The Metabolism of Nicotinic Acid I. PURIFICATION AND PROPERTIES OF 2,5DIHYDROXYPYRIDINE OXYGENASE

SUMMARY Purification and crystallization of an enzyme which catalyzes the oxidation of 2,5-dihydroxypyridine, an intermediate in nicotinic acid catabolism, are described. The labile enzyme is stabilized by dithiothreitol. Activity lost by dialysis or purification procedures is restored by incubation with dithiothreitol and ferrous sulfate. The enzyme is inhibited by sulfhydryl reagents and iron-chelating agents. The crystalline enzyme, on polyacrylamide electrophoresis with dithiothreitol, yields a single major band and a region of minor diEuse bands. In the absence of dithiothreitol the intensity of the minor bands is increased. On acrylamide gels containing sodium dodecyl a single band with mobility corresponding to a molecular weight of 39,500 is obtained. A single region of enzyme activity corresponding to a molecular weight of 242,000 is obtained on sucrose gradients containing dithiothreitol. A portion of this activity is shifted to a region corresponding to lower molecular weight when dithiothreitol is omitted. The of nicotinic by

agents. The crystalline enzyme, on polyacrylamide electrophoresis with dithiothreitol, yields a single major band and a region of minor diEuse bands.
In the absence of dithiothreitol the intensity of the minor bands is increased. On acrylamide gels containing sodium dodecyl sulfate, a single band with mobility corresponding to a molecular weight of 39,500 is obtained.
A single region of enzyme activity corresponding to a molecular weight of 242,000 is obtained on sucrose gradients containing dithiothreitol. A portion of this activity is shifted to a region corresponding to lower molecular weight when dithiothreitol is omitted.
The metabolism of nicotinic acid by Pseudomonas putida' has been investigated by Behrman and Stanier (2). They found that nicotinic acid was oxidized to 6-hydroxynicotinic acid, which was converted by oxidative decarboxylation to 2,5dihydroxypyridine.
A partially purified enzyme catalyzed the oxidation of 1 mole of 2,5-dihydroxypyridine to equivalent amounts of maleamic acid and formic acid, with consumption of 1 mole of oxygen.
The enzyme was labile and complete loss of activity occurred within 48 hours at 0". Activity lost on dialysis was restored by addition of ferrous iron. These properties suggested that this enzyme may be a dioxygenase similar to pyrocatechase (3). but later reclassified (1).
In this paper, we report the crystallization and further char acterization of the 2,5-dihydroxypyridine-oxidizing enzyme. An accompanying paper (4) deals with the oxidation products of 2,5-dihydroxypyridine and with oxygen 18 experiments, which establish that the enzyme is an oxygenase. For convenience, we anticipate these data and will, in this paper, refer to the enzyme as 2,5-dihydroxypyridine oxygenase. After 15 min of incubation at room temperature, an aliquot (0.5 to 0.10 ml) of the reactivated enzyme solution was used for assay.
Enzyme Assay-Enzyme activity was determined by following the disappearance of 2,5-dihydroxypyridine (e = 5200 M-I cm-l) at 320 rnp with a Cary recording spectrophotometer, model 15. The reaction was carried out at 25" in a 3-ml quartz cuvette with a l-cm light path and was initiated by the addition of substrate.
The reaction mixture contained 0.50 pmole of 2,5-dihydroxypyridine, 50 pmoles of sodium phosphate, pH 8.0, and 0.05 to 0.10 ml of reactivated enzyme solution, in a total volume of 2.7 ml.
One unit of activity is defined as the amount of enzyme which catalyzes the disappearance of 1.0 pmole of substrate in 1 min.
For testing substrate specificity, enzyme activity was followed by use of the Clarke Oxygen Electrode ( Gel Electrophoresis-Analytical disc gel electrophoresis was done with 50 mM Tris-glycine buffer, pH 9.3, and 7.5% acrylamide-separating gel (7). Protein in a 20% sucrose solution was layered on the gels. Electrophoresis was for approximately 2 hours at 3 ma per tube. The procedure of Weber and Osborn (8) was used to determine the molecular weight of the enzyme subunit. Gels and buffer contained 0.10% sodium dodecyl sulfate and electrophoresis was at 8 ma per gel.
Sucrose Gradient Centrijugation-Linear gradients from 5 to 20% sucrose (9) were made in a final volume of 12 ml of 50 mM sodium phosphate, pH 7.5. Centrifugation was for 16 hours at 286,000 X g. Fractions of 0.30 ml were collected. Each fraction was diluted to 1.0 ml, reactivated as described above, and assayed. This procedure yielded approximately 3 g of cells, wet weight, per liter of medium. All subsequent procedures were carried out at about 4' and centrifugations were at 9000 rpm for 15 min, unless otherwise stated.
Approximately 0.1 mg of deoxyribonuclease was added to reduce viscosity, and the suspension was passed through a French pressure cell under 16,000 p.s.i. of pressure, followed by centrifugation to remove broken cells and debris. High Speed Centrifugation-The crude cell free extract was centrifuged in the Beckman model L centrifuge at 89,000 x g for 2 hours to remove particulate matter. Ammonium Suljafe Fractionation-The high speed supernatant solution was diluted to 1000 ml with 20 mM sodium phosphate, pH 7.5, and the dithiothreitol concentration was adjusted to 5.0 mM. Solid amonium sulfate (265 g) was added with stirring, and the pH of the solution was adjusted to 7.5 with 5 N NaOH.
After 1 hour of continued stirring, the precipitate was removed by centrifugation and discarded.
To the supernatant solution, 125 g of solid ammonium sulfate were added and the pH was again adjusted to 7.5. After 1 hour, the precipitate was removed by centrifugation, resuspended in 100 ml of 20 mM sodium phosphate, pH 7.5, and dialyzed for 12 hours against three changes (4 liters each) of 10 mM sodium phosphate, pH 7.5.
DEAE-cellulose Column Chromatography-The DEAE-cellulose (Cellex-D, 0.77 meq per g) was prepared by suspending 100 g of the dry powder in 3 liters of 0.10 N NaOH containing 44 g of NaCl and 10 g of EDTA.
After holding the slurry for 1 hour at room temperature, the DEAE-cellulose was collected by filtration and resuspended in 3 liters of 0.25 h' SaOH. After 15 min, this material was filtered and resuspended in 3 liters of 0.10 N NaOH containing 44 g of NaCl.
The DEAE-cellulose was then washed once with 3 liters of 2 M NaCl, followed by four similar washes with 20 mM sodium phosphate, pH 7.5. Fine particles were removed by suspending the DEAE-cellulose in 3 liters of 20 mM sodium phosphate, pH 7.5, and decanting. This procedure was repeated four times.
The DEAE-cellulose was then equilibrated with 0.10 M sodium phosphate, pH 7.5, at 4". The column (4.5 x 35 cm) was packed under a pressure of 75 cm of H20 and washed with 4 liters of 0.10 M sodium phosphate, pH 7.5.
The dialyzed ammonium sulfate fraction was applied, and the column was washed with 1200 ml of the equilibration buffer. The enzyme was eluted with a linear concentration gradient established between 700 ml of 0.10 M sodium phosphat,e, pH 7.5, and 700 ml of 0.30 M potassium phosphate, pH 7.5, both containing 5.0 mM dithiothreitol. The column was eluted at a flow rate of 120 ml per hour and M-ml fractions were collect&d. Fractions with specific activity greater than 12, which emerged between 600-and 800-ml effluent volume, were pooled and dialyzed for 12 hours against three changes (4 liters each) of 20 mM sodium phosphate, pH 7.0. Hydroxylapatite Column Chromatography-The dialyzed DEAE-cellulose fraction was applied to an hydroxplapatite column (2.0 x 15 cm) equilibrated with 20 mM sodium phosphate, pH 7.0. The column was washed with 200 ml of 60 lllM sodium phosphate, pH 7.0, followed by elution of the enzyme with 90 mM sodium phosphate, pH 7.0, containing 3.0 mM dithiothreitol.
Elution was at 75 ml per hour and 6-ml fractions were collected.
Fractions with specific activity of at least 33 were pooled, placed in dialysis tubing, and concentrated to approximately 5 mg of protein per ml by covering the tubing with polyethylene glycol. This material was then dia.lyzed against 20 mM sodium phosphate, pH 7.5, for 12 hours.
Crystallization-Finely ground ammonium sulfate was added to the concentrated hydroxylapatite fraction until it became turbid and remained so for at least 5 min.
The precipitate was removed by centrifugation. This procedure was repeated one or two more times until addition of ammonium sulfate produced a silver sheen.
The solution was then dialyzed against 50 mM sodium phosphate, pH 7.5, for 6 hours, and then concentrated as described above to 5 to 10 mg of protein per ml.
The solution was further concentrated by suspending the dialysis tubing in air for 24 hours.
The crystals which formed (Fig. 1) were centrifuged, washed once with cold buffer, and dissolved in 50 IYIM sodium phosphate, pH 7.5, containing 5.0 mM dithiothreitol. Results of the enzyme purification procedure are presented in Table I.

Homogeneify
and Subunit Structure-Acrylamide gel electrophoresis of the crystalline enzyme solution in the absence of dithiothreitol yielded a single major band and a region of faster moving diffuse protein material ( Fig. 2A). When dithiothreitol was incorporated into the electrophoresis buffer and gels, the diffuse region was reduced in intensity (Fig. 2B). Protein from the major band in gels run without dithiothreitol was eluted from six gels, concentrated, and reanalyzed electrophoretically. This material yielded major and minor diffuse regions (Fig. 2C), indicating that enzyme from the major band dissociates to yield C, re-electrophoresis subunits of lower molecular weight. When the crystalline enzyme was electrophoretically analyzed in the presence of 0.10% sodium dodecyl sulfate, a single protein band was obtained (Fig. 2E). The molecular weight of this polypeptide was approximately 39,500 (Fig. 3). Centrifugation of the crystalline enzyme in a sucrose gradient containing dithiothreitol yielded a single region of enzyme activity corresponding to a molecular weight of 242,000 (Fig. 4A). When dithiothreitol was omitted, the peak corresponding to molecular weight 242,000 was reduced, and a broad region of enzyme activity corresponding to lower molecular weight was obtained (Fig. 4B). When Tris-glycine, pH 9.3, with no dithiothreitol was used, enzyme activity was observed only in the lower molecular weight region of the gradient (Fig. 4C).
Enzyme from the lower molecular weight region of the sucrose gradient (Fig. 4B) was concentrated and electrophoretically analyzed (Fig. 20). Both major and minor bands were obtained, indicating reassociation of the low molecular weight material.
Enzyme Reactivation-Initial studies showed that enzyme activity was completely lost within 48 hours at 0". Treatments such as dialysis, Sephadex gel filtration, or any of the purification procedures listed in Table I resulted in marked loss of enzyme activity.
Incubation of inactivated enzyme with the reducing agents, sodium borohydride, glutathione, ascorbate, or dithiothreitol, caused varying degrees of reactivation. Of these compounds, dithiothreitol was the most effective.
Behrman and Stanier (2) reported reactivation of their enzyme preparations by addition of ferrous ions to the a.%ay of the major Drotein band (75 LL~) eluted from gels similar to A. No dithioihreitol was added: Dyprotein (50 pg)lfrom low molecular weight region of sucrose gradient (see Fig. 4C). E, crystalline enzyme (40 pg) with 0.10% sodium dodecyl sulfate in gel and buffers.
An arrow indicates the position of the marker dye in each gel.  Fig. 5 presents the results of incubating dialyzed high speed supernatant solution at 25" with different concentrations of dithiothreitol or ferrous sulfate prior to assay. Prior incubation of the dialyzed or partially purified enzyme with dithiothreitol and ferrous sulfate together is more effective than prior incubation with either compound alone (Fig. 6). The combination of dithiothreitol and ferrous sulfate in the prior incubation mixture had a stabilizing effect upon enzyme activity (Fig. 6). This made possible a reproducible assay if the enzyme was initially incubated at 25" for 15 to 30 min immediately before testing. Optimal concentrations of these activating agents were found to be 10 mM dithiothreitol and 0.25 mM ferrous sulfate.
Although the dialyzed or partially purified enzyme could be In each case, 0.57 mg of enzyme was applied to the gradient in a volume of 0.40 ml. A, gradient prepared in 50 mM sodium phosphate, pH 7.5, containing 5.0 mM dithiothreitol.
Hemoglobin (O-O), monitored at 410 rnp, was used as the molecular weight standard. B, 50 mM sodium phosphate, pH 7.5, with no dithiothreitol.
C, 50 mM Tris-glycine, pH 9.3, with no dithiothreitol. reactivated to a considerable extent with either dithiothreitol or ferrous sulfate alone, the crystalline enzyme required both dithiothreitol and ferrous sulfate together for reactivation. Stability-When the enzyme was stored at 0" in 20 InM sodium phosphate, pH 7.5, then reactivated and assayed, its half-life was approximately 2 to 3 days (Fig. 7). The presence of dithiothreitol in the buffer extended the half-life to about 2 weeks. solution dialyzed against 20 m&r sodium phosphate, pH 7.5, and diluted to 2.0 mg protein per ml in the same buffer, was first incubated at 25" in the presence of 10 mM dithiothreitol, 0.25 mM ferrous sulfate, or both of these, in a total volume of 2.0 ml. At the times indicated, 0.10 ml was assayed as described under "Mat.erials and Methods." Enzyme first incubated in buffer without dithiothreitol and ferrous sulfate remained completely inactive.
Addition of ferrous sulfate did not protect the enzyme, whereas dithiothreitol and ferrous sulfate together were detrimental on long term incubation at 0". The enzyme retained 807; of its activity when stored for 3 months at -40". the crystalline material indicated that it did not contain bound ferric iron. ~11 Optima-The purified enzyme, like the crude preparation tested by Behrman and Stanier (2) was most active when assayed at pH 8.0. The optimal pH for reactivation and that at which the enzyme was most stable was found to be pH 7.5. Maximal activity was obtained in sodium phosphate buffer (20 to 100 mM). Tris, Tris-glycine, and Tris-maleate were inhibitory.
Inhibitors-The enzyme was sensitive to the sulfhydryl reagents p-chloromercuribenzoate and N-ethylmaleimide when these compounds were added directly to the reaction mixture prior to adding the substrate (Table III).
It was less sensitive to iodoacetamide and was not inhibited by iodoacetate. The enzyme was inhibited to some extent by the metal-chelating agents EDTA and KCN, and it was very sensitive to a,&dipyridyl and o-phenanthroline. The enzyme was also very sensitive to HzOz.
K&&O, and K,Fe(CN)G oxidized the substrate in the absence of enzyme.

DISCUSSION
The properties of the 2,5-dihydroxypyridine-oxidizing enzyme are similar to those of dioxygenases which catalyze the cleavage of the aromatic ring (10-13). It is unstable in the absence of dithiothreitol, sensitive to oxidizing agents, and is activated by ferrous inns and reducing agents. This enzyme does not contain heme or flavin and does not require an external reduced cofactor in stoichiometric amounts. The reaction does not proceed anaerobically in the presence of electron acceptors. The rise and subsequent decrease in activity observed when the enzyme is incubated with ferrous sulfate suggests that the iron readily dissociates from the enzyme or is lost by auto-oxidation or formation of an insoluble product.  I Vol. 246,Ko. 11 da11 (14) have shown that ascorbate maintains ferrous iron in a soluble form in phosphate buffer, pH 7.0. The marked stabilization of the enzyme activity upon prior incubation with ferrous ions and dithiothreitol together suggests that the dithiothreitol may be functioning in a similar fashion as ascorbate.
necessary for activity or whether the subunits themselves are enzymatically active.