Eur. J . Biochem. 180,205-211 (1989) 0 FEBS 1989

Purification and characterization of 6-pyruvoyl tetrahydropterin synthase from salmon liver Thomas HASLER and Hans-Christoph CURTIUS

Division of Clinical Chemistry, Department of Pediatrics, University of Zurich

(Received August 17/November 17, 1988) - EJB 88 0978

Salmon liver was chosen for the isolation of 6-pyruvoyl tetrahydropterin synthase, one of the enzymes involved in tetrahydrobiopterin biosynthesis. A 9500-fold purification was obtained and the purified enzyme showed two single bands of 16 and 17 kDa on SDS/PAGE. The native enzyme (68 kDa) consists of four subunits and needs free thiol groups for enzymatic activity as was shown by reacting the enzyme with the fluorescent thiol reagent N-(7-dimethylamino-4-methylcoumarinyl)-maleimide. The enzyme is heat-stable up to 80 “C, has an isoelectric point of 6.0-6.3, and a pH optimum at 7.5. The enzyme is Mg2+-dependent and has a Michaelis constant for its substrate dihydroneopterin triphosphate of 2.2 pM. The turnover number of the purified salmon liver enzyme is about 50 times as high as that of the enzyme purified from human liver. It does not bind to the lectin concanav- alin A, indicating that it is free of mannose and glucose residues.

Polyclonal antibodies raised against the purified enzyme in Balb/c mice were able to immunoprecipitate enzyme activity. The same polyclonal serum was not able to immunoprecipitate enzyme activity of human liver 6-pyruvoyl tetrahydropterin synthase, nor was any cross-reaction in ELISA tests seen.

6-Pyruvoyl tetrahydropterin synthase (PPH4S) is one of the enzymes involved in the biosynthesis of tetrahydro- biopterin (BH,) (Scheme l), the cofactor of the aromatic amino acid hydroxylases. Inherited PPH4S deficiency is the most frequent cause of atypical phenylketonuria, where BH4 biosynthesis is impaired [l]. Changes in the metabolism of BH, have also been observed in different neurological dis- orders such as Parkinson’s disease, Alzheimer’s disease, dystonia and others [2 - 51.

The biosynthetic pathway as shown in Scheme 1 is now generally accepted, but the in vivo role of the two-side-chain reductases, sepiapterin reductase and 6-pyruvoyl tetrahy- dropterin reductase (PPH4R), is still not clear. In addition there was only indirect proof for the postulated structure of the biosynthetic intermediate 6-pyruvoyl tetrahydropterin (PPH4) [6 - 101. The production of this intermediate is only possible through enzymatic action of PPH4S on di- hydroneopterin triphosphate (NH2TP).

PPH4S was purified from human liver to apparent homogeneity by Takikawa et al. [ll]. However, he was able to isolate only small amounts of enzyme activity (30 m u ) and

Correspondence to H.-Ch. Curtius, Division of Clinical Chem- istry, Department of Pediatrics, University of Zurich, Steinwies- strasse 75, CH-8032 Zurich, Switzerland

Abbreviations. BH4, tetrahydrobiopterin or 6-(~-erythro-l’,2’-di- hydroxypropyl)-5,6,7,8-tetrahydropterin; NH2TP, dihydroneopterin triphosphdte or 6-(o-erythro-1’,2’,3’-trihydroxypropyl)-7,8-dihy- dropterin 3’-triphosphate; PPH4, 6-pyruvoyl tetrahydropterin; ABTS, 2,2’-azino-bis(3-ethyl-benzthiazoline-6-sulfonic acid); BSA, bovine serum albumin; DACM, N-(7-dimethylamino-4-methyl- coumarinyl) maleimide; PPH4S, 6-pyruvoyl tetrahydropterin synthase; PPH4R, 6-pyruvoyl tetrahydropterin reductase.

Enzymes. GTP cyclohydrolase I (EC 3.5.4.16); sepiapterin re- ductase (EC 1.1 .I .I 53); dihydropteridine reductase (EC 1.6.99.7).

protein (300 pg partially pure enzyme) from 500 g human liver. Attempts to produce monoclonal antibodies against the human enzyme using either partially pure antigen or antigen from SDSjPAGE for immunization have not been successful [12]. In order to produce high amounts of the very labile PPH4, especially necessary for determination of the kinetic parameters of the two reductases, sepiapterin reductase and PPH4R (Scheme l), and for further immunization trials, another enzyme source had to be found.

In a comparative study we determined the activities of the enzymes involved in BH4 biosynthesis in the livers of 11 different species [ 131. We purified PPH4S from salmon liver [14] and used the purified enzyme for the immunization of mice. The results of the enzyme purification and characteriza- tion, as well as an immunization trial, are presented in this paper. Nuclear magnetic resonance data of the enzymatic conversion of NH2TP to PPH4, and fast-atom-bombardment/ secondary-ion mass spectrometry (FAB/MS) on PPH4 using the salmon enzyme have been published elsewhere [15, 161. Determination of K, and V,,, values for sepiapterin re- ductase and PPH4R are now underway and will be published later.

MATERIALS AND METHODS

Materials

After catch salmon livers were kept for 3-4 days at O’C, and afterwards at -70°C. NH2TP was prepared en- zymatically from GTP by immobilized GTP cyclohydrolase I isolated from Escherichia coli. BH, was from Dr B. Schircks Laboratories (Jona, Switzerland). Concanavalin A (type IV), bovine serum albumin (BSA) (A-7030), chymotrypsin (type I-S), and horseradish peroxidase (type VI) were from Sigma

PPH4S assay For the measurement of PPH4S activities, the unstable

product of this enzymatic reaction was immediately reduced by an excess of sepiapterin reductase to the stable BH4, which was quantified using HPLC with electrochemical detection. The incubation mixture contained 10 pl enzyme, 100 mM Tris/HCl, pH 7.4, 20 pM NH,TP, 10 mM dithioerythritol, 8 mM MgCI2, 1 mM NADPH, and 3 mU sepiapterin re- ductase in a final volume of 125 pl. After incubation at 37°C for 15 min in the dark, the reaction was stopped by addition of 25 pl 0.2 M EDTA. 10 pl of the supernatant after centrifugation was analyzed by HPLC. I U enzyme activity was defined as the amount catalyzing the production of 1 pmol BH4/min.

206

r

bn OH

L

p3

PPHq

NRDPH 7 / NAOPH \

Scheme 1. Proposedpathway for the biosynthesis of tetrahydrobiopterin (BH4) ,from guanosine triphosphate (GTP). GTP is converted to dihydroneopterin triphosphate (NH,TP) by GTP cyclohydrolase I (GTPCH). Triphosphate is then eliminated together with an intramo- lecular shift of a double bond by 6-pyruvoyl tetrahydropterin synthase (PPH,S) to yield 6-pyruvoyl tetrahydropterin (PPH4). Reduction of the side-chain carbonyl groups by 6-pyruvoyl tetrahydropterin re- ductase (PPH4R) and sepiapterin reductase (SR) or sepiapterin re- ductase alone finally leads to tetrahydrobiopterin

(St Louis, USA). NADPH and 2,2'-azino-bis(3-ethylbenz- thiazoline-6-sulfonic acid) (ABTS) were purchased from Boehringer (Mannheim, FRG). N-(7-Dimethylamino-4- methylcoumariny1)-maleimide (DACM) was from Fluka (Buchs, Switzerland). An Immun-Blot assay kit, goat anti- [mouse IgG (H + L)] - horseradish-peroxidase conjugate, a protein assay kit, and Tween-20 were from Bio-Rad (Richmond, USA). Staphylococcus nureus V8 protease was from Miles Scientific (Naperville, USA). ELISA Fastbinder plates were obtained from Costar (Cambridge, USA). Fractogel TSK Butyl-650s was a product of Merck (Darm- stadt, FRG) and Ultrogel AcA 44 was from LKB (Bromma, Sweden). The FPLC ion-exchange column Mono Q was from Pharmacia (Uppsala, Sweden) and an HPLC hydroxyapatite HPHT column from Bio-Rad. S. aureus was a gift from Dr J. Fluckiger (Department of Biochemistry, University of Ziirich, Switzerland). All other chemicals were obtained from com- mercial sources.

Purijication of PPH4S All steps were carried out in a coldroom at 4°C. Homogenization. 300-g aliquots of salmon liver were

thawed out and homogenized in 900ml 50mM potassium phosphate buffer, pH 7.0. The homogenate was then centri- fuged for 60 rnin at 27000 x g. The supernatant was filtered through cheese cloth to remove fat.

Ammonium sulfate fractionation. PPH4S was precipitated between 40 - 55% saturation and then centrifuged for 30 min at 27000 x g. The precipitate was dissolved with 150 ml 50 mM potassium phosphate buffer, pH 7.0, and dialyzed overnight against 5 1 of the same buffer.

Heut treatment. The enzyme solution in the dialysis bag was exposed to 80 "C for 5 rnin in a beaker containing the same dialysis buffer with the appropriate temperature. Immediately afterwards, the dialysis bag was moved to a reservoir of ice- cold buffer. The thick precipitate was then removed by centrifugation at 48000 x g for 15 min.

Hydrophobic interaction chromatography. To the super- natant of the previous step solid ammonium sulfate was added up to 30% saturation. The pH was adjusted to 7.0 by addition of 1 M NaOH and the enzyme solution was applied to a column of Fractogel TSK Butyl-650s (2.6 x 26 cm) equili- brated with 50 mM potassium phosphate buffer, pH 7.0, and ammonium sulfate at 30% saturation. The column was then washed with 40 ml of the same buffer before a gradient from 30% to 0% ammonium sulfate saturation was started. The gradient volume was 620 ml. Active fractions were combined and concentrated to less than 15 ml by Amicon Diaflo PM-10 ultrafiltration. During this concentration step, ammonium sulfate was removed and replaced by 50 mM potassium phos- phate buffer, pH 7.0.

Gel filtration. Approximately 5 mg BSA was added to the concentrated solution of the last step. Gel filtration was carried out on an Ultrogel AcA 44 column (2.6 x 90 cm) equil- ibrated with 50 mM potassium phosphate buffer, pH 7.0. The active fractions eluting together with BSA were combined and concentrated by Amicon Diaflo YM-10 ultrafiltration to a final volume of less than 1 ml. During this concentration step the phosphate buffer was replaced by 20 mM triethanolamine/ acetate, pH 7.5.

Ion-exchange FPLC. The concentrated eluate from gel fil- tration was applied to a Mono Q H R 5 /5 ion-exchange column equilibrated with 20 mM triethanolamine/acetate, pH 7.5, containing 0.05% Tween-20. The enzyme was eluted using a linear gradient from 0 to 0.5 M sodium acetate in the same buffer. The gradient volume was 15 ml. Protein elution was monitored in the ultraviolet at 280 nm and protein peaks were collected.

207

Table 1. Purijication of salmon liver PPHJ 1 U enzyme activity catalyzes the production of 1 pmol PPH4/min from NHzTP at 37°C. Only 66% and 89% of the enzyme was applied to the last two chromatographic steps (steps 6 and 7, respectively)

Step Total Total Specific Recovery Purification activity protein activity

mU 1. Homogenate 38 100 2. Ammonium sulfate (40- 55%) 27 500 3. Heat treatment (80°C) 19700 4. Hydrophobic interaction chromatography 14600 5. Gel filtration 3 3 300 6. Ion exchange 6100 7. Hydroxyapatite 2 700

mg 79 000 25 000

1700 78 19 1.6 0.62

mU/mg 0.482 1.10

11.6 187 700

3800 4600

%

100 72 52 38 35 24 12

-fold 1 .o 2.3

24 389

1450 7880 9500

Hydroxyapatite chromatography on HPLC. The buffer of the enzyme solution was exchanged with Amicon Diaflo YM-10 ultrafiltration to 10 mM potassium phosphate buffer, pH 6.8, containing 10 pM CaCI, and 0.05% Tween-20. The hydroxyapatite HPHT column was equilibrated with the same buffer and the enzyme applied in a volume of about 1 ml. Elution was carried out using a linear gradient of 10 - 350 mM phosphate at constant pH in a total volume of 15 nil.

Glycoprotein detection Glycoprotein was detected on nitrocellulose after Western

blotting of proteins separated by SDS/PAGE. The detection method using concanavalin A and horseradish peroxidase was performed as published by Hawkes in 1982 [17].

Reaction of enzyme -SH groups with DACM The reaction buffer was a 50 mM potassium phosphate

buffer, pH 7.0, containing 0.05% Tween-20. 10 pg purified PPH4S was reduced by 10 mM dithioerythritol for 5 min at 37 "C in 135 p120 mM triethanolamine/acetate, pH 7.5. In the control experiment, dithioerythritol was omitted. Dithio- erythritol was removed by desalting on a small Sephadex (3-25 column (0.5 x 6 cm) equilibrated with reaction buffer. From the eluting protein fraction (250 p1) an aliquot was taken for the enzyme activity assay. 10 mM DACM was added to a final concentration of 50 pM and the samples were incu- bated for 1 h at room temperature. The reaction was stopped by addition of 25 pl 100 mM dithioerythritol and an aliquot was taken for the enzyme assay. DACM-dithioerythritol and unreacted dithioerythritol were removed by another Sephadex (3-25 step and the eluting protein (300 pl) was submitted to protein determination (Bio-Rad protein assay) and DACM incorporation measurement. For the latter, 50 pl of eluate was added to 2.5 ml 50 mM Tris/HCl buffer, pH 9.0, containing 1 YO SDS. Fluorescence (398/465 nm) was measured after 30 min and 24 h at 37°C in the dark. Fluorescence yield was calibrated using DACM-ovalbumin and comparing A380 in 50 mM potassium phosphate buffer, pH 7.0, with its fluores- cence in 50 mM Tris/HCl, pH 9.0, containing 1% SDS (cjSO = 19800 M- ' cm-I).

Peptide mapping The peptide mapping on SDSjPAGE was carried out as

described by Cleveland et al. [18]. A 12.5-20% gradient gel was used and the two enzyme bands (16 and 17 kDa) were separately digested with either 50 pg/ml V8 protease or 50 pg/ ml chymotrypsin on the stacking gel.

Antibody production and screening

Immunization of Balblc mice. For the first immunization, 10 pg purified PPH4S in Freund's complete adjuvant were subcutaneously administered to two female Balb/c mice. After 21 and 42 days, two further subcutaneous administrations of 20 pg PPH4S were given in Freund's incomplete adjuvant. The mice were boosted on day 66 by injection of 40 pg PPH4S in 0.9% NaCl intraperitoneally. 7 days later, serum was taken from the tail vein and submitted to the different immuno- assays.

ELISA. Purified human or salmon PPH4S (300 ng/well) was applied to Costar Fastbinder microtiter plates in 0.1 M NaHC03, pH 9.6. The plates were stored for at least 24 h at 4°C in a humid chamber. After washing twice with 10 mM phosphate buffer, pH 7.2, containing 0.8% NaCl and 0.05% Tween-20 (NaC1/Pi/Tween), the sera of the immunized mice and of a control mouse were serially diluted in NaC1/Pi/Tween containing 0.1 Yn BSA. After a 3-h incubation period, the wells were washed three times with NaC1/Pi/Tween. The wells were then incubated with the second antibody, goat anti-[mouse IgG (H)] - peroxidase conjugate (diluted 1 : 2000) for an ad- ditional 3 h period, again washed with NaC1/Pi/Tween (three times), and finally incubated with ABTS substrate in 0.1 M NaH2P04 containing 0.012% H202.

Dot blot assay. Partially purified human and salmon PPH4S (0.5 pg) was dotted on nitrocellulose paper strips. After drying, free binding sites were blocked by immersing the papers in 10 mM Tris, pH 7.4, with 0.9% NaCl (Tris/ NaCI) containing 2% BSA for 1 h. The papers were then washed three times with Tris/NaCl containing 0.03% Tween- 20 (Tris/NaCl/Tween) and incubated with the diluted sera (1:5000 in Tris/NaCl containing 2% BSA) for 2 h . After washing three times with Tris/NaCI, the second antibody, goat anti-[mouse IgG (H + L)] - peroxidase conjugate (1 : 3000 in Tris/NaCl containing 2% BSA), was added to the papers for 1.5 h. Following extensive washing (once with Tris/NaCl/ Tween, three times with Tris/NaCl), antibody binding was visualized using 4-chloro-1 -naphthol in Tris/NaCl containing

Immunoblot assay. Electrophoretic transfer of the sepa- rated proteins from SDSjPAGE to nitrocellulose was per- formed according to the manufacturer's instructions (Bio- Rad). Western blots were then tested for antibody binding as described for the dot blot assay.

Immunoprecipitation. Partially purified human and salmon PPH4S containing less than 20 ng enzyme was incubated in diluted mouse sera (1 :300 in NaCl/Pi) overnight at 4°C. In a

0.015% H2Oz.

208

Fig. 1.12.5% SDSjPAGEofsulmon liver PPH4S. (A) Molecular mass standards; (B) and (C) are two different fractions of the cnzyme after hydroxyapatite chromatography, representing different peaks in the chromatogram (Fig. 2)

5 0.05 n Q

I I

Time (mn) Fig. 2. Elution profile of PPH,S,from hydroxyupatite H P H T column. The gradient was started after 2 min and stopped after 17 min. The flow rate was 1 miimin

parallel experiment, S. uureus was also added, and the samples were slowly agitated during the incubation period. Enzyme activity in the supernatant was then measured after centrifuga- tion.

RESULTS AND DISCUSSION

Purification of P P H J

Starting to purify PPH4S from salmon liver, we tried to follow the purification scheme of Takikawa et al. for the human liver enzyme [ll]. We soon had to recognize that

Fig. 3. Peptide mapping of the two bands of pi~rijied salmon liver PPH4S. Gradient SDSjPAGE from 12.5 - 20% polyacrylamide. (A-C) Addition of 500 ng V8 protease: (A) alone, (B) with 17-kDa band and (C) with 16-kDa band of PPH4S; (D) 17-kDa band and (E) 16-kDa band without addition of protease; (F-H) addition of 500 ng chyrnotrypsin: (H) alone, (F) with 17-kDa band and (G) with 16-kDa band

during their purification the two enzymes showed some important differences. A dramatic loss of activity in the last purification steps caused particularly poor recovery rates of the salmon enzyme. The influence of different additives on enzyme stability was tested. BSA (0.1 mg/ml) and the nonionic detergent Tween-20 (0.05%) were able to stabilize the enzyme, especially during chromatographic or freezing steps. Dithio- erythritol(1 mM), EDTA (0.1 mM), MgC12 (1 mM), and KCl (0.1 5 M) had no influence, whereas glycerol (50%) partially stabilized the enzyme. We attribute these activity losses to nonspecific adsorption of the enzyme to plastic, glass or column material, as they only occur in dilute solutions of highly purified enzyme. Therefore, in the last three purifi- cation steps either BSA (gel filtration) or Tween-20 (ion-ex- change and hydroxyapatite chromatography) were added.

Table 1 presents the final purification scheme starting from 800 g salmon liver. Hydrophobic interaction chromatog- raphy was introduced as a powerful additional step. Ion- exchange and hydroxyapatite chromatography were run on FPLC columns. The overall purification factor was 9500, with 12% recovery, and 620 pg PPH4S having a total activity of 2.7 U was isolated. The purified enzyme showed two bands (16 and 17 kDa) on SDSjPAGE without other contaminants (Fig. 1). The elution profile of the last chromatographic step (Fig. 2 ) indicated a heterogeneity of the purified enzyme. However, the different elution peaks showed identical SDS/ PAGE patterns. The entire procedure was repeated twice and in one case the last step was omitted (10000-fold purification

209

Fig. 4. Antibody titrution of sera with ELISA. (U) Uncoated; (H) coated with 3 pg/ml partially purified human PPH4S; (S) coated with 3 pg/ml partially purified salmon PPH4S; (N) no serum added; (C) incubated with serum of a control mouse; (A) incubated with serum of mouse A; (B) incubated with serum of mouse B; (1 :25- 1 :48125) serial dilutions of the sera

with 16% recovery). Attempts to separate the two bands using isoelectric focussing and reversed-phase HPLC after complete reduction of disulfide bonds failed (data not shown).

Peptide mapping

In order to prove our assumption of a close relationship between the two bands, we decided to carry out a peptide mapping on SDSjPAGE according to Cleveland et al. [18]. The digestion with V8 protease and chymotrypsin gave differ- ent fragmentation patterns, but in both cases the fragments of the 16 and 17 kDa bands were identical (Fig. 3). Most probably the 16-kDa band is a proteolytic degradation prod- uct of the 17-kDa band. We also found that the ratio of the two bands varies between liver batches. We were not able to prevent proteolysis, as the conditions of processing and transportation of these livers were beyond our control. How- ever, the specific activities of PPH4S were fairly constant and the ratio of the two bands did not affect enzyme activity.

Characterization of salmon liver PPH4S

The molecular mass of the salmon liver PPH4S as deter- mined by gel filtration on AcA 44 was found to be 68 kDa, a value which is substantially lower than the 83 kDa published for the human enzyme by Takikawa et al. [I 11. However, this result comes close to the 63 kDa reported for human PPH4S by Heintel et al. [19]. The subunit molecular mass was deter- mined on 12.5% SDS/PAGE. The two bands with 16 and 17 kDa are in accordance with a four-subunit structure of the enzyme. Isoelectric focussing on Sephadex G-200 superfine using Servalyt T4-9 [20] revealed an isoelectric point of 6.0-6.3. This value is much higher than the 4.2-4.4 of the human liver enzyme (Heintel et al. [19]). The activity optimum was at pH 7.5 and the enzyme needs Mgz+ ions for catalytic activity. The K, value for the substrate NH2TP was measured

Fig. 5. Dat blot assay of sera. 0.5 kg partially purified human (H) and salmon (S) PPH4S were applied to nitrocellulose. The spots were then incubated in diluted sera of a control mouse (C) and of the two immunized mice (A, B)

Table 2. Inzmunoprecipitation of salmon liver PPHJ vvith polyclonul mouse antisera The enzyme (15 ng) was incubated overnight in diluted sera (1 :300) or buffer with or without addition of S. aureus. Results are expressed as a percentage of enzyme activity compared to untreated PPH4S

Sample Activity after incubation with

buffer control anti- anti- serum serum 1 serum 2

YU

Sera only 300 100 62 38 Sera and S. aureus 90 110 10 5

using the assay described in Materials and Methods. We de- cided to use this assay procedure, where the enzymatic product PPH4 is immediately converted to BH4 by the action of sepiapterin reductase, instead of the direct quantification of PPH4 by HPLC. The instability of PPH4, especially at low concentrations, might influence the apparent K, value. Fur- thermore, PPH4 elutes as a broad peak from HPLC reversed- phase columns, and accurate quantification of low amounts is therefore not ensured. We obtained a K, value of 2.2 pM and a maximal turnover number of 4.6 U/mg purified PPH4S. The specific activity of the pure salmon enzyme is about 50 times higher and the K, value for NH2TP 5 times lower than the corresponding values of the purified human liver enzyme [ll]. The salmon enzyme is also very heat-stable: 75% of the original activity were recovered after 5 min at 80°C.

Takikawa et al. [21] suspected that the human enzyme is a glycoprotein. They performed a periodic acid/Schiff stain on a nondenaturating Davis polyacrylamide gel and found a weak band at the same place where the PPH4S activity was

210

measured. We used the concanavalin A/horseradish peroxi- dase method described by Hawkes [17], as the sensitivity of the periodic acid/Schiff stain is very low. Concanavalin A is a lectin with a very broad binding specificity for glycoproteins (mannose, glucose, N-acetylglucosamine, and sorbose resi- dues [22]). On Western blots after SDSjPAGE we were able to show that concanavalin A was not binding to the salmon enzyme.

Loss of catalytic activity due to the formation of intramolecular disuvide bonds

The literature yields controversial information about the involvement of essential thiol groups in the catalytic activity of PPH4S. Milstien and Kaufman, working with a rat enzyme preparation, reported that the addition of a disulfide-reducing agent (dithioerythritol) is not necessary for catalytic activity providing the assay is carried out under absolutely anaerobic conditions [23]. Takikawa et al. [I 11, working with purified human liver PPH4S, showed that preincubation with dithio- erythritol resulted in maximal catalytic activity, and therefore assumed that intramolecular disulfide bonds had to be cleaved to activate the enzyme. We obtained the same results as Takikawa by preincubating the salmon liver enzyme with dithioerythritol. Furthermore, we reacted the preincubated enzyme with the fluorescing - SH reagent, N-(7-dimethyl- amino-4-methyl coumarinyl) maleimide (DACM) [24], and found incorporation of 0.8 mol DACM/mol enzyme. As a control, enzyme without dithioerythritol preincubation was exposed to DACM under the Same conditions and only 0.2 mol DACM/mol enzyme was incorporated. The incorpo- ration of roughly one molecule DACM per molecule PPH4S resulted in a 50% reduction of catalytic activity. Catalytic activity was reduced to less than 5% by using a 10-fold higher concentration of DACM. These results demonstrate the influ- ence of essential -SH groups on the catalytic activity of this enzyme. It remains to be shown whether this influence represents a direct involvement in the enzyme mechanism or an allosteric effect of changing the enzyme’s three-dimensional structure.

Production ojantibodies against PPH4S The immunization of Balb/c mice with either purified hu-

man liver enzyme or the same enzyme cut from SDSjPAGE did not result in a specific antibody response against human

In this paper we present the results of an immunization trial with the purified salmon liver PPH4S. The sera of two Balb/c mice, immunized with this enzyme as described in Materials and Methods, were assayed with ELISA, dot-blot, Western blot, and immunoprecipitation assays. Both sera had high antibody titers (1 : 15000 and 1 :48000), as estimated with ELISA (Fig. 4). There was no cross-reaction with the human enzyme. The result in the dot-blot assay was essentially the same, except that there was a weak cross-reaction of the serum having the higher antibody titer with the human antigen (Fig. 5). The same antibodies did not react on Western blots (not shown), indicating that the denaturation of the enzyme during SDS/PAGE results in a loss of the antigenic deter- minant(s). Immunoprecipitation is still the most specific assay for detection of antibodies against an enzyme. In this assay both sera were able to immunoprecipitate salmon PPH4S activity (Table 2). The fact that S. aureus enhances immuno- precipitation indicates that antibodies of the IgG type are

PPH4S [12].

predominantly responsible for the antibody response. The sera were not able to immunoprecipitate human PPH4S ac- tivity (data not shown). Although monoclonal antibodies raised against human GTP cyclohydrolase I were able to cross-react with the corresponding enzyme from E. coli [25], such a cross-reaction between PPH4S from salmon and that from human liver was not obtained. It seems that the enzyme regions preserved during evolution are not or only very weakly immunogenic in mice.

We are grateful to U. Redweik for the preparation of dihy- droneopterin triphosphate and to M. Killen for reading the manu- script. This work was supported by the Swiss National Science Foun- dation, project no. 3.613-0.84 and 3.395-0.86.

REFERENCES 1 .

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17. 18.

19.

Niederwieser, A., Ponzone, A. & Curtius, H.-Ch. (1985) J . Inherit. Metab. Dis. 8 (Suppl. l ) , 28-33.

Lovenberg, W., Levine, R. A., Robinson, D. S., Ebert, M., Williams, A. C. & Calne, D. B. (1979) Science (Wash. DC)

Kay, A. D., Milstien, S., Kaufman, S., Creasey, H., Haxby, J. V., Cutler, N. R. & Rapoport, S. J. (1986) Arch. Neurol. 43, 996- 999.

LeWitt, P., Miller, L., Levine, R., Lovenberg, W., Newman, R., Papavasiliou, A,, Rayes, A,, Eldridge, R. & Burns, R. (1986) Neurology 36, 760 - 764.

Levinc, R. A. (1987) in Unconjugatedpterins and related biogenic amines (Curtius, H.-Ch., Blau, N. & Levine, R. A,, eds) pp. 315-326, Walter de Gruyter, Berlin.

Switchenko, A. C. & Brown, G. M. (1985) J . Bid. Chem. 260,

Curtius, H.-Ch., Heintel, D., Ghisla, S., Kuster, T., Leimbacher, W. & Niederwieser, A. (1985) Eur. J . Biochem. 148, 413- 419.

Masada, M., Akino, M., Sueoka, T. & Katoh, S. (1985) Biochim. Biophys. Acta 840, 235-244.

Smith, G. K. & Nichol, C. A. (1986) J . Bid. Chern. 261, 2725- 2737.

Milstien, S. & Kaufman, S. (1985) Biochem. Biophys. Rrs. Commun. 128, 1099-1107.

Takikawa, S., Curtius, H.-Ch., Redweik, U., Leimbachcr, W. & Ghisla, S. (1986) Eur. J. Biochem. 161, 295-302.

Hasler, Th., Schoedon, G., Leimbacher, W. & Curtius, H.-Ch. (1989) in Pteridines and related biogenic amines (Levine, R. et al., eds) Lakeshore Publishing Company, Grosse Pointe, in the prcss.

Haslcr, Th. & Niederwiescr, A. (1986) in Chemistry and biology ofpteridines 1986 (Cooper, B. A. & Whitehead, V. M., eds) pp. 319-322, Walter de Gruyter, Bcrlin.

Hasler, Th., Niederwieser, A. & Curtius, H.-Ch. (1987) in Unconjugated pterins and biogenic amines (Curtius, H.-Ch., Blau, N. & Levine, R. A,, eds) pp. 81 -87, Walter de Gruyter, Berlin.

Ghisla, S., Tdkikawa, S., Steinerstauch, P., Hasler, Th. & Curtius, H.-Ch. (1986) in Chemistry und biology of pteridines 1986 (Cooper, B. A. & Whitehead, B. M., eds) pp. 299 - 303, Waltcr de Gruyter, Berlin.

Richter, W. J., Raschdorf, F., Dahinden, R., Kuster, Th., Niederwieser, A., Leimbacher, W. & Curtius, H.-Ch. (1987) in Unconjugated pterins and related biogenic amines (Curtius, H.- Ch., Blau, N. & Levine, R. A,, eds) pp. 39-46. Walter de Gruyter, Berlin.

204,624 - 626.

2945 - 2951.

Hawkes, R. (1982) Anal. Biochem. 123, 143-146. Cleveland, D. W., Fischer, S. G., Kirschner, M. W. & Laemmli,

U. K.(1976)J. B i d . Chenz.252, 1102-1106. Heintel, D., Lcimbacher, W., Redweik, U., Zagalak, B. & Curtius,

H.-Ch. (1985) Biochem. Biophys. Res. Commun. 127, 213- 219.

21 1

20. Radola, B. J . (1974) Biochirn. Biophys. Acta 386, 181 -195. 24. Yamamoto, K., Sekine, T. & Kanaoka, Y. (1977) Anal. Biochem.

25. Schoedon, G., Niederwieser, A. & Curtius, H.-Ch. (1987) in Unconjuguted pterins and related hiogenic amines (Curtius, H.- Ch., Blau, N. & Levine, R. A., eds) pp. 99 - 104, Walter de Gruyter, Berlin.

21. Takikawa, S., Curtius, H.-Ch., Redweik, U. & Ghisla, S. (1986) 79,83 -94. Biochem. Biophys. Res. Commun. 134, 646 -651.

22. Kristiansen, T. (1974) Methods Enzymol. 34, 332-341. 23. Milstien, S. & Kaufman, S. (1986) in Chemistry and biology of

pteridines 1986 (Cooper, B. A. & Whitehead, V. M., eds) pp. 169-181, Walter de Gruyter, Berlin.