Catecholamines and Serotonin Are Differently Regulatedby TetrahydrobiopterinA STUDY FROM 6-PYRUVOYLTETRAHYDROPTERIN SYNTHASE KNOCKOUT MICE*

Received for publication, March 13, 2001, and in revised form, August 3, 2001Published, JBC Papers in Press, August 21, 2001, DOI 10.1074/jbc.M102237200

Chiho Sumi-Ichinose‡, Fumi Urano§, Risa Kuroda‡, Tamae Ohye§, Masayo Kojima§,Masahiro Tazawa‡, Hiroaki Shiraishi‡, Yasumichi Hagino‡, Toshiharu Nagatsu§,Takahide Nomura‡, and Hiroshi Ichinose§¶

From the ‡Department of Pharmacology, School of Medicine and the §Division of Molecular Genetics,Institute for Comprehensive Medical Science, Fujita Health University, Toyoake, Aichi 470-1192, Japan

(6R)-L-erythro-5,6,7,8-Tetrahydrobiopterin (BH4) is anessential cofactor for tyrosine hydroxylase (TH), trypto-phan hydroxylase, phenylalanine hydroxylase, and ni-tric-oxide synthase. These enzymes synthesize neuro-transmitters, e.g. catecholamines, serotonin, and nitricoxide (NO). We established mice unable to synthesizeBH4 by disruption of the 6-pyruvoyltetrahydropterinsynthase gene, the encoded protein of which catalyzesthe second step of BH4 biosynthesis. Homozygous micewere born at the almost expected Mendelian ratio, butdied within 48 h after birth. In the brain of homozygousmutant neonates, levels of biopterin, catecholamines,and serotonin were extremely low. The number of THmolecules was highly dependent on the intracellularconcentration of BH4 at nerve terminals. Alteration ofthe TH protein level by modulation of the BH4 content isa novel regulatory mechanism. Our data showing thatcatecholaminergic, serotonergic, and NO systems weredifferently affected by BH4 starvation suggest the pos-sible involvement of BH4 synthesis in the etiology ofmonoamine-based neurological and neuropsychiatricdisorders.

Catecholamines, e.g. dopamine (DA),1 norepinephrine (NE),and epinephrine, and serotonin (5-hydroxytryptamine; 5HT)are well known to be involved in many aspects of brain func-tion, including mood, addiction, reward, and sleep. Alterationin the metabolism of these monoamines leads to abnormalitiesin behavior and neuropsychiatric states. For example, in-

creased dopaminergic neurotransmission in the mesolimbicsystem induces hallucinations, illusions, and delusions, whichare the major symptoms of schizophrenia (1). Depletion ofmonoamines may be the cause of depression, because drugsthat can increase the level of monoamines and blockers ofnorepinephrine or serotonin uptake show therapeutic effects inpatients with depression (2). Abnormal monoamine metabo-lism is also suspected to be involved in Tourette’s syndrome,obsessive-compulsive disorder, Rett syndrome, and infantileautism (3, 4).

Biosynthesis of catecholamines and serotonin is mainly reg-ulated by the activity of the hydroxylation reaction catalyzedby tyrosine hydroxylase (TH) in catecholamine synthesis (5)and by tryptophan hydroxylase (TPH) in 5HT synthesis (6).The TH activity is tightly regulated by several factors (forreview, see Ref. 7 and references therein). Short term regula-tion is achieved by phosphorylation of the enzyme and byfeedback inhibition with the end products. Long term regula-tion is mainly governed at the transcriptional level.

(6R)-L-erythro-5,6,7,8-Tetrahydrobiopterin (BH4) is an es-sential cofactor for both TH and TPH. Although many cofactorssuch as folic acid and cobalamin (vitamin B12) are required tobe dietary supplemented, mammals have a de novo biosyn-thetic pathway for BH4. In this pathway, guanosine triphos-phate (GTP) is converted to 7,8-dihydroneopterin triphosphateby GTP cyclohydrolase 1 (GCH), and 7,8-dihydroneopterintriphosphate is then converted to 6-pyruvoyltetrahydropterinby pyruvoyltetrahydropterin synthase (PTPS). This tetrahy-dropterin is subsequently converted to BH4 by sepiapterinreductase (Fig. 1A). The intracellular concentration of BH4 ismainly regulated by the activity of GCH. Factors that canstimulate GCH activity, for example, phytohemagglutinin (8,9), interferon-� (10), and tumor necrosis factor-� (11), can in-crease the level of BH4.

BH4 also acts as the cofactor for phenylalanine hydroxylase(12) and nitric-oxide synthase (NOS) (13, 14). Phenylalaninehydroxylase in the liver is essential for metabolism of theessential amino acid phenylalanine, and converts it to tyrosine.Nitric oxide (NO) has a variety of physiological roles. NO syn-thesized by neural NOS (nNOS) in neuron functions as a neu-rotransmitter, whereas that generated by NOS in endothelialcells serves as a vasodilator. Additionally, NO generated byinducible NOS (iNOS) in macrophages acts as a chemical me-diator, and iNOS activity in these cells is strongly stimulatedby lipopolysaccharide and cytokines.

There are two types of inherited disorders that are caused bya genetic defect in BH4-synthesizing enzymes. One is atypicalhyperphenylalaninemia (malignant type of phenylketonuria),

* This work was supported in part by a grant-in-aid from the Grants-in-aid Program for Encouragement of Young Scientists from the Min-istry of Education, Science, and Culture of Japan (to C. S.-I.); grants-in-aid from the Ministry of Education, Science, and Culture of Japan,the Uehara Memorial Foundation for Life Sciences, and the Yamanou-chi Foundation for Research on Metabolic Disorders (all to H. I.); and bygrants from the Ministry of Health and Welfare of Japan and the FujitaHealth University. The costs of publication of this article were defrayedin part by the payment of page charges. This article must therefore behereby marked “advertisement” in accordance with 18 U.S.C. Section1734 solely to indicate this fact.

¶ To whom correspondence should be addressed. Tel.: 81-562-93-9391; Fax: 81-562-93-8831; E-mail: hichi@fujita-hu.ac.jp.

1 The abbreviations used are: DA, dopamine; NE, norepinephrine; 5HT,5-hydroxytryptamine; kb, kilobase pair(s); TH, tyrosine hydroxylase;TPH, tryptophan hydroxylase; BH4, (6R)-L-erythro-5,6,7,8-tetrahydro-biopterin; GCH, GTP cyclohydrolase 1; PTPS, pyruvoyltetrahydropterinsynthase; NOS, nitric-oxide synthase; iNOS, inducible nitric-oxide syn-thase; nNOS, neural nitric-oxide synthase; DRD, dopa-responsive dysto-nia; L-DOPA, L-3,4-dihydroxyphenylalanine; ES, embryonic stem; HPLC,high performance liquid chromatography; AADC, aromatic L-amino aciddecarboxylase.

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 276, No. 44, Issue of November 2, pp. 41150–41160, 2001© 2001 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

This paper is available on line at http://www.jbc.org41150

by guest on May 13, 2018

http://ww

w.jbc.org/

Dow

nloaded from

which is inherited as an autosomal recessive trait. Mutationsin the BH4-synthesizing or BH4-recycling enzymes such asGCH, PTPS, or dihydropteridine reductase, cause this disease.In the patients, neurological symptoms, e.g. convulsion, drows-iness, and disturbance of muscle tone, are caused by decreasedlevels of neurotransmitters; e.g. catecholamines, 5HT, and NO,in addition to the excess amount of phenylalanine. The other isdopa-responsive dystonia (DRD; also called Segawa’s disease).We found in 1994 that autosomal dominant mutations in theGCH gene cause this disease (15). This disease is characterizedby childhood-onset dystonia with remarkable responsiveness tolow doses of L-3,4-dihydroxyphenylalanine (L-DOPA) (16, 17).The DRD patients have both mutated and wild-type alleles ofthe GCH gene, and the resultant partial depletion of BH4causes a shortage in DA in the nigrostriatal dopaminergicneurons in the brain of DRD patients (15). In these patients,there are no symptoms other than dystonia, and dysfunction ofserotonergic neurons is not obvious, even though BH4 is acofactor for both TH and TPH.

In this present study, for the first time, we established BH4-deficient mice by disrupting the Ptps gene to investigate theeffects of BH4 depletion on the animals and the involvement ofBH4 in regulating neural systems. Our results indicated thatBH4 distinctively regulates the synthesis of catecholaminesaltering the amount of TH protein and that of 5HT, and sug-gested that BH4 metabolism may be deeply involved in thepathophysiology of Parkinson’s disease, stress responses, andneuropsychiatric disorders.

EXPERIMENTAL PROCEDURES

Generation of a Targeted Mutation in 6-PyruvoyltetrahydropterinSynthase Gene—A 17.2-kb mouse Ptps genomic fragment was isolatedfrom a 129/SVJ genomic library by using mouse Ptps cDNA as a probe.The DNA fragment was revealed to contain all exons of the mouse Ptpsgene and 4.6-kb 5�-flanking and 4.6-kb 3�-flanking sequences. To con-struct the targeting vector containing a mouse Ptps 9.5-kb genomicfragment, we replaced 0.6-kb KpnI-BamHI fragment containing exon 1,in which the initiation codon is located, with a neomycin resistance geneunder the promoter of phosphoglycerate kinase (Pgk). MC1-diphtheriatoxin A fragment and pBlueScript (�) sequence were ligated to the 5�end of the vector. The vector was linearized and electroporated into129/SVJ embryonic stem (ES) cells. ES cells were cultured in mediumcontaining G418 for selection. Clones in which homologous recombina-tion occurred were identified by Southern hybridization (see detailsbelow), amplified, and injected into C57BL/6 blastocysts, which werethen transplanted into pseudopregnant female mice. Germline trans-mission occurred from one chimera, and we established C57BL/6 back-ground line by repetitive crossing to compare their phenotype with thatof 129/SVJ-C57BL/6 hybrid mice. Animals were handled according tothe Guidelines for the Care and Use of Laboratory Animals in FujitaHealth University established by the Education and Research Center ofAnimal Models for Human Diseases of Fujita Health University.

Identification of Genotype—DNA was extracted from the cells or micetail, digested with SacI, and analyzed by Southern hybridization usinga 0.45-kb EcoRI-EcoRI fragment, as a 3� outside probe, as indicated inFig. 1B.

Northern Hybridization—Total RNA was extracted from tissues withTrizol reagent (Life Technologies, Inc.). Eight �g (for Ptps) or 20 �g (forTh) of total RNA was electrophoresed and blotted onto a nylon mem-brane. Specific RNA was detected by using mouse Ptps cDNA or mouseTh cDNA (18) as a probe. The same membranes were dehybridized andreprobed with mouse glyceraldehyde 3-phosphate dehydrogenase(G3pdh) cDNA as a control.

Biochemical Analysis—Mouse brains were homogenized in 10 vol-umes of buffer containing 25 mM Tris-HCl (pH 7.4), 1 mM EDTA, 1 mM

EGTA, 1 mM dithiothreitol, and 10% glycerol. Protease inhibitors, i.e. 1�g/ml pepstatin A, 2 �g/ml leupeptin, and 0.5 �M phenylmethylsulfonylfluoride, were added before the TH assay. Protein concentration wasdetermined by the method of Bradford (19).

For the assay of biopterin and neopterin, protein was removed byadding perchloric acid, and pteridines were then oxidized with 1% I2

and 2% KI in 1 N HCl for 60 min in the dark at room temperature, and

analyzed by high performance liquid chromatography (HPLC) withfluorescence detection (20).

Catecholamine and 5HT levels were analyzed by HPLC electrochem-ical and HPLC fluorescence detection, respectively.

TH activity was determined based on the formation of L-DOPA fromL-tyrosine by HPLC electrochemical detection. The reaction mixturecontained 200 mM sodium acetate buffer (pH 6.0), 100 mM 2-mercapto-ethanol, 0.2 mg/ml catalase, 0.2 mM L-tyrosine, and 1 mM BH4. Themixture was incubated for 10 min at 37 °C. The reaction was stopped bythe addition of perchloric acid, and L-DOPA was extracted by an alu-mina column (21). TPH activity was determined by an HPLC fluoro-metric system monitoring the enzymatically formed 5-hydroxytrypto-phan. The reaction mixture contained 0.12 M HEPES buffer (pH 7.6),0.4 mg/ml catalase, 1 mM NSD-1015 (an aromatic L-amino acid decar-boxylase (AADC) inhibitor), 2 mM dithiothreitol, 0.4 mM ferrous ammo-nium sulfate, and 0.4 mM L-tryptophan. 6-Methyltetrahydropterin (500�M) was added to start the reaction, which was carried out for 10 min at30 °C, and then stopped by the addition of perchloric acid (22). AADCactivity was determined by measuring the enzymatically formed DA byHPLC electrochemical detection. The reaction mixture contained 0.03 M

sodium phosphate buffer (pH 7.2), 0.1 mM pargyline (a monoamineoxidase inhibitor), 0.17 mM ascorbic acid, 1 mM L-DOPA, and 0.01 mM

pyridoxal phosphate. The mixture was incubated for 20 min at 37 °C(23).

For measurement of NOS, the homogenate was centrifuged at13,000 � g for 5 min, and the supernatant was used as an enzymesource. The reaction mixture contained 25 mM Tris-HCl (pH 7.4), 3 �M

BH4, 1 �M flavin adenine dinucleotide, 1 �M flavin adenine mononu-cleotide, 1 mM NADPH, 1 mCi/ml [14C]arginine, and 0.6 mM CaCl2, andwas incubated at 25 °C for 60 min. The reaction was stopped by theaddition of EDTA, and enzymatically synthesized 14C-labeled citrullinewas quantified by a scintillation counter (24) (NOS Detect assay kit,Stratagene).

For quantification of amino acids, the homogenate was deproteinizedwith sulfosaritilate, and analyzed by an L-8500 amino acid analyzer(Hitachi).

Immunohistochemistry—Immunohistochemistry was performed bythe avidin-biotin peroxidase complex method (25). Ether-anesthetizednewborn mice were fixed with 4% paraformaldehyde. The brains weredissected and sliced at 40 �m, and their adrenal glands and kidneys, at25 �m, with a cryostat after having been rinsed with sucrose. Sectionswere pretreated with 0.5% hydrogen peroxide (H2O2), blocked with 5%normal swine serum or 5% normal goat serum, and incubated withrabbit anti-TH serum (26), rabbit anti-AADC serum (26), rabbit anti-GCH serum (27) and mouse anti-TPH monoclonal antibody (Ab-1, 0nco-gene, Cambridge) diluted to 1:100,000, 1:20,000, 1:1000, and 1 ng/�lrespectively, followed by biotinylated goat anti-rabbit IgG or biotiny-lated anti-mouse IgG and avidin-biotin peroxidase complex (VectastainABC kit, Funakoshi). Then the sections were visualized by a reactionwith 3,3�-diaminobenzidine tetrahydrochloride and 0.003% H2O2. Atleast two mice were analyzed.

Western Blotting—Antibodies were made against purified bovine TH(26), purified bovine AADC (23), and synthetic peptide for rat GCH (27).Monoclonal antibody against TPH was purchased from Oncogene.Brain homogenates containing 30 �g of protein were separated by 10%SDS-polyacrylamide gel electrophoresis, blotted onto nitrocellulosemembranes, and reacted with the respective first antibodies, and thenwith goat anti-rabbit or anti-mouse IgG conjugated with horseradishperoxidase (1:3000) for 1 h. Immunoreactive bands were detected by useof an ECL-Plus system (Amersham Pharmacia Biotech).

Electrocardiogram of Mice—Electrocardiograms of newborn micewere recorded on the day of delivery. Electrodes were attached to theright fore and left hind limbs, and surface electrocardiogram was mon-itored with an eight-channel oscilloscope using a high gain preamplifier(Nihon Koden, Tokyo, Japan). n � 15 (�/�), 37 (�/�), and 15 (�/�).

Administration of BH4—Ascorbic acid was dissolved to 0.25% in0.9% NaCl. 6R-BH4, which had generously been donated by SuntoryInc. (Osaka, Japan), was dissolved in 0.25% ascorbic acid. The solutionwas administered to mice intraperitoneally. For examination of theacute effect of BH4, mice were sacrificed 1 h after the injection, andfor that of the chronic effect, animals were injected with the drug oncea day for 7 days. Then, on the following day, they were sacrificed forbiochemical and Western blot analyses. For immunohistochemicalanalysis, mice were injected with the drug once a day for 7 days,and 3 h after the last injection their brains were fixed with 4%paraformaldehyde.

6-Pyruvoyltetrahydropterin Synthase Knockout Mice 41151

by guest on May 13, 2018

http://ww

w.jbc.org/

Dow

nloaded from

RESULTS

Ptps Knockout Mice Were Born but Died within 48 h ofBirth—The Ptps gene was inactivated by homologous recombi-nation in ES cells by use of a targeting vector, in which a 0.6-kbKpnI-BamHI Ptps genomic fragment containing exon 1 and apart of the 5� promoter region was replaced with the phospho-glycerate kinase (Pgk)-neomycin resistance gene (Fig. 1B). We

identified three homologously recombinant clones out of 157129/SVJ ES cells by Southern hybridization. We injected thoseinto C57BL/6 blastocysts and obtained five chimeric mice. Thegenotype of the mice was also determined by Southern hybrid-ization (Fig. 1C). Germline transmission occurred in one chi-mera; thus, we established C57BL/6 background mice by repet-itive crossing and compared their phenotype with that of 129/

FIG. 1. Biosynthetic pathway ofBH4, targeting strategy, and geno-type identification of Ptps knockoutmice. A, biosynthetic pathway of BH4. Denovo biosynthetic pathway of BH4 inmammalian cells is indicated. 6-Pyru-voyltetrahydropterin synthase is boxed,and GTP cyclohydrolase 1, the rate-limit-ing enzyme, is indicated by the asterisk.B, mouse Ptps genomic locus and the tar-geting vector. Mouse Ptps gene consistingof 6 exons is depicted on the top. Codingregion and noncoding region are indicatedby open box and closed box, respectively.For construction of the targeting vector, aKpnI-BamHI 0.6-kb genomic fragmentcontaining exon 1 and a part of the 5�promoter region of the Ptps gene was re-placed with a neomycin resistance gene(shaded box). MC1-diptheria toxin A frag-ment ( hatched box) and pBlueScript wereligated to the 5� end. The mutant allele isdepicted on the bottom. Location of 3� out-side probe is indicated above the wild-type allele, and DNA fragments detectedby Southern hybridization in wild-type al-lele and mutant allele are indicated aboveeach allele. B, BamHI; K, KpnI; S, SacI.C, Southern hybridization analysis ofgenotypes. Genomic DNA extracted frommouse tail was digested with SacI andhybridized with a 3� probe. The 18-kbband corresponds to the wild-type allele,and the 9-kb one corresponds to the mu-tant allele. Genotypes of mice are indi-cated on the top of the panels (homozy-gotes, �/�; heterozygotes, �/�; wild type,�/�). D, Northern hybridization analysisof Ptps mRNA expression in the liver andthe brain. Tissues and genotypes are in-dicated on the top of the panel. Total RNAwas extracted from the liver and the brainof each mouse, electrophoresed, and blot-ted. The membrane was hybridized withmouse Ptps cDNA probe (PTPS; upperpanel) and then re-hybridized with mouseG3pdh probe (G3PDH; lower panel).

6-Pyruvoyltetrahydropterin Synthase Knockout Mice41152

by guest on May 13, 2018

http://ww

w.jbc.org/

Dow

nloaded from

SVJ-C57BL/6 hybrid mice. Homozygous mice of 129/SVJ-C57BL/6 hybrids were born roughly according to the Mende-lian ratio (wild type; 26.7%, heterozygote; 51.2%, homozygote;22.2%, n � 424) without gross anomaly in major organs (datanot shown). This means that mutant mice could survive inutero during embryogenesis; however, all homozygotes diedwithin 48 h after birth. On the other hand, genotypes of new-born mice of C57BL/6 background in the fourth to fifth gener-ation were identified as follows: wild type, 26.6%; heterozygote,

53.2%; homozygote, 20.3% (n � 79), and all homozygotes alsodied within 48 h. Thus all experiments were done with 129/SVJ-C57BL/6 hybrids.

Biopterin, Catecholamines, and Serotonin Were Severely Re-duced in the Brain of Newborn Ptps Knockout Mice—Northernhybridization using total RNA extracted from the brain and theliver of newborn mice showed that homozygotes were nullmutants, for Ptps mRNA was not detected (Fig. 1D). Biochem-ical analysis revealed a dramatic decrease in the level of biop-terin (6.3% of wild type), the oxidized form of BH4, and theaccumulation of neopterin (17.6 times higher than wild type), ametabolite of 7,8-dihydroneopterin triphosphate (Fig. 2A).Since BH4 is an essential cofactor for TH and TPH, we alsoquantified catecholamines and 5HT in the brain of newbornmice. As we expected, the levels of DA, NE, and 5HT were alldecreased in homozygotes (14.6, 6.3, and 7.8% of wild type,respectively). There was no significant difference between het-erozygotes and wild-type mice (Fig. 2B). These results showthat in vivo synthesis of catecholamines and 5HT was dimin-ished by the lack of BH4. The phenylalanine content in bothbrain and liver of homozygotes was 2 times higher than that ofthe wild type (Fig. 2C).

TH Activity Was Severely Impaired by Depletion of BH4—Weassayed the activity of TH, AADC, TPH, GCH, and NOS in thebrain of newborn mice. AADC decarboxylates L-DOPA and5-HTP to form DA and 5HT, respectively, and co-exists withTH and TPH in catecholaminergic and serotonergic neurons.Although both TH and TPH are BH4-dependent enzymes, onlyTH activity was severely impaired in the homozygotic mutantmice (7.0% of wild type; Table I). AADC, GCH, and TPH activ-ities were not altered at all. Activity of NOS was significantlydecreased (72% of wild type), but far less than that of TH(Table I).

TH Immunoreactivity Was Reduced in the Nerve Terminalsbut Not in the Cell Bodies—Next, we examined the localizationof TH protein by using an immunohistochemical technique.Dopaminergic and noradrenergic neural cell bodies are locatedin the substantia nigra, ventral tegmental area, and locusceruleus. Those neural cell bodies in homozygotic mutant miceas well as in wild-type ones were stained with anti-TH antibodyequally (Fig. 3, A and B; locus ceruleus, data not shown), andwere also stained with anti-AADC antibody (data not shown).Although there was no significant difference in TH staining oncell bodies, we found a striking difference in the terminalregion. In the brain of wild-type mice, nerve fibers werestrongly stained with anti-TH antibody, and many positivevaricosities were easily seen at higher magnification in thestriatum, to which dopaminergic neurons in the substantianigra project (Fig. 3, C and E). In the same region of thehomozygotic mutant mice, nerve fibers were not stained; in-stead, a few unusual TH-immunopositive cells were found inthe striatum (Fig. 3, D and F). In order to examine the presenceof dopaminergic fibers, we stained the same sections with anti-

FIG. 2. Pteridine, monoamine, and phenylalanine contents inthe tissues of newborn mice. A, neopterin (black column) and biop-terin (hatched column). B, DA (black column), NE (hatched column),and 5HT (white column) contents in the brain of newborn mice. C,phenylalanine contents in their brain (black column) and liver (whitecolumn). Genotypes are indicated below the columns. Values are themeans � S.E., n � 6. Significant difference from the correspondingvalue for wild-type is indicated by asterisks: *, p � 0.05; **, p � 0.01(Student’s t test).

TABLE IEnzyme activities in the brains of newborn knockout and wild-type mice

Values are indicated by the mean � S.E. TH, AADC, TPH, and GCH activities are indicated by pmol/min/mg protein, and that of NOS is indicatedby cpm/min/mg protein. Values in parentheses are the enzyme activity as a percentage of that of the wild-type mice.

Genotype n TH AADC TPH GCH NOS

pmol/min/mg protein cpm/min/mg protein

�/� 5 4.12 � 0.47 685 � 30 19.9 � 1.7 2.87 � 0.14 211,000 � 12,000�/� 6 4.07 � 0.11 671 � 21 16.7 � 1.4 3.10 � 0.22 236,000 � 21,000

(99) (98) (84) (108) (112)�/� 4 0.29 � 0.08a 627 � 21 21.0 � 2.0 2.86 � 0.20 152,000 � 15,000b

(7) (92) (105) (100) (72)a Significant difference from the corresponding values of wild-type, p � 0.01 (Student’s t test).b Significant difference from the corresponding values of wild-type, p � 0.05 (Student’s t test).

6-Pyruvoyltetrahydropterin Synthase Knockout Mice 41153

by guest on May 13, 2018

http://ww

w.jbc.org/

Dow

nloaded from

AADC antibody. This time nerve fibers in the striatum werepositively stained in homozygotes similarly as in the wild type(Fig. 3, G and H). These results indicate that the amount of THprotein was markedly decreased only in nerve terminals, butnot in cell bodies. These findings were identical to those ob-tained from the other homozygous mouse.

A reduction in TH protein in the brain of homozygous new-born mice was also shown by Western blotting (Fig. 9, lanes 1and 2), although we could not see a big difference in the mRNAlevel of Th by Northern blot analysis (Fig. 4). This Western blotanalysis clearly confirmed that the reduction in TH activitywas caused by decrease in the amount of the TH proteinamount itself.

We also stained brain sections of newborn mice with anti-GCH and with anti-TPH antibody. Neural cell bodies in thesubstantia nigra, ventral tegmental area, and locus ceruleus ofhomozygotes were immunopositively stained with anti-GCHantibody as in the wild type (data not shown). Serotonergic cellbodies in the dorsal raphe nucleus of the homozygous micewere also well stained with anti-TPH antibody in the Ptps-nullmice as well as in the wild type (data not shown).

Function of the Peripheral Nervous System Was Impaired inPtps Knockout Mice—NE is a neurotransmitter of the periph-eral sympathetic neurons, and NE and epinephrine are syn-thesized in adrenomedullary cells as hormones. Since cardio-vascular function, respiratory rate, metabolism, and bodytemperature are regulated by both sympathetic and parasym-pathetic nervous systems, physical catastrophe could be causedby failure of the peripheral nervous systems as well as bydysfunction of the central nervous system in Ptps knockoutmice. So we recorded the surface electrocardiogram of newbornhomozygotic mutant mice and compared their heart rates withthose of their littermates. Except for one ventricular rhythm,all homozygous mutants showed sinus rhythms that tendedtoward bradycardia (64% of the wild type, Fig. 5, A and B).Next, with anti-TH antibody we immunostained the adrenalgland and the kidney where many terminals of sympatheticnerves exist. Adrenomedullary cells in the homozygotic mu-tants were immunopositive as well as those in the wild type,but sympathetic nerve terminals in the kidney of homozygotesshowed less positive staining for TH compared with those of thewild-type animals (Fig. 6). These results indicate that TH pro-tein was decreased selectively in nerve terminals and that

FIG. 3. Immunostaining of TH and AADC in the brain of wild-type and homozygous mutant mice. A, C, E, and G, wild-type mice(�/�); B, D, F, and H, homozygous mutants (�/�). Cell bodies ofdopaminergic neurons in the midbrain of the homozygote mutant werestained with anti-TH antibody (B) to the same extent as those of thewild type (A). In the striatum of the wild-type mouse (C), many neuralfibers with varicosities were reactive with anti-TH antibody, which wasespecially evident at higher magnification (E). Neural fibers were notstained with anti-TH antibody in the same region of the homozygousmutant mouse (D), and unusual TH-immunopositive cells could be seenin the mutant striatum (F). Some of these are indicated by arrows. Thestriatum of the homozygous mutant was reactive with anti-AADC an-tibody (H) to the same extent as that of the wild type (G). Asterisks inC and D indicate the portions that are shown at higher magnificationsin E and F, respectively. VTA, ventral tegmental area; SN, substantianigra; CP, caudate-putamen; Cx, cortex. Scale bars in B, D, and H are500 mm, and the one in F is 50 mm.

FIG. 4. Northern blotting of TH. Genotypes are indicated at the topof the panels. Northern hybridization of Th mRNA using total RNAextracted from the brain of newborn mice. Upper panel, Th; lower panel,G3pdh.

6-Pyruvoyltetrahydropterin Synthase Knockout Mice41154

by guest on May 13, 2018

http://ww

w.jbc.org/

Dow

nloaded from

dysfunction of peripheral nervous system could thus be in-volved in the cause of death for homozygous mutant mice.

Effect of Acute BH4 Administration—To explore the acuteeffect of administration of BH4 on catecholamine and 5HTlevels in the brain, we injected newborn mice with BH4 at adose of 50 mg/kg body weight by the intraperitoneal route onday 0 (postnatal day 0), and sacrificed them 60 min later.Monoamine contents and the monoamine-synthesizing enzymeactivities in their brain were analyzed and compared withthose of vehicle-injected mice (0.25% ascorbic acid used to pro-tect BH4 from oxidization). The biopterin content in the brainsof both BH4-injected wild type and homozygous mutants was6–7 times higher than that of the ascorbic acid-injected wild-type mice (Fig. 7A). This finding indicates that BH4 passedthrough the immature blood-brain barrier well in neonates.Despite the great elevation in the amount of BH4, the DAcontent in the brain was not altered at all in the wild-type mice,whereas the 5HT content was slightly increased (1.4 times). Inthe Ptps knockout mice, the 5HT level in the brain of BH4-injected animals was markedly elevated to more than 10 times

the level for the ascorbic acid-injected ones, reflecting a recov-ering up to 73% of the level for the wild type injected withascorbic acid (Fig. 7C). In contrast, the increase in DA levels inthe Ptps-null mice was only 1.5-fold the original level (Fig. 7B).TH activity was increased �3-fold in the homozygous mutantsby 60 min after the administration of this dose of BH4, whereasthe TPH activity was not altered (Fig. 7, D and E).

Effect of Repeated BH4 Administration—We tried to rescuethe Ptps knockout mice by multiple intraperitoneal injectionsof BH4 (Table II). We administered BH4 at 5 or 50 mg/kg oncea day intraperitoneally seven times, and then sacrificed themice on the next day (postnatal day 7) after the last injection.The knockout mice could survive up to postnatal day 7, al-though growth retardation was obvious. The body weight of thehomozygous mutants given the low and high doses was only 29and 64%, respectively, of that of their heterozygous littermates.Accumulation of phenylalanine in the brain of homozygotesrescued with 5 mg/kg BH4 was observed, being 22.3 timeshigher than that for the heterozygote given the same dose ofBH4 (Table II).

FIG. 5. Electrocardiographic analy-sis of the newborn mice. A, typical sur-face electrocardiograms of the littermatesare depicted. �/�, wild type; �/�, het-erozygous; �/�, homozygous. Examplesof P, Q, R, S and T waves are indicated onthe top panel. B, heart rate calculatedfrom the electrocardiograms. Values arethe means � S.E., n � 15 (wild type), 37(heterozygote), and 15 (homozygote). Sig-nificant difference from the correspondingvalue for wild-type is indicated by aster-isks: *, p � 0.01 (Student’s t test).

6-Pyruvoyltetrahydropterin Synthase Knockout Mice 41155

by guest on May 13, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Then we analyzed the content of pteridines and monoamines,as well as TH, GCH, and TPH activities, in the brain. Thebiopterin content in the brain of homozygous mice was not soimproved by the 5 mg/kg BH4 injection, and the TH activityremained low. In contrast, the biopterin content in the brain ofhomozygotes rescued with 50 mg/kg BH4 was 44% of that of theheterozygotes used as a control. By this subchronic adminis-tration of BH4, TH activity was recovered to a greater extentthan the DA level in the brain. In the homozygotes we observed50% of the TH activity found for the 50 mg/kg BH4-injectedheterozygous mice, although the DA level remained very low(16% of the level for heterozygous mice). The 5HT content inthe brain of the homozygous mutants was not remarkablyimproved 1 day after the last injection. The activity of TPH andGCH tended to increase in the homozygous mice rescued with5 mg/kg BH4, but the TH activity was still low (Table II).

To examine whether TH protein was recovered in the stria-tum, we stained with anti-TH antibody brain sections of micerescued by 7 days’ administration of 50 mg/kg BH4. The mousebrains were fixed 3 h after the last injection. Strikingly, thecaudate-putamen of the rescued homozygous mice was quitereactive with the anti-TH antibody as was that of the heterozy-gous control mice (Fig. 8, A–C). Further, we confirmed theincrease in the amount of TH protein in the BH4-rescuedhomozygous mutant mice by Western blot analysis. As shownin Fig. 9, we detected a substantial immunoreactive band of THafter repeated administration of 50 mg/kg BH4 (lane 5), al-though only a faint band was observed before the administra-tion of BH4 (Fig. 9, lane 2). The amount of other proteins,including AADC, GCH, and TPH, did not show significantalteration between heterozygous mice and homozygous mutantmice administered BH4. The data also showed that TPH pro-tein was not decreased in the neonatal brain, in contrast to thecase for TH.

DISCUSSION

For the first time, we established mice in which de novobiosynthesis of BH4 was impaired by disruption of the Ptpsgene, whose product catalyzes the conversion of 7,8-dihydrone-opterin triphosphate to 6-pyruvoyl-5,6,7,8-tetrahydropterin.Homozygous mutants were born nearly according to the Men-delian law, but all of them died within 48 h. In the brain ofnewborn Ptps-null mutants, only a trace amount of biopterinwas present in the brain, neopterin derived from 7,8-dihydro-neopterin triphosphate accumulated, and catecholamine and5HT levels were markedly reduced. It was surprising that THactivity was selectively diminished compared with that of TPHand NOS, for BH4 is an essential cofactor for those enzymes asfor TH. The reduction of TH activity was accompanied by adecreased amount of the enzyme only at the nerve terminals.Whereas a single administration of BH4 to the null mice led toa great increase in the 5HT content, 1 h after the administra-tion, the DA content remained low. These results suggest dif-ferential action of BH4 between the regulation of cat-echolamines and that of 5HT in the brain.

We observed accumulation of phenylalanine in the brain andliver of homozygous mutant mice. Most cases of hyperphen-ylalaninemia are caused by a genetic defect in phenylalaninehydroxylase. However, in �10% of the cases it is caused bygenetic defects in biopterin metabolism. More than 400 pa-tients were reported to have this “atypical” form of hyperphen-ylalaninemia, and �40% of them had the genetic defect in thePTPS gene (28). The present Ptps knockout mouse is an animalmodel for this type of atypical hyperphenylalaninemia.

PTPS Knockout Mice Were Not Embryonically Lethal—De-spite the dramatic change in BH4 and monoamine levels in thebrain, homozygous mutants were born according to the Men-delian ratio, and no gross anomaly was seen by microscopicexamination of major organs including the brain. The present

FIG. 6. Immunostaining of TH in theadrenal gland and kidney. A and B,the adrenal gland of a wild type (�/�) anda homozygous mutant (�/�) mouse, re-spectively. The adrenal medulla in the ho-mozygous mutant mouse was positivelystained with anti-TH antibody as well asthe wild-type one. C and D, the kidney ofwild type and homozygous mutant, re-spectively, at higher magnification. Sym-pathetic nerve terminals were wellstained with anti-TH antibody in the kid-ney of the wild-type mouse (C), whereasthose of the homozygous mutant wereweakly stained (D). Scale bar in B, 100�m; in D, 50 �m.

6-Pyruvoyltetrahydropterin Synthase Knockout Mice41156

by guest on May 13, 2018

http://ww

w.jbc.org/

Dow

nloaded from

result thus indicates that the Ptps knockout mouse is not anembryonically lethal. Previous studies on Th knockout mice(26, 29) and dopamine �-hydroxylase (Dbh) knockout mice (30)showed that catecholamines are essential for embryonic devel-opment, because these knockout mice died in utero at a latestage of embryonic development or shortly after birth. Dbhknockout mice die at embryonic day 11 or 12 if the mother wasalso Dbh-null. Our findings raise the possibility that a minorside pathway for BH4 biosynthesis may exist, because it wasreported that BH4 hardly passed through placenta (31).

We succeeded in rescuing homozygous mutants from deathfor up to 7 days by intraperitoneal administration of BH4. BH4,being a hydrophilic molecule, is believed to hardly permeateinto the blood-brain barrier. However, in the neonatal stage,peripherally administered BH4 could pass through the imma-ture blood-brain barrier well and was incorporated into cells toincrease neurotransmitter levels (Fig. 7). Their increase byadministration of BH4 was more remarkable in the Ptps knock-out mice than in the wild-type mice. It is possible that a BH4transporter or carrier protein to maintain the BH4 concentra-tion constant in the cell exists.

Mice lacking nNos (32), eNos (33), and iNos (34) were estab-lished by gene targeting. Hypertrophy of the pyloric sphincter

and the circulate muscle layer along with the lack of NADPH-diaphorase-containing neurons occurred in the nNos knockoutmice (32). eNos mutant mice were hypertensive, and acetylcho-line-induced relaxation was absent in their vascular system(33). Hypotension caused by endotoxin in wild-type mice wasnot observed in mice deficient in iNos, and antibacterial andanti-tumor activities in the latter were poor (34). Since we didnot subfractionate the brain homogenate, the NOS activity thatwe assayed was that of mixed types of NOS. However, themajor subtype of NOS in the brain is nNOS in neurons, andiNOS activity remained low before induction with immunolog-ical stimulation including cytokines. Thus, the NOS activitythat we assayed would mainly reflect that of nNOS. In thepresent study, the NOS activity was slightly reduced by thelack of BH4 in vitro, but no phenotype relating to NO systemswas evident in vivo. We could not determine whether in vivoNO production was impaired or not, because the levels of NOmetabolites such as nitrite and nitrate were too low to bequantified under the unstimulated condition. However, as theKm value (0.02–0.03 �M) of NOS against BH4 is very low (35),the binding between BH4 and NOS protein may be so strongthat a trace amount of BH4 may be sufficient for constitutiveNOS activity. It is likely that depleted BH4 may affect theimmune response or host defense mechanisms when iNOSinduction is required. We are currently planning an experi-ment to examine this possibility.

Cause of Death in PTPS Knockout Mice—Th knockout mice,which were embryonic lethal but could be rescued by adminis-tration of L-DOPA or L-threo-dihydroxyphenylserine, died ap-parently of cardiovascular failure as shown by congestion in theliver and vessels and by an abnormal electrocardiogram (26,29). Although the cause of neonatal death in our Ptps knockoutmice remains unclear, the severe decrease in catecholaminelevels, bradycardia, and the marked reduction in TH immuno-reactivity in sympathetic nerve terminals suggest that failurein the peripheral nervous system in addition to abnormality inthe brain could be a causative factor leading to death. BH4could be maternally supplied in utero; however, it would begradually depleted after birth. This assumption is further sup-ported by the prolonged survival period with administration ofBH4. The mortality of Ptps knockout mice resembled that ofPhox2a (36) or Nurr1-deficient mice (37). In the former, thelocus ceruleus and subsets of sympathetic, parasympathetic,and sensory ganglia were absent; and in the latter, TH-positivecells in the ventral midbrain were not generated. DA-deficientmice were reported to be hypoactive and showed a feedingdisturbance after weaning (38). However, as thinning out oflittermates did not seem to be very helpful for survival ofhomozygous mutants of Ptps knockout mice (data not shown),simple competition with their littermates was not a criticalfactor. The phenylalanine content in the brain of homozygousmutant neonates was only 2 times higher than that of the wildtype. The contribution of hyperphenylalaninemia to the earlydeath seems to be low, as it is known that the symptom ofhuman phenylketonuria is not remarkable at the neonatalstage, but becomes worse by intake of phenylalanine in themilk. Further, the level of 5HT also decreased in Ptps knockoutmice, and NO system could be impaired by BH4 starvation.Although symptoms based on dysfunction of 5HT and NO sys-tems were not observed in human PTPS deficiency, these fac-tors also may have contributed to the death of knockout mice.

TH-positive Cells in the Striatum—Interestingly, we foundTH-positive cells in the striatum of the BH4-deficient mice,which cells were not detectable in the wild-type mice. AADC-positive cells were also not visible, and GCH-immunopositivecell bodies were not detected in the striatum of wild-type and

FIG. 7. Acute effect of intraperitoneal administration of BH4on the brain of newborn mice. Biopterin (A), DA (B), serotonin(5HT) (C), TH activity (D), and TPH activity (E) in the brain of newbornmice at 60 min after an intraperitoneal injection of ascorbic acid (blackcolumn) or 50 mg/kg BH4 (hatched column). Genotypes are indicatedbelow the columns. Values are the means � S.E., n � 5; significantdifference from the corresponding value for each genotype is indicatedabove the columns (Student’s t test).

6-Pyruvoyltetrahydropterin Synthase Knockout Mice 41157

by guest on May 13, 2018

http://ww

w.jbc.org/

Dow

nloaded from

homozygous mice. It is possible that TH- and/or AADC-positivecells in the wild type were masked by the extensively stainednerve terminals. However, such intrinsic TH-positive cellswere also observed in the striatum of rats with 6-hydroxydo-pamine lesions (39) and monkeys treated with 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (a dopaminergic neurotoxinthat induces parkinsonism) (40), and these cells were reportedto be increased in number after the treatment. In the lattercase, the cells were reported to be doubly positive when stained

with anti-TH and anti-DA transporter antibodies, and theywere further reactive with anti-glutamate decarboxylase(GAD67) antibodies. TH-positive neurons in the striatum werealso observed in DA transporter knockout mice, in which theDA content and TH protein were reduced by 95 and 90%,respectively (41). Physiological and pathological meanings ofthese cells are still unclear; however, they could represent anendogenous mechanism compensating for DA deficiency in thestriatum and may be common in BH4-and/or DA-deficientstates of the human brain.

Differential Regulation of Monoaminergic Neurons by BH4—The fact that catecholaminergic, serotonergic, and NO systemswere differently affected by BH4 starvation was clearly shownby our results. In the catecholaminergic system, depletion ofBH4 resulted in a reduced level of TH protein (accompanied byreduced TH activity) at nerve terminals. A reduction in THimmunoreactivity was observed in the striatum and sympa-thetic nerve terminals, whereas cell bodies of catecholaminer-gic neurons in the substantia nigra, ventral tegmental area,and locus ceruleus, as well as adrenomedullary cells, were wellstained with the anti-TH antibody. In the serotonergic system,however, TPH activity was not affected by the severe reductionin the BH4 level. These differences were clearly reflected byresponses to acute BH4 administration. In the Ptps knockoutmice, the 5HT content in the brain was dramatically increased60 min after the administration, although the DA content wasonly slightly elevated. These differences were also observed toa lesser extent even in normal animals, which could synthesizeBH4 (Fig. 7, B and C).

The BH4 concentration in the cell had been thought to be nothigh enough to saturate TH (42–44). However, our resultsshowed that acute administration of BH4 to the wild-type micewas not effective to increase the DA content in the brain,

TABLE IIEffect of repeated intraperitoneal administration of BH4 and ascorbic acid on the brain biochemistry of knockout mice

Mice were injected with 0.25% ascorbic acid (AA) and 5 or 50 mg/kg BH4. Values are the means � S.E. Body weight is indicated in grams.Phenylalanine content is expressed as nmol/mg protein. Neopterin, biopterin, DA, and serotonin (5HT) are expressed as pmol/mg protein. TH, TPH,and GCH activities are indicated by pmol/min/mg protein. Values in parentheses are percentages to those of their heterozygous littermates.

Genotype n Body weight Phenylalanine Neopterin Biopterin DA TH 5HT TPH GCH

AA�/� 4 4.63 � 0.09 1.6 � 0.2 0.77 � 0.04 9.25 � 0.32 16.4 � 1.3 14.9 � 0.6 1.50 � 0.11 21.1 � 1.3 5.22 � 0.26

BH4(5 mg/kg)

�/� 4 5.88 � 0.19 1.3 � 0.04 0.80 � 0.05 9.86 � 0.29 19.9 � 1.3 10.6 � 0.8 1.08 � 0.30 22.7 � 0.8 5.32 � 0.35�/� 3 1.73 � 0.20a 29 � 10b 17.8 � 2.5a 1.64 � 0.38a 0.85 � 0.3a 1.49 � 0.2a NDc 34.9 � 0.8a 6.60 � 0.57b

(29) (2230) (2230) (17) (4) (14) (0) (154) (124)

BH4(50 mg/kg)

�/� 5 4.42 � 0.04 1.6 � 0.2 0.93 � 0.08 19.1 � 2.48 17.2 � 0.9 13.4 � 0.9 0.39 � 0.01 22.9 � 2.3 5.08 � 0.18�/� 3 2.83 � 0.34a 5.4 � 1.3a 13.5 � 0.4a 8.48 � 0.40 2.73 � 0.9a 6.69 � 0.66a 0.02 � 0.01a 24.0 � 5.1 5.60 � 0.09

(64) (338) (1450) (44) (16) (50) (5) (105) (110)a Significant difference from the corresponding values for heterozygotes, p � 0.01 (Student’s t test).b Significant difference from the corresponding values for heterozygotes, p � 0.05 (Student’s t test).c Not detected.

FIG. 8. Immunohistochemical stain-ing of TH in the striatum of mice res-cued with BH4 administration. A, anascorbic acid (vehicle)-administered het-erozygous mouse (�/�); B, a 50 mg/kgBH4-administered heterozygous mice(�/�); C, a 50 mg/kg BH4-administeredhomozygous mouse (�/�). These injec-tions were done for 7 days. CP, caudate-putamen; Cx, cortex. Scale bar, 500 �m.

FIG. 9. Western analysis using the brain homogenate of micerescued with BH4 administration. Thirty �g of protein was appliedto each lane and stained with specific antibodies for TH, AADC, GCH,and TPH, respectively, as indicated at the left side of panels. Lane 1,wild type (�/�) newborn; lane 2, homozygous (�/�) newborn; lane 3,ascorbic acid (vehicle)-administered heterozygous mouse (�/�); lane 4,50 mg/kg BH4-administered heterozygous mouse (�/�); lane 5, 50mg/kg BH4-administered homozygous mouse (�/�). These injectionswere done for 7 days.

6-Pyruvoyltetrahydropterin Synthase Knockout Mice41158

by guest on May 13, 2018

http://ww

w.jbc.org/

Dow

nloaded from

although the biopterin content in the brain was elevated up to7-fold of that of the vehicle-injected mice (Fig. 7). Instead, BH4depletion caused a reduction in the content of TH protein.These findings suggest that BH4 regulates the catecholaminesynthesis through altering the amount of TH molecules, espe-cially at nerve terminals.

One possible explanation for the reduction only at nerveterminals would be that TH forms a stable complex with BH4and/or catecholamines, and that only the stable complex iscarried to nerve terminals through axonal flow. We showedpreviously that human TH shows regulatory kinetic propertiesfor BH4 (45). Flatmark et al. (46) also observed the negativecooperativity of BH4 binding to TH. On the other hand, TH hasbeen reported to make a stable and inactive complex withcatechol compounds (47). An alternative possibility is that THis carried to the terminals with or without BH4. TH protein andmRNA levels are normal in neural cell bodies because that iswhere the protein is synthesized. A shortened half-life wouldonly be revealed at nerve terminals where TH accumulates. Weobserved an additional immunoreactive band of small molecu-lar size in the brain homogenate of Ptps-null mutants by West-ern blot analysis (data not shown), which could represent de-graded TH protein. Although more study is required to clarifythe molecular mechanism, these data strongly suggest that theturnover of TH protein is very rapid in the absence of thebiopterin cofactor.

A similar reduction in the level of TH protein only in thestriatum was shown also in patients with DRD (Segawa’s dis-ease). DRD is caused by disturbed BH4 synthesis (15). Autop-sied brains of patients with the disease showed a markedreduction in the content of TH protein in the striatum (48, 49).

Our present results would explain the reason why only dys-function of nigrostriatal dopaminergic neurons is apparent,and why serotonergic systems can function properly in theDRD patients. Because of decreased synthesis of BH4, theamount of TH protein carried to the nerve terminal is reduced.At the nerve terminal (striatum), although a small amount ofBH4 is present there, the ability to synthesize DA is not suffi-cient due to the decreased number of the TH molecules. In thecase of serotonergic neurons, the stability of the TPH protein isnot affected by depletion of BH4. Thus, the number of TPHmolecules at the nerve terminal is the same as the normal one,and a sufficient amount of 5HT can be synthesized, althoughthe rate of de novo synthesis of BH4 is partly impaired. Thepresence of a recycling system powered by dihydropteridinereductase may help to facilitate the monoamine synthesisthrough re-redox of quinonoid-dihydrobiopterin, which isformed from the reaction with aromatic amino acid hydroxyl-ases, to BH4.

Reduction in the Level of TH Protein in the Striatum WasReversed by BH4 Administration—Our results also showedthat the reduction in the TH protein level was reversible bysupplementation with BH4. Intraperitoneal administration of50 mg/kg BH4 increased TH activity in the brain even after 1 h,and this increase was remarkable in mice continuously injectedwith 50 mg/kg BH4 for 7 days. This is a novel regulatorymechanism for TH protein by BH4.

In the brain of homozygous mutant mice, DA and 5HT levelswere very low (16 and 5%, respectively, of those of ascorbicacid-treated heterozygous mice), although the BH4 level waselevated up to 44% of the control 24 h after the last of therepeated injections of 50 mg/kg BH4 (Table II). The low levelsof 5HT and DA would be explained by the time after the lastinjection. We think that the local concentration of BH4 inmonoaminergic neurons would not be as high as 44% of thecontrol, because BH4 incorporated into monoaminergic neu-

rons might be consumed during the 24 h before the analysiswas made.

There have been several reports describing the dissociationbetween Th mRNA levels and expression of the enzyme activity(7). For example, a transgenic mouse bearing extra copies of thehuman TH gene expressed the transgene mRNA in dopamin-ergic neurons at levels 50 times over that of the endogenousgene. Despite this increase in the amount of mRNA, TH immu-noreactivity and enzyme activity in the striatum were alteredby only 2–3-fold (50). A part of the dissociation between ThmRNA and the protein could be explained by the stabilizingeffect of the cofactor.

In summary, our results showed that the amount of THprotein at the nerve terminal is governed by the intracellularconcentration of BH4, that transient elevation in the BH4content increases the biosynthesis of 5HT but not that of cat-echolamines, and that a continuous elevated level of BH4 leadsto an increase in the catecholamine levels. We think that thispaper opens a new insight into regulation of monoamines byBH4.

Acknowledgments—We are indebted to Drs. Hirohide Sawada andKazuhiro Nishii for excellent technological help with the mice, includ-ing in vitro fertilization. We appreciate significant discussions withDrs. Ikuko Nagatsu, Mutsushi Matsuyama, Kayo Ogino, and TadayoshiHata. Finally, we thank Atsumi Tsubata, Kumiko Asano, Chie Goto,Yuko Isobe, Yuko Suenaga, and Tomoko Watanabe, as well as ToshimichiHoriuchi for helping with parts of our experiments.

REFERENCES

1. Carlsson, A. (1988) Neuropsychopharmacology 1, 179–1862. Hirschfeld, R. M. (2000) J. Clin. Psychiatry 61, 4–63. Anderson, G. M. (1987) Med. Biol. 65, 67–744. Cook, E. H., and Leventhal, B. L. (1996) Curr. Opin. Pediatr. 8, 348–3545. Nagatsu, T., Levitt, M., and Udenfriend, S. (1964) J. Biol. Chem. 239,

2910–29176. Lovenberg, W., Jaquier, E., and Sjoerdsma, A. (1967) Science 155, 217–2197. Kumer, S. C., and Vrana, K. E. (1996) J. Neurochem. 67, 443–4628. Blau, N., Joller, P., and Atares, M. (1985) Clin. Chim. Acta 148, 47–529. Ziegler, I., Schott, K., Lubbert, M., Herrmann, F., Schwulera, U., and Bacher,

A. (1990) J. Biol. Chem. 265, 17026–1703010. Schoedon, G., Troppmair, J., Fontana, A., Huber, C., Curtius, H. C., and

Niederwieser, A. (1987) Eur. J. Biochem. 166, 303–31011. Werner, E. R., Werner-Felmayer, G., Fuchs, D., Hausen, A., Reibnegger, G.,

Yim, J. J., and Wachter, H. (1991) Biochem. J. 280, 709–71412. Kaufman, S. (1959) J. Biol. Chem. 234, 2677–268213. Tayeh, M. A., and Marletta, M. A. (1989) J. Biol. Chem. 264, 19654–1965814. Kwon, N. S., Nathan, C. F., and Stuehr, D. J. (1989) J. Biol. Chem. 264,

20496–2050115. Ichinose, H., Ohye, T., Takahashi, E., Seki, N., Hori, T., Segawa, M., Nomura,

Y., Endo, K., Tanaka, H., Tsuji, S., Fujita, K., and Nagatsu, T. (1994) Nat.Genet. 8, 236–242

16. Segawa, M., Ohmi, K., Itoh, S., Aoyama, M., and Hayakawa, H. (1971) Shinryo(Tokyo) 24, 667–672

17. Segawa, M., Hosaka, A., Miyagawa, F., Nomura, Y., and Imai, H. (1976) inAdvances in Neurology (Eldrige, R., and Fahn, S., eds) Vol. 14, pp. 215–233,Raven Press, Ltd., New York

18. Ichikawa, S., Sasaoka, T., and Nagatsu, T. (1991) Biochem. Biophys. Res.Commun. 176, 1610–1616

19. Bradford, M. M. (1976) Anal. Biochem. 72, 248–25420. Fukushima, T., and Nixon, J. C. (1980) Anal. Biochem. 102, 176–18821. Nagatsu, T., Oka, K., and Kato, T. (1979) J. Chromatogr. 163, 247–25222. Friedman, P. A., Kappelman, A. H., and Kaufman, S. (1972) J. Biol. Chem.

247, 4165–417323. Ichinose, H., Kojima, K., Togari, A., Kato, Y., Parvez, S., Parvez, H., and

Nagatsu, T. (1985) Anal. Biochem. 150, 408–41424. Bredt, D. S., and Snyder, S. H. (1994) Annu. Rev. Biochem. 63, 175–19525. Nagatsu, I., Yamada, K., Sakai, M., and Karasawa, N. (1993) in Methods in

Neurotransmitter and Neuropeptide Research (Parvez, S. H., Naoi, M.,Nagatsu, T., and Parvez, S., eds) pp. 151–183, Elsevier Science PublishersB. V., Amsterdam

26. Kobayashi, K., Morita, S., Sawada, H., Mizuguchi, T., Yamada, K., Nagatsu, I.,Hata, T., Watanabe, Y., Fujita, K., and Nagatsu, T. (1995) J. Biol. Chem.270, 27235–27243

27. Nagatsu, I., Ichinose, H., Sakai, M., Titani, K., Suzuki, M., and Nagatsu, T.(1995) J. Neural Transm. 102, 175–188

28. Blau., N, Barnes, I., and Dhondt, J. L. (1996) J. Inher. Metab. Dis. 19, 8–1429. Zhou, Q.-Y., Qualfe, C. J., and Parmiter, R. D. (1995) Nature 374, 640–64330. Thomas, S. A., Matsumoto, A. M., and Parmiter, R. D. (1995) Nature 374,

643–64631. Blau, N., Kierat, L., Matasovic, A., Leimbacher, W., Heizmann, C. W.,

Guardamagna, O., and Ponzone, A. (1994) Clin. Chim. Acta 226, 159–16932. Huang, P. L., Dawson, T. M., Bredt, D. S., Snyder, S. H., and Fishman, M. C.

(1993) Cell 75, 1273–1286

6-Pyruvoyltetrahydropterin Synthase Knockout Mice 41159

by guest on May 13, 2018

http://ww

w.jbc.org/

Dow

nloaded from

33. Huang, P. L., Huang, Z., Mashino, H., Bloch, K. D., Moskowitz, M. A., Bevan,J. A., and Fishman, M. C. (1995) Nature 377, 239–242

34. MacMicking, J. D., Nathan, C., Hom, G., Chartrain, N., Fletcher, D. S.,Trumbauer, M., Stevens, K., Xie, Q., Sokol, K., Hutchinson, N., Chen, H.,and Mudgett, J. S. (1995) Cell 81, 641–650

35. Pastor, C. M,. Williams, D., Yoneyama, T., Hatakeyama, K., Singleton, S.,Naylor, E., and Billiar, T. R. (1996) J. Biol. Chem. 271, 24534–24538

36. Morin, X., Cremer, H., Hirsh, M.-R., Kapur, R. P., Goridis, C., and Brunet, J.-F.(1997) Neuron 18, 411–423

37. Zetterstrom, R. H., Solomin, L., Jansson, L., Hoffer, B. J., Olson, L., andPerlman, T. (1997) Science 276, 248–250

38. Zhou, Q.-Y., and Palmiter, R. D. (1995) Cell 83, 1197–112039. Tashiro, Y., Sugimoto, T., Hattori, T., Uemura, Y., Nagatsu, I., Kikuchi, H.,

and Mizuno, N. (1989) Neurosci. Lett. 97, 6–1040. Betarbet, R., Turner, R., Chockkan, V., DeLong, M. R., Allers, K. A., Walters,

J., Levey, A. I., and Greenamyre, J. T. (1997) J. Neurosci. 17, 6761–676841. Jaber, M., Dumartin, B., Sagne, C., Haycock, J. W., Roubert, C., Giros, B.,

Bloch, B., and Caron, M. G. (1999) Eur. J. Neurosci. 11, 3499–3511

42. Kettler, R., Bartholini, G., and Pletscher, A. (1974) Nature 249, 476–47843. Levine, R. A., Miller, L. P., and Lovenberg, W. (1981) Science 214, 919–92144. Miwa, S., Watanabe, Y., and Hayaishi, O. (1985) Arch. Biochem. Biophys. 239,

234–24145. Nasrin, S., Ichinose, H., Hidaka, H., and Nagatsu, T. (1994) J. Biochem.

(Tokyo) 116, 393–39846. Flatmark, T., Almås, B., Knappskog, P. M., Berge, S. V., Svebak, R. M.,

Chehin, R., Muga, A., and Martınez, A. (1999) Eur. J. Biochem. 262,840–849

47. Okuno, S., and Fujisawa, H. (1985) J. Biol. Chem. 260, 2633–263548. Rajput, A. H., Gibb, W. R., Zhong, X. H., Shannak, K. S., Kish, S., Chang, L. G.,

and Hornykiewicz, O. (1994) Ann. Neurol. 35, 396–40249. Furukawa, Y., Nygaard, T. G., Gutlich, M., Rajput, A. H., Pifl, C., DiStefano,

L., Chang, L. J., Price, K., Shimadzu, M., Hornykiewicz, O., Haycock, J. W.,and Kish, S. J. (1999) Neurology 53, 1032–1041

50. Kaneda, N., Sasaoka, T., Kobayashi, K., Kiuchi, K., Nagatsu, I., Kurosawa, Y.,Fujita, K., Yokoyama, M., Nomura, T., Katsuki, M., and Nagatsu, T. (1991)Neuron 6, 583–594

6-Pyruvoyltetrahydropterin Synthase Knockout Mice41160

by guest on May 13, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Hiroshi IchinoseTazawa, Hiroaki Shiraishi, Yasumichi Hagino, Toshiharu Nagatsu, Takahide Nomura and Chiho Sumi-Ichinose, Fumi Urano, Risa Kuroda, Tamae Ohye, Masayo Kojima, Masahiro

MICEKNOCKOUTSTUDY FROM 6-PYRUVOYLTETRAHYDROPTERIN SYNTHASE

Catecholamines and Serotonin Are Differently Regulated by Tetrahydrobiopterin: A

doi: 10.1074/jbc.M102237200 originally published online August 21, 20012001, 276:41150-41160.J. Biol. Chem. 

  10.1074/jbc.M102237200Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/276/44/41150.full.html#ref-list-1

This article cites 49 references, 14 of which can be accessed free at

by guest on May 13, 2018

http://ww

w.jbc.org/

Dow

nloaded from