Rat liver pyruvate kinase: Influence of ligands on activity and fructose 1,6-bisphosphate binding

ARCHNES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 232, No. 1, July, pp. 202-213, 1934

Rat Liver Pyruvate Kinase: Influence of Ligands on Activity and Fructose 1,6-Bisphosphate Binding’ JAMES Lkpartment

of

Biochemistry, Received

B. BLAIR’ West Virginia October

ROBERT

AND University

G. WALKER3

Medicul Center, Morgantown,

21, 1983, and in revised

form

March

West Virginia

26506

16, 1934

The ability for various ligands to modulate the binding of fructose 1,6-bisphosphate (Fru-1,6-P,) with purified rat liver pyruvate kinase was examined. Binding of Fru-l,& Pz with pyruvate kinase exhibits positive cooperativity, with maximum binding of 4 mol Fru-1,6-Pz per enzyme tetramer. The Hill coefficient (nu), and the concentration of Fru-l,&Ps giving half-maximal binding [FBP&, are influenced by several factors. In 150 mM Tris-HCl, 70 mM KCl, 11 mM MgS04 at pH 7.4, [FBP]i,s is 2.6 PM and nn is 2.7. Phosphoenolpyruvate and pyruvate enhance the binding of Fru-1,6-Pz by decreasing [FBP]i,z. ADP and ATP alone had little influence on Fru-1,6-P, binding. However, the nucleotides antagonize the response elicited by pyruvate or phosphoenolpyruvate, suggesting that the competent enzyme substrate complex does not favor Fru-1,6-Pz binding. Phosphorylation of pyruvate kinase or the inclusion of alanine in the medium, two actions which inhibit the enzyme activity, result in diminished binding of low concentrations of Fru-1,6-Pz with the enzyme. These effecters do not alter the maximum binding capacity of the enzyme but rather they raise the concentrations of Fru-1,6-Pz needed for maximum binding. Phosphorylation also decreased the nn for Fru-1,6-Pe binding from 2.7 to 1.7. Pyruvate kinase activity is dependent on a divalent metal ion. Substituting Mn2+ for Mgz+ results in a 60% decrease in the maximum catalytic activity for the enzyme and decreases the concentration of phosphoenolpyruvate needed for half-maximal activity from 1 to 0.1 m&I. As a consequence, Mn2+ stimulates activity at subsaturating concentrations of phosphoenolpyruvate, but inhibitis at saturating concentrations of the substrate or in the presence of Fru-1,6-Ps. Both MS+ and Mn2+ diminish binding of low concentrations of Fru-1,6-P2; however, the concentrations of the metal ions needed to influence Fru-1,6-P2 binding exceed those needed to support catalytic activity. Pyruvate kinase (ATP:pyruvate phosphotransferase, EC 2.7.1.40) is an important regulatory enzyme of hepatic carbohydrate metabolism. The primary isozyme of pyruvate kinase in the liver (Type L) is distinct from the isozymes found in glycolytic tissues, and is subject to regulation by several mechanisms. Early studies by Tanaka and co-workers (1,2) demonstrated

that liver pyruvate kinase is subject to allosteric regulation by several metabolites, making its activity responsive to the supply of nutrients to the liver. Several sugarphosphates are capable of stimulating its activity, the most potent of which is Frul,6,-P2.4 alanine and other amino acids inhibit the activity of liver pyruvate kinase ‘Abbreviations used: CAMP, adenosine-3’,5’-monophosphate; DTT, dithiothreitol; Fru-1,6-Pa, fructose1,6-hisphosphate; [FBP]i,a, the concentration of Fru1,6-Pa giving one-half maximal binding; [PEP]i,a, the concentration of P-enolpyruvate giving half-maximal activity; nu, the Hill coefficient; P-enolpyruvate, phosphoenolpyruvate.

’ This investigation was supported by a grant from the National Institutes of Health, United States Public Health Service (AM 17664). a To whom reprint requests should be sent. * Current address: Department of Chemistry, Marietta College, Marietta, Ohio. 0003-9861/34 Copyright All rights

$3.00

0 1984 by Academic Press. Inc. of reproduction in any form reserved.

202

LIVER

PYRUVATE

and this inhibition is overcome by Fru-1,6Pz (3-6). More recently, several laboratories have demonstrated that liver pyruvate kinase is subject to phosphorylation by CAMP-dependent protein kinases, and that phosphorylation in response to hormonal stimulation of the liver decreases the activity under physiological conditions (3, 7-9). Although the kinetic properties of liver pyruvate kinase have been extensively examined, very little is known about the direct interaction of various ligands with the enzyme. Fru-1,6-Pz has been shown to bind very tightly with the enzyme; however, characterization of this binding using equilibrium dialysis or gel-filtration techniques has not been extensive (10-12). Nitrocellulose adsorption techniques have recently been shown suitable for studying interaction of Fru-1,6-Pz with the enzyme (9,13). Using this technique, El-Maghrabi et aZ. (13) demonstrated that Fru-1,6-Pz binding is enhanced by the substrate Penolpyruvate and antagonized by the allosteric inhibitor alanine. We have also studied Fru-1,6-Pa binding with rat liver pyruvate kinase, and have expanded previous studies concerning the interaction of various ligands with the enzyme.The results of our investigation are contained in this report. A preliminary report of this work has been presented (14). MATERIALS

AND

METHODS

Crystalline rat liver pyruvate kinase was prepared by the procedure of Harada et al. (15) using the modifications previously described (9). The enzyme had a specific activity of 150-200 units/mg. It was stored as an amorphous precipitate in 2.6 M ammonium sulfate containing 50 @M Fru-1,6-P, and 30 mM Z-mercaptoethanol in a sealed tube under a nitrogen atmosphere. The properties and activity of the enzyme are stabilized under this condition; however, an approximate 50% decrease in the specific activity was found over a B-month period. This decrease in activity had no marked influence on Fru-1,6-P, binding. Solutions of pyruvate kinase used for kinetic and binding studies were prepared fresh each day. To remove Fru1,6-P, for kinetic and binding studies, the precipitated enzyme was extensively washed by centrifugation and resuspension in fresh 2.6 M ammonium sulfate containing 2-mercaptoethanol (three times). The precipitate was then dissolved in 10 mM phosphate buffer

KINASE

203

(pH 7.0) and reprecipitated with ammonium sulfate (2.6 M final concentration). The final precipitate was collected by centrifugation and dissolved in the appropriate buffer (see below). No attempts were made to remove ammonium sulfate from the preparation, and the final concentration of ammonium sulfate in the binding analysis was approximately 5 to 10 mM. Concentrations of pyruvate kinase in working solutions was determined by measuring the absorbance at 280 nm, using an extinction coefficient of 0.5 ml/ w (9). Basal phosphate content of the enzyme preparation used in this investigation, as determined by the method of Stull and Buss (16), was 0.8 mol phosphate/ 250,000 g protein (approximately 0.2 mol/subunit). Maximum phosphorylation of the enzyme was attained by incubating pyruvate kinase with CAMPdependent protein kinase from pig heart as previously described (9). As previously reported, maximum phosphorylation does not give phosphorylation of each of the subunits for reasons yet unresolved. The phosphate content of the phosphorylated form of the enzyme used in this study was 2.9 mo1/250,000 g. Phosphorylation resulted in changes in the kinetic properties of pyruvate kinase identical to that previously reported (9). Kinetic analysis of pyruvate kinase was carried out at 25°C and pH 7.5 in a standard buffer containing 150 mM Tris-HCI, 70 mM KCl, and 11 mM MgSO, by coupling formation of pyruvate to the oxidation of NADH using lactate dehydrogenase (1’7). The concentrations of ADP and P-enolpyruvate employed are provided in the description of individual experiments. In some studies, MgSO, was deleted and replaced with MnSO, at indicated concentrations. Binding of Fru1,6-Pr with pyruvate kinase was carried out using the nitrocellulose adsorption technique previously described (9), with the exception that the binding buffer was changed to be similar to that buffer used for kinetic analysis. The binding buffer at pH 7.4 contained 150 mM Tris-HCl, 70 mM KCl, 11 mM MgSOI, and 1 mM D’IT. In some studies, MgSO, was omitted as noted. Pyruvate kinase was dissolved to a concentration ranging from 0.6 to 0.9 FM (2 to 4 j&M binding sites) and incubated for 10 min at 0°C with varying concentrations of [U-‘%]Fru-1,6-P* (Amersham, 239 mCi/mmol) and other effectors as noted. The equilibrium mixture of free Fru-1,6-P, and the Fru-1,6Pr-pyruvate kinase complex (0.025 ml) was then rapidly quenched into 10 ml of 10 mM potassium phosphate (pH 7.0) containing 1 mM DTT. The solution was rapidly filtered through type HA, 2.5-cm nitrocellulose filter discs (Millipore). Tbe discs were washed two times with 5 ml of the phosphate buffer to remove free Fru-1,6-P,. Dilution and washing of the enzyme bound to the disc was accomplished within 20-25 s. The filter discs were then dried and counted. To assess the time required to achieve equilibrium,

204

BLAIR

AND WALKER

the binding of Fru-1,6-P* with pyruvate kinase was measured at various times after mixing the reagents. In the Tris-KCl-Mg buffer, maximum binding was observed within 3 to 6 min at all concentrations of Fru-1,6-P* examined. To assure equilibrium was achieved, the standard incubation period used was 10 min. In many of the individual experiments presented, additional samples were taken at 15 min to assure that the binding was not changing with time. To determine to what extent the Fru-1,6-Ps-pyruvate kinase complex may dissociate during the dilution and nitrocellulose binding steps, the binding was examined at various time periods after diluting equilibrium mixtures into cold 10 mnr phosphate-l mM Dl”I at pH 7.0. In these studies, the amount of complex remaining bound to the nitrocellulose filter was found to be constant for at least 15 min. Fifteen seconds was the shortest time period examined, which represents the minimum time required for filtering the sample and washing the disc. These observations suggest that the rate of dissociation of the complex is quite slow under the conditions applied. Binding studies were not successful when high-ionic-strength buffers, which allow rapid dissociation of the complex, were used for adsorption and washing of the discs. Analysis of the binding and kinetic parameters were accomplished by computer fitting to the Hill equation using the microcomputer analysis described by Knack and Rohm without weighing individual points (18). This curve-fitting analysis provides an independent estimate for three parameters: maximum binding (B-); the concentration of Fru-1,6-P* giving halfmaximal binding ([FBPl&; and the Hill coefficient (nn). Estimates of four intrinsic binding constants for the sequential model of Edsall and Wyman (19, 20) were also made for comparison with the recent report of El-Maghrabi et aL (13). Preliminary estimates of the four intrinsic binding constants were computer adjusted to give the best “least-squares” fit to the actual data. The Hill analysis of Knack and Rohm (18) was routinely applied for the present study, since it gives an independent estimate of the maximum number of binding sites and does not make any assumptions as to the stoichiometry of binding. All reagents were of the highest grade commercially available. The concentrations of free metal ions present for kinetic studies carried out in the presence of high concentrations of ADP were calculated using the dissociation constant of 0.1 mM for MnADP (21) and 0.36 mM for MgADP (22), at the ionic strength and pH conditions applied in this investigation. The results of binding experiments are presented as the means of duplicate determinations- within a single experiment unless noted. Each finding was confirmed in at least one additional experiment unless otherwise noted. Results from kinetic and binding isotherms analyzed by computer fitting are presented as the best fit f SE obtained from the fit.

RESULTS

General features of Fru-1,6-P,

binding.

The binding of Fru-1,6-Pz to rat liver pyruvate kinase exhibits a sigmoidal response with increasing concentrations of Fru-l,& Pz (Fig. 1). Maximal binding was found to be nearly 4 mol Fru-1,6-PJ250,OOO g pyruvate kinase, representing one binding site per subunit. [FBP]1,2 and nn for Fru1,6-Pz binding are markedly influenced by a number of conditions, which are discussed in detail below. Analysis of Fru-l,& Pz binding in the absence of added effecters using the Hill model (18) gave a value of 2.6 PM for [FBPlllz, with an nn of 2.7 (Table I). Determination of the individual intrinsic binding constants assuming four independent binding sites by the method of Edsall and Wyman (19, 20) gave the following values for the apparent intrinsic binding constants: K{ = 0.05 PM-‘, Ki = 0.08 PM-‘, Kk = 0.94 PM-‘, and Kk = 6.5 PM-~. The successive increments in the apparent intrinsic association constants de-

I’

l ’

0.2

2

4

6

L-ALANINE

e

IO

25

FIG. 1. Concentration dependence of Fru-1,6-P, binding. Rat liver pyruvate kinase was incubated-with varying concentrations of radiolabeled Fru-l&Pa. The bound protein and free radioligand were separated by adsorption to nitrocellulose disca as described under Materials and Methods. The amounts of Fru-1,6Pa (FBP) bound at the various concentrations of free Fru-1,6-Ps (FBP free) are presented. Binding was carried out in the complete binding buffer containing.no additions, 3 mm P-enolpyruvate (PEP), and 3 mrd balanine as indicated. Each point represents the mean of duplicate determinations.

LIVER

PYRUVATE TABLE

INFLUENCE

Buffer

OF VARIOUS

LICANDS

Enzyme (D or P)

205

KINASE I

ON Fru-1,6-P,

BINDING

WITH PYRUVATE

lFBPl,/2

B mm (mol/mol)

Effector

+ k f +

0.1 0.1 0.3 0.2

KINASE”

(Wd) 2.6 1.1 10.8 1.3

2 0.1 * 0.1 IIZ 0.8 +- 0.1

nH

Tris-KCI-Mg

D

None 3 mM P-enolpyruvate 3 mM Alanine 0.5 mM Cxaiate

4.0 3.9 4.4 3.3

2.7 2.9 2.6 2.1

+ 1+ +

0.4 0.7 0.4 0.5

Phosphate

D

None

3.7 It 0.4

0.93 * 0.07

3.2 f 0.6

Tris-KCI-Mg

P

None 3 mM P-enoipyruvate 3 mM Alanine

4.0 + 0.2 4.4 IL 1.6 4.4 k 1.2

5.4 -I- 0.6 1.2 * 0.1 18.0 + 6.7

1.7 f 0.3 2.0 +- 0.3 1.8 +- 0.4

“The binding of varying concentrations of Fru-1,6-Pz with dephosphorylated (D) or phosphorylated (P) enzyme was determined as described in Fig. 1 in the presence of the indicated effecters. The binding studies were conducted in the standard binding buffer described under Materials and Methods (Tris-KCl-Mg) or in 10 mM potassium phosphate (Phosphate). The binding results were analyzed by the Hill model, as described under Materials and Methods, and the maximum binding (B,,,,), concentration of Fru-1,6-P* giving one-half maximal binding [FBP],,, and the Hill coefficient (nu) are presented. The results represent the best fit +- SE obtained by computer curve fitting. Similar results were obtained in at least two separate experiments under each condition.

termined by this latter method are consistent with the positive cooperativity denoted by the Hill coefficient of 2.7. No evidence for negative cooperativity at higher concentrations of Fru-1,6-PZ was obtained in this study. To characterize changes in the Fru-l,&Pz interactions elicited by various effecters in the remainder of this report, the Hill analysis was applied and [FBPll,z and the nn are presented (Table I). I@uence of substrates and products on Fru-1,6-Pz binding. Substrates and products of the pyruvate kinase-catalyzed reaction influence the binding of Fru-1,6-Pz. The influence of substrates on the binding at subsaturating concentrations of Fru-1,6Pz (1.5-2.4 PM, total concentration) is presented in Table II. P-enolpyruvate and pyruvate enhance Fru-1,6-Pz binding. A full analysis at varying concentrations of Fru1,6-P, demonstrated that 3.0 InM P-enolpyruvate enhances binding of the activator only at low concentrations of Fru-1,6-Pz due to a shift in the apparent affinity for the ligand (Fig. 1 and Table I). Under the conditions applied in the present study, Penolpyruvate had no significant influence on the Hill coefficient for Fru-1,6-Pz bind-

ing, and it did not alter maximum binding at saturating concentrations of the ligand (Table I). The shift in the concentration TABLE

II

INFLUENCE OF SUBSTRATES AND PRODUIXS ON Fru1,6-P, BINDING WITH LIVER PYRUVATE KINASE’

Addition NO”C

3 mM 5 mM 5 mM 5 mu 10 mM 3 rn~ 5 mM 3 mM

P-suO~pyruVate

pyruvate ADP ATP ATP P-enolpyruvate + 5 m&i ADP Pyruvate + 5 mrd ATP P-enolpyruvate + 5 rnM ATP

Fru-1,6-F’s binding (9 control) 100 289 f 41 197 t 11 111 -+ 6 87 + 4 74b 172b 114 * 13 114b

’ Purified rat liver pyruvate kinase wss incubated in the standard binding buffer (Tris-KCl-Mg) with subsaturating concentrations of radiolabeled Fru-1,6-P% (1.5 to 2.4’rr.f total concentration) and the indicated effecters. Binding of Fru1,6-Ps with the enzyme was assessed ss described under Materials and Methods, and is expressed as the percentage of the binding observed in control conditions receiving no effector. The results represent the means f SE of three or more observations or the average of duplicate determinations, as indicated. *Mean of duplicate determinations.

206

BLAIR

AND

dependence of Fru-1,6-P, binding in response to 5 mM pyruvate was found to be similar to that seen with 3.0 MM P-enolpyruvate (not shown). Alone, ATP or ADP had little influence on the binding of Fru1,6-Pz. However, ATP and ADP did decrease the ability for P-enolpyruvate and pyruvate to enhance Fru-1,6-Pz binding (Table II). Oxalate is a competitive inhibitor of pyruvate kinase from several sources, including the liver enzyme (23, 24). At saturating concentrations of Fru-1,6-Pz (50 PM), oxalate competes with P-enolpyruvate with an apparent Ki of 26 PM oxalate (Fig. 2). In the absence of Fru-1,6-Pa, the response of pyruvate kinase activity to oxalate is complex and depends on both the concentration of oxalate and the concentration of P-enolpyruvate (Fig. 3). Low concentrations of oxalate activate pyruvate kinase activity at subsaturating concentrations of P-enolpyruvate (0.09 PM, Fig. 3A). As oxalate concentration is increased, however, it becomes an inhibitor. When the influence of oxalate on Fru-1,6-Pz binding was examined, it was found to enhance binding of Fru-1,6-Pz at subsaturating 20

‘A ’

1

#

16

I

I I

2

3 I/[PEP]

4

5

mM-’

FIG. 2. Inhibition of liver pyruvate kinase by oxalate. The kinetic activity of pyruvate kinase was determined at 3.0 rnM ADP, 50 nrd Fru-1,6-P*, 11 mM MgS04, and varying concentrations of P-enolpyruvate (PEP) and oxalate (none, 0.25, 0.63, and 1.89 m?d) as indicated. Sodium oxalate was added with equal molar additions of MgSO, to maintain the free Me concentrations constant. The insert shows the replot of the slope at each oxalate concentration, used to estimate Kc.

WALKER

0.2 [OXALATE]

75

0.4 mM

[OXALATE]

mM

FIG. 3. Influence of oxalate on activity and Fru-l,BPe binding. (A) Pyruvate kinase activity was determined at 0.9 mM P-enolpyruvate, 3.0 mM ADP, 11 mM MgSO,, and indicated concentrations of oxalate in the absence and presence of 50 WM Fru-1,6-Ps (FBP). Sodium oxalate was added with equal molar amounts of MgSOl to maintain free M$+ concentrations constant. (B) The binding of Fru-1,6-P, (FBP) with pyruvate kinase was determined at various concentrations of oxalate, as indicated. The binding was conducted in the standard binding buffer containing 11 mM MgSO, (see Materials and Methods), and sodium oxalate was added with equal molar additions of MgSOI to maintain the free Mgs+ concentrations.

concentrations of the activator (Fig. 3B) by shifting the saturation curve to the left as does P-enolpyruvate (not shown). A complete analysis of Fru-1,6-Pz binding in the presence of 0.5 mM oxalate revealed that oxalate decreases [FBPll,z with no marked change in maximal binding of Fru1,6-Pz and only a slight decrease in nn (Table I). These findings suggest that oxalate binds to the active site and brings about cooperative conformational changes similar to those elicited by P-enolpyruvate. InfZuence of Alanine on Fru-1,6-P, binding. Alanine is a potent allosteric inhibitor of liver pyruvate kinase (3-6). The amino acid decreases the apparent affinity of the enzyme for P-enolpyruvate determined kinetically, and Fru-1,6-Pz will overcome the inhibitory actions of alanine (3). As shown in Figs. 1 and 4 and Table I, alanine antagonizes the binding of Fru-1,6-Pz with liver pyruvate kinase. Three micromolar alanine raised [FBP]i,a from 3 PM Fru-1,6PZ to greater than 10 PM. The response to alanine of Fru-1,6-Pz binding with pyruvate kinase is dependent both on the concentration of Fru-1,6-Pz and the concentration of alanine (Fig. 4). Concentrations of alanine as low as 0.5 InM significantly

LIVER

[L-ALANINE]

PYRUVATE

mM

FIG. 4. Inhibition of Fru-1,6-P* binding by alanine. The binding of Fru-1,6-P, (FBP) was determined in the standard Tris-KCl-Mg buffer at 2 and 5 PM total concentrations of Fru-1,6-P, and varying concentrations of L-alanine, as shown.

diminish Fru-1,6-Pz binding with the enzyme. We have consistently noted, however, that very high concentrations of alanine (5 to 10 InM) are unable to completely abolish binding of Fru-1,6-P2 with the enzyme, and that the basal binding of Fru1,6-Pz at high concentrations of alanine is dependent on Fru-1,6-Pa concentrations (Fig. 4). We feel these findings are consistent with an inhibitory allosteric site for alanine which is distinct from the activating Fru-1,6-P2 site. InJIuence of phosphorylation on Fru-1,6Pz binding. Phosphorylation of rat liver pyruvate kinase has been demonstrated to decrease the activity determined in the absence of Fru-1,6-P2 and to increase the concentration of Fru-1,6-P2 needed to fully activate the enzyme (3, 7-9). Phosphorylation also alters binding of Fru-1,6-P2 with the enzyme. In the present investigation, purified rat liver pyruvate kinase was phosphorylated with CAMP-dependent protein kinase to achieve a maximal level of phosphorylation. Phosphorylation was observed to increase [FBP]1,2 from 3 to 5 pM Fru-1,6-P2. The nn for Fru-1,6-P2 binding was observed to decrease from 2.7 to 1.7 as a consequence of phosphorylation (Table I). The phosphorylated enzyme still binds 4 mol Fru-1,6-P&etramer and retains the responsiveness to P-enolpyruvate and alanine (Table I).

207

KINASE

Ir&ence of divalent metal ions on kinetic activity and Fru-1,6-Ps binding. Divalent metal ions are essential for pyruvate kinase activity. Substituting 6 mM Mn2+ for 11 mM Mga+ (saturating concentrations for either ion, see below) in the standard assay buffer has complex consequences on the kinetic properties of the enzyme. First, the V,, in the presence of Mn2+ is approximately 50% of the V,,, found with M$+ (Table III). The divalent metal ion also influences the apparent affinity of the enzyme for P-enolpyruvate. In the presence of Mg2+ [PEP]1,2 is approximately 1 mM in the absence of Fru-1,6-P2, and addition of the activator decreases that value to approximately 0.1 mM. However, with Mn2+ as the sole divalent metal ion, [PEP&/a in the absence of Fru-1,6-P2 is 0.1 mM and the sugarP does not markedly alter that value (Table III). Thus, in the presence of Mn2+, FruTABLE KINETIC

KINASE

PROPERTIES WITH

III

OF RAT LIVER PYRUVATE

MAGNESIUM

AND Metal

Parameter

Mga’

MANGANESE“

ion M”e+

Vmu

110 & 3 unita/mg

WWm n” (-FBP)

1.10 f 0.10 mrd 1.80 + 0.20

0.12 f 0.01 rnM 1.7 ?I 0.10

46 k 1 unita/mg

U’Wm n” (+FBP)

0.10 f 0.01 rnM 0.94 -+ 0.06

0.08 + 0.01 1.20 * 0.30

W*+hn %

1.80 + 0.50 rn~ 0.95 + 0.08

mM

0.013 f 0.10 rnM 1.10 + 0.10

‘The activity of rat liver pyruvate kinase in the absence and presence of 50 pbd Fru-1,6-P, (FBP) was determined at varying concentrations of P-enolpyruvate and 3.0 mx ADP with 11 mM MgSO, or 6 mrd MnSO, as the sole added divalent cation. The maximum catalytic activity (V,,), concentration of P-enolpyruvate giving one-half maximal activity [PEP&n, and the Hill cosfficient (nu) under various conditions were determined by curve-fitting to the Hill equation as described under Materials and Methods. The response to varying free concentrations of MgSO, or &SO, was examined at 0.9 mM P-enolpyruvate and 3.0 mu ADP. The concentration of free metal ion at which one-half maximal activity is observed (Me*%, was also determined by curve-fitting procedures. The specific activity of the enzyme used for this study was 110 units/mg, determined under standard conditions. The results are presented as the best fit + SE obtained from the curve-fitting analysis.

208

BLAIR

AND

1,6-Pz is not a strong activator of hepatic pyruvate kinase. The response of pyruvate kinase activity to increasing concentrations of free Mgzt or free Mn2+ at 0.9 mM P-enolpyruvate and 3 mM ADP appears to be hyperbolic (nn = 1, Table III and Fig. 5). In the absence of Fru-1,6-Pz , under these conditions, maximal activity is achieved only with relatively high concentrations of free Mgz (Fig. 5A). On the other hand, 50 to 100 PM free Mn2+ is sufficient to achieve maximal activity. Figure 5B shows the dependence of pyruvate kinase activity on increasing concentrations of Mn2+ to 0.5 mM. Raising free Mn2+ to as high as 7.5 mM had no further consequences on the kinetic activity from that observed at 0.5 InM Mn2+ (not shown). The apparent difference in the response of pyruvate kinase activity to varying concentrations of Mg2+ or Mn2+, as shown in Fig. 5, are somewhat complex since the V,,, and [PEP&l2 for the enzyme differs with the two metal ions. For example, under the conditions presented in Fig. 5 (0.9 mM P-enolpyruvate), P-enolpyruvate concentrations are saturating in

[ktg”]

rnM

[htn”]

rnM

FIG. 5. Dependence of pyruvate kinase activity on Mge’ or Mn’+. The activity of pyruvate kinase was determined at 0.9 mM P-enolpyruvate and 3.0 mM ADP in the absence and presence of 56 PM Fru-1,6-P* (FBP) and increasing concentrations of MgSO, or MnSO,, as indicated. The concentrations of Mgr+ and Mna’ represent the free concentrations calculated from the total metal ion concentration and the reported dissociation constants for the metal ion-ADP complex (see Materials and Methods). The activity is presented relative to that observed at 10 mM MgSO, in the presence of Fru-1,6-Pa. The specific activity of the enzyme preparation used in this experiment was 150 units/ ma

WALKER

0

2

4

0

2 extra

4 [Me]

0

2

4

mM

FIG. 6. Interaction of MnZf with pyruvate kinase in the presence of 11 mM Mp. Pyruvate kinase activity was determined at 3.0 mM ADP and varying concentrations of P-enolpyruvate (PEP) as indicated in (A) through (C). The standard assay medium contained 11 mM MgSO, in addition to the indicated extra addition of MgSO, or MnSOl (Me). Each condition was examined in the absence (-) and presence (+) of 50 PM Fru-1,6-Pa (FBP) as indicated.

the presence of Mn2+ but not with Me in the absence of Fru-l,&P2. Despite this complexity, those studies suggest that very low concentrations of Mn2+ may have unique consequences on the kinetic properties of the enzyme. To further explore the consequences of Mn2+ on the enzyme kinetic properties, studies were carried out in which Mn2+ was added in the presence of high concentrations (11 InM) Mgzf (Fig. 6). In the presence of Mp, the response to Mn2+ was found to be dependent both on the concentration of P-enolpyruvate and Fru-1,6-P2. At 0.28 mM P-enolpyruvate (Fig. 6A) Mn2+ stimulates pyruvate kinase activity in the absence of Fru-1,6-P2, but inhibits the activity in the presence of the activator. At 2.8 mM P-enolpyruvate (Fig. 6C), Mn2+ is an inhibitor of the enzyme activity, even in the absence of Fru-1,6-P2. Equivalent additions of MgSOl had no influence, indicating the Mn2+ effect is not simply due to changes in ionic strength or sulfate concentrations. The observation that Mn2+ causes a decrease in [PEP]1j2 similar to that elicited by Fru-1,6-P2 might suggest that Mn2’

LIVER

PYRLJVATE

causes a conformational change in the enzyme similar to that elicited by Fru-1,6Pz. To test that possibility, we examined the influence of divalent metal ions on Fru1,6-Pz binding. We previously demonstrated that Fru-1,6-Pz binding to liver pyruvate kinase is observed in a dilute (10 mM) phosphate buffer in the absence of any divalent metal ion (see Ref. (9) and Table I). In 10 mM phosphate at pH 7.0, (FBPh,z is approximately 1 j&M, a value lower than that observed for the Tris-KClMg buffer used in the studies reported here. Addition of 5 mM Mgz+ with 1.5 to 2.4 pM of Fru-1,6-Pz in the phosphate buffer at pH 7.0 had little influence on Fru-1,6-Pz binding; however, at pH 7.4 a small inhibition of Fru-1,6-Pz binding was observed with 5 mM M$+ (Table IV). In the Tris-KC1 buffer, however, both 5 mM Mgzf and 5 mM Mn2+ decrease binding at subsaturating concentrations of Fru-1,6-Pz with the enzyme at pHs 7.0 and 7.4. Raising the pH from 7.0 to 7.4 in the Tris-KC1 medium was found to be accompanied by an increase in binding at subsaturating concentrations of Fru-1,6-P, (Table IV). The influence of the divalent metal ions on Fru1,6-P2 binding was not due simply to the concomitant addition of sulfate in these experiments, since equivalent addition of N&SO4 was found to have no influence. Interpretation of the influence of pH and metal ions on Fru-1,6-P, binding with pyruvate kinase in these studies is complicated in that several species of the sugarP are present within the pH range examined. The acid dissociation constants (Kk and K!!) are dependent on ionic strength, and the apparent dissociation constants for magnesium salts of Fru-l,BPz depend on the ionic forms of Fru-1,6P2 (25). McGilvery (25) reported the Kd for MgFru-1,6-PZ4- is 2 X 10e3 M-‘, and the Kd for MgHFru-1,6-P;is 6.6 X 10e3 M-‘. The acid dissociation constants for Fru-1,6-Pz extrapolated to zero ionic strength are pK3 = 6.43 and pK4 = 7.23 (Ref. (25)). Using this information as a rough guide, one can estimate that, upon raising Mti+ concentrations to 10 mM, between 60 and 30% of the Fru-1,6-Pz will become complexed with magnesium. It is possible that the inhi-

209

KINASE TABLE

IV

INFLUENCE OF DIVALENT METAL IONS OF Fru-1,6-P, BINDING LI

Buffer

pH

Addition

Fru-1,6-P, bound (W control)

7.0 7.0

None (control) 5 mat MgSO,

100 (2052 epm) 98

pi

7.4

None (control) 5 mM MgSO,

100 (2018 cpm) 75

Tris-KC1

7.0

None (control) 5 mM MgSO, 5 nm MnSO,

100 (1073 cpm) 72 35

T&-KC1

7.4

None (control) 5 mre MgSO, 10 mM MgSO, 1 mad MnSO, 5 mbI MnSO, 5 mM NagSO, 10 rn~ N&O, 3 mrd PEP” 3 m&r PEP’ + 10 ml4 MgSO, 3 m&r PEPb + 1 mM MnSO,

100 (2939 cpm) 75 42 97 56 93 95 194 79 171

n Binding of Fru-l,6Pe was carried out as indicated in Table II, except that the standard binding buffer ~88 replaced with divalent metal ion-free 10 mbr phosphate (Pi) or 150 mM TrisHCI plus 70 mbr KC1 (Tris-KCI), each containing 1 mbf Dl”T and at the indicated pH. Results represent the averages of duplicate determinations which were confirmed in at least one additional experiment. For comparison of the influence of various buffer systems on Fru-1,6-Pz binding, the amount of radioactive Fru-1,6-P* bound in each control is presented. b PEP = P-enolpyruvate.

bition of Fru-1,6-P, binding with pyruvate kinase by high concentrations of Mgzf or Mn2+ (Table IV) is due to formation of Me2+Fru-1,6-P2, which binds weakly or not at all compared to uncomplexed forms of the sugar-P. This interpretation is supported by the observation that Mgz+ inhibition of Fru-1,6-P, to pyruvate kinase in the 10 mM phosphate buffer is enhanced upon raising the pH from 7.0 to 7.4 (Table IV), since raising the pH in this range will enhance formation of MgFru-1,6-P2. That interpretation may be oversimplified, however, in that raising the pH from 7.0 to 7.4 in the Tris-KC1 buffer (Table IV) had little influence on the inhibition of binding by Mga+. A greater understanding of the

210

BLAIR

AND

interaction of various ionic species of Fru1,6-Pz will require more detailed studies. It should be emphasized, however, that the ability for either M@;2+or Mn2+ to influence Fru-l,&Pz binding with pyruvate kinase requires relatively high concentrations of the metal ions compared to those needed to support the catalytic activity of the enzyme (Fig. 5). Although divalent metal ions are required for catalytic activity, the present investigation indicates that interaction of P-enolpyruvate with pyruvate kinase does not require the divalent metal ion. This conclusion is supported by the results presented in Table IV, showing that the ability for P-enolpyruvate to enhance Fru-l,&Pz binding with the enzyme can be demonstrated in the absence of any divalent metal ion. Detailed studies on the influence of metal ions at various subsaturating concentrations of P-enolpyruvate, however, were not carried out in this study. DISCUSSION

The potent influence of Fru-1,6-Pa on the kinetic properties of liver pyruvate kinase from several species is now well characterized. However, reports concerning the direct binding of Fru-1,6-Pz with the enzyme are limited. Early investigations with the rabbit liver enzyme indicated that Fru1,6-PZ binds to a single site per enzyme tetramer, with no cooperativity being observed (10). More detailed studies with the pig liver enzyme indicated up to four Fru1,6-Pz binding sites per tetramer, and demonstrated that P-enolpyruvate enhances binding of the allosteric activator (11, 12). More recent studies applying the nitrocellulose binding procedure for studying binding have obtained results consistent with four binding sites per tetramer of the rat liver enzyme, and have indicated positive cooperativity in that binding of Fru-1,6-Pz (9, 13). The present findings also support a strong positive cooperativity in Fru-l,&Pz binding with rat liver pyruvate kinase, since the Hill coefficient for binding is 2.7 or greater (Table I, Fig. 1). No evidence for negative cooperativity was found (see Results), which is

WALKER

not consistent with the recent report by El-Maghrabi and co-workers (13). Those investigators found positive cooperativity for binding of Fru-1,6-P:! at low concentrations of the ligand. However, at high concentrations of Fru-1,6-Pz, negative cooperativity was detected by a decrease in the fourth intrinsic binding constant when binding was analyzed by the sequential model of Edsall and Wymann (13). The basis for this discrepancy between the present findings and those of El-Maghrabi et al. (13) concerning possible negative cooperativity in the binding of Fru-1,6-Pz is not presently understood. It should be noted, however, that the binding studies were carried out under different conditions by the two groups, and that a number of factors including cation concentrations influence binding of the Fru-1,6-Pz. The presence of alanine and the phosphorylation of pyruvate kinase, both of which diminish the enyzme activity, antagonize binding of low concentrations of Fru-1,6-Pz (Table I and Fig. 4). The influence of alanine on Fru-1,6-Pz binding was anticipated from the known heterotropic interactions of the two ligands with respect to the kinetic properties of the enzyme (l6). The decreased affinity of the enzyme for Fru-1,6-Pz after phosphorylation was also consistent with kinetic observations that phosphorylation raises the concentration of Fru-1,6-Pz needed for full activation (3). It is important to note the broad range in apparent affinity of pyruvate kinase for Fru-1,6-Pz which can be generated by appropriate manipulation of the phosphorylation state and effector concentrations. For example, nearly an l&fold difference in [FBP]l,B is noted beween the dephosphorylated form of the enzyme in the presence of P-enolpyruvate from that for the phosphorylated form in the presence of alanine (Table I). The [PEP]l,B for hepatic pyruvate kinase in the absence of Fru-1,6Pz is considerably higher than physiological concentrations of P-enolpyruvate in the liver. It might be suggested that, physiologically, the enzyme activity is absolutely dependent on Fru-1,6-Pe. Since alanine and phosphorylation greatly decrease the apparent affinity of the enzyme

LIVER

PYRUVATE

for Fru-l,&Pz, an increase in the phosphorylation state accompanied by uptake of alanine by the liver during gluconeogenic conditions may serve as an effective means for preventing activation of pyruvate kinase by physiological concentrations of Fru-1,6-Pz when P-enolpyruvate is needed for glucose synthesis. The present findings substantiate previous reports (12,13) that the three-carbon substrate, P-enolpyruvate, enhances the binding of low concentrations of Fru-1,6Pz. This positive influence of P-enolpyruvate is consistent with the kinetically observed heterotropic interactions of Penolpyruvate and Fru-1,6-P,. The present investigation extends previous reports to show that other three-carbon carboxylic acids which bind at the active site also enhance the binding of Fru-1,6-Pz. Oxalate, which has been recognized as a potent competitive inhibitor of pyruvate kinase from other sources (23,24), provides a particularly interesting example. In the presence of high concentrations of Fru-1,6-Pz, oxalate is a straightforward competitive inhibitor of the rat liver enzyme (Fig. 2). In the absence of Fru-1,6-Pz, however, oxalate may actually stimulate the enzyme activity at subsaturating concentrations of P-enolpyruvate (Fig. 3). The interpretation of this observation is that binding of oxalate to one subunit of the tetramer brings about a cooperative conformational change in adjoining subunits, increasing the affinity for P-enolpyruvate (and oxalate) and hence increasing kinetic activity at low, subsaturating concentrations of the two ligands. In the presence of Fru-1,6-Pz, however, the enzyme apparently exists completely in an active conformation and oxalate acts simply as a competitive inhibitor. Such an activation by competitive inhibitors has been kinetically observed for other enzymes as well (26). The present demonstration that oxalate enhances the binding of Fru-l,&Pz with liver pyruvate kinase and increases the apparent affinity for the activator provides direct evidence that it brings about a conformational change which is transmitted through the subunits. Interestingly, the product of the pyruvate kinase reaction, pyruvate, also

KINASE

211

enhances Fru-1,6-Pz binding with the enzyme (Table II). The kinetic consequences of pyruvate at subsaturating concentrations of P-enolpyruvate was not determined in the present study, since we routinely assay pyruvate kinase activity by coupling the formation of pyruvate to the oxidation of NADH using lactate dehydrogenase. However, if pyruvate acts similarly to oxalate in competing at the active site, it is possible that conditions will be found in which the product (pyruvate) actually stimulates pyruvate formation from P-enolpyruvate. This possibility remains to be tested in an appropriate assay system. The potential physiological significance of pyruvate activation of liver pyruvate kinase remains to be established; however, it is interesting to note that administration of high concentrations of pyruvate to isolated rat hepatocytes is accompanied by a high rate of “futile” recycling of P-enolpyruvate to pyruvate through pyruvate kinase (27). It is possible that the high concentrations of pyruvate used to attain such recycling result in activation of pyruvate kinase by pyruvate itself. The latter suggestion, however, can only remain speculation at this time; but, the consequences of pyruvate on the kinetic activity of the enzyme remains an interesting topic for future investigation. The influence of the nucleotide substrates ADP and ATP on Fru-1,6-P, binding is somewhat more complex. In agreement with El-Maghrabi et al. (13), we find that the nucleotides alone have little influence on Fru-1,6-Pz binding (Table II). However, the present study extends past studies to show that both nucleotides antagonize the ability for the three-carbon carboxylic acids to cause enhanced binding of the activator (Table II). A full interpretation of the consequences of ADP and ATP interaction with the liver enzyme may not be possible at this time. It does seem clear, however, that, in the absence of the three-carbon substrate, neither ATP nor ADP preferentially induce a particular conformational state of the enzyme (since Fru-1,6-Pz binding is not influenced). However, the formation of a ternary complex

212

BLAIR

AND

of the enzyme with both substrates may favor the conformational state having a low affinity for Fru-1,6-Pz. It is also possible that binding of the substrates is ordered, and that ATP or ADP binding first requires binding of the carboxylic acid. It is interesting to note that similar findings concerning the influence of ATP and ADP on the conformation of the muscle isozyme have been reported from NMR studies (23). For that isozyme ATP and ADP alone had small effects, but they antagonized the conformational changes about the active site elicited by substrate analogs such as oxalate and P-lactate. Although more study is warranted in this area, it is clear that ATP does not have the strong inhibitory actions on Fru-1,6-Pe binding as does alanine (Table II and Figs. 1 and 3). Both alanine and ATP have been considered allosteric inhibitors of liver pyruvate kinase. The present findings concerning ATP actions on the enzyme, however, are more in agreement with the recent suggestion by El-Maghrabi et al. (13) that ATP inhibits the enzyme activity by binding only to the active site and not at a distinct allosteric site. It is well established that pyruvate kinase from most sources requires both a divalent and a monovalent cation for activity. The importance of divalent metal ions to the regulation of the enzyme by Fru-1,6-Pz was investigated in this study. Concentrations of M$+ in the millimolar range are required to support activity of the enzyme; however, no added divalent metal ion was found necessary for binding of Fru-1,6-Pz with the enzyme nor for the ability for P-enolpyruvate to enhance Fru1,6-Pz binding. In fact, high concentrations of both MS+ or Mn2+ antagonize Fru-1,6Pz binding (Table IV). The influence of divalent metal ions on Fru-1,6-Pe binding may be complex since the metal ion may bind both with the enzyme and the hexose-P. One interpretation of the inhibition of Fru-l,&Pz binding by high concentrations of both Mti+ and Mn2+ is that the metal ion-Fru-1,6-P2 complex has a weak affinity for the enzyme compared with free Fru-1,6-P2. A full evaluation of that possibility will require more

WALKER

detailed studies on the metal ion complexes with Fru-l,&P2 under the various conditions applied in this investigation. Despite that uncertainty, it is clear that the inhibition of Fru-1,6-Pz binding with pyruvate kinase by high concentrations of divalent metal ions is distinct from the kinetic consequences of the metal ions. Whereas high concentrations of both Mgz+ and Mn2+ inhibit Fru-1,6-P2 binding (Table IV), very low concentrations of Mn2+ have a specific and pronounced effect on the kinetic properties of the enzyme, even in the presence of M@;2+(Fig. 6). The substitution of Mn2+ for Mg2 results in an increase in the apparent affinity for P-enolpyruvate and a decrease in the maximal catalytic activity. Thus, even in the presence of saturating concentrations of Me, low concentrations of Mn2+ stimulate pyruvate kinase activity at subsaturating concentrations of P-enolpyruvate due to the increase in apparent affinity of the enzyme for Penolpyruvate. However, at saturating concentrations of P-enolpyruvate, Mn2+ is inhibitory due to the V,, effect. This observation concerning the rat liver enzyme is similar to that observed with the pigeon liver enzyme by Gabrielli and Baldi (29). The present investigation extends their observations to show, however, that Mn2+ is inhibitory in the presence of Fru-1,6-P2 even when saturating concentrations of Mgz+ are present. The consequences of Frul,6-Pz addition on the enzyme activity are also small when Mn2+ is the sole divalent metal ion. The interpretation of these observations is that the major stimulatory actions of Mn2+ at subsaturating P-enolpyruvate concentrations in the absence of Fru-1,6-P2 is due to a decrease in the apparent K, for P-enolpyruvate. In the presence of Fru-1,6-P,, however, Mn2+ does not support as high a catalytic activity at saturating substrate levels as does M$+. The reason that Mn2+ does not support a high catalytic activity is not presently known. The present findings support a dual role for divalent metal ions in the catalytic mechanism, as has been suggested by others for the muscle isozyme (30-32). The present findings might further suggest that one role of the divalent metal ion involves

LIVER

PYRUVATE

the interaction of P-enolpyruvate with the enzyme (Table III). Since divalent metal ions are not required for the P-enolpyruvate induced conformational change (as judged by Fru-1,6-P, binding), the divalent metal ion influence on the kinetic response to varying concentrations of P-enolpyruvate (Table III) may require participation of a second metal ion-nucleotide complex. Further studies into the influence of divalent metal ions on P-enolpyruvate binding may be necessary to further resolve this problem. ACKNOWLEDGMENTS The authors wish to thank Dr. Mary Wimmer, Dr. C. Larry Harris, and Dr. Tom Nowak for discussions and suggestions during preparation of this manuscript.

213

KINASE

C. (1971) Hoppc-Se&d Chem 352,453-458. 12. KUTZBACH, C., BISCHOFBERGER, H., HESS, B., AND ZIMMERMANN-TELSCHOW, H. (1973) Hoppe-Seyler’s 2. Physiol Chem 354.1473-1489.

11. HESS, B., AND KUTZBACH,

2. Physid

13. EL-MAGHRABI, M. R., CLAUS, T. H., MCGRANE, M. M., AND PILKIS, S. J. (1982) J. Biol ‘%em. 257,233~240. 14. BLAIR, J. B., AND WALKER, R. G. (1983) Fed Proc 42, 2081. 15. HARADA, K., SAHEKI, S., WADA, K., AND TANAKA, T. (1978) B&him Biophys. A& 524,327~389. 16. STULL, J. T., AND Buss, J. E. (1977) J. Biol. Chem 252, 851-857. 17. BLAIR, J. B., CIMBALA, M. A., FOSTER, J. L., AND MORGAN, R. A. (1976) J. Biol Chem 251,37563762. 18. KNACK, I., AND ROHM, K.-H. (1981) Hoppdeyler’s 2. Physid Chem 362, 1119-1130. 19. EDSALL, J. T., AND WYMAN, J. (1958) Biophysical Chemistry, Vol. 1, Academic Press, New York. 20. PONTREMOLI, S., GRAZI, E., AND ACCORSI, A. (1968)

Biochemistry REFERENCES 1. TANAKA, T., HARANO, Y., SUE, F., AND MORIMURA, H. (1967) J. Biochem (Tokyo) 62,71-91. 2. TANAKA, T., FIJMIAKI, S., AND MORIMLJRA, H. (1967)

Biochem. Biophys. Res. Commun 29,444-449. 3. BLAIR, J. B. (1980) in The Regulation of Carbohydrate Formation and Utilization in Mammals (Venezaile, C. M., ed.), pp. 121-151, Univ. Park Press, Baltimore. 4. HALL, E. R., AND COTTAM, G. L. (1978) Int. J.

Biochem 9, 785-793. 5. SEUBERT, W., AND SCHONER,

Curr.

P., MARCO,

J. Biochem

25.

Top.

Cell Reg. 3, 237-267. 6. LORENTE,

B&hem

R., AND SOLS, A. (1970)

27.

Eur.

13,45-54.

7. ENGSTROM, L. (1978) Curr. Top. Cell. Reg. 13, 2951. 8. EL-MAGHRABI, M. R., HASTON, W. S., FLOCKHART, D. A.,, CLAUS, T. H., AND PILKIS, S. J. (1980) J.

Biol. Ckem. 255, 668-675. 9. BLAIR, J. B., CIMBALA, M. A., AND JAMES, M. A. (1982) J. Biol. Chem 257, 7595-7602. 10. IRVING, M. G., AND WILLIAMS, J. F. (1973) Biochem J. 131,303-313.

Biophys. Res. Ccnnmun. 85,774-779.

24. But,

26. W. (1971)

7,1655-1661.

21. MILDVAN, A. S., AND COHN, M. (1966) J. BioL Ch.enz. 241, 1178-1193. 22. O’SULLIVAN, W. J., AND SYITHERS, G. W. (1979) in Methods in Enzymology (Purich, D. L., ed.), Vol. 63, pp. 257-294, Academic Press, New York. 23. But, H., DEMAUGRE, F., AND LEROUX, J.-P. (1978)

28. 29. 30.

H. A., DEMAUGRE, F., MONCION, A., AND LERoux, J.-P. (1981) Biochimie 63, 595-602. MCGILVERY, R. W. (1965) Biochemistry 4, 19241930. MONOD, J., WYMANN, J., AND CHANGEUX, J. P. (1965) J. Mol. Biol 12, 88-118. ROGNSTAD, R. (1975) Biochem. Biophys. Res. Cornmun 63, 900-905. NOWAK, T. (1978) .I Biol. Chem 253, 1998-2004. GABRIELLI, F., AND BALDI, S. (1973) Experimsntia 29, 1342-1343. NOWAK, T., AND SUELTER, C. (1981) Mol. Ce.!.!

Bioch.em. 35, 65-75. 31. KAYNE, F. J. (1973) in The Enzymes (Boyer, P., ed.) Vol. 8, pp. 353-382, Academic Press, New York. 32. BAEK, Y. H., AND NOWAK, T. (1982) Arch Biochem

Biophys. 217,491-497.