THE JOURNAL OF BIOLOQICAL CHEMISTRY Vol. 246. No. 16, Issue of August 25, pp. 5085-5092, 1971

Printed in U.S.A.

Arginine Racemase of Pseudomonas graveolens

I. PURIFICATIOK, CRYSTALLIZATION, AKD PROPERTIES*

(Received for publication, December 31, 1970)

TAKAMITSU YORIFUJI$ AND KOICHI OGATA

From the Department of Argicultural Chemistry, Kyoto University, Kyoto, Kycto-Fu 606, Japan

KFNJI SODAS

From the Institute for Chemical Research, Kyoto University, Uji, Kyoto-Fu 611, Japan

SUMMARY

Arginine racemase has been purified approximately 5400- fold and crystallized from an extract of Pseudomonas graveo- lens. The puritication procedure consists of sonic disruption, ammonium sulfate fractionation, treatment with 1-butanol, DEAE-cellulose and DEAE-Sephadex column chromatog- raphy, and Sephadex G-150 gel filtration, followed by crystal- lization. The purit?ed enzyme is homogeneous by the criteria of ultracentrifugation(&,, = 5.2 S) and disc gel electrophore- sis. The molecular weight is 167,000, assuming a partial specific volume of 0.74.

The enzyme is yellow in solution and exhibits an absorption maximum at 420 mp. The absorption spectrum is shifted neither by varying pH of the enzyme nor by addition of argi- nine. One mole of pyridoxal 5’-phosphate is tightly bound per 42,000 g of enzyme. The apoenzyme was obtained by dialysis against phenylhydrazine solution. Addition of pyridoxal 5’-phosphate caused return of the 420 rnp absorp- tion peak and reconstitution of activity. Reduction of argi- nine racemase with sodium borohydride shifted the 420 rnp peak to about 315 mp, and destroyed the racemase activity. The enzyme catalyzes racemization of lysine, arginine, E-N-acetyllysine, ornithine, 2,3 -diaminopropionate, homoargi- nine, 2,4-diaminobutyrate, ethionine, citrulline, homocitrul- line, b-N-acetylornithine, theanine, glutamine, andmethionine in this order, when examined at pH 10.0. Maximum activi- ties for arginine, lysine, theanine, and ornithine are found in the pH regions of 9.0 to 10.6, 7.5 to 10.6, 7.0 to 10.0, and 6.5 to 9.0, respectively. The following Michaelis con- stants were determined: D-arginine, 1.0 X 10e3 M, and pyridoxal 5’-phosphate, 4.0 x 1O-7 M. The enzyme activity was inhibited by hydroxylamine and D- and L-penicillamine.

Ample evidence has been obtained for the occurrence of D-

amiiio acids in bacteria and animals, and for t,heir presence in

* h preliminary report of part of this work has been presented (1).

1 Present address, Department of Agricultural Chemistry, Shinshu University, Ina, Nagano-Ken 396, Japan.

5 To whom all correspondence regarding this work should be addressed.

cell wall materials and antibiotics, as reviewed by Meister (2) and Corrigan (3). There has been increasing interest in the properties and physiological functions of amino acid racemases since this is the most plausible mechanism of interconverting D- and n-amino acids. In recent years, a few amino acid race- mases have been purified to homogeneity or near homogeneity; hydroxyproline 2-epimerase from Pseudomonas striata (4)) alanine racemase from Pseudomonas sp. (5) and from Pseudomonas putida (6), proline racemase from Clostridium sticklandii (7), and amino acid racemase with low substrate specificity from P. striatu (8).

We have found a new amino acid racemase catalyzing the conversion of either D- or L-arginine to the racemate in the cell- free extract of Pseudomonas gruveolens (9). In the present paper, the preparation of crystalline arginine racemase (EC class 5.1. l), and some of its properties are described.

EXPERIMENTAL PROCEDURE

,Vuterials

n-Arginine-HCl and L-arginine-HCl were obtained from Gen- eral Biochemicals, Chagrin Falls, Ohio, and from Tanabe Seiyaku Company, Osaka, respectively. L-Homoarginine-HCI was prepared by the method of Odo and Ichikawa (10). The other L-amino acids and n-amino acids were products of Kyowa Hakko Kogyo Company, Tokyo, and Ajinomoto Company, Tokyo, respectively. n-Cycloserine was obtained from Shionogi Seiyaku Company, Osaka; D- and n-penicillamine were from Calbiochem; pyridoxal 5’-phosphate was from Dainippon Seiyaku Company, Osaka; DEAE-cellulose was from Midori Juji Company, Osaka; pyridoxamine 5’-phosphate, pyridoxalLHC1, and pyridoxamine- HCl were from Sigma; DEAE-Sephadex A-50, and Sephadex G-150 was from Pharmacia, Uppsala. Pyridoxal 5’-phosphate and pyridoxamine 5’-phosphate were chromatographically purified by the procedure of Peterson and Sober (11). The other chemicals were analytical grade reagents.

Glycine-KCl-KOH buffer was made up by adding 0.1 M KOH to a mixture containing 0.1 M glycine and 0.1 M KC1 to adjust the pH to 10.0. KC1 and KOH were replaced by NaCl and NaOH in glycine-NaCl-NaOH buffer. Arginase was extracted from bovine liver and partially purified according to the proce- dure of Hunter and Dauphinee (12). This preparation con- tained enough manganese ion to activat,e the enzyme.

5085

by guest on January 23, 2020http://w

ww

.jbc.org/D

ownloaded from

5086 Crystalline Arginine Racemase Vol. 246, Xo. 16

Methods

Microorganism and Conditions of Culture

P. graveolens IF0 3460, which was obtained from Institute for Fermentation, Osaka, Japan, was grown in a medium com- posed of 1.0% peptone, 0.2% yeast extract, 0.05% n-arginine- HCl, 0.5% NaCl, and 0.1% K2HPOI. The pH was adjusted to 7.2 with 3 N NaOH. The cultures were carried out in a lo- liter jar fermentor containing 7.5 liters of the medium at 30” for 20 hours under aeration (8 liters per mm). The cells were harvested by centrifugation and washed twice with 0.85% NaCI. The usual yield of cells was approximately 10 g, wet weight, per liter of the medium. The washed cells were stored frozen at - 20” until used.

Enzyme Assay

Procedure A-The standard assay system consisted of 10 pmoles of n-arginine-HCl adjusted to about pH 10, 40 pmoles of glycine-KCI-KOH buffer (~1-1 lO.O), 10 nmoles of pyridoxal 5’-phosphate, and enzyme in a final volume of 1.0 ml. Enzyme was replaced by water in a blank. The mixture was incubated at 37” for 20 min. The reaction was stopped by immersing the tubes in boiling wat,er for 3 min. After cooling, 3.0 mg of the arginase preparation, 0.2 pmole of NH20H-HCl, and water were added to the mixture to a final volume of 2.4 ml, followed by incubation at 37” for 50 min. The slight formation of urea from n-arginine by this arginase preparation was observed in the absence of NH120H-HCl. The arginase reaction was ter- minated by addition of 0.6 ml of 25y0 trichloroacetic acid. After the precipitate was removed by centrifugation, urea re- leased from n-arginine was determined by a modification of the method of Archibald (13) as follows. To 1.0 ml of the clear supernatant, 1 .O ml of 1 y0 diacetyl monoxime dissolved in 5% acetic acid and 3.0 ml of 3.7 N HCl were added, and the mixture was heated in a test tube with a glass stopper in boiling water for 40 min. After cooling, the absorbance was measured at 470 mp. The amount of n-arginine formed was calculated by reference to a standard curve prepared with authentic n-arginine. Citrulline and semicarbazide interfered with the determination of urea by this procedure.

Procedure B-The standard reaction mixture contained 80 pmoles of n-amino acid adjusted to about pH 10, 100 pmoles of glycine-KCl-KOH buffer (pH lO.O), 10 nmoles of pyridoxal 5’.phosphate, and enzyme in a final volume of 2.0 ml. After incubation at 37” for 20 min, the racemization was terminated by addition of 0.2 ml of 50% trichloroacetic acid. The mixture was centrifuged if necessary. The optical rotation at 320 to 350 rnp was measured at 20” on a Yanagimoto recording spec- tropolarimeter, model 185A. The values were corrected for blanks without enzyme or without substrate. Procedure B was used chiefly for substrate specificity study. Good agreement between these procedures was found.

Protein Determination

Protein was determined by the method of Lowry et al. (14) with crystalline egg albumin as a standard; with most column fractions, protein elution patterns were estimated by the 280 nm absorption. Concentrations of the purified enzyme were derived from the absorbance at 280 mp. The extinction CO-

efficient (E& = 9.3) was obtained by absorbance and dry weight determinations.

Dejkition of Units and Specijk Activity

One racemase unit was defined as the amount of enzyme re- quired to convert 1 pmole of L (or n)-amino acid into the D

(or L) isomer per min. Specific activity was expressed as units per mg of protein.

Spectrophotometry

Spectrophotometric measurements were made with a Shimadzu IVIPS5OL recording spectrophotometer or with a Beckman DU spectrophotomet.er with a l.O-cm light path.

Ultracentrifugal Analysis

A Spinco model E ultracentrifuge was used throughout. The molecular weight of enzyme was determined by the ultracen- trifugal sedimentation equilibrium method according to the procedure of Yphantis (15). The experiments were carried out in a Spinco model E ultracentrifuge equipped with schlieren optics. In order to perform the experiment on four samples of different initial concentrations ranging from 0.1 to 0.4%, an eight-channel centerpiece and an An-D rotor were used. The rotor was centrifuged at 8766 rpm at 10” for 8 hours and 20 min. Schlieren patterns were photographed with Fuji pan- chromatic process plate at intervals of 30 min to compare and make sure equilibrium was established.

RESULTS

Puriifcation of Enzyme

The purification procedure of the enzyme previously reported (1) was modified to obtain the better yield of crystals as follows. All operations were performed at O-5”, unless otherwise specified.

Step 1: Preparation of Cell Extract-The washed cells (about 1800 g, wet weight) were suspended in 5 liters of 0.01 M potas- sium phosphate buffer, pH 7.3, and disrupted in 500.ml portions by treatment for 30 min in a 19-kc Kaijo Denki ultrasonic dis- integrator.

Step 2: Ammonium sulfate Fractionation-To the turbid mix- ture containing the extract and cell debris (about 240 g of pro- tein) was added ammonium sulfate to 3Oa/, saturation. The precipitate was collected by centrifugation at 15,000 x g for 15 min. After washing with 2 liters of 0.01 M potassium phosphate buffer, pH 7.3, containing 30% saturation of ammonium sulfate, the precipitate was suspended in 0.01 M potassium phosphate buffer, pH 7.3, and dialyzed for 22 hours against two changes of the same buffer (200 volumes).

Step 3: I-Butanol Treatment-The enzyme contained in the precipitate from Step 2 was solubilized as follows. To a l-liter portion of the suspension, cooled to O-2”, were added slowly 250 ml of chilled I-butanol ( -5”) under vigorous stirring. After standing for 30 min under stirring, the mixture was brought to 35% saturation with ammonium sulfate. The aqueous layer obtained by centrifugation was dialyzed for 22 hours against four changes of 0.01 M potassium phosphate buffer, pH 7.3, containing 2 X 1O-5 M pyridoxal 5’-phosphate (200 volumes). The dialyzed enzyme was concentrated by addition of ammonium sulfate (7570 saturation). The precipitate was dissolved in a small volume of the dialysis buffer and again dialyzed for 18 hours against 200 volumes of the same buffer.

Step 4: DEAE-cellulose Column Chromatography-The enzyme solution was applied to a DEAE-cellulose column (5.5 X 45 cm)

by guest on January 23, 2020http://w

ww

.jbc.org/D

ownloaded from

Issue of August 25, 1971 T. Yorifuji, K. Ogata, and K. Soda 5087

EFFLUENT ( ML I

FIG. 1. Chromatography of arginine racemase on DEAE- Sephadex. The flow rate was approximately 0.8 ml per min and 5-ml fractions were collected. 0, absorbance at 280 rnp; A,, arginine racemase activity determined by Procedure A. Buffer I: 0.01 M potassium phosphate buffer, pH 7.3, containing 0.1 M sodium chloride and 2 X 10e6 M pyridoxal 5’-phosphate. Buffer II: 0.01 M potassium phosphate buffer, pH 7.3, containing 0.15 M sodium chloride and 2 X 10mr’ M pyridoxal 5’-phosphat,e. Other conditions are given in the text.

equilibrated with 0.01 M potassium phosphate buffer, pH 7.3, containing 2 X 1O-5 M pyridoxal 5’-phosphate. After the col- umn was washed thoroughly with the buffer containing 0.07 M

sodium chloride, the enzyme was eluted with the buffer supple- mented with 0.14 M sodium chloride. The active fractions were pooled (1800 ml), concentrated by addition of ammonium sulfate (757; saturation), and dissolved in 0.01 M potassium phosphate buffer, pH 7.3, containing 2 X lop5 M pyridoxal 5’-phosphate, followed by dialysis for 18 hours against 200 volumes of the same buffer.

Step 5: DEAE-Sephadex Column Chromatography-The en- zyme solution was applied to a DEAE-Sephadex A-50 column (3 X 50 cm) equilibrated with 0.01 M potassium phosphate buffer, pH 7.3, containing 0.1 M sodium chloride and 2 X lop5 M pyridoxal 5’-phosphate. After the column was washed with the same buffer, the enzyme was eluted with the buffer supple- mented with 0.15 M sodium chloride. Fig. 1 shows a typical elution profile. The active fractions were pooled and concen- trated by ammonium sulfate precipitation (75 ye saturation). The enzyme was dissolved in 0.01 M potassium phosphate buffer, pH 7.3, containing 2 X 1O-5 M pyridoxal 5’-phosphate.

Step 6: Sephadex G-150 Column Chromatography-The enzyme solution was placed on a Sephadex G-150 column (1 x 90 cm) buffered with 0.01 M Tris-HCl buffer, pH 7.5, containing 2 X 1O-5 M pyridoxal 5’-phosphate, and eluted with the same buffer. The elution pattern is shown in Fig. 2. The active fractions were combined and concentrated by addition of ammonium sulfate (75% saturation). The enzyme was dissolved in a small volume of 0.01 M Tris-HCl buffer, pH 7.5, containing 2 X 1O-6 M pyridoxal 5’-phosphate. The insoluble materials were re- moved by centrifugation.

Step 7: Crystallization-Ammonium sulfate was added to the enzyme solution, maintaining the pH at 7.2 to 7.5 with 3 N am- monium hydroxide, until a faint turbidity was obtained. On standing overnight, the insoluble materials were discarded by centrifugation. To the clear supernatant, a minimum amount of ammonium sulfate was added. The crystals were first visible

0 0 0 20 40 60 80 100 120 140

EFFLUENT ( ML 1

FIG. 2. Chromatography of arginine racemase on Sephadex G-150. The flow rate was approximately 0.25 ml per min and 5-ml fractions were collected. 0, absorbance at 280 rnp; A, arginine racemase activity determined by Procedure A. Other conditions are given in the text.

TABLE I Summary of puri$cation of arginine racemase

1. 2.

3. 4.

ml w

Crude extract. 5,000 240,000 Ammonium sulfate frac-

%

0.2355,200100

tionation...............3,500 98,200 0.4544,200 80 LButanol treatment.. 320 8,200 4.5 36,900 67 DEAE-cellulose chroma-

tography. 75 560 46 25,800 46.8 DEAE-Sephadex chroma-

tography. 12 67.4 350 23,600 42.8 Sephadex G-150 chroma-

tography. 8 19.0 1150 21,800 39.6 Crystallization. 2 14.0 1250 17,500 31.7

after the solution was maintained at 4” for about 12 hours. Maximum yield of crystalline enzyme was obtained after 3 days. The crystals appear as needles with a yellow color. A protocol of the purification is presented in Table I.

Stability of Enzyme

The crystalline enzyme can be stored at 4” as a suspension in 0.01 M potassium phosphate buffer, pH 7.3, containing 10-d M pyridoxal 5’-phosphate and 60% saturated ammonium sulfate without loss of activity for periods of over 6 months. When the enzyme solution was heated to 50” at various pH values, the enzyme was found stable between pH 6.8 and 10.0 (Fig. 3).

Purity and Molecular Weight

The purity of the crystalline enzyme and its sedimentation coefficient were determined with a Spinco model E ultracentrifuge equipped with a phase plate as a schlieren diaphragm. The schlieren patterns (Fig. 4) obtained in a sedimentation velocity experiment indicate the presence of a single component. The sedimentation coefficient of the protein peak, calculated for water at 20” and zero protein concentration, was 5.2 S. The molecular weight of arginine racemase was determined by the sedimentation equilibrium method of Yphantis (15) as described above. Assuming a partial specific volume of 0.74, a molecular weight of 167,000 f. 5,000 was obtained.

by guest on January 23, 2020http://w

ww

.jbc.org/D

ownloaded from

Crystalline Arginine Racemase Vol. 246, No. 16

PH

FIG. 3. Effect of pH on the enzyme stability. The enzyme preparation was heated at 50” for 10 min in the following buffer (final, 0.04 M), and diluted four times with water. The enzyme activity was then assayed by Procedure A. A, acetate buffer; 0, Tris-mnleate buffer; l , Tris-HCl buffer; 0, glycine-KCl-KOH buffer.

FIG. 4 (left). Sedimentation pattern of crystalline arginine racemase. Protein concentration, 0.6$& in 0.01 M potassium phosphate buffer, pH 7.3. Pictures were taken at bar angle of 65”. A, 28 min after achieving top speed (59,780 rpm) ; B, 56 min. Temperature was 20”.

FIG. 5 (right). Disc gel electrophoresis of arginine racemase. Electrophoresis was conducted at a current of 2 ma for 2 hours in Tris-glycine buffer, pH 9.0. The direction of migration is from the cathode (lop of photo) to the anode. Ot,her conditions are given in the text.

Disc gel electrophoresis in 7.5% polyacrylamide gel was per- formed by a modification of the procedure of Davis (16). The enzyme (24 pg) was applied on the top of spacer gel in 1 M su-

crose. After the run, protein was stained with 1.0% Amido schwarz in 7% acetic acid. The crystalline enzyme migrated as a single band as shown in Fig. 5.

-06

- 0.5

s -04 5

E x

-03 $

-02

-0 I

0”’ 260 300 350 400

0 450

WAVELENGTH t m,u I

FIG. 6. Absorption spectra of arginine racemase. Curve A, a 0.22% solution of holoenzyme in 0.05 M potassium phosphate buffer, pH 7.3, was used. The same spectrum was obtained, when the buffer was replaced by Tris-maleate buffer, pH 5.0, or glycine-KCl-KOH buffer, pH 10.2. Curve B, The same concentra- tion of apoenzyme in 0.01 M potassium phosphate buffer, pH 7.3 was used. Curve C, the same concentration of enzyme reduced with sodium borohydride and dialyzed was used. Other condi- tions are given in the text;.

Spectrophotometric Behavior and Pyridoxal 5’-Phosphate Content

As shown in Fig. 6, the absorption spectrum of the enzyme in the visible region is characterized by maximum at 420 mp, indicating that the formyl group of the bound pyridoxal 5’- phosphate forms an azomethine link to an amino group of the protein, as in other pyridoxal 5’-phosphate enzymes so far in- vestigated. An absorbance ratio of the enzyme at 280 and 420 mp is 5:l. Although the absorbance at 420 ml decreased gradually during storage of the enzyme, the original absorbance ratio was obtained by addition of pyridoxal 5’-phosphate and dialysis. No appreciable spectral shift occurred on varying the pH between 5.4 and 10.0. The absorption spectrum was not affected by addition of either n-arginine or L-arginine.

The amount of pyridoxal 5’-phosphate present in the enzyme was determined in duplicate experiments with two different enzyme samples according to the procedure of Wada and Snell (17). After the dialyzed enzyme was kept at room temperature for 30 min in the presence of 0.1 N HCl to release the bound coenzyme, the amount of free pyridoxal 5’-phosphate was ana- lyzed with phenylhydrazine reagent. An average pyridoxal 5’-phosphate content of 1 male/42,000 g of protein was obtained, indicating that 4 moles of pyridoxal 5’-phosphate are bound t,o 1 mole of enzyme protein in the holoenzyme.

Resolution and Reconstitution of Arginine Racemase

Pyridoxal5’-phosphate was not required for maximum activity of the enzyme, even after the enzyme was exhaustively dialyzed. This finding suggests that the cofactor is tightly bound to the protein moiety of enzyme. Resolution of the racemase was carried out by treatment of the enzyme with phenylhydrazine as follows. The enzyme was dialyzed at 2” overnight against 0.01 M potassium phosphate buffer, pII 7.3, containing 2 rnnf phenylhydrazine hydrochloride. The enzyme solution was brought to 75% saturation with ammonium sulfate. The en- zyme collected by centrifugation was dissolved in 0.01 M potas- sium phosphate buffer, pH 8.0, and dialyzed against the same

by guest on January 23, 2020http://w

ww

.jbc.org/D

ownloaded from

Issue of August 25, 1971 T. Yorijuji, K. Ogata, and K. Soda 5089

buffer for 12 hours. The enzyme thus treated had almost no detectable activity in the absence of added pyridoxal 5’-phos- phate, and no longer exhibited an absorption maximum at 420 rnp (Curve B in Fig. 6). The activity can be restored by about 855; by addition of pyridoxal 5’-phosphate. This cannot be replaced by pyridoxamine 5’-phosphate, pyridoxal, pyridox- amine, and pyridoxine. The Michaelis constant for pyridoxal 5’-phosphate was estimated t.o be 4.0 x 1OW M. FAD, FMN, and riboflavin showed no influence on the activity.

Reduction with sodium borohydride affects both the absorp- tion spectrum and the activity of arginine racemase. The en- zyme was treated with 5 ITIM sodium borohydride at about 0’ for 5 min by the dialysis method of Matsuo and Greenberg (18), and was then dialyzed for 18 hours against two changes of 0.01 M potassium phosphate buffer, pII 7.3. Treatment with sodium borohydride resulted in a loss of the 420 rnp absorption peak coupled with an appearance of the shoulder at about 315 1nl.l (Curve C in Fig. 6). This shoulder did not change on further dialysis against 0.01 M potassium phosphate buffer, pH 7.3, for 20 hours. Reduced enzyme was catalytically inactive, and the addition of pyridoxal 5’.phosphate did not reverse the inactiva- tion. These findings suggest that borohydride reduces to a secondary amine the aldimine type Schiff base between the formyl group of pyridoxal 5’phosphate and an amino group of the enzyme protein.

Substrate Xpeci$city

The ability of this racemase to catalyze the racemization of various amino acids is presented in Table II. In addition to arginine, lysine, &r-acetyllysine, canavanine, ornithine, 2,3- diaminopropionate, homoarginine, 2,4-diaminobutyrate, ethio- nine, citrulline, homocitrulline, 6-N-acetylornithine, theanine (y-glutamylethylamide), glutamine, and methionine were race- mized by the racemase in this order when the assays were carried

TABLE II

Su,bslrate specificity of arginine racemase

The enzyme activity, unless otherwise indicated, was det,er- mined by measuring the change in optical rotation at 350 rng, according to Procedure B. The values below 1 are expressed as 0.

Substrate

I,ysine ................ ilrginine .............. c-A-Acetyllysine .......... Ornithine .............. X,3-Diaminopropionic acid. Homoarginine. ............. Canavanine. ............... 2,4-Diaminobutyric acid. .. Ethionine. .............. Citrulline. ................. Homocitrulline. ............ 6-N-Acetylornithine . ....... Theanine ...... ........... Glutamine. .............. Methionine. ............... Alanine ....................

- R a

-

110 100

86 44 40 25 19 18 13 13 12 12 11

7 40 0

:elati~ ctivit Substrate

Asparagine 00 Valine. 0” Leucine 0” Isoleucine 0" Histidine. 0" Phenylalanine ob Aspartic acid. 0 Glutamic acid. . 0 Serine.............. Oc Threonine Oc Hydroxyproline.. 0 Proline 0 NY-Nitroarginine. 0a a-N-Acetyllysine.. 0 a-N-Acetylornithine 0

dative ctivity

a The optical rotation was measured at 320 rnp. * The activit.y was determined with n-amino acid oxidase (20). = The optical rotation was measured at 280 mp.

out at pH 10.0 (glycine-KCl-KOH buffer), although the optimum pH values for the racemization of some amino acids were lower than 10.0 as mentioned below. Diaminomonocarboxylic acids such as lysine, ornithine, 2,4-diaminobutyrate, and 2, S-diamino- propionate showed high reactivity. e-N-Acetyllysine and 6-N- acetylornithine were racemized by the enzyme, while the cr-N- acetylated amino acids were inert.

The racemization of phenylalanine was assayed spectrophoto- metrically with n-amino acid oxidase and 3-methyl-2-benzo- thiazolone hydrazone hydrochloride as described previously (19, 20), because the optical rotation of pl~enylnlanine was very low under the conditions used.

Eject of pll

The enzyme when examined in the presence of Tris-male&e, Tris-HCI, glycine-NaCl-NaOII, glycine-KCl-KOH, and NaOH- Na2HP04 buffers has an optimum reactivity in the pH range of 9.0 to 10.6 for n-arginine racemization as shown in Fig. 7. Race- mizing activity was a function of pII as shown in Fig. 8 for the other three substrates, L-lysine, n-ornithine, and L-theanine.

Maximum activities for lysine and theanine were observed in the pH regions similar to that for arginine, of 7.5 to 10.6 and

7.0 to 10.0, respectively, but the enzyme is most active for orni-

G I

$ 16-

3

8 I.& l

E

: : 0.8-

5

z o’4

0

(--y-y)

7 8 9 IO II 12 PH

FIG. 7. Effect of pH on racemization of arginine. The react.ion mixture contained 10 pmoles of pyridoxal 5’-phosphate, 30 ng of enzyme, and 40 pmoles of the following buffer in a final volume of 1.0 ml. 0, Tris-maleate buffer; 0, Tris-HCl buffer; A, glycine- NaCl-NaOH buffer; A, glycine-KCl-KOH buffer; n , Na2HP04- NaOH buffer. The enzyme activities were assayed by Procedure A.

16-

8-

0’ b ” 3 ” 3 I I ” a 10 ’ I 7 8 9 IO II 7 8 9 IO II 7 8 9 IO II

PH

FIG. 8. Effect of pH on racemization of lysine, ornithine, and theanine. The reaction mixture contained 80 rmoles of n-lysine (A), L-ornithine (B) or L-theanine (C), 10 nmoles of pyridoxal 5’.phosphate, 600 ng of enzyme, and 100 &moles of the following buffer in a final volume of 2.0 ml. 0, Tris-acetate buffer; l , potassium ph0sphat.e buffer; A, Tris-HCl buffer; A, glycine-KCl- KOH buffer; 0, NaHC03-Na&03 buffer. The enzyme activities were determined by Procedure B.

by guest on January 23, 2020http://w

ww

.jbc.org/D

ownloaded from

5090 Crystalline Arginine Racemase Vol. 246, No. 16

FIG. 9. Effect of arginine on arginine racemase activity. The reaction mixture consisted of the indicated concentration of D-arginine-HCl, 40 pmoles of glycine-KCI-KOH buffer (pH 10.0)) 10 nmoles of pyridoxal 5’-phosphate, and 35 ng of enzyme in a final volume of 1.0 ml. The enzyme activity was determined by Procedure A. The velocity was expressed as amount of L-arginine formed (micromoles) for 20 min under the conditions used. The reciprocal velocity was plotted against the reciprocal concentra- tion of D-arginine.

I / I I I

3.0

z 2.0

1.0

i___ 0

., A

.A\ I

0 ‘.

0 1.0 2.0 3.0 4.c 5.0

I/S ( X I O-6 M I

FIG. 10. Effect of pyridoxal5’-phosphate on arginine racemase activity. Apoenzyme (30 ng) and 10 moles of n-arginine were used in the reaction mixture. Other conditions are shown in Fig. 9. The reciprocal velocity was plotted against the re- ciprocal concentration of pyridoxal 5’-phosphate.

thine between pH 7.5 and 8.5, and the activity decreases mark- edly above pH 9.0. The enzyme has almost the same activity for lysine and ornithine when assayed at each optimum pH.

Kinetics

Fig. 9 shows a plot of the reciprocal of reaction velocity against the reciprocal of substrate concentration. The apparent Mi- chaelis constant for n-arginine was calculated to be 1.0 x 10e3 M. The constant for the cofactor, pyridoxal 5’-phosphate, was also estimated to be 4.0 X 10-T M (Fig. 10).

Inhibitors

Nutritional studies show that Streptococcus jaecalis is capable of growing under conditions whereby vitamin Bc or D-alanine is the growth-limiting factor; in such a system vitamin Bg may be utilized solely for the formation of D-alanine, which is indis- pensable for growth of the organism, from the L isomer (22). Subsequent studies on nonenzymatic racemization of amino acids in the presence of pyridoxal suggested that during the enzymatic racemization the amino acids may react with the pyridoxal derivatives to form a Schiff base. A reaction mech- anism has been proposed on the basis of these studies that is similar to that put forth for transamination, i.e. the reversible formation of the ketimine intermediate, but without the hy- drolysis leading to transamination (23).

The effects of various inhibitors upon the rate of racemization However, there is diversity in the cofactor requirement of the of arginine are shown in Table III. The first group of inhibitors several amino acid racemases studied thus far, although the is amino acids and amines. The nonsubstrate amino acids and possibility that such diversity may be at least partially ascribed amines tested produce no inhibition, although L-ornithine and to the impurities involved in the enzyme preparations cannot be L-homoarginine, which are also racemized by the enzyme, have, excluded. Glutamate racemase purified approximately 500-fold as expected, an inhibitory effect on the racemization of arginine. from Lactobacillus jermenti exhibits absorption maxima at 273, Arginine racemase is inhibited by hydroxylamine and D- and 360, and 450 mp, and is not activated by pyridoxal5’-phosphate, L-penicillamine which are typical inhibitors for pyridoxal 5’- but by FAD, which was shown to be present in the enzyme by phosphate enzymes, while n-cycloserine and isonicotinic acid means of paper chromatography, paper electrophoresis, and its hydrazide are not inhibitory. The enzyme activity is partially activity for apo+amino acid oxidase (24, 25). Recently, Diven decreased by addition of p-chloromercuribenzoate (1 mM). None (26) reported that glutamate racemase purified from the same

TABLE III Effect of inhibitors on arginine racemase activity

The enzyme was first incubated with inhibitors at 37” for 10 min. Racemization was then initiated by addition of L-arginine and determined by Procedure A. The inhibitors tested showed no effect on the assay with arginase.

Compound Tinal concentratior Relative activity

None.......................... L-Aspartic acid. . . L-Glutamic acid. . . . . . . . L-Valine....................... L-Homoarginine L-Ornithine. Agmatine Putrescine..................... Isonicotinic acid hydrazine D-Cycloserine.. Hydroxylamine. L-Penicillamine . D-Penicillamine. p-Chloromercuribenzoate. FAD FMN . Riboflavin.

I

?ndd

1.0 1.0 1.0 5.0 3.0 1.0 1.0 1.0 1.0 1.0 5.0 5.0 1.0 0.1 0.1 0.1

100 100 100 100 89 67

100 104 99

101 49 72 75 86 96 96 96

i

of the flavin compounds shows any significant effect on the en- zyme activity.

The degree of inactivation of the enzyme by L-ornithine in- creased with prolonged preliminary incubation time. The pres- ence of pyruvate during the preliminary incubation with L-

ornithine resulted in protection against the inactivation of enzyme. The mechanism of inactivation by ornithine is de- scribed in the following paper (21).

DISCUSSION

by guest on January 23, 2020http://w

ww

.jbc.org/D

ownloaded from

Issue of August 25, 1971 T. Yorifuji, K. Ogata, and K. Soda 5091

organism was inhibited by the presence of FMN, FAD, ribo- flavin, and some analogues of FAD, although the prosthetic group of the enzyme has not been established. Hydroxyproline 2-epimerase has been purified to homogeneity or near homo- geneity from P. striata (4). No cofactors of the epimerase have been detected on the basis of the absorption spectrum of the enzyme, the phosphorus content, fluorescence assays for pyridine or flavin nucleotides, and inhibitor action, although the activity of enzyme inhibited by flavin analogues is in part restored by addition of FAD and FMN. In proline racemase purified to near homogeneity from C. sticklandii, no evidence of cofactor requirement could be obtained, and the structures of inhibitors and substrate suggest the possible involvement of a metal ion at the active center (7). Diven, Scholz, and Johnston (27) pro- vided evidence that both FAD and pyridoxal 5’-phosphate function as the prosthetic groups in the purified alanine racemase from Bacillus subtilis, and proposed two possible mechanisms involving both cofactors. On the other hand, the highly purified alanine racemase obtained from P. putida (6) and L. fermenti (28) was shown not to contain any flavin coenzyme, but pyridoxal 5’phosphate. The amino acid racemase with low substrate specificity, purified and crystallized from the cell-free extract of P. striata, also requires pyridoxal 5’-phosphate as a cofactor (8).

The studies described here deal with the purification, the crystallization, and the characterization of arginine racemase. The racemase has been purified approximately 5400.fold from a sonic extract of P. graveolens which had been grown in a medium containing n-arginine to induce the enzyme; the enzyme was crystallized by addition of ammonium sulfate. The crystalline enzyme is homogeneous by the criteria of ultracentrifugation and disc gel electrophoresis. On the basis of the characteristic spectrum of the enzyme having an absorption maximum at 420 mp, and in view of the borohydride reduction of the enzyme, it is suggested that pyridoxal 5’-phosphate is firmly bound to an amino group of protein (probably an c-amino group of lysine residue by analogy with other pyridoxal 5’-phosphate enzymes) through a Schiff base (29). The determination of pyridoxal 5’-phosphate shows that 4 moles of the cofactor are bound per mole of enzyme. The apoenzyme is obtained by dialysis against phenylhydrazine solution. The resolution of the cofactor from the enzyme is reversible, as shown by the spectral and activity studies. FAD, FMN, and riboflavin have no influence on the activity of enzyme. These findings rule out the possibility that a flavin compound functions as a prosthetic group in arginine racemase.

The amino acid racemases so far well characterized, e.g. proline racemase (7) and hydroxyproline 2-epimerase (4), have a high substrate specificity with the exception of the amino acid race- mase with low substrate specificity from P. striata (8). It was reported previously in a preliminary note that arginine was the exclusive substrate for arginine racemase, and that other amino acids tested are inactive as a substrate under the conditions used, when the enzyme was assayed manometrically with D-

amino acid oxidase (1). The substrate specificity of this race- mase, however, was re-examined in detail here by a polarimetric procedure, using a large amount of enzyme. This sensitive procedure can determine the racemization of amino acids which are insusceptible to D- or n-amino acid oxidase as well. In addition to arginine, various amino acids, most of which are very poor substrates for amino acid oxidases, were shown to be racemized by arginine racemase. The sensitive polarimetric

assay, for example, has revealed the racemization of methionine, which could not be detected by the manometric method (1). It is interesting that all good substrates are diaminomonocar- boxylic acids, e.g. lysine, ornithine, and arginine. Although no activity was observed with cY-N-acetylated derivatives of lysine and ornithine, e-N-acetyllysine and 6-N-acetylornithine were- racemized by this enzyme. This finding suggests that a-amino group of the substrate is required to be free for this enzymatic racemization. y-Glutamyl amide compounds (theanine and glutamine) and a-amino-y-alkylthiobutyric acids (ethionine and methionine) can be racemized, although slowly. The y-glutamyl carboxyl group must be blocked for racemization to occur, be- cause glutamate is completely inert, while glutamine and the- anine, which is an important constituent in Japanese tea leaves (30), and in the mushroom Xerocomus badius (31), can be a substrate. It seems of interest that all of the substrates have a group containing either nitrogen or sulfur in addition to the a-amino group, e.g. amino group in lysine, guanidino group in arginine, and methylthio group in methionine, although the relationship between the structure of substrate and the enzyme activity has not been well elucidated.

In view of the substrate specificity this enzyme might be more properly designated “lysine racemase.” However, “arginine racemase” is still used here since the enzyme has been so designated since its discovery, and in order to avoid possible confusion between this enzyme and lysine racemase. Lysine racemase has so far not been fully purified and the substrate specificity has not been made clear (32, 33).

Acknowledgments-The authors thank Dr. T. Yamamoto- Dr. T. Tochikura, and Dr. H. Yamada for their helpful dis- cussions. Thanks are also due Dr. II. Utiyama for ultracen. trifugation studies.

1.

2.

3. 4. 5.

6.

7.

8.

9.

10.

11.

12.

13. 14.

15. 16. 17. 18.

19.

REFERENCES

YORIFUJI, T., OG~T~, K., AND SODA, K., Biochem. Biophys RES. Commun.. 34. 760 (1969).

MEISTER, A., Bibch~mistr~/ of tie amino acids, Academic Press. New York, 19G5, p. 1131 .

CORRIG~N. J. J.. Science. 164. 142 (1969). ADAMS, E.j AND NORTON,‘~. L:, J. B‘iol. f?hem., 239,1525 (1964). FREE, C. A., JULIUS, M., AND BARRY, G. T., Ped. Proc., 27,

785 (1968). Rosso, G., TAKASHIMA, K., AND ADAMS, E., Biochem. Biophys.

Res. Commun., 34, 134 (1969). CARDINALE, G. J., AND ABELES, It. I-I., Biochemistry, 7, 3970

(1968). SODA, K., AND OSUMI, T., Biochem. Biophys. Res. Commun.,

36, 363 (1969). SODA, K., YORIFUJI, T., AND Ooxrl~, K., Biochim. Biophys.

Acta, 146, 606 (1967). ODO, K., AND ICHIICAWA, E., Nippon Kagaku Zasshi, 76, 228

(1955). PETERSON, E. A., AND SOBER, H. A., J. Amer. Chem. Sot., 76,

169 (1954). HUNTER, A., AND DAUPHINEE, J. A., J. Biol. Chem., 86, 627

(1930). ARCHIBALD, R. M., J. Biol. Chem., 156, 121 (1944). LOWRY, 0. H., ROSEBROUGH, N. J., F~RR, A. L., AND RANDALL,

R. J., J. Biol. Chem., 193, 265 (1951). YPHANTIS, D. A., Ann. N. Y. Acad. Sci., 88, 586 (1960). DAVIS, B. J., Ann. N. Y. Acad. Sci., 121, 404 (1964). WADA, H., AND SNF:LL, E. E., J. Biol. Chem., 236,2089 (1961). MATSUO, Y., AND GREENBERG, D. M., J. Biol. Chem., 234, 507

(1959). SOD.&, K., Anal. Biochem., 26, 228 (1968).

by guest on January 23, 2020http://w

ww

.jbc.org/D

ownloaded from

5092 Crystalline Arginine Ra.cemase Vol. 246, No. 16

20. SODA, K., OSUMI, T., YORIFUJI, T., AND OGATA, K., Agr. Biol. Chem. (Tokyo), 33,424 (1969).

21. YORIFUJI, T., MISONO, H., AND SODA, K., J. Biol. Chem., 246,5093 (1971).

22. OLIVARD, J., AND SNELL, E. E., J. Biol. Chem., 213,203 (1955). 23. METZLER, D. IL, IKAWA, M., AND SNELL, I$ E., J. Amer. Chem.

Sot., 76, 648 (1954). 24. TANAKA, M., KATO, Y., AND KINOSHITA, S., Nippon Nogeika-

gaku K&hi, 36, 1378 (1961). 25. TANAKA, M., KATO, Y., AND KINOSHITA, S., Nippon Nogeika-

gaku K&hi, 36, 1381 (1961). 26. DIVEN, W. F., Biochim. Biophys. Acta, 191, 702 (1969).

27. DIVEN, W. F., SCHOLZ, J. J., AND JOHNSTON, R. B., Biochim. Biophys. Acta, 86, 322 (1964).

28. JOHNSTON, M. M., AND DIVEN, W. F., J. Biol. Chem., 244,5414 (1969).

29. FASELLA, l’., Annu. Rev. Biochem., 36, 185 (1967). 30. SAKATO, .Y.,.Nippon Nogeikagaku Kaishi, 23,262 (1950). 31. CASIMIR. J.. JADOT. J.. AND RENARD. M.. Biochim. Biophws.

Ada, 39, 4G2 (196b). ’ I _ -

32. HUANG, H. T., KITA, D. A., AND DAVISSON, J. W., J. Amer. Chem. Sot., 80, 1006 (1958).

33. ICHIWARA, A., FURIYA, S., AND SUDA, M., J. Biochem. (Tokyo), 48, 277 (1960).

by guest on January 23, 2020http://w

ww

.jbc.org/D

ownloaded from

Takamitsu Yorifuji, Koichi Ogata and Kenji SodaCRYSTALLIZATION, AND PROPERTIES

: I. PURIFICATION,Pseudomonas graveolensArginine Racemase of

1971, 246:5085-5092.J. Biol. Chem. 

  http://www.jbc.org/content/246/16/5085Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/246/16/5085.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on January 23, 2020http://w

ww

.jbc.org/D

ownloaded from