1521-009X/43/5/735–742$25.00 http://dx.doi.org/10.1124/dmd.115.063594DRUG METABOLISM AND DISPOSITION Drug Metab Dispos 43:735–742, May 2015Copyright ª 2015 by The American Society for Pharmacology and Experimental Therapeutics

Activation and Deactivation of 1-Methyl-4-Phenyl-1,2,3,6-Tetrahydropyridine by Cytochrome P450 Enzymes and

Flavin-Containing Monooxygenases in Common Marmosets(Callithrix jacchus)

Shotaro Uehara, Yasuhiro Uno, Takashi Inoue, Norie Murayama, Makiko Shimizu, Erika Sasaki,and Hiroshi Yamazaki

Laboratory of Drug Metabolism and Pharmacokinetics, Showa Pharmaceutical University, Machida, Tokyo, Japan (S.U., N.M.,M.S., H.Y.); Pharmacokinetics and Bioanalysis Center, Shin Nippon Biomedical Laboratories, Ltd., Kainan, Wakayama, Japan (Y.U.);Department of Applied Developmental Biology, Central Institute for Experimental Animals, Kawasaki, Kanagawa, Japan (T.I., E.S.);

and Keio Advanced Research Center, Keio University, Minato-ku, Tokyo, Japan (E.S.)

Received January 28, 2015; accepted March 3, 2015

ABSTRACT

The potential proneurotoxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine(MPTP) induces Parkinson-like syndromes in common marmosets,other primates, and humans. MPTP is metabolically activated to1-methyl-4-phenyl-2,3-dihydropyridinium and 1-methyl-4-phenylpyridiniumions (MPDP+ and MPP+, respectively) by desaturation reactions.MPTP is deactivated to 4-phenyl-1,2,3,6-tetrahydropyridine (PTP) byN-demethylation and is also deactivated to MPTP N-oxide. The roles ofcytochrome P450 (P450) enzymes and flavin-containing monooxygenases(FMOs) in the oxidative metabolism of MPTP-treated marmosetsare not yet fully clarified. This study aimed to elucidate P450- andFMO-dependent MPTP metabolism in marmoset liver and brain.Rates of MPTP N-oxygenation in liver microsomes were similar tothose in brain microsomes from 11 individual marmosets (substrateconcentration, 50 mM) and were correlated with rates of benzydamine

N-oxygenation (r = 0.75, P < 0.05); the reactions were inhibited bymethimazole (10 mM). MPTP N-oxygenation was efficiently mediatedby recombinantly expressedmarmoset FMO3. Rates of PTP formation byMPTP N-demethylation in marmoset liver microsomes were correlatedwith bufuralol 19-hydroxylation rates (r = 0.77, P < 0.01) and weresuppressed by quinidine (1 mM), thereby indicating the importance ofmarmoset CYP2D6 in PTP formation. MPTP transformations to MPDP+

and MPP+ were efficiently catalyzed by recombinant marmoset CYP2D6and human CYP1A2. These results indicated the contributions of multipledrug-metabolizing enzymes to MPTP oxidation, especially marmosetFMO3 in deactivation (N-oxygenation) and marmoset CYP2D6 for bothMPTP deactivation and MPTP activation to MPDP+ and MPP+. Thesefindings provide a foundation for understanding MPTP metabolism andfor the successful production of preclinical marmoset models.

Introduction

Cytochrome P450 (P450; EC 1.14.14.1) enzymes have beencharacterized with respect to drug metabolism and disposition(Johansson and Ingelman-Sundberg, 2011). Research has focused onthe P450s as well as on another monooxygenase family, the flavin-containing monooxygenases (FMOs; EC 1.14.13.8) (Yamazaki andShimizu, 2013), which are involved in the oxidation of a variety ofcompounds associated with pharmacological and/or toxicologicaleffects in humans. The potential proneurotoxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) degenerates dopaminergic neuronsand causes parkinsonism in primates, including common marmosets,cynomolgus monkeys, and humans (Langston and Ballard, 1983;Ballard et al., 1985; Davis et al., 1997). Several metabolic pathways of

MPTP have been reported and are summarized in Fig. 1. The 1-methyl-4-phenyl-2,3-dihydropyridinium ion (MPDP+) (Fig. 1) isreportedly an unstable compound that undergoes further oxidation toform the active toxic 1-methyl-4-phenylpyridinium ion (MPP+) (Chibaet al., 1985). The metabolic transformation of MPTP to its toxicmetabolite MPP+ via intermediate MPDP+ is mediated by monoamineoxidase or CYP2D6 in human brains (Langston and Ballard, 1983;Ballard et al., 1985; Davis et al., 1997). On the other hand, MPTP isalso metabolized to non-neurotoxic 4-phenyl-1,2,3,6-tetrahydropyridine(PTP) and MPTP N-oxide (Fig. 1) by CYP2D6 and FMO, respectively,in human livers (Bajpai et al., 2013; Herraiz et al., 2013).The common marmoset (Callithrix jacchus) is a member of the

New World nonhuman primate family Callitrichidae (Abbott et al.,2003; Mansfield, 2003; Carrion and Patterson, 2012; Okano et al.,2012; Tokuno et al., 2012). The species has attracted considerableattention as a potentially useful animal model in fields such asneuroscience and drug toxicology (Mansfield, 2003) because of itssize, availability, unique biologic characteristics (Abbott et al., 2003),and evidence of crossreactivity with human cytokines and hormones

This research was supported in part by the Ministry of Education, Culture,Sports, Science and Technology of Japan [Grant-in-Aid for Scientific Research].This work also resulted from “Construction of System for Spread of PrimateModel Animals”, under the Strategic Research Program for Brain Sciences of theMinistry of Education, Culture, Sports, Science and Technology of Japan.

dx.doi.org/10.1124/dmd.115.063594.

ABBREVIATIONS: FMO, flavin-containing monooxygenase; HPLC, high-performance liquid chromatography; MPDP+, 1-methyl-4-phenyl-2,3-dihydropyridinium ion; MPP+, 1-methyl-4-phenylpyridinium ion; MPTP, 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine; P450, cytochrome P450;PCR, polymerase chain reaction; PTP, 4-phenyl-1,2,3,6-tetrahydropyridine; RT, reverse transcription.

735

at ASPE

T Journals on February 15, 2021

dmd.aspetjournals.org

Dow

nloaded from

(Okano et al., 2012). In particular, MPTP-treated marmosets havebeen used for preclinical studies on Parkinson disease (Eslamboli,2005; Jenner, 2009). The parkinsonian marmosets have showncharacteristic neural degeneration of nigrostriatal dopamine neuronsand behavioral signs such as moving tremors, immobility, musclerigidity, positional dysfunction, and L-dopa–induced dyskinesia (Andoet al., 2012, 2014). However, we do not yet have a comprehensiveunderstanding of common marmoset drug-metabolizing enzymesrelated to MPTP metabolism, partly because of the absence of anavailable reference genome sequence or enzyme characteristics.Although we recently carried out identification and quantitativeanalysis of tissue-specific mRNA transcripts of P450s and FMOs inthe marmoset as an animal model in drug development (Shimizu et al.,2014), limited numbers of P450 isoforms (Uno et al., 2011) and noFMOs in the marmoset have been investigated in terms of their drug-metabolizing activities thus far.This study combined analyses to identify novel common marmoset

FMO1/FMO3 and P450 enzymes that have been unreported until nowand to investigate MPTP oxidative metabolism. We report herein thatmarmoset FMO3 mainly contributed to deactivation of MPTP and thatmarmoset CYP2D6 was responsible for both deactivation and metabolicactivation of MPTP to MPP+. These findings should prove an importantresource for future biomedical research and will facilitate use of thecommon marmoset as an animal preclinical parkinsonism model.

Materials and Methods

Animals. Adult common marmosets (aged .2 years) were purchased fromCLEA Japan (Tokyo, Japan). The animals were maintained in cages (400 �610 � 1578 mm) at 24�C–27�C and 40%–60% relative air humidity with a 12-hour/12-hour light/dark cycle and had free access to a balanced diet (CMS-1M;CLEA Japan) with added vitamins and water. This study was approved by theanimal ethics committees and gene recombination experiment safety managementcommittees of the Central Institute for Experimental Animals and was performedin accordance with the 2006 Science Council of Japan Guidelines for ProperConduct of Animal Experiments. Animal care was conducted in accordance withthe recommendations of the 2011 Institute for Laboratory Animal ResourcesGuide for the Care and Use of Laboratory Animals. Brain and liver samples werecollected from 11 common marmosets (five males and six females, aged 2–6 years)after euthanasia by exsanguination under ketamine (60 mg/kg) and isoflurane deepanesthesia as previously described (Shimizu et al., 2014).

Chemicals and Enzymes.MPTP, MPP+, 7-ethoxyresorufin, chlorzoxazone,S-mephenytoin, warfarin, ketoconazole, sulfaphenazole, quinidine, and ticlopidinewere purchased from Sigma-Aldrich (St. Louis, MO). Coumarin, midazolam,

and a-naphthoflavone were from Wako Pure Chemical Industries (Osaka,Japan). Paclitaxel, bufuralol hydrochloride, PTP, and MPTP N-oxide wereobtained from Toronto Research Chemicals (Toronto, Canada), Corning (NewYork, NY), Tokyo Chemical Industry (Tokyo, Japan), and KNC Laboratories(Kobe, Japan), respectively. Microsomal fractions from marmoset tissue sam-ples were prepared as previously described (Yamazaki et al., 2002, 2014).Microsomal samples from a baculovirus-insect cell line (Supersomes) ex-pressing human CYP1A1, CYP2A6, CYP2B6, CYP2C8, CYP2C9, CYP2C19,CYP2D6, CYP2E1, CYP3A4, and CYP3A5 were purchased from Corning.The other chemicals and reagents used were obtained in the highest gradecommercially available.

Molecular Cloning of cDNAs for Marmoset P450s and FMOs. To obtaincDNAs of marmoset P450s and FMOs, reverse transcription (RT)–polymerasechain reaction (PCR) was performed using brain and liver total RNAs of marmosets.Total RNAwas isolated from each tissue using RNeasy Mini Kits (Qiagen, Valencia,CA) according to the protocols in the manufacturer’s instructions. Then, a first-strandcDNA was prepared by RT reaction at 50�C for 1 hour in a mixture containing 1 mgtotal RNA, oligo(dT), and SuperScript III RT reverse transcriptase (Invitrogen,Carlsbad, CA). PCR was performed using KOD-Plus-Neo DNA polymerase(Toyobo, Osaka, Japan) with the RT product as described by the protocols in themanufacturer’s instructions. Full-length cDNAs encoding marmoset P450s andFMOs were amplified by PCR with the following oligonucleotide primers:59-CAGATGGCATTGTCCCAGTTTGTTC-39 and 59-CGGTGTCTTCCTCAC-TGGAAGGAG-39 for CYP1A2, 59-ATGGAGCTCACCGTCTTCCTCTTC-39and 59-CTGGATGACCCGGAATCTCTTGAC-39 for CYP2B6, 59-GCTC-ATAGTTGTCTTAGTAAGAAGAGTAGGCTTCAAT-39 and 59-TCAGACAG-GAATGAAGCAGATCTGGTACGG-39 for CYP2C8, 59-CAGGGGTGTCCAGAGGAGTTCAGT-39 and 59-TCTAGCGGGGCACAGCACAAAG-39 forCYP2D6, 59-ATGTCTGCCCTCGGCATGACTGTG-39 and 59-CATATGCAAA-GAAAGGAATAGGTTTGAGGAA-39 for CYP2E1, 59-CGGAGGAGAGAGA-TAAGGAAGGATAGTAGT-39 and 59-CTTAGGGAAATTCAGGCTCCACTTACA-39 for CYP3A4, 59-CAAACAGCAGTACACAACTGAAAGGAGA-39and 59-CTGGGGCACAGCTTTCTTGAAACA-39 for CYP3A5, 59-ACAGCAGC-ACACAACTGAAAGGAAA-39 and 59-CTGGGGCACAGCTTTTTTGATGC-39for CYP3A90, 59-GAGAACATGGGCAAGCGAGTTG-39 and 59-CCTTTAG-GAAATCTTTTACTCATAGGAAAATCAG-39 for FMO1, and 59-ATGGGGAA-GAAAGTGGCCATCA-39 and 59-ATGATGATTAGGTCAACACAAGGAAAACAG-39 for FMO3. PCR was performed using an Applied BiosystemsGene Amp PCR system 9700 (Applied Biosystems, Foster City, CA)consisting of an initial denaturation at 94�C for 2 minutes and 35 cycles of98�C for 10 seconds, 65�C for 30 seconds, and 68�C for 2 minutes,followed by a final extension at 68�C for 7 minutes. The PCR productswere cloned into pCR4 vectors using a TOPO Cloning Kit (Invitrogen)according to the manufacturer’s instructions. Sequencing of the inserts wascarried out using an ABI PRISM BigDye Terminator v3.0 Ready ReactionCycle Sequencing Kit (Applied Biosystems) with an ABI PRISM 3730DNA Analyzer (Applied Biosystems).

Fig. 1. Metabolic activation (desaturation reac-tions) and deactivation (N-demethylation andN-oxygenation) pathways of MPTP.

736 Uehara et al.

at ASPE

T Journals on February 15, 2021

dmd.aspetjournals.org

Dow

nloaded from

Recombinant Marmoset P450s and FMOs. Recombinant marmoset P450s(CYP1A2, CYP2B6, CYP2C8, CYP2D6, CYP2E1, CYP3A4, CYP3A5, andCYP3A90) and FMOs (FMO1 and FMO3) were heterologously expressed inEscherichia coli using expression plasmids. For P450s, the N terminus modifica-tion was conducted by PCR using the following forward and reverse primers:59-GGAATTCCATATGGCTCTGTTATTAGCAGTTTTTTTCTCAGCCACAG-AGCTTCTCCT-39 and 59-GCTCTAGACGGTGTCTTCCTCACTGGAAGG-39for CYP1A2, 59-GGAATTCCATATGGCTCTGTTATTAGCAGTTTTTTTT-GCACTCCTCACAGGCCTTTTGC-39 and 59-GCTCTAGACTTCAGCGGG-GCAGGAA-39 for CYP2B6, 59-GGAATTCCATATGGCTCTGTTATT-AGCAG-TTTTTCTCTGTCTCTCTTTTTTGCTTCTCTTTTCAC-39 and59-GCTCTAGATCAGACAGGAATGAAGCAGATCTGGTA-39 for CYP2C8,59-GGAATTCCATATGGCTCTGTTATTAGCAGTTTTTCTGGCTGTGACAGTGG-39and 59-GCTCTAGACTAGCGGGGCACAGCACA-39 for CYP2D6,59-GGAATTCCATATGGCTCTGTTATTAGCAGTTTTTGCCCTGCTG-ATATGGGCGGCCATTCTCCTGCT-39 and 59-GCTCTAGACGAGTGT-GTCCTCCTCACACTGAT-39 for CYP2E1, 59-GGAATTCCATATGGCTC-TGTTATTAGCAGTTTTTGTGGAAACCTGGCTTCTCCTG-39 and59-GCTCTAGAGGAAATTCAGGCTCCACTTACAGTC-39 for CYP3A4, and59-GGAATTCCATATGGCTCTGTTATTAGCAGTTTTTGTGGAAACCTGGCTTCT-CCTG-39 and 59-GCTCTAGATCATTCTCCACTTAGGGTTCCATCTC-39 forCYP3A5/90, respectively. The NdeI and XbaI sites (underlined) in the forwardand reverse primers, respectively, were used for subcloning of the product intopCW vectors that contained human NADPH-P450 reductase cDNA (Uno et al.,2010). Membrane preparation and measurement of P450 protein and reductasecontents in each sample were performed as previously described (Uehara et al.,2010). Expected drug oxidation activities of marmoset P450 enzymes wereconfirmed with typical human P450 probe substrates (Yamazaki et al., 2002;Uehara et al., 2011).

For FMOs, PCR was carried out with the following primers: 59-ATGGGCAAGC-GAGTTGCCATTGT-39 and 59-CCGCTCGAGCCTTTAGGAAATCTTTTACTCA-TAGGAAAATCAG-39 for FMO1, and 59-ATGGGGAAGAAAGTGGCCATCA-39and 59-CCGCTCGAGTTAGGTCAACCCAAGGAAAACAGCAA-39 for FMO3.After restriction enzyme digestion using XhoI (the restriction site is underlined),the PCR products were subcloned to pET30 vectors (Novagen, Madison, WI)to provide a 6� His-tag at the N terminus (Uno et al., 2013). Protein expressionand E. coli membrane preparations were performed as previously described(Yamazaki et al., 2014) in a similar manner to that used for human FMO1 andFMO3 preparations. The final amount of recombinant FMO proteins in bacterialmembranes was normalized to the flavin adenine dinucleotide contents (Yamazakiet al., 2014). Expected drug oxygenation activities of marmoset FMO1/FMO3 wereconfirmed with the typical FMO probe substrate benzydamine (Yamazaki et al.,2014).

MPTP Oxidation Catalyzed by Marmoset P450s and FMOs. The fourmetabolites of MPTP, namely PTP, MPTP N-oxide, MPDP+, and MPP+ (Fig.1), were determined as previously described (Herraiz et al., 2013) with minormodifications. Briefly, reaction mixtures of 0.25 ml contained enzymesources (microsomal preparations or E. coli membranes), 100 mM potassiumphosphate buffer (pH 7.4), an NADPH-generating system (0.5 mM NADP+,5 mM glucose 6-phosphate, 5 mM MgCl2, and 1 U/ml glucose-6-phosphatedehydrogenase), and 1–2000 mM MPTP, unless otherwise specified. Afterincubation for 10 minutes at 37�C, the reaction was terminated by theaddition of 25 ml 60% (v/v) perchloric acid. The linearity for incubation timeand protein concentrations was confirmed under these conditions. Thesupernatant obtained by centrifugation (10,000 � g, 5 minutes) was analyzedby high-performance liquid chromatography (HPLC) equipped with UV (at290 nm) and fluorescence detection (excitation at 332 nm and emission at344 nm) with a Mightysil RP-18 GP Aqua column (4.6 � 150 mm, 5 mm;Kanto Chemical, Tokyo, Japan) at a flow rate of 1.0 ml/min using a mobilephase consisting of 0.05 M ammonium acetate (pH 5.5) and acetonitrile. Thegradient conditions for elution were 1%–25.6% acetonitrile (0–8 minutes),25.6%–72% acetonitrile (8–15 minutes), 72% acetonitrile (15–20 minutes), and1% acetonitrile (20–30 minutes). PTP and MPTP N-oxide were quantified on thebasis of their standard curve peak areas with F332/344. MPDP+ was quantified onthe basis of the standard curve peak areas at A290 of MPP+.

The activities of 7-ethoxyresorufin O-deethylation, coumarin 7-hydroxylation,paclitaxel 6a-hydroxylation, warfarin 7-hydroxylation, S-mephenytoin49-hydroxylation, bufuralol 19-hydroxylation, chlorzoxazone 6-hydroxylation,

midazolam 19-hydroxylation, and benzydamine N-oxygenation in brain or livermicrosomes were determined as previously described (Yamazaki et al., 2002;Uehara et al., 2011). Drug and MPTP oxidation analyses were carried out intriplicate determinations. The kinetic analysis of MPTP oxidations was performedusing a nonlinear regression analysis in a Michaelis–Menten model (Kaleida-Graph; Synergy Software, Reading, PA).

Results

MPTP Oxidation by Marmoset Liver and Brain Microsomes.Representative HPLC chromatograms of MPTP metabolites inmarmoset liver microsomes are shown in Fig. 2, A and B, after 10-minute incubation of MPTP (50 mM) in the presence of an NADPH-generating system. The four metabolites of MPTP shown in Fig. 1could be detected with UV (at 290 nm) and fluorescence detection(excitation at 332 nm and emission at 344 nm) in an analyticalreverse-phase HPLC system. To elucidate the metabolic pathwaysof MPTP in the marmoset, the formation rates of PTP, MPTP N-oxide, MPDP+, and MPP+ mediated by liver and brain microsomesfrom 11 individual marmosets were determined (Fig. 3). Althoughrates of MPTP N-oxygenation in brain microsomes were similar tothose in liver microsomes from marmosets at a substrate concen-tration of 50 mM, formation rates of PTP, MPP+, and MPDP+ inbrain microsomes were lower than those in liver microsomes.MPDP+ and MPP+ formation rates in liver microsomes (Fig. 3, Band D) were approximately 2- and 4-fold faster, respectively, thanPTP formation rates (Fig. 3A). The formation rates of four MPTPmetabolites in individual marmosets varied less than 5-fold withinthe liver and brain microsomes analyzed in this study (Fig. 3);larger interindividual variations in the marmoset were observed forMPP+ formation in the brain (4.7-fold). Interestingly, PTPformation rates (Fig. 3A) correlated with MPDP+ formation rates(Fig. 3B) in marmoset liver microsomes (r = 0.77, n = 11, P , 0.05).Correlation analyses revealed that MPTP N-demethylation (PTP

formation) activities were correlated with activities of midazolam 19-hydroxylation (r = 0.92, n = 10, P , 0.001), bufuralol 19-hydroxylation (r = 0.77, n = 10, P , 0.01), and 7-ethoxyresorufinO-deethylation (r = 0.73, n = 10, P , 0.05) (Table 1). Significantcorrelation coefficients (P , 0.05) were also observed betweenMPDP+ and MPP+ formations and warfarin 7-hydroxylation activities,between MPP+ formation and midazolam 19-hydroxylation activities,and between MPTP N-oxygenation and benzydamine N-oxygenation(Table 1). No correlations were seen between MPTP metaboliteformation and coumarin 7-hydroxylation, paclitaxel 6a-hydroxylation,S-mephenytoin 49-hydroxylation, or chlorzoxazone hydroxylationactivities under these conditions.The effects of P450 and FMO inhibitors (a-naphthoflavone,

sulfaphenazole, ticlopidine, quinidine, ketoconazole, and methimazole) onMPTP oxidation in liver microsomes were investigated at substrateconcentrations of 50 mM MPTP (Fig. 4). Quinidine (1–10 mM),a CYP2D inhibitor, strongly inhibited MPTP N-demethylation (PTPformation) (Fig. 4A) and moderately suppressed MPP+ formation (Fig.4D) in marmoset liver microsomes. Ticlopidine (2–20 mM), a CYP2Cinhibitor, moderately suppressed MPTP N-demethylation (Fig. 4A) andMPP+ formation (Fig. 4D) in liver microsomes. a-Naphthoflavone (0.2–5mM), a CYP1A inhibitor, partly suppressed MPDP+ formation (Fig. 4B)in liver microsomes. Methimazole (10–100 mM), a FMO substrate andinhibitor, strongly suppressed MPTP N-oxygenation in liver micro-somes (Fig. 4C). By contrast, inhibitory effects of ketoconazole (0.02–0.2 mM) or sulfaphenazole (1–20 mM), inhibitors of CYP3A4 andCYP2C9, respectively, were not seen on MPTP oxidation under theseconditions.

MPTP Activation by Marmoset CYP2D6 737

at ASPE

T Journals on February 15, 2021

dmd.aspetjournals.org

Dow

nloaded from

MPTP Metabolism by Recombinantly Expressed P450s andFMOs. MPTP (50 mM) was incubated with recombinant marmosetFMO1 and FMO3 and marmoset CYP1A2, CYP2B6, CYP2C8, CYP2D6,CYP3A4, CYP3A5, and CYP3A90, along with human FMO1 andFMO3 and human CYP1A2, CYP2A6, CYP2B6, CYP2C8, CYP2C9,

CYP2C19, CYP2D6, CYP2E1, CYP3A4, and CYP3A5. MarmosetFMO3 (Fig. 2C) catalyzed MPTP N-oxygenation (14.5 nmol/min pernmol FMO, Table 2) more efficiently than FMO1 did; a similar patternwas evident for human FMO3 (78.0 nmol/min per nmol FMO) versusFMO1. Marmoset CYP2D6 (Fig. 2D) efficiently catalyzed PTP

Fig. 2. Representative HPLC chromatograms of MPTPoxidized by marmoset liver microsomes (A and B) andrecombinant FMO3 (C) and CYP2D6 (D). MPTP and itsmetabolites were analyzed with a reverse-phase liquidchromatography system using fluorescence and UVdetection as described in Materials and Methods.

Fig. 3. Variation in formation rates of PTP (A),MPDP+ (B), MPTP N-oxide (C), and MPP+ (D) fromMPTP by liver and brain microsomes of 11 individualmarmosets. MPTP (50 mM) was incubated with liverand brain microsomes (0.50 mg/ml; black and whitebars, respectively) from 11 individual marmosets for10 minutes in the presence of an NADPH-generatingsystem. The individual marmosets were numbered inthe increasing order of MPP formation rates in livermicrosomes.

738 Uehara et al.

at ASPE

T Journals on February 15, 2021

dmd.aspetjournals.org

Dow

nloaded from

formation (6.6 nmol/min per nmol P450) in a similar manner to humanCYP2D6 (3.5 nmol/min per nmol P450) (Fig. 5). Marmoset CYP2D6also mediated MPDP+ and MPP+ formation (2.0, and 5.7 nmol/minper nmol P450, respectively), but the rates of MPDP+ and MPP+

formation mediated by human CYP2D6 were much lower (Fig. 5).Marmoset CYP1A2 and CYP2C8 also mediated the formation of PTP,MPDP+, and MPP+, but to a lesser extent than CYP2D6. In addition,human CYP1A2 showed high activities for the formation of MPDP+

and MPP+, whereas human CYP2D6 showed much lower activitiesunder these conditions.Kinetic parameters for MPTP oxidation by marmoset and human

liver microsomes and recombinant CYP2D6 and FMO3 were determined.

Apparent Km values for PTP formation by liver microsomes frommarmosets and humans were 451 and 577 mM, respectively; forrecombinant marmoset and human CYP2D6, Km values were 71 and125 mM, respectively (Table 3). The calculated Vmax/Km values forPTP formation were 0.95 and 0.23 ml/min per mg protein and 220 and92 ml/min per nmol P450, respectively. Marmoset CYP2D6 showeda roughly 2-fold Vmax/Km value for PTP formation compared with thatof human CYP2D6. Apparently consistent Km values (162 mM and137 mM) for MPTP N-oxygenations by marmoset liver microsomesand marmoset FMO3, respectively, were obtained in a similar mannerto those in humans. Apparently consistent Km values (66 mM and72 mM) for MPP+ formation by marmoset liver microsomes and

TABLE 1

Correlation coefficients between MPTP oxidation and probe drug oxidation activities in liver microsomes from 11 individual marmosets

Probe Reaction Activity P450 Isoform or FMOaCorrelation Coefficient for Rates of Formation of MPTP Metabolites

PTP MPTP N-Oxide MPDP+ MPP+

7-Ethoxyresorufin O-deethylation CYP1A 0.73* 20.77 0.52 0.56Coumarin 7-hydroxylation CYP2A 20.16 20.37 20.17 20.15Paclitaxel 6a-hydroxylation CYP2C 0.29 20.64 0.62 0.57Warfarin 7-hydroxylation CYP2C 0.60 20.79 0.68* 0.70*S-Mephenytoin 49-hydroxylation CYP2C 0.30 20.60 0.29 0.27Bufuralol 19-hydroxylation CYP2D 0.77** 20.40 0.38 0.47Chlorzoxazone hydroxylation CYP2E 20.27 0.38 20.45 20.46Midazolam 19-hydroxylation CYP3A 0.92*** 20.65 0.62 0.73*Benzydamine N-oxygenation FMO 0.63 0.75* 0.51 0.62

aThe P450 isoform or FMO mainly responsible for catalyzing the probe reactions.*P , 0.05; **P , 0.01; ***P , 0.001.

Fig. 4. Effects of chemical inhibitors on the formation ratesof PTP (A), MPDP+ (B), MPTP N-oxide (C), and MPP+ (D)from MPTP in marmoset liver microsomes. Percentages of thecontrol (without inhibitors) are shown. a-Naphthoflavone,sulfaphenazole, ticlopidine, quinidine, ketoconazole, andmethimazole, respectively, inhibit CYP1A, CYP2C9, CYP2C,CYP2D, CYP3A4, and FMO.

MPTP Activation by Marmoset CYP2D6 739

at ASPE

T Journals on February 15, 2021

dmd.aspetjournals.org

Dow

nloaded from

marmoset CYP2D6 were observed with a high Vmax/Km value (150 ml/minper nmol CYP2D6), but the Km value for MPP+ formation in humanliver microsomes was much lower than that of human CYP2D6,resulting in a low Vmax/Km value (8.5 ml/min per nmol) for humanCYP2D6. In terms of kinetic parameters for MPDP+ formation, asimilar pattern of differences between marmosets and humans wasseen to that for MPP+ formation.

Discussion

Oxidation of potential proneurotoxin MPTP has been reported invarious animal species, including mice, rats, and humans (Chiba et al.,1990; Coleman et al., 1996; Yoshihara et al., 2000). In vitro studiesusing rat liver microsomes have shown approximately 20 times fasterMPTP N-oxygenation rates than MPTP N-demethylation rates (Cash-man and Ziegler, 1986). In this study, fast MPTP N-oxygenationreactions in vitro were confirmed in marmoset liver and brainmicrosomes (Fig. 3). These facts suggest that MPTP N-oxygenationmight be the major detoxification pathway in most animal species,including common marmosets. MPTP N-oxygenation in marmoset livermicrosomes was likely mediated by FMO3 because of suppression of

the reaction by methimazole (an inhibitor for FMO, Fig. 4) and thehigh metabolic capacity of marmoset FMO3 compared with FMO1(Table 3). In terms of FMO-mediated reactions, apparent Km valuesfor NADPH-dependent drug oxygenations were generally in the rangeof approximately 50–100 mM (Yamazaki et al., 2014), presumablybecause FMO exhibits a stable 4a-flavin hydroperoxide intermediatecapable of oxygenating both nucleophiles and electrophiles in itscatalytic cycle, even in the absence of an oxygenatable substrate(Jones and Ballou, 1986).In this study, the main roles of liver microsomal CYP2D6 and

FMO3 in MPTP oxidation in common marmosets were demonstrated(Fig. 5; Table 3). Marmoset liver microsomes were the main focus ofthis study to investigate P450-dependent metabolic activation ofMPTP, because efficient catalytic activities of liver microsomes withrespect to MPTP oxidation were seen compared with the generallylower activities for marmoset brain microsomes. It could be noted thatour correlation analysis using 10 marmoset liver microsomes withrelatively fewer interindividual variations might not be as predictive asthe chemical inhibitor and recombinant enzyme experiments. Severallines of evidence in this study using marmoset liver microsomes withcorrelation or chemical inhibition studies and experiments withrecombinantly expressed P450 enzymes and FMO3 suggested thatthe MPTP N-demethylation, MPTP N-oxygenation, and MPDP+ andMPP+ formation reactions shown in Fig. 1 were mediated by multipledrug-metabolizing enzymes, such as CYP1A, CYP2C, CYP2D, andFMO in marmoset liver microsomes, but these reactions were mainlycatalyzed by marmoset CYP2D6 or FMO3. Kinetic analysis of MPP+

formation revealed that the apparent Km values for marmoset livermicrosomes and recombinant marmoset CYP2D6 were consistent(Table 3), although minor roles of other P450 isoforms could not beruled out. It should be noted that occasional species differences wereseen with regard to the roles of P450 enzymes between marmosets andhumans (Fig. 5); human CYP1A2 seemed to have high capacity forMPDP+ and MPP+ formation under these conditions. Althoughmonoamine oxidase exists in another cellar fraction that was not

TABLE 2

MPTP N-oxygenation catalyzed by marmoset and human FMO1 and FMO3

Marmoset and human recombinant FMOs (20 pmol/ml) were incubated with 200 mM MPTPat 37�C for 10 minutes in the presence of an NADPH-generating system. Formations of PTP,MPDP+, and MPP+ by FMO were below the detection limits (for MPDP+ and MPP+,,0.1 nmol/minper nmol FMO; for PTP, ,0.01 nmol/min per nmol FMO).

FMO MPTP N-Oxygenation

nmol/min per nmol FMO

MarmosetFMO1 2.9FMO3 14.5

HumanFMO1 4.2FMO3 78.0

Fig. 5. Formation of PTP (A and D), MPDP+ (B and E), and MPP+ (C and F) from MPTP mediated by marmoset (A–C) and human (D–F) recombinant P450 enzymes.Marmoset and human recombinant P450 enzymes (20 pmol/ml) were incubated with 50 mM MPTP at 37�C for 10 minutes in the presence of an NADPH-generating system.

740 Uehara et al.

at ASPE

T Journals on February 15, 2021

dmd.aspetjournals.org

Dow

nloaded from

tested in this study, these results collectively suggested that in themarmoset, CYP2D6 is mainly responsible for both deactivation ofMPTP to PTP and metabolic activation of MPTP to MPP+ to producethe preclinical marmoset model for Parkinson-like syndromes. In ourpreliminary study, we were unable to isolate CYP2D30 cDNA frommarmoset livers by RT-PCR, and we thus excluded CYP2D30 fromfurther analysis. This was consistent with a reported finding thatCYP2D30 was not isolated from another source of marmosets usingthe same CYP2D30 primer sets (Hichiya et al., 2004).Our findings should provide a foundation for understanding MPTP

metabolism and the successful production of preclinical marmosetparkinsonism models. Although some differences were seen in thisstudy in terms of major roles of drug-metabolizing enzymes involvedin the MPTP activation and deactivation pathways in humans andmarmosets, marmosets are again recognized as a good preclinicalmodel for Parkinson disease, as previously reported (Ando et al.,2012, 2014). A variety of genetic polymorphisms of CYP2D6 causinggene duplication, reduced function, or loss of function in humans arewell known (Kiyotani et al., 2010). Genetic polymorphisms ofCYP2D enzymes in cynomolgus and rhesus macaques (Uno et al.,2014) were also recently demonstrated. In this study, the interindi-vidual variations for MPTP oxidations in marmoset liver microsomes(Fig. 3) were not as high as expected, presumably because no geneticvariations for marmoset CYP2D6 were confirmed in our preliminarystudy using these 11 marmosets. To effectively produce the Parkinsonpreclinical animal model, it will be of interest to find more efficientindividual marmosets (i.e., rapid metabolizers of CYP2D6 substrates)in further studies.In conclusion, these results indicated the contributions of multiple

drug-metabolizing enzymes in MPTP oxidation, especially marmosetFMO3 in MPTP deactivation (N-oxygenation). In the marmoset,CYP2D6 was responsible for biotransformations including MPTPdeactivation (N-demethylation) and metabolic activation of MPTP (bydesaturation reactions), leading to active neurotoxic compound MPP+.

Authorship ContributionsParticipated in research design: Uehara, Uno, Yamazaki.Conducted experiments: Uehara, Uno, Murayama, Shimizu.Contributed new reagents or analytic tools: Inoue, Sasaki.Performed data analysis: Uehara, Uno, Yamazaki.Wrote or contributed to the writing of the manuscript: Uehara, Uno,

Yamazaki.

References

Abbott DH, Barnett DK, Colman RJ, Yamamoto ME, and Schultz-Darken NJ (2003) Aspects ofcommon marmoset basic biology and life history important for biomedical research. Comp Med53:339–350.

Ando K, Inoue T, and Itoh T (2014) L-DOPA-induced behavioral sensitization of motor activityin the MPTP-treated common marmoset as a Parkinson’s disease model. Pharmacol BiochemBehav 127:62–69.

Ando K, Obayashi S, Nagai Y, Oh-Nishi A, Minamimoto T, Higuchi M, Inoue T, Itoh T,and Suhara T (2012) PET analysis of dopaminergic neurodegeneration in relation to im-mobility in the MPTP-treated common marmoset, a model for Parkinson’s disease. PLoSOne 7:e46371.

Bajpai P, Sangar MC, Singh S, Tang W, Bansal S, Chowdhury G, Cheng Q, Fang JK, MartinMV, and Guengerich FP, et al. (2013) Metabolism of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine by mitochondrion-targeted cytochrome P450 2D6: implications in Parkin-son disease. J Biol Chem 288:4436–4451.

Ballard PA, Tetrud JW, and Langston JW (1985) Permanent human parkinsonism due to 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP): seven cases. Neurology 35:949–956.

Carrion R, Jr and Patterson JL (2012) An animal model that reflects human disease: the commonmarmoset (Callithrix jacchus). Curr Opin Virol 2:357–362.

Cashman JR and Ziegler DM (1986) Contribution of N-oxygenation to the metabolism of MPTP(1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine) by various liver preparations. Mol Pharmacol29:163–167.

Chiba K, Horii H, Kubota E, Ishizaki T, and Kato Y (1990) Effects of N-methylmercaptoimidazoleon the disposition of MPTP and its metabolites in mice. Eur J Pharmacol 180:59–67.

Chiba K, Peterson LA, Castagnoli KP, Trevor AJ, and Castagnoli N, Jr (1985) Studies on themolecular mechanism of bioactivation of the selective nigrostriatal toxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine. Drug Metab Dispos 13:342–347.

Coleman T, Ellis SW, Martin IJ, Lennard MS, and Tucker GT (1996) 1-Methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) is N-demethylated by cytochromes P450 2D6, 1A2 and 3A4—implications for susceptibility to Parkinson’s disease. J Pharmacol Exp Ther 277:685–690.

Davis KD, Taub E, Houle S, Lang AE, Dostrovsky JO, Tasker RR, and Lozano AM (1997)Globus pallidus stimulation activates the cortical motor system during alleviation of parkin-sonian symptoms. Nat Med 3:671–674.

Eslamboli A (2005) Marmoset monkey models of Parkinson’s disease: which model, when andwhy? Brain Res Bull 68:140–149.

TABLE 3

Kinetic parameters for MPTP oxidation by marmoset and human liver microsomes and recombinant CYP2D6 and FMO3

Kinetic parameters were calculated by nonlinear regression analysis (mean 6 S.E., 10 points of substrate concentrations within the range 1–2000 mM).

Kinetic ParameterMarmoset Human

Liver Microsomes CYP2D6 FMO3 Liver Microsomes CYP2D6 FMO3

PTP formationKm (mM) 451 6 99 71 6 7 N.A. 577 6 116 125 6 23 N.A.Vmax (nmol/min per mg) 0.43 6 0.03 — — 0.13 6 0.01 — —

Vmax (min21) — 15.6 6 0.4 ,0.01 — 11.5 6 0.6 ,0.01Vmax/Km (ml/min per mg) 0.95 — — 0.23 — —

Vmax/Km (ml/min per nmol) — 220 N.A. — 92 N.A.MPTP N-oxide formation

Km (mM) 162 6 21 N.A. 137 6 14 173 6 69 N.A. 102 6 19Vmax (nmol/min per mg) 0.52 6 0.02 — — 0.22 6 0.02 — —

Vmax (min21) — ,0.1 23 6 1 — ,0.1 104 6 5Vmax/Km (ml/min per mg) 3.2 — — 1.3 — —

Vmax/Km (ml/min per nmol) — N.A. 0.17 — N.A. 1.0MPDP+ formation

Km (mM) 125 6 19 74 6 23 N.A. 103 6 6 276 6 96 N.A.Vmax (nmol/min per mg) 0.23 6 0.01 — — 0.66 6 0.01 — —

Vmax (min21) — 5.5 6 0.4 ,0.1 — 1.0 6 0.1 ,0.1Vmax/Km (ml/min per mg) 1.8 — — 6.4 — —

Vmax/Km (ml/min per nmol) — 74 N.A. — 3.6 N.A.MPP+ formation

Km (mM) 66 6 18 72 6 18 N.A. 53 6 10 331 6 110 N.A.Vmax (nmol/min per mg) 0.23 6 0.02 — — 0.38 6 0.02 — —

Vmax (min21) — 11 6 1 ,0.1 — 2.8 6 0.3 ,0.1Vmax/Km (ml/min per mg) 3.5 — — 7.2 — —

Vmax/Km (ml/min per nmol) — 150 N.A. — 8.5 N.A.

N.A., not available; —, not calculated.

MPTP Activation by Marmoset CYP2D6 741

at ASPE

T Journals on February 15, 2021

dmd.aspetjournals.org

Dow

nloaded from

Herraiz T, Guillén H, and Galisteo J (2013) Metabolite profile resulting from the activation/inactivation of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine and 2-methyltetrahydro-b-carboline by oxidative enzymes. Biomed Res Int 2013:248608.

Hichiya H, Kuramoto S, Yamamoto S, Shinoda S, Hanioka N, Narimatsu S, Asaoka K, Miyata A,Iwata S, and Nomoto M, et al. (2004) Cloning and functional expression of a novel marmosetcytochrome P450 2D enzyme, CYP2D30: comparison with the known marmoset CYP2D19.Biochem Pharmacol 68:165–175.

Jenner P (2009) From the MPTP-treated primate to the treatment of motor complications inParkinson’s disease. Parkinsonism Relat Disord 15 (Suppl 4):S18–S23.

Johansson I and Ingelman-Sundberg M (2011) Genetic polymorphism and toxicology—withemphasis on cytochrome p450. Toxicol Sci 120:1–13.

Jones KC and Ballou DP (1986) Reactions of the 4a-hydroperoxide of liver microsomal flavin-containing monooxygenase with nucleophilic and electrophilic substrates. J Biol Chem 261:2553–2559.

Kiyotani K, Shimizu M, Kumai T, Kamataki T, Kobayashi S, and Yamazaki H (2010) Limitedeffects of frequent CYP2D6*36-*10 tandem duplication allele on in vivo dextromethorphanmetabolism in a Japanese population. Eur J Clin Pharmacol 66:1065–1068.

Langston JW and Ballard PA, Jr (1983) Parkinson’s disease in a chemist working with 1-methyl-4-phenyl-1,2,5,6-tetrahydropyridine. N Engl J Med 309:310.

Mansfield K (2003) Marmoset models commonly used in biomedical research. Comp Med 53:383–392.

Okano H, Hikishima K, Iriki A, and Sasaki E (2012) The common marmoset as a novel animalmodel system for biomedical and neuroscience research applications. Semin Fetal NeonatalMed 17:336–340.

Shimizu M, Iwano S, Uno Y, Uehara S, Inoue T, Murayama N, Onodera J, Sasaki E,and Yamazaki H (2014) Qualitative de novo analysis of full length cDNA and quantitativeanalysis of gene expression for common marmoset (Callithrix jacchus) transcriptomes usingparallel long-read technology and short-read sequencing. PLoS One 9:e100936.

Tokuno H, Moriya-Ito K, and Tanaka I (2012) Experimental techniques for neuroscience researchusing common marmosets. Exp Anim 61:389–397.

Uehara S, Murayama N, Nakanishi Y, Zeldin DC, Yamazaki H, and Uno Y (2011) Immuno-chemical detection of cytochrome P450 enzymes in liver microsomes of 27 cynomolgusmonkeys. J Pharmacol Exp Ther 339:654–661.

Uehara S, Murayama N, Yamazaki H, and Uno Y (2010) A novel CYP2A26 identified incynomolgus monkey liver metabolizes coumarin. Xenobiotica 40:621–629.

Uno Y, Iwasaki K, Yamazaki H, and Nelson DR (2011) Macaque cytochromes P450: nomenclature,transcript, gene, genomic structure, and function. Drug Metab Rev 43:346–361.

Uno Y, Shimizu M, and Yamazaki H (2013) Molecular and functional characterization of flavin-containing monooxygenases in cynomolgus macaque. Biochem Pharmacol 85:1837–1847.

Uno Y, Uehara S, Kohara S, Murayama N, and Yamazaki H (2010) Cynomolgus monkeyCYP2D44 newly identified in liver, metabolizes bufuralol, and dextromethorphan. Drug MetabDispos 38:1486–1492.

Uno Y, Uehara S, Kohara S, Murayama N, and Yamazaki H (2014) Polymorphisms of CYP2D17in cynomolgus and rhesus macaques: an evidence of the genetic basis for the variability ofCYP2D-dependent drug metabolism. Drug Metab Dispos 42:1407–1410.

Yamazaki H, Nakamura M, Komatsu T, Ohyama K, Hatanaka N, Asahi S, Shimada N,Guengerich FP, Shimada T, and Nakajima M, et al. (2002) Roles of NADPH-P450 reductaseand apo- and holo-cytochrome b5 on xenobiotic oxidations catalyzed by 12 recombinant humancytochrome P450s expressed in membranes of Escherichia coli. Protein Expr Purif 24:329–337.

Yamazaki H and Shimizu M (2013) Survey of variants of human flavin-containing mono-oxygenase 3 (FMO3) and their drug oxidation activities. Biochem Pharmacol 85:1588–1593.

Yamazaki M, Shimizu M, Uno Y, and Yamazaki H (2014) Drug oxygenation activities mediatedby liver microsomal flavin-containing monooxygenases 1 and 3 in humans, monkeys, rats, andminipigs. Biochem Pharmacol 90:159–165.

Yoshihara S, Harada K, and Ohta S (2000) Metabolism of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) in perfused rat liver: involvement of hepatic aldehyde oxidase asa detoxification enzyme. Drug Metab Dispos 28:538–543.

Address correspondence to: Hiroshi Yamazaki, Laboratory of Drug Metabo-lism and Pharmacokinetics, Showa Pharmaceutical University, 3-3165 Higashi-tamagawa Gakuen, Machida, Tokyo 194-8543, Japan. E-mail: hyamazak@ac.shoyaku.ac.jp

742 Uehara et al.

at ASPE

T Journals on February 15, 2021

dmd.aspetjournals.org

Dow

nloaded from