Purification and properties of yeast nicotinamide adenine dinucleotide synthetase.

Abstract Yeast NAD synthetase was purified 2,000-fold. The purified enzyme gave one protein band on disc gel electrophoresis and was found to be monodisperse on high speed equilibrium ultracentrifugation with an apparent molecular weight for the "native" enzyme of 630,000. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis showed the presence of apparent nonidentical subunits with molecular weights of 80,000 and 65,000. The amino acid composition of "native" enzyme revealed no unusual amino acid residues. The specificity of the substrates, nucleoside triphosphate requirement, and divalent and monovalent metal ion requirements were studied. Nicotinate adenine dinucleotide, but not nicotinate mononucleotide, is the amide acceptor, and glutamine or free ammonia is the amide donor. The enzyme is specific for ATP, exhibiting a stoichiometric cleavage into AMP and PPi during the amidation. Mg2+ and K+ are required for enzymatic activity. Mn2+ or Co2+ can replace Mg2+ to a certain extent. NH4+ is as effective as K+ in stimulating enzymatic activity. The enzyme displays a broad peak of activity between pH 6.2 and 8.4, with maximal activity being observed around pH 7.6 when l-glutamine is used as the amide donor. However, when ammonium chloride is employed as the substrate, the enzymatic activity exhibits a rather narrow peak between pH 8.4 and 8.8. The apparent Km values for nicotinate adenine dinucleotide, ATP, and l-glutamine at pH 7.6 were found to be 1.9 x 10-4, 1.7 x 10-4, and 5 x 10-3 m, respectively. The Km for NH4Cl at pH 8.6 is 6.4 x 10-3 m. The catalytic constant was calculated to be approximately 1,260 moles of NAD produced per min per mole of enzyme, based on a molecular weight of 630,000. When the enzyme was previously incubated with azaserine, nicotinate adenine dinucleotide, and ATP, glutamine-dependent activity was progressively inhibited as a function of prior incubation time, whereas with NH4Cl as the amide donor there was an initial 20% inhibition and then NH4Cl-dependent activity remained constant. p-Chloromercuribenzoate and heavy metals inhibit both glutamine- and NH4Cl-dependent activity. Hydroxamate analogues were formed when (nicotinate adenine dinucleotide + ATP) and/or glutamine were incubated with enzyme and hydroxylamine. Hydroxamate analogues were not formed from nicotinate adenine dinucleotide unless ATP was also present. Hydroxamate analogue formation was additive when glutamine, nicotinate adenine dinucleotide, and ATP were added.

The purified enzyme gave one protein band on disc gel electrophoresis and was found to be monodisperse on high speed equilibrium ultracentrifugation with an apparent molecular weight for the "native" enzyme of 630,000. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis showed the presence of apparent nonidentical subunits with molecular weights of 80,000 and 65,000.
The amino acid composition of "native" enzyme revealed no unusual amino acid residues.
The specificity of the substrates, nucleoside triphosphate requirement, and divalent and monovalent metal ion requirements were studied.
Nicotinate adenine dinucleotide, but not nicotinate mononucleotide, is the amide acceptor, and glutamine or free ammonia is the amide donor.
The enzyme is specific for ATP, exhibiting a stoichiometric cleavage into AMP and PPi during the amidation.
Mg2+ and K+ are required for enzymatic activity.
Mn2+ or Co2+ can replace Mg2+ to a certain extent.
NH4f is as effective as K+ in stimulating enzymatic activity. The enzyme displays a broad peak of activity between pH 6.2 and 8.4, with maximal activity being observed around pH 7.6 when L-glutamine is used as the amide donor.
However, when ammonium chloride is employed as the substrate, the enzymatic activity exhibits a rather narrow peak between pH 8.4 and 8.8. The apparent K, values for nicotinate adenine dinucleotide, ATP, and L-glutamine at pH 7.6 were found to be 1.9 X 10-q 1.7 X 10w4, and 5 X 10e3M, respectively.
The K,,, for NH&l at pH 8.6 is 6.4 x low3 M. The catalytic constant was cakulated to be approximately 1,260 moles of NAD produced per min per mole of enzyme, based on a molecular weight of 630,000.
When the enzyme was previously incubated with azaserine, nicotinate adenine dinucleotide, and ATP, glutaminedependent activity was progressively inhibited as a function of prior incubation time, whereas with NH&l as the amide donor there was an initial 20% inhibition and then NH&ldependent activity remained constant. p-Chloromercuribenzoate and heavy metals inhibit both glutamine-and NH&l-dependent activity. Hydroxamate analogues were formed when (nicotinate adenine dinucleotide + ATP) and/or glutamine were incubated with enzyme and hydroxylamine. Hydroxamate analogues were not formed from nicotinate adenine dinucleotide unless ATP was also present.
Hydroxamate analogue formation was additive when glutamine, nicotinate adenine dinucleotide, and ATP were added.
In 1958, Preiss and Handler (I) demonstrated the existence of NAD synthetase in yeast and partially purified the enzyme. NAD synthetase cat,nlyzes the amidation of nicotinate adenine dinucleotide (1, 2) as shown in the reaction: Nicotinate adenine dinucleotide + glutamine or NH3 + ATP ---t NAD + glutamate + AMP + PPi The reported apparent K, for N"ADr is in the order of lop4 M. In contrast, the intracellular N"AD concentration determined from various sources, such as Ehrlich ascites cells (3)) Escherichia coli (4), and rat liver (5) varies from lC+ to 1OF M. Although NaSD concentration in yeast has not been measured, the indications are that it is low (6). These discrepancies may indicate that the enzyme is under some type of regulatory control or is present in a great excess over the rate-limiting reactions in NAD synthesis, N"JIX phosphoribosyltransferase, and quinolinate phosphoribosyltransferase (I, 7). In addition, yeast NAD synthetase utilizes both L-glutamine and NH&l as substrates at pH 7.5 and pH 8.6, respectively.
The pK of NHd+ is 9.2 (8), and the rate of react,ion is not enhanced by the concomitant presence of saturated concentrations of NH&l and L-glutamine at pH 7.4. Studies on the physicochemical properties of yeast NAD synthetnse and reinvestigations of its catalytic properties in the homogeneous state may provide possible clues to answer the above problems and also provide a way leading to further explorations of the nature of the enzyme at the molecular level. It is for t,bese reasons that studies on NAD synthetase from yeast were initiated. Assay of Enzymatic dctivity-NAD synthetase activity was determined by measuring the rate of NAD formation at 37". The reaction mixture for standard assays, except when obhermise stated, contained 1 mM N"AD, 2 MM ATP, 20 mM L-glutamine, 5 m&f MgC12, 56 rnM KCl, (0.04%, BSA when needed), 50 mxf Tris-Cl buffer, pH 8.0, and enzyme in a total volume of 0.25 to 0.5 ml. The reaction mixture was incubated for 30 or 60 min. After incubation, the reaction was terminated by heating in a boiling water bath for 30 to 60 s and the reaction mixture was centrifuged at 800 x g for 2 min. NAD was determined on the clear supernatant material by employing alcohol dehydrogenase (9) or fluorescence (10). Synthesis of &&strafe-N"AD was synthesized by the direct exchange reaction of the nicotinamide moiety of NAD with free nicotinate catalyzed by fresh beef spleen microsomal NAD glycohydrolase (NADase) (EC 3.2.2.5) (11). The incubation procedure and isolation of N"AD were those described by Honjo et al. (12). In order to obtain maximum N"AD formation from the NAD-nicotinate exchange reaction, it is critical to determine the optimurn amount of enzyme needed as well as length of incubation time. This can best be done by performing small scale pilot experiments.
By taking these precautions, it is possible to obtain NaAD with a yield as high as 70%.
Disc filectrophoresis-For ordinary polyacrylamide gel electrophoresis in the absence of SDS, the Canalco gel system was used except that the porosity of the gels was adjusted to 5.6% or 4% and the gels contained 1 x 10ea M Cleland's reagent.
A Buchler polyanalytic electrophoresis apparatus was employed. Protein (50 to 100 ~1) containing 25% glycerol was layered on top of the gel. Electrophoresis was run at 1.25 ma per tube until the tracking dye line was observed.
The current flow was t,hen adjusted to 2.5 ma per tube until the blue line was 1 cm from the bottom of the tube. The gels were fixed in 12% t'richloroacetic acid for 30 min, then stained in Coomassie blue for 1 hour or longer, and then destained in 7.5% acetic acid. For polyacrylamide gel electrophoresis in the presence of SDS, the systems used were the same as those described by Weber and Osborn (13).
When appropriate, the gels were removed from the tubes and sliced into 3-mm segments and assayed for enzymatic a.ctivity. The assay system was essentially the same as described in stand-ard NAD synthetase assay except that the incubation time was 3 hours or longer. gnalysis of Labeled Nucleotides-'4C-labeled A%IP, ADP, and ATP were separated by descending paper chromatographic techniques with Solvent I (isobutyric acid-concentrated NH,OHwater, pH 3.7, 66:1:33) or Solvent v (ethanol-l M ammonium acetate, pH 3.8, 5:2) as the solvent system. The spots corresponding to AMP, ADP, and ATP were cut into srnall pieces, placed in counting vials with 6 ml of t'oluene mixture containing 0.05yc 1,4-(2-(4-methyl-5-oxazolyl))benzene and 0.6% 2,5-diphenylosazole, and quantitated in a liquid scintillation spectrometer.
Two The radioactive spots identified as N&,1/N, NMN, N"AD, and NAD were cut out and qua.ntitated as described above.
PPi was quantitated by measuring the orthophosphate produced by t,he hydrolysis of pyrophosphate in the presence of crystaIline inorganic pyrophosphatase. Inorganic phosphate was determined by the method of Fiske and SubbaRow (14).
ATP-%PP+ and ATP-[14C]AMP Exchange-In the ATPm3'PPi exchange st'udies, 32PPi at a final concentration of lOV, 2 x 1p3, 10s3, or lo-' DI was added to the following sets of reaction vessels: +N"AD -GluNHz, +N"AD -GluNHz -K+, +N"AD + GluNHs, -NaAD + GluNH2, or -N"AD -GluNH2. The concentrations of all reactants were the same as described in the NAD synthetase assay. The reactions took place at pH 7.6 or 8.6. In other experiments, NAD synthetase was previously incubated with azaserine ('2 x lo+ M) in the presence of ATP, N"AD, and cations at pH 8.0 for 1 hour to inhibit glutaminedependent activity before 32PPi (2 X lO+ M) was added. In the case of the ATP-[14C]AMP exchange studies, [14C]AMP (10V3 M) was added to a series of reaction vessels similar to those employed in the ATP-PPi exchange studies. Activity was measured in the presence or absence of PPi (lo-* M) at pH 7.6. Paper chromatography was employed to separate radioactive PPi and ATP with Solvent III in the descending technique. Separation and quantitation of radioactive ATP and AMP was performed as described previously.
Protein was measured by the method of Lowry et al. (15), with crystalline BSA as a standard.
Amounts of Tris-Cl, pH 7.4, Cleland's reagent, EDTA, and KC1 equal to those present in samples being assayed were added to the BSA solutions before the 13SA standard curve was run. Protein was also determined by measuring the absorbance at 280 nm.
Afoolecular Weight Determination-The molecular weight of native NAD sgnthetase was determined according to the high speed equilibrium t,echnique of Yphantis (16). IJltracentrifugation was carried out with a Spinco model E analytical ultracentrifuge equipped with an electronic speed control and Rayleigh interference optical system. During the centrifugation, the temperature was maintained at 281.4" K. Centrifuge runs were performed in a six-channel centerpiece at initial protein concentrations of 0.65 to 0.33 mg per ml and column heights were 3 mm. The fringe patterns were measured with a Nikon microcomparat,or.
Data were processed by computer programs developed by Small and Resnick (17). A partial specific volume of 0.74 ml per mg was assumed for the calculation of molecular weight..
For molecular weight det,erminations of t'he subunits, 10 y0 polyacrylamide gel in the presence of O.lyO SDS (as described above) was used. Proteins of known molecular weight were used t,o construct a molecular weight st,andard curve. The mobiIit,y of ea.& species of protein was calculated a.ccording to Weber and Osborn (I 3). The molecular weights of the subunits of the native enzyme were determined by a comparison with the molecular weight standard curve. Amino ,&id Analysis-Yeast, NAD synthetase was suspended in const,ant boiling hydrochloric acid (approsimately 6 N) in a hydrolysis tube. The tube was thoroughly evacuated. After evacuation, the tube was sealed and heated to 100" for 24 hours. The hydrolysate was taken rapidly to dryness with a rotary evaporator.
LAnmlo acid analysis was performed by the method of Spackmau et al. (18) with a Beckman amino acid analyzer, model 120 C, having a recorder equipped with expanded range card so that the full scale on t,he recorder is 0.1 absorbance.
Formation of Hydroxamate Anulogues-Yeast NAD synthetase was incuba.ted with 200 1llM NHtOH, 5 mM MgClz, 56 mM KCI, 50 nu+f Tris, pH 8.0, and the various sub&rates for 60 min at 37" in a total volume of 0.55 ml. The reaction was terminated by the addition of 0.15 ml of 8% trichloroacet,ic acid containing 3.370 FeCl3 and 2 N HCl.
The hydrossmate analogues formed were measured as the ferric comples according to the procedure of Hartman (19).

RESULTS
Purifiration oJ NAZI X@hetase-All purification steps were carried out at 4". All of the buffer solutions used in t,he enzyme purifications contained 1 nlM Cleland's reagent and 1 rnM EDTA. Cleland's reagent could not be replaced by 2Oc/, glycerol.
The yeast cells, suspended iu 50 mM Tris, pH 7.4 (1 g wet weight per ml of buffer), were disrupted by employing a French press at a pressure of 16,000 psi., 4,960 ml of the disrupted cells being obtained by this t,reatment.
The broken cell suspension was centrifuged at 14,000 x g for 40 min and the pellet was washed once with the same buffer. The resultant turbid supernatant material was recentrifuged at 40,000 x g for 1 hour in a Spinco ultracentrifuge. NAD synthetase activity was recovered in the slightly turbid supernatant material.
To the supernatant material, solid ammonium sulfate (28.2 g per 100 ml) was added slowly with constant stirring.
After the last addition of the salt, the suspension was stirred for 30 min and then kept at 4" overnight.
The precipitate was collected by eentrifugation at 27,000 X g for 30 min with a Sorvall centrifuge and dissolved in 0.05 M Tris-Cl, pII 7.4. The insoluble material was removed by rentrifugation.
Most of the NAD synthetase activity was recovered in this fraction (0 to 40% saturation of ammonium sulfate).
The specific activity of the enzyme is highest in the fraction obtained at 20 to 300/, saturation, only a residual enzymatic activity being observed in fractions higher than 40% saturation.
Our experience is that the sa.me degree of final purity can be obtained using fractions from either 20 to 30% or 0 to 40% saturation.
The column (2.5 x 56 cm) was developed with a linear gradient of KC1 (0.04 to 0.26 M). A single peak of enzymatic activity was obtained.
Tile active fractions were pooled and concentruted to 20 1111 usiilg an Amicon Diafio cell with a PM30 or PM10 membntire. The active material was dialyzed against 100 volumes of the buffer containing 25 m&f potassium phosphat.e, pH 7.5, for 3 to 4 hours. The insoluble inactive mat,erials were removed by centrifugation before the enzyme preparation was applied t,o t,he nest column. The active material was then applied to a hydrosylapatite column which had been equilibrated with 0.1 M potassium phosphate, pH 7.4, containing Cleland's reagent and EDTX. After the proteins were adsorbed on the hydroxylapatite, the column was further washed with 200 ml of washing buffer. The column (1.6 X 15 cm) was developed with a linear gradient of potassium phosphate (0.025 to 0.2 M). The active fractions were pooled and concentrated to 4 ml using an Amicon Diafto rell with a PM-10 membrane.
This material was then dialyzed against 250 volumes of the buffer cont,aining 50 IIIM Tris-Cl, $1 7.6, and 100 or 350 m&f KC1 for 4 hours. The insoluble inactive material was removed by centrifugation.
The active material from hydroxylapatite fraetionat,ion was applied to a Sepharose 4B column which had been equilibrated with 50 mM Tris-Cl, pH 7.6, and 100 or 350 mM KCl. Elution was carried out with the same buffer. Two-milliliter fractions were collected at a flow rate of approximately 10 ml per hour. The fractions which contained a single protein band coincident with NAD synthetase activity on disc gel electrophoresis at pH 9 were pooled.
The pooled material was concentrated using an Amicon Diaflo cell with a PM-10 membrane.
A single enzymatic peak was usually observed and a V,/Vo = 2 to 2.2 was usually obtained. Attempts to locate a,nother enzymatic peak at higher elution volumes were unsuccessful.
A summary of the purification of NAD synthetase from bakers' yeast is presented in Table I. This enzyme was purified 2000fold with a specific activity of 2 pmoles per min per mg of protein when freshly prepared.
Criteria of Homogeneity-The active N&4D synthetase obt'ained from a Sepharose 4B column was purified to a state of apparent homogeneity according to the following criteria.
When purified NAD synthetase was submitted t'o gel electrophoresis (4% gel) using different protein c!oncent,rations, a single protein band was detected as shown in Fig. 1 (Fig. 2).

IV&cochemical Properties of NdD Synthetase
Xtability-The stability of NM) synthetase is dependent upon the degree of purity and protein concentration.
NAD synthetase from ammonium sulfate fractionation with a protein concentration of 40 mg per ml can be stored at -70" for more than 3 months with loss of only about 30% of the erlzymatic activity. The homogeneous enzyme with a protein concentration of 0.2 mg l,er ml loses 60y0 activity during 10 days of storage at 4". This inactivation cannot be prevented by storing at -70" or FIG. 1. Electrophoresis of NAD synthetase obtained after Sepharose 4B column chromatography (Table I) employing 5, 10, and 20 mg of protein.
Conditions of gel electrophoresis and NAD synthetase assay in the gel are described in the text. 2. Molecular weight determination of NAD synthetase employing ultracentrifugation.
The method of Yphantis (16) was employed. A plot of In C versus (radius)2 was obtained from this method. Protein concentration was 0.65 mg per ml. Details are described in the text and Table II. 4797 -20". However, the rate of inactivation can be greatly decreased by the addition of 13SA at a concentration great.er than 10 mg per ml. Repeated freezing and thawing accelerate the enzymatic inactivation.
Attempts to prevent inactivation of the pure enzyme by drying, changing the ionic strength, pH, addition of 1 mM Cleland's reagent, storing the enzyme in 30% glycerol, in 70% ammonium sulfate at neutral pH, or in 1 M urea were unsuccessful.
Partially inactivated enzyme can be reactivated to the original activity by the addition of 1 to 2% mercaptoethanol.
However, once enzymatic activity is totally lost, mercaptoethanol addition has no effect. Exposure of NSD synthetase to various pH levels at 4" overnight demonstrated that no loss of activity, occurred between pH 6 and 10. At pH 5, 50% of the initial activity was lost. This also occurred when the enzyme preparation was incubated at pH 5 for 5 to 10 min at 37".
MoLecu2ar Weight-The molecular weight of active, pure NAD synthetase obtained by the high speed equilibrium method of Yphantis (16) is summarized in Table II. Three expressions of molecular weight, i.e. weight average, number average, and Z average, are almost identical.
It is not known at present whether a molecular weight of 630,000 represents a molecular weight of a minimum active unit or a molecular weight of an aggregate from minimum active units.
One milligram of homogeneous yeast NAD synthetase catalyzes the formation of 2 pmoles of NAD per min when N&AD and L-glutamine are used as the substrates at pH 7.4 at 37". Since the molecular weight of NAD synthetase as isolated is approximately 630,000 (Table  II), the minimum cat.alytic constant is calculated to be approximately 1,260 moles of NAD produced per min per mole of NAD synthetase under the assay conditions described.
The rate of enzyme catalysis is not affected by using either 50 MM Tris-Cl or 10 IllM potassium phosphate as the buffer in the incubation system at pH 7.4. SDS-treated NAD synthetase, run on a disc gel (5.6%) containing SDS, was resolved into two bands which were stained in equal intensity by Coomassie blue (Fig, 3A). The SDStreated protein was incubated at 37" for 3 hours in 0.01 M sodium Speed, temperature, and time were 7164 rpm, 281.4" K, and 41 hours, respectively. The molecular weight calculations were processed employing a FORTRAN computer program (17). Subunit molecular weight determinations were determined employing SDS gel electrophoresis (13) as described in the text. Enzyme was incubated with 1% SDS and lDjO p-mercaptoethanol at 37" for 3 hours before being applied to the gels. The molecular >veight was calculated by the method described in Fig. 4 Native enzyme migrated as a single species in Isolencine. . '. 305 (329) ordinary 5.6oje gel (Fig. 3B). In order to estimate the approxi-Leucme.. . . . . . . 509 mate sizes of these polypeptides, the native enzyme was thor-Tyrosine 159 (156) oughly treated with 1% SDS and 1 y. P-mercaptoethanol. This Phenylalanine. . .

(202)
SDS-treated enzyme was then subjected to electrophoresis in a 10% polyacrylamide gel containing 0.1% SDS. The mobilities of these bands were compared with those of a number of standard addition, high amounts of aspartate and glutamate were obproteins of known molecular weight under the same conditions.

served. Asparagine and glutamine may have contributed to
The results are shown in Fig. 4. The molecular weights for the the high levels of glutamate and aspartate since a high percentage of ammonia was recovered after hydrolysis. Since the amount subunits, estimated by interpolation, were 80,000 and 65,000. of ammonia formed was not quantitated, no estimate of the The molecular weights thus estimated have &lO"j, deviation amount of asparagine and glutamine can be made.
in weight according to Weber and Osborn (13). Amino Acid Composition- Table III shows the amino acid Catalytic Properties oj NA D Synthetase composition of NAD synthetase. Because of the scarcity of pN optimum studies are presented in Fig. 5. NAD synthetase pure enzyme, only the results obtained after hydrolysis for 24 activity displays a broad peak of activity between pH 6.2 and hours in acid are presented. NO correction was made on unstable 8.4, with maximal activity b&g observed around pH 7.6 when residues such as serine and threonine.
Tryptophan was not n-glut~amine is used as the amide donor. However, when ammodetermined.
No unusual amino acids were detected in this nium chloride is employed as the substrate, the enzymatic acanalysis.
A rather high percentage of hydrophobic residues tivity exhibits a rather narrow peak between pH 8.4 and 8.8.
such as leucine, isoleucine, alanine, and lysine were noted.
In Enzymatic activity between pH 7 and 8 is directly correlated formation of ADP or AMP. However, when L-glutamine is omitted from the incubation mixture, small amounts of AMP are formed. When the complete incubation mixture is incubated with yeast NAD synthetase, a significant amount of AMP is formed. As shown in Table V, 1 pmole of [14C]AMP and PPi is produced per pmole of NAD synthesized.
No trace of inorganic phosphate is found during the reaction under the standard assay conditions. This is in agreement with the report of Preiss and Handler (I). In E. coli, the products of ATP cleavage are also found to be AMP and PPi (2).   at a 5 mn/r concentration. Mn2+ at 1 mM has some activity. However, when the concentration of Mn" is raised to 5 mM, NAD formation is inhibited.
In the absence of divalent metal ions, no NAD formation is detected under the assay conditions.
In the presence of 5 MM Mg*, K+ is required for enzymatic activity with glutamine as the substrate. Na+ has very weak activity.
Other monovalent metal ions tested; Li+ and Cs+, had no activity.
When NH&l is employed as the amide donor, K+ is not required for enzymatic activity. NH4+ can replace K+ when glutamine is the donor (Table VI) Tau~n  VI  TABLE   VII   Stim~rlalion   of LA-AD synthetase  011 ammonium  ion  Formation  of hytlroramate  analogues  in n-AD  synthetase  reaction The enzyme material from the hydroxylapntite step (Table 1) was used in these studies. The concent,rations of NH&l, gliitamine, and K+employedwere 2.1 X ~OWM, 2.1 X 10e2~, and 5.6 X 1OW M, respect,ively.
The concentrations of other reactants and assay conditions were as described in the text. The enzymatic reactions were carried out at the pH levels indicated. pH 6.2 is the lowest pH at which good enzymatic activity is retained. At this pH most of the NH3 is in the form of NHa+.
The incubation was carried out at 37" for 60 min.
The reaction lvvas terminated by the addition of 0.15 ml of 8% trichloroacetic acid containing 2 N HCl and 3.3ye FeC&.
The hydroxamate analogues formed were measured as a ferric complex by the procedure of Hartman tions of n-glutamine, NH&l, ATP, or N"AD. The apparent K, values for N"AD, ATP, n-glutamine, and NH&l at pH 7.6 were found to be 1.9 X 10v4 M, I.7 X low4 M, 5 X 10L3 M, and I.5 X 10-i M, respectively.
The apparent K, value for NH&J at pH 8.6 is 6.4 X 10e3 M. The apparent K, values were obtained from computer-generated calculations using the program of Hanson et al. (20) based on the method of Bliss and James (21). The K, values for ATP, N"AD, and n-glutamine were previously determined by Preiss and Handler (1) and the values were of the same order as found here.
Hydroxamate analogues are formed when NAD synthetase and (N"AD + ATP) and/or glutamine are incubated with hydroxylamine (Table  VII). ITnder the same conditions hydroxamate analogues were not formed with N"BD unless ATP was also present,. Hydroxamate analogue formation was additive when glutamine, N"AD, and ATP were added. Attempts were made to test whether or not, homogeneous yeast NAD synthetase catalyzes ATP-PPi or ATP-AMP exchange reactions.
We have exhaustively explored this area with negntive results.
Inhibition of NAD Synthetase Activity by illetal Ions-In the presence of 5 IrIM Mg2+ and 56 mM K+, heavy metal ions were tested to see whether or not these ions inhibit NAD synthetase activity.
Other met'al ions, such as Cazf, Fez+ or S1"+, are not inhibitory. NAD formation in these inhibition studies was measured by fluorescence, since heavy metal ions inhibit alcohol dehydrogenase activity (22). Inhibition oj NAD Synthetase by Glutamine Analogues-It was previously reported by Preiss and Handler (1) that prior incubation of yeast NAD synthetase with azaserine, N"AD, and ATP markedly accentuates the inhibition of NAD formation when n-glutamine is used as the amide donor.
These workers observed that omission of ATP or N"9D during prior incubation results in no demonstrable azaserine inhibition.
These findings were confirmed in the present study.
In addition, it was demonstrated that a similar effect with the glutamine analogue 6-diazo- 5-oxo-r-norleucine was observed.
In the presence of 6-diazo-5oxo-r-norleucine, prior incubation was not essential to produce inhibition.
That the inhibition observed with 6-diazo-5-oxo-nnorleucine may be irreversible in nature is suggested by the findings shown in Fig. 6. In the control experiment (in the absence of 6-diazo-5-oxo-n-norleucine), a linear relationship of NhD formation versus incubation time is obtained, whereas in the presence of 10e4 M 6.diazo-5.oxo-nnorleucine a complete inhibition is exhibited after 20 min incubation. The apparent irreversible inhibition by azaserine or 6-diazo-5oxo-r-norleucine is probably due to the fact that azaserine or 6.diazo-5-oxo-n-norleucine covalently links to the enzyme, presumably at the glut,amine site. It has been reported that the enzymes requiring glutamine (23-25) are also inhibited by glutamine analogues, and studies clearly demonstrate that azaserine covalently binds to the cysteine residues of the enzymes (26). This could well be the case for yeast NAD synthetase. Another possible explanation is that azaserine or 6.diazo-5-oxo-n-norleucine could irreversibly denature yeast NAD synthetase.
However, this possibility is ruled out by the finding presented in Fig. 7. The figure represents effects of prior incubation time on the inhibition of NAD synthetase by azaserine in the presence of ATP and N"hD at pH 8.0. The results show that with L-glutamine as the amide donor (assayed at pH 7.6), NAD synthetase activity is progressively inhibited as a function of prior incubation time, whereas with NH4Cl as the amide donor (assayed at pH 8.6), t,here is an initial 20% inhibition and then the NAD synthetase activity remains constant between 20 min and 150 min of preliminary incubation. These findings indicate that glutamine and ammonium chloride could act at different sites of the enzyme.
Inhibition of NAD Xynthetase Activity by p-Chloromercuri-  6. The effect of time on NAD synthetase activity in the presence and absence of 6-diazo-5-oxo-n-norleucine.
Enzymatic material from the Sepharose 4B step (Table I) having an initial specific activity of 0.9 pmole per min per mg of protein was used. 6-diazo&oxo-n-norleucine (0.1 mM) was added to the reaction mixture before the addition of enzyme. The incubation was carried out at 37". The reaction was terminated by boiling for 1 min at the time indicated.
NAD formation was measured by fluorescence as described in the text. X-X, + 6-diazo5-oxo-n-norleutine (1 X 1OF M); 0-O control (-6-diazo-5oxo-n-norleucine). I  I  I  I  I  I  I  0  10  20  30  40  50  60  150   PRIOR INCUBATION  TIME (mink   FIG. 7. Effect of prior incubation on the inhibition of NAD synthetase by azaserine. Enzymatic material from the Sepharose 4B step (Table I) was used. Incubations with L-glutamine and ammonium chloride were carried out at pH 7.6 and 8.6, respecttively. The tubes were incubated at 37" for 60 min. The reaction was terminated by boiling for ) min in a water bath. NAD formed was measured employing alcohol dehydrogenase as described in the text. loO$& is equal to the amount. of NAD formed in the absence of azaserine.
benzoic Acid and Diisopropyl Fluorophosphafe-Prior incubation of NAD synthetase with diisopropyl fluorophosphate (5 x lop4 M) for 2 hours at room temperature had no effect on enzymatic activity, whereas 100% enzymatic activity was lost during prior incubation of enzyme with lop4 M p-chloromercuribenzoic acid. Approximately 40y0 of the enzymatic activity is inhibited at a concentration of 10e5 M p-chloromercuribenzoic a,cid. These results were obtained using glutamine or NH&l as the substrate at pH 7.4 and 8.6, respectively. DISCUSSION NAD synthetaee has been investigated in yeast (1)) E. coli (2)) and rat liver (1). However, a purification procedure resulting in 4801 a homogeneous NAD synthetase has not previously been reported.
In the present study, this enzyme was purified from yeast approximately 2000-fold t)o a state of apparent homogeneity.
Physicochemical studies demonstrated that, the apparent molecular weight of the active enzyme is 630,000 and the molecular weights of its subunits are 80,000 and 65,000. An enzyme having a molecular weight of that value is rather unusual although there a,re some examples, e.g. glutamine synthetase (27), P-galactosidase (28), and RNA polymerase (29). Whether the molecular weight obtained from equilibrium studies represents a state of aggregation of the minimum active units or a minimum unit is not yet clear. Gel filtration studies do not contradict these observations.
It was clearly demonstrated that the amidation of N"AD requires ATP, Mg2+, and K+ and is associated with ATP cleavage to AMP and PPi (Table IV).
A small amount of AMP was obtained during the incubation in the presence of N*AD, ATP, W+, and K+. Both ATP and N"AD are necessary to promote the inhibition of amidation by azaserine when n-glutamine was employed as the amide donor.
ATP is also required for the amidation of N"AD when NH&l is employed as the substrate. The enzyme catalyzes the formation of a hydroxamate analogue of NAD in the presence of hydroxylamine which is not formed in the absence of ATP and Mg"+.
All of these findings taken together indicate that an enzyme-bound activated substrate is probably formed preceding the amidation step. This postulated enzyme-bound activated substrate complex may be analogous to those observed in the activation of acetate and amino acids in which the enzyme-acyl-AMP or enzyme-amino acyl-AMP complexes have been isolated (30)(31)(32).
On the other hand, Lagerkvist (34) reported that XMP aminase (pigeon liver) in the presence of the amino donor catalyzes the annnation of XMP to GMP wit.h a stoichiometric cleavage of ATP to AMP and PPi. XMP aminase from bone marrow behaves in a similar manner (35). In these cases, ATP-PPi exchange was not demonstrated. Experiments employing 102-XMP showed that **O is incorporated into the AMP in the presence of the amino donor.
We have not been able to demonstrate a significant exchange reaction between ATP and PPi or ATP and AMP.
A residual ATP-PPi exchange has been detected. However, since the purified NAD synthetase still contains a trace of NMN adenylyltransferase activity, it is not known whether this very slight exchange between ATP and PPi is catalyzed by NMN adenylyltransferase or by NAD synthetase.
The flus of materials through the various pathways of metabolism proceeds in a precise way according to the needs of the organism.
For this to occur, metabolic pathways must be controlled.
Regulation can be exerted at many levels. Studies on the individual enzymes catalyzing a series of consecutive reactions are one of the means to understand the metabolic regulation.
The 2,000-fold purified NAD synthetase has a specific activity of 2 pmoles per min per mg of protein (Table I). The catalytic constant based on the specific activity and a molecular weight of 630,000 was found to be 1,260 moles per min per mole of enzyme. The K, values for N"AD and L-glutamine were det.ermined to be 1.9 X lop4 and 5 X lo-' M, respectively. The intracellular NAD concentration in yeast is about. 5 x lop4 i+r (36), which is approximately 0.5 pmole per g wet weight tissue. The turnover time of hepatic NAD is about 2 hours (37). Utilizing these data and assuming that yeast cells contain 70% water and 30% soluble proteins, one concludes that yeast NAD synthetase is capable in vivo of producing approximately 216 pmoles of NAD per g wet weight of yeast per day. Furthermore, if one assumes that the NAD turnover time in resting yeast cells is similar to that observed in hepatic tissue (2 hours), then yeast cells must synthesize about 6 pmoles per g wet weight per day in order to maintain the above intracellular NAD concentrations. Since yeast NAD synthetase can synthesize 216 pmoles per g wet weight per day under optimum assay conditions, the yeast cell contains more than 30 times the amount of NAD synthetase needed to maintain normal levels of NAD.
The intracellular NaAD concentration is 10hs to 1OV M (3, 38, 39). From the measured rate and K, values, it can be calculated that the expected activity of NAD synthetase at endogenous substrate concentrations would be close to the NAD turnover rate. Thus, we feel that there is no need to postulate that NAD synthetase is under regulatory control.