THE JOURNAL OF BIOLOGICAL CHEMISTRY CC) 19% by The American Society of Biological Chemists, Inc

Val. 259, No. 2, Issue of November 25. pp. 14121-14127,1984 Printed in U.S.A.

Purification and Properties of the Enzymes from Drosophila melanogaster That Catalyze the Conversion of Dihydroneopterin Triphosphate to the Pyrimidodiazepine Precursor of the Drosopterins"

(Received for publication, May 7, 1984)

Gregory J. WiederrechtS and Gene M. Brown From the Department of Biology, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139

The enzyme system responsible for the conversion of 2-amino-4-oxo-6-(~-erythro-1',2',3'-trihydroxypro- pyl)-7,8-dihydropteridine triphosphate (dihydroneop- terin triphosphate or H2-NTP) to 2-amino-4-oxo-6- acetyl-7,8-dihydro-3H,SH-pyrimido~4,5-~~-~1,4]di- azepine (pyrimidodiazepine or PDA), a precursor to the red eye pigments, the drosopterins, has been puri- fied from the heads of Drosophila melanogaster. The PDA-synthesizing system consists of two components, a heat-stable enzyme and a heat-labile enzyme. The heat-stable enzyme can be replaced by sepiapterin syn- thase A, a previously purified enzyme required for the Mg2+-dependent conversion of Hz-NTP to an unstable compound that appears to be 6-pyruvoyltetrahydrop- terin (pyruvoyl-H4-pterin). The heat-labile enzyme, purified to near-homogeneity and termed PDA syn- thase (M, = 48,000), catalyzes the conversion of py- ruvoyl-H4-pterin to PDA in a reaction requiring the presence of reduced glutathione. Because PDA is two electrons more reduced than pyruvoyl-H4-pterin, the reducing power required for this transformation is probably supplied by glutathione. The PDA-synthesiz- ing system requires the presence of another thiol-con- taining compound such as 2-mercaptoethanol when in- cubations are performed aerobically. However, under anaerobic incubation conditions 2-mercaptoethanol is no longer required. Evidence is presented to indicate that the Drosophila eye color mutant, sepia, is missing PDA synthase.

Five red eye pigments, the drosopterins, occur in the eyes of Drosophila rnelanogaster. The drosopterins contain a pterin ring within their structures (1-5) and at least two of the drosopterins also contain a pyrimidodiazepine ring as part of their structures (6, 7). H,-NTP' (dihydroneopterin triphos-

*This work was supported in part by Research Grant 2-R01- AM03442-24 from the National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduer- tisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

$Supported by National Institutes of Health Training Grant GM07287 and by a Predoctoral Fellowship from Johnson and John- son. Present address, Division of Chemistry and Chemical Engineer- ing, 147-75, California Institute of Technology, Pasadena, CA 91125. ' The abbreviations and trivial names used are: Hz-NTP (dihydro-

neopterin triphosphate), 2-amino-4-0~0-6-(D-erythro-l',2',3'-trihy- droxypropy1)-7&dihydropteridine triphosphate; PDA, 2-amino-4- oxo-6-acety1-7,8-dihydro-3H,9H-pyrimido[4,5-b]-[1,4]diazepine; py- ruvoyl-H4-pterin (pyruvoyltetrahydropterin), 2-amino-4-oxo-6-pyru- voyl-5,6,7,8-tetrahydropteridine; sepiapterin, 2-amino-4-oxo-6-lac- toyl-7,8-dihydropteridine; tetrahydrobiopterin, 2-amino-4-oxo-6-(~- erythro-l',2'-dihydroxypropyl)pteridine; Pipes, 1,4-piperazinedietha- nesulfonic acid GSH, glutathione.

phate) is a precursor of the drosopterins (8-10) and is prob- ably required for the formation of the pterin portion of the drosopterin ring system. A compound that has been identified (9, 11) as PDA is also required for the enzymatic synthesis of the drosopterins and is probably the precursor of the pyrimi- dodiazepine portion of the drosopterin ring system.

Dihydroneopterin triphosphate is converted to PDA, in yields as high as 40%, by enzyme fractions prepared from Drosophila heads (12). Reduced GSH, another thiol-contain- ing compound such as 2-mercaptoethanol, and M&+ are also required for this conversion. Based upon an analysis of various PDA-deficient eye color mutants of Drosophila, evidence was presented (12) to indicate that three mutants, purple, sepia, and clot, are deficient in the conversion of H2-NTP to PDA.

In this paper we report on the purification and properties of the PDA-synthesizing enzyme system. Two enzymes, se- piapterin synthase A (or Enzyme A), which is present in limited quantities in the mutant, purple (12, 13), and PDA synthase, an enzyme missing in sepia flies, are clearly required for the conversion of H,-NTP to PDA. Evidence is presented to support a reaction pathway in which the product of action of Enzyme A on H,-NTP (tentatively identified as pyruvoyl- H4-pterin (14) is converted to PDA in the presence of GSH and PDA synthase.

EXPERIMENTAL PROCEDURES

MateriuLs-[U-"C]GTP was obtained from Amersham-Searle; Pipes from Calbiochem-Behring; Ultrogel AcA 44 from LKB; reduced glutathione, standard proteins used for molecular weight determina- tions by nondenaturing and denaturing gel electrophoresis, and aga- rose/hexane/guanosine 5"triphosphate for the purification of GTP cyclohydrolase I from Sigma; ultrapure ammonium sulfate from Schwarz-Mann; DEAE-Sepharose, phenyl-Sephsrose, Polybuffer Ex- changer 94, Polybuffer 74, and standard proteins used in the deter- mination of molecular weights by molecular sieve chromatography from Pharmacia; Wheaton micro vials, serum stoppers, and serum bottles for anaerobic incubations from Aldrich; 3" paper for de- scending chromatography from Whatman; and Affi-Gel blue and materials for electrophoresis from Bio-Rad. [U-"C]HZ-NTP was pre- pared enzymatically from [U-"CIGTP as described earlier (9).

Methods-Breeding populations of wild-type D. melanogaster, strain Oregon-R, were maintained as described previously (12). Egg collection (15), the growth and collection of pupae (9), and the collection and storage of young adult flies (12) were performed as previously described.

The preparation of the heat-treated (75 "C for 2 min) ammonium sulfate fraction required for the biosynthesis of drosopterin, isodro- sopterin, and neodrosopterin has been described (9). The preparation of crude extracts and of ammonium sulfate fractions containing PDA- synthesizing activity and the preparation of boiled extracts of Dro- sophila heads have also been described (12).

The composition of the reaction mixtures prepared to measure enzymatic production of PDA from Hz-NTP varied, depending upon the experiment, and will be described as each experiment is presented. Except where noted, incubations were performed at room temperature

14121

14122 Pyrirnidodiazepine Synthase from D. melanogaster in the dark for 2 h. After incubation, the reaction mixtures were placed in a boiling water bath for 10 min and then subjected to centrifugation to remove precipitated protein. The high-performance liquid chromatography assay previously described (12) was used to measure PDA synthesis except when large numbers of fractions were assayed during enzyme purification. For the latter purpose, the de- scending chromatography assay described by Krivi and Brown (13) was used, slightly modified as follows. A portion (0.15 ml) of an incubated reaction mixture was removed and applied to a strip of Whatman 3MM paper (3 X 57 cm). An amount (0.1 ml) of boiled extract of Drosophila heads was also applied to the strip to provide a carrier quantity of PDA. After development, the position of PDA was determined by placing a UV lamp beneath the strip and locating the PDA zone, which quenches UV light. The RF of PDA in this system is 0.56. The PDA zone was cut from the paper strip and the paper was cut into small pieces, placed into a scintillation vial, and gently shaken with 4 ml of 2% NH40H to elute PDA from the paper. After addition of 10 ml of scintillation fluid to each vial, the contents were mixed, and the amount of radioactivity was determined in a Packard 460 refrigerated scintillation spectrometer.

When performing PDA synthase assays, it was not possible to use the likely substrate, pyruvoyl-H4-pterin, due to the extreme instability of this compound. Therefore, the substrate used in incubation mix- tures containing PDA-synthesizing activity was Hz-NTP. A source of Enzyme A was always added to the incubation mixtures to convert H,-NTP to the substrate used directly by PDA synthase. Because under these conditions it was not possible to quantitate the exact amount of pyruvoyl-H,-pterin, an arbitrary definition of "units of PDA synthase activity" was devised. One unit of PDA synthase activity is defined as that amount of enzyme required to make 1 nmol of PDA from H,-NTP in a 2-h incubation period at 22 "C in the presence of an excess of Enzyme A.

Electrophoresis on polyacrylamide gels was performed as described by Davis (16) and gel electrophoresis in the presence of sodium dodecyl sulfate was performed as described by Laemmli (17). Anaer- obic incubations were performed as described (18). Protein solutions were concentrated with the use of an Amicon pressure cell equipped with a PM-10 membrane. Protein concentrations were determined by the method of Bradford (19) with the use of bovine 7-globulin as the protein standard.

RESULTS

Purification of PDA Synthase Previous results from this laboratory have suggested that

a t least three enzymes may be involved in the conversion of H2-NTP to PDA: Enzyme A, which is present in limiting quantities in the mutant, purple; an enzyme present in lim- iting quantities in the mutant, clot; and an enzyme which is missing in the mutant, sepia. Theoretically, activity of any one of these enzymes could be assessed by incubating frac- tionated protein from wild-type flies with a crude extract prepared from the mutant deficient in the enzyme being assayed. However, Enzyme A has already been partially pu- rified in this laboratory (14) and clot is so "leaky" that any purification scheme involving complementation with clot ex- tracts would be plagued by a high background level of PDA synthesis. The decision was made to purify the enzyme miss- ing in sepia flies because sepia is a "null" mutant and comple- mentation with extracts prepared from sepia flies would not have a high background level of PDA synthesis.

Preliminary experiments demonstrated that when an am- monium sulfate fraction having PDA-synthesizing activity was prepared from the heads of wild-type flies and then fractionated by gel filtration on an Ultrogel AcA 44 column, no single eluant fraction could support the synthesis of PDA. However, when crude extracts prepared from the heads of sepia flies were also included in reaction mixtures, a single peak of PDA-synthesizing activity from the column was evi- dent. This activity peak was interpreted to be due to an enzyme present in wild-type flies, but missing in the sepia mutant.

Since the sepia mutant does not grow as fast or to the same population density as the wild-type flies, it would have been somewhat inconvenient to maintain the large population of sepia flies for the preparation of enough sepia crude extract needed for the assays during purification of the enzyme miss- ing in sepia flies. This problem was solved with the observa- tion that either purified Enzyme A or a protein fraction (that protein precipitating between 40 and 70% ammonium sulfate saturation) prepared from a heat-treated (75 "C for 2 min) crude extract of wild-type fly heads (and which contained Enzyme A activity) could substitute for the crude extract of sepia flies. The heat-labile enzyme missing in sepia flies is the one we have designated as PDA synthase. All column chromatography procedures described below were performed at 4 "C. Incubation mixtures (total volume, 0.2 ml) prepared to assess enzymatic activity in column fractions contained 0.05 ml of the fraction being assayed, 10.4 HM [U-'4C]H2-NTP (90 mCi/mmol), 2.5 mM MgC12, 5 mM GSH, 0.11 mg of the heat-stable (75 "C for 2 min) ammonium sulfate fraction (to provide Enzyme A activity), and 71 mM 2-mercaptoethanol.

Step 1: Fractionation on Ultrogel AcA 44-A dialyzed am- monium sulfate fraction (25 ml) prepared as described (12) from a crude extract (168 ml) of wild-type fly heads was subjected to centrifugation (1 h a t 40,000 X g) to remove insoluble material. The resulting supernatant solution was applied to a column (2.8 X 110 cm) of Ultrogel AcA 44 that had been equilibrated with a buffer solution containing 50 mM Pipes and 10% glycerol (pH 6.5). This solution will hereafter be referred to as the standard buffer. The column was developed with the standard buffer at a flow rate of 24 ml/h and 4.8-ml fractions were collected. The elution profile of protein from this column is shown in Fig. 1. Active fractions (73-99) were combined (total volume, 131 ml) and dialyzed overnight against 6 liters of a buffer solution containing 50 mM imidazole and 10% glycerol (pH 6.0) at 4 "C.

Step 2: Fractionation on DEAE-Sepharose-The dialyzed material from Step 1 was applied to a column (2.8 x 16.5 cm) of DEAE-Sepharose that had been equilibrated with a buffer solution containing 50 mM imidazole and 10% glycerol (pH 6.0) at 4 "C. The column was washed with 150 ml of equili- bration buffer and then developed with a linear gradient (total volume, 1 liter) of 0-500 mM KC1 (in the equilibration buffer). Fractions (11.0 ml each) were collected at a flow rate of 35 ml/h. Enzyme activity eluted between 225 and 275 mM KCI. The elution profile of protein and enzymatic activity from this column are shown in Fig. 2. Active fractions were com- bined (total volume, 125 ml) and dialyzed overnight against 6 liters of standard buffer that was 15% saturated with ammo- nium sulfate.

Step 3: Fractionation on Phenyl-Sepharose C1-4B"The di- alyzed protein solution from the previous step was applied to a column (1.5 x 15 cm) of phenyl-Sepharose C1-4B that had been equilibrated with the dialysis buffer described in Step 2. The column was washed with 100 ml of the dialysis buffer and then developed with a linear gradient (total volume, 300 ml) with the dialysis buffer of Step 2 as the starting buffer and standard buffer containing 70% (v/v) ethylene glycol as the final buffer. The column was then washed with 100 ml of the final buffer. Fractions (88 drops each) were collected at a flow rate of 1 drop/5 s. Activity eluted with the major peak of protein (as determined by optical density at 280 nm) begin- ning at about 40% ethylene glycol concentration. Fractions containing activity were combined and dialyzed overnight against 6 liters of standard buffer.

Step 4: Fractionation on Affi-Gel Blue-The dialyzed pro- tein solution from Step 3 (total volume, 295 ml) was applied

Pyrimidodiazepine Synthase from D. melanogaster 14123

FRACTION NUMBER

FIG. 1. Fractionation on a column of Ultrogel AcA 44. Protein concentration was followed by UV absorbance at 280 nm. Enzyme activity was assayed as W-labeled PDA synthesized as described under “Experi- mental Procedures.”

056

A 104

8 - 032

z d 6 E 024

- 72

- 56

24

8

FRACTION NCMBER

FIG. 2. Fractionation on a column of DEAE-Sepharose. Protein concentration and enzyme activity were determined as described in the legend to Fig. 1.

to a column (1.5 x 9.5 cm) of Affi-Gel blue that had been equilibrated with standard buffer. The protein solution was applied to the column at a flow rate of 13 ml/h. Following a wash with 100 ml of standard buffer, the column was then developed with 150 ml of standard buffer containing 1.5 M

KCl. Protein eluting during the salt wash (total volume of solution, 122 ml) contained enzyme activity and was dialyzed overnight against 6 liters of standard buffer.

Step 5: Fractionation on Polybuffer Exchanger 94-The dialyzed protein solution from Step 4 was concentrated to 4.7 ml by ultrafiltration. The concentrate was dialyzed for 4 h against 300 ml of a buffer solution containing 8-fold diluted Polybuffer 74 and 10% glycerol (pH 4.0), after which it was subjected to centrifugation (15 min at 40,000 x g) to remove

insoluble material. The resulting supernatant solution was applied to a column (1 x 35 cm) of Polybuffer Exchanger 94 that had been equilibrated with a buffer solution containing 25 mM imidazole and 10% glycerol (pH 7.4). The column was developed with 300 ml of the buffer containing 8-fold diluted Polybuffer 74 and 10% glycerol (pH 4.0). Fractions (4.8 ml each) were collected at a flow rate of 20 ml/h. The elution profile of protein from this column is shown in Fig. 3. Active fractions (45-53) eluting between pH 4.6 and 4.2 were com- bined, concentrated to 1 ml by ultrafiltration, and dialyzed overnight against 1 liter of standard buffer.

Table I presents a summary of the purification of PDA synthase from Drosophila melanogaster. The overall purifica- tion from heads is 2,170-fold (more than I9,000-fold from

14124

0 QD (u

0 0 - 0.04 z W

6 oz Q 0.0;

I C

Pyrimidodiazepine Synthase from D. melanogaster

28

20

12

4

4 12 20 28 36 44 52 60 68 76

\ ' l ' l l l l l l l l l ' l l

Pyrimidodiazepine Synthase from D. melanogaster

FRACTION NUMBER FIG. 3. Fractionation on a column of Polybuffer Exchanger 94. Protein concentration and enzyme

activity were determined as described in the legend to Fig. 1.

TABLE I Summary of the purification of PDA synthase

Enzyme ~ o t a ~ specific Re'ative preparation activitp activit9 ::?:;

~ ~~ ~~~

~ ~~ ~~

units unitslmg Crude extract 1400 0.37 1.0 Ammonium sulfate fraction 1140 0.66 1.8 AcA 44 eluate 1040 2.5 6.7 DEAE-Sepharose eluate 1020 31 85.0 Phenyl-Sepharose eluate 1330 240 640 Affi-Gel blue eluate 890 490 1320 Polybuffer Exchanger 94 eluate 470 800 2170 . ."

"One unit of activity is that amount of enzyme required for the production of 1 nmol of PDA from H,-NTP a t 22 "C under standard assay conditions.

~~ ~~

Units of enzyme/mg of protein.

whole flies) with an overall yield of 34%. When the purified protein was subjected to electrophoresis on polyacrylamide gels, one major protein band and one or two minor diffuse bands were evident (Fig. 4A). The purification of PDA syn- thase has been reproduced many times and the elution profiles of protein from the various columns are completely reproduc- ible. PDA synthase purified through the Affi-Gel blue step is stable a t -80 "C for a t least 9 months. The enzyme purified through the final step is stable for 1-2 weeks a t -80 "C; no attempts have been made to stabilize it.

Some Properties of the PDA-synthesizing System

Since neither the heat-treated (75 "C for 2 min) ammonium sulfate fraction nor the purified PDA synthase alone can catalyze the synthesis of PDA from H2-NTP, the work re- ported below was performed with saturating levels of the two enzyme preparations.

The enzymatic synthesis of PDA from H,-NTP was found to be linear with time up to 10 min of incubation. Optimal activity of the enzyme system was observed a t 30 "C. The enzymatic synthesis of PDA from Hz-NTP exhibited typical Michaelis-Menten kinetics. The K,,, value for H2-NTP was determined to be 8.5 p~ and the K,,, value for GSH is 1.8 mM. The PDA-synthesizing system is dependent upon the presence of Mg'", supplied as 5 mM MgCl,. When incubations are performed with no attempt to provide anaerobic conditions,

A .. ".T

B

j .O

U X

I. 0

FIG. 4. Electrophoretic behavior of purified PDA synthase. A, electrophoresis of purified PDA synthase on a polyacrylamide gel. Electrophoresis was performed on a 7.5% polyacrylamide gel as described under "Experimental Procedures" with 50 pg of protein purified through the final step. The gels were stained with Coomassie Blue to locate bands of protein. R, electrophoresis of purified PDA synthase on a polyacrylamide gel containing sodium dodecyl sulfate. A portion of the purified enzyme (20 pg) was subjected to electropho- resis on a 10% polyacrylamide gel that also contained sodium dodecyl sulfate (lone I ) . The six major bands in lone 2 (from top to bottom) correspond to the following standard proteins and their molecular weights: bovine serum albumin (67,000), chicken egg albumin (45.000). glyceraldehyde-3-phosphate dehydrogenase (36,000), t r w - sinogen (24,000), soy bean tr-ypsin inhibitor (20,100), and lactalbumin (14,200). Protein hands were located by staining with Coomassie Blue.

the PDA-synthesizing system is dependent upon the presence of another thiol-containing compound, routinely supplied as 71 mM 2-mercaptoethanol (12).

Pyrimidodiazepine Synthase from D. melanogaster 14125

TABLE I1 Order of action of enzymes required for PDA synthesis

Incubation mixtures (total volume, 0.2 ml) contained 10.4 WM [U-"C]H,-NTP (81 mCi/mmol), 1.34 units of Enzyme A, 1.4 units of PDA synthase, 5 mM GSH, 2.5 mM MgClz, 5 mM EDTA (pH 6.5) where indicated, and 71 mM 2-mercaptoethanol where indicated. All incubations were performed anaerobically for 1 h at 30 "C in sealed serum vials. When split incubations were performed, each incubation was for 30 min. The first 30-min incubation was terminated with the addition of EDTA to inhibit the action of Enzyme A (14). A 0.15-ml volume of each reaction mixture was analyzed by high-performance liquid chromatography for the production of PDA.

Components present Additional components Radioactive Conversion added prior to second

incubation PDA made o:yb:r during first incubation

Enzyme A, PDA synthase, 2-mercaptoethanol, MgCl,, GSH Enzyme A, PDA synthase, MgCl,, GSH Enzyme A, PDA synthase, 2-mercaptoethanol, MgCl,, GSH, EDTA Enzyme A, 2-mercaptoethanol, MgCL, GSH PDA synthase, 2-mercaptoethanol, MgC12, GSH Enzyme A, 2-mercaptoethanol, MgClz Enzyme A, 2-mercaptoethanol

3

3 00 4 00 3 00 4 00

1

0

WAVELENGTH (nm)

FIG. 5. Conversion of the product of the action of Enzyme A on H,-NTP to PDA by PDA synthase. A reaction mixture in an anaerobic cuvette was prepared to contain, per 0.75 m146 p M HZ- NTP, 7 mM Pipes (pH 7.5), 7% glycerol, 5 mM MgC12, 7 mM 2- mercaptoethanol, and 120 units of Enzyme A. Hz-NTP and the other incubation components were separately degassed for 1 h. After de- gassing, the substrate and enzyme solutions were anaerobically sy- ringed into a sealed anaerobic cuvette and the formation of the intermediate was allowed to proceed for 30 min at 30 "C. EDTA was then anaerobically added to the cuvette to yield a EDTA,"?' ratio of 2.0. A degassed solution (0.5 ml) containing 36 mM Pipes (pH 6.5), 7% glycerol, 12.5 mM GSH, and 55.6 units of PDA synthase purified through the Affi-Gel blue step was anaerobically added to the sealed cuvette. The incubation was allowed to proceed at 30 "C and spectra were recorded on a Perkin-Elmer model 557 double wavelength, double beam spectrophotometer at the timed intervals shown: 1, 0 min; 2, 4 min; 3, 8 min; 4, 12 min; 5, 16 min; and 6, 24 min. The spectrophotometer was background corrected for the buffer compo- nents and the protein solution containing Enzyme A and PDA synthase.

cpm None 57,620 None 56,120 None 4,600 EDTA, PDA synthase 49,510 EDTA, Enzyme A 6080 EDTA, PDA synthase, GSH 109,340 EDTA, ME&, PDA svn- 3080

Yo

20.7 20.2

1.6 17.8 2.2

39.3 1.1

thase, GsH ~'

TABLE I11 Effect of the addition of Enzyme A and PDA synthase on PDA-

synthesizing activity in crude extracts of purple, sepia, and wild-type flies

Incubation mixtures (0.2 ml each) contained 10.4 WM [U-'4C]H2- NTP (81 mCi/mmol), about 1 mg of protein from the crude extract indicated, 5 mM GSH, 70 mM 2-mercaptoethanol, 2 mM MgCl,, 1.2 units of Enzyme A (where indicated), and 1.1 units of PDA synthase (where indicated). An amount (0.15 ml) from each incubation mixture was assayed by high-performance liquid chromatography. All values have been normalized for differences in protein concentration be- tween the crude extracts.

Source of enzymes Radioactive Conversion of PDA made to

Wild-type" crude extract Wild-type crude extract +

Enzyme A Wild-type crude extract +

PDA synthase Sepia crude extract Sepia crude extract + Enzyme A Sepia crude extract + PDA

Purple crude extract 10,560 Purple crude extract + Enzyme A 49,870 Enzyme A + PDA synthase 107,400

synthase

Oregon-R strain.

cpm 67,300 96,630

80,700

2,930 3,830

62,970

% 24.2 34.8

29.0

1.0 1.4

22.6

3.8 17.9 38.6

Further experiments showed that Enzyme A (purified by 750-fold) could completely replace the heat-treated ammo- nium sulfate fraction. This indicates that only two enzymes, Enzyme A and PDA synthase, are required for the conversion of Hz-NTP to PDA in the presence of Mg', GSH, and 2- mercaptoethanol. The molecular weight of Enzyme A is known to be 82,000 (13). The molecular weight of PDA synthase was estimated to be 48,000 based upon a comparison of its rate of elution (detected as enzyme activity) from a column (1.2 X 111 cm) of Ultrogel AcA 44 with the rates exhibited by the following standard proteins and their molec- ular weights: bovine serum albumin (67,000), ovalbumin (44,0001, chymotrypsinogen A (25,000), and RNase A (13,700). Analysis of the purified enzyme by electrophoresis on 7.5, 10, and 15% polyacrylamide gels revealed one major protein band whose molecular weight was also estimated to be 48,000 by comparison with the behavior of the following standard proteins and their molecular weights: bovine serum

14126 Pyrimidodiazepine Synthase from D. melanogaster

FIG. 6. Probable pathway for the formation of PDA in D. melanogas- ter. Pa is triphosphate.

P DA

albumin (dimer, 134,000; monomer, 67,000), chicken egg al- bumin (45,000), and carbonic anhydrase (29,000). The method of Hedrick and Smith (20) was used in this determination. Finally, when the purified enzyme was subjected to electro- phoresis on polyacrylamide gels containing sodium dodecyl sulfate, one major band and one minor band was evident (Fig. 4B, column 1). The molecular weight of the major band was estimated to be 24,000 (Fig. 4B). All of these results considered together indicate that the molecular weight of the native enzyme is 48,000 and that it consists of two polypeptide chains of identical molecular weight.

Pathway of PDA Synthesis

Previous results from this laboratory have shown that the product of the action of Enzyme A on Hz-NTP is an extremely labile compound that can be temporarily stabilized when incubations are performed anaerobically (14). Based upon its ultraviolet spectrum (which is typical of that of a tetrahy- dropterin) and upon unpublished results' from this laboratory, the labile compound has tentatively been identified as pyru- voyl-H4-pterin (14). With the use of purified Enzyme A and purified PDA synthase we performed split anaerobic incuba- tions in order to determine the order of action of the two enzymes during the enzymatic conversion of H2-NTP to PDA. The results from these split incubations are given in Table I1 and show that Enzyme A and Mg2+ were required in the first incubation for the conversion of Hz-NTP to a product (py- ruvoyl-Hi,-pterin). GSH was not required during this trans- formation. During the second incubation, pyruvoyl-H4-pterin was converted to PDA in the presence of PDA synthase, GSH, and EDTA (added to inhibit the further action of Enzyme A). When the order of action of these enzymes was reversed, no synthesis of PDA was evident. Table I1 also shows that 2- mercaptoethanol is not required under anaerobic incubation conditions.

Previous work from this laboratory has shown that pyru- voyl-H4-pterin can be enzymatically converted to sepiapterin and to H4-biopterin in the presence of NADPH (14). The results from the split incubation experiment described above indicated that pyruvoyl-H4-pterin is also a precursor of PDA. Further evidence that pyruvoyl-H4-pterin can be converted to PDA in the presence of PDA synthase was obtained by following the reaction spectrophotometrically. The product of the action of Enzyme A on Hz-NTP was prepared in an anaerobic cuvette as described (14). Enzyme A allowed the conversion of Hz-NTP (which has a peak at 330 nm, typical

A. C . Switchenko and G. M. Brown, unpublished results.

Pyruvoyl- H 4 - Pterin

0 Synthase

of dihydropterins) to the product, pyruvoyl-H4-pterin, which has a peak at 300 nm (typical of tetrahydropterins). After the addition of excess EDTA to stop the reaction catalyzed by Enzyme A, excess PDA synthase and GSH were anaerobically added to the cuvette and UV spectra of the mixture were then recorded during a second incubation at the various time intervals indicated in Fig. 5. The spectrum at 0 min (curve 1) was taken at the conclusion of the reaction catalyzed by Enzyme A. At later times, after the addition of PDA synthase and GSH, absorbance at 300 nm decreased while absorbance at 265 nm and 384 nm increased. The final UV spectrum recorded (curve 6) was identical to the UV spectrum of PDA (9). When GSH was omitted from the reaction mixture, no change in the tetrahydropterin spectrum of the intermediate was evident.

PDA Synthesis in Eye Color Mutants Purple (pr ) , clot (cl) , and sepia (se) flies have little or no

PDA content and are deficient in the enzymatic synthesis of PDA from H,-NTP (9, 12). Purple is probably the structural gene for Enzyme A because when purified Enzyme A is added Zo purple extracts the wild-type level of PDA and sepiapterin synthesis is achieved (12,13). Because sepia flies have a lower level of PDA-synthesizing activity relative to wild-type flies, purified Enzyme A and PDA synthase were added to extracts prepared from the heads of this mutant to determine if it was deficient in either of these enzymes. The results from this experiment are shown in Table 111. The addition of purified PDA synthase to crude head extracts of sepia flies increased the PDA-synthesizing activity 22-fold to the level observed in wild-type flies. Addition of purified Enzyme A to sepia ex- tracts had no effect on PDA synthesis. This suggests that sepia* codes for the structural gene for PDA synthase.

DISCUSSION

In an earlier paper (12), analyses of various eye color mutants of Drosophila for PDA content and for the ability to synthesize PDA led to the suggestion that at least three enzymes are involved in the conversion of H,-NTP to PDA: one, called sepiapterin synthase A (or Enzyme A), that is present in limited amounts in the mutant, purple; a second enzyme, which we have named PDA synthase in the present paper, that is missing in the mutant, sepia; and a third enzyme in which the mutant, clot, is deficient. From the information supplied in the present paper, it seems clear that Enzyme A and PDA synthase are the only enzymes needed for the conversion of Hz-NTP to PDA as long as GSH is also avail- able. The question that arises is the nature of the deficiency

Pyrimidodiazepine Synthase from D. melanogaster 14127

in clot. The phenotypes of clot and sepia are identical in that neither contains either the drosopterins or PDA and both contain large amounts of sepiapterin (12). In contrast to sepia, clot extracts are capable of supporting the enzymatic conver- sion of Hz-NTP to PDA (12) despite the fact that clot flies contain no PDA or drosopterin. Clot has been described as leaky since the ability of extracts to support the synthesis of PDA is not as great as extracts prepared from wild-type flies (12). All of these observations suggest that the defect in clot i s probably only indirectly concerned with the conversion of

The split incubations performed under anaerobic conditions indicate conclusively that sepiapterin synthase A functions first to convert H2-NTP to a product that has been tentatively identified as 6-pyruvoyl-H~pterin (14) and that this product is then the substrate for PDA synthase in a reaction that requires the presence of GSH. These reactions are summa- rized in Fig. 6. Since pyruvoyl-H4-pterin is at a higher oxida- tion level than PDA, it seems reasonable to conclude that reducing power needed for the transformation is supplied as GSH. The conversion of pyruvoyl-H4-pterin to PDA can be envisioned to involve a number of chemical reactions includ- ing the opening of the 6-membered pyrazine ring, some rear- rangements, ring closure to form the 7-membered diazepine ring, and reduction. The evidence suggests that all of these probable reactions are catalyzed by a single enzyme, PDA synthase.

It has already been established that the enzyme designated as sepiapterin synthase A (or Enzyme A) is important in the biosynthesis of sepiapterin and tetrahydrobiopterin since the evidence is clear that the product of its action on Hz-NTP, pyruvoyl-H,-pterin, is an intermediate in the enzymatic pro- duction of these substances (14). The observation in this paper that this same enzymatic product is also an intermedi- ate in the biosynthesis of PDA strengthens the view that sepiapterin synthase A and the product of its action are of general importance in the biosynthesis of a variety of pterin compounds.

The reason for the need for a thiol compound such as 2- mercaptoethanol for PDA synthesis when care is not taken to maintain anaerobic conditions is not completely clear, although a likely possibility is that the thiol stabilizes pyru- voyl-H,-pterin against oxidative destruction. Consistent with this likely explanation is the unpublished observation3 from this laboratory that pyruvoyl-H4-pterin is stable in air only

A. C. Switchenko and G. M. Brown, unpublished observations.

Hz-NTP to PDA.

in the presence of high concentrations of 2-mercaptoethanol. Our results indicate that the mutant, sepia, is missing PDA

synthase. This suggests that sepia+ is the locus for this en- zyme, although gene dosage experiments remain to be per- formed to confirm this hypothesis.

Acknowledgments-We are grateful to Arthur C. Switchenko for purified Enzyme A and for assistance in performing the anaerobic incubations and to Dr. William H. Orme-Johnson and his colleagues for the use of their facilities for maintaining anaerobic conditions.

REFERENCES 1. Viscontini, M., and Mohlmann, E. (1959) Helu. Chim. Acta 42,

2. Pfleiderer, W. (1970) in Chemistry and Biology of Pteridines, Proceedings of the Fourth International Symposium on Pteri- dines, 1969 (Iwai, K., Akiro, M., Goto, M., and Iwanami, Y., eds) pp. 7-19, International Academic Press Printing Co., Ltd., Tokyo

3. Schlobsch, H., and Pfleiderer, W. (1972) Helv. Chim. Acta 56, 2518-2525

4. Rokos, K., and Pfleiderer, W. (1975) Chem. Ber. 108,2728-2736 5. Rokos, K., and Pfleiderer, W . (1975) in Chemistry and Biology of

Pteridines, Proceedings of the Fifth International Symposium, 1975 (Pfleiderer, W., ed) pp. 931-945, Walter de Gruyter, Berlin

6. Theobald, N., and Pfleiderer, W. (1977) Tetrahedron Lett. 10, 84-844

7. Theobald, N., and Pfleiderer, W. (1978) Chem. Ber. 111, 3385- 3402

8. Dorsett, D.. Yim. J. J.. and Jacobson. K. B. 11979) Biochemistrv

1679-1683

. , 18,2596-2600

9. Wiederrecht. G. J.. Paton. D. R.. and Brown. G. M. (1981) J. Biol. Chem. 256; 10399-10402 '

10. Wiederrecht, G. J., Paton, D. R., and Brown, G. M. (1983) in Chemistry and Biology of Pteridines, Proceedings of the Seventh International Symposium, 1982 (Blair, J. A., ed) pp. 391-395, Walter de Gruyter, Berlin

11. Jacobson, K. B., Dorsett, D., Pfleiderer, W., McCloskey, J. A., Sethi, S. K., Buchanan, M. V., and Rubin, I. B. (1982) Biochem-

12. Wiederrecht, G. J., Paton, D. R., and Brown, G. M. (1984) J.

13. Krivi, G. G., and Brown, G. M. (1979) Biochem. Genet. 17, 371-

14. Switchenko, A. C., Primus, J. P., and Brown, G. M. (1984)

15. Fan, C. L., Hall, L. M., Skrinska, A. J., and Brown, G. M. (1976)

16. Davis, B. J. (1964) Ann. N. Y. Acad. Sci. 121,404-427 17. Laemmli, U. K. (1970) Nature (Lond.) 227,680-685 18. Shriver, D. F. (1979) The Manipulation of Air-Sensitiue Com-

19. Bradford, M. M. (1976) Anal. Biochem. 72,248-254 20. Hedrick, J . L., and Smith, A. J. (1968) Arch. Biochem. Biophys.

. - ~ , ~

istry 2 1,5700-5706

Biol. Chem. 259, 2195-2200

390

Bwchem. Biophys. Res. Commun. 120, 754-760

Biochem. Genet. 14, 271-280

pounds pp. 3-37 and 139-205, McGraw-Hill, Inc., New York

126,155-164