In the kinetic analysis, apparent Km values were commonly in the ␮M range and substrate enantioselectiv-
ity of R-PL < S-PL was observed in both Km and Vmax values for the formation of the three metabolites from
PL enantiomers. The activity to produce NDP tended to be higher for the monkey enzymes, particularly
CYP2D17, than for the human enzyme. These results indicate that in the oxidation of PL enantiomers
by CYP2D enzymes, stereoselectivity is similar but regioselectivity is different between humans and
monkeys.
Received 16 September 2010
Received in revised form
22 November 2010
Accepted 8 December 2010
Available online 22 December 2010
Keywords:
Propranolol enantiomer
CYP2D6
CYP2D17
CYP2D19
Regioselectivity
Stereoselectivity
© 2010 Elsevier Ireland Ltd. All rights reserved.
1. Introduction
extrapolated to humans. In this context, monkeys such as cynomol-
gus monkeys, rhesus monkeys and marmosets are superior to
Since various human tissues and recombinant enzymes have
become available, the metabolic profiles of drug candidates can be
predicted fairly accurately in experiments in vitro. However, toxic
and pharmacokinetic profiles of drug candidates should still be
evaluated in vivo using experimental animals and the data obtained
non-primates such as rodents, rabbits, and dogs as experimental
animal species. Cynomolgus monkeys are rather big (5–10 kg in
body weight) as compared with marmosets (200–300 g), which
affects handling and feeding. Though cynomolgus monkeys and
rhesus monkeys have been used extensively in research into drug
metabolism and toxicology, relatively little data has been obtained
from marmosets.
Drug metabolism is divided into phase I reactions consisting
of oxidation, reduction and hydrolysis and phase II reactions con-
sisting of various kinds of conjugation. The oxidation catalyzed by
cytochrome P450s (CYPs) makes up about 80% of phase I reac-
tions [1]. CYPs compose a superfamily of hemethiolate enzymes,
and over 10,000 CYPs from animals, birds, fish, plants, microor-
Abbreviations: CYP, cytochrome P450; PL, propranolol; X-OH-PL, X-hydroxypro-
pranolol; NDP, N-desisopropylpropranolol; 4-OH-BTL, 4-hydroxybunitrolol; PVDF,
polyvinylidene difluoride; PCR, polymerase chain reaction; SRS, substrate recogni-
tion site.
∗
Corresponding author. Tel.: +81 86 251 7942; fax: +81 86 251 7942.
E-mail address: shizuo@pharm.okayama-u.ac.jp (S. Narimatsu).
0009-2797/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.cbi.2010.12.014

S. Narimatsu et al. / Chemico-Biological Interactions 189 (2011) 146–152
147
Table 1
ganisms etc. are known to exist [2]. Major isoenzymes of CYP1,
Primers used for the amplification of CYP2D17 cDNA.
2 and 3 are responsible for drug metabolism in humans, namely
CYP1A1/2, -2A6, -2B6, 2C8, -2C9, -2C19, -2D6, -2E1 and -3A4/5
[3]. CYP2D6 is clinically important because it contributes as the
major enzyme to the oxidation of 15% of clinically prescribed
medicines [4], though it accounts for only about 2% of all hep-
atic CYPs [5]. CYP2D6 shows extensive genetic polymorphism, and
some 80 allelic variants have been reported to date [6], resulting
in variation in drug-response phenotypes such as poor, intermedi-
ate, extensive and super-extensive metabolizers [4]. Cynomolgus
monkeys and marmosets also have many CYPs [2] including CYP2D
enzymes; CYP2D17 for cynomolgus monkeys [7] and CYP2D19 [8]
and CYP2D30 [9] for marmosets.
Propranolol (PL) is a classical adrenoceptor blocking agent used
clinically to treat arrhythmia and hypertension. PL has an asym-
metric carbon atom in its side-chain, yielding the enantiomers R-PL
and S-PL. Though S-PL has much more pharmacological activity as
a ␤-blocker than R-PL [10], PL is given as a racemate. PL undergoes
extensive metabolism in humans as shown in Fig. 1. For exam-
ple, it is oxidized at the aromatic 4- and 5-positions mainly by
CYP2D6 yielding 4-hydroxypropranolol (4-OH-PL) and 5-OH-PL,
respectively, whereas the oxidation of the PL side-chain is mainly
catalyzed by CYP1A2 giving N-desisopropylpropranolol (NDP) [11].
The oxidative metabolites as well as the parental compound are the
substrates for UDP-glucuronosyltransferases and sulfotransferases
[12,13]. PL is thus a useful substrate to study species differences in
the regio- and stereoselective metabolism by CYP and conjugation
enzymes.
Recently, we examined the oxidation of PL enantiomers by
microsomal fractions from cynomolgus monkey and marmoset liv-
ers, and compared it with that by a human liver microsomal fraction
[14]. As a result, we obtained experimental evidence that CYP2D
enzymes are involved not only in the ring hydroxylation at the
4- and 5-positions but also in the side-chain N-desisopropylation
in the monkey liver microsomal fractions [14]. In the present
study, we expressed cynomolgus monkey CYP2D17 and marmoset
CYP2D19 as well as human CYP2D6 in yeast cells, and compared
the oxidation of PL enantiomers by yeast cell microsomal fractions
among monkeys and humans.
Primer
Sequence
Tm (◦C)
Length
(bp)
CYP2D17-F1
CYP2D17-R1
CYP2D17-F2
CYP2D17-R2
ATGGAGCTAGATGCACTGGTGCCCCTGGC
CTAGCGGGGCACAGCACAAAGCTCATAGG
AAGCTTAAAAAAATGGAGCTAGATGCACTG
AAGCTTTCTAGCGGGGCACAGCACA
69.0
69.0
59.2
63.9
29
29
30
25
Underlined and italic letters indicate restriction enzyme sites and the Kozak
sequence, respectively.
monkeys were from Ina Research Co. Ltd. (Ina, Japan). Microsomal
fractions from cynomolgus monkey and marmoset livers were pre-
pared according to published methods [15]. Pooled human liver
microsomal fractions were obtained from BD Biosciences (San Jose,
CA).
2.2. Cloning of cDNA encoding CYP2D17
Total RNA was extracted from cynomolgus monkey liver using
the RNeasy minikit and QIA shredder according to the manufac-
turer’s instructions. The total RNA was reverse-transcribed to cDNA
using the RNA PCR kit v3.0. The full-length cDNA encoding CYP2D17
was amplified by PCR from single-strand cDNA templates using
CYP2D17-F1 and -R1 as primers (Table 1). These primers were
designed based on the nucleotide sequence in the flanking regions
of CYP2D17 cDNA (GenBank accession number Q29488). The PCR
reaction mixture contained 1× PCR buffer, 0.2 mM dNTPs, each
primer at 0.2 ␮M, 1.5 mM MgSO4, and 1 U of KOD plus DNA poly-
merase in a final volume of 50 ␮L. The PCR consisted of 30 cycles
with denaturation at 94 ◦C for 15 s, annealing at 65 ◦C for 30 s and
extension at 68 ◦C for 100 s. The PCR product was isolated and
purified by agarose electrophoresis, and the 5ꢁ- and 3ꢁ-ends of the
coding region were sequenced in both the forward and reverse
directions. The full-length cDNA obtained was modified by ampli-
fication with CYP2D17-F2 and -R2 as primers (Table 1). The PCR
was performed with a similar reaction medium to that as described
above though the annealing temperature was 58 ◦C. The PCR prod-
uct was isolated and purified by agarose electrophoresis. The PCR
product was introduced into the pGEM-T vector after A-tailing and
sequenced in both the forward and reverse directions. A DNA frag-
ment corresponding to CYP2D17 was cut from the pGEM-T vector
with HindIII, and subcloned into pGYR1 digested with the same
restriction enzyme. The insert of the plasmid was sequenced to ver-
ify the correct orientation with respect to the promoter for pGYR1.
Construction of the expression plasmids containing each of CYP2D6
and CYP2D19 was described previously [9].
2. Materials and methods
2.1. Materials
PL enantiomers as hydrochlorides were obtained from
Sigma–Aldrich (St. Louis, MO); 4-OH-PL and 5-OH-PL as hydrochlo-
rides from C/D/N Isotopes Inc. (Quebec, Canada); NDP as a
hydrochloride from AstraZenaca (Cheshire, England); and 4-
hydroxybunitrolol (4-OH-BTL) as a hydrochloride from Nippon
Boehringer Ingelheim Co. (Hyogo, Japan). The RNeasy Mini kit,
QIA shredder, and MiniElute Gel Extraction kit were purchased
from Qiagen (Heiden, Germany). The RNA PCR kit v3.0, DNA
ligation kit v2.1, Taq DNA polymerase, calf intestinal alkaline
phosphatase and HindIII were from Takara Bio (Ohtsu, Japan);
pGEM-T vector and T4 DNA ligase from Promega (Madison, WI);
KOD-plus DNA polymerase from Toyobo (Osaka, Japan); Quantum
Prep Plasmid Miniprep kit and polyvinylidene difluoride (PVDF)
membrane from BioRad (Hercules, CA); and BigDye terminator
cycle sequencing reaction kit v3.1 from Applied Biosystems (Foster
City, CA). Horse radish peroxidase-conjugated anti-rabbit IgG was
obtained from ICN Pharmaceuticals Inc. (Costa Mesa, CA). Enhanced
chemiluminescence-plus reagents were from GE Healthcare Bio-
Sciences Inc. (Little Chalfont, UK). Livers from adult male common
marmosets were supplied by Professor Atsuro Miyata of Kagoshima
University (Kagoshima, Japan). Livers from adult male cynomolgus
2.3. Expression of CYP2D enzymes
Saccharomyces cerevisiae AH22 was transformed with pGYR1
containing each of the CYP2D cDNAs by the lithium acetate method,
and the culture of yeast transformants thus obtained was per-
formed as described previously [16]. Microsomal fractions were
prepared from yeast cells expressing the CYP2D enzymes by meth-
ods reported previously [16]. The fractions were diluted to a protein
concentration of 5.0 mg/mL with 100 mM potassium phosphate
buffer (pH 7.4) containing 0.4% Emulgen 911 and 20% glycerol.
Total holo-CYP content was measured spectrophotometrically by
the method of Omura and Sato [17] using 91 mM−1 cm−1 as the
absorption coefficient. Appropriate portions of the yeast cell micro-
somal fractions together with microsomal fractions from human,
cynomolgus monkey and marmoset livers were subjected to
sodium dodecyl sulfate-polyacrylamide gel electrophoresis using
a 10% slab gel. Following the electrophoresis, proteins on the gel
were electroblotted to a PVDF membrane, and analyzed by Western

148
S. Narimatsu et al. / Chemico-Biological Interactions 189 (2011) 146–152
Fig. 1. Primary oxidation pathways of PL in humans. *Asymmetric carbon atom.
blotting according to published methods [18] employing polyclonal
antibodies raised against CYP2D6 (rabbit antiserum) prepared in
this lab as the primary antibody and peroxidase-conjugated goat
anti-rabbit IgG as the secondary antibody.
program Prism v5.0 (GraphPad Software, San Diego, CA). Intrinsic
clearance (CLint) was determined as the ratio Vmax/Km. All values are
expressed as the mean S.D. for three separate experiments with
independent preparations. Statistical comparisons were made with
Student’s t-test or Tukey’s test using Prism v5.0, and a difference
was considered significant when the p-value was <0.05.
2.4. Enzyme assay
The oxidative activities of PL enantiomers in the yeast cell
microsomal fractions were measured as described previously [14].
Briefly, an ice-cold reaction mixture containing 10 mM G-6-P, 1 IU
of G-6-P dehydrogenase, 1 mM NADP+, 5 mM MgCl2, 0.2–200 ␮M
R-PL or S-PL and 100 mM potassium phosphate buffer (pH 7.4) in
a final volume of 500 ␮L was preincubated at 37 ◦C for 1 min. The
reaction was started by adding the microsomal fraction (0.08, 0.05
and 0.5 mg protein/mL for CYP2D6, CYP2D17 and CYP2D19, respec-
tively) and stopped by adding 1 mL of 1 M aqueous NaOH containing
sodium bisulfite (25 mg/mL) as an antioxidant and mixing vigor-
ously. Then, 4-OH-BTL (200 pmol) as the internal standard was
added, and PL and metabolites were extracted into ethyl acetate
(4 mL) by shaking at room temperature. The organic layer was evap-
orated, and the residue was dissolved in 100 ␮L of the mobile phase
of HPLC described below. Calibration curves were made by adding
known amounts of PL metabolites to the incubation medium and by
extracting the metabolites without incubation in the same manner
as above. Protein concentrations were measured by the method
of Lowry et al. [19] using bovine plasma albumin as a standard.
The HPLC system consisted of an LC-10AD pump, a RF-10A fluores-
cence detector, a CTO-10A column oven, an SIL-10A injector and a
CBM-10A integrator (Shimadzu Co., Kyoto, Japan). The conditions
were as follows: column, Inertsil ODS-3 (4.6 mm i.d. × 250 mm,
5 ␮m, GL Science Co. Tokyo, Japan); column temperature, 30 ◦C;
mobile phase, 10 mM ammonium acetate buffer (pH 4.0)/methanol
(59:41 by volume); flow rate, 1.0 mL/min; detection, fluorescence
at 310/380 nm for excitation/emission wavelengths. In preliminary
experiments, the linearity in the metabolite formation was con-
firmed for protein concentrations and incubation time.
2.6. Molecular modeling
The homology models of CYP2D17 and CYP2D19 were
constructed using Swiss Model (http://swissmodel.exspasy.org/)
employing the crystallographic data of CYP2D6 (PDB ID, 2F9Q).
Hydrogen atoms were added to the models using the Biopolymer
module of Insight II software (Molecular Simulation Inc., San Diego,
CA). A heme moiety and six peptides as substrate recognition sites
(SRSs) were extracted; from 101 to 123 as SRS1, 205 to 223 as SRS2,
236 to 247 as SRS3, 296 to 311 as SRS4, 367 to 376 as SRS5 and 479
to 485 as SRS6. Energy optimization of the models was performed
using Insight II/Discover as described previously [20]. The active
site models were drawn using RasMol v2.6-ucb-1.0 [20].
3. Results
3.1. Expression of CYP2D enzymes in yeast cells
Yeast cell microsomal fractions expressing CYP2D6 and
CYP2D17 yielded typical reduced carbon monoxide (CO)-difference
spectra, while the CYP2D19 fraction showed a spectrum having a
low Soret peak at 450 nm (Fig. 2). The amounts of the recombinant
enzymes were 39.1 11.5, 63.3 18.8 and 4.38 0.83 pmol/mg
protein for CYP2D6, CYP2D17 and CYP2D19, respectively. The con-
centration of CYP2D19 was significantly lower than that of CYP2D6
or CYP2D17. Western blot analysis using the polyclonal antibody
against CYP2D6 gave a single protein band for each sample (human,
cynomolgus monkey and marmoset liver microsomal fractions, and
microsomal fractions from yeast cells expressing human, cynomol-
gus monkey and marmoset recombinant CYP2D enzymes) with a
similar migration profile (Fig. 3). The protein bands in the liver
microsomal fractions of cynomolgus monkeys and marmosets were
strongly stained whereas staining of the protein band of the human
2.5. Data analysis
Kinetic parameters (apparent Km and Vmax values) were esti-
mated by analyzing Michaelis–Menten plots using the computer

S. Narimatsu et al. / Chemico-Biological Interactions 189 (2011) 146–152
149
Fig. 2. Typical reduced CO-difference spectra of microsomal fractions from yeast
cells expressing CYP2D6, CYP2D17 and CYP2D19. The microsomal protein concen-
tration was 10 mg/mL.
sample was relatively weak. In contrast, the protein band for the
marmoset enzyme was weak compared with the bands for the
human and cynomolgus monkey enzymes.
3.2. Oxidation activity of PL enantiomers by recombinant CYP
enzymes
Employing PL enantiomers (each 100 ␮M) as substrates, func-
tions of the three recombinant CYP2D enzymes were examined. As
shown in typical HPLC chromatograms (Fig. 4), three PL metabolites
were eluted at the retention times of 4.6 min for 5-OH-PL, 5.6 min
for 4-OH-PL and 10.3 min for NDP, which were formed from both PL
enantiomers. The internal standard (4-OH-BTL) and the substrates
(PL enantiomers) were eluted at the retention times of 3.9 min
and 15.8 min, respectively. Under the fluorescence HPLC condi-
tions employed (excitation/emission wavelengths = 310/380 nm),
the peak heights ratio of 5-OH-PL:4-OH-PL:NDP in the same
amounts is 10:1:4, which was calculated from the calibration
curves. This is due to the difference in the fluorescence intensities
among the three metabolites.
Fig. 4. Typical HPLC chromatograms showing the formation of 4-OH-PL, 5-OH-PL
and NDP from PL enantiomers by human, cynomolgus monkey and marmoset CYP2D
enzymes. Thin line, formed from S-PL; thick line, formed from R-PL. (A) HM-CYP2D6,
(B) CYP2D17 and (C) CYP2D19 as enzyme sources. IS, internal standard (4-OH-BTL).
The substrate concentration used was 100 ␮M. HPLC conditions are given in Section
2.
In the case of CYP2D6 (Fig. 4A), the peak heights of the metabo-
lites formed were 5-OH-PL > 4-OH-PL > NDP. The peak heights for
all three metabolites showed a tendency of R-metabolites < S-
metabolites (Fig. 4A and Table 2), meaning that the oxidation
activities of S-PL are higher than those of R-PL. In the case of
CYP2D17 (Fig. 4B), the peak heights of the metabolites formed were
NDP > 4-OH-PL > 5-OH-PL. The peak heights for R-5-OH-PL and R-
NDP tended to be higher than those of corresponding S-metabolites,
whereas the peak heights for 4-OH-PL enantiomers were almost the
same (Table 2). In the case of CYP2D19 (Fig. 4C), the peak heights of
the metabolites were 5-OH-PL > 4-OH-PL > NDP, and the R/S ratio
in the metabolite formation was similar to the case of CYP2D17
(Table 2).
Table 2
Comparison of metabolites formed from PL enantiomers by recombinant CYP2D
enzymes.
Metabolite
Retention time (min)
Peak heighta
R/S ratiob
CYP2D6
R-5-OH-PL
S-5-OH-PL
R-4-OH-PL
S-4-OH-PL
R-NDP
4.62
4.62
5.59
5.59
10.28
10.28
103
151
138
203
34
0.68
0.68
0.57
S-NDP
60
CYP2D17
R-5-OH-PL
S-5-OH-PL
R-4-OH-PL
S-4-OH-PL
R-NDP
4.62
4.62
5.59
5.59
10.28
10.28
233
332
216
207
164
418
0.70
1.04
0.39
S-NDP
CYP2D19
R-5-OH-PL
S-5-OH-PL
R-4-OH-PL
S-4-OH-PL
R-NDP
4.62
4.62
5.59
5.59
10.28
10.28
140
222
93
93
31
0.63
1.00
0.50
Fig. 3. Western blot analysis of microsomal fractions from human and monkey liv-
ers, and from yeast cells expressing human and monkey recombinant enzymes.
Typical results for pooled microsomes from three independent preparations are
shown. The microsomal protein levels were 20 ␮g/lane for liver microsomes
and 0.5 ␮g/lane for yeast cell microsomes. HLM, human liver microsomes; CLM,
cynomolgus liver microsomes, MLM, marmoset liver microsomes, 2D6, CYP2D6;
2D17, CYP2D17; 2D19, CYP2D19. Conditions are given in Section 2.
S-NDP
62
a
Peak height values represent the detector response (␮V) in the HPLC (Fig. 3).
R/S ratio was calculated by dividing the peak height of the R-metabolite with
b
that of the corresponding S-metabolite.

150
S. Narimatsu et al. / Chemico-Biological Interactions 189 (2011) 146–152
Fig. 5. Typical Mechaelis–Menten plots for the formation of 4-OH-PL, 5-OH-PL and NDP from PL enantiomers by human, cynomolgus monkey and marmoset CYP2D enzymes.
(A)–(C) shows CYP2D6, CYP2D17 and CYP2D19, respectively. Circles, 4-OH-PL; triangles, 5-OH-PL; squares, NDP; open symbols and broken lines, formed from S-PL; closed
symbols and solid lines, formed from R-PL. Experimental conditions are given in Section 2.
3.3. Kinetic analysis of the oxidation of PL enantiomers by
recombinant CYP2D enzymes
by a polyclonal CYP2D6 antibody and quinidine, a specific inhibitor
of CYP2D6 [14]. In the present study, we cloned cDNA encoding
CYP2D17 from the total RNA fractions from cynomolgus monkey
liver, and expressed its protein in the yeast cell expression system.
Then, the oxidation of PL enantiomers by the recombinant CYP2D17
was compared with that by the human CYP2D6 and marmoset
CYP2D19 expressed in the same system as described previously
[9].
The formation of all metabolites from PL enantiomers fitted to
the Michaelis–Menten plots (Fig. 5). These plots demonstrated that
the major metabolite was 4-OH-PL for all CYP2D enzymes, and NDP
formed as a second major metabolite only for CYP2D17. The kinetic
parameters are summarized in Table 3. There was no significant
difference in the apparent Km values among the three recombinant
enzymes. From the view point of regioselectivity, the Vmax values
for the formation of 4-OH-PL and 5-OH-PL from both PL enan-
tiomers were significantly higher for CYP2D19 than for CYP2D6 and
CYP2D17. Furthermore, the Vmax value for the formation of NDP
from S-PL was significantly higher for CYP2D17 than for CYP2D6.
The CLint values for the formation of 5-OH-PL by CYP2D19 were
significantly higher than the corresponding values for the human
and cynomolgus monkey enzymes. On the other hand, the CLint
values for the formation of 4-OH-PL by the marmoset enzyme
were significantly higher than for CYP2D17. From the view point
of stereoselectivity, substrate enantioselectivity of R-PL < S-PL was
observed for the Km and Vmax values for the formation of all metabo-
lites from PL enantiomers by all of the recombinant enzymes.
Functional CYP levels in yeast cell microsomal fractions were
determined based on reduced CO-difference spectra. The level of
the CYP2D19 holo-protein was significantly lower than levels of the
human and cynomolgus monkey CYP2D enzymes. Similar results
were obtained in our previous study in which the level of CYP2D19
was much lower than that of CYP2D6 [9]. As shown in Fig. 2, the
peak heights at 450 and 420 nm were almost the same in the
reduced CO-difference spectrum for the recombinant CYP2D19.
A similar tendency was observed for the other two recombinant
CYP2D19 samples (data no shown). Considering the absorbance
coefficients of 91 and 111 mM−1 cm−1 for cytochromes P450 and
P420, respectively, in mammalian liver microsomes [17,22], the
total contents of functional (P450) and denatured (P420) CYP2D19
in yeast cell microsomal fractions were estimated to be less than
15% and 25% those of functional CYP2D17 and 2D6, respectively.
To search for a possible cause, we constructed a homology model
of CYP2D19 by Swiss Model on the website employing crystal-
lographic data of CYP2D6 in the present study. The active-site
structure consisting of six SRSs and the heme moiety was extracted
from the whole conformation of CYP2D19 and carefully compared
with that of CYP2D6 The possibility arises that an amino acid
residue, leucine, at position 119 may cause the low efficiency in the
expression of the holo-protein for CYP2D19. That is, the amino acid
at the corresponding position is valine for CYP2D6 and CYP2D17,
both of which yield typical reduced CO-difference spectra (Fig. 2).
The side chain (isobutyl group) of leucine is larger than that (iso-
propyl) of valine as shown in Fig. 6, indicating that appropriate
4. Discussion
We previously examined the oxidative metabolism of PL enan-
tiomers by liver microsomal fractions from cynomolgus monkeys
and marmosets, and found that the kinetic profiles were consid-
erably different between the two monkey species [14]. However,
we obtained some lines of experimental evidence that not only PL
aromatic ring hydroxylation at the 4- and 5-positions but also side-
chain N-desalkylation was mediated by monkey CYP2D enzymes,
because the formation of the three metabolites from PL enan-
tiomers by the monkey liver microsomal fractions was inhibited

S. Narimatsu et al. / Chemico-Biological Interactions 189 (2011) 146–152
151
Table 3
Kinetic parameters for the oxidation of PL enantiomers by microsomal fractions from yeast cells expressing human, cynomolgous monkey and marmoset CYP2D enzymes.
Metabolite
Km (␮M)
Vmax (pmol/min pmol CYP)
CLint (␮L/min pmol CYP)
R
S
R/S
R
S
R/S
R
S
R/S
CYP2D6
4-OH-PL
5-OH-PL
NDP
1.25 0.41
1.06 0.73
2.52 1.22
2 .43 0.68
2.56 0.65
3.52 0.78
0.51
0.62
0.78
24.2 5.7
1.60 0.42
1.39 0.50
34.9 7.4 0
2.20 0.53
2.30 0.66
0.69
0.73
0.61
20.1 3.9
1.11 0.38
0.57 0.09
14.6 1.6
0.86 0.02
0.66 0.19
1.38
1.28
0.87
CYP2D17
4-OH-PL
5-OH-PL
NDP
2.99 0.38*
2.88 0.33
3.14 0.30
5.04 1.22*
4.51 1.07
5.25 1.22*
0.59
0.64
0.60
32.0 8.5
2.49 0.68
4.70 2.06
35.2 11.5
4.30 1.47
14.6 6.40*
0.91
0.58
0.32
11.1 4.5
0.89 0.36
1.51 0.70
7.26 4.32
1.04 0.61
3.12 2.23
1.46
0.86
0.49
CYP2D19
4-OH-PL
5-OH-PL
NDP
2.30 0.71
2.32 0.73
4.54 3.55
3.97 1.28
3.81 1.22
6.32 2.66
0.58
0.61
0.72
57.0 10.4*,‡
7.13 2.27*,‡
4.23 0.92*,#
68.3 9.6*,‡
12.8 3.8*,‡
9.64 1.64#
0.83
0.56
0.44
25.4 3.2*
17.8 3.0‡,#
3.41 0.51*,‡
1.63 0.41*,‡
1.42
0.90
0.76
3.07 0.07*,‡
1.24 0.59*,‡
Each value represents the mean S.D. for three separate experiments with three independent preparations. Significant differences were calculated with Tukey’s test for
species difference or Student’s t-test for enantioselectivity.
*
p < 0.05 compared with CYP2D6.
p < 0.05 compared with R-PL.
p < 0.05 compared with CYP2D17.
#
‡
incorporation of the heme moiety may be disturbed with the bulky
side-chain of Leu-119 during the biosynthesis of the CYP2D19 pro-
tein.
formation of 4-OH-PL and 5-OH-PL, and (5) the NDP-forming activ-
ity, particularly from S-PL, was much higher in CYP2D17 than in
CYP2D6. Furthermore, the kinetic profiles of the PL oxidation by
the recombinant enzymes were similar to those by liver microso-
mal fractions from humans, cynomolgus monkeys and marmosets
obtained in our previous studies [14], i.e., the Km values for the
recombinant enzymes were close to the values of low-Km phases
for the liver microsomal fractions. However, S/R ratios in the Km val-
ues were slightly different between the recombinant enzymes and
the liver microsomal fractions [14], which may be due to involve-
ment of some other CYP enzyme(s) in addition to CYP2D6, CYP2D17
or CYP2D19 in PL oxidation in each liver microsomal fraction.
Employing the crystal structure of CYP2D6, we constructed
homology models of CYP2D19 and CYP2D17 by the Swiss Model,
and performed docking simulations on a personal computer using
the Insight II/Discover program. We did not obtain any potential
docking model, which can explain why the NDP-forming activity is
higherin CYP2D17 and/orCYP2D19 than in CYP2D6. Further studies
are necessary to elucidate mechanism(s) causing the species dif-
ferences in DNP formation between human and monkey enzymes
together with the possibility of the existence of another CYP2D
enzyme in marmoset livers.
In the Western blot analysis of liver microsomal fractions
(Fig. 3), however, marmoset liver microsomes as well as cynomol-
gus monkey liver microsomes showed a thick protein band that
immunochemically reacted with the polyclonal antibody raised
against CYP2D6. Recently, Uno et al. [21] succeeded in cloning a
cDNA encoding a novel CYP2D44 from cynomolgus monkey liver
and characterized its enzymatic properties. We also demonstrated
previously the expression of two CYP2D genes in marmoset liv-
ers, i.e., CYP2D19 and CYP2D30 were found to be expressed in
marmoset livers supplied from Kagoshima University and from
Kyoto University, respectively [9]. In the present study, we used
marmoset livers supplied from Kagoshima University, and so the
possibility arises that an unknown CYP2D enzyme, like cynomolgus
monkey CYP2D44, may express together with CYP2D19 in mar-
moset livers obtained from Kagoshima University.
In general, the kinetic profiles for the oxidation of PL enan-
tiomers were similar among the three recombinant enzymes; (1)
apparent Km values were around several ␮M, (2) substrate enan-
tioselectivity of R-PL < S-PL was observed in both the Km and Vmax
values for the three metabolites formed from PL enantiomers, and
(3) the most abundant metabolite was commonly 4-OH-PL for
all of the recombinant enzymes. On the other hand, considerable
differences were observed in the following points; (4) compared
with CYP2D6 and CYP2D17, CYP2D19 had higher activity for the
In summary, cynomolgus monkey, marmoset and human CYP2D
enzymes were expressed in a yeast cell system, and the regio- and
stereoselective oxidations of PL enantiomers by yeast cell micro-
somal fractions were compared. In efficiency of expression in the
system, the ranking was CYP2D6 ꢀ CYP2D17 ꢀ CYP2D19. PL enan-
Fig. 6. Comparison of the active-site structure between the CYP2D19 wild type having leucine (A) and its mutant having valine (B) at position 119. The active-site structure
consists of six SRSs and the heme moiety. A space-filling model is used to depict the amino acid residue at position 119.

152
S. Narimatsu et al. / Chemico-Biological Interactions 189 (2011) 146–152
tiomers were mainly oxidized to 4-OH-PL, followed by 5-OH-PL
and NDP by all of the enzymes. In kinetic analysis, apparent Km
values were similar and substrate enantioselectivity of R-PL < S-PL
was observed in both Km and Vmax values for the three metabolites
formed from PL enantiomers. The NDP-forming activity tended to
be higher in the monkey enzymes, particularly CYP2D17, than in
the human enzyme. These results indicate that in the oxidation
of PL enantiomers by CYP2D enzymes, stereoselectivity is similar
but some differences exist in regioselectivity between humans and
monkeys.
G.T. Tucker, S.W. Ellis, Cloning and functional expression of a novel marmoset
cytochrome P450 2D enzyme, CYP2D30: comparison with the known mar-
moset CYP2D19, Biochem. Pharmacol. 68 (2004) 165–175.
[10] A. Barrret, V. Cullum, The biological properties of the optical isomers of pro-
pranolol and their effects on cardiac arrhysmias, Br. J. Pharmacol. 34 (1968)
443–455.
[11] Y. Masubuchi, S. Hosokawa, T. Horie, S. Ohmori, M. Kitada, T. Suzuki, S. Nari-
matsu, Cytochrome P-450 isozymes involved in propranolol metabolism in
human liver microsomes: the role of CYP-2D6 as ring-hydroxylase and CYP1A2
as N-desisopropyrase, Drug Metab. Dispos. 22 (1994) 909–915.
[12] B. Silber, N.H. Holford, S. Riegelman, Stereoselective disposition and glu-
curonidation of propranolol in human, J. Pharm. Sci. 71 (1982) 699–704.
[13] T. Walle, U.K. Walle, D.R. Knapp, E.C. Conradi, E.M. Bargar, Identification of
major sulfate conjugates in the metabolism of propranolol in dog and man,
Drug Metab. Dispos. 11 (1983) 344–349.
[14] T. Shimizudani, K. Nagaoka, N. Hanioka, S. Yamano, S. Narimatsu, Compara-
tive study of the oxidation of propranolol enantiomers in hepatic and small
intestinal microsomes from cynomolgus and marmoset monkeys, Chemico-
Biol. Interact. 183 (2010) 67–78.
Conflict of interest
The author declares that there are no conflicts of interest.
[15] S. Narimatsu, M. Gotoh, Y. Masubuchi, T. Horie, S. Ohmori, M. Kitada, T.
Kageyama, K. Asaoka, I. Yamamoto, T. Suzuki, Stereoselectivity in bunitrolol
4-hydroxylation in liver microsomes from marmosets and Japanese monkeys,
Biol. Pharm. Bull. 19 (1996) 1429–1433.
[16] S. Narimatsu, M. Oda, H. Hichiya, T. Isobe, K. Asaoka, N. Hanioka, S. Yamano,
S. Shinoda, S. Yamamoto, Molecular cloning and functional analysis of
cytochrome P450 1A2 from Japanese monkey liver: comparison with marmoset
cytochrome P450 1A2, Chemico-Biol. Interact. 152 (2005) 1–12.
[17] T. Omura, R. Sato, The carbon monoxide-binding pigment of liver microsomes
I. Evidence for its hemoprotein nature, J. Biol. Chem. 239 (1964) 2370–2378.
[18] F.P. Guengerich, P. Wang, N.K. Davidson, Estimation of isozymes of microsomal
cytochrome P-450 in rats, rabbits, and humans using immunochemical stain-
ing coupled with sodium dodecyl sulfate-polyacrylamide gel electrophoresis,
Biochemistry 21 (1982) 1698–1706.
[19] O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, Protein measurement with
the Folin phenol reagent, J. Biol. Chem. 193 (1951) 265–275.
[20] K. Masuda, K. Tamagake, Y. Okuda, F. Torigoe, D. Tsuzuki, T. Isobe, H. Hichiya,
N. Hanioka, S. Yamamoto, S. Narimatsu, Change in enantioselectivity in bufu-
ralol 1ꢁꢁ-hydroxylation by the substitution of phenylalanine-120 by alanine in
cytochrome P450 2D6, Chirality 17 (2005) 37–43.
[21] Y. Uno, S. Uehara, S. Kohara, N. Murayama, H. Yamazaki, Cynomolgus monkey
CYP2D44 newly identified in liver, metabolizes bufuralol and dextromethor-
phan, Drug Metab. Dispos. (2010), Fast forward, published on May 25 as
doi:10.112/dmd.110.033274.
[22] T. Omura, R. Sato, The carbon monoxide-binding pigment of liver microsomes II.
Solubilization, purification and properties, J. Biol. Chem. 239 (1964) 2379–2385.
References
[1] W.E. Evans, M.V. Relling, Pharmacogenomics: translating functional genomics
into rational therapeutics, Science 286 (1999) 487–491.
[2] The home page of Dr. David Nelson; http://drnelson.uthsc.edu/
CytochromeP450.html.
[3] S. Rendic, F.J. Di Carlo, Human cytochrome P450 enzymes: a status report sum-
marizing their reactions, substrates, inducers and inhibitors, Drug Metab. Rev.
29 (1997) 413–580.
[4] U.M. Zanger, M. Turpeinen, K. Klein, M. Schwarb, Functional pharmacogenet-
ics/genomics of human cytochrome P450 involved in drug biotransformation,
Anal. Bioanal. Chem. 392 (2008) 1093–1108.
[5] T. Shimada, H. Yamazaki, M. Mimura, Y. Inui, F.P. Guengerich, Interindividual
variations in human liver cytochrome P-450 enzymes involved in the oxidation
of drugs, carcinogens and toxic chemicals: studies with liver microsomes of 30
Japanese and 30 Caucasians, J. Pharmacol. Exp. Ther. 270 (1994) 414–423.
[6] Home page of the human cytochrome P450 (CYP) allele nomenclature commit-
tee. http://www.cypalleles.ki.se/cyp2d6.htm.
[7] D.C. Mankowski, K.J. Laddison, P.A. Christpherson, S. Ekins, D.J. Tweedie, M.P.
Lawton, Molecular cloning, expression, and characterization of CYP2D17 from
cynomolgus monkey liver, Arch. Biochem. Biophys. 372 (1999) 189–196.
[8] T. Igarashi, T. Sakuma, M. Isogai, R. Nagata, T. Kamataki, Marmoset liver
cytochrome P450s: study for expression and molecular cloning of their cDNAs,
Arch. Biochem. Biophys. 339 (1997) 85–91.
[9] H. Hichiya, S. Kuramoto, S. Yamamoto, S. Shinoda, N. Hanioka, S. Narimatsu,
K. Asaoka, A. Miyata, S. Iwata, M. Nomoto, T. Satoh, K. Kiryu, N. Ueda, S. Naito,