THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 256, No. 20. Issue of October 25, pp. 10335-10339, 1981 Printed in U S A .

Biosynthesis of Porphyrin Precursors in Mammals IDENTITY OF ALANINE:y,G-DIOXOVALERATE AMINOTRANSFERASE WITH ALAN1NE:GLYOXYLATE AMINOTRANSFERASE*

(Received for publication, January 5, 1981)

Tomoo Noguchi and Ryosuke Mori From the Department of Biochemistry, Kyushu Dental College, Kokura, Kitakyushu 803, Japan

Alanine: y,S-dioxovalerate aminotransferase had been purified from bovine liver mitochondria, and the capacity of this enzyme to form 8-aminolevulinic acid had been suggested to be f a r greater than that of S- aminolevulinate synthase (EC 2.3.1.37) from the same mitochondria (Varticovski, L., Kushner, J. P., and Burnham, B. F. (1980) J. Biol. Chem 255,3742-3747). In the present study, alanine: y,S-dioxovalerate amino- t ransferase and alanine-glyoxylate aminotransferase (EC 2.6.1.44) were co-purified to homogeneity from bo- vine liver mitochondria. The ratio of the two activities remains constant during purification and is unchanged by a variety of t reatments of the purified enzyme. Ala- nine:y,S-dioxovalerate aminotransferase activity is competitively inhibited by glyoxylate. Some kinetic data are presented. These results show that the two activities are associated with the same protein. The enzyme is much higher in the glyoxylate aminotrans- fe rase activity than in the dioxovalerate aminotrans- ferase activity. The purified enzyme has a molecular weight of approximately 240,000 with four identical subunits and an isoelectric point of 5.4. The ratio of the y,S-dioxovalerate aminotransferase activity to the gly- oxylate aminotransferase was determined with ala- nine:glyoxylate aminotransferase preparations from various mammal ian liver and kidney.

The formation of 6-aminolevulinic acid is the first step of the biosynthetic pathway of tetrapyrroles, leading to heme, chlorophyll, vitamin BI2, and other tetrapyrroles (1). It is generally accepted that 6-aminolevulinic acid is formed by 6- aminolevulinate synthase (EC 2.3.1.37) which catalyzes the condensation of glycine and succinyl CoA (2, 3).

However, recently an additional biosynthetic pathway of 6-aminolevulinic acid has been reported to be present in mammalian liver by Varticovski et al. (4); they have isolated and partially characterized alanine:y,6-dioxovalerate amino- transferase from bovine liver mitochondria, which catalyzes a transamination reaction between L-alanine and y,&dioxoval- erate yielding 6-aminolevulinic acid and pyruvate. The capac- ity of this enzyme to form 6-aminolevulinic acid appears to be far greater than that of 6-aminolevulinic acid synthase from the same mitochondria, suggesting the possibility that the enzyme plays a role in the biosynthesis of 6-aminolevulinic acid in vivo (4).

On the other hand, evidence continues to accumulate that in plant leaves, 6-aminolevulinic acid is synthesized exclu-

* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

sively via a transamination reaction between L-alanine and y,6-dioxovalerate (5-8); 6-aminolevulinic acid synthase has not been demonstrated in plant leaves (5, 6). Alanine:y,G-dioxo- valerate aminotransferase has been reported to be present in extracts of Zea mays (9), chlorella (lo), and Rhodopseudo- monas ( 11).

In the present report, we describe the identity of alanine: y,S-dioxovalerate aminotransferase with alanine-glyoxylate aminotransferase (EC 2.6.1.44) in mammalian liver and kid- ney.

EXPERIMENTAL PROCEDURES

Materials-y,S-Dioxovalerate was synthesized from 3,5-dibromo- levulinic acid as described by Varticovski et al. (4). Sources of other materials used are as previously described (12-14).

Enzyme Assays-Alanine:y,S-dioxovalerate aminotransferase was assayed at pH 7.0, as described by Varticovski et al. (4). &Aminole- vulinic acid formed was converted into 2-methyl-3-carbethoxy-4-(3- propionic acid)pyrrole by the method of Mauzerall and Granick (15) and Ehrlich chromophor measured at 553 nm. A1anine:glyoxylate aminotransferase was assayed at pH 8.0, as previously described (16, 17). Pyruvate formed was determined in the presence of Tris by using D-lactate dehydrogenase (EC 1.1.1.28). The assay mixture of amino- transferases contained, unless specified otherwise, 50 mM L-alanine, 2 mM 2-oxo acid, 40 PM pyridoxal 5’-phosphate, 0.1 M potassium phosphate buffer, and enzyme preparation in a total volume of 0.4 ml. Glutamate dehydrogenase (EC 1.4.1.3) (18), catalase (EC 1.11.1.6) (19), and acid phosphatase (EC 3.1.3.2) (20) were assayed as described in the cited references. A unit of enzyme activity is defined as the amount of enzyme that catalyzes a formation of product or a decrease in substrate of 1 pmol/min at 37 “C.

Organelle Separation-Bovine liver was cut into small pieces and homogenized with 4 volumes of 0.25 M sucrose in 20 mM glycylglycine, pH 7.5, for 10 s in a blender. The crude homogenate was squeezed through Miracloth and centrifuged at 300 X g for 5 min. Most of alanine:y,S-dioxovalerate aminotransferase and a1anine:glyoxylate aminotransferase activities were recovered in the supernatant frac- tion. The supernatant was layered over a 50-ml linear sucrose density gradient (24-54%, w/w) in 20 mM glycylglycine, pH 7.5, in amounts corresponding to 0.8 g of liver. The gradient was centrifuged at 58,000 X g for 3 h with a swing bucket rotor (RPS 25-2) in a 55P preparative ultracentrifuge (Hitachi, Tokyo). Fractions (2.5 ml) were collected from the bottom of the tube.

Purification of Alanine:y,S-dioxoualerate Aminotransferase and A1anine:glyoxylate Aminotransferase-All manipulations were car- ried out at 0-4 “C, and potassium phosphate buffer, pH 7.5, containing 10% glycerol, was used throughout. At each stage in the purification, alanine:y,S-dioxovalerate aminotransferase and a1anine:glyoxylate aminotransferase activities were determined. Both activities were always found in the same fractions through all purification steps.

Bovine liver (80 g) was cut into very small pieces with scissors and homogenized for 20 s in a blender with 4 volumes of 0.25 M sucrose in 5 mM buffer. The crude homogenate was filtered through six layers of coarse cheesecloth and centrifuged at 500 x g for 5 min to sediment nuclei and any whole cells. The postnuclear fraction was centrifuged at 10,OOO X g for 30 min and the sediment (the crude mitochondrial fraction) was homogenized with 320 ml of 5 mM phosphate buffer in a Waring blendor for 1 min. After sonication for 30 s at 10 kHz, the

10335

10336 Alanine:y,S-dioxoualerate Aminotransferase in Mammals

homogenate was centrifuged at 100,000 X g for 60 min and the precipitate discarded.

The supernatant (mitochondrial extract) was warmed rapidly to 70 “C, after which it was quickly chilled in a cold water bath. The precipitate was removed by centrifugation a t 5,000 X g for 20 min.

Solid (NHdhSOs was added to the resulting supernatant to 45% saturation. After being left for 30 min, the precipitate was removed by centrifugation and discarded. (NH4)2SOs was added to the super- natant to 658 saturation, and the precipitate was collected by cen- trifugation after 30 min. The precipitate was dissolved in a minimum volume of 5 mM buffer and desalted by dialysis against the same buffer overnight, and the insoluble materials were removed by cen- trifugation.

Nondiffusible solution was applied to a column (4 x 15 cm) of DEAE-cellulose equilibrated with 5 mM buffer. After washing with 500 ml of the same buffer, the enzyme was eluted with a 1000-ml linear gradient of 5-250 mM buffer. Fractions (20 ml) were collected a t a flow rate of 100 ml/h. The active fractions (fractions 35 to 42) were pooled and concentrated by ultrafiltration.

The concentrated enzyme solution was subjected to isoelectric focusing on a pH 3.0-10 Pharmalyte gradient, as described by Vester- berg and Swenson (21). Fractions (2 ml) were collected; activity and pH determinations were performed immediately. The focusing re- sulted in the detection of a single peak (PI 5.4) with y,S-dioxovalerate and glyoxyalte aminotransferase activities coincident. The active fractions were pooled, concentrated by ultrafiltration, and applied to a column of Sephacryl S-200 (2.5 X 100 cm), equilibrated with 50 mM buffer. The column was eluted with the same buffer a t a flow rate of 25 ml/h. The effluent was collected in 3-ml fractions. The active fractions (fractions 83 to 94) were pooled, concentrated by ultrafiltra- tion, and diluted with water, containing 10% glycerol, to adjust the buffer concentration to 5 mM.

The enzyme solution was applied to a hydroxylapatite column ( 2 X 5 cm) equilibrated with 5 mM buffer. After washing with 200 ml of the same buffer, the enzyme was eluted with a 500-ml linear gradient of 5-250 mM buffer. Fractions (20 ml) were collected at a flow rate of 25 ml/h. The active fractions (fractions 8 to 16) were collected and concentrated by ultrafiltration.

Purifieation of A1anine:gtyoxyEate Aminotransferase Isoenzymes 1 and 2 from Various Mammalian Liver and Kidney-Alanine: glyoxylate aminotransferase isoenzyme 1 was purified from rat, mouse, dog, cat (17) and human liver (22), and dog and cat kidney (23), and a1anine:glyoxylate aminotransferase isoenzyme 2 from rat and mouse liver (17) and rat, monkey, and pig kidney (23), as de- scribed in the cited references. The isoenzymes 2 of monkey, chicken, and pig liver and chicken kidney were purified as described for rat kidney (23).

Other Methods-Other methods were carried out as previously described (12-14, 17).

RESULTS AND DISCUSSION

A representative sedimentation in a sucrose density gra- dient for the homogenate from bovine liver is present in Fig. 1. The peroxisomes and mitochondria were separated; the peroxisomes, marked by catalase (EC 1.11.1.6), were a t a density of about 1.25 g / d , and the mitochondria, marked by glutamate dehydrogenase, a t a density of about 1.18 g/ml (Fig. 1). Acid phosphatase as a lysosomal marker was distrib- uted over a broad density range with a peak of about 1.19 g/ ml, and in the soluble top fraction from presumably broken lysosomes (not shown). Alanine:y,G-dioxovalerate aminotrans- ferase and a1anine:glyoxylate aminotransferase activities showed distribution profdes nearly identical with that of glu- tamate dehydrogenase but not with those of catalase and acid phosphatase activities; most of the homogenate activity of glutamate dehydrogenase and the two aminotransferase activ- ities was recovered in the mitochondrial fraction. These re- sults show that the two aminotransferases are associated with the mitochondria.

The mitochondrial extract was prepared after sucrose den- sity gradient centrifugation of the liver homogenate (see Fig. 1) and subjected to isoelectric focusing on a pH 3.0-10 Phar- malyte gradient. Alanine:y,S-dioxovalerate aminotransferase and a1anine:glyoxylate aminotransferase were identically fo-

cused, showing a single activity peak with PI 5.4 (Fig. 2). The activity ratio, glyoxylate aminotransferase/y,S-dioxovalerate aminotransferase, was about 60. These data suggest that the two aminotransferase activities are from the same protein.

Purification of alanine:y,&dioxovalerate aminotransferase and a1anine:glyoxylate aminotransferase from the mitochon- drial fraction of bovine liver was carried out to obtain further evidence on the identity of these enzymes, as described under “Experimental Procedures.” Results are summarized in Table I. Both enzyme activities were found in the same fractions in all purification steps. About 870-fold purification was achieved with a recovery of about 4.2% for both enzyme activities. The activity ratio, y,S-dioxovalerate aminotransferase/glyoxylate aminotransferase, remained constant during the purification. These activities were not increased by the addition of pyri- doxal 5”phosphate (0-40 ELM), showing the presence of the holoenzyme.

The purified enzyme (in 50 mM potassium phosphate buffer, pH 7.5, containing 10% glycerol) may be stored at -20 “C for a t least 8 weeks without loss of either activity. Little or none of each activity was lost when the enzyme was stored a t 0-5 “C for 2 weeks.

Heating the purified enzyme at 60 “C for different lengths of time produced equivalent losses of both activities (Fig. 3).

The two aminotransferase activities, each to the same de- gree, were inhibited by carbonyl reagents, namely isonicotinic acid hydrazide, hydroxylamine, potassium cyanide, and sem- icarbazide (Table 11); the inhibition is probably due to the

3oooc

FRACTION NUMBER

FIG. 1. Subcellular distribution of alanine:y,6-dioxovalerate aminotransferase and alanine:glyoxylate aminotransferase in bovine liver. The postnuclear supernatant was prepared from bovine liver and subjected to sucrose density gradient centrifugation, as described in the text. 0, alanine:y,S-dioxovalerate aminotransferase ( A VT ); 0, a1anine:glyoxylate aminotransferase (AGT); A, catalase; A, glutamate dehydrogenase.

1.0 - 20

. . p 0 . 5

I- 0 - 5 10

<

0- 0 20 40

FRACTION NUMBER

FIG. 2. Isoelectric focusing of the mitochondrial extract from bovine liver. The mitochondrial fractions (fractions 8 to 11) were pooled after sucrose density gradient centrifugation of bovine liver homogenate as described in Fig. 1, dialyzed against 5 mM potassium phosphate buffer, pH 7.5, sonicated at 10 kHz for 30 s and then centrifuged a t 105,000 X g for 60 min. 0, alanine:y,S-dioxovalerate aminotransferase ( A VT); 0, a1anine:glyoxylate aminotransferase (AGT); . . . ., pH value.

Alanine: y,S-dioxovalerate Aminotransferase in Mammals 10337

TABLE I Purification of alanine:y,d-dioxovalerate aminotransferase and

a1anine:glyoxylate aminotransferase from bovine liver Details of purification and assay methods are described in the text.

Alanine:y,G-dioxovalerate aminotransferase

w units Illiiii. .f& %

Postnuclear frac- 10,260 8.93 0.87 1 100 58.2

Mitochondrial 2,000 4.40 2.20 2.6 49.3 59.1

Heat treatment 354.0 3.89 11.0 12.6 43.6 57.3

DEAE-cellulose 21.1 1.13 53.6 61.6 12.7 61.3 Isoelectric focus- 1.2 0.78 652 749 8.7 60.8

tion

extract

(NHASO, 105.6 1.91 18.1 20.8 21.4 60.4

ing and Se- phacryl S-200

" A1anine:glyoxylate aminotransferase/alanine: y,G-dioxovalerate Hydroxylapatite 0.5 0.38 753 866 4.3 61.4

aminotransferase.

iS0

TIME (mln)

FIG. 3. Effect of heat denaturation of the enzyme. The puri- fied enzyme was maintained at 60 "C for different lengths of time before assays of alanine:y,G-dioxovalerate aminotransferase (A VT, 0) and a1anine:glyoxylate aminotransferase (ACT, 0). Details of assays are given in the text. a, specific activity of the untreated enzyme.

TABLE I1 Effects of carbonyl reagents on the enzyme

Details of assay methods are given in the text. Relative activity

r n M

Control Isonicotinic acid hydrazide 3

12 Semicarbazide 5

10 Potassium cyanide 5

10 Hydroxylamine 0.2

0.4

100 100 39.5 40.8 11.3 13.1 41.0 42.3 10.3 11.2 76.9 74.3 17.9 15.9 66.7 68.3 51.3 52.4

binding of the inhibitor with the aldehyde group of the coen- zyme, pyrifoxal 5"phosphate.

On polyacrylamide gel disc electrophoresis at pH 8.9 in 7% gel, the purified enzyme migrated toward the anode as a single protein band with both activities coincident (Fig. 4). Sedimen- tation in a sucrose density gradient centrifugation showed a single peak with both activities coincident (not shown).

The activity ratio, y,G-dioxovalerate aminotransferase/gly- oxylate aminotransferase, was unchanged after polyacryl- amide disc electrophoresis in 7% gel at pH 8.9, sucrose density gradient centrifugation, isoelectric focusing on a pH 3.0-10 Pharmalyte gradient and Sephacryl S-200 gel electrophoresis of the purified enzyme.

The molecular weight was estimated as 238,000 f 4,500 by Sephacryl S-200 gel filtration and as 245,000 f 5,600 by sucrose density gradient centrifugation; it had been estimated to be approximately 240,000 by gel filtration by Valticovski et al. (4). Polyacrylamide gel disc electrophoresis in the presence of sodium dodecyl sulfate yielded a molecular weight of 58,000 f 3,000, suggesting that the enzyme consists of four identical subunits.

The reverse transamination reaction between glycine (40 mM) and pyruvate (2 mM) as substrates was examined with the purified enzyme, as previously described (17). However, we cannot demonstrate the reverse reaction.

A double reciprocal plot of the initial velocity (L') against glyoxylate or y,G-dioxovalerate concentration at a series of fixed concentrations of alanine yielded a set of parallel lines, one for each concentration of alanine (Figs. 5-8); plots of 1/ y,d-dioxovalerate against l /u at varying y,S-dioxovalerate had been reported to yield a series of parallel lines.

These data are consistent with the ping-pong reaction

30 t i2

0 2 4 6 DISTANCE FROM ORIGIN (cm)

FIG. 4. Polyacrylamide gel disc electrophoresis of the en- zyme. The purified enzyme preparation (protein, 50 pg) was subjected to electrophoresis at pH 8.9 in 7V. gel. Of a pair of polyacrylamide gels, one was stained, and the other was sliced and assayed for enzyme activity. 0, alanine:y,G-dioxovalerate aminotransferase (A VT); 0, al- anine:glyoxylate aminotransferase (ACT).

0.12 -

0.08 - I - e - t

0.04 -

ALANINE 2mM ALANINE 3mM ALANINE 4mM

FIG. 5. Lineweaver-Burk double reciprocal plots of initial velocity of a1anine:glyoxylate aminotransferase activity against glyoxylate concentration at a series of fixed alanine concentrations. Experimental procedures are described in the text.

10338 Alanine: y,&dioxoualerate Aminotransferase in Mammals

mechanism characteristic transaminases. The K, values were obtained by the method of Velick and Vavra (25). From data of Figs. 5 and 7, K, values of 4.0 mM for alanine and 0.67 mM for glyoxylate were calculated. The data in Figs. 6 and 8 yielded a K , for alanine of 2.5 m~ and for y,6-dioxovalerate of

Alanine:y,G-dioxovalerate aminotransferase activity of the enzyme was competitively inhibited by glyoxylate (Fig. 9).

We have described that two forms of alanine:glyoxylate aminotransferase are present in mammalian liver and kidney. One, designated isoenzyme 1, has serine:pyruvate aminotrans- ferase activity; the other, designated isoenzyme 2, has not (17, 22, 23). Isoenzyme 1 is larger than isoenzyme 2 in the molec- ular weight. Both isoenzymes are present in the liver of rats, mice, dogs, and cats. Human liver contains only isoenzyme 1, and rat, pig, and monkey kidney only isoenzyme 2. Recently

0.3 mM.

8 - ALANINE 1mM

ALANJNE lOmM

0 I I 5 10

l/(DOVAXmM)-l

FIG. 6. Lineweaver-Burk double reciprocal plots of initial velocity of alanine:y,d-dioxovalerate aminotransferase activ- ity against y.8-dioxovalerate concentration at a series of fixed alanine concentrations. Experimental procedures are described in the text. DOVA, y,S-dioxovalerate.

- 0.041

ol 0.6 1.0

lI(ALANINE)(mM)-l

FIG. 7. Lineweaver-Burk double reciprocal plots of initial velocity of a1anine:glyoxylate aminotransferase activity against alanine concentration at a series of fixed glyoxylate concentrations. Experimental procedures are described in the text.

.c DOVA 0.1 mM

DOVA 0.2 mM

DOVA 0.4 n*l DOVA 0.3 mM

?

1 I 0.6 1.0 0

l/(ALANME)(mM)-l

FIG. 8. Lineweaver-Burk double reciprocal plots of initial velocity of a1anine:y.d-dioxovalerate aminotransferase activ- ity against alanine concentration at a series of fixed y,b-diox- ovalerate concentrations. DOVA, y.6-dioxovalerate.

7 1 /. QLYOXYLATE 1 mM

*u 10

lI(DOVA)(mM)"

FIG. 9. Lineweaver-Burk double reciprocal plots of initial velocity of a1anine:y.b-dioxovalerate aminotransferase activ- ity against y,d-dioxovalerate concentration in the presence and absence of glyoxylate. DOVA, y&dioxovalerate.

TABLE I11 Alanzne:y,S-dioxovalerate aminotransferase activity of alanine

glyoxylate aminotransferase isoenzymes from various mammalian liver and kidney

Procedures for purification of a1anine:glyoxylate aminotransferase isoenzymes 1 and 2 and assay methods are as described in the text. None of the isoenzyme 1 preparations (rat, mouse, dog, cat, and human liver and dog and cat kidney) had alanaine:y,S-dioxovalerate aminotransferase activitv.

A1anine:glyoxylate aminotransferase isoenzyme 2 Liver Kidney

Species Activity ratio" Species Activity

ratio"

Pig 90.7 Rat 27.2 Monkey 63.9 Monkey 29.1 Rat 80.6 Chicken 69.5 Bovine 61.4 Pig 56.7 Mouse 58.5 Chicken 61.8

Alanine: glyoxylate aminotransferase/alanine : y , S- dioxovalerate aminotransferase.

we found that pig, bovine, and chicken liver contain only isoenzyme 2, and monkey liver contains both isoenzymes.' In the present study, alanine:y,S-dioxovalerate aminotransferase activity of the two isoenzymes were determined with enzyme preparations from various mammalian liver and kidney. Re- sults are summarized in Table 111. All of isoenzyme 2 prepa- rations from livers (pig, monkey, rat, chicken, bovine, mouse) and kidneys (rat, monkey, chicken, and pig) had dioxovalerate aminotransferase activity. All of these enzyme preparations were much higher in the glyoxylate aminotransferase activity than in the y,6-dioxovalerate aminotransferase activity. In each case, the two activities were not separated by isoelectric focusing on a pH 3.0-10 Pharmalyte gradient, Sephacryl S- 200 gel filtration, sucrose density gradient centrifugation, and polyacrylamide gel disc electrophoresis in 7% gel a t pH 8.9. Furthermore the ratio of the two activities was unchanged after these treatments. In contrast, none of the isoenzyme 1 preparations (rat, mouse, dog, cat, and human liver and dog and cat kidney) had the y,S-dioxovalerate aminotransferase activity.

Recently, Varticovski et al. (14) have reported that bovine iiver mitochondria contains alanine:y,&dioxovalerate amino- transferase which catalyzes the formation of 6-aminolevulinic acid via a transamination, and the capacity of this enzyme to synthesize 6-aminolevulinic acid appears to be far greater than the capacity of 6-aminolevulinic acid synthase from the same source, suggesting the possibility that the y,&dioxovalerate aminotransferase plays a role in the biosynthesis of &amino-

' T. Noguchi and Y. Takada, unpublished observations.

AEanine:y,&dioxovalerate Aminotransferase in Mammals 10339

levulinic acid in vivo. They reported that the enzyme had high specificity for a pair of L-alanine and y,S-dioxovalerate; they did not examine glyoxylate as amino acceptor. In the present study, alanine:y,S-dioxovalerate aminotransferase and alanine:glyoxylate aminotransferase were co-purified from bo- vine liver mitochondria to homogeneity as judged by poly- acrylamide gel disc electrophoresis. The ratio of the two activities remained constant during purification and was un- changed by a variety of treatments of the purified enzyme. Alanine:y,S-dioxovalerate aminotransferase activity of the en- zyme was competitively inhibited by glyoxylate. These results show that bovine liver alanine:y,S-dioxovalerate aminotrans- ferase is identical with a1anine:glyoxlate aminotransferase. The enzyme is unusual in that it can catalyze both an a-keto acid-a-amino acid transaminase reaction and an w-keto acid- a-amino acid transminase reaction. Furthermore, the identity of the two enzymes was shown in livers and kidneys of various mammalian species; a1anine:glyoxylate aminotransferase is present only in liver and kidney of mammalian species (17). Varticovski et al. (4) have not been able to demonstrate reversibility of alanine y,S-dioxovalerate aminotransferase from bovine liver. The reaction of a1anine:glyoxylate amino- transferase lies so far toward glycine formation that we have not yet been able to demonstrate formation of alanine in reaction mixtures containing glycine and pyruvate with en- zyme preparations from dog, cat, rat, and mouse (17) liver and the present enzyme preparation from bovine liver.

Metzler et al. (24) have reported that transamination be- tween glyoxylate and many a-amino acids occurs nonenzy- matically, and the equilibrium strongly favors the conversion of glyoxylate to glycine. They have explained this unfavorable reversibility thermodynamically. The inability to detect re- versibility of a1anine:glyoxylate and alanine:y,S-dioxovalerate aminotransferase reactions of the present preparation might also be explained by this mechanism.

Studies on the presence of specific y,S-dioxovalerate ami- notransferase in mammals and roles of a1anine:glyoxylate aminotransferase in the formation of 6-aminolevulinic acid and tetrapyrroles in vivo are required.

Acknowledgment-We are grateful to K. Etoh for her skillful secretarial assistance.

REFERENCES 1. Granick, S., and Sassa, S. (1971) in Metabolic Regulation (Vogel,

H. J., ed) Vol. 5, pp. 77-141, Academic Press, New York 2. Marver, H. S., Collins, A., Tschudy, D. P., and Rechcigl, M., Jr.

(1966) J . Biol. Chem. 241,4323-4329 3. Bottomley, S. S., and Smithee, G. A. (1968) Biochim. Biophys.

4. Varticovski, L., Kushner, J. P., and Burnham, B. F. (1980) J. Biol.

5. Beale, S. I., Gough, S. P., and Granick, S . (1975) Proc. Natl. Acad.

6. Jurgenson, J. E., Beale, S. I., and Truxler, R. F. (1976) Biochem.

7. Kannangra, C. G., and Gough, S. P. (1978) Carlsberg Res. Com-

8. Beale, S . I., and Castelfranco, P. A. (1973) Biochem. Biophys.

9. Lohr, J . B., and Friedmann, H. C. (1976) Biochem. Biophys. Res.

10. Gassman, M., Pluscec, J., Bogorad, L. (1968) Plant Physiol. 43,

11. Neuberger, A., and Turner, J. M. (1963) Biochim. Biophys. Acta

12. Noguchi, T., and Takada, Y. (1978) J. Biol. Chem. 253,7598-7600 13. Noguchi, T., Takada, Y., and Fujiwara, S . (1979) J . Bzol. Chem.

14. Noguchi, T., and Takada, Y. (1980) J. Bzol. Chem. 255,9844-9847 15. Mauzerall, D., and Granick, S. (1956) J. Biol. Chem. 219,435-446 16. Rowsell, E. V., Snell, K., Carnie, J . A., and Rowsell, K. V. (1972)

Biochem. J. 127, 155-165 17. Noguchi, T., Okuno, E., Takada, Y., Minatogawa, Y., Okai, K.,

and Kido, R. (1978) Biochem. J . 169, 113-122 18. Beaufay, H., Bendall, B. S., and de Duve, C. (1959) Biochem. J.

73,623-628 19. Aebi, H. (1974) in Methods of Enzymatic Analysis (Bergmeyer,

H. U., ed) Vol. 2, pp. 673-684, Academic Press, New York 20. Bergmeyer, H. U. (1974) in Methods of Enzymatic Analysis

(Bergmeyer, H. U., ed) Vol. 1, pp. 417-522, Academic Press, New York

21. Vesterberg, O., and Swenson, H. (1966) Acta Chem. Scand. 20, 820-824

22. Noguchi, T., and Takada, Y. (1979) Arch. Biochem. Biophys. 196, 645-647

23. Takada, Y., and Noguchi, T. (1980) Comp. Biochem. Physiol. 65B, 133-138

24. Metzler, D. E., Olivard, J., and Snell, E. E. (1954) J. Am. Chem.

25. Velick, S. F., and Vavra, J . (1962) J. Biol. Chem. 237, 2109-2122

Acta 159,27-37

Chem. 255,3742-3747

Sci. U. S. A. 72, 2719-2723

Biophys. Res. Commun. 69, 149-157

mun. 43, 185-194

Res. Commun. 52, 143-149

Commun. 69,908-913

1411-1414

67,345-348

254, 5272-5275

SOC. 76,644-648