Studies on the Metabolism of the Benzene Ring of …zymes. The liver was removed immediately after death, packed in ice, trimmed of connective tissues and gall bladder, washed with

THE JOURNAL OF BIOLOGICAL CHE~STRY Vol. 240, No. 2, February 1965

Printed in U.S.A.

Studies on the Metabolism of the Benzene Ring of

Tryptophan in Mammalian Tissues

II. ENZYMIC FORMATION OF a-AMINOMUCONIC ACID FROM 3-HYDROXYANTHRANILIC ACID*

ARATA ICHIYAMA,~ SHIGENOBU NAKAMURA, HITOSHI KAWAI, TASUKU HONJO, YASUTOMI NISHIZUKA, OSAMU HAYAISHI, AND SIRO SESOH$

From the Department of Medical Chemistry, Kyoto University Faculty of Medicine, Kyoto, Japan

(Received for publication, June 30, 1964)

In preceding reports from this laboratory, a-hydroxyanthranilic acid was shown to be converted to glutaric acid and COz by crude extracts of cat liver (l-3). Evidence was also presented that ol-amino-p-carboxymuconic e-semialdehyde, the primary oxidation product of 3-hydroxyanthranilic acid by the oxygenase (4), is an obligatory intermediate in this conversion (2,3). How- ever, all previous experiments in vitro have indicated that quinolinic acid (5) and picolinic acid (6) are the stable products thus far obtained from oc-amino-P-carboxymuconic e-semialdehyde in mammalian liver. The enzyme responsible for the formation of picolinic acid was first reported by Mehler (6) and was referred to as picolinic carboxylase (7). However, since picolinic acid was not degraded to COZ even in V&JO (7, 8), the physiological role of the enzyme as well as the metabolic pathway of a-amino- @-carboxymuconic e-semialdehyde to glutaric acid have re- mained obscure.

In this communication, we wish to describe the conversion of Lu-amino-fi-carboxymuconic e-semialdehyde to cr-aminomuconic acid catalyzed by picolinic carboxylase and a nicotinamide adenine dinucleotide-linked aldehyde dehydrogenase (Scheme 1). Since the primary product of the picolinic carboxylase- catalyzed reaction, possibly oc-aminomuconic e-semialdehyde, is extremely unstable and cyclizes rapidly to picolinic acid, oc-hydroxymuconic e-semialdehyde was employed as substrate for the purification of the dehydrogenase. The enzyme is present in the liver and kidney of several mammals and will be referred here- after to as oc-hydroxymuconic e-semialdehyde dehydrogenase.

EXPERIMENTAL PROCEDURE

Chemicalsol-Hydroxymuconic e-semialdehyde was prepared enzymically from catechol by the use of metapyrocatechase as described by Kojima, Itada, and Hayaishi (9). The product was employed after repeated crystallization from ethyl acetate and

* This investigation has been supported in part by research grants from the National Institutes of Healt,h (CA 04222), The Rockefeller Foundation, the Jane Coffin Childs Memorial Fund for Medical Research, the Squibb Institute for Medical Research, and the Scientific Research Fund of the Ministry of Education of Japan.

t Recipient of a Sigma Chemical Company Postgraduate Fel- lowship.

$ Permanent address, The Institute of Food Chemistry, Doji- manaka, Kitaku, Osaka, Japan.

ethanol (decomposed at 215O). ol-Hydroxymuconic acid was prepared chemically by alkaline hydrolysis of diethyl y-oxalocrotonate synthesized according to the method of Lapworth (10). The product was purified by recrystallization from diox- ane-benzene and ethanol-benzene (decomposed at 190”). L-3- Hydroxykynurenine, the benzene ring of which was uniformly labeled with i4c, was prepared enzymically from nn-tryptophan as described previously (3). We are indebted to Dr. L. M. Henderson for generous samples of carboxyl-lGlabeled and uniformly 3H-labeled 3-hydroxyanthranilic acid. nn-Trypto- phan, uniformly labeled in the benzene ring with i4C, and catechol, uniformly labeled with 3H, were obtained from the Radio- chemical Centre, Amersham, Buckinghamshire, England. 3-Hydroxyanthranilic acid, catechol, ar-aminoadipic acid, and other chemicals were obtained from commercial sources. Pro- tamine sulfate was a product of Sigma Chemical Company. DEAE-cellulose was a product of Serva Entwicklungslabor, Heidelberg, and was treated according to the method of Peterson and Sober (11). Calcium phosphate gel was prepared according to the method of Colowick (12) and hydroxylapatite by the method of Tiselius, HjertCn, and Levin (13).

Biological Materials-Cat liver was used as the source of en- zymes. The liver was removed immediately after death, packed in ice, trimmed of connective tissues and gall bladder, washed with cold 0.85% KCl, and then used immediately or stored at -20”.

3-Hydroxyanthranilic acid oxygenase was partially purified from the aqueous extracts of rat liver acetone powder. All subsequent manipulations were conducted at 4”. The powder, 2 g, was extracted with 20 ml of a lop5 M solution of FeS04 (4, 14, 15). To the crude extract (18 ml), 6 ml of a 0.4y0 solution of protamine sulfate, pH 7, were added with continuous stirring, and inactive precipitates were removed by centrifugation. To the supernatant solution (23 ml), 19 ml of a saturated solution of ammonium sulfate, pH 7, were added with stirring. After 5 minutes the resulting precipitate was removed by centrifugation. To the supernatant solution, an additional 15.5-ml por- tion of a saturated ammonium sulfate solution was added. After stirring for 5 minutes, the precipitate was collected by centrifugation and dissolved in 11 ml of a 1O-5 M solution of FeS04. To the ammonium sulfate fraction (11.5 ml), 23 ml of calcium phosphate gel suspension (30 mg, dry weight, per ml) were added with stirring. After 10 minutes the gel was collected

740

by guest on April 13, 2020

http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

February 1965 Ichiyama et al. 741

- nonenzymic

--------------+ N

OH a-Amino-@-carboxy- Quinolinic 3-Hydroxy- muconic d semialdehyde acid

anthranilic acid

Picolinic carboxylase

COOH

a-Aminomuconic e-semialdehyde

Picolinic acid

NAD

.:

a-Hydroxymuconic e- semialdehyde

NADH dehydrogenase

COOH NH,

a-Aminomuconic acid SCHEME 1. Formation of quinolinic, picolinic, and or-aminomuconic acids from 3-hydroxyanthranilic acid

by centrifugation and washed once with 20 ml of distilled water. The enzyme was then eluted twice from the gel, each time with 10 ml of 0.05 M K&CPO,. The combined eluates (20 ml) were made 1O-5 M with respect to FeS04. By the above procedures the enzyme was purified about 5-fold with an over-all yield of about 15%, and was essentially free from kynureninase, picolinic carboxylase, quinolinate transphosphoribosylase (IS), a-hydroxymuconic e-semialdehyde dehydrogenase, oc-aminomuconic acid reductase,’ and a-ketoglutarate dehydrogenase. The purified enzyme was unstable and prepared shortly before use.

Kynureninase was purified from nn-tryptophan-adapted cells of PseudomonasJluorescens ATCC 11250 according to the method of Hayaishi and Stanier (17), except that the dialyzed second ammonium sulfate fraction was further purified by DEAE- cellulose column chromatography. By this procedure the specific activity was increased about 2-fold and the enzyme could be separated from quinolinate transphosphoribosylase. Crystalline metapyrocatechase (18, 19) was kindly supplied by Dr. M. Nozaki. Crystalline heart muscle lactic dehydrogenase was obtained from Sigma Chemical Company.

Enzyme Assay-The activity of picolinic carboxylase was conveniently assayed by measurement of the decrease of absorbance of cr-amino-P-carboxymuconic e-semialdehyde at 360 rnp as described by Mehler (6). However, as discussed below, this assay method may not be exactly consistent with the activity of picolinic carboxylase. Decarboxylation alone does not seem

1 The enzyme catalyzes the formation of or-ketoadipic acid from a-aminomuconic acid in the presence of either NADPH or NADH. a-ketoadipic acid is then oxidatively decarboxylated to glutaryl- CoA by the oc-ketoglutarate dehydrogenase complex. The details concerning the metabolism of a-aminomuconic acid will be pub- lished elsewhere.

to affect the absorbance significant.ly, and the decrease in absorbance is probably due to the nonenzymic cyclization of the primary reaction product, a-aminomuconic t-semialdehyde, to picolinic acid. Thus a parallel assay was performed by measurement of the radioactive COZ evolved from Lu-amino$-carboxymuconic e-semialdehyde which was generated in situ from carboxyl-14C-labeled 3-hydroxyanthranilic acid by the oxygenase.

For the optical assay, the reaction mixture initially contained 60 m,umoles of 3-hydroxyanthranilic acid, 200 pmoles of Tris- acetate buffer, pH 8.0, and an excess quantity of a purified preparation of 3-hydroxyanthranilic acid oxygenase (approximately 0.4 mg of protein) in a total volume of 2.9 ml. As shown in Fig. 1, the reaction was followed spect.rophotometrically at 360 rnp at 24”. After t,he formation of ol-amino-/%carboxymuconic c-semialdehyde was complete as judged by its absorption at 360 rnp, 0.1 ml of picolinic carboxylase was added. The decrease in absorbance at 360 rnp was followed at 30-second intervals against a blank incubation that contained all ingredients except the substrate. These data were corrected for the spontaneous decrease of absorbance due to the formation of quinolinic acid.

For the isotopic assay, the incubation was carried out as above except t.hat all ingredients were placed in a Thunberg tube fitted to a quartz cell with a light path of 1 cm, and carboxyl-lGlabeled 3-hydroxyanthranilic acid was employed as substrate. The side arm of the tube contained 0.2 ml of 0.5 N NaOH. The reaction was stopped by the addition of 0.5 ml of 2 N acetic acid. This brought the reaction mixture to pH 4 and facilitated the libera- tion of radioactive Cot. Under these conditions, oL-amino-p- carboxymuconic c-semialdehyde did not decompose nonenzymi- tally to a-hydroxymuconic c-semialdehyde with the elimination of the /%carboxyl and amino groups, as described below. After


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

742 Tryptophan Metabolism. II Vol. 240, ivo. 2

TABLE I Optical properties of various compounds

Compound and solvents

3-Hydroxykynurenine HzO. 0.02 M phosphate, pH 7.3. 0.1 N HCl. _.

3-Hydroxyanthranilic acid 0.02 M phosphate, pH 7.3.. 0.1 N HCl. t. 0.05 N NaOH.

cr.Amino-P-carboxymuconic E-semialdehyde

pH 7.5 to 8.5. ._ _. a-Aminomuconic acid

pH 7.5 to 8.5. a-Hydroxymuconic e-semi-

aldehyde Phosphate, pH 7.5. 0.1 N HCl

a-Hydroxymuconic acid 0.1 N HCl. 0.1 N NaOH, _.

Quinolonic acid Phosphate, pH 6.8.

Picolinic acid Phosphate, pH 6.8.

hmsx Cd Refer- ence

268.5 (7,800), 370 (4,000) 268.5 (9,840), 370 (4,240) 251.5 (8,750), 315 (2,650)

20

314 (3,240) 297 (3) 120) 230 (18,600), 331 (4,270)

20 21

360 (45,000)*

325 (16,500)t

375 (33,000) 315 (14,000)

9

304 (18,000) 350 (23,700)

268 (3,400) 6

264 (3,500) 6

* This value was calculated by repeated careful comparison Chromatographic ProceduresPaper chromatography and

with the absorbance of 3-hydroxyanthranilic acid and that of partition chromatography of organic acids were carried out as

a-hydroxymuconic l -semialdehyde (see the text). described previously (3). t This approximate value was obtained by comparison with the

absorbance of a-amino-fi-carboxymuconic c-semialdehyde and that of a-hydroxymuconic acid (see the text).

RESULTS

standing for 2 hours at 30”, a O.l-ml aliquot of the NaOH was placed on a steel disk and dried, and the radioactivity was determined as an infinitely thin sample.

Purification and Properties of Picolinic Carboxylase

One unit of picolinic carboxylase activity was defined as the amount which catalyzed the release of 1 Mmole of i4CO2 or caused a decrease in absorbance at 360 nm of 15 per minute. Since the molar extinction coefficient of a-amino-P-carboxymuconic 6.semialdehyde is approximately 4.5 x lo4 under these conditions (Table I), 1 unit of the enzyme activity corresponds to the disappearance of 1 Mmole of the substrate per minute. Specific activity was defined as units per mg of protein.

Lu-Hydroxymuconic t-semialdehyde dehydrogenase was rou- tinely assayed by measurement of the decrease in absorbance of a-hydroxymuconic e-semialdehyde at 375 rnM in the presence of NAD. The standard assay system contained 90 mpmoles of a-hydroxymuconic e-semialdehyde, 0.5 pmole of NAD, 200 hmoles of Tris-acetate buffer, pH 8.0, and the enzyme in a final volume of 3.0 ml. The decrease in absorption at 375 nm was followed spectrophotometrically at 24” at 30-second intervals.

Comparison of Optical and Isotopic Assays-In order to eluci- date whether the optical assay represents a reaction catalyzed by picolinic carboxylase, a comparison was made between the optical and isotopic assays. As shown in Fig. 2, both the decrease in absorption at 360 nm and the evolution of radioactive COz were linear with time until about half of the substrate was removed. The two methods were in parallel with each other under any conditions tested, except that the disappearance of absorbance showed a small time lag. The rate of the reaction was directly proportional to the quantities of enzyme employed. These results indicate that the primary decarboxylation reaction is the rate-limiting step and that the expected intermediate cyclizes very rapidly to picolinic acid. The optical assay may therefore be safely used as a routine assay for picolinic carboxylase activity.

One unit of the dehydrogenase activity was defined as the amount which caused an optical density decrease of 11 per minute under the standard assay conditions. Since the molar extinction coefficient of the substrate is approximately 3.3 x lo4 under these conditions (9), 1 unit of the enzyme activity corresponds to the oxidation of 1 pmole of a-hydroxymuconic c-semialdehyde per minute. Specific activity was defined as units per mg of protein.

Purijication of Picolinic Carboxylase-Picolinic carboxylase has been demonstrated in the liver and kidney of many species of animals (6). As reported by Suhadolnik et al. (26), the enzyme activity in cat liver was shown to be 30 to 50 times greater than that in rat liver. Thus the enzyme was purified from cat liver. All manipulations were carried out, in the cold (04”) unless otherwise noted.

Frozen cat liver, 45 g, was thawed in 180 ml of 0.14 M KC1 and homogenized in a Waring Blendor for 3 minutes. The homogenate was centrifuged for 15 minutes at 20,000 x g.

Step 1: To the supernatant solution (178 ml), 36 ml of a 0.4%

Determinations-3-Hydroxyanthranilic acid was determined by its absorption at 297 rnp in 0.1 N HCl or by the absorption at 360 rnp after conversion to a-amino-P-carboxymuconic .+sem- aldehyde by the oxygenase. n-3-Hydroxykynurenine, oc-amino- fi-carboxymuconic e-semialdehyde, cr-hydroxymuconic t-semialdehyde, oc-hydroxymuconic acid, quinolinic acid, and picolinic acid were determined spectrophotometrically. Maximal ab- sorbances and molar extinction coefficients of these compounds are listed in Table I.

Radioactive CO, evolved was determined as previously described (22). Protein was determined by the method of Lowry et al. (23), with bovine serum albumin as a standard. Absorption spectra were measured with a Cary recording spectrophotometer, model 15. All absorption measurements were carried out with a Shimadzu spectrophotometer. Radioactivities of 14C- and 3H- labeled samples were determined by the use of a Nuclear-Chicago gas flow counter with or without window and with a Packard Tri-Carb liquid scintillation spectrometer, as previously described (3). Radioactivity on paper was determined by direct paper strip counting with the scintillation spectrometer (24) or after elution on steel disks with the gas flow counter.

Paper Electrophoresis-High voltage paper electrophoresis was carried out on Whatman No. 1 filter paper (15 x 55 cm) with pyridine-acetic acid-water (5.0 : 3.4 : 90.0), pH 5.0, as previously described (3). Pyridinecarboxylic acids were detected by spraying with alcoholic N-(1-naphthyl)ethylenediamine dihydro- chloride, followed by exposure to cyanogen bromide (25). Amino acids were detected by the color reaction with ninhydrin.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


solution of protamine sulfate were added slowly. The mixture was stirred for an additional 5 minutes and the precipitate was removed by centrifugation.

Step 2: The supernatant solution (202 ml) was warmed rapidly to 58” with continuous stirring and maintained for 1 minute at this temperature. The enzyme solution was then cooled rapidly to 1-2” in an ice-salt bath, and the denatured protein was removed by centrifugation.

Step 3: To the enzyme solution (192 ml), 72 ml of peroxide-free cold acetone (-30”) were added slowly with mechanical stirring (30% final concentration of acetone). The mixture was maintained at - 2 to - 10” in a Dry Ice-methanol bath. The resulting precipitate was removed by centrifugation at -10”. To the supernatant solution, additional cold acetone was added in the same manner to give a 50% final concentration of acetone. After 3 minutes, the precipitate was collected by centrifugation at - 10” and dissolved in 30 ml of 0.0075 M potassium phosphate buffer, pH 7.5. Insoluble material was removed by centrifugation.

Step 4: The enzyme solution (36 ml) was adsorbed on a DEAE- cellulose column (diameter, 2.5 cm; length, 5 cm) which had been equilibrated with 0.0075 M potassium phosphate buffer, pH 7.5. After the column was washed with 100 ml of the same buffer, the enzyme was eluted with 180 ml of 0.025 M potassium phosphate buffer, pH 7.5.

Step 5: To the eluate (180 ml), 220 ml of a saturated ammonium sulfate solution, pH 7, were added with stirring. After 15 minutes the precipitate was removed by centrifugation. To the supernatant solution, additional 200 ml of the ammonium sulfate solution were added. After 15 minutes the precipitate was collected by centrifugation and dissolved in 13 ml of 0.0075 ti potassium phosphate buffer, pH 7.5. The solution was dialyzed overnight against 2 liters of the same buffer.

Step 6: The enzyme solution (14 ml) was adsorbed on a hy- droxy1apatit.e column (diameter, 1.6 cm; length, 2.5 cm) equilibrated with 0.0075 M potassium phosphate buffer, pH 7.5. After the column was washed with 20 ml of 0.1 M potassium phosphate buffer, pH 7.5, the enzyme was eluted with 20 ml of 0.2 M potassium phosphate buffer, pH 7.5.

By the above procedures the enzyme was purified more than loo-fold with an over-all yield o 10 to 20% (Table II). The enzyme preparation thus obtained was free from kynureninase, 3-hydroxyanthranilic acid oxygenase, quinolinate transphosphoribosylase, oc-hydroxymuconic t-semialdehyde dehydrogenase, a-aminomuconic acid reductase, and a-ketoglutarate dehydrogenase. The purified enzyme was stable and could be stored for several months at -20” without significant loss of activity.

Nonenxymic Spectral Changes of cr-Amino-fi-carboxymwonic e-Semialdehyde under Acidic Conditions-The absorption spectrum of oc-amino-P-carboxymuconic t-semialdehyde, with maximum absorption at 360 rnp, did not change within the pH range from 7.5 to 13.0. The molar extinction coefficient was calculated to be 45,000 by repeated careful comparison with the absorption of 3-hydroxyanthranilic acid and that of ar-hydroxymuconic c-semialdehyde (see below). At pH 4.5, the absorption maximum shifted to 345 rnp. The process was reversible, and the transient absorption spectrum was observed between pH 5.5 and

4.9 (Fig. 3A). In confirmation of the previous observations (9, 15), the absorption maximum was shown to shift irreversibly

to 315 rnp below pH 3. It takes about 25 minutes for this reac-

tion as shown in Fig. 3B. In agreement with the previous re-

ports by Mehler (27) and by Dagley, Evans, and Ribbons (28), the compound giving the absorption maximum at 315 rnp below

G > 0.6 I- 5 z w 0.4 0

2 0 0.2

is

0 5 IO 15 0 2 4 6 8 1012

MINUTES MINUTES

FIG. 1 (left). Optical assay for picolinic carboxylase. Incuba- tion was carried out under the standard assay conditions. The reaction was started by the addition of 3-hydroxyanthranilic acid at the Jirst arrow. Picolinic carboxylase (Step 5, Table II) was added at the second arrow as indicated. Blanks contained all ingredients except the substrate.

FIG. 2 (right). Release of WOZ and change of absorption at 360 rnp as a function of time. Incubation was carried out under the standard assay conditions, except that 90 mpmoles of carboxyl-%-labeled 3-hydroxyanthranilic acid (90,000 c.p.m. per @mole) were used as substrate. The reaction was started by the addition of picolinic carboxylase (Step 5, Table II, 0.15 mg). a- - -A, radioactivity released as ‘%Oz; O-O, decrease in optical density at 360 mp.

TABLE II

Summary of purijkation of picolinic carboxylase from cat liver

Crude extract, 1. Protamine sulfate 2. Heat treatment, 3. Acetone.. 4. DEAE-cellulose 5. Ammonium sulfate 6. Hydroxylapatite.

olume f ktivity Protein Specific activity Yield

ml units mg unit/mg %

178 25.00 6586.0 0.0038 100.0 202 24.45 5496.0 0.0045 97.8 192 20.16 2976.0 0.0068 80.6

36 15.00 1494.0 0.0100 60.0 180 12.53 209.0 0.0600 50.1

14 6.50 52.5 0.1238 26.0 22 4.91 6.5 0.7554 15.6

300 350 400 300 350 400

WAVE LENGTH (mu)

FIG. 3. Effect of pH on the absorption spectrum of a-amino- p-carboxymuconic E-semialdehyde. A, pH was adjusted by the addition of 0.5 N HCl as indicated; B, recordings were taken at 1 minute (Curve l), 10 minutes (Curve 2), and 25 minutes (Curve 3) after the solution was made 0.1 N with respect to HCl. After 30 minutes the solution was neutralized to pH 8.0 with 1 N NaOH. Blanks contained all ingredients except the substrate.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

744 Tryptophan Metabolism. II Vol. 240, No. 2

Compounds

None.... .._............ MgCl2 NaF .._............ KCN

p-Chloromercuribenzoate. p-Chloromercuribenzoate. Iodoacetate...................... n-Cysteine....................... GSH 10-4 M p-chloromercuribenzoate +

lop3 M L-cysteine. lo-“ M p-chloromercuribenzoate +

10-3~ GSH....................

a-Hydroxymuconic c-semialdehyde...........................

a-Hydroxymuconic e-semialdehyde...........................

Concentration P * 1 rctivity’ ielative xtivity

%

4.0 100 5.3 132 4.0 100 1.6 40

M

2 x 10-4 1 x 10-S 1 x 10-s

1 x 10-d 1 x 10-S 5 x 10-Z 1 x 10-a 1 x 10-a

1 x 10-G

3.3 x lo-6

TABLE III

E$ect of various compounds on activity of picolinic carboxylase

The activity was measured spectrophotometrically under the standard assay conditions except that 200 pmoles of potassium phosphate buffer, pH 7.0, were used and the inhibitors were added as indicated. Reaction was started by the addition of picolinic carboxylase (Step 6, Table II), and the decrease in density at 360 rnp between the 90. and 150~second readings was taken as enzyme activity.

0 0 2.6 64 3.0 75 5.6 140 4.0 100

4.5 112

3.2 80

0 0

0.6 16

* The values are expressed as milliunits per 0.1 ml of picolinic carboxylase.

pH 3 was identified as oc-hydroxymuconic e-semialdehyde by comparison with an authentic sample prepared from catechol by the action of metapyrocatechase. It was converted enzymically to cu-hydroxymuconic acid in the presence of a-hydroxymuconic e-semialdehyde dehydrogenase and NAD, and the dicarboxylic acid thus obtained was further identified by spectrophotometric criteria, paper chromatography. and electrophoresis as described below.

Spectral Change of cr-Amino-@-carboxymuconi d’emialdehyde by Picolinic Carboxylase-As reported by Mehler (6), when the purified picolinic carboxylase was added to the reaction mixture containing c-r-amino-&carboxymuconic e-semialdehyde, the absorption maximum at 360 rnH decreased rapidly with concomitant formation of picolinic acid (Fig. 4). A small shift of the absorption maximum to 364 rnp was observed during the reaction. This suggests the transient formation of an intermediate, a-aminomuconic e-semialdehyde. However, all attempts to isolate this intermediate have been unsuccessful and picolinic acid was shown to be the only product. Picolinic acid was isolated by Dowex I-formate column chromatography and identified by paper chromatography and paper electrophoresis as previously described (29). The reaction proceeded stoichiomet- rically and was essentially irreversible.

E$ects of pH and Substrate Concentration-As reported by Mehler (6), picolinic carboxylase exhibited a broad range of maximal activity and stability between pH 6 and 9.5.

The K, value for cr-amino-p-carboxymuconic e-semialdehyde,

as calculated from t.he Lineweaver-Burk plot (30), was of the order of 1O-6 M.

Inhibitor and Activator StudiesThe purified enzyme preparation (Step 6, Table II) was dialyzed for 12 hours against a large volume of 0.005 M potassium phosphate buffer, pH 7.5, and tested for activators and inhibitors. The addition of boiled extracts (5%) or yeast c0ncentrat.e (0.5%) showed little effect on the activity of picolinic carboxylase. As shown in Table III, MgClz (2 X 10m4 M) caused a slight acceleration of the rate of the reaction. However, the enzyme was not inhibited by either NaF or EDTA at concentrations up to 1 x 10-z M. The activity was not affected by other metal-chelating agents, such as a,(~‘- dipyridyl, o-phenanthroline, and 8-hydroxyquinoline at 1 x 10-3 M. Among the metals tested, only CuS04 and HgCl* exhibited an inhibitory effect. 3-Hydroxyanthranilic acid metabolites, such as picolinate, quinolinate, niacin, nicotinamide, NAD, cu-ketoadipate, and glutarate, as well as cY-hydroxymuconic acid, did not influence the rate of the reaction at 2 x lop4 M. How- ever, cr-hydroxymuconic e-semialdehyde inhibited the reaction to about 15% of the rate in buffer alone, even at a low concentration of 3.3 x lo+ M. The enzyme activity was also inhibited by p-chloromercuribenzoate. This inhibition was prevented by n-cysteine. These results suggest that a sulfhydryl group might be involved in this reaction.

Purijication and Properties of a-Hydroxymuconic t-Xemialdehyde Dehydrogenase

As described above, a-amino-@-carboxymuconic e-semialdehyde was shown to be converted nonenzymically to or-hydroxymuconic c-semialdehyde under acidic conditions. The latter compound was characterized by its absorption maximum at 375 nm at neutral and alkaline pH values (9, 27, 28, 31). In a preceding report from this laboratory, a-hydroxymuconic e-semialdehyde was shown to be oxidized to a-hydroxymuconic acid by an NAD-linked specific aldehyde dehydrogenase obtained from o-cresol-adapted cells of Pseudomonas (32). Concurrently with Henderson and MitchelF we have observed the presence of a similar enzyme in the liver and kidney of several species of mammals. The enzyme was purified about 50- to loo-fold from cat and rat livers, and from o-cresol-adapted cells of Pseudomonas. Purification and properties of the cat liver enzyme are described below.

Purification-All manipulations were carried out at 2-4” unless otherwise noted. In a representative run of purification, frozen cat liver (40 g) was thawed in 160 ml of 0.14 M KC1 containing lop4 M EDTA, and homogenized for 3 minutes in a Waring Blendor. The homogenate was centrifuged for 15 minutes at 20,000 x g.

Step 1: The crude extract (165 ml) was brought to pH 7.0 with 1 M K2HP04, and 55 ml of a 0.4% solution of protamine sulfate were added with stirring. The precipitate that formed was removed by centrifuga,tion.

Step 2: The supernatant solution (211 ml) was warmed rapidly to 49” with continuous stirring in a water bath. After this temperature was maintained for 3 minutes, the solution was cooled to 1-2” in an ice-salt bath, and the resulting precipitate was removed by centrifugation.

Step 3: To the enzyme solution (205 ml), 88 ml of peroxide- free, cold acetone (-30”) were added slowly with mechanical

2 L. M. Henderson, personal communicat,ion.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


stirEing (30% final concentration of acetone). The temperature of the mixture was controlled at -2 to -10” in a Dry Ice- methanol mixture. The resulting precipitate was immediately removed by centrifugaticn at -10”. To the supernatant solution, additional cold acetone was added in the same manner to give a 40% final concentration of acetone. After 3 minutes the precipitate was collected by centrifugation at - 10” and dissolved in 45 ml of 0.0075 M potassium phosphate buffer, pH 7.0, containing 10m4 M EDTA. Insoluble materials were removed by centrifugaticn.

Step 4: The enzyme solution (51 ml) was passed through a DEAE-cellulose column (diameter, 3.5 cm; length, 5 cm), equilibrated with 0.0075 M potassium phosphate buffer, pH 7.0, containing lop4 M EDTA. Only inactive proteins were adsorbed on the column. The column was washed with 80 ml of the buffer, and the effluent and washings were combined.

Step 5: To the combined solution (108 ml), 108 ml of a saturated ammonium sulfate solution, pH 7, were added slowly with continuous stirring. After 15 minutes, the precipitate was removed by centrifugation. To the supernatant solution, an additional 55 ml of the ammonium sulfate solution were added. After 15 minutes, t.he precipitate was collected by centrifugation and dissolved in 30 ml of 0.01 M potassium phosphate buffer, pH 7.0, containing lop4 M EDTA.

Step 6: To the enzyme solution (31 ml), 31 ml of a calcium phosphate gel suspension (30 mg, dry weight, per ml) were added with mechanical stirring. After 10 minutes, the gel was collected by centrifugation; it was washed once with 20 ml of distilled water and then twice with 0.05 M KBHPO4,30 ml each time. The enzyme was then eluted twice from the gel, each time with 13 ml of 0.2 M KSHPOI. The eluates were combined and brought to pH 7.0 by the addition of 1 M KHZPO,. EDTA was added to a final concentration of lop4 M.

By the above procedures, the enzyme was purified 90- to lOO- fold with an over-all yield of 60 to 70% (Table IV). The enzyme preparation thus obtained was free from kynureninase, 3-hydroxyanthranilic acid oxygenase, picolinic carboxylase, quinolinate transphosphoribosylase, cr-aminomuconic acid reductase, and a+ketoglutarate dehydrogenase. The enzyme was relatively unstable. Approximately 50$& of the activity was lost after storage at - 10” for a week.

Product of Reaction-When cr-hydroxymuconic c-semialdehyde was incubated with NAD and the purified enzyme, the absorption at 375 rnp decreased rapidly with concomitant formation of a new compound with an absorption maximum at 295 nip, as shown in Fig. 5. No reaction occurred in t.he absence of either NAD or the enzyme. The maximal absorbance of the reaction product shifted to 305 nip under acidic conditions (below pH 3.5)

and to 350 rnp under alkaline conditions (above pH 10.5) (Fig. 5). These spectral changes of the reaction product were reversible and identical with that of a synthetic sample of cu-hydroxymu-

conic acid (Table I). In order to confirm that the product is oc-hydroxymuconic

acid, catechoFH, 0.2 pmole (7 X lo6 c.p.m. per pmole), was

incubated with a purified preparation of metapyrocatechase (about 10 p/g) and 200 pmoles of potassium phosphate buffer,

pH 8.0, in a total volume of 20 ml. When the formation of cY-hydroxymuconic e-semialdehyde was complete as judged by its

absorpticn at 375 rnp, 1 pmole of NAD and 0.28 mg of cr-hydroxymuconic t-semialdehyde dehydrogenase (Step 6, Table IV) were added. The mixture was further incubated for a.n addi-

TABLE IV

Summary of purijkation o.f a-hydroxymuconic e-semialdehyde dehydrogenase fhm cdt liver

Crude extract.. 1. Protamine sulfate. 2. Heat treatment. 3. Acetone. 4. DEAE-cellulose. 5. Ammonium sulfate. 6. Calcium phosphate gel.

Vol- ume Activity Protein Specific

activity Yield

ml units mg unit/mg %

165 5.000 6105 0.0008 100.0 211 4.795 4431 0.0011 95.9 205 4.100 3280 0.0013 82.0

51 11.592 1173 0.0100 231.8 108 6.775 572 0.0118 135.5

31 3.875 89 0.0435 77.5 25 3.364 44 0.0765 67.3

E

2 0.6

n 0.5

2

0.7 i

0.4

f 0.3

c% 0.2 0.1 k

z 0.8 g 0.7

9” 0.6 J 0.5

4 0.4

E 0.3

O 0.2

0.1

WAVE LENGTH (mp) WAVE LENGTH ( mp 1

FIG. 4 (left). Spectral change of a-amino-p-carboxymuconic e-semialdehyde in the presence of picolinic carboxylase. The reaction mixture (3.0 ml) contained (in micromoles) Tris-acetate buffer, pH 8.0, 200; 3-hydroxyanthranilic acid, 0.063; 3-hydroxyanthranilic acid oxygenase, about 0.4 mg; and picolinic carboxvl- ase (Step 5, Table II), 0.075 mg. Incubation was carried out”at 24”. The reaction was started bv the addition of nicolinic carboxylase. Recordings were taken at 30 seconds (G&e 1), 1 minute (Curue .2), 3 minutes (Curve S), 6 minutes (Curve 4)) 10 minutes (Curve 5), 15 minutes (Curve 6), and 20 minutes (Curve 7) after the addition of picolinic carboxylase.

FIG. 5 (right). Spectral change of ar-hydroxymuconic e-semialdehyde in the presence of the dehydrogenase and NAD. The reaction mixture (3.0 ml) contained (in micromoles) Tris-acetate buffer, pH 8.0, 100; NAD, 0.2; a-hydroxymuconic e-semialdehyde, 0.09; pyruvate, 5; lactic dehydrogenase, 0.02 mg; and cu-hydroxymuconic e-semialdehyde dehydrogenase (Step 6, Table IV), 0.3 mg. Incubation was carried out at 24”. The reaction was started by the addition of the dehydrogenase. Recordings were taken at 30 seconds (Curve 1), 1 minute (Curve 2), 2 minutes (Curve S), 4 minutes (Curve 4), and 7 minutes (Curve 5) after the addition of the dehydrogenase. After 10 minutes, an aliquot of the mixture was made 0.1 N with respect to NaOH (-.-., pH 13), and another 0.1 N with respect to HCl (-.-. , pH 1).

tional 20 minutes at 24”. At this point, a-hydroxymuconic e-semialdehyde disappeared completely. After the addition of 100 mg of an authentic sample of nonradioactive oc-hydroxymuconic acid, the reaction was stopped by the addition of 40 ml of a saturated 2,4-dinitrophenylhydrazine solution (in 2 N HCl). After standing for 2 hours at 24”, the resulting precipitate was collected by centrifugation, washed with 20 ml of cold water, and then crystallized several times from ethanol-water and ethyl acetate-n-hexane. The specific activity of each sample re- mained constant (7200 c.p.m. per mg). The final product


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


TABLE V

Xtoichiometry of dehydrogenation reaction

The details of the experimental conditions are described in the text.

System

Complete ..................... -Substrate. ................ -NAD .....................

Complete with boiled enzymet.

a-Hydroxy- muconic 6

semialdehyde disappeared*

m/.moles

45.9

0

0.1 0

a-Hydroxy- NADH muconic acid aP-

appeared* peared

mpmoles mpmles

46.3 46.4 0 0 0.2 0.3 0 0

* The values were corrected for the interference of NADH

formed. t Heated for 10 minutes in a boiling water bath.

0.10

2 0.08

12 Ic) 0.06

z

d 0.04

.

0

0.02 4

r 5

PH

FIG. 6. Effect of pH on the activity and the stability of au-hydroxymuconic e-semialdehyde dehydrogenase. The activity (-) was measured under the standard assay conditions in 0.1

M buffers at various pH values. Potassium phosphate buffer (0) was used between pH 5.5 and 8.0, Tris-acetate buffer (A) in the range of 7.5 to 9.5, and glycine-NaOH buffer (m) in the range of 10.0 to 10.5. drogenase.

- - -, effect of pH on the stability of the dehy- Enzyme (Step 6, Table IV) was kept for 8 hours at

various pH at 4”, and the activity was determined under the standard assay conditions with 0.2 M Tris-acetate buffer, pH 8.0. C- - -0, stored in potassium phosphate buffer; A, in Tris-acetate buffer; 0, in glycine-NaOH buffer.

TABLE VI

Relative activity of aldehyde clehydrogenase with various substrates

The reaction mixture contained potassium phosphate buffer, pH 8.0, 100 pmoles; NAD, 0.27 pmole; aldehydes as indicated, 0.1 pmole; and the purified preparation of ol-hydroxymuconic E- semialdehyde dehydrogenase (Step 6, Table IV), 0.18 mg, in a final volume of 3.0 ml. Incubation was carried out at 24”. Oxi- dation of ti-hydroxymuconic e-semialdehyde was followed at 375 mp. Activities for other aldehydes were assayed by the increase of the optical density at 340 rnp.

Substrate I Relative rates

a-Hydroxymuconic e-semialdehyde Formaldehyde Acetaldehyde n-Propylaldehyde n-Butylaldehyde . Crotonaldehyde.

% 100 50

7 0

32 0

thus obtained had a melting point (185.5-187”) and an elemental analysis corresponding to the hydrazone derivative of oc-hy-

droxymuconic acid. The radioactive hydrazone derivative

was further identified by paper chromatography in five different solvent systems. RF values of the radioactive compound and

the authentic sample on Whatman No. 1 paper were identical and as follows: with tert-amyl alcohol-ethyl alcohol-O.1 N NaHC03 (5:1:4, upper layer), 0.55 and 0.78; with isopropyl alcohol- ammonia-water (20:1:2), 0.13; with n-butyl alcohol-ethyl alcohol-water (7: 1:2), 0.24 and 0.31; with ethyl alcohol-ammonia-water (20: 1:4), 0.64; and with tert-amyl alcohol-ethyl alcohol-water (5: 1:4, upper layer), 0.38 and 0.50, respectively. Two figures given for a hydrazone indicate Rg values for its stereoisomers.

Xtoichiometry of Reaction-The complete reaction mixture initially contained 91 mpmoles of cr-hydroxymuconic c-semialdehyde, 1 pmole of L-cysteine, 100 pmoles of potassium phosphate buffer, pH 8.0, and 0.13 mg of a purified preparation of c+hydroxymuconic c-semialdehyde dehydrogenase (Step 6, Table IV)

in a total volume of 3.0 ml. The mixture was incubated at 24”, and the reaction was followed spectrophotometrically. After 5 minutes, the amounts of a-hydroxymuconic acid formed and of oc-hydroxymuconic e-semialdehyde disappeared were cal-

culated from the increase in absorbance at 296 rnp and the decrease in that at 375 rnp, respectively. NADH produced was calculated from the decrease in absorbance at 340 rnp upon the addition of pyruvic acid and crystalline heart muscle lactic dehydrogenase. As shown in Table V, the appearance of a- hydroxymuconic acid and NADH was exactly matched by the disappearance of oc-hydroxymucnoic e-semialdehyde.

E$ect of pH and Substrate Concentration-The effects of pH on the activity and stability of the dehydrogenase are given in Fig. 6. The optimal pH for activity in 0.1 M Tris-acetate and potassium phosphate buffer was found between pH 7.5 and 8.0. The enzyme exhibited a relatively broad range of maximal stability between pH 7.0 and 9.0.

The K, values for a-hydroxymuconic e-semialdehyde and for NAD, as calculated from the Lineweaver-Burk plot (30), were approximately 1.6 X 10e5 and 1.9 X lo+ M, respectively.

Substrate SpeciJcity-In the dehydrogenation reaction of or-hydroxymuconic t-semialdehyde, NADP could not serve as a hydrogen acceptor. The preparation of most purified enzyme also oxidized several other aldehydes, including formaldehyde, acetaldehyde, n-butyraldehyde, and a-methyl-oc-hydroxymuconic esemialdehyde, but not as rapidly as a-hydroxymuconic e-semialdehyde (Table VI). Propionaldehyde and crotonaldehyde in comparable concentrations did not lead to the formation of NADH in this system.

Inhibitor and Activator StudiesThe enzyme was inhibited by p-chloromercuribenzoate (100% at 2 X lop4 M, 52$& at 10-S M).

This inhibition was completely reversed by the addition of either n-cysteine or reduced glutathione at a concentration of 10-3 M.

This suggests that the dehydrogenase might be a sulfhydryl enzyme. Among the heavy metals tested, Co++, NF, Hg++, and C&f inhibited the enzyme almost completely at 2 x 10-4 M. Metal-chelating agents, such as EDTA, Q ,a’-dipyridyl, o-phenanthroline, and 8-hydroxyquinoline, at 2 x 10-J M,

showed no effect on the rate of reaction. As shown in Table IV, acetone fractionation of the enzyme

resulted in a a-fold increase of the total activity. However,


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


neither acetone (5%) nor boiled extract (5%) influenced the rate of the reaction. The activity was not affected by 3-hydroxyanthranilic acid metabolites, including picolinate, quinolinate, a-ketoadipate, and glutarate, at 1 X 10P3 M.

Formation of oc-Aminomuconic Acid from or-Amino-P-carboxymuconic ~-Semialdehyde by Combined Action of Picolinic

Carboxylase and ol-Hydroxymuconic t-Xemialdehyde Dehydrogenase

As described above, the @-carboxy group of a-amino-& carboxymuconic t-semialdehyde was shown to be decarboxylated to yield picolinic acid (6). The formation of the pyridinecarboxylic acid, however, appears to take place in two steps with the formation of an acyclic, unsaturated aldehyde as an intermediate. Instead of cyclization to picolinic acid, the aldehyde was shown to be oxidized to a+aminomuconic acid in the presence of NAD and a-hydroxymuconic c-semialdehyde dehydrogenase. When the purified picolinic carboxylase was added to a reaction mixture containing a-amino-@arboxymuconic t-semialdehyde, the absorption peak at 360 rnp decreased rapidly with a concomitant formation of picolinic acid as shown in Fig. 4. In addition to picolinic carboxylase, when purified a-hydroxymuconic e-semialdehyde dehydrogenase and NAD were added to the reaction mixture, a new compound was formed which showed an absorption maximum at 325 rnp (Fig. 7). When NAD was omitted from the reaction mixture, or when the preparation of the dehydrogenase heated at 100” for 5 minutes was employed, no compound other than picolinic acid accumulated in the reaction mixture and the spectral change was the same as that shown in Fig. 4.

Product of Reaction-Because of its instability, the intermediate compound has not yet been isolated, but chemical studies were performed to characterize this compound. The absorption spectrum of the product with a maximum at 325 rnp did not change within the pH range of 5.0 to 9.0. Under alkaline pH (0.1 N with respect to NaOH), the maximal absorbance shifted to 333 rnp and the process was reversible. Under acidic conditions (below pH 4.5), however, the maximal absorbance shifted irreversibly to 305 mp. It took about 20 minutes for the com- pletion of this spectral change. Thereafter, the absorption maximum shifted promptly to 297 rnp under neutral conditions (pH 5.5 to 9.5) and to 350 rnp at alkaline pH (above 10.5). These spectral changes of the latter product, which was obtained by acidification of the 325 rnp-absorbing compound, were exactly the same as those of a-hydroxymuconic acid (Table I and Fig.

5). In order to confirm that the acidified product of the 325 m,u-

absorbing compound is ol-hydroxymuconic acid, 0.33 pmole of 3-hydroxykynurenine, uniformly labeled in the benzene ring with 1% (247,000 c.p.m. per kmole), was incubated with 300 pmoles of Tris-acetate buffer, pH 8.0, and a purified preparation of Pseudomonas kynureninase (1.76 mg of protein) in a total volume of 3 ml at 24”. The reaction was followed spectrophotometrically. When the formation of 3-hydroxyanthranilic acid was complete (about 20 minutes), the mixture was diluted to 14 ml with 0.05 M Tris-acetate buffer, pH 8.0, and 0.6 ml of 3-hydroxyanthranilic acid oxygenase (2.1 mg of protein) was added. After about 3 minutes, the optical density at 360 mp of ar-amino- P-carboxymuconic r-semialdehyde reached its maximum value (theoretical, 1.017; found, 1.006). To the reaction mixture, 2

0.8

z 0.7

# 0.6

Q 0.5

d 0.4 0 k 0.3

0 0.2

0.1

WAVE LENGTH (my 1

FIQ. 7. Spectral change of or-amino-p-carboxymuconic e-semialdehyde by the combined action of picolinic carboxylase and oc-hydroxymuconic c-semialdehyde dehydrogenase. The reaction mixture (3.0 ml) contained (in micromoles) Tris-acetate buffer, pH 8.0, 200; 3-hydroxyanthranilic acid, 0.062; 3-hydroxyanthranilic acid oxygenase, about 0.4 mg; NAD, 0.2; pyruvate, 5; lactic dehydrogenase, 0.02 mg; picolinic carboxylase (Step 5, Table II), 0.075 mg; and rw-hydroxymuconic e-semialdehyde dehydrogenase (Step 6, Table IV), 0.4 mg. Incubation was carried out at 24”. The reaction was started by the addition of picolinic carboxylase. Recordings were taken at 1 minute (Curve 1), 3 minutes (Curve 2), 5 minutes (Curve 5), 8 minutes (Curve 4), 13 minutes (Curve 6), and 20 minutes (Curve 6’) after the addition of picolinic carboxylase.

kmoles of NAD, 30 pmoles of pyruvate, 0.2 mg of crystalline heart muscle lactic dehydrogenase, 0.058 mg of the purified preparations of picolinic carboxylase, and 1.09 mg of a-hydroxymuconic e-semialdehyde dehydrogenase were added (final volume, 16 ml). The mixture was further incubated at 24”. When the formation of the 325 rnp-absorbing compound was complete, the reaction was stopped by the addition of 0.8 ml of 5 N HzS04 and the precipitate was removed by centrifugation. The shift of the maximal absorption from 325 rnp to 305 rnp was complete within 20 minutes. The supernatant solution was concentrated to about 2 ml under reduced pressure, then chromatographed on a silicic acid column (diameter, 1.2 cm; length, 50 cm) after the addition of 20 pmoles of an authentic sample of or-hydroxymuconic acid as carrier. Elution was carried out with an n-butyl alcohol-chloroform system (33). Fractions (10 g each) were titrated with 0.01 N NaOH and assayed for radioactivity. A major radioactive peak appeared which coincided with the titration peak of a-hydroxymuconic acid.

The radioactive oc-hydroxymuconic acid thus obtained was treated with an equal volume of a saturated 2,4-dinitrophenylhydrazine solution (in 2 N HCl). The resulting hydrazone derivative was extracted with ethyl acetate. The organic layer was washed twice with cold distilled water and then re- extracted with 0.1 N NaHC03. The aqueous layer was washed once with ethyl acetate, acidified with 6 N HCl, and then extracted again with a small quantity of ethyl acetate. The radioactive hydrazone derivative obtained was identified as that of ac-hydroxymuconic acid by paper chromatography in five differ-


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


ent solvent systems described a.bove, and by paper electrophoresis.

In another set of experiments with 2.2 Mmoles of 3-hydroxykynurenine-i4C as substrate, the radioactive 325 rnp-absorbing compound was prepared in the same manner, and the reaction was stopped finally by the addition of 15 ml of 10 N KOH. After the addition of 1.3 mg of an authentic samples of cy-aminoadipic acid, catalytic hydrogenation was carried out for 20 hours at 24” in the presence of Raney nickel. The mixture was neutralized to pH 7 with 20% perchloric acid. The mixture was treated with activated charcoal and evaporated to dryness under reduced pressure. The dried residue was extracted five times with 80% ethanol, and an aliquot of the combined extracts was subjected to paper chromatography. A radioactive spot was detected and shown to coincide with that of a-aminoadipic acid in two different solvent systems. The RF value of a-aminoadipic acid with phenol-HZ0 (8:2) was 0.40, and with n-butyl alcohol-pyridine-Hz0 (I: 1: 1) it was 0.18. Radioactive a-aminoadipic acid was further identified by high voltage paper electrophoresis. The mobility of the amino acid at pH 5.0 (2000 volts, 50 minutes) was 3.2 cm to the anode. Furt,her evidence for the identity was provided by repeated crystallization from ethanol and water to constant specific activity (100 c.p.m. per mg). The final product thus obtained had the same melting point (212”) as an authentic sample of oc-aminoadipic acid. The total radioactivity recovered as a-aminoadipic acid was about 60% of that of the original substrate. As a control for this experiment, ol-hydroxymuconic acid was reduced in the same manner, but no ar-aminoadipic acid was formed as judged by paper chromatographic and paper electrophoretic criteria. The results described above indicate that the product of the combined reaction of picolinic carboxylase and a-hydroxymuconic t-semialdehyde dehydrogenase is a-aminomuconic acid, and that the primary product of the picolinic carboxylase- catalyzed reaction is probably oc-aminomuconic e-semialdehyde.

DISCUSSION

Early experiments in viva have provided evidence that the benzene ring of tryptophan is rapidly degraded to COZ via 3-hydroxyanthranilic and glutaric acids (34-39). However, it was impossible for a long time to demonstrate in vitro a degradative pathway of 3-hydroxyanthranilic acid which could account for the observation obtained by st.udies in viva. Using cat liver extracts, we have recently demonstrated the conversion of 3-hydroxyanthranilic acid to glutaric acid and have shown that cr-amino-/%carboxymuconic t-semialdehyde is an obligatory intermediate in this conversion (l-3). The latter compound is converted to quinolinic acid as was shown first by Henderson and Hirsch (5), and further to niacin ribonucleotide in the presence of PP-ribose-P (16, 29, 40). The cyclizat.ion of the acyclic intermediate is, however, generally regarded to be nonenzymic (6, 14), and picolinic carboxylase has been known as the only enzyme which reacts with ol-amino-P-carboxymuconic c-semialdehyde (6). The product obtained so far has been picolinic acid. The latter pyridinecarboxylic acid is, however, metaboli- cally inert and is excreted in the urine as the glycine conjugate when administered to mammals (7, 8).

The experimental evidence presented in this report leads us to the reasonable postulate that the formation of picolinic acid takes place in two steps: enzymic decarboxylation of the p-car- boxy1 of cr-amino-P-carboxymuconic e-semialdehyde and a

subsequent cyclization to picolinic acid. The term “picolinic carboxylase” was originally proposed by Mehler to designate the enzyme which catalyzes the decarboxylation of cu-amino-& carboxymuconic e-semialdehyde to form picolinic acid. How- ever, it would be more reasonable to name this enzyme a-amino- P-carboxymuconic t-semialdehyde decarboxylase.

Although little difference was observed between the rate of decarboxylation and the decrease in absorption at 360 nip, a small shift of the absorption maximum at 360 rnp of a-amino-p- carboxymuconic t-semialdehyde to 364 rnp was observed during the picolinic carboxylase-catalyzed reaction. This suggests the existence of a transient acyclic intermediate, possibly (Y- aminomuconic e-semialdehyde. We also found that the unstable, acyclic intermediate is metabolized in the presence of an NAD-linked aldehyde dehydrogenase and NAD to ar-aminomuconic acid, rather than being cyclized nonenzymically to picolinic acid. a-Aminomuconic acid was identified by various methods described in the present paper. All attempts to isolate cr-aminomuconic acid from the reaction mixture, however, were unsuccessful since this acid is readily converted to a-hydroxymuconic acid under acidic conditions.

Subsequent experiments have shown that o+aminomuconic acid is reductively deaminated’to oc-ketoadipic acid in the presence of either NADH or NADPH. The keto acid is then oxidatively decarboxylated to glutaryl-CoA.1 The detailed experimental results will be described in a following report in this series.

Several experiments were performed to ascertain whether or not a single enzyme was responsible for the oxidation of the aldehyde moiet,ies of a-aminomuconic and cu-hydroxymuconic e-semialdehydes. Although it is not possible to reach a con- clusion with regard to the purity of the enzyme, there is no indication at present that the activity is separable. a-Hydroxy- muconic t-semialdehyde dehydrogenase also reacts with n-butyraldehyde, formaldehyde, and &methyl-a-hydroxymuconic e-semialdehyde to some extent, but not with propionaldehyde. Acetaldehyde was a poor substrate. The nonspecific aldehyde dehydrogenase of the liver reported by Racker (41) oxidizes various aldehydes, including acetaldehyde and propionaldehyde, at various rates, and acetaldehyde is the most efficient substrat’e. Steroid-sensitive liver aldehyde dehydrogenase reported by Maxwell and Topper (42) also oxidizes n-propionaldehyde. These results, together with the difference in the optimal pH and heat stability, indicate that these dehydrogenases belong to different entities. NADP does not serve as a hydrogen acceptor. The aldehyde dehydrogenases purified from yeast (43), Acetobacter suboxydans (44) and Pseudomonas jluorescens (45, 46) react with NADP as well as N9D. il different type of NAD- linked dehydrogenase found in Clostridium by Burton and Stadtman (47) requires CoA in addition to NAD as an essential cofactor.

In a preliminary report from our laboratory, cr-hydroxymuconic e-semialdehyde and c+hydroxymuconic acid (a tautomer of y-oxalocrotonate) were proposed to be intermediates in the catabolic pathway of 3-hydroxyanthranilic acid (1). The postu- lation was based on the results of i4C trapping experiments with ol-hydroxymuconic acid as a cosubstrate. Although Lu-hydroxymuconic t-semialdehyde is metabolized rapidly to glutaryl- CoA and COZ via a-hydroxymuconic acid by another liver enzyme system, no available evidence has been obtained to indicate that ac-hydroxymuconic e-semialdehyde is enzymically


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


formed from 3-hydroxyanthranilic acid. The incorporation of the radioactivity of 3-hydroxyanthranilic acid-% as a starting material into a-hydroxymuconic acid is probably due to the nonenzymic conversion of c+aminomuconic acid to a-hydroxymuconic acid under the acidic conditions employed earlier (1).

The common difficulty of the previous studies on the degradative pathway of 3-hydroxyanthranilic acid in vitro can be ascribed to the extreme instability of the primary product of the picolinic carboxylase-catalyzed reaction, This intermediate cyclizes rapidly to picolinic acid unless enzymically dehydrogenated. The relative ratios of the format,ion of quinolinic, picolinic, and oc-aminomuconic acids depend mainly upon the relative activities of 3-hydroxyanthranilic acid oxygenase, picolinic carboxylase, and cY-hydroxymuconic e-semialdehyde dehydrogenase. The quantitative aspects concerning the metabolism of tryptophan will be described elsewhere.

SUMMARY

In addition to picolinic acid, a new compound is produced from cY-amino-P-carboxymuconic t-semialdehyde by the combined action of picolinic carboxylase and a nicotinamide adenine dinucleotide-linked specific aldehyde dehydrogenase. Although this compound could not be isolated in a pure form (since it is converted readily to a-hydroxymuconic acid under acidic conditions), available evidence indicates that the compound is a- aminomuconic acid. Picolinic carboxylase and the aldehyde dehydrogenase were both purified about loo-fold from cat liver extracts and were characterized. The purified preparation of the latter enzyme also reacted with several other aldehydes, including formaldehyde and n-butyraldehyde, to some extent. As judged by its substrate specificity and pH optimum, however, the enzyme appeared to be different from the nonspecific aldehyde dehydrogenases previously described by other investiga- tors. Nicotinamide adenine dinucleotide phosphate did not serve as a hydrogen acceptor in this reaction.

Acknowledgments-We are grateful to Drs. T. Suzuki and K. Hayashi, Department of Biochemistry, Kyoto University Faculty of Pharmaceutical Science, and Dr. H. Nozaki, Depart- ment of Organic Chemistry, Kyoto University Faculty of Technology, for their many stimulating discussions and helpful suggestions.

REFERENCES

1. GHOLSON, I~. K., NISHIZUKA, Y., ICHIYBMA, A., KAWAI, H., NAKAMURA, S., AND HAYAISHI, O., J. Biol. Chem., 237, PC2043, (1962).

2. ICHIYAMA, A., KAWAI, H., NAK~MURA, S., HONJO, T., IKEDA, M., NISHIZUKA, Y., HAYAISHI, O., BND SENOH, S., Syrn- posia on Enzyme Chem. (Japan), 15, 198 (1963). -

3. NISHIZUKA. Y.. ICHIYAMA. A.. GHOLSON. R. K.. AND HAYAISHI. O., J. Biol. Chem., 240; 733 (1965). ’ ’

4. WISS, O., Proceedings of the International Symposium on enzyme chemistry (Tokyo-Kyoto), 1957, Maruzen Company, Ltd., Tokyo, 1958, p. 200.

5. HENDERSON, I,. M., AND HIRSCH, H. M., J. Biol. Chem., 181, 667 (1949).

6. MEHLER, A. H., J. Biol. Chem., 218, 241 (1956). 7. MEHLER, A. H., AND MAY, E. L., J. Biol. Chem., 223, 449

(1956). 8. SENDJU, Y., J. Biochem. (Tokyo), 7, 273 (1927). 9. KOJIMA, Y., ITADA, N., AND HAYAISHI, O., J. Biol. Chem.,

236, 2223 (1961). 10. LAPWORTH, A., J. Chem. Sot., 79, 1265 (1901). 11. PETERSON, E. A., AND SOBER, H. A., in S. P. COLOWICK AND

N. 0. KAPLAN (Editors), Methods in enzymology, VoZ. V, Academic Press, Inc., New York. 1962. n. 3.

12. COLOWICK, S. P., in S. @. C~LOWIC& BND’~~. 0. KAPLAN (Edi- tors), Methods in enzymology, Vol. I, Academic Press, Inc., New York, 1955, p. 97.

13. TISELIUS, A., HJERTI~N, S., END LEVIN, o., Arch. Biochem. Biophys., 66, 132 (1956).

14. LONG, C. L., HILL, H. N., WEINSTOCK, I. M., AND HENDER- SON, L. M., J. Biol. Chem., 211,405 (1954).

15. MIY.~KE, A., BOKMAN, A. H., END SCHWEIGERT, B. S., J. Biol. Chem., 211, 391 (1954).

16. NAKAMURA, S., IKEDA, M., TSUJI, H., NISHIZUKA, Y., AND HAYAISHI, O., Biochem. and Biophys. Research Communs., 13, 285 (1963).

17. HAYAISHI, O., AND STANIER, R. Y., J. Biol. Chem., 196, 735 (1952).

18. NOZAKI, M., KAGAMIYAMA, H., AND HAYAISHI, O., Biochem. and Biophys. Research Communs., 11, 65 (1963).

19. NOZAKI, M., KAGAMIYAMA, H., AND HAYAISHI, O., Biochem. Z., 338, 582 (1963).

20. DALGLIESH, C. E., Biochem. J., 52, 3 (1952). 21. DANNENBERG, H., Z. Naturforsch., Pt. b, 4, 327 (1949). 22. HAYAISHI, O., NISHIZUKA, Y., TaTIBANA, M., TAKESHITA,

M., AND KUNO, S., J. Biol. Chem., 236, 781 (1961). 23. LOWRY, 0. H., ROSEBROUGH, N. J., FARR, A. L., AND RANDALL,

R. J., J. Biol. Chem., 193, 265 (1951). 24. WANG, G. H., AND JONES, D. EL, Biochem. and Biophys. Re-

search Communs., 1, 203 (1959). 25. JOHNSON, B. C., AND LIN, P. H., J. Am. Chem. Sot., 75, 2971

(1953). 26. SUHADOLNIK, R. J., STEVENS, C. O., DECKER, R. H., HENDER-

SON, L. M., AND HANKES, L. V., J. Biol. Chem., 228, 973 (1957).

27. MEHLER, A. H., Proceedings of the Fourth International Con- gress of Biochemistry, Vienna, 1958, Vol. 18, Pergamon Press, New York, 1960, p. 164.

28. DAGLEY, S., EVANS, W. C., AND RIBBONS, D. W., Nature, 188, 560 (1960).

29. NISHIZUKA, Y., AND HAYAIsHI, O., J. Biol. Chem., 238, 3369 (1963).

30. LINEWEAVER, H., AND BURK, D., J. Am. Chem. Sot., 56, 658 (1934).

31. DAGLEY, S., AND STOPHER, D. A., Biochem. J., 73, 169 (1959). 32. NISHIZUKA, Y., ICHIYAMA, A., NAKAMURA, S., AND HAYAISHI,

O., J. Biol. Chem., 237, PC269 (1962). 33. VARNER, J. E., in S. P. COLOWICK AND N. 0. KAPLAN (Edi-

tors), Methods in enzymology, Vol. III, Academic Press, Inc., New York, 1957, p. 397.

34. HENDERSON, L. M., AND HANKES, L. V., J. Biol. Chem., 222, 1069 (1956).

35. DALGLIESH, C. E., AND TBBECHIAN, H., Biochem. J., 62, 625 (1956).

36. GHOLSON, R. K., RAO, D. R., HENDERSON, L. M., HILL, R. J.. AND K~EPPE, It. E., J. Biol. Chem., 230; 179 (1958). ’

37. GHOLSON. R. K.. AND HENDERSON. L. M.. Biochim. et Bionhus. Acta, 36, 424 (1958).

1 Y

38. GHOLSON, R. K., HENDERSON, L. M., MOURKIL)ES, G. A., HILL, R. J., AND KOEPPE, R. E., J. Biol. Chem., 234, 96 (1959).

39. GHOLSON, R. K., HANKES, L. V., ANU HENDERSON, L. M., J. Biol. Chem., 235, 132 (1960).

40. NISHIZUKA, Y., AND HAYAISHI, O., J. Biol. Chem., 238, PC483 (1963).

41. RACKER, E., J. Biol. Chem., 177, 883 (1949). 42. MAXWELL, E. S., AND TOPPER, Y. J., J. Biol. Chem., 236, 1032

(1961). 43. BLACK, S., Arch. Biochem. Biophys., 34, 86 (1951). 44. KING, T. E., AND CHELDELIN, V. H., J. Biol. Chem., 220,177

(1956). 45. JAKOBY, W. B., J. Biol. Chem., 232, 75 (1958). 46. JAKOBY, W. B., AND SCOTT, E. M., J. Biol. Chem., 234,937

(1959). 47. BURTON, R. M., AND ST~DTMAN, E. R., J. Biol. Chem., 202,

873 (1953).


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Nishizuka, Osamu Hayaishi and Siro SenohArata Ichiyama, Shigenobu Nakamura, Hitoshi Kawai, Tasuku Honjo, Yasutomi

FROM 3-HYDROXYANTHRANILIC ACID-AMINOMUCONIC ACIDαTissues: II. ENZYMIC FORMATION OF

Studies on the Metabolism of the Benzene Ring of Tryptophan in Mammalian

1965, 240:740-749.J. Biol. Chem.

http://www.jbc.org/content/240/2/740.citation

Access 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/240/2/740.citation.full.html#ref-list-1

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


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/content/240/2/740.citation

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;240/2/740&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/240/2/740.citation

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=240/2/740&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/240/2/740.citation

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/240/2/740.citation.full.html#ref-list-1

http://www.jbc.org/

Documents

Studies on the Metabolism of the Benzene Ring of …zymes. The liver was removed immediately after death, packed in ice, trimmed of connective tissues and gall bladder, washed with