Preparation and properties of a homogeneous aromatic l -amino acid decarboxylase from hog kidney

ARCHIVES

OF

BIOCHEMISTRY

Preparation

and Acid JAMES

AND

141, 356-367 (1970)

BIOPHYSICS

Properties

of a Homogeneous

Decarboxylase

from

Hog

G. CHRISTENSON,2 WALLACE AND SIDNEY UDENFRIEND

Roche Institute

of Molecular

Received

Biology,

Nutley,

August

17, 1970

Aromatic

L-Amino

Kidney’ DAIRMAN,

New Jersey OYllO

Preparations of hog kidney aromatic L-amino acid decarboxylase have been obtained which are 99% pure as indicated by disc gel electrophoresis, sedimentation, and immunological techniques. The purified enzyme has a molecular weight of 112,000 daltons and is associated with about 0.9 mole of tightly bound pyridoxal phosphate. Nevertheless, added pyridoxal phosphate stimulated the activity up to fivefold. The enzyme decarboxylated 3,4-dihydroxyphenylalanine, 5-hydroxytryptophan, tryptophan, phenylalanine, and tyrosine at readily measurable rates. Slow decarboxylation of histidine also occurred. The enzyme was inhibited by sulfhydryl reagents and by certain metal ions, but sulfhydryl compounds and chelating agents had little or no effect. Certain other physical and chemical properties of the enzyme were studied.

DOPA decarboxylase (L-3,4-dihydroxyphenylalanine carboxy-lyase, EC 4.1.1.26) was the first enzyme in the pathway from tyrosine to norepinephrine to be described (1). Early work (see, for example, Ref. 2) suggested that DOPA decarboxylase and 5-hydroxytryptophan decarboxylase were distinct enzymes and the Ezyme Commission assigned the number 4.1.1.28 to L-5hydroxytryptophan carboxy-lyase. However, the weight of later evidence favored the hypothesis that a single enzyme acted on both substrates (see the review by Sourkes, Ref. 3). Lovenberg et al. (4), using a partially purified preparation from guinea pig kidney, concluded not only that DOPA and 5-hydroxytryptophan were decarboxylated by the same enzyme, but that the specificity

of the enzyme was sufficiently broad to utilize all the naturally occurring aromatic amino acids, including histidine, and certain a-methyl derivatives as substrates. They proposed the name “aromatic L-amino acid decarboxylase,” but some controversy still surrounds the activity of the enzyme toward p-tyrosine and histidine (5-7). More recently, Streffer (8) has confirmed earlier observations that DOPA and 5-hydroxytryptophan were competitive inhibitors, each for the other, and Coulson et al. (9) have reported that in four electrophoretitally separated forms of the decarboxylase, DOPA and 5-hydroxytryptophan activities were not separated. All of the above investigations, however, were limited because highly purified enzyme preparations were not available. This paper reports the purification to homogeneity of aromatic L-amino acid decarboxylase from hog kidney and some of its physical, chemical, and enzymological properties.

1 A preliminary report of these findings w&s presented at the meeting of the Federation of American Societies of Experimental Biology, Fed. Proc. as, 867 (1970). e The material presented in this paper is taken in part from a dissertation to be submitted by J. G. C. to the Department of Biochemistry of the City University of New York in partial fulfillment of the requirements for the degree of Doctor of Philosophy.

MATERIALS

AND

METHODS

Streptomycin sulfate, ammonium sulfate (“ultra pure”), and amino acids (all L-isomers, unless otherwise specified) were obtained from Mann Re356

AROMATIC

L-AMINO

search Labs. Omnifluor and radioactive compounds were the products of New England Nuclear, unless otherwise specified. 1-“C!-L-Tyrosine was freed of possible contamination with DOPA by treatment with alumina, according to the method of Crout (10); other compounds were used without further purification. Pyridoxal phosphate was supplied by Hoffman-La Roche, Basle, Switzerland. NCS3 was obtained from Amersham-Searle. Polyethylene glycol-6000 was purchased from Matheson, Coleman and Bell. Alumina Cr gel and hydroxylapatite (Bio-Gel HTP) were from Bio-Rad Labs. The ultrafiltration apparatus was the product of Amicon Corp. and the variable gradient maker was from Phoenix Precision Instrument Co. Pyridoxal phosphate was assayed by the fluorometric method of Adams (11). Protein was assayed by the method of Lowry et al. (12), except in monitoring column effluents, when the optical density at 280 nm was taken &s a measure of protein concentration. Histidine decarboxylase activity was assayed both by a radiometric method similar to that described below for DOPA decarboxylase and by the fluorometric method of Lovenberg et al. (4). Polyacrylamide gel electrophoresis. Analytical polyacrylamide gel disc electrophoresis was performed by a modification of the method of Davis (13) or of Williams and Reisfeld (14). Bothstacking and resolving gels were photopolymerized in the presence of 5 or 2.5 rg/ml, respectively, of riboflavin. Electrophoresis was at a constant current of 2 mA per gel column and was ended when the bromothymol blue tracking dye reached the anode end of the gel. The gels were stained for at least 3 hr in a solution of Coomassie brilliant blue R250 prepared by diluting a 1% solution l:u) with 12.5y0 trichloroacetic acid. They were destained overnight in 5&75 ml/gel of 12.5% trichloroacetic acid and stored permanently in 7.5% acetic acid. Relative amounts of protein in the bands were estimated by scanning the destained gels at 580 nm in a Gilford Model 240 recording spectrophotometer equipped with aMode 24101inear transport device. Assay for DOPA decarboxylase activity. This assay is based on “CO2 evolution from 1-‘4CDOPA and makes use of apparatus and procedures previously described by Rhoads and Udenfriend (15) and by Ellenbogen et al. (16) for other enzymatic decarboxylations. Assays were performed in a 21 X 145-mm tube having a closed side arm, about 50 mm from the bottom, to which was added 0.3 ml of 35% trichloroacetic acid. The preincubation mixture contained, in a total volume of 1.25 ml: a All abbreviations are standard IUPAC-IUB abbreviations. NCS is a trademark of the Amersham-Searle Corp. for a 0.6 N solution of quaternary ammonium base in toluene.

ACID

DECARBOXYLASE

357

120 pmoles of sodium phosphate, pH 7.0; 0.105 pmole of pyridoxal phosphate; 15 pmoles of 2mercaptoethanol; and enzyme. This was preincubated at 37” for 15 min, then cooled in an ice bath. Strips of Whatman 3 MM paper (about 15 x 23 mm) were suspended from rubber stoppers by a stainless steel hook, wetted with 0.05 ml of NCS, and allowed to dry for 10-20 min. Then 0.25 ml of &i of 1-14C-~~-DOPA 0.018 M DOPA containing0.1 in 0.02 N HCl was added to each tube. The blank was the same, except that no enzyme was added. The tubes were stoppered with the paper strip centered in the tube and were incubated at 37” for 15 min. After incubation, the tubes were placed in an ice bath and the trichloroacetic acid in the side arm w&9 immediately tipped in and mixed. The tubes were then incubated at 37” for at least 30 min, after which the strips were removed and counted in 10 ml of Omnifluor-dioxane scintillation mixture. The evolution of ‘“CO2 was linear within experimental error as a function of time and enzyme concentration. Although Vogel (17) reported significant nonenzymatic decarboxylation of DOPA under similar conditions, our blanks were consistently low. One unit of activity is defined as that amount of enzyme which produces 1 nmole of COZ per minute under the specified conditions. The other amino acids except 5-hydroxytryptophan were assayed in a similar manner substituting Tris-HCl buffer, pH 8.5, for the sodium phosphate. Histidine decarboxylation was also assayed by the above method but was at the limit of its sensitivity. The fluorometric procedure of Lovenberg et al. (4) was found to be satisfactory and was therefore used. Assay for 6-hydroxytryptophan decarboxylase activity. Because carboxyl-labeled 5-hydroxytryptophan was not available, another type of assay was developed. This assay is based on the standard extraction assay for serotonin (18), modified to permit extraction of serotonin in a single step with relatively little contamination by 5-hydroxytryptophan. The preincubation mixture contained, in a total volume of 1.25 ml : 125 pmoles of Tris-HCl, pH 8.5; 0.105 pmole of pyridoxal phosphate; 15 pmoles of 2-mercaptoethanol; and enzyme. This was preincubated at 37” for 15 min, and cooled in an ice bath. Then 0.25 ml of 0.018 M L-5-hydroxytryptophan (Calbiochem), containing 0.5 pCi of 3-I%-DL-5-hydroxytryptophan (Amersham-Searle), in 0.02 N HCl was added to each tube. The blank was the same, but without enzyme. Incubations were carried out at 37” for 25 min, after which the reaction was stopped by placing the tubes in a boiling water bath for 2 min. They were then cooled on ice, 0.16 g of solid sodium carbonate added, and the tubes extracted with 1.5 ml of a mixture of benzene and butanol. This sol-

358

CHRISTENSON,

DAIRMAN,

vent mixture was prepared by mixing equal volumes of 1-butanol and benzene, then washing the solution with equal volumes of 0.5 N sodium hydroxide, 0.5 N hydrochloric acid, three times with distilled water, and finally 0.08 M Tris-HCI buffer, pH 8.5. Control tubes containing 0.05 bCi of 3-Wserotonin (creatinine sulfate complex, AmershamSearle) and 3 X 10m3M 5-hydroxytryptophan, but no radioactive 5-hydroxytryptophan or enzyme, were also included in order to determine the efficiency of the serotonin extraction, which is about SO%, compared with 0.27, for 5-hydroxytryptophan. A portion (0.5-1.0 ml) of the organic phase was counted in 10 ml of Omnifluor-dioxane scintillation mixture. Analytical ultracentrifugation. Samples of enzyme were dialyzed overnight at 5” against 1000 vol of 0.005 M sodium phosphate, pH 7.2, containing 0.2 M sodium chloride and 0.01 M 2-mercaptoethanol. Centrifugation was performed in a Spinco Model E analytical ultracentrifuge equipped with ultraviolet absorption optics and a photoelectric scanner. A sample of the dialysate was used as blank in all cases. For determination of the molecular weight by the Archibald (19) method, the optical densities at 280 nm from the meniscus or the bottom of the cell to the plateau region were fitted to third-order polynomials as a function of radial distance, using a computer program based on the method of least squares.4 The protein concentrations at the meniscus and bottom were determined by substitution of the radial distances of the meniscus and bottom into these equations and the partial derivatives of concentration with respect to radial distance at the meniscus and bottom were estimated by differentiation and substitut.ion. Amino acid analysis. A solution of enzyme containing 0.60 mg/ml of protein was dialyzed exhaustively against distilled water in the cold room. Two or three milliliters of twice-distilled, constant boiling hydrochloric acid which had been flushed with nitrogen was added to vials containing 0.2 or 0.3 ml of the dialyzed enzyme solution. The vials were then repeatedly evacuated and refilled with oxygen-free nitrogen and sealed under vacuum. Hydrolysis was in a forced-draft oven at 108-111” for 23, 48, or 92 hr, after which the vials were opened and dried in vacua over sodium hydroxide pellets. Cystine and cysteine were determined by the performic acid oxidation method of Moore (20). Tryptophan w&s determined essentially by the method of Duggan and Udenfriend (21) on a sample hydrolyzed in 5 N sodium hydrox4 The computer program used for curve fitting, known as POLFIT, is available through the Genera1 Electric Time Sharing Service.

AND

UDENFRIEND

ide in a polypropylene tube. Acid hydrolysates were analyzed with a Beckman Model 120C instrument. Immunology. About 1 mg of 99+% pure enzyme was subjected to electrophoresis by the method of Davis (13), loading about 1OOpg of protein on each of 10 gels. The gels were stained with 8-anilinonaphthalene sulfonate according to the method of Hartman and Udenfriend (22). Only one fluorescent band per gel was visible under ultraviolet light. The fluorescent bands (34 mm wide) were cut out of the gels, extruded through an l&gauge needle, and allowed to stand in an equal volume of physiological saline overnight in the cold room. An equal volume of complete Freund’s adjuvant was added and thoroughly mixed. The entire mixture was injected subcutaneously into a young, female goat at four sites in the neck region. The antiserum used in this work was taken 3 weeks after injection. RESULTS

Puri$cation

of the Enzyme

,411 procedures were carried out at O&5”, unless otherwise specified. Usually, fresh kidneys were used, but the activity appears fairly stable when the kidneys were stored at -20” for up to 2 weeks. Preliminary treatment. Three hog kidneys (300400 g total wet weight) were defatted, minced, and homogenized in a Waring Blendor for 1 min with approximately 3 vol of 0.005 M sodium phosphate buffer, pH 7.2, containing 0.01 M 2-mercaptoethanol. The crude homogenate was centrifuged at 25,000g for 30 min and the supernatant filtered through cheesecloth. Freshly prepared 20% streptomycin sulfate was added dropwise to the supernatant to bring the concentration of streptomycin sulfate to 0.8 %. The solution was stirred for 15-20 min then centrifuged at 20,OOOgfor 15 min. Ammonium sulfate fractionation. The streptomycin sulfate supernatant (990 ml) was brought to “32% saturation” by the dropwise addition of 466 ml of a neutralized saturated solution of ammonium sulfate. After stirring for a further 20 min, the solution was centrifuged at 25,000g for 20 min and the precipitate discarded. The supernatant was brought to “49 % saturation” by the dropwise addition of one-third its volume of saturated ammonium sulfate. After stirring and centrifuging as before, the pre-

AROMATIC

L-AMINO

ACID

cipitate was resuspended in about 165 ml of 0.05 M sodium phosphate, pH 7.2, containing 0.01 M 2-mercaptoethanol. Heat treatment in the presence of substrates. The redissolved “32-49 % ammonium sulfate fraction” was made up to contain 0.067 M sodium phosphate, pH 7.2, 0.02 M 2-mercaptoethanol, 0.6 M potassium chloride, and 7 X lo+ M pyridoxal phosphate. Sixmilliliter aliquots of this solution were added to 16 X 150-mm tubes containing 0.6 ml of 2 X 10B3M 5-hydroxytryptophan. The tubes were then placed in a 50” water bath for 6 min and immediately cooled in an ice bath. When cold, the precipitated protein was removed by centrifugation at 30,OOOgfor 20 min. The supernatant (about 280 ml) was dialyzed overnight against two changes of 4 1. each of 0.005 M sodium phosphate pH 7.2, containing 0.01 M 2-mercaptoethanol. Alumina Cy gel adsorption. The dialyzed “heat supernatant” (about 310 ml having an optical density at 280 nm of about 20) was adjusted to pH 5.8 with 0.2 N acetic acid and centrifuged at 20,OOOgfor 20 min. To the clear supernatant was added a 3 % suspension of alumina Cr gel to a final concentration of 0.4 mg dry weight of gel per optical density unit. This was stirred for 20 min, then centrifuged and the supernatant discarded. The sedimented gel was resuspended in about 150 ml of 0.01 M 2-mercaptoethanol, centrifuged, and the supernatant discarded. The washed gel was then resuspended in about 200 ml of 0.1 M sodium phosphate buffer, pH 7.2, containing 0.01 M 2-mercaptoethanol, centrifuged, and the clear, yellow “alumina CT gel eluate” was retained. Polyethylene glycol-6000 precipitation. A 40 % solution of polyethylene glycol-6000 was added dropwise to the alumina CT gel eluate to a final concentration of 12.5 %. After stirring for 20 min, the precipitated protein was sedimented at 40,OOOg for 45 min. To the 290 ml of supernatant was added 140 ml of a suspension of DEAE-cellulose (about 27 mg/ml). After stirring for 20 min, the DEAE-cellulose was isolated on a coarse sintered glass funnel and washed with 100 ml of 0.01 M sodium phosphate buffer, pH 7.2, containing 0.01 M 2-mercaptoethanol. The adsorbed material was then

DECARBOXYLASE

359

eluted from the DEAE-cellulose by suspending it in 100 ml of 0.05 M sodium phosphate buffer, pH 7.2, containing 0.6 M sodium chloride and 0.01 M 2-mercaptoethanol, then sedimenting the DEAEcellulose at 25,000g for 20 min. The “PEGDEAE eluate” was dialyzed against two changes of 2 1. each of 0.05 M sodium phosphate buffer, pH 7.2, containing 0.01 M 2-mercaptoethanol. DEAE-Sephadex fractionation. The dialyzed PEG-DEAE eluate was loaded on a 2.5 X 30-cm column of DEAE-Sephadex A-50, which had been equilibrated with 0.05 M sodium phosphate buffer, pH 7.2, containing 0.01 M 2-mercaptoethanol, and rinsed on with two small aliquots of the same buffer. The column was developed at a flow rate of 8-10 ml/hr with a linear gradient of O-O.5 M sodium chloride in a total volume of 1000 ml of the same buffer solution. Fractions of about 10 ml were collected. The fractions containing peak activity, which eluted at about 0.25 M NaCl, were combined. Hydroxylapatite column chromatographyPreparation of hydroxylapatite. Twenty grams of Bio-Gel HTP were suspended in about 400 ml of distilled water at room temperature with very gentle stirring and the fines decanted. This was repeated at least five times. The material was then resuspended and decanted in about 400 ml of cold 0.01 M sodium phosphate buffer, pH 7.2, containing 0.01 M 2-mercaptoethanol a total of three times. It was then poured into a 1.5-cm diameter column to obtain a packed bed height of 25-30 cm and allowed to settle. The column was washed with at least 500 ml of 0.01 M sodium phosphate buffer, pH 7.2, containing 0.01 M 2-mercaptoethanol over a period of at least 24 hr. Chromatography. The pooled DEAESephadex fractions were concentrated by ultrafiltration to about 10 ml and dialyzed overnight against 1 1. of 0.01 M sodium phosphate buffer, pH 7.2, containing 0.01 M 2-mercaptoethanol. The dialyzed material was then loaded on a freshly prepared 1.5 X 27-cm hydroxylapatite column, rinsed in with two or three small aliquots of starting buffer, and eluted with a concave gradient formed by a variable gradient maker. The chambers of the

360

CHRISTENSON,

DAIRMAN, TABLE

SUMMARY

OF PURIFICATION

step

Crude supernatant 3249% Ammonium sulfate fraction Heat supernatant Alumina CT gel eluate Polyethylene glycol-6OOQ precipitation DEAE-cellulose eluate Chromatography on DEAE-Sephadex Chromatography and rechromatography on Hydroxylapatite

AND

UDENFRIEND

I

OF DECARBOXYLASE Volume

(ml)

987 183 283 176 95 50 45

Totaf

FROM

p;Ttein m

2.31 X 5.07 x 3.04 x 1.22 x 396 34.7 4.1

lo4 103 103 103

HOG

KIDNEYS

Total PurificaS&iC actjvity actlvlty tion ReETy 0 ‘“$$’ (units/m.&] (fold)

604 353 214 209 138 68 36

26.1 69.6 70.3 171 373 1960 8670

2.7 2.7 6.6 14.3 75.0 332

58 35 35 23 11 5.9

a These fiaures renresent average values obtained with several preparations, based on 350 g wet The assays were carried out with L-DOPA as subweight of defatted hog kidney as starting material strates under the usual conditions.

gradient mixer contained 90 ml each of the following concentrations of pH 7.2 sodium phosphate buffer: chamber 1, 0.01 nf; chambers 2, 3, 4, 0.05 M; chamber 5, saturated (made up to 0.3 hr at room temperature and allowed to crystallize in the cold room; the actual concentration of the mother liquor was about 0.2 M). All buffers contained 0.01 M 2-mercaptoethanol. Fractions of about 5 ml were collected. The fractions containing peak activity, which eluted at about 0.09 M sodium phosphate, were combined, concentrated to about 10 ml, dialyzed against 0.01 M sodium phosphate, pH 7.2, containing 0.01 31 2-mercaptoethanol, and the chromatography was repeated on a fresh column. Purity of the Enzyme. A typical purification is summarized in Table I. The purifications obtained for each of the last four preparations averaged about 300-fold with recoveries from 5 to 14%. The final specific activities ranged from 7500 to 9500 units/ mg. The purified enzyme was relatively stable during storage having a half-life of about 6 weeks at 0”. The enzyme prepared as described was 97-100% homogeneous, as estimated by polyacrylamide gel disc electrophoresis, staining, and scanning at 580 nm, as described in Materials and Methods. Relative amounts of protein in the bands were estimated by integrating the corresponding peaks in the trace. Typical electrophoresis

patterns are shown in Fig. 1. Preparation A was obtained as described, except that the enzyme was subjected to gel filtration on Sephadex G-100 before the DEAESephadex chromatography step. Note that the slow contaminant present in Preparation A was absent in Preparation B and that the two fast contaminants present in B were absent in A. This suggests that the major band represents the enzyme. Although only a few percentage of the enzyme activity could be recovered5 after the electrophoresis, the activity which was recovered coincided with the major band shown in Fig. 1. The presence or absence of a given contaminant probably depends upon minor differences in the choice of fractions to be saved. Although the pure enzyme appeared as a rather broad band in polyacrylamide gel disc electrophoresis, it yielded antibody which gave a single precipitin line against even the crudest preparation on Ouchterlony double diffusion (Fig. 2). If the pure enzyme had contained more than a single protein, one would have expected to find two or more precipitin lines or to see precipitin lines which crossed, rather than merged. A control experiment using normal goat serum instead 6 Enzyme activity was recovered by extruding l-mm gel slices through an Wguage needle in the presence of three times the normal amount of sodium phosphate buffer, pH 7.0, directly into assay tubes and assaying in the usual way.

AROMATIC

L-AMINO

ACID

FIG. 1. Polyacrylamide gel disc electrophoresis of purified aromatic L-amino acid decarboxylase. Direction of migration was from top to bottom. From left to right: (1) 20 pg of Preparation A at pH 8.3; running time 2 hr 20 min. The contaminant running just behind the major band represents about 37, of the total protein. (2) 10 pg of Preparation B at pH 8.3; running time 1 hr 30 min. The two contaminants running ahead of the major band, combined, represent less than 1% of the total protein. The mark near the lower end of the gel shows the position of the tracking dye. (3) 20 fig of Preparation A at pH 7.5; running time 1 hr 20 min. (4) 3Opg of Preparation B at pH 7.5; running time 1 hr 20 min. The opaque material at the top of the pH 7.5 gels is the stacking gel, which adheres tightly in this system.

of antiserum showed no reaction with any of the antigen preparations. On velocity sedimentation at 52,000 rpm and 13” or 33”, a single, symmetrical boundary of optical density was obtained, as shown in Fig. 3. Further sedimentation data which are in agreement with the above criteria of purity are reported in a subsequent section. Properties of the PuriJied Enzyme Requirement of a free sulfhydryl group. Fellman (23) and Lovenberg et al. (4) re-

DECARBOXYLASE

361

FIG. 2. Ouchterlony double diffusion experiment. The center well contained 50 ~1 of goat anti-enzyme serum. The outer walls contained, clockwise : 1 o’clock, early fractions from a DEAESephadex column containing protein, but no enzyme activity (about 12 pg of protein); 3 o’clock, 400 pg of heat-treated supernatant; 5 o’clock, 200 pg of alumina CT gel eluate; 7 o’clock, DEAESephadex fractions containing peak enzyme activity (about 20 pg of protein); 9 o’clock, about 100 pg of polyethylene glycol-6000 supernatant, eluted from DEAE-cellulose; 11 o’clock, 4 fig of purified enzyme, estimated to be 970jo homogeneous. Development was for approximately 40 hr at 5”.

FIG. 3. Rapid sedimentation of aromatic Lamino acid decarboxylase. Rotor speed was 52,COO rpm, temperature 13’. Other conditions were as specified in Materials and Methods. The enzyme sample was estimated to be 97y0 homogeneous by disc gel electrophoresis.

ported inhibition of their preparations of DOPA decarboxylase when assayed in the presence of sulfhydryl reagents. The effects of pretreatment with some sulfhydryl reagents and sulfhydryl compounds on the purified enzyme are summarized in Table II. This study differs from the previous studies in that the excess sulfhydryl reagents were

362

CHRISTENSON,

DAIICMAN,

destroyed before assaying. That p-chloromercuribenzoate resulted in only partial inhibition may be due to the reversibility of mercaptide formation in the presence of excess 2-mercaptoethanol. Eflect of pyricloxal phosphate. The purified enzyme was found to contain 0.7-1.1 mole of pyridoxal phosphate per 112,OOOgof protein. TABLE

II

EFFECTS OF SOME SULFHYDRYL REAGENTS SULFHYDRYL COMPOUNDS ON THE DECARBOXYLATION OF DOPAR

COPevolved (nmoles)

Treatment Control (incubated with P-Mercaptoethanol Dithiothreitol p-Chloromercllribenzoate Iodoacetamide N-Ethylmaleimide

ANL)

H#)

255 261 271 61 18 6

Q A sample of the enzyme was dialyzed free of 2-mercaptoethanol. Aliquots of the dialyzed enzyme were then incubated in the presence of the compounds shown in the Table for 1 hr at room temperature in 8 X lo-$ M sodium phosphate buffer, pH 7 (except p-chloromercuribenzoate, in which case the buffer was 0.012 M sodium phosphate, pH 7.8). The incubation mixtures were then chilled on ice and a IO-fold excess of cold 0.05 M 2-mercaptoethanol was added to destroy unreacted sulfhydryl reagents. The treated enzyme was then assayed by the usual procedure except that no ftlrther 2-mercaptoethanol was added.

f2 0

10-7

ANl)

UI)F:SFRIEKI)

The exciiation and emission spectra of the fluorescent derivative of t,he enzyme-bound material agreed well with those of pyridoxal phosphate standards and with published spectra (11). The enzyme catalyzed significant decarboxylation without exogenous pyridoxal phosphate, but the addition of 1 X lo-’ to 1 X 1O-3M pyridoxal phosphate stimulated the activit)y two- to fivefold, with a broad maximum in the range of 5 X 10e6 to 1 X 1O-4 M (Fig. 4). To test, whet,her inhibition of DOPA decarboxylation by pyridoxal phosphate at concentrations above 1 X 1OV M was due to t,he formation of the tetrahydroisoquinoline cyclization product (24), t,he effect of pyridoxal phosphate upon the decarboxylation of phenylalanine was also tested. The 1at)ter does not form a cyclic product with pyridoxal phosphate. With phenylalanine as substrate, there was twoto threefold stimulation throughout the range of 1 X lo-’ to 1 X 1O-3 M pyridoxal phosphate, reaching a plateau at. approximately 1 X 1O-6 M. Thus, inhibition of the decarboxylation of DOPA by the higher pyridoxal phosphate concent,rations is probably due to formation of the tetrahydroisoquinoline derivative. The experiment, with phenylalanine also shows that the sites responsible for act.ivat,ion by exogenous cofactor approach saturation at, 1OV M pyridoxal phosphate. Attempts to resolve the holoenzyme com-

10-6 10-s Pyridoxal Phosphate (Ml

10-4

lo- 3

FIG. 4. Dependence upon pyridoxal phosphate concentration of the decarboxylation of DOPA and phenylalanine. The concentrations of DOPA and phenylalanine were 0.003 and 0.017 M, and the incubation times were 15 and 25 min, respectively.

AROMATIC 5-

L-AMINO

ACID

363

DECARBOXYLASE TABLE

Buffers: b and

r,Phosphate

Iq and 3 ond

n ,Glycine l ,citrate

III

KINETIC PARAMETERS OF AROMATIC L-AMINO ACID DECARBOXYLASE FROM HOG KIDNEYS &It CM)

Sub&rat@

VU% bloles 1 min /mg)

o-

DOPA 5..Hydroxytryptophan Phenylalanine Tryptophan Tyrosine

DH

FIG. 5. Dependence upon pH of the decarboxylation of phenylalanine and tryptophan by aromatic L-amino acid decarboxylase. The enzyme preparation used was estimated to be about 90% homogeneous by disc gel electrophoresis. Assays were performed in the presence of 0.08 M buffer and either 0.01 M tryptophan containing 0.2 pCi of 1-i4C-tryptophan or 0.017 M phenylalanine containing 0.5 pCi of 1-W-phenylalanine. Incubation time was 25 min. The buffers used were sodium citrate for pH 5.5 and 6.0; sodium phosphate for pH 6.0-7.5; Tris-HCl for pH 7.59.0; and sodium glycinate for pH 9.c10.5.

plex have not been successful. An attempt to bind more cofactor to the enzyme by incubation with pyridoxal phosphate followed by dialysis increased the pyridoxal phosphate content of the enzyme by about 50%, but there was no significant change in the stimulation afforded by exogenous pyridoxal phosphate. pH Dependence. The effect of pH on the decarboxylation of phenylalanine and tryptophan is shown in Fig. 5. DOPA and its decarboxylation product dopamine are unstable at alkaline pH, so DOPA decarboxylation was always assayed at pH 7.0 (4, 25, 26). Substrate specificity and kinetics. Results of the studies of the specificity and kinetics of the enzyme were in general agreement with those of Lovenberg et al. (4). The enzyme was found to decarboxylate DOPA, 5hydroxytryptophan, phenylalanine, tryptophan, and tyrosine at readily measurable rates. No substrate inhibition was observed

1.9 1 4.2 1.0 8.4

x x X x x

10-d lcr4 lo10-z 10-g

8900 850 590 230 30

a Kinetic measurements were done in the assay systems described in Materials and Methods; DOPA was assayed at pH 7.0, all others at pH 8.5. The enzyme preparation used was essentially homogeneous, over 99% pure. The concentration of carrier amino acid was varied while the amount of radioactive substrate was held constant. K,,, and V=., were calculated from plots of substrate concentration/velocity vs. substrate concentration fitted by the method of least squares using unweighted data. 6 In a preliminary study (see footnote 1) W-CO2 was evolved from uniformly labeled alanine. In subsequent studies this could not be repeated with other sources of radioactive labeled alanine. Furthermore radioactive ethylamine was not formed.

with DOPA at concentrations as high as 6 X 1O-3 M. A summary of the kinetic parameters obtained with each amino acid is given in Table III. The activity of the enzyme toward histidine was very low but detectable by both radiometric and fluorometric assays. Kinetic measurements were not made, but the histidine and tyrosine activities were compared. At substrate concentrations of 2 X 10m3M, the rate of decarboxylation of tyrosine was 2.0 X 10m2 nmole/min while the rate of histidine decarboxylation was 2.5 X 10e3nmoles/min. None of the other naturally occurring amino acids was decarboxylated to any detectable extent. Efects of cations. No evidence was found to suggest that metal ions serve an important function in the decarboxylation of DOPA. The ions Fe2+, Fe3+, K+ were either weakly inhibitory or without effect, while Mg2+, Caz+, and A13+enhanced the activity by only 10% or less. Cu2+, Zn2+, and Hg2+ were strongly inhibitory.6 Furthermore, the che6 The concentration 1 x lo-3M.

of

metal

ions

used

was

364

CHRISTENSON,

DAIRMAN,

lating agents sodium diethyldithiocarbacupferron, 1, lomate, 2,2’-bipyridine, phenanthroline, sodium EDTA, and 1,5diphenylcarbohydrazide had no significant effect at concentrations of 1 X 10m4M. Amino acid analyses. Amino acid analyses were performed as described in Materials and Methods. The results are shown in Table IV. No significant, time-dependent destruct’ion of serine and threonine was observed, although this is generally the case under the conditions of hydrolysis which were used. However, time-dependent losses of proline, lysine, and arginine were observed and such losses have been reported TABLE AMINO

ACID

ACID

Grams Amino

IV

OF AROMATIC DECARBOXYLASE

COMPOSITION

acid

Aspartic acida Threonine Serine Prolineb Glutamic acid0 Glycine Alanine Valinec Half-Cystined Methionine Isoleucine” Leucinec Tyrosine Phenylalanine Lysineb Histidine Arginine” Tryptophan”

Time

of residue/100 protein;

L-AMINO

g

of hydrolysis

23 hr

48 hr

92 hr

6.80 3.06 4.29 4.28 12.46 4.36 6.87 5.74 2.08 2.62 3.92 11.98 4.30 7.63 5.28 3.06 8.20 3.07

6.62 3.31 4.40 4.04 12.50 4.58 7.19 5.88 2.67 4.14 12.06 3.94 7.86 4.89 3.29 7.70 -

6.82 3.22 4.42 3.82 12.91 4.96 7.46 5.77 2.85 4.20 12.93 4.09 6.84 4.35 2.92 7.53 -

&;i$;;/ p&in: “Best” integral V&e

64 35 55 50 107 89 110 64 21 23 41 125 28 55 48 25 58 18

a These figures include both free and amidated residues of aspartic and glutamic acids. b Values for proline, lysine, and arginine were determined by extrapolation to zero time of hydrolysis . c Values for valine, isoleucine, and leucine were taken at the maximum time of hydrolysis. d Cystine and cysteine were determined as cysteic acid as described in Materials and Methods. e Tryptophan was determined fluorometrically from a basic hydrolysate.

AND

UDENFRIEND

previously (27). Accordingly, the values for these three amino acids were determined by extrapolation to zero time of hydrolysis, assuming first-order kinetics. The recovery of methionine was not improved by the addition of one part per 2000 of 2-mercaptoethanol (28) or when it was determined as methionine sulfone in the performic acidoxidized sample. The excitation and emission spectra of the basic hydrolysate agreed well with those of the tryptophan standards and with published results (21). Recoveries of amino acids after acid hydrolysis accounted for 113-127% of the total protein as estimated by the Lowry (12) method. Thus, the enzyme protein concentrations used in this work may be low by a corresponding factor. Ultracentrifuge studies. The sedimentation constant corrected to standard conditions, S 20,~, was found to be 5.82 f 0.03 based on four determinations. The sensitivity of the absorption optics is such that this value, determined at a protein concentration of about 0.2 mg/ml, is a good estimat’e of the sedimentation constant at infinite dilution. The molecular weight of the enzyme was calculated from sedimentation equilibrium using a preparation estimated to be greater than 97 % homogeneous by disc gel electrophoresis. The rotor speed and temperature were 10,000 rpm and 4”. A molecular weight of 112,000 daltons was calculated using a partial specific volume of 0.742 ml/g, calculated from the amino acid composition by the method of Cohn and Edsall (29). Early in the run, 128 min after reaching speed, estimates of the molecular weight at the meniscus and at the bottom of the cell were made by the Archibald (19) technique, as described in Materials and Methods. The were molecular weights so determined 108,000 and 107,000 daltons, respectively, in good agreement mutually, and with the molecular weight obt#ained from sedimentation equilibrium. DISCUSSION

The result#s reported above demonstrate conclusively for the first time that a single homogeneous protein is capable of decarboxylating both DOPA and 5-hydroxytryp-

AROMATIC

L-AMINO

ACID

tophan, as well as the other aromatic amino acids and hi&dine. In view of this broad specificity and of the stereospecificity previously determined by Lovenberg et al. (4), we have used the term “aromatic L-amino acid decarboxylase.” Although Fellman (23) has reported the preparation of a homogeneous decarboxylase with similar properties from bovine adrenal medulla, serious questions must be raised concerning the purity of Fellman’s preparation. For example, his most highly purified preparation was less than 20-fold purified from a high-speed supernatant, compared with over 300-fold purification in the present study. Furthermore, Fellman reported a specific activity of only 158 ~1. COz/hr/mg of protein, or about 120 nmoles/min/mg, compared with over 8600 nmoles/min/mg in the present study. Fellman’s criteria of purity were the presence of a single schlieren peak during velocity sedimentation and a single protein peak in paper and free boundary electrophoresis. The staining procedures used were not described. By current standards, none of these is a particularly reliable or sensitive indicator of homogeneity. We cite the following evidence for the homogeneity of our preparation: presence of a single band in polyacrylamide gel disc electrophoresis in two different buffer systems; demonstration that t,he electrophoretically pure enzyme contains only a single antigenic species by Ouchterlony double diffusion; presence of a single symmetrical boundary of optical density during velocity sedimentation; linearity of plots of In c as a function of r2 at sedimentation equilibrium; and agreement of molecular weights determined at the meniscus and at the bottom of the cell by the Archibald method. Although the enzyme preparations were homogeneous by all the above criteria, it, was surprising to obtain such homogeneity after only 300-fold purification of a mammalian enzyme. The possibility that the degree of purification is misleading due to the presence of an activator in the crude material or an inhibitor in the purified enzyme was investigated. This was ruled out by showing that the activities of purified enzyme and crude supernatent were additive. Taken at face value, the data suggest that the decarboxy-

DECARBOXYLASE

365

lase represents approximately 0.3% of the soluble protein of hog kidney. Kidneys from other species are also known to be rich in this enzyme (2, 5). The significance of such high concentration of this enzyme in the kidney is not apparent. A turnover number for DOPA of about 1000 mole/mole of enzyme may be calculated, based on the average specific activity of homogeneous enzyme. It is interesting to compare this figure with the turnover number of 1060, reported by Friedman and Kaufman (30) for dopamine /3-hydroxylase, the third enzyme in the norepinephrine biosynthetic pathway. Although the purification procedure we used was quite straightforward, some comments on the steps involving heat and polyethylene glycol are in order. Ear&y experiments with relatively broad ammonium sulfate cuts suggested that significant purification could be achieved by heating at 55” for 6-10 min in the presence of substrate and cofactor. However, this was not borne out after the ammonium sulfate fractionation procedure had been refined. Nevertheless, the heating step was retained in the final purification scheme, since the later chromatographic steps seemed to be more effective with a heat-treated preparation. The use of polyethylene glycol as a precipitant for proteins has not received widespread attention, although Poison et al. (31) have developed detailed procedures for its use in the fractionation of serum proteins and Janssen and Ruelius (32) have used polyethylene glycol-6000 as the only reagent in their elegant purification of a fungal alcohol oxidase. We found the more concentrated solutions too viscous for convenient handling at cold room temperatures. However, with the batchwise use of DEAE-cellulose to remove the enzyme from a 12.5% polyethylene glycol-6000 solution, the technique is simple and effective. It is generally accepted that pyridoxal phosphate is the cofactor for the enzyme. The nature of the binding of the coenzyme, however, has not been clear. It is possible that some of the conflicting results are due to differences among species. Clark et al. (2) and Lovenberg et al. (33) found the coenzyme to be tightly bound to the enzyme

366

CHRISTENSON,

DAIRMAN,

from guinea pig kidney and could not demonstrate significant stimulation by the addition of pyridoxal phosphate. Awapara et al. (5) and Buzard and Nytch (34) (using rat liver respectively) and kidney preparations, found that adding pyridoxal phosphate resulted in significant stimulation of their preparations, but they did not separate the coenzyme from the apoenzyme. Fellman’s from bovine adrenal (23) preparations medulla exhibited an obligatory requirement for the cofactor. Werle and Aures (26) were able to resolve their enzyme preparation (from guinea pig kidney) by extensive dialysis against EDTA and found a K, of 9 X lo-* M for the apoenzyme-coenzyme complex. Thus, the cofactor appears to be tightly, but not irreversibly bound. Our results are consistent either with this hypothesis or with the proposal of Awapara et al. (5) that a portion of the coenzyme is tightly bound, while another portion is easily removed. Further studies of this problem are in progress. Although it is clear that at least one sulfhydryl group is required for activity, the mechanism of its involvement is not known. While sulfhydryl groups have been implicated in the binding of coenzyme in certain pyridoxal phosphate-dependent enzymes (for example, aspartate aminotransferase, Ref. 35), this may be due to stabilization of a favorable conformation, rather than to direct interaction with the cofactor, as was concluded by Tate and Meister (36) in the case of aspartate B-decarboxylase. Given the molecular weight of 112,000 and the sedimentation constant of 5.82 X lo-l3 see, one may calculate a diffusion coefficient of 4.88 X lo-’ cm2/sec. This in turn leads to a frictional ratio, flfmin, of 1.36. This ratio is a function both of the shape of the protein and of its degree of solvation. Assuming a reasonable degree of hydration, 0.2 g H,O/g protein, one may calculate that the frictional ratio due to shape alone, f/f0 , is about 1.26. For a prolate or oblate ellipsoid of revolution, this value corresponds to an axial ratio of approximately five. Even for a rather high degree of hydration, say 0.5, the axial ratio is about 3.5. Thus, the enzyme appears to be a relatively asymmetric molecule.

AND UDENFRIEND

Although the plot of In c vs. r2 for sedimentation equilibrium at 4” was linear, at 22” the plot was concave. This nonlinear behavior most likely indicates an associatingdissociating system at elevated temperatures. Molecular weights calculated from the slope of the curve obtained at 22’ ranged from 30,000 near the meniscus to greater than 90,000 near the bottom of the cell. Polyacrylamide gel electrophoresis in the presence of the denaturant sodium dodecyl sulfate, essentially according to Weber and Osborn (37), gave three bands, corresponding to molecular weights of about 57,000, 40,000, and 21,000. It should be noted that gel electrophoresis without denaturation yielded broad bands with enzyme which was characterized as homogeneous by all the other criteria. This also suggests that purified preparations contain multiple forms of the enzyme. ACKNOWLEDGMENTS We thank Dr. Paul Bartl for the use of the analytical ultracentrifuge. Thanks are also due to Miss Barbara Segen and Messrs. James Stone and Daniel Luk for their skillful technical assistance. REFERENCES 1. HOLTZ, 2. 3. 4. 5. 6. 7. 8.

P., HEISE,

R., AND LUDTKE, K., Arch. Ezp. Pathol. Pharmakol. 191,87 (1938). CLARK, C. T., WEISSBACH, H., AND UDENFRIEND, S., J. Biol. Chem. 210,139 (1954). SOURKES, T. L., Pharmacol. Rev. 18, 53 (1966). LOVENBERG, W., WEISSBACH, H., AND UDENFRIEND, S., J. Biol. Chem. 237, 89 (1962). AWAPARA, J., SANDMAN, R. P., AND HANLY, C., Arch. Biochem. Biophys. 63, 520 (1962). HAGEN, P., Brit. J. Pharmacol. 18,175 (1962). COULSON, W. F., HENSON, G., AND JEPSON, J. B., Biochim. Biophys. Acta166.135 (1968). STREFFER, C., Biochim. Biophys. Acta 139,

193 (1967). 9. COULSON, W. F., BENDER, D. A., AND JEPSON, J. B., Biochem. J. 116, 63P (1969). 10. CROUT, R. J., Stand. Methods CZin. Chem. 3, 62 (1961). 11. ADAMS, E., Anal. Biochem. 31, 118 (1969). 12. LOWRY, 0. H., ROSEBROUGH, N. J., FARR, A. J., AND RANDALL, R. J., J. BioZ. Chem. 193, 265 (1951). 13. DAVIS, B. J., Ann. N.Y. Acad. Sci. 121, 404 ww. 14. WILLIAMS, D. E., AND REISFELD, R. A., Ann. N.Y. Acad. Sci. I!& 373 (1964).

AROMATIC

L-AMINO

15. RHOADS, R. E., AND UDENFRIEND, S., Proc. Nat. Acad. Sci. U.S. 60, 1473 (1968). 16. ELLENBOGEN, L., MARKLEY, E., AND TAYLOR, R. J., Biochem. Pharmacol. 18, 683 (1969). 17. VOGEL, W. H., Naturwissenschaften 66, 462 (1969). 18. UDENFRIEND, S., WEISSBACH, H., AND CLARK, C. T., J. Biol. Chem. 216,337 (1955). 19. ARCHIBALD, W. J., J. Phys. Chem. 61, 1204 (1947). 20. MOORE, S., J. Biol. Chem. 238, 235 (1963). 21. DUGGAN, D. E., AND UDENFRIEND, S., J. Biol. Chem. 393, 313 (1956). 22. HARTMAN, B. K., AND UDENFRIEND, S., Anal. Biochem. 30. 391 (1969). 23. FELLMAN, J. H., Enzymologia 20, 366 (1959). 24. SCHOTT, H. F., AND CLARK, W. G., J. BioZ. Chem. 196, 449 (1952). 25. SCHALES, O., AND SCHALES, S., Arch. Biochem. Biophys. 24, 83 (1949). 26. WERLE, E., AND AURES, D., Z. Physiol. Chem. 316, 45 (1959). 27. DUGGAN, E. L., Methods Enzymol. 3.492 (1957).

ACID

DECARBOXYLASE

367

28. KEUTMANN, H. T., AND POTTS, J. T., Anal. Biochem. 29, 175 (1969). 29. COHN, E. J., AND EDSALL, J. T., “Proteins, Amino Acids, and Peptides,” p. 370. Reinhold, New York (1943). 30. FRIEDMAN, S., AND KAUFMAN, S., J. BioZ. Chem. 240, PC552 (1965). 31. POLSON, A., POTGIETER, G. M., LARGIER, J. G., MEARS, G. E. F., AND JOUBERT, F. J., Biochim. Biophys. Acta 82, 463 (1964). 32. JANSSEN, F. W., AND RUELIUS, H. W., Biochim. Biophys. Acta 161, 330 (1968). 33. LOVENBERG, W., BARCHAS, J., WEISSBACH, H., AND UDENFRIEND, S., Arch. Biochem. Biophys. 103, 9 (1963). 34. BUZARD, J. A., AND NYTCH, P., J. BioZ. Chem. 234, 884 (1959). 35. TURANO, C., GIARTOSIO, A., AND FASELLA, P., Arch. Biochem. Biophys. 104, 524 (1964). 36. TATE, S. S., AND MEISTER, A., Biochemistry 7, 3240 (1968). 37. WEBER, K., AND OSBORN, M., J. BioZ. Chem. 244, 4406 (1969).