Adenosine Triphosphate Sulfurylase from Penicillium chrysogenum

The molecular weight of Penicillium chrysogenum ATPsulfurylase (ATP-sulfate adenylyl transferase, EC 2.7.7.4) has been shown to be between 425,000 and 440,000. Other physical parameters determined were: ~~0,~ = 13.0, Stokes’ radius = 72 A, Dzovw = 2.94 X 1OV cm3.sec-l, and t = 0.733 cm3 .g-l. The carboxymethylated enzyme breaks down to subunits having a molecular weight of approximately 56,000 in the presence of 0.1% sodium dodecyl sulfate. Titration of the enzyme with 5,5’-dithiobis(Z-nitrobenzoic acid) reveals that there are eight reactive sulfhydryl equivalents per 440,000 g. This, together with the molecular weight of the subunit and amino acid analysis, suggests that the enzyme is an octamer containing one free sulfhydryl and four disulfides per protomer. Kinetic studies have shown that the actual substrate for the reaction is the ATP-Mg2complex and that free ATP is a competitive inhibitor with respect to both ATP-Mgzand MoOd2(K; = 0.6 to 1.25 mM). The K, for ATP-Mg2at saturating MoOd2is 4.6 x 10e5 M. The K, for Mo042at saturating ATP-Mg2is 1.5 X 10V4 M. In the reverse direction, the K,,, values for adenosine 5’-phosphosulfate and PPi are 7.1 X 10e6 and 7.7 X 10M5 M, respectively. Initial velocity studies and isotope exchange experiments show that the mechanism of the reaction is of the sequential type. The enzyme is inhibited by adenosine 5’-phosphosulfate (Ki = 4 x 10W5 M) and by sulfide. The inhibition by sulfide is sigmoidal. Sulfide also changes the molybdate concentration dependence from a hyperbolic to a sigmoidal form. The changes in the level of ATP-sulfurylase when mycelia are grown on different sulfur sources suggest that the synthesis of the enzyme is repressed by methionine, or some close metabolite of methionine.

The carboxymethylated enzyme breaks down to subunits having a molecular weight of approximately 56,000 in the presence of 0.1% sodium dodecyl sulfate. Titration of the enzyme with 5,5'-dithiobis(Z-nitrobenzoic acid) reveals that there are eight reactive sulfhydryl equivalents per 440,000 g. This, together with the molecular weight of the subunit and amino acid analysis, suggests that the enzyme is an octamer containing one free sulfhydryl and four disulfides per protomer. Kinetic studies have shown that the actual substrate for the reaction is the ATP-Mg2-complex and that free ATP is a competitive inhibitor with respect to both ATP-Mgzand MoOd2-(K; = 0.6 to 1.25 mM). The K, for ATP-Mg2at saturating MoOd2-is 4.6 x 10e5 M. The K, for Mo042at saturating ATP-Mg2-is 1.5 X 10V4 M. In the reverse direction, the K,,, values for adenosine 5'-phosphosulfate and PPi are 7.1 X 10e6 and 7.7 X 10M5 M, respectively. Initial velocity studies and isotope exchange experiments show that the mechanism of the reaction is of the sequential type.
The inhibition by sulfide is sigmoidal.
Sulfide also changes the molybdate concentration dependence from a hyperbolic to a sigmoidal form.
The changes in the level of ATP-sulfurylase when mycelia are grown on different sulfur sources suggest that the synthesis of the enzyme is repressed by methionine, or some close metabolite of methionine.
*This work was supported by National Science Foundation Research Grants GB-7736 and GB-19243.
Part I of this study is described in Reference 11.
1 The research described in this paper was taken from a thesis submitted by the author to the Graduate School of the University of California, Davis, in partial fulfillment of the requirements for the Ph.D. degree in Biochemistry (1). Present  In this reaction adenosine 5'-phosphosulfate is formed from ATP and sulfate (2). The enzyme is ubiquitous and has been studied from several sources (3)(4)(5)(6)(7)(8)(9).
However, scant attention has been paid to the physical and kinetic properties of the enzyme and only the yeast enzyme has been purified to homogeneity (10). The level of ATP-sulfurylase has been shown to be under metabolic control in yeast and in bacteria (5)(6)(7)(8).
Almost nothing is known about the kinetic and regulatory properties of ATPsulfurylase of filamentous fungi.
For this reason we were prompted to examine ATP-sulfurylase from Penicillium chqsogenum as part of our broader study on the regulation of sulfur metabolism in this organism. The purification and initial characterization of the enzyme have been reported (11).

MATERIALS AND METHODS
Chemicals-Except as noted, all chemicals were of the highest purity obtainable from commercial sources. APS (lithium salt) was synthesized by the procedure of Cherniak and Davidson (12). The identity of the product was established by its participation in the reverse reaction of ATP-sulfurylaee.
This reaction was also used to determine the concentration of APS in the presence of the AMP which was always a contaminant in the chemical synthesis.
All other nucleotides were titrated to pH 6.5 and their concentrations determined from their absorbance in the ultraviolet region (13 merged culture (14) and the purification of ATP-sulfurylase from this organism (11) have been described previously. For the physical studies described in this paper, a homogeneous preparation was used. For kinetic studies the eluate from the second DEAE-Sephadex column described in the purification procedure (11) was used. The concentration of the purified enzyme was determined from its absorbance index (E:iT at 278 mp) of 8.71 determined previously (11). Enzyme Assays-The assay procedure used in this study was a modification of the molybdolysis assay of Wilson and Bandurski (15) and has been described previously (11). The standard assay was carried out in a total volume of 0.1 ml.
However, in experiments where the ATP concentration was low this resulted in an assay which was nonlinear with time because of the progressive reduction in the concentration of ATP. Consequently the reaction was scaled up to either 1 or 2 ml.
The concentrations of the components (except for the enzyme) were unchanged.
The specific activity of the enzyme was the same whether assayed in a total volume of 0.1 or 2.0 ml.
For convenience, enzyme activity is reported in terms of micromoles of Pi released per min in the molybdolysis assay. Because the product of the ATP-sulfurylase reaction is inorganic pyrophosphate, true specific activities (in terms of international units) are half the reported values.
The enzyme was assayed in the reverse direction (ATP synthesis) with the spectrophotometric coupled assay described previously (11). In some experiments ATP-sulfurylase was assayed in the forward direction with %04" as the substrate. This assay (described below) depends on the incorporation of radioactive sulfate into APS and (if APS-kinase is present) PAPS. The nucleotides are then adsorbed onto charcoal leaving the unreacted Kz~~SO~ in the wash. Inorganic pyrophosphatase is added to the assay mixture in order to remove pyrophosphate from the reaction since the equilibrium is far in the direction of ATP and sulfate. However, even very large excesses of pyrophosphatase could not produce an assay linear with time or protein when purified ATP-sulfurylase was assayed because of inhibition of ATP-sulfurylase by APS.
In crude extracts containing APS-kinase the assay is more linear with both time and protein.
The radioactive assay was carried out in a total volume of 0. 3  with Schlieren optics and a temperature control system set to 20". Two standard 12-mm single sector cells with quartz windows were used, one being equipped with a 1" wedge window.
The enzyme solution was exhaustively dialyzed against 0.01 M Tris (titrated to pH 8.0 with HCI and containing 2 X lop4 M EDTA and 0.1 M NaCl) before use. The protein concentration ranged from 0.5 to 6.0 mg per ml. Glass photographic plates were used and the moving boundary distances were measured by the position of maximum ordinate by use of a Bausch and Lomb microcomparator.
The slope of the log z versus time curve was determined by the method of least squares with an Olivetti Programma 101 computer. The observed value of S was converted to the s20,u, according to Schachman (16). Values for the density and viscosity of the solvent were obtained from standard tables for 0.1 M NaCl (17).
Discontinuous Gel Electrophoresis-Discontinuous gel electrophoresis was carried out at several polyacrylamide concentrations according to the procedure described by Hedrick and Smith (18). Several proteins of known molecular weight were used to calibrate the gels. The logarithm of the relative mobility of each protein was plotted against gel concentration and the slope of this line was then plotted against the molecular weight. This calibration curve was used to determine the molecular weight of ATP-sulfurylase.
Disc gel electrophoresis was also performed in the presence of SDS by a modification of procedures described by Shapiro, Vifiuela, and Maize1 (19) and Weber and Osborn (20). The gel and buffer system used was the same as that used for electrophoresis of the unmodified proteins. ATP-sulfurylase and the standard proteins used were carboxymethylated before electrophoresis (21). The proteins were then dialyzed against the upper gel buffer to which 0.1% SDS had been added.
The samples were applied above the upper gel buffer in 20% sucrose in the normal way.
The sample solution and the upper reservoir buffer both contained 0.1% SDS.
At the completion of electrophoresis the dye front was marked with a fine wire and the gels were stained with Coomassie blue according to the procedure of Weber and Osborn (20). The relative mobility of the proteins was then measured and plotted against their molecular weight on a logarithmic scale. Gel filtration-Sephadex G-200, fine (Pharmacia), was allowed to swell for several days at 4" in 0.05 M Tris base adjusted to pH 8.0 with HCl and containing lop3 M EDTA. This buffer was used for all experiments.
The gel was poured into a column (Pharmacia), 2.5 x 40 cm, and allowed to settle while a slow flow rate was maintained.
All operations were performed at 4". Sample solutions contained 10% sucrose and were applied by layering under the buffer solution just above the gel. The column eluate was monitored at 254 rnp with an LKB Uvicord detector.
Constant flow rates (8 to 12 ml per hour) were maintained with a Mariotte flask. The elution volumes of several proteins of known molecular weight were used to calibrate the column according to the method described by Andrews (22 The position of ATP-sulfurylase was determined by measurement of its enzymatic activity with the molybdolysis assay. There was no dependence of elution volume on protein concentration over the range of 0.1 to 1.0 mg per ml. The excluded volume (V,) of the column was determined with blue dextran 2000 and the total volume (Vi) with tritiated water. The data obtained above were also interpreted according to Ackers (25)

Physical Characteristics Sedimentation
Velocity-The szO,W of ATP-sulfurylase was determined at several protein concentrations between 0.5 and 6.0 mg per ml.
This data was then p'otted and extrapolated to zero protein concentration ( Fig. 1). A value of 13.0 S was obtained for the s20,W of the enzyme at infinite dilution.
Molecular Weight from Gel Filtration-The molecular weight was determined by gel filtration on a Sephadex G-200 column calibrated with pure proteins of known molecular weight. The results, shown in Fig. 2, indicate a molecular weight of 440,000 for the enzyme.
An interesting observation made during the course of these experiments was that if ATP-sulfurylase was applied to the Sephadex G-200 column together with the blue dextran 2000, the enzyme activity peak coincided with the blue dextran peak, i.e. the enzyme was eluted at the excluded volume. Presumably this resulted from the binding of the enzyme to the blue dextran but the phenomenon was not further investigated. A Sephadex G-200 column was also used to calculate the Stokes' radius of ATP-sulfurylase according to the method of Ackers (25). Table I shows the gel pore radius determined from the elution volumes of ferritin, bovine serum albumin, yeast alcohol dehydrogenase, catalase, and myoglobin. Also shown are the elution volumes for blue dextran 2000 and tritiated water, which were used to determine the column parameters V,, Vt, and Vi. The gel pore radius calculated from this data showed good agreement for proteins with a wide range of Stokes' radii.
The elution volume of ATP-sulfurylase from this column was 75.5 ml which corresponds to a Stokes' radius of 72 A. The Stokes' radius of ATP-sulfurylase was then used to calculate the free diffusion coefficient for the enzyme according to the based on a molar absorbance index of 13,600 at 412 rnp (26). Eight sulfhydryl equivalents per 440,000 g were titrated.
In the presence of SDS, no additional groups became available. This suggests that there is only one free sulfhydryl equivalent per 56,000 g of subunits.
Since amino acid analysis gives a value of 9 half-cystine residues per 56,000 g (11) it is probable that the enzyme contains four disulfide bonds and one free sulfhydryl per subunit.
This result also supports the conclusion obtained by gel electrophoresis in the presence of SDS that the enzyme is an octamer.

Kinetic Properties
Interaction with ATP and Magnesium-It has previously been established that a divalent cation (MgZf, Mn*, or Cot+) is necessary for the activity of ATP-sulfurylase (11). The results shown in Fig. 5A show that the substrate for the reaction is the 1: 1 complex of ATP and magnesium. Fig. 5A also shows that free ATP is an inhibitor of the enzyme. Fig. 5B shows a Dixon plot of the data from Fig. 5A indicating that the inhibition by free ATP is competitive with respect to molybdate. Double reciprocal plots of these data (not shown) were linear and extrapolated to the same Vm,, for all free ATP concentrations, as expected for a competitive inhibitor. Fig. 5C shows the results of an experiment in which ATP was increased at several fixed concentrations of MgC12 and at a single molybdate concentration (1.0 InM). Once again the inhibition by free ATP is obvious and a Dixon plot at several ATP-Mgzconcentrations (i.e. Mg* concentrations) shows that free ATP is a competitive inhibitor with respect to ATP-Mg" (Fig. 50). Double reciprocal plots of these data (not shown) were linear and extrapolated to the same V,,, for all free ATP concentrations. Similarly, a replot of the slopes of the Dixon plot (Fig. 50) against the reciprocal of the ATP-Mg2-concentration was linear with an intercept at the origin. The ATP-Mg" concentration in these experiments was taken to be equal to either the MgC& or the total ATP concentration depending on which was limiting in each experiment (the K, for ATP-Mg2- The free ATP concentration was calculated from the difference between this value and the total ATP concentration. Increasing the concentration of Mg2+ at a constant concentration of ATP-Mg" had no effect on the rate of the reaction. The fact that the reciprocal plots are not parallel but intersect to the left of the vertical axis reveals that there is no irreversible step between the addition of the two substrates (36). Since the release of a product before interaction with the second substrate would constitute an irreversible step, this result suggests that the mechanism of the reaction requires the involvement of both substrates prior to the release of any products.
1)ouble reciprocal replots of both the slopes (not shown) and the y axis intercepts from Fig. 6, A and B, were linear.
The double reciprocal replots of the y axis intercepts from both primary reciprocal replots are shown in Fig. 6C This mechanism is shown in the following reactions.
ATP + enzyme + enzyme-AMP + PP; Enzyme-AMP + SOd2-$ enzyme + APS (2) This mechanism is of the "ping-pang" type rather than the sequential type and is inconsistent with the initial velocity   Fig. 7. In the absence of added sulfate Sulfide is the end product of the pathway of sulfate activation there was no detectable exchange of 32PPi into ATP. The and reduction and, as such, would be a likely feedback inhibitor addition of 2.0 mM sulfate did enable the exchange to take place controlling the first step in the sequence. Inhibition by APS but the rate at which 32PPi entered ATP could be accounted for is also to be expected since it is a product of the reaction with the by the reaction of the APS produced with "PPi to give labeled physiological substrate.
None of the sulfur compounds tested ATP.
(n-cysteine, n-cystine, n-methionine, n-cysteine sulfinic acid, The mechanism proposed by Levi and Wolf (4) also predicts n-cysteic acid, reduced glutathione, Na2S03, Na&05, and choan exchange between the sulfur atoms of APS and sulfate in the line-o-sulfate) had any effect on the reaction. Similarly, Labsence of other substrates (Reaction 2). We were unable to serine, n-homoserine, 0-acetyl-n-serine, and O-euccinyl-nndemonstrate this exchange. This result, together with the data homoserine had no effect. obtained from the initial-velocity experimen&, suggests that the Inhibition by APL&-The effect of APS on the reaction is mechanism of the reaction catalyzed by the P. chysogenum shown in Fig. 8. When this data was plotted in double reciprocal ATP-sulfurylase involves the combination of both ATP-Mg2+ form (inset, Fig. 8 The inhibition of ATP-sulfurylase by APS was also evident when K235S04 was used as the substrate and the AF3% produced was adsorbed to Norit. This reaction was never linear either with time or protein concentration even in the presence of large amounts of inorganic pyrophosphatase. The inhibition must have been caused by APS produced during the reaction since pyrophosphate is continually The enzyme activity was determined as described previously (11). The APS concentration used when PPi was the varied substrate was 0.5 mM, while the PPi concentration used when APS was the varied substrate was 5.0 mM. The inset shows the double reciprocal plot of these data. removed.
The concentration of APS produced in these experiments was between 0.02 and 0.05 mM, i.e. in the same range as the Ki for APS determined with molybdate as substrate. Inhibition by SulJide- Fig.  QA shows the effect of sulfide on the molybdate dependence of the reaction rate.
In the absence of sulfide a normal hyperbolic saturation curve is obtained. Sulfide inhibits the reaction and changes the saturation curve to a sigmoidal form.
The double reciprocal plots of velocity versus substrate concentration (inset, Fig. 9A) are concave upward but all extrapolate to the same V,,,. This is characteristic of an allosteric inhibitor of the K type as described by Monod,Wyman,and Changeux (39).
The Hill plots (40) from the data in Fig.  9A are shown in Fig. 9B. Sulfide at 1.5 mu changes the interaction coefficient (n) for molybdate from 1 (in the absence of inhibitor) to 1.54. The change in the SO.s for molybdate is even more marked, increasing from 0.65 mu in the absence of inhibitor to 4.5 mu at 1.5 mu sulfide.
The dependence of the reaction rate on sulfide concentration at several molybdate concentrations is shown in Fig. 9C. trations tested the inhibition curve for sulfide was sigmoidal with an interaction coefficient between 1.30 and 1.55. The greatest effect of molybdate was on the I0.s for sulfide. At high molybdate (5.0 mM) the IO.5 for sulfide was 4.02 IIIM but at a lower molybdate concentration (0.5 mM) the IO.5 for the inhibitor was 0.46 mM. These results suggest that sulfide may play a significant role in the control of sulfate activation in P. chrysogenum. Reaction in Direction of ATP Synthesis-Preliminary studies were carried out on the reaction of ATP-sulfurylase in the nonphysiological direction (ATP synthesis) with the assay procedure described previously (11). The concentration dependence for each substrate in the presence of saturating amounts of the alternate substrate was examined and the results are shown in Fig. 10. The K, for APS was 0.0071 mM and the K, for PPi was 0.077 mM.
The reciprocal plots for PPi concentration dependence (inset, Fig. 10) are linear even at high substrate concentrations (up to 2.0 111~) showing that PPi does not inhibit ATP-sulfurylase.

ATP-sulfurylase Levels in Mycelia
Grown on Various Sulfur sources-P. chrysogenum was grown with several different sulfur compounds as the sole sulfur source and the enzyme activity of the mycelia determined with the molybdolysis assay. The initial concentration of the sulfur source in the medium was 10 pmoles of sulfur per ml.
Sulfur sources at this concentration are not depleted in a 24-hour period. 2 The results of this experiment are shown in Table II.
The most dramatic difference in enzyme level was shown by the mycelia grown on methionine.
The level of the enzyme was about 15% of that present in mycelia grown on thiosulfate (the sulfur source yielding the highest level of enzyme).
If methionine-grown mycelium were washed and resuspended in media without a sulfur source, the level of ATP-sulfurylase rose to 30 to 40 enzyme units per g, dry weight, after 10 to 12 hours.
The combined activating enzyme levels were also determined in these crude extracts with the radioactive a5S042-assay described under "Materials and Methods." The relative activities on different sulfur sources were the same as shown with the molybdolysis assay (Table II) although the actual values were much lower because of the slower rate of reaction with sulfate as substrate.
These results are consistent with a control mechanism involving the repression and  (Fig. 10). It is possible that the actual substrate in divided into two parts and weighed. One of these was used to this direction is the APS-Mg complex and that free APS is indetermine the dry weight after 8 hours at 100". ATP-sulfurylase hibitory as was shown for free *4TP in the forward direction. In activity was determined in the second piece after extraction by the experiment shown in Fig. 10  The experiments described in this report suggest that the The molecular weight of ATP-sulfurylase was shown by sev-amount of ATP-sulfurylase present in the mycelia is controlled by the intracellular concentration of some sulfur metabolite reeral methods to be between 425,000 and 440,000. This is greater than the molecular weight of 100,000 calculated for the yeast en-lated to methionine.
When the organism is grown on sulfur sources which provide high levels of methionine, or a close zyme from the data of Robbins and Lipmann (10). Levi and Wolf (4) have shown that rat liver ATP-sulfurylase is very large, metabolite of methionine, the synthesis of ATP-sulfurylase is repressed. with an approximate molecular weight of 900,000. These re-Sulfur starving of a culture previously grown on methiosults suggest that the ATP-sulfurylases from the three sources nine results in a 6-to 7-fold increase in the level of ATP-sulfurylase activity. are unrelated proteins. This result is similar to that obtained in the present work and suggests that similar mechanisms are involved in the control of ATP-sulfurylase synthesis in yeast and filamentous fungi.