THE JOURNAL OB BIOLOGICAL CHEMISTRY Vol. 248, No. 24, Issue of December 25, PP. 8534-8540, 1973

Printed in U.S.A.

Concerted Feedback Inhibition

PURIFICATION AND SOME PROPERTIES OF ASPARTOKINASE FROM PSEUDOMONAS FLUORESCENS*

(Received for publication, December 1, 1972)

STEPHEN M. DUNGAN$ AND PRASANTA DATTA~

From the Department of Biological Chemistry, University of Michigan, Ann Arbor, Michigan 48104

SUMMARY

Aspartokinase of Pseudomonas fluorescens has been purified by heat treatment, ammonium sulfate fractionation, DEAE-cellulose chromatography, and by successive gel filtra- tion steps on Sephadex G-ZOO in the presence and absence of feedback modifiers. The 700-fold purified enzyme was judged to be greater than 90% pure. Sedimentation velocity centrifugation yielded an a~,,+, value of 6.7 S, a Dw,.~ value of 5.3 x lo-' cm2 s-l, and a molecular weight of 133,000. The Stokes radius of the native enzyme in 50 mu potassium phosphate buffer, pH 7.5, containing 200 mu KC1 was found to be about 43 A. A diffusion coefficient of 5.2 x lo-’ cm2 s-l was calculated from the Stokes-Einstein equation. A molecular weight of 137,000 was also estimated from the gel filtration data. The molecular weight of the subunit was 43,000.

The enzyme had an absolute requirement for ATP and a divalent cation. No other nucleoside phosphates served as the phosphate donor. Addition of 200 mu KC1 increased the reaction rate by about 40 % ; sodium and lithium ions had no effect.

The activity of this enzyme was inhibited by concerted feedback by two pairs of amino acid end products: lysine plus threonine and methionine plus threonine. Individually, lysine and methionine were slightly stimulatory ;’ threonine showed a slight activation at low concentrations, but was weakly inhibitory at concentrations over 10 mm Concerted inhibition by low threonine and high lysine combination was compensated by methionine; whereas, methionine did not counteract when enzyme activity was severely decreased by high concentrations of threonine plus lysine. A possible physiological significance of the dual role of methionine is discussed.

* This research was supported by Grant GB 29016X from the National Science Foundation, and was taken from a dissertation submitted by S. M. D. to the Graduate School of the University of Michigan in partial fulfillment of the requirements for the degree of Doctor of Philosophy.

$ Predoctoral Trainee of the United States Public Health Ser- vice, Grant GM 00187. Present address, Department of Biochem- istry, University of Miami School of Medicine, Miami, Fla. 33152.

5 To whom inquiries should be addressed.

Aspartokinase (ATP:~aspartate 4-phosphotransferase EC 2.7.2.4) is the first enzyme of the branched biosynthetic path- way leading to the synthesis of four amino acids of the aspartate family: lysine, methionine, threonine, and isoleucine (1). The enzyme catalyzes the phosphorylation of L-aspartic acid by ATP to produce aspartyl P-phosphate (2). Recent studies have revealed the existence of multiple metabolic feedback patterns of this key enzyme in a variety of microorganisms (3). In Escherichia coli and Salmonella typhimurium, isoenzymes of aspartokinase, each regulated by one metabolite, appear to be the principal mode of control (4, 5) ; in other organisms such as Rhodopseudomonas capsulata and in several species of Bacillus, concerted action of two metabolites controls the activity of a single enzyme (6-8). In preliminary experiments (3, 9) we found that the enzyme activity in crude extracts from Pseudom- onas jluorescens was severely inhibited by the concerted action of low concentrations of L-lysine plus L-threonine; individually, lysine was a mild activator and threonine was weakly inhibitory. Furthermore, the enzyme activity was also inhibited by the combination of L-threonine and L-methionine (3, 9). Although, in metabolic terms, concerted feedback inhibition has proven to be an important physiological device for the control of early enzymes (3, g-11), relatively little information is available on the mechanism of this phenomenon at the molecular level. In this communication we describe the purification, some general properties, and regulation of the aspartokinase from P. jluores- tens. In the accompanying paper (12), we present experimental evidence to support the notion that the mechanism of concerted feedback inhibition may involve modifier-induced formation of inactive enzyme oligomers. A preliminary report of some of these findings has been published (see Ref. 13).

EXPERIMENTAL PROCEDURE

Materials

All ingredients of the bacterial growth medium were obtained from Difco Laboratories. Chromatographically pure amino acids and amino acid analogs were supplied by Sigma Chemical Co., Calbiochem, and Cycle Chemicals. ATP was supplied by Sigma. Yeast alcohol dehydrogenase, catalase, P-galactosidase, and peroxidase were purchased from Worthington Biochemical Corp.; bovine serum albumin was obtained from Miles Labora- tories.

3534

by guest on May 8, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Methods

Aspartokinase Activity Assays-Enzyme activity was routinely assayed by the method of Black (2). The standard reaction mixture, in a final volume of 1.0 ml, contained in micromoles: hydroxylamine hydrochloride, 608; MgC12, 10; KCl, 200; ATP, 20; and n-aspartic acid, 50. One unit of enzyme was defined as the amount producing 1 pmole of aspartohydroxamate in 60 min at 25’ as calculated from the molar extinction coefficient of 625 determined with authentic ,&aspartohydroxamate. All values reported with this procedure were within the linear range of the assay.

In some instances, a coupled assay using pyruvate kinase and lactate dehydrogenase as described by Wampler and Westhead (14) was employed. The standard reaction mixture contained 50 mM imidazole-HCl buffer, pH 7.6, 1 mM ATP, 20 XnM L-as-

partic acid, 200 mM KCl, 10 mM MgC12, 2 mM phosphoenolpyru- vate, 0.3 mM NADH, 5 pg of lactate dehydrogenase, and 50 pg of pyruvate kinase in a final volume of 1.0 ml. The initial veloc- ities were linear up to 2.5 pg of protein.

Protein Determination-Protein was routinely assayed by the method of Lowry et al. (15).

Polyacrylamide Gel Electrophoresis-Analytical disc gel elec- trophoresis was carried out in 7.5% gels according to the method described by Brewer and Ashworth (16). Protein bands were located by staining with Coomassie blue. For an in situ activity stain, the gel was incubated in a reaction mixture for the hy- droxamate assay; after 15 to 30 min, it was transferred to a tube containing the FeC13 color reagent, and a faint brown band could be seen which faded rapidly.

For preparative gel electrophoresis, a 7.5% gel, 2.5 x 7 cm, was pretreated electrophoretically at 5 ma for 1 hour in 0.074 M TrisL0.192 M glycine buffer, pH 7.4, containing 0.25 mM DTTl and 1 mM each of n-threonine and n-lysine. The gel was eluted with 0.074 M Tris-HCl, pH 7.2, containing 10 mM P-mercapto- ethanol and 1 mM each of n-lysine and L-threonine.

The procedure for routine subunit studies and molecular weight calibrations using sodium dodecyl sulfate-polyacrylamide gels was that of Weber and Osborn (17).

Sedimentation Velocity and Dij’usion Constant Measurements- Sedimentation velocity experiments were carried out in a Beck- man model E analytical ultracentrifuge using the An-D rotor. All sedimentation coefficients were converted to SZO,~ values after correction for temperature and solvent viscosity. Diffusion coefficients were measured following the standard technique (18).

Stokes Radius Determination-Stokes radius was determined on calibrated Sephadex G-200 columns as described (19, 20). The location of the protein peaks were established by assaying for their activities (14, 21).

Strain and Growth Conditions-A wild type strain of P. &ores- ccns (P-14) was kindly supplied by Dr. Ronald H. Olsen of the Department of Microbiology, the University of Michigan. A stock culture was maintained on nutrient slants. Large scale cultures were grown at 25” in a New Brunswick Fermacell Fer- mentor in the minimal salts medium of Vogel and Bonner (22) supplemented with 0.2% glucose.

RESULTS

Enzyme PuriJication

Preliminary experiments revealed that aspartokinase was stable at 25”. Therefore, except as noted, all purification steps were carried out at room temperature.

1 The abbreviation used is: DTT, dithiothreitol.

8535

Crude Extract-P. jkorescens cells (800 to 900 g, wet weight) were suspended in 50 mM potassium phosphate buffer, pH 7.5, containing 10 mM /3-mercaptoethanol, 1 mM n-lysine, and 1 mM n-threonine (Buffer A) at 1 ml of buffer per g of cells. The slurry was passed once through a Manton-Gaulin homogenizer at 9,000 p.s.i. The resulting homogenate was treated with ribonuclease A and deoxyribonuclease I, 2 mg each, stirred for 1 hour, and centrifuged at 48,000 X g for 60 min at 4’. Unbroken cells, resuspended in the same buffer, were sonically disrupted (Branson Sonifier) and centrifuged as above. The two super- natant fluids were combined.

Heat Treatment-The crude extract was divided into 150-ml portions and heated in a shaking water bath for 4 min at 53”. After chilling in ice water the suspension was centrifuged at 39,000 X g for 20 min at 4’.

Ammonium Sulfate Fractionation-The pH of the supernatant fluid following the heat step was adjusted to pH 7.5 by dropwise addition of 1 N KOH. Twenty-one grams of solid ammonium sulfate were added to each 100 ml of heat-treated supernatant fluid over a period of 1 hour. The solution was stirred for 1 hour, centrifuged at 27,000 x g for 40 min, and the precipitate was discarded. For each 100 ml of the supernatant fluid, 6.3 g of ammonium sulfate were added and the suspension was stirred and centrifuged as above; the precipitate was discarded. To precipitate the enzyme, 6.5 g of ammonium sulfate were added to each 100 ml of the supernatant fluid obtained from the previous step. After stirring and centrifugation, the precipitate was resuspended in a minimal volume of Buffer A, and was dialyzed against two changes of 50 volumes of the same buffer.

Ion Exchange Chromatography-A DEAE-cellulose column (4 x 56 cm) was poured and equilibrated in Buffer A. The dialyzed ammonium sulfate fraction was layered on the column at a flow rate of 60 ml per hour and a linear 3000-ml gradient from 0 to 0.4 M KC1 in Buffer A was applied. Flow rate during elution was 160 ml per hour and 15-ml fractions were collected. Elution profiles of enzyme activity and 280nm absorbing ma- terial for this step are shown in Fig. 1. Fractions with high enzyme activity were pooled and concentrated in a Diaflo ultra- filtration apparatus.

t

1

40 60 80 100 120 140 160

Fraction Number

FIG. 1. Chromatography of aspartokinase on DEAE-cellulose. Dialyzed ammonium sulfate fraction (105 ml) was applied on a DEAE-cellulose column (56 X 4 cm) and the enzyme was eluted at 25” with a linear KC1 gradient (from 0 to 0.4 M KC1 in Buffer A). Flow rate during elution was 160 ml per hour and 15-ml fractions were collected. Enzyme activity is expressed as micromoles of aspartohydroxamate per hour per ml. 0, Am; 0, enzyme ac- tivity; inset, a magnified profile of the activity eluted between Fractions 30 and 55.

by guest on May 8, 2018

http://ww

w.jbc.org/

Dow

nloaded from

8536

TABLE I Purijication of Pseudomonas Jluorescens aspartokinase

Unit9

Per ml Total

Fraction and step

I. Crude extract. . . . . . 63 II. Heat, treatment.. . . . 64

III. Ammonium sulfate.. . . 720 IV. DEAE-cellulose (con-

centrated). . 4,330 V. Sephadex G-200 I

(concentrated). . 5,520 VI. Sephadex G-200 II

(concentrated). . 5,780 VII. Sephadex G-200 III

Pool A. .._.._..._.. 4,140 Pool B. . . . . . . . . 1,100

VIII. Preparative gel elec- trophoresis (con- centrated). . 300

97,500 91,200 76,300

56,300

51,900

41,600

25,700 5,400

900 0.45 670

Proteinb concen- tration

Specific activity

wdml

32 24 68

63 70

8.4 660

4.8 1,200

2.9 1,430 0.82 1,340

0 Micromoles of aspartohydroxamate per hour at 25”. b Estimated by the method of Lowry et al. (15).

Sephadex G-200 I-Sephadex G-200 was allowed to swell in 50 mM potassium phosphate buffer, pH 7.5, containing 10 mM /3-mercaptoethanol, 100 mM KCl, and 5 mM each of L-lysine and I,-threonine (Buffer B). Two columns (3.2 x 95 cm), poured and equilibrated in the same buffer, were connected in tandem. Concentrated material obtained from the DEAE-cellulose step was applied to the column and eluted with Buffer B at a flow rate of 20 ml per hour; 3-ml fractions were collected. Enzy- matically active fractions were pooled and concentrated in the Diaflo apparatus.

Sephadex G-200 II-Two columns (90 X 2.5 cm) were equili- brated and run in tandem with Buffer C (same as Buffer B but without L-threonine and L-lysine) . Concentrated material from G-200 I was applied and eluted at 15 ml per hour; fraction volume was 2 ml. Peak fractions were concentrated as above.

Xephadex G-200 III-The same two columns and buffer system employed for G-200 I were used in this step; flow rates and other conditions were as described above. Fractions with maximum specific activities were pooled and concentrated (Pool A). Sev- eral other fractions with slightly lower specific activities were also pooled (Pool B). The results of one such purification pro- cedure are summarized in Table I.

Preparative Gel Electrophoresis-For some experiments concen- trated enzyme solution from Sephadex G-200 III step, Pool B, was dialyzed against 10 mM potassium phosphate buffer, pH 7.5, containing DTT, L-lysine, and L-threonine, at 1 mM each. The sample was then subjected to electrophoresis as described under “Methods.”

The elution profile given in Fig. 1 shows that a very small fraction (< 4%) of the total activity applied on the DEAE-cellu- lose column did not adhere to the ion exchanger. The enzyme activity of this early eluting fraction was found to be almost to- tally insensitive to concerted feedback inhibition by L-threonine plus L-lysine, and to inhibition by L-threonine or L-2,4&a- minopimelic acid. It is not known whether the minor compo- nent was a desensitized form of the aspartokinase or an isoen- zyme similar to that observed in Bacillus stearothermophilus (23) and Bacillus subtilis (24).

4

I 1

I l.AJl I\ 1 1 .JIl .I 0 I2 3 4 56 7 8 9 IO

distcnce from top (ortitrory stole)

IFIG. 2. Polyacrylamide gel pattern of aspartokinase. Sixty micrograms of protein from Pool A, Table I, were subjected to gel electrophoresis as described under “Methods.” The gel was stained with Coomassie blue and a densinometric scan at 500 nm was recorded. Top of the gel is shown on the left. Band 1, high molecular weight impurity; Band d, active aspartokinase peak; Band 3, dissociation product of the enzyme; Band 4, dye front.

Charaderization of Enzyme

Estim.ation of Purity-Enzyme samples were analyzed on analytical polyacrylamide gels in Tris-glycine buffer system, pH 9.5 (16). A densitometric scan of a gel obtained with Pool A from Sephadex G-200 III step and stained with Coomassie blue, as shown in Fig. 2, revealed a major band (Band 8) and a smear of protein-positive material migrating faster than this band. Using the activity stain, the major band was shown to correspond with aspartokinase activity. The minor band (Band 1) and the smeared area did not have enzyme activity. Both Bands 1 and S, and some of the protein-positive material present in the smeared area were absent from the gels that were obtained with enzyme purified by preparative gel electrophoresis at pH 7.4. Furthermore, when the active enzyme (Band 2) was sliced off, homogenized, and subjected to re-electrophoresis at pH 9.5, several low molecular weight components were visible in addition to the main band. These results suggested extensive dissociation of the enzyme during gel electrophoresis.2

Electrophoresis on cellulose acetate paper (25) of the Pool A material from Sephadex G-200 III step revealed only one protein band with no indication of any dissociation product being formed. A photograph of the schlieren pattern during sedimentation velocity centrifugation (see Fig. 3) also did not reveal any small molecular weight contaminants. Based on these results, we conclude that the enzyme was highly purified; from the densito- metric scan shown in Fig. 2, and assuming that the low molecular weight material was indeed dissociation product of the native enzyme, we estimated that the enzyme was greater than 90% pure.

Molecular Weight and Stokes Radius-The molecular weight of the native enzyme was determined by sedimentation velocity centrifugation. Fig. 3 shows the schlieren pattern of the enzyme at 7 mg per ml. A slight decrease of the sedimentation coefficient on increasing protein concentrations was observed; an &+, value

9 Several variations in the condition of gel electrophoresis such as the presence or absence of amino acid modifiers and DTT, replacement of Tris with imidazole or sodium phosphate buffers at different pH values, elimination of ammonium persulfate as the catalyst, and prior reduction of the enzyme with 0.1 M DTT did not significantly alter the gel pattern. Addition of 40 mM L- aspartate, or 20 mM ATP plus 10 mM MgClz to both gel and elec- trophoresis buffer, however, reduced the extent of enzyme break- down to a large degree.

by guest on May 8, 2018

http://ww

w.jbc.org/

Dow

nloaded from

8537

0.6

I 1

30 40 50 60 Stokes’ radius (i )

FIG. 4. Determination of Stokes radius of aspartokinase. En- zyme from Pool A, Table I (420 pg), was mixed with 200 pg of yeast alcohol dehydrogenase, 75 fig of horseradish peroxidase, and 100 pg of rabbit muscle pyruvate kinase in a total volume of 0.15 ml and applied to a Sephadex G-200 column (86 X 1.4 cm) equilibrated at 25” with 50 mM potassium phosphate buffer, pH 7.5, containing 200 mM KC1 and 0.02% sodium azide. The proteins were eluted with the same buffer at a flow rate of about 5 ml per hour and 0.75-ml fractions were collected. The location of protein peaks were established by assaying for their activities as given under “Methods.” KD values were calculated according to the method given by Siegel and Monty (20). In separate experiments, it was established that 0.02% sodium azide did not influence the elution

FIG. 3. Sedimentation velocity pattern of aspartokinase. profile or the enzyme activities of these proteins. AK = asparto-

Purified enzyme (7.0 mg per ml, Step VII, Pool A, Table I) ex- kinase; PK = pyruvate kinase; ADH = yeast alcohol dehydro-

tensively dialyzed against 50 mM potassium phosphate buffer, genase; PEROX = horseradish peroxidase.

pH 7.5, containing 100 mM KC1 and 10 mM &mercaptoethanol was sedimented in a single sector cell at 20”. Photograph was taken TABLE II 64 min after the rotor had reached two-thirds of the top speed of 52,000 rpm (bar angle GO’). Sedimentation was from left to right.

Requirements for aspartokinase reaction The complete reaction mixture contained in a final volume of 1

ml the following: 608 pmoles of hydroxylamine, 10 pmoles of of 6.70 was calculated from extrapolation of the data at infinite MgClz, 200 Mmoles of KCl, 20 pmoles of ATP, and 50 pmoles of dilution. The Dto,, value at a single protein concentration of L-aspartic acid. In Experiment 1, 3 rg of protein from G-200 I

7 mg per ml was found to be 5.3 x lo-’ cm2 s-l. Assuming a step were incubated for 20 min at 25’. In Experiment 2, 8.4 rg

partial specific volume of 0.74 g per cc, a molecular weight of of protein from a different enzyme preparation were incubated

133,000 was calculat,ed. for 15 min at 25”. Wherever present the concentration of MnClr

Gel filtration on calibrated Sephadex G-200 was employed to was 10 mM and 200 mM LiCl or NaCl was used.

determine the Stokes radius of the aspartokinase molecule. At 25”, in 50 mM potassium phosphate buffer, pH 7.5, containing

system Aspart&&rcpnate

200 mM KCl, the Stokes radius of the enzyme was 42.8 A (Fig. 4). fmoles When the KC1 concentration of the gel filtration buffer was de- creased, the Stokes radius of the protein molecule increased sig-

Experiment I Complete............................... 0.98

nificantly; extrapolation to zero ionic strength gave a value of Omit KCl.. . . . . 0.61 46.6 A. Omit MgClt.. . . . . . 0.03

The molecular weight of the native enzyme was also estimated Omit ATP. . . <O.Ol from the Stokes radius as obtained from the Sephadex G-200 ex- Omit KC1 and MgClz. . . 0.02

periment. Using the Stokes-Einstein equation (20), a D~o,~ of Experiment 2 5.2 x lo-’ cm2 s-l was calculated from the Stokes radius; this Complete............................... 1.11

diffusion coefficient, with the .s&,,~ value of 6.7 S, yielded a molec- Omit MgCIz but add MnCls. . . 1.16

ular weight of 137,000 for the native enzyme. Omit KC1 but add LiCl . 0.61

Subunit dfolecular Weight-Purified aspartokinase obtained Omit KC1 but add NaCl. , 0.71

after preparative gel electrophoresis was subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (17). A sin- gle band corresponding to subunit molecular weight of 43,000

K+ from the reaction mixture decreased the activity by about 40%. The requirement for fiIg2+ could be replaced by Mnfi;

was observed. no other cations tested, including Zs@‘, Cazf, and Co*, could Requirements for Reaction-Table II shows that the phospho- substitute either Mg2f or &In*. In the presence of Mg2f, Li+,

rylation of L-aspartate by the P. Jluorescens aspartokinase was or Na+ had no effect. absolutely dependent, on ATP and a divalent cation; omission of In separate experiments, the enzyme was.found to be highly

by guest on May 8, 2018

http://ww

w.jbc.org/

Dow

nloaded from

8538

specific for both the donor and the acceptor of the phosphate group. Analogs of aspartate such as n-glutamate, nn-a-methyl aspartate, N-acetyl aspartate, and n-aspartate were inactive. ATP was the only known phosphate donor for the phosphoryla- tion reaction; ADP, AMP, UTP, CTP, and GTP were ineffective.

Control of Aspartolcinase Activity

Concerted Feedback by Amino Acid Pairs-In contrast to the other aspartokinases from a variety of bacterial species the ac- tivity of the P. jtuorescens aspartokinase was subject to concerted feedback inhibition by two pairs of amino acid end products. The data summarized in Table III show that n-threonine, at low concentration, was a,mild activator and exerted a weak inhibi- tory effect only at concentrations over 10 mM. The other end products of the pathway, n-lysine, n-methionine, and n-isoleucine, showed a small but reproducible stimulatory effect. However, a combination of n-lysine plus n-threonine strongly inhibited the phosphorylation reaction. Concerted feedback inhibition by n-threonine plus n-methionine was not nearly as strong as the effect of the n-lysine plus n-threonine combination.

Control experiments in Table III indicate no concerted feed- back inhibition by the other possible pairs of amino acids. That the regulation of enzyme activity by various amino acids was not an artifact of the purification procedure is illustrated in Table IV. The results show that the activation-inhibition patterns seen with crude extracts were virtually unchanged as the enzyme was purified through the Sephadex G-200 step.

Fig. 5 depicts the concentrations of n-lysine and n-threonine required for concerted feedback inhibition. It is clear that a drastic reduction in activity occurred at very low levels of these amino acids; at 0.25 mM each, the enzyme activity was inhibited by approximately 70 7..

Structural Specificity of Inhibiting Amino Acids--The results of concerted inhibition studies with structural analogs of n-threo- nine in the presence and absence of the natural amino acids are summarized in Table V. Only L-threonine amide and n-threo- nine methyl ester could replace n-threonine for the concerted feedback inhibition when present in combination with either n-lysine or n-methionine. Individually, these analogs were in-

TABLE III

Concerted feedback inhibition of aspartokinase activity Enzyme activities were measured using standard hydroxamate

assay. Each assay contained 4.2 pg of protein (Fraction VI, G-200, II, Table I) and the amino acids as indicated. Incubation was for 10 min at 25”. In the tube with no amino acid, 0.51 wale of aspartohydroxamate was produced under these conditions.

None.......................................... 100 n-Threonine (5 mM). . . . . . 111 n-Threonine (BORAX)........................... 88 n-Threonine (20 mM). . : :. . 67 L-Lysine (5 mM) . 112 n-Methionine (5mmM)........................... 136 n-Isoleucine (5 mM) . . 118 n-Threonine + L-lysine (5 mM each). . 4 L-Threonine + n-methionine (5 mM each). . 67 n-Methionine + n-lysine (5 mM each). . . 146 n-Methionine + n-isoleucine (5 mM each). . 130 n-Lysine + n-isoleucine (5 rnM each). . . . . 120 n-Threonine + n-isoleucine (5 mM each). . . 92

hibitory, but the effect was less than that observed with equi- molar concentration -threonine. It is clear&,hat the (y-car- boxy1 of n-threonine not required for inhibition; any other alteration of the ammo &id structure destroyed the&hibitory potential.

To determine the structural specificity of L-lysine, the follow- ing compounds were tested at 5 rnM each, ^&her alone or in ‘the presenceaof 0.25 mM n-threonine: D-lJ&k, cadavarine, n-arginine, L-2,4-diaminopimelic acid,“and S-@-aminoethyl)-n- cysteine. The only lysine analog that showed concerted inhibi- tion was S-@-aminoethyl)-n-cysteine indicatingrthat. the substi- tution of a sulfur atom for a methylene group in.%@ side chain reduced but did not eliminate binding on the enzyme molecule.

TABLE IV

Ej’ect of enzyme purification on feedback control of aspartokinase activity

The standard hydroxamate assay was used; where ,indio&ed each amino acid was present at 5 mM. The enzyme fromsephadex G-200 I step was diluted 500-fold to reduce the concentration of the amino acids to 5 PM; other enzyme samples were dialyzed against 50 mM potassium phosphate buffer, pH 7.5, prior to assay. The protein concentration, time of incubation, and the amount of aspartohydroxamate formed with no amino acid added were as follows: crude extract: 1.5 mg, 15 min, 0.68 pmole; heated super- natant: 1.1 mg, 30 min, 1.01 pmoles; ammonium sulfate fraction: 270 pg, 10 min, 0.54 Fmole; G-200 I: 9 pg, 10 min, 0.87 pmole.

Relative enzyme activity

Additions

None........................ . n-Threonine . L-Lysine . n-Methionine . n-Threonine + n-lysine L-Threonine + n-methionine. n-Methionine + n-lysine. . . .

100 113 106 122 20 81

127

-- 100 100 123 106 106 103 120 116 24 18 98 82

128 108

100 119 115 136

5 74

145

- 0 0.4 0.8 1.2 1.6 2.0 L-Threonine or L-lysine concentration (mM)

FIG. 5. Concerted feedback inhibition of aspartokinase L-lysine plus L-threonine. l , progressive inhibition with

by in-

creasing n-threonine concentration at a constant level (0.25 mna) of n-lysine. 0, progressive inhibition with increasing n-lysine concentration at a constant level (0.25 mM) of n-threonine. Cou- pled assay with 2.1 pg of protein (Fraction VI, G-200 II, Table I) were used. One hundred per cent activity was equal to 0.015 pmole of aspartohydroxamate formed per min.

by guest on May 8, 2018

http://ww

w.jbc.org/

Dow

nloaded from

TABLE V

Specijicity of A-threonine for concerted feedback inhibition

The standard coupled assay with 2.1 pg of protein from Sepha- dex G-200 II was used. For Experiment 1, the concentrations of L-threonine and the amino acid analogs were 5 mM; for Experi- ment II, the concentrations were 2 mM. In controls, with no amino acid added, 0.034 pmole of aspartohydroxamate per min was produced under these conditions.

Threonine analogs

None. L-Threonine . n-Threonine n-Threonine amide. n-Threonine methyl ester N-Methyl-L-threonine 0-Methyl-L-threonine. . Glycylthreonine L-Serine.................

T I-

Relative enzyme activity

Experiment I I No

lysine

100 30

58 54

-

( I

-

-

With b.25 mhl Aysine -

102 3

94 5 6

102 100 98

105

_-

I

--

-

Experiment II

No nethio-

nine

100 52

83 74

With 5rnM

-methio- nine

119 31

115 62 57

119 115 114 119

1401 5 mM L-LYS -I

‘“L ;mM L-THRT5mM L-LY; ,

2 4 6 8 IO

L-Methionine concentration (mM)

FIG. 6. Effects of n-methionine on the concerted feedback in- hibition by n-lysine plus n-threonine. One hundred and fi f ty micrograms of protein from the DEAE-cellulose step (Table I) were used in the standard hydroxamate assay. Incubation was for 30 min at 25”. Amino acid concentrations were as indicated. One hundred per cent activity was equal to 0.90 pmole of asparto- hydroxamate formed.

Modulation of Concerted Feedback Inhibition by z-Methionine- Although L-lysine plus n-threonine, and n-methionine plus L-

threonine inhibited enzyme activity in a concerted manner (see Table III), inclusion of all three amino acids in the reaction mixture revealed a compensatory role of methionine in modulat- ing the degree of inhibition by the n-lysine plus n-threonine com- bination. The results are summarized in Fig. 6. With increas- ing concentrations of n-methionine, concerted inhibition by 0.1 mu n-threonine plus 5 mM n-lysine was almost totally compen-

In addition to the inhibitory effect of L-methionine in concert with n-threonine, the sulfur-ammo acid also exhibited a com- pensatory role on the concerted feedback inhibition elicited by the n-threonine plus n-lysine pair (Fig. 6). This effect was seen only when the concentration of n-threonine was low (in combina- tion with high n-lysine) ; under this condition, addition of 10 mru n-methionine almost completely counteracted the concerted in- hibition. Although the exact molecular mechanism of the “re- versal” by n-methionine of the n-threonine plus n-lysine inhibi- tion is not known, when compared with other over-all control patterns seen in a variety of bacterial species (3), the compensa- tory phenomenon appears to have some physiological significance. Since n-threonine is an end product by itself as well as a precursor of another end product n-isoleucine (l), it is reasonable to expect that at low L-threonine concentration and at high concentrations of both n-lysine and n-methionine, a severe deprivation of L-

threonine (also n-isoleucine) would slow down the cellular growth unless sufficient amounts of the precursors for n-threonine bio- synthesis were made available by relieving the concerted feed- back of aspartokinase. When n-threonine is present in excess (in addition to excess n-lysine and n-methionine), no limitation of L-threonine is predicted, and the synthesis of all these end products should be efficiently regulated by normal feedback mechanisms of the early enzymes. (Excess n-threonine would

sated; the strong inhibitory effect of 1.0 mM threonine plus 5 rnM be adequate to supply both n-threonine and L-isoleucine.) The

8539

lysine was completely unaffected even up to 10 mM n-methionine. A possible physiological significance of the dual role of methio- nine will be discussed below.

DISCUSSION

Several lines of evidence indicate that the P. Jluofescens as- partokinase is quite distinct from other kinases studied thus far. The activity of this enzyme is not associated with the homoserine dehydrogenase activity; upon chromatography on DEAE-cellu- lose these two activities could be easily separated (26, also see Ref. 27). From the data presented in this article, it appears that the enzyme isolated from this bacterium has an approximate molecular weight of 135,000; gel electrophoresis in sodium dodecyl sulfate showed a single band corresponding to a subunit molecular weight of 43,000, suggesting that the enzyme is composed of three subunits of equal molecular size. The aspartokinase from Ba- cillus polymyxa, which is also regulated by concerted feedback inhibition, appears to be a tetramer (molecular weight 116,000) composed of two types of nonidentical subunits of 47,000 and 17,000 (28).

In the context of regulation of enzyme activity, the P. jluores- tens aspartokinase shows some properties that are distinguishable from the other aspartokinase subject to concerted feedback in- hibition. In Pseudomonas putida (29) the enzyme activity was strongly inhibited by n-lysine or by n-threonine; when added together at low concentrations, n-lysine plus L-threonine brought about drastic inhibition. Similar results were obtained with the B. polymyxa enzyme (30). In contrast, n-threonine inhibited the activity of P. jlwrescens aspartokinase only slightly and n-lysine had no inhibitory effect (Table III). When n-lysine and L-threonine were present ‘together the enzyme activity was severely inhibited (cf. Table III and Fig. 5). The important difference between these aspartokinases is, therefore, that in the case of P. Jluorescens, only one member of the pair involved in the concerted feedback inhibition shows a weak inhibitory effect, and the second member of the pair is slightly stimulatory. This particular combination is similar to the concerted feedback pat- tern reported in R. capsulata (6).

by guest on May 8, 2018

http://ww

w.jbc.org/

Dow

nloaded from

8540

dual role of n-methionine, i.e. its ability to inhibit enzyme activity in concert with n-threonine as well as its capacity to compensate the concerted feedback inhibition by n-threonine plus n-lysine, therefore, appears to be an important physiological device for the regulation of the activity of the first enzyme of the aspartate metabolism in P. JEuorescens.

Achxowledgments-We thank Mr. John Trojanowski for run- ning the analytical ultracentrifuge, Mr. Joseph A. Gingras for constructing the preparative polyacrylamide gel apparatus, and Dr. M. Hunter for valuable discussions.

REFERENCES

1. MEISTER, A. (1965) Biochemistry of the Amino Acids, 2nd Ed, Vol. 2, p. 674, Academic Press, New York

2. BLACK, S. (1962) Methods Enzymol. 6, 821 3. DATTA, P. (1969) Science 166,556562 4. STADTMAN, E. R., COHEN, G. N., LEBRAS, G. & DE ROBICHON-

SZULMAJSTER, H. (1961) J. Biol. Chem. 236, 2033-2038 5. COHEN, G. N. (1969) Cm-r. Top. Cell. Regul. 1,183-231 6. DATTA, P. & GEST, H. (1964) Proc. Nat. Acad. Sci. U. S. A.

62, 1004-1009 7. PAULUS, H. & GRAY, E. (1964) J. Biol. Chem. 239,400&-4009 8. KURAMITSU, H. K. (1968) Biochim. Biophys. Acta 167, 643-

645 9. DATTA, P., DUNGAN, S. M. & FELDBERG, R. S. (1973) in Ge-

netics of Industrial 2Microorganisms (VANEK, Z., HOSTALEK, Z. & CUDLIN, J., eds) pp. 177-193, Academia, Prague

10. BURLANT, L., DATTA, P. & GEST, H. (1965) Science 148, 1351- 1353

11. STADTMAN, E. R. (1970) The Enzymes (BOYER, P. D., ed) pp. 397-459, Academic Press. New York

12. DUNGAN,‘S. M. & DATTA, ‘P. (1973) J. BioZ. Chem. 248, 8541- 8546

13. DUNGAN, S. k. & DATTA, P. (1972) Fed. Proc. 31,423 14. WAMPLER, D. E. & WESTHEAD, E. W. (1968) Biochemistru 7.

1661-1671 ” .

15. LOWRY, 0. H., ROSEBROUGH, N. J., F.~RR, A. L. & RANDALL, R. J. (1951) J. Biol. Chem. 193,265-275

16. BREWER, J. M. & ASHWORTH, R. B. (1969) J. Chem. Educ. 46, 41-45

17. WEBER, K. & OSBORN, M. (1969) J. Biol. Chem. 244,4406-4412 18. SCHACHMAN, H. K. (1959) Ultracentrifugation in Biochemistry,

Academic Press, New York 19. ACKERS, G. K. (1964) Biochemistry 3, 723-730 20. SIEGEL, L. M. & MONTY, K. J. (1966) Biochim. Biophys. Acta

112,346-362 21. DATTA, P. (1970) J. BioZ. Chem. 246, 5779-5787 22. VOGEL, H. J. & BONNER, D. M. (1956) J. Biol. Chem. 218,

97-m 23. KURAMITSU, H. K. & YOSHIMURA, S. (1971) Arch. Biochem.

Biophys. 147, 683-691 24. ROSNER, A. & PAULUS, H. (1971) J. BioZ. Chem. 246,2965-2971 25. Gelman Procedures, Techniques, and Apparatus for Electro-

phoresis (1968) Gelman Instrument Company, Ann Arbor 26. DUNGAN, S. M. (1972) Ph.D. thesis, University of Michigan 27. BOTHWELL, M. A. & DATTA, P. (1971) Biochim. Biophys. Act&

236, 1-13 28. BISWAS, C., GRAY, E. & PAULUS, H. (1970) J. BioZ. Chem. 246,

49004906 29. ROBERT-GERO, M., POIRET, M. & COHEN, G. N. (1970) Bio-

chim. Biophys. Acta 206, 17-30 30. PAULUS, H. & GRAY, E. (1968) J. BioZ. Chem. 243,1349-1355

by guest on May 8, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Stephen M. Dungan and Prasanta DattaASPARTOKINASE FROM PSEUDOMONAS FLUORESCENS

Concerted Feedback Inhibition: PURIFICATION AND SOME PROPERTIES OF

1973, 248:8534-8540.J. Biol. Chem. 

  http://www.jbc.org/content/248/24/8534Access 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/248/24/8534.full.html#ref-list-1

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

by guest on May 8, 2018

http://ww

w.jbc.org/

Dow

nloaded from