Ketopantoate Hydroxymethyltransferase - jbc.org · Ketopantoate’ hydroxymethyltransferase (5,10-methylene- tetrahydrofolate:a-ketoisovalerate hydroxymethyltransferase) * These studies

THE JOURNAL OE‘ BIOLOGICAL CHEMISTRY Vol. 251, No. 12, Issue of June 25, pp. 3786-3793, 1976

Printed in U.S.A.

Ketopantoate Hydroxymethyltransferase II. PHYSICAL, CATALYTIC, AND REGULATORY PROPERTIES*

(Received for publication, December 29, 1975)

SUE GLENN POWERS AND ESMOND E. SNELL

From the Department of Biochemistry, University of California, Berkeley, California 94720

Some physical, catalytic, and regulatory properties of ketopantoate hydroxymethyltransferase (5,lO-methylenetetrahydrofolate:a-ketoisovalerate hydroxymethyltransferase) from Escherichia coli are described. This enzyme catalyzes the reversible synthesis of ketopantoate (Reaction l), an essential

precursor of pantothenic acid. f

HC(CH,),COCOO- + 5,10-methylene tetrahydrofolate f HOCH,C(CH,),COCOO- + tetrahydrofolate (1)

It has a molecular weight by sedimentation equilibrium of 255,000, a sedimentation coefficient (.s~~,~) of 11 S, a partial specific volume of 0.74 ml/g, an isoelectric point of 4.4, and an absorbance, A%T%Y’, of 0.85. Polyacrylamide gel electrophoresis in sodium dodecyl sulfate and amino acid analyses give a subunit molecular weight of 27,000 and 25,700, respectively; both procedures indicate the presence of 10 identical subunits. The NH,-terminal sequence is Met-Tyr . . The enzyme is stable and active over a

broad pH range, with an optimum from 7.0 to 7.6. It requires Mg2+ for activity; Mn’+, Co’+, Zn*+ are progressively less active. The enzyme is not inactivated by borohydride reduction in the presence of excess substrates, i.e. it is a Class II aldolase. Reaction lf is partially inhibited by concentrations of formaldehyde (0.8 mM) and tetrahydrofolate (0.38 mM) below or near the K, values; apparent KM values are 0.18, 1.1, and 5.9 mM for tetrahydrofolate, a-ketoisovalerate, and formaldehyde, respectively. For Reaction lr, apparent K, values are 0.16 and 0.18 mM, respectively, for ketopantoate and tetrahydrofolate, and the saturation curves for both substrates show positive cooperativity. Forward and reverse reactions occur at similar maximum velocities (V N 8 pmol of ketopantoate formed or decomposed per max - min per mg of enzyme at 37”). Only I-tetrahydrofolate is active in Reaction 1; d-tetrahydrofolate, folate,

and methotrexate were neither active nor inhibitory. However, l-tetrahydrofolate was effectively replaced with conjugates containing 1 to 6 additional glutamate residues; of these, tetrahydropteroylpenta-, tetra-, and triglutamate were effective at lower concentrations than tetrahydrofolate itself; they were also the predominant conjugates of tetrahydrofolate present in E. coli. a-Ketobutyrate, Lu-ketovalerate, and cu-keto-P-methylvalerate replaced Lu-ketoisovalerate as substrates; pyruvate was inactive as a substrate, but like isovalerate, 3-methyl-2-butanone and D- or L-valine, inhibited Reaction 1. The transferase has

regulatory properties expected of an enzyme catalyzing the first committed step in a biosynthetic pathway. Pantoate (2 50 PM), pantothenate (z 500 ELM) and coenzyme A (above 1 mM) all inhibit; the V max is decreased, K, is increased, and the cooperativity for substrate (ketopantoate) is enhanced.

Catalytic activity of the transferase is thus regulated by the products of the reaction path of which it is one component; transferase synthesis is not repressed by growth in the presence of pantothenate.

Ketopantoate’ hydroxymethyltransferase (5,10-methylenetetrahydrofolate:a-ketoisovalerate hydroxymethyltransferase)

* These studies were supported in part by Grants AM 01448 and AI 01575 from the National Institutes of Health. The results have been abstracted from the Ph.D. dissertation of S. G. Powers (University of California, 1974) and were reported in part at the 1974 meeting of the American Society of Biological Chemists (1).

1 The terminology used is: ketopantoate for 2-keto-4-hydroxy-3,3- dimethylbutyric acid, pantoate for 2,4-dihydroxy-3,3-dimethylbutyric acid and dansyl for the 5-dimethylaminonaphthalene sulfonyl group; conjugates of tetrahydrofolate (tetrahydropteroylglutamate) are abbreviated as H,Pte(Glu),, where n = 1 to 7.

has been purified to homogeneity from Escherichia coli K12, and shown (2) to catalyze the reaction:

f oc-Ketoisovalerate + 5,10-methylenetetrahydrofolate 7

ketopantoate + tetrahydrofolate. (1)

This reaction is the first committed step of pantothenate biosynthesis, and is thus a likely control point for the pathway. The properties of the homogeneous enzyme were studied in order to learn more about its mode of action, the condensation reaction it catalyzes, and about the regulation of pantothenate biosynthesis.

3786

by guest on August 12, 2019

http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Properties of Ketopantoate Hydroxymethyltransferase 3787

EXPERIMENTAL PROCEDURE

Materials-Ketopantoate, pantoate, pantothenate, formaldehyde, and tetrahydrofolate were prepared as previously described (2). The folate conjugates, pteroyldiglutamate through pteroylheptaglutamate, and several 5methyltetrahydropteroylpolyglutamate standards labeled with tritium on the 3’, 5’ positions of the p-aminobenzoyl residue. were a generous gift of Dr. E. L. R. Stokstad. The preparative procedure of Blakley (3) was modified for conversion of small quantities of folate conjugates to their tetrahydro derivatives. PtO, (1 mg) in 0.4 ml of H,O was placed in the central reservoir of a 14.ml Warburg vessel, the folate derivate (about 0.5 rmol) in 0.5 ml of 0.1 M

NaHCO, was placed in the side arm, and the apparatus was evacuated and filled with hydrogen three times to reduce the catalyst. The folate derivative was then tipped into the main reservoir and the uptake of hydrogen followed manometrically. When reduction was complete, 0.1 ml of 1 M sodium ascorbate, pH 7, was added, the catalyst was filtered off, and the concentration of tetrahydropteroyl derivative was determined by assay with formyltetrahydrofolate synthetase (4) and from its absorbance at 298 nm (c = 25,000 M-I). The two determinations agreed well, indicating that the synthetase acted on each of the conjugates tested.

Molecular weight standards were obtained commercially as follows: n-amino acid oxidase, chymotrypsinogen, &galactosidase, leucine aminopeptidase, lysozyme, and yeast alcohol dehydrogenase (Worth- ington); enolase (Boehringer-Mannheim); pepsin and trypsin (Nutri- tional Biochemicals); thyroglobulin (Sigma); and catalase (Calbio- them).

Density gradient grade sucrose and p-amino[l-“Clbenzoic acid (52 mCi/mmol) were obtained from Schwarz/Mann; sequenation grade phenylisothiocyanate from Eastman; sequenal grade trifluoroacetic acid and dansyl chloride (10% w/v, in acetone) from Pierce Chemical Co.: and Cheng-Chin polyamide layer sheets from Gallard-Schle- singer. All other reagents were obtained commercially and used without further purification.

Preparation and Assay of Ketopantoate Hydroxymethyltransferase -This enzyme was purified and assayed as described in the accompanying paper (2).

Protein Determinations-For impure enzyme preparations, protein was determined by the method of Lowry et al. (5) with bovine serum albumin as standard. The concentration of homogeneous enzyme was calculated from its absorbance. A solution containing 1 mg/ml was found to have an absorbance of 0.85 at 280 nm by the specific refractive index increment method (6).

Ultracentrifugal Studies-Ultracentrifugal experiments were carried out in a Beckman-Spinco model E analytical ultracentrifuge equipped with a rotor temperature indicating unit. Double sector cells (12 mm) with quartz windows were used. The sedimentation velocity experiments were performed at 60,000 rpm, and the progress of the sedimenting boundary was followed at 280 nm using the photoelectric scanner at 4.min intervals. Sedimentation equilibrium experiments were performed for 22 hours at 12,000 rpm, and the boundary positions were again determined with the absorption optical system at 280 nm. The calculations were carried out as described by Schachman (7). Viscosities were measured with a 2-ml Ostwald viscometer, and densities with a pycnometer.

Density Gradient Centrifugation-The procedure was that of Mar- tin and Ames (8). The buffer was 50 IIIM potassium phosphate, pH 6.8, and the gradient was 5 ml of 5 to 20% sucrose, with 0.2 ml of 60% sucrose as a cushion. Standards were 0.4 mg of thyroglobulin (19.2 S) and 0.2 mg of catalase (13.3 S). The gradient was centrifuged at 4” and 65,000 rpm for 5 hours in the SW 65 rotor of a Beckman model L ultracentrifuge. Thyroglobulin was detected by its absorbance at 280 nm, and catalase by its absorbance at 280 and 405 nm. Ketopantoate hydroxymethyltransferase (10 pg; specific activity, 5.4) was co-sedimented and its position was detected by assay of individual fractions.

Gel Filtration Chromatography-A column of Bio-Gel A-5m (1 x 70 cm) was eluted at 4’ with pH 6.8 buffer containing 100 rn~ potassium phosphate, 1 rnrvr EDTA, and 0.5 IIIM dithiothreitol. Fractions (1.2 ml) were collected at a flow rate of 0.3 ml/min. Molecular weight standards of thyroglobulin (2 mg), catalase (1 mg), and P-galactosidase (0.5 mg) were assayed, respectively, by absorbance at 280 nm, absorbance at 405 nm, and enzymatic activity (9). Ketopantoate hydroxymethyltransferase (0.1 mg; specific activity, 6.0) was co-chromatographed and its position was determined by activity measurements. Blue dextran was added to mark the void volume.

Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis-The

polyacrylamide slab gel technique, the sodium dodecyl sulfate buffer system, and the staining procedure used were those described by Ames (10).

Isoelectric Focusing-Ketopantoate hydroxymethyltransferase (0.5 mg; specific activity, 0.8) was dialyzed overnight against 1% glycine before focusing at 4’ in a llO-ml LKB 8100 Ampholine electrofocusing column. The pH range was 3 to 6, the linear density gradient was 0 to 50% sucrose, and the electrofocusing solution was 1% in ampholytes. The voltages were 300 V for 8 hours, 450 V for 14 hours, and 600 V for 2 hours. Fractions (1 ml) of the pH gradient were collected and the pH was determined at 4”. Protein concentrations were determined from the absorbance at 280 nm, and the position of the enzyme by assaying for activity.

RESULTS

Physical Properties of Ketopantoate Hydroxymethyltransferase

Sedimentation Coefficient

The szo,w for the homogeneous enzyme in pH 6.8 buffer (0.1 M potassium phosphate, 1 mM EDTA, and 0.5 mM dithiothreitol) was determined to be 11 S by sedimentation velocity

experiments in the analytical ultracentrifuge and was constant at protein concentrations from 0.1 to 1.5 mg/ml. In a sucrose density gradient (8) the transferase sedimented slightly slower than catalase (.s~~.~ = 13.3 S), consistent with its behavior in

the analytical ultracentrifuge.

Molecular Weight of Native Transferase

Parallel sedimentation equilibrium experiments in H,O and D,O (11) with enzyme concentrations of 1.0, 0.4, and 0.2 mg/ml gave nearly identical values of 255,000 for the molecular weight and 0.74 ml/g for the partial specific volume. The plots of In y versus x2 were linear (Fig. IA), confirming the homogeneity of the enzyme sample, the molecular weight was also determined

, /

OL

nc

- Ch

,

02

B I I

-6

04 06 oa

Moblhty

FIG. 1. Molecular weight determinations of native (A) and denatured (B) ketopantoate hydroxymethyltransferase. A, native enzyme (0.4 mg/ml; specific activity, 5.2) in buffer containing 10 ins KPO,, 100 IIIM KCl, 1 IIIM EDTA, and 0.5 mvt dithiothreitol, pH 6.8, was centrifuged 22 hours at 22” and 12,000 rpm (see “Experimental Procedure”). z, the distance in centimeters from the axis of rota- tion, y, the concentration of protein. B, the same transferase preparation and several standard proteins were dissociated by boiling for 90 s in 1% sodium dodecyl sulfate/2.5% 2.mercaptoethanol. Samples containing 2.0 pg of protein were then applied to an 8% polyacrylamide gel containing 0.1% dodecyl sulfate and subjected to electrophoresis at pH 8.8. Subunit molecular weights of standard proteins are from Weber and Osborn (13). KPH, ketopantoate hydroxymethyltransferase; ADH, alcohol dehydrogenase.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

3788 Properties of Ketopantoate Hydroxymethyltransferase

by column chromatography on Bio-Gel A-5m (12); a higher but less accurate value of 285,000 was obtained.

Subunit Molecular Weight of Ketopantoate Hydroxymethyltransferase

On electrophoresis in sodium dodecyl sulfate-polyacrylamide gels the denatured enzyme migrated as a single band (see Fig. 3 in the accompanying paper (2)) implying that it is composed of subunits of identical or closely similar molecular weight. Comparisons with standard proteins (Fig. lls) gave a subunit molecular weight of 27,000, corresponding to approximately 10 subunits/molecule of native enzyme.

NH2-terminal Analysis

The NH, terminus and the penultimate residues of the transferase were determined by a modified dansyl-Edman degradation procedure (14). Dansyl methionine sulfone was the only a-dansylated amino acid to appear on chromatograms following the first cycle; O,N-bis(dansyl)tyrosine was the only a-substituted amino acid after the second cycle. The result establishes a unique sequence of NH2.Met-Tyr . . ., strongly indicating that the enzyme consists of identical subunits, and that no major protein contaminants are present.

Amino Acid Analysis

The amino acid composition of the transferase (Table I) provides excellent evidence for ten identical subunits, since the number of residues per 25,000 daltons are very close to integral values. The presence of only 1 residue each of tryptophan, methionine, and isoleucine indicates that the minimum molecular weight of the subunit is 25,000; any other permissible molecular weight must be a multiple of 25,000 which is

TABLE I

Amino acid composition of ketopantoate hydroxymethyltransferase The transferase (0.5 mg) was hydrolyzed for 20 hours at 110’ in 3 N

p-toluenesulfonic acid containing 0.2% 3-(2-aminoethyl)-indole (15). Since tryptophan is not destroyed under these conditions (15), all amino acids except half-cystine could be determined in this hydroly- sate. Half-cystine was determined as cysteic acid from a separate analysis in which 0.2 mg of enzyme was oxidized with performic acid, then hydrolyzed for 22 hours at 110” in 6 N HCl(16). Both hydrolysates were analyzed on a Beckman 120C amino acid analyzer (17).

Amino Mel per Nearest acid 25,000 g integer

Lysine 15.0 15 Histidine 4.05 4 Arginine 12.1 12 Tryptophan 1.18 1 Aspartic acid” 29.2 29 Threonine 17.0 17 Serine 10.9 11 Glutamic acid” 33.6 34 Proline 11.4 11 Glycine 30.9 31 Alanine 32.1 32 Cysteine 3.62 4 Valine 1.95 2 Methionine 1.06 1 Isoleucine 0.99 1 Leucine 23.1 23 Tyrosine 6.6 7 Phenylalanine 6.9 7

“These values include both free and amidated residues.

inconsistent with direct measurements. When the residue weights of the component amino acid residues are summed, the calculated molecular weight is 25,742. When the enzyme was denatured with 2% sodium dodecyl sulfate, 3.5 sulfhydryl residues/25,000 daltons were titrated with 5,5’-dithiobis(2- nitrobenzoic acid) (18). Thus there are no disulfide bonds in the protein.

Electron Microscopy

While the transferase is too small for electron microscopy to provide definitive evidence with respect to subunit structure or numbers, its micrographs (Fig. 2) are consistent with a lo-subunit structure in which the oligomer consists of two layers of pentamers. A top view of this structure (solid arrow) resembles a pentamer with 5-fold symmetry, while a tetram- eric structure (open arrow) can be interpreted as a side view showing two subunits from each layer.

Isoelectric Point and Stability to pH

Upon isoelectric focusing (19), only 25% of the applied transferase activity was recovered, but the isoelectric pH value was determined to be 4.4. Separate storage experiments showed that the enzyme was stable between pH 5.0 and 10, but was rapidly inactivated below pH 4.5.

Catalytic Properties of Ketopantoate Hydroxymethyltransferase

Effect of Temperature and pH

The rate of the enzymatic conversion of ketopantoate to oc-ketoisovalerate (Reaction lr) increased with temperature to 70°, remained approximately constant between 70 and 80”, then decreased rapidly above 80” as the enzyme became denatured. An Arrhenius plot of the data between 20 and 70’ was linear and gave an activation energy of 6.05 kcal/mol. The pH activity curve displays a broad maximum at pH 7.0 to 7.6 with half-maxima at pH 6 and pH 9; the enzyme is inactive below pH 5. Within the pH range studied, there is no nonenzymic formation of formaldehyde from ketopantoate during the assay period; however, above pH 10 the nonenzy- mat,ic reaction becomes rapid.

FIG. 2. Electron micrograph of homogeneous ketopantoate hydroxymethyltransferase after negative staining with potassium phos- photungstate, pH 4.8. The magnification is 512,500-fold.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


Role of Metal Ions Kinetic Parameters for Conversion of Ketopantoate to

Activity of the transferase is dependent on the presence of a divalent metal ion. In crude preparations, its activity is reduced to about 15% of maximum by dialysis for 3 hours at 4” against 10 mM EDTA/0.25 M Tris, pH 9.0, and activity could be restored to initial levels by addition of Mg2+. All of the buffers used during purification of the enzyme contained EDTA, and as purification progresses, the activity without added Mg2+ gradually decreases; the homogeneous enzyme is only 4 to 10% as active without as it is with added Mg’+. The activating effect of various divalent ions for the latter preparation is shown in Fig. 3. Mg*+ (0.1 mM) is most active followed by MnZ+, Ni*+,

Co*+ and Zn2+. , Cuz+ and Fez+ were ineffective.

When the transferase was treated with 10 mM sodium

Formaldehyde and ol-Ketoisovalerate (Reaction lr)

borohydride in the presence of a 1000.fold molar excess of a-ketoisovalerate with or without excess tetrahydrofolate, no inactivation occurred. Apparently, therefore, no Schiff base formation occurs between enzyme and substrate; and the transferase is thus a Class II aldolase (20).

For Ketopantoate and Tetrahydrofolate-Plots of the dependence of transferase activity on ketopantoate concentration (Fig. 5A) indicate positive cooperativity. The ratio of ligand concentrations required for 90% to that required for 10% of the

maximum velocity (i.e. [S],.,/[S],., = R,) is 15, whereas this ratio is 81 for curves following Michaelis-Menten kinetics (21). The Lineweaver-Burk plot (Fig. 5B) curves upward, and the Hill plot (not shown) is linear with a slope of 1.5. The concentration of ketopantoate needed for half-saturation (S,.,) is 0.16 mM. Similar plots showed that tetrahydrofolate also binds cooperatively to the transferase. In this case, R, is 16, and the Hill coefficient is 1.9. The S,, for dl-tetrahydrofolate is 0.18 mM.

Kinetic Parameters for Synthesis of Ketopantoate (Reaction If)

Both free formaldehyde (above 0.8 mM, Fig. 4) and tetrahydrofolate (above 0.38 mM; inset, Fig. 4) inhibit transferase activity, and the inhibition by formaldehyde could not be overcome by adding excess tetrahydrofolate. Thus, saturating levels of methylene tetrahydrofolate (i.e. formaldehyde plus tetrahydrofolate) could not be used in the assay mixture. Kinetic parameters for ketopantoate formation are therefore uncertain and it could not be determined whether a-ketoisovalerate and tetrahydrofolate bind cooperatively to the transferase. Apparent K, values from Fig. 4 are 5.9 mM for formaldehyde, 1.1 mM for oc-ketoisovalerate, and 0.16 mM for tetrahydrofolate. Separate determinations with homogeneous transferase in which tetrahydrofolate (0.5 mM) and HCHO (0.5

mM) were at constant initial concentrations and Lu-ketoisovalerate was varied gave an apparent maximum velocity for ketopantoate formation of 8.25 pmol min-’ mg-‘, similar to that observed for breakdown of ketopantoate to HCHO and cu-ketoisovalerate (V,,,,, = 7.3 pmol minm’ mgm’; see following

section). Both values are considered minimum estimates, and the extrapolated value for the V,,,,, of ketopantoate formation where HCHO is the varied substrate (Fig. 4) gives a much higher value of 66 pmol min-’ mg- I.

For Conjugates of Tetrahydrofolate-The major forms of folate that occur naturally contain from 2 to 7 glutamate residues (22). Synthetic samples of pteroyldiglutamate through pteroylheptaglutamate (23) were reduced to their tetrahydro derivatives (see “Experimental Procedure”) and tested for activity (Table II). The maximum velocity of the reaction was not affected by the extent of conjugation but the

3 I I I I I 06

= /' /. I

-05 0

6 I

-04 ; .-

T

. -03 TF a- KV

I .’ I -\’ ,:’

0 / I [HqPteGlu]-‘, mM-’

I 1 I 2 3 4 5

[a-Ketoisovolerote]-‘or [HCHO]-‘, TIM-’

FIG. 4. Ketopantoate synthesis as a function of formaldehyde, ol-ketoisovalerate (a-KV) and tetrahydrofolate (H,PteGlu) concentrations. Standard concentrations (0.5 rnM HCHO, 0.5 mM tetrahydrofolate, and 5 rnM cu-ketoisovalerate) of substrates were used except for the one being varied. The amount of enzyme per reaction mixture (1.0 ml) varied as follows: for HCHO, 2.5 Mg (specific activity, 4.8); for cu-ketoisovalerate, 3 pg (specific activity, 6.2); and for tetrahydrofolate, 8 fig (specific activity, 3.5).

FIG. 3. Comparative effects of divalent metal ions as activators for ketopantoate hydroxymethyltransferase. The transferase (2.3 pg; specific activity, 4.4 after addition of optimal Mg*+) was assayed by following Reaction lr for 20 min after addition of the indicated metal .

[Ketopantoate], mM

FIG. 5. Dependence of transferase activity in Reaction lr on ketopantoate concentration. A, formaldehyde formed after 30 min in the standard assay (transferase, 1 pg; specific activity, 4.8) with varying

Ions (as chlorides). amounts of ketopantoate. B, double reciprocal plot of the data m A.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


TABLE II

Affinities of ketopantoate hydrorymethyltransferase for tetrahydrofolate and its conjugates

Activity of the transferase (1 pg; specific activity, 4.8) was determined by following HCHO formation (see “Experimental Procedure”) from ketopantoate (7 mM) after 30 min in the presence of varying amounts of each conjugate (cf. Fig. 4).

Compound0 K”, S 05

(rnM) (mM14 H,PteGlu 0.33 0.18 H,PteGlu, 0.25 0.20 H,PteGlu, 0.18 0.10 H,PteGlu, 0.10 0.10 H,PteGlu, 0.10 0.10 H,PteGlu, 0.17 0.10 H,PteGlu, 0.29 0.20

“Tetrahydrofolate and its conjugates are abbreviated as H,PteGlu, , where n is the number of glutamic acid residues. The observed V,,,., was 6.7 pmol of formaldehyde formed/min/mg of enzyme in all cases.

* so.5 was calculated by assuming (since V,,,,, obtained from Lineweaver-Burk plots was the same for all of the conjugates) that the substrate saturation curves would all approach the same maximum activity as that obtained with tetrahydrofolate itself.

tetra- and pentaglutamates exhibited the lowest K, for the enzyme. However, these reduced conjugates, like tetrahydrofolate itself, show cooperative binding to the transferase, and Lineweaver-Burk plots are linear only because the small

amount of conjugates available permitted us to obtain data points only at low substrate concentrations. A more appropri- ate parameter is the concentration of ligand needed for half-saturation (S,.,). The calculated values of S,, (Table II) indicate that the most efficient conjugates are H,PteGlu, through H,PteGlu,.

Predominant Tetrahydrofolate Conjugates in Escherichia coli-It would be of interest to determine whether those tetrahydrofolate conjugates that showed the lowest K, values for the transferase were also those that predominated in E. coli. For this purpose we first grew E. coli K12 with various levels of p-amino[“C]benzoate to label the tetrahydrofolate derivatives, extracted washed cells with reducing buffer, and

passed the extract over Sephadex G-25 (24). This procedure in our hands did not separate the conjugates sufficiently to

permit their individual identification; however, the results clearly showed that growth with 0.02 and 0.2 pM p-amino- benzoate gave an identical size distribution of conjugates, while growth with higher concentrations of precursor altered this distribution toward the lower molecular weight conjugates. For further studies, we therefore grew the organism with

0.2 FM p-amino[“C]benzoate, and adopted the modified extraction procedure of Buehring et al. (25), in which the con-

jugates are converted to their more stable 5-methyl derivatives before separation. The derivatized extract, as well as

tritiated standards, was chromatographed over both Sephadex G-25 (Fig. 6) and DEAE-cellulose (data not shown). In both cases, the major conjugate found in the extract was 5-methyl- tetrahydropteroylpentaglutamate, followed in order of decreas- ing concentrations by the tetra-, tri-, and hexaglutamate. Thus the corresponding unmethylated conjugates are the forms that normally occur in E. coli. As shown previously (Table II) they are also the forms with greatest affinity for ketopantoate hydroxymethyltransferase.

Fraction

FIG. 6. Sephadex G-25 column chromatography of the &methyltet- rahydrofolate conjugates present in derivatized extracts from Esche- richia coli K-12. Cells were grown with aeration at 37” for 18 hours in 100 ml of M-9 medium (26) supplemented with 0.2 rmol ofp-amino[l- “C]benzoate/liter. Cells were harvested by centrifuging, washed three times in M-9 medium, then extracted by the procedure of Buehring et al. (25) which converts the tetrahydrofolate derivatives to their 5-methyl derivatives. The *‘C-labeled extract (2 ml) and tritium- labeled 5methyltetrahydropteroylpolyglutamate standards were placed on a Sephadex G-25 column (0.75 x 200 cm) and eluted with 0.1 M KP0,/0.2 M mercaptoethanol buffer, pH 7.0. Fractions (1.65 ml) were collected at a flow rate of 14 ml/hour and a O.&ml portion of each fraction was counted. The arrows indicate the peak positions of folate and the polyglutamate standards; glu, = 5.methyltetrahydropteroyl- hexaglutamate; glu, = 5-methyltetrahydropteroyltriglutamate.

Substrate Specificity of Ketopantoate Hydroxymethyltransferase

Except for the activity of tetrahydrofolate conjugates just

discussed, the transferase showed an absolute requirement for tetrahydrofolate under the assay conditions used. Neither folate nor methotrexate, at concentrations up to 200 times that

of tetrahydrofolate, inhibited the enzyme. To determine the configurational specificity for tetrahydrofolate, the synthetic dl-compound was resolved via its 5,10-methylene derivative (27) as described by Kaufman et al. (28). Only the 1 isomer was active, and the d isomer did not inhibit at concentrations equimolar with the 1 isomer.

The specificity toward a-ketoisovalerate (Table III) was less rigid. cr-Ketobutyrate, a-keto-fl-methylvalerate, and cY-keto-

valerate all served as formaldehyde acceptors, each giving rise to a different product, which although not further characterized, are presumed to be the corresponding hydroxymethylated ketoacids. Isovalerate, D- and L-valine, 3-methyl-2-butanone and pyruvate are inactive as substrates, but inhibit ketopanto-

ate formation from a-ketoisovalerate (Table IV).

Regulatory Properties of a-Ketopantoate Hydroxymethyltransferase

Apparent Absence of Repression-Induction Effects

The transferase level in cells of E. coli W was not decreased

by growth in the presence of pantothenate (2). In more extensive tests, E. coli K12 was grown in nutrient broth or in Vogel-Bonner citrate minimal medium (31) supplemented with 0.5% glucose or with 0.5% glucose plus pantothenate ranging from 0.2 PM to 50 mM. The levels of transferase remained constant under these conditions, indicating that

enzyme levels are not repressed by excess pantothenate or any of the components of nutrient broth.

The transferase level also was not increased by growth under


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


TABLE III

Specificity of ketopantoate hydrozymethyltransferase for its keto acid substrates

Kinetic data were obtained by following incorporation of [%]HCHO into product by the usual assay for ketopantoate formation (see “Experimental Procedure”), but using the ketoacid substrates listed. Formation of separate products from each substrate was further verified by incubating transferase (25 fig; specific activity, 3.5), ketoacid (5 mM), [“CIHCHO (0.5 mM), tetrahydrofolate (0.5 mM), and MgSO, (1 mM) for 3 hours at 37” and pH 6.8. The reaction mixture (1.0 ml) was evaporated to about 0.05 ml, then subjected to paper electrophoresis (Whatman No. 1 paper) for 45 min at 50 mA and 1500 V in pyridine/acetic acid/acetone/H,O, pH 4.2 (29). Substrates were detected with bromcresol green (30) and the products by radioautogra- phy. Control radioautograms of reaction mixtures (enzyme omitted) showed only the spot near the origin due to [YZIHCHO.

Substrate

a-Ketoisovalerate a-Ketobutyrate a-Ketovalerate a-Keto-P-methyl

valerate

Kinetic Electrophoretic parameters migration

K”, nlai V Substrate “C-product

pm01 rnM min-1 cm cm

mg-’

1.0 2.8 10.0 8.5 2.9 4.0 10.5 9.5

25. 2.0 8.5 7.0 5.9 0.43 9.0 1.5

TABLE IV

Inhibition of ketopantoate formation from a-ketoisoualerate by related compounds

The standard reaction mixture (see “Experimental Procedure”) for the synthetic direction (Reaction lfl was used with 8 pg of transferase (specific activity, 6.2). The concentration of a-ketoisovalerate was 5 mM; in the absence of inhibitors 0.38 pmol of ketopantoate were formed during the 10.min incubation period.

Compound added (5 mM) % inhibition

None 0 Isovalerate 39 Pyruvate 38 3.Methyl-2.butanone 27 L-Valine 23 n-Valine 16

conditions that limit availability of pantothenate since: (a) a pantothenate auxotroph, Salmonella typhimurium pan C, blocked in pantothenate synthesis at a point subsequent to

ketopantoate formation and cultured on growth-limiting concentrations (0.013 PM) of pantothenate, contained transferase levels essentially identical with the parent strain; and (b) the level of this enzyme in E. coli K12 was not increased during growth on a concentration of D-serine (2 mM) that partially inhibited growth. Cosloy and McFall (32) showed that growth inhibition under the latter conditions resulted from inhibition of pantothenate biosynthesis.

Inhibition of Ketopantoate Hydroxymethyltransferase by End Products of the Biosynthetic Pathway

Inhibition of Reaction lr by Pantothenate and Pan- to&e-The transferase-catalyzed breakdown of ketopantoate was inhibited by both pantothenate (Fig. 7A) and pantoate (Fig. 7B). With 10 mM pantothenate, the cooperativity of the ketopantoate saturation curve is greatly enhanced, the V,,, is

50FGJ 1 B Parmate m4

[Ketopantoote], mu

FIG. 7. Inhibition by pantothenate (A) or pantoate (B) of the conversion of ketopantoate to cu-ketoisovalerate and formaldehyde (Reaction lr) by ketopantoate hydroxymethyltransferase. Concentra- tions of pantothenate or pantoate are given on each curve. Reaction mixtures contained transferase (5.7 pg; specific activity, 4.8), ketopantoate and inhibitors as indicated, tetrahydrofolate (0.5 rmol), and MgSO, (1.0 rmol) in 1.0 ml of 0.1 M potassium phosphate buffer, pH 6.8, and were incubated 10 min at 37” before formaldehyde was determined.

decreased, and the K, is increased. Similar but less marked effects are observed with 2.5 and 5.0 mM pantothenate, and the saturation curves exhibit intermediary plateau regions. Ki- netic analyses (33) indicate that for such plateaus to be

produced, an enzyme must have more than two substrate- binding sites, and the relative magnitude of either the catalytic or the binding constants of these sites must first decrease and then increase as the substrate concentration is increased. That is, the enzyme must show negative, then positive cooperativity in the plateau region (33). The enhancement of positive cooperativity by pantothenate can be seen more clearly in Hill plots (not shown) of these same data: the Hill coefficient for the low substrate portion of the saturation curve increases from

1.5 to 3.0 as the pantothenate concentration increases. Panto- ate (Fig. 7B) is considerably more effective as an inhibitor than pantothenate but its effects otherwise are similar, i.e. V,,, is decreased and K, is increased. However, the plateau regions are less well defined, and the enhancement of cooperativity is less pronounced than with pantothenate.

Because a-ketobutyrate was such a good substrate for the transferase (Table III), the extent of pantoate and pantothenate inhibition of its utilization was studied to determine whether it would provide a possible mechanism for discrimi- nating against this substrate in uiuo. However, the transferase- catalyzed reactions with a-ketobutyrate and a-ketoisovalerate

displayed almost identical inhibition patterns. One step in the purification of the transferase involves

treatment with 4 M urea. Since such harsh treatment might alter the allosteric properties of the protein, a modified preparative procedure that did not use urea was devised to

partially purify the transferase. The ketopantoate saturation curve for this preparation in the absence and presence of

pantothenate was nearly identical with that shown in Fig. 7. Inhibition of Ketopantoate Synthesis (Reaction If) by Pan-

toate, Pantothenate, and Coenzyme A-The enzymatic synthesis of ketopantoate was inhibited by the same levels of pantoate and pantothenate, 50 and 500 MM, respectively, that inhibited the breakdown of ketopantoate. Although complete

saturation curves could not be obtained because of the previously noted inhibition by formaldehyde and tetrahydrofolate, the inhibition patterns (Fig. 8) are qualitatively similar to


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


0 I mM pantoate

FIG. 8. Inhibition by pantoate of the transferase-catalyzed formation of ketopantoate. The reaction mixtures contained transferase (3 pg; specific activity, 6.2), [I’CIHCHO (0.5 rmol), tetrahydrofolate (0.5 wmol), ol-ketoisovalerate and pantoate as indicated, and M&O, (1.0 nmol) in 1.0 ml of 0.1 M KPO,; pH 6.8. [“ClKetopantoate formed was assayed (2) after 10 min incubation at 37”.

those for the breakdown reaction. Coenzyme A at concentrations above 1 mM also inhibited transferase activity.

DISCUSSION

Like all other tetrahydrofolate-requiring enzymes studied to date, ketopantoate hydroxymethyltransferase requires the 1 isomer of tetrahydrofolate, and does not utilize the d form. All of the naturally occurring conjugates of tetrahydrofolate WSte(Glu),, where n = 1 to 7) were effective, but it is of interest that those with highest affinity for the enzyme were the same as those identified here as predominant in E. coli, i.e. those where n = 5, 4, and 3. This assignment of the distribution of tetrahydrofolate conjugates in E. coli differs from those previously reported. Viswanathan et al. (34) reported that only tetrahydropteroyl mono-, di-, and triglutamates were present. However, they failed to include reducing agents at all stages of the extraction procedure. Under these conditions, reduced folates are partially degraded and then appear at positions on DEAE-cellulose chromatography corresponding to reduced conjugates containing fewer glutamic acid residues (25). Kozl- off and Lute (35) reported that tetrahydropteroyl triglutamate was the major form in E. coli but that conjugates containing up to 6 glutamate residues were also present. The technique employed, however, did not give clean separations, so that some ambiguity remained. Furthermore, to introduce suffi- cient label into the conjugates, the bacteria were grown with comparatively high concentrations (14 PM) of [“Clp- aminobenzoic acid. In Lactobacillus casei and Streptococcus fuecalis, growth with large amounts of folic acid results in formation of polyglutamates with fewer glutamic acid residues than those normally found (25), and we have reported here that high concentration of p-aminobenzoic acid (>0.2 pM) similarly perturb the normal distribution of folate conjugates in E. coli. The improved conditions adopted herein for growth of cells, their extraction, and separation of the tetrahydrofolate conjugates provide the most reliable estimate currently available concerning the distribution of these conjugates in E. coli.

Ketopantoate hydroxymethyltransferase does not show absolute specificity for the ketoacid substrate. However, substrate activity does require a carboxyl group and an cY-carbonyl function, and there are also constraints on the hydrophobic

portion of the ketoacid: although one methyl group of cy- ketoisovalerate can be replaced by CJI, (cY-keto-&methylvalerate) or H (a-ketobutyrate), and the remaining methyl group in cu-ketobutyrate can be extended by one carbon atom (a-ketovalerate), both methyl groups cannot be replaced by H (pyruvate). Lu-Ketoisovalerate analogs in which one of these three structural requirements is missing (3-methyl-2-butanone, isovalerate, pyruvate) are inhibitors of the enzyme reaction, as is valine, the corresponding amino acid.

The available data indicate that regulation of the transferase is accomplished by end product control rather than by repression or induction. Because of the positive cooperativity for Reaction lr, ketopantoate would accelerate its own breakdown were it to accumulate. More importantly, the transferase is inhibited by pantoate at concentrations greater than 50 FM, by pantothenate above 500 PM, and by coenzyme A at concentrations greater than 1 mM, and this inhibition is characterized by both a reduced maximum velocity of synthesis and an increased K,. The pantoate inhibition seems to occur at a sufficiently low concentration to be physiologically significant. Maas (36) observed that the in vitro rate of pantothenate production by the pantoate$-alanine ligase of E. coli is 100 times faster than the maximal in uiuo rate. Since the ligase has a K, for pantoate of approximately 2 mM, and the transferase studied here is inhibited by s/40 of this concentration, the in vivo rate of pantothenate synthesis may well be maintained at a low level by control of the transferase activity. Several aspects of this control need further investigation, however. For example, formaldehyde and tetrahydrofolate have been used to generate the methylenetetrahydrofolate required as substrate in Reac- tion lf. Both compounds inhibit at concentrations near their K, values, and this complicates assessment of the significance of inhibition by pantoate. It would be interesting to investigate kinetics of the reaction at the steady state concentrations (now unknown) of methylenetetrahydrofolate that occur in cells. A somewhat similar problem is presented by the fact that a-ketobutyrate is almost as good a substrate for the transferase as a-ketoisovalerate. It is not clear how, or whether, the former substrate is discriminated against in ho.

REFERENCES

1. Powers, S. G., and Snell, E. E. (1974) Fed. Proc. 33, 871 Abstr. 2. Teller. J. H.. Powers. S. G.. and Snell. E. E. (1976) J. Biol. Chem.

251;3780-3785 3. 4.

Blakley, R. L. (1957) Biochem. J. 65,331-342 Rabinowitz, J. C., and Pricer, W. E., Jr. (1962) J. Biol. Chem. 237,

2898-2902 5. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J.

6. (1951) J. Biol. Chem. 193, 265-275

Richards, E. G., Teller, D. C., and Schachman, H. K. (1968) Biochemistry 7, 1054-1076

7. Schachman, H. K. (1957) Methods Enzymol. 4, 32-103 8. Martin, R. G., and Ames, B. N. (1961) J. Biol. Chem. 236.

1372-1379 9. Wallenfels, K. (1962) Methods Enzymol. 5, 212-213

10. Ames, G. F.-L. (1974) J. Biol. Chem. 249,634-644 11. Edelstein, S. J., and Schachman, H. K. (1967) J. Biol. Chem. 242,

306-311 12. Piez, K. A. (1968) Anal. Biochem. 26, 305-312 13. Weber, K., and Osborn, M. (1969) J. Biol. Chem. 244, 4406-4412 14. Weiner, A. M., Platt, T., and Weber, K. (1972) J. Biol. Chem. 247,

3242-3251 15. Liu, T.-Y. (1972) Methods Enzymol. 25, 44-55 16. Hirs, C. H. W. (1967) Methods Enzymol. 11, 59-62 17. Spackman, D. H. (1967) Mehods Enzymol. 11, 3-15 18. Habeeb, A. F. S. A. (1972) Methods Enzymol. 25, 457-464 19. Vesterberg, 0. (1971) Methods Enzymol. 22, 389-412 20. Rutter, W. J. (1964) Fed. Proc. 23, 1248-1257


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


21. Levitzki, A., and Koshland, D. E., Jr. (1969) Proc. N&l. Acad. Sci. U. S. A. 62, 1121-1128

22. Baugh, C. M., and Krumdieck, C. L. (1971) Ann. N. Y. Acad. Sci. 186, 7-28

23. Baugh, C. M., Stevens, J. C., and Krumdieck, C. L. (1970) B&him. Biophys. Acta 212, 116-125

24. Shin, Y. S., Buehring, K. U., and Stokstad, E. L. R. (1972) J. Biol. Chem. 247, 7266-7269

25. Buehring, K. U., Tamura, T., and Stokstad, E. L. R. (1974) J. Biol. Chem. 249, 1081-1089

26. Adams, M. H. (1959) Bacteriophages 446, Interscience Publish- ers, Inc., New York

27. Osborn, M. J., Talbot, P. T., and Huennekens, F. M. (1960) J. Am.

Chem. Sot. 82, 4921-4927 28. Kaufman, B. T., Donaldson, K. O., and Keresztesy, J. C. (1963) J.

Biol. Chem. 238,1498-1500 29. Crestfield, A. M., and Allen, F. W. (1955) Anal. Chem. 27,422-423 30. Brown, F., and Hall, L. P. (1950) Nature 166, 66-67 31. Vogel, H. J., and Bonner, D. M. (1956) J. Biol. Chem. 218,97-106 32. Cosloy, S. D., and McFall, E. (1973) J. Bacterial. 114, 685-694 33. Teipel, J., and Koshland, D. E., Jr. (1969) Biochemistry 8,

4656-4663 34. Viswanathan, G., Amin, P. M., and Noronha, J. M. (1970) Indian

J. Biochem. 7, 226-230 35. Kozloff, L. M., and Lute, M. (1973) J. Viral. 11, 630-636 36. Maas, W. K. (1952) J. Biol. Chem. 198,23-32


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

S G Powers and E E Snellproperties.

Ketopantoate hydroxymethyltransferase. II. Physical, catalytic, and regulatory

1976, 251:3786-3793.J. Biol. Chem.

http://www.jbc.org/content/251/12/3786Access 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/251/12/3786.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/251/12/3786

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;251/12/3786&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/251/12/3786

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=251/12/3786&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/251/12/3786

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

http://www.jbc.org/content/251/12/3786.full.html#ref-list-1

http://www.jbc.org/

Ketopantoate Hydroxymethyltransferase - jbc.org · Ketopantoate’ hydroxymethyltransferase (5,10-methylene- tetrahydrofolate:a-ketoisovalerate hydroxymethyltransferase) * These studies

Documents