1521-009X/44/1/8–15$25.00 http://dx.doi.org/10.1124/dmd.115.067561DRUG METABOLISM AND DISPOSITION Drug Metab Dispos 44:8–15, January 2016Copyright ª 2015 by The American Society for Pharmacology and Experimental Therapeutics

Molecular Cloning, Tissue Distribution, and FunctionalCharacterization of Marmoset Cytochrome P450 1A1, 1A2, and 1B1

Shotaro Uehara,1 Yasuhiro Uno,1 Takashi Inoue, Erika Sasaki, and Hiroshi Yamazaki

Laboratory of Drug Metabolism and Pharmacokinetics, Showa Pharmaceutical University, Machida, Tokyo, Japan (S.U., H.Y.);Pharmacokinetics and Bioanalysis Center, Shin Nippon Biomedical Laboratories, Ltd., Kainan, Wakayama, Japan (Y.U.);

Department of Applied Developmental Biology (T.I.) and Center of Applied Developmental Biology (E.S.), Central Institute forExperimental Animals, Kawasaki, Japan; and Keio Advanced Research Center, Keio University, Minato-ku, Tokyo, Japan (E.S.)

Received October 1, 2015; accepted October 23, 2015

ABSTRACT

The commonmarmoset (Callithrix jacchus), a NewWorldmonkey, haspotential to be an animal model for drug metabolism studies. In thisstudy, we identified and characterized cytochrome P450 (P450) 1A1and 1B1 in addition to the known P450 1A2 in marmosets. MarmosetP450 1A1 and 1B1 cDNA contained open reading frames encoding512 and 543 amino acids, respectively, with high sequence identities(90%–93%) to other primate P450 1A1s and 1B1s. A phylogenetic treebased on amino acid sequences showed close evolutionary relation-ships among marmoset, macaque, and human P450 1A and 1Benzymes. By mRNA quantification and immunoblot analyses in fivemarmoset tissues, P450 1A1 was mainly expressed in lungs andsmall intestines, and P450 1A2 was expressed predominantly inlivers. In contrast, P450 1B1 was expressed in all tissues tested.Marmoset P450 1A1, 1A2, and 1B1 heterologously expressed

in Escherichia coli catalyzed 7-ethoxyresorufin O-deethylation,7-ethoxycoumarinO-deethylation, and phenacetinO-deethylation, sim-ilar to those of humans and cynomolgus monkeys. Notably, marmosetP450 1A1 and 1A2 more efficiently catalyzed 7-ethoxyresorufinO-deethylation than thoseof thehumanhomologs,butwerecomparableto those of the cynomolgus monkey homologs. Additionally, marmosetP450 1B1 preferentially catalyzed estradiol 4-hydroxylation; however, ratP450 1B1 more favorably catalyzed estradiol 2-hydroxylation, indicatingthat the estradiol hydroxylation specificity of marmoset P450 1B1 wassimilar to those of human and cynomolgus monkey P450 1B1. Theseresults indicated thatmarmoset P450 1Aand1Benzymeshad functionalcharacteristics similar to those of humans and cynomolgus monkeys,suggesting that P450 1A and 1B–dependent metabolism was similaramong marmosets, cynomolgus monkeys, and humans.

Introduction

Non-human primates, especially Old World monkeys, such ascynomolgus monkeys (Macaca fascicularis) and rhesus monkeys(Macaca mulatta) have been used to estimate the drug efficacy andtoxicity due to their evolutionary closeness to humans. Commonmarmosets (Callithrix jacchus), belonging to NewWorld monkeys, areattractive as small non-human primate models in drug metabolismstudies due to genetic closeness, small body size, early sexual maturity,and high reproductive efficacy (Orsi et al., 2011; Sasaki, 2015).Human cytochrome P450 (P450) genes, encoding typical drug

metabolizing enzymes, are composed of 57 functional genes and 58pseudogenes. The P450 enzymes, especially members of the 1, 2, and3 families, play important roles in the metabolism of xenobiotics,including various drugs and some endogenous compounds. HumanP450 1 enzymes include P450 1A1, 1A2, and 1B1. Human P450 1A1mRNA is expressed in many tissues such as lung and trachea (Biècheet al., 2007). In contrast, human P450 1A2 mRNA is predominantly

expressed in liver. Human P450 1A and 1B enzymes are involved in7-ethoxyresorufin O-deethylation, 7-ethoxycoumarin O-deethylation,benzo(a)pyrene hydroxylation, and estradiol hydroxylation (Shimadaet al., 1997; Lee et al., 2003). In marmosets, the only P450 1 enzymeidentified to date is P450 1A2, which catalyzes metabolic activa-tion of 2-amino-3-methyl-imidazo[4,5-f]quinolone and 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline, 7-ethoxyresorufin O-deethylation,and phenacetin O-deethylation (Sakuma et al., 1997; Narimatsu et al.,2005).Recently, Shimizu et al. (2014) and Uehara et al. (2015a,b,c,d)

identified marmoset P450 2A, 2C, 2D, and 3A genes and character-ized enzymatic properties of these proteins. Most marmoset P450shave high sequence identities (.85%) to the homologous humanP450s at the amino acid level. In particular, marmoset P450 3A, 2D,and 2C enzymes metabolize their respective typical human P450probe substrates, but substrate selectivity of marmoset P450 2Aenzyme partly differs from that of the human homologous enzyme.The molecular analysis of marmoset P450 enzymes is useful inunderstanding drug metabolism in marmosets; however, marmosetP450 1A and 1B enzymes have not been identified and fullycharacterized. In this study, we found marmoset P450 1A1 and 1B1genes by analysis of the marmoset genome sequence and we report theresults of cDNA analysis, tissue distribution at mRNA and proteinlevels, and enzymatic properties of marmoset P450 1A1, 1A2,and 1B1.

This work resulted from “Construction of System for Spread of Primate ModelAnimals” under the Strategic Research Program for Brain Sciences of the JapanAgency for Medical Research and Development. S.U. was also supported partlyby the Japan Society for the Promotion of Science Grant-in-Aid for YoungScientists B [15K18934].

1S.U. and Y.U. contributed equally to this work.dx.doi.org/10.1124/dmd.115.067561.

ABBREVIATIONS: P450, cytochrome P450; PCR, polymerase chain reaction; RT, reverse transcription; TCDD, 2,3,7,8-tetrachlorodibenzo-p-dioxin.

8

at ASPE

T Journals on June 30, 2020

dmd.aspetjournals.org

Dow

nloaded from

Materials and Methods

Chemicals and Enzymes. 7-Ethoxyresorufin, resorufin, 7-ethoxycoumarin,2-hydroxyestradiol, and 4-hydroxyestradiol were purchased from Sigma-Aldrich(Tokyo). Phenacetin, acetaminophen, 7-hydroxycoumarin, and b-estradiol werepurchased from Wako Pure Chemicals (Osaka, Japan). Marmoset (pooled from10 subjects), cynomolgus monkey (pooled from 10 subjects), and human livermicrosomes (pooled from 150 subjects) were purchased from Corning LifeSciences (Woburn, MA). Brain, lung, liver, kidney, and small intestine tissuesamples from eight marmosets (four males and four females, 2–6 years of age) wereobtained from the Central Institution for Experimental Animals (Kawasaki, Japan).This study was approved by the Institutional Animal Care and Use Committee.Pooled marmoset tissue microsomes were prepared as described previously(Uehara et al., 2015d). Recombinant rat P450 1B1 was purchased from Nosan(Yokohama, Japan). Anti-human P450 1A1 (H-70), 1A2 (D-3), and 1B1 (H-105)antibodies, and anti-human protein disulfide isomerase antibodies (H-160) werepurchased from Santa Cruz Biotechnology (Santa Cruz, CA). All other chemicalsused were the best grade available commercially.

P450 cDNA Cloning. Total RNA was extracted from marmoset andcynomolgus monkey livers using the RNeasy Mini Kit (Qiagen, Valencia,CA), and the first-strand cDNA was synthesized using SuperScript III reversetranscription (RT) reverse transcriptase (Invitrogen, Carlsbad, CA) and oligo(dT)primers (Invitrogen). The cDNAs, encodingmarmoset P450 1A1, 1A2, and 1B1,and cynomolgus monkey P450 1B1, were amplified from the first-strand cDNAtemplates by polymerase chain reaction (PCR) with KOD-Plus-Neo DNA

polymerase (Toyobo, Osaka, Japan) in an ABI GeneAmp PCR System 9700thermocycler (Applied Biosystems, Foster City, CA). The PCR primers used weredesigned based on the genome sequences of marmoset and cynomolgus monkey,including cjCYP1A1 (5rt1) 59-CCAAGATCCCTACACTGATCATGCTT-39 andcjCYP1A1 (3rt1) 59-CTAAGAGCGCAGCTGCATTTGGAA-39 for marmosetP450 1A1, cjCYP1A2 (5rt1) 59-CAGATGGCATTGTCCCAGTTTGTTC-39and cjCYP1A2 (3rt1) 59-CGGTGTCTTCCTCACTGGAAGGAG-39 for mar-moset P450 1A2, cjCYP1B1(5rt2) 59-GACAACGCCAAGAGACTCAAG-39and cjCYP1B1(3rt2) 59-GAATTTTACTCCTCATCTCCCAAG-39 for marmo-set P450 1B1, and mfCYP1B1(5rt1) 59-GCAAAGGTCCCAGTTCCTTC-39and mfCYP1B1(3rt1) 59-CACACCTCACCTGACGAACA-39 for cynomolgusmonkey P450 1B1. The PCR products were cloned into pCR4 vectors using aZero Blunt TOPO PCR Cloning Kit (Invitrogen) or pGEM-T easy vectors usinga pGEM-T Easy Vector System (Promega, Madison, WI), and insert sequenceswere verified by DNA sequencing using an ABI PRISM BigDye Terminatorv3.0 Ready Reaction Cycle Sequencing Kit (Applied Biosystems) with an ABIPRISM 3730 DNA Analyzer (Applied Biosystems).

mRNA Quantification. Using the total RNA extracted from brain, lung,liver, kidney, and small intestine as described previously, the first-strand cDNAwas synthesized with random primers (Invitrogen) and SuperScript III RTreverse transcriptase (Invitrogen). Real-time PCR was performed with an ABIPRISM 7500 sequence detection system (Applied Biosystems) using PowerSYBRGreen PCRMaster Mix (Applied Biosystems). The primers used for real-time PCRwere cjCYP1A1 (5qrt1) 59-GCCCAATTCTTCCTTGAACAC-39 andcjCYP1A1 (3qrt1) 59-ATGTGCCCCTTCTCAAAGGTT-39 for marmoset P450

Fig. 1. Multiple amino acid sequence alignment of marmoset P450 1A enzymes. Marmoset (cj) P450 1A1 and 1A2 amino acid sequences were aligned with those of thecynomolgus monkey (mf) and human (h) homologs. The broken and solid lines represent the putative heme-binding region and substrate recognition site (SRS), respectively.Identical amino acids are indicated by asterisks.

Characterization of Marmoset P450 1A1, 1A2, and 1B1 9

at ASPE

T Journals on June 30, 2020

dmd.aspetjournals.org

Dow

nloaded from

1A1, cjCYP1A2 (5qrt1) 59-CACTATCAGGACTCTGACAAGAACAG-39 andcjCYP1A2 (3qrt1) 59-CCCTGCTCCAAAGATGTCGT-39 for marmosetP450 1A2, and cjCYP1B1 (5ex1a) 59-TGGATTTGGCGGACGTACC-39 andcjCYP1B1 (3ex2a) 59-ACATAGTGCAGGTTGGGCTG-39 for marmoset P4501B1. The real-time PCR conditions were as follows: 95�C for 10 minutes,followed by 40 cycles of 95�C for 15 seconds and 60�C for 1 minute. All PCRreactions were carried out in triplicate, and the expression level of each P450mRNA was normalized to the expression of 18S rRNA.

Immunoblotting. The protein expressions of P450 1A1, 1A2, and 1B1, andprotein disulfide isomerase in five marmoset tissues were verified by immuno-blotting using anti-human P450 1A1, 1A2, and 1B1 antibodies, and anti-proteindisulfide isomerase antibodies, respectively. Pooled tissue microsomes (40 mg)or recombinant P450 proteins (1 pmol Eq) were loaded on a 10% sodium dodecylsulfate-polyacrylamide gel, separated by electrophoresis, and transferred to apolyvinylidene difluoride membrane (Merck Millipore, Billerica, MA). Themembranes were blocked with Tris-buffered saline containing 0.05% Tween 20and 0.5% nonfat dry milk for 30 minutes at room temperature, and wereincubated with anti-human P450 1A1 (1:5000), anti-human P450 1A2 antibodies(1:5000), or anti-human P450 1B1 antibodies (1:1000) at room temperature for1 hour. Then, the membranes were washed three times for 10 minutes andincubated with anti-rabbit and mouse IgG secondary antibody (Santa CruzBiotechnology; 1:5000) at room temperature for 20 minutes. Protein bands weredetected using the ECL Prime Western Blotting Detection System (GE Health-care, Buckinghamshire, United Kingdom) according to the manufacturer’sinstructions.

Heterologous Protein Expression in Escherichia coli. Recombinantmarmoset P450 1A1, 1A2, and 1B1, and cynomolgus monkey P450 1B1 wereprepared in E. coli using pCW vector (coexpressed with NADPH-P450reductase) as described previously (Uehara et al., 2015d). The cDNAs weremodified N terminus by PCR amplification using primers containing therestriction sites of the NdeI and XbaI sites (underlined). The PCR primerswere cjCYP1A1 (5exp1bov) 59- GGAATTCCATATGGCTCTGTTATTAG-CAGTTTTTGTCACGGAGCTTCTCCTGGCCTCT-39 and cjCYP1A1 (3exp1)59-GCTCTAGACTAAGAGCGCAGCTGCATTTGGAA-39 for marmoset P4501A1, cjCYP1A2 (5exp1bov) 59-GGAATTCCATATGGCTCTGTTATTAG-CAGTTTTTTTCTCAGCCACAGAGCTTCTCCT-39 and cjCYP1A2 (3exp1)59-GCTCTAGACGGTGTCTTCCTCACTGGAAGG-39 for marmoset P4501A2, cjCYP1B1 (5exp1bov) 59- CATATGGCTCTGTTATTAGCAGTTTTT-GACCCTTGGCCGCTAAAC-39 and cjCYP1B1 (3exp1) 59- TCTAGAT-TATTGGCAAATTTCCTCGGCTTG-39 for marmoset P450 1B1, andmfCYP1B1 (5exp1bov) 59- CATATGGCTCTGTTATTAGCAGTTTTT-GACCCTTGGCCTCTAAAC-39 and mfCYP1B1 (3exp1) 59- TCTAGAT-TATTGGCAAGTTTCCTCAG-39 for cynomolgus monkey P450 1B1. Formarmoset P450 1A1 and 1A2, PCR products were cloned into pCWvectors afterrestriction enzyme reaction. For marmoset and cynomolgus monkey P450 1B1,PCR products were first cloned into pGEM-T easy vectors, and then subclonedinto pCW vectors after restriction enzyme reaction. Insert sequences wereverified by DNA sequencing as described previously. Expression plasmids ofhuman P450 1A1, 1A2, and 1B1, and cynomolgus monkey P450 1A1 and 1A2proteins with NADPH-P450 reductase were prepared as described previously

Fig. 2. Multiple amino acid sequence alignment of marmoset P450 1B enzyme. Marmoset (cj) P450 1B1 amino acid sequences were aligned with those of the cynomolgusmonkey (mf) and human (h) homolog. The broken and solid lines represent the putative heme-binding region and substrate recognition site (SRS), respectively. Identicalamino acids are indicated by asterisks.

10 Uehara et al.

at ASPE

T Journals on June 30, 2020

dmd.aspetjournals.org

Dow

nloaded from

(Yamazaki et al., 2002; Uno et al., 2011). Expression of these P450 proteins wascarried out as described previously (Yamazaki et al., 2002; Uno et al., 2011). Themembrane fractions from E. coli DH5a expressing P450 proteins were preparedby sonication and centrifugation, and the concentrations of P450 protein andNADPH-P450 reductase were measured as described previously (Yamazakiet al., 2002).

Enzymatic Characterization and Kinetic Analysis. 7-EthoxyresorufinO-deethylation, 7-ethoxycoumarin O-deethylation, and phenacetin O-deethylationactivity of recombinant proteins and livermicrosomeswere determined for enzymaticcharacterization as described previously (Uehara et al., 2015c,d). For thekinetic analysis of 7-ethoxyresorufin O-deethylation, an incubation mixturecontaining 1.0 pmol Eq of recombinant P450 protein or 0.04 mg livermicrosomes, an NADPH-generating system (0.25 mMNADP+, 2.5 mM glucose6-phosphate, and 0.25 units/ml glucose 6-phosphate dehydrogenase), and 0.05–10 mM 7-ethoxyresorufin in 100 mM potassium phosphate buffer (pH7.4) in afinal volume of 0.25 ml were incubated at 37�C for 10 minutes. For estradiol2- and 4-hydroxylation, the incubation mixture contained 10 pmol Eq ofrecombinant P450 protein or 0.5 mg liver microsomes, an NADPH-generatingsystem, and 100mMestradiol in 100mMpotassium phosphate buffer (pH 7.4) ina final volume of 0.25ml. After 3-minute preincubation at 37�C, the reaction wasinitiated by adding NADPH and incubated at 37�C for 30 minutes. Linearity formetabolite formations with incubation time and protein concentrations wasconfirmed under these conditions. After the reactions were terminated with ethylacetate and centrifuged at 2500g for 5 minutes the supernatants were evaporatedin a rotary evaporator under vacuum and the dry residues were dissolved inthe high-performance liquid chromatography mobile phase. The samples weresubjected to high-performance liquid chromatography in an analytical C18column (Mightysil, 5 mm, 150 � 4.6 mm; Kanto Chemical, Tokyo, Japan) withelution by 33% acetonitrile/0.1% acetic acid, and the metabolites were detectedwith a UV detector at 280 nm. Kinetic parameters were estimated from the fittingto the Michaelis-Menten equation using the Kaleidagraph program (SynergySoftware, Reading, PA).

Sequence Analysis. The homology of amino acid sequences was analyzedusing BLAST (National Center for Biotechnology Information Bethesda, MD).The multiple sequence alignment was performed with the Genetyx system(Software Development, Tokyo, Japan). The phylogenetic analysis wasconducted by the neighbor-joining method using DNASIS Pro (HitachiSoftware, Tokyo, Japan). The P450 amino acid sequences used were obtainedfrom GenBank; marmoset P450 1A2 (NP_001191363); human P450 1A1(NP_000490), 1A2 (NP_000752), 1B1 (NP_000095), and 2A6 (NM_000762);cynomolgus monkey P450 1A1 (BAA04500) and 1A2 (BAA33789); rhesusmonkey P450 1A1 (AAT49262) and 1B1 (NP_001253797); Japanese monkey

P450 1A2 (BAE16271); dog P450 1A1 (P56590), 1A2 (NP_001008720), and1B1 (NP_001153156); pig P450 1A1 (NP_999577) and 1A2 (NP_001153086);guinea pig P450 1A1 (NP_001166411) and 1A2 (NP_001166165); rabbit P4501A1 (NP_001164543) and 1A2 (NP_001164592); rat P450 1A1 (NP_036672),1A2 (NP_036673), and 1B1 (NP_037072); and mouse P450 1A1 (NP_034122),1A2 (NP_034123), and 1B1 (NP_034124). The amino acid sequences ofmarmoset P450 1A1 and 1B1 and cynomolgus monkey P450 1B1 were deducedfrom the cDNAs identified in this study.

Results

cDNA Cloning and Sequence Analysis of Marmoset P450 1A1,1A2, and 1B1. To identify P450 1A1 and 1B1 in marmosets, we firstsearched against the marmoset genome database using the human P4501A1 and 1B1 cDNA sequences, which resulted in the discovery ofunpublished marmoset P450 1A1 and 1B1 gene sequences. We clonedthe coding sequence of marmoset P450 1A1 (1536 base pairs) and 1B1(1629 base pairs) cDNAs from the marmoset liver by RT-PCR withgene-specific primers. Analysis of the deduced amino acids of P4501A1 (512 residues) and 1B1 (543 residues) showed that the primarystructure of these P450s had well-conserved sequence regions includingsix substrate recognition sites and the heme-binding region (Figs. 1and 2). The amino acid sequences of marmoset P450 1A1 and 1B1possessed 90% and 92% to 93% identities with the homologous humanand cynomolgus and rhesus monkey P450 enzymes, and the sequence

TABLE 1

Similarity of the deduced amino acid sequences of marmoset P450 1A1 and 1A2 with other P450 1A enzymes

Species P450 FormIdentity

Accession NumberMarmoset P450 1A1 Marmoset P450 1A2

% %

Human 1A1 90 70 NP_0004901A2 74 88 NP_000752

Cynomolgus monkey 1A1 90 71 BAA045001A2 75 91 BAA33789

Rhesus monkey 1A1 90 71 NP_001035328Dog 1A1 82 72 P56590

1A2 70 80 NP_001008720Pig 1A1 82 71 NP_999577

1A2 71 80 NP_001153086Rabbit 1A1 76 70 NP_001164543

1A2 71 78 NP_001164592Guinea pig 1A1 77 69 NP_001166411

1A2 68 77 NP_001166165Rat 1A1 79 68 NP_036672

1A2 68 74 NP_036673Mouse 1A1 79 68 NP_034122

1A2 67 72 NP_034123Marmoset 1A2 74 — NP_001191363

—, 100

TABLE 2

Similarity of the deduced amino acid sequences of marmoset P450 1B1 with otherP450 1B1 enzymes

Species Identity Accession Number

%

Human 92 NP_000095Cynomolgus monkey 93 —

a

Rhesus monkey 93 NP_001253797Dog 86 NP_001153156Rat 81 NP_037072Mouse 82 NP_034124

aDeduced amino acid sequences determined by this study were used.

Characterization of Marmoset P450 1A1, 1A2, and 1B1 11

at ASPE

T Journals on June 30, 2020

dmd.aspetjournals.org

Dow

nloaded from

identities were higher than those of other species (Tables 1 and 2). Inthis study, previously reported Marmoset P450 1A2 cDNA was alsoisolated to compare the enzymatic function with marmoset P450 1A1and 1B1. A phylogenetic comparison of P450 1A and 1B enzymes ofvarious animal species (Fig. 3) demonstrated that marmoset P450 1Aand 1B enzymes had evolutionary closeness with human P450 1A and1B enzymes, respectively, similar to cynomolgus monkey and rhesusmonkey P450 1A and 1B enzymes, but unlike dog, rabbit, rat, or mouseP450 1A and 1B enzymes.Tissue Expression Pattern of Marmoset P450 1A1, 1A2, and

1B1 mRNA. To determine the distribution of P450 1A1, 1A2, and 1B1mRNA in marmoset tissues, real-time PCR was carried out in brain,kidney, liver, lung, and small intestine (Fig. 4). Marmoset P450 1A1mRNA was mainly expressed in small intestine, lung, and brain.Marmoset 1B1mRNAwas expressed in most of the extrahepatic tissues(kidney. lung. brain. small intestine. liver). Marmoset P450 1A2mRNA expression was most abundant in liver. Next, P450 1A1, 1A2,and 1B1 protein expressions were detected in five marmoset tissues byimmunoblotting using anti-human P450 1A1, 1A2, and 1B1 antibodies.Recombinant marmoset P450 1A1, 1A2, and 1B1 were selectivelydetected with anti-human P450 1A1, 1A2, and 1B1 antibodies,respectively (Fig. 5A). Marmoset P450 1A1 protein was expressed inlung and small intestine, and at much lower levels in liver and kidney(Fig. 5B). Marmoset P450 1B1 protein was extensively expressed inlung, liver, kidney, and small intestine, and at a much lower level inbrain (Fig. 5B). An approximately 54-kDa band was detected in livermicrosomes by immunoblotting using anti-human P450 1A2 antibodies(Fig. 5B), indicating that marmoset P450 1A2 protein was predominantlyexpressed in liver. These results suggest a predominant expression ofmarmoset P450 1A2 protein in liver and more abundant expression ofmarmoset P450 1A1 and 1B1 proteins in the extrahepatic tissues.Enzymatic Properties of Marmoset P450 1A1, 1A2, and 1B1. To

determine the catalytic properties of marmoset P450 1A1, 1A2, and

1B1, metabolic activities were measured using the typical humanP450 1A and 1B substrates, 7-ethoxyresorufin, 7-ethoxycoumarin, andphenacetin for recombinant P450 1A1, 1A2, and 1B1 proteins, andmarmoset, cynomolgus monkey, and human liver microsomes. Mar-moset P450 1A1, 1A2, and 1B1 enzymes, and marmoset liver micro-somes catalyzed 7-ethoxyresorufin O-deethylation, 7-ethoxycoumarinO-deethylation, and phenacetin O-deethylation, similar to those ofhumans and cynomolgus monkeys (Table 3), indicating that the substrateselectivity of P450 1A and 1B enzymes was apparently conserved amongmarmosets, cynomolgus monkeys, and humans. Kinetic analysis for7-ethoxyresorufinO-deethylation showed that marmoset liver microsomeshad higher a Vmax value over theKm value (2.5ml/min/mg) compared with

Fig. 3. Phylogenetic analysis of amino acid sequences from marmosetP450 1A and 1B enzymes. The phylogenetic tree was constructed withthe neighbor-joining method using P450 1A1, 1A2, and 1B1 aminoacid sequences of marmoset (cj), human (h), cynomolgus monkey(mf), rhesus monkey (mm), Japanese monkey (mfu), dog (d), pig (p),guinea pig (gp), rabbit (rab), rat (r), and mouse (m). Human P450 2A6was used as an out group. The scale bar indicates 10% replacementof an amino acid per site.

Fig. 4. P450 1A1, 1A2, and 1B1 mRNA expression in marmoset tissues.Expression of marmoset P450 1A1, 1A2, and 1B1 mRNAs was measured in brain,lung, liver, kidney, and small intestine by real-time RT-PCR. The P450 mRNAexpression level was normalized to the 18S rRNA level. Each value represents themean 6 S.D. of three independent experiments.

12 Uehara et al.

at ASPE

T Journals on June 30, 2020

dmd.aspetjournals.org

Dow

nloaded from

those of humans and cynomolgus monkeys (Table 4). Similarly,marmoset P450 1A1 and 1A2 enzymes more efficiently catalyzed7-ethoxyresorufin O-deethylation compared with those of humans andcynomolgus monkeys. These findings indicated that marmoset P4501A and 1B enzymes catalyzed typical human P450 probes.A previous study showed that the estradiol hydroxylation specificity

of P450 1B1 enzyme differed between those of humans and rats

(Nishida et al., 2013). To determine the estradiol hydroxylationspecificity of marmoset P450 1B1, estradiol 2- and 4-hydroxylationactivities were compared with rat, cynomolgus monkey, and humanP450 1B1. Marmoset P450 1B1 preferentially catalyzed estradiol4-hydroxylation, similar to those of humans and cynomolgus monkeys,but rat P450 1B1 more favorably catalyzed estradiol 2-hydroxylation(Table 5). Therefore, estradiol hydroxylation specificity of marmosetP450 1B1 enzyme was similar to those of human and cynomolgusmonkey P450 1B1.

Discussion

Common marmosets have been used in preclinical studies for drugdevelopment due to various advantages as an experimental animal.However, P450s, which are major drugmetabolizing enzymes, have notbeen fully identified and analyzed in marmosets. In a previous study,the tissue-specific expression patterns of P450 1-3–like genes wereshown (Shimizu et al., 2014). In this study, we newly cloned P450 1A1and 1B1 cDNAs from marmoset livers by RT-PCR using the gene-specific primers designed based on the marmoset genome sequences.Marmoset P450 1A1, 1A2, and 1B1 had high sequence identities(.88%) to the homologous human P450 at the amino acid level.Physiologic analysis also indicated the evolutionary closeness ofmarmoset P450 1A and 1B to those of humans, cynomolgus monkeys,and rhesus monkeys (Fig. 3).Marmoset P450 1A1, 1A2, and 1B1 catalyzed 7-ethoxyresorufin

O-deethylation, 7-ethoxycoumarin O-deethylation, and phenacetinO-deethylation (Table 3), which are mediated by P450 1As in humans.Among the P450 1A and 1B enzymes in marmosets, cynomolgusmonkeys, and humans, marmoset P450 1A1 most efficiently catalyzed7-ethoxyresorufin O-deethylation. Marmoset P450 1A2 had a higherVmax value over the Km value (21 ml/min/nmol) than that of humanP450 1A2 (9.1 ml/min/nmol), which might account for the higher Vmax

value over the Km value (2.5 ml/min/mg) for 7-ethoxyresorufinO-deethylation in marmoset liver microsomes than that of human livermicrosomes (0.070ml/min/mg) (Table 4). Marmoset P450 1B1 catalyzed7-ethoxyresorufin O-deethylation, 7-ethoxycoumarin O-deethylation,and phenacetin O-deethylation at a much lower rate than human andcynomolgus monkey P450 1B1, suggesting that the contribution ofmarmoset P450 1B1 to overall hepatic drug metabolism is very limited.Interestingly, among primates relatively close to cynomolgus or rhesusmonkeys, marmoset P450 1A2 has metabolic properties different fromJapanese monkey P450 1A2, which has a high sequence identity (99%)to cynomolgus monkey P450 1A2 (Narimatsu et al., 2005). Marmoset

Fig. 5. Immunoblot analysis of P450 1A1, 1A2, and 1B1 proteins in marmosettissues. Immunoblotting analysis was carried out using anti-human P450 1A1, 1A2,and 1B1 antibodies to detect P450 1A1, 1A2, and 1B1 proteins in marmoset tissues,respectively. (A) The cross-reactivity of anti-human P450 antibodies was verifiedusing recombinant marmoset P450 (1 pmol of P450 per lane). (B) Marmoset P4501A1, 1A2, and 1B1 proteins were detected in marmoset tissues (40 mg microsomalproteins per lane). Protein disulfide isomerase was used as a loading control.

TABLE 3

Catalytic activities by recombinant P450 1A1, 1A2, and 1B1 enzymes and liver microsomes from marmosets, cynomolgus monkeys, and humans

Activities were measured at substrate concentrations of 2.0, 100, and 100 mM for 7-ethoxyresorufin O-deethylation, 7-ethoxycoumarin O-deethylation, andphenacetin O-deethylation, respectively. Each value represents the mean from two independent analyses in duplicate determinations.

P450 Form Ethoxyresorufin O-Deethylation 7-Ethoxycoumarin O-Deethylation Phenacetin O-Deethylation

min21 nmol/min/mg min21 nmol/min/mg min21 nmol/min/mg

Marmoset P450 1A1 38.5 54.1 24.3Marmoset P450 1A2 14.4 71.0 25.9Marmoset P450 1B1 0.3 1.0 0.2Cynomolgus monkey P450 1A1 4.5 10.9 3.7Cynomolgus monkey P450 1A2 4.4 18.0 5.7Cynomolgus monkey P450 1B1 7.1 5.9 1.9Human P450 1A1 23.5 87.8 8.2Human P450 1A2 4.4 8.3 13.9Human P450 1B1 7.1 2.3 0.3Marmoset liver microsomes 0.31 0.23 0.54Monkey liver microsomes 0.29 0.33 1.34Human liver microsomes 0.07 0.29 0.54

Characterization of Marmoset P450 1A1, 1A2, and 1B1 13

at ASPE

T Journals on June 30, 2020

dmd.aspetjournals.org

Dow

nloaded from

P450 1A2 had a high Km value in 7-ethoxyresorufin O-deethylationand a low Km value in phenacetin O-deethylation, whereas Japanesemonkey P450 1A2 has a high Km value in both. These results suggestthat marmoset P450 1A enzymes have enzyme characteristics similarto human and cynomolgus monkey P450 1As with some functionaldifferences between primate species, despite their high sequenceidentities. Therefore, further functional studies of P450 1A and 1Benzymes between primate species would be interesting.Marmoset P450 1A1 mRNA and protein were mainly expressed in

lung and small intestine among the five tissues analyzed (Figs. 4 and 5).In humans, P450 1A1 mRNA is mainly expressed in the extrahepatictissues such as lung, trachea, and small intestine (Bièche et al., 2007).Marmoset P450 1B1 mRNA was mainly expressed in most of theextrahepatic tissues analyzed, including kidney and lung, and at a muchlower level in liver. Similarly, P450 1B1 protein was also expressed inall these five tissues (Fig. 5). In humans, P450 1B1 mRNA is expressedin the extrahepatic tissues, particularly in prostate, uterus, trachea, lung,and kidney (Bièche et al., 2007). A previous report showed an abundantexpression of marmoset P450 1A2 mRNA in liver (Sakuma et al.,1997), similar to human P450 1A2 mRNA (Bièche et al., 2007). P4501A2 is expressed much more abundantly than P450 1A1 in marmosetliver (Sakuma et al., 1997), similar to that of humans (Shimada et al.,1994). In contrast, cynomolgus monkey P450 1A1 mRNA is expressed

in liver much more abundantly than P450 1A2 mRNA (Sakuma et al.,1998), indicating that marmosets and humans might share themechanisms of the P450 1A basal transcriptional regulation in liver,unlike cynomolgus monkeys. These findings indicated that tissuedistribution of marmoset P450 1A and 1Bwas similar to those of humanP450 1A and 1B.The P450 1A gene family is transcriptionally induced by a polycyclic

aromatic hydrocarbon such as 2,3,7,8-tetrachlorodibenzo-p-dioxin(TCDD) and 3-methylcholanthren through the aryl hydrocarbon re-ceptor (Nebert and Dalton, 2006). P450 1A1, expressed in human liverat quite a low level, is inducible upon exposure to TCDD (Drahushuket al., 1998). Marmoset P450 1A1 and 1A2–like protein expressions inliver are greatly induced by TCDD, but not by phenobarbitone,rifampicin, or clofibric acid (Schulz et al., 2001). In human andcynomolgus monkey livers, P450 1A1 is greatly induced by TCDD,but P450 1A2 shows only modest induction (Edwards et al., 1994;Drahushuk et al., 1996). Additionally, P450 1A2 mRNA is induced by3-methylcholanthrene in marmosets, cynomolgus monkeys, and humans(Sakuma et al., 1997, 1998; Rhodes et al., 2011). These findings suggestthat P450 1A genes might share the mechanisms of the transcriptionalactivation in marmosets, cynomolgus monkeys, and humans.Human P450 1A1, 1A2, and 1B1 enzymes predominantly catalyze

estradiol 2- and 4-hydroxylation, but with different metabolite ratios

TABLE 4

Kinetic parameters of 7-ethoxyresorufin O-deethylation by recombinant P450 1A and 1B enzymes, and liver microsomesfrom marmosets, cynomolgus monkeys, and humans

Kinetic parameters were calculated by nonlinear regression analysis (mean 6 S.E.).

Enzyme Source Km Vmax Vmax Vmax/Km Vmax/Km

mM min21 pmol/min/mg ml/min/nmol ml/min/mg

Marmoset P450 1A1 1.0 6 0.3 47 6 5 47Marmoset P450 1A2 0.78 6 0.16 16 6 1 21Marmoset P450 1B1 1.0 6 0.2 0.69 6 0.05 0.69Cynomolgus monkey P450 1A1 0.13 6 0.03 5.9 6 0.3 45Cynomolgus monkey P450 1A2 0.42 6 0.04 5.3 6 0.2 13Cynomolgus monkey P450 1B1 0.29 6 0.07 7.3 6 0.5 25Human P450 1A1 1.2 6 0.2 33 6 2 28Human P450 1A2 0.56 6 0.25 5.1 6 0.7 9.1Human P450 1B1 0.31 6 0.06 2.8 6 0.1 9.0Marmoset liver microsomes 0.11 6 0.01 0.27 6 0.01 2.5Monkey liver microsomes 0.38 6 0.07 0.35 6 0.02 0.92Human liver microsomes 1.2 6 0.2 0.08 6 0.01 0.070

TABLE 5

Estradiol 4- and 2-hydroxylation activities by recombinant P450 1A and 1B enzymes, and liver microsomes frommarmosets, cynomolgus monkeys, humans, and rats

Activity was measured at a substrate concentration of 100 mM. Each value represents the mean of triplicate determinations.

P450 Form Estradiol 4-Hydroxylation Estradiol 2-Hydroxylation

pmol/min/nmol P450 pmol/min/mg pmol/min/nmol P450 pmol/min/mg

Marmoset P450 1A1 54 6 10 244 6 13Marmoset P450 1A2 79 6 5 501 6 34Marmoset P450 1B1 81 6 24 24 6 9Cynomolgus monkey P450 1A1 55 6 3 82 6 1Cynomolgus monkey P450 1A2 60 6 2 221 6 5Cynomolgus monkey P450 1B1 141 6 17 101 6 3Human P450 1A1 57 6 1 85 6 24Human P450 1A2 185 6 24 7040 6 215Human P450 1B1 90 6 22 36 6 15Rat P450 1B1 174 6 47 977 6 203Marmoset liver microsomes 8.0 6 0.6 185 6 9Monkey liver microsomes 6.5 6 0.9 283 6 6Human liver microsomes 9.3 6 0.5 266 6 7Rat liver microsomes 13 6 1 1460 6 78

14 Uehara et al.

at ASPE

T Journals on June 30, 2020

dmd.aspetjournals.org

Dow

nloaded from

(Lee et al., 2003). Marmoset P450 1A1 and 1A2 enzymes pre-dominantly catalyzed estradiol 2-hydroxylation similar to those ofhumans and cynomolgus monkeys (Table 5). Additionally, P450 1A2enzymes had higher estradiol 2-hydroxylation activity than P450 1A1enzymes in marmosets, cynomolgus monkeys, and humans. In contrast,P450 1B1 enzymes of marmosets, cynomolgus monkeys, and humanspredominantly catalyzed estradiol 4-hydroxylation, unlike that of rats(Table 5). V395 in human P450 1B1 could account for the differentialestradiol hydroxylation specificities (Nishida et al., 2013). V395 inmarmoset P450 1B1might contribute to specific estradiol 4-hydroxylationsimilar to that of human P450 1B1. These findings indicated that theestradiol hydroxylation specificity of P450 1A1, 1A2, and 1B1 issimilar among marmosets, cynomolgus monkeys, and humans.In conclusion, marmoset P450 1A and 1B share high sequence

identities with the homologous human and cynomolgus monkey P450s.Marmoset P450 1A1 was mainly expressed in lung and small intestine,whereas marmoset P450 1A2 was predominantly expressed in liver.Marmoset P450 1B1 was expressed in most of the extrahepatic tissuesanalyzed. Marmoset 1A1, 1A2, and 1B1 enzymes catalyzed typical hu-man P450 1A and 1B probe reactions, 7-ethoxyresorufinO-deethylation,7-ethoxycoumarin O-deethylation, and phenacetin O-deethylation.Additionally, marmoset P450 1B1 preferentially catalyzed estradiol4-hydroxylation similar to human and cynomolgus monkey P450 1B1,but not estradiol 2-hydroxylation, which is catalyzed by rat P450 1B1.These findings indicated a resemblance of marmoset P450 1A1, 1A2,and 1B1 in metabolic properties among cynomolgus monkeys andhumans. The suitability of this marmoset species for drug metabolismstudies based on the structure and enzymatic properties was experi-mentally confirmed with regard to P450 1A and 1B enzymes. Theinformation presented in this study will help in the understanding ofP450 1A and 1B–dependent drug metabolism in pharmacokinetic ortoxicokinetic studies with marmosets.

Acknowledgments

The authors thank Dr. NorieMurayama and Dr. Makiko Shimizu for technicalhelp and Lance Bell for advice on the writing of the manuscript in English.

Authorship ContributionsParticipated in research design: Uehara, Uno, Yamazaki.Conducted experiments: Uehara, Uno.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

Bièche I, Narjoz C, Asselah T, Vacher S, Marcellin P, Lidereau R, Beaune P, and de Waziers I(2007) Reverse transcriptase-PCR quantification of mRNA levels from cytochrome (CYP)1,CYP2 and CYP3 families in 22 different human tissues. Pharmacogenet Genomics 17:731–742.

Drahushuk AT, McGarrigle BP, Larsen KE, Stegeman JJ, and Olson JR (1998) Detection ofCYP1A1 protein in human liver and induction by TCDD in precision-cut liver slices incubatedin dynamic organ culture. Carcinogenesis 19:1361–1368.

Drahushuk AT, McGarrigle BP, Tai HL, Kitareewan S, Goldstein JA, and Olson JR (1996)Validation of precision-cut liver slices in dynamic organ culture as an in vitro model forstudying CYP1A1 and CYP1A2 induction. Toxicol Appl Pharmacol 140:393–403.

Edwards RJ, Murray BP, Murray S, Schulz T, Neubert D, Gant TW, Thorgeirsson SS, BoobisAR, and Davies DS (1994) Contribution of CYP1A1 and CYP1A2 to the activation of het-erocyclic amines in monkeys and human. Carcinogenesis 15:829–836.

Lee AJ, Cai MX, Thomas PE, Conney AH, and Zhu BT (2003) Characterization of the oxidativemetabolites of 17b-estradiol and estrone formed by 15 selectively expressed human cyto-chrome P450 isoforms. Endocrinology 144:3382–3398.

Narimatsu S, Oda M, Hichiya H, Isobe T, Asaoka K, Hanioka N, Yamano S, Shinoda S,and Yamamoto S (2005) Molecular cloning and functional analysis of cytochrome P450 1A2from Japanese monkey liver: comparison with marmoset cytochrome P450 1A2. Chem BiolInteract 152:1–12.

Nebert DW and Dalton TP (2006) The role of cytochrome P450 enzymes in endogenous sig-nalling pathways and environmental carcinogenesis. Nat Rev Cancer 6:947–960.

Nishida CR, Everett S, and Ortiz de Montellano PR (2013) Specificity determinants of CYP1B1estradiol hydroxylation. Mol Pharmacol 84:451–458.

Orsi A, Rees D, Andreini I, Venturella S, Cinelli S, and Oberto G (2011) Overview of themarmoset as a model in nonclinical development of pharmaceutical products. Regul ToxicolPharmacol 59:19–27.

Rhodes SP, Otten JN, Hingorani GP, Hartley DP, and Franklin RB (2011) Simultaneous as-sessment of cytochrome P450 activity in cultured human hepatocytes for compound-mediatedinduction of CYP3A4, CYP2B6, and CYP1A2. J Pharmacol Toxicol Methods 63:223–226.

Sakuma T, Hieda M, Igarashi T, Ohgiya S, Nagata R, Nemoto N, and Kamataki T (1998)Molecular cloning and functional analysis of cynomolgus monkey CYP1A2. BiochemPharmacol 56:131–139.

Sakuma T, Igarashi T, Hieda M, Ohgiya S, Isogai M, Ninomiya S, Nagata R, Nemoto N,and Kamataki T (1997) Marmoset CYP1A2: primary structure and constitutive ex-pression in livers. Carcinogenesis 18:1985–1991.

Sasaki E (2015) Prospects for genetically modified non-human primate models, including thecommon marmoset. Neurosci Res 93:110–115.

Schulz TG, Thiel R, Neubert D, Brassil PJ, Schulz-Utermoehl T, Boobis AR, and Edwards RJ(2001) Assessment of P450 induction in the marmoset monkey using targeted anti-peptideantibodies. Biochim Biophys Acta 1546:143–155.

Shimada T, Gillam EM, Sutter TR, Strickland PT, Guengerich FP, and Yamazaki H (1997)Oxidation of xenobiotics by recombinant human cytochrome P450 1B1. Drug Metab Dispos25:617–622.

Shimada T, Yamazaki H, Mimura M, Inui Y, and Guengerich FP (1994) Interindividual variationsin human liver cytochrome P-450 enzymes involved in the oxidation of drugs, carcinogens andtoxic chemicals: studies with liver microsomes of 30 Japanese and 30 Caucasians. J PharmacolExp Ther 270:414–423.

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.

Uehara S, Inoue T, Utoh M, Toda A, Shimizu M, Uno Y, Sasaki E, and Yamazaki H (2015a)Simultaneous pharmacokinetics evaluation of human cytochrome P450probes, caffeine,warfarin, omeprazole, metoprolol and midazolam, in common marmosets (Callithrixjacchus). Xenobiotica, in press.

Uehara S, Uno Y, Hagihira Y, Murayama N, Shimizu M, Inoue T, Sasaki E, and Yamazaki H(2015b) Marmoset cytochrome P450 2D8 in livers and small intestines metabolizes typicalhuman P450 2D6 substrates, metoprolol, bufuralol and dextromethorphan. Xenobiotica 45:766–772.

Uehara S, Uno Y, Inoue T, Kawano M, Shimizu M, Toda A, Utoh M, Sasaki E, and Yamazaki H(2015c) Novel marmoset cytochrome P450 2C19 in livers efficiently metabolizes human P4502C9 and 2C19 substrates, S-warfarin, tolbutamide, flurbiprofen, and omeprazole. Drug MetabDispos 43:1408–1416.

Uehara S, Uno Y, Inoue T, Sasaki E, and Yamazaki H (2015d) Substrate selectivities and catalyticactivities of marmoset liver cytochrome P450 2A6 differed from those of human p450 2a6.Drug Metab Dispos 43:969–976.

Uno Y, Uehara S, Murayama N, and Yamazaki H (2011) CYP1D1, pseudogenized in human, isexpressed and encodes a functional drug-metabolizing enzyme in cynomolgus monkey.Biochem Pharmacol 81:442–450.

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 recombinanthuman cytochrome P450s expressed in membranes of Escherichia coli. Protein Expr Purif 24:329–337.

Address correspondence to: Dr. Hiroshi Yamazaki, Showa PharmaceuticalUniversity, 3-3165 Higashi-tamagawa Gakuen, Machida, Tokyo 194-8543, Japan.E-mail: hyamazak@ac.shoyaku.ac.jp

Characterization of Marmoset P450 1A1, 1A2, and 1B1 15

at ASPE

T Journals on June 30, 2020

dmd.aspetjournals.org

Dow

nloaded from