THE JOURNAL OF BIOLOGICAL CHEMWI’HY

Vol. 251. No. 20. Issue of October 25, pp. 6244-6251. 1976 Printed in U.S.A.

Differences in Phosphofructokinase Regulation in Normal and Tumor Rat Thyroid Cells

(Received for publication, May 13, 1976)

MARIA F. MELDOLES,* VINCENZO MACCHIA, AND PAOLO LACCETTI

From the Centro di Endocrinologia ed Oncologia Sperimentale de1 Consiglio Nazionale delle Ricerche, Istituto di Patologia Generale dell’Universita di Napoli, II Facolta di Medicina, 80131, Naples, Italy

The kinetic and molecular properties of a phosphofructokinase derived from a transplantable rat thyroid tumor lacking regulatory control on the glycolytic pathway were studied. The properties of the near-purified enzyme (specific activity 140 units/mg) were compared with those of phosphofructokinase from normal rat thyroid (specific activity 134 units/mg). The electrophoretic mobilities and gel elution behavior of these two enzymes were almost similar. The thyroid tumor phosphofructokinase showed, however, a greater degree of size and/or shape heterogeneity in the presence of ATP than the normal thyroid enzyme, as determined by gel filtration and sucrose density gradient centrifugation.

Kinetic studies below pH 7.4 showed a sigmoid response curve for both enzymes when the velocity was determined at 1 mM ATP with varying levels of fructose-6-P. The interaction coefficient, however, was 4.2 and 2.6 for normal and tumor thyroid phosphofructokinase, respectively. Ammonium sulfate decreased the cooperative interactions with the substrate fructose-6-P in both enzymes. The thyroid tumor enzyme, however, was less sensitive to the inhibition by ATP and by citrate. The reversal of citrate inhibition by cyclic 3’:5’-adenosine monophosphate was also less effective with the thyroid tumor phosphofructokinase, while the protective effect of fructose-6-P was stronger. The difference in citrate inhibition between tumor and normal thyroid enzyme was not strongly affected by varying the MgCl* concentration up to 10 mM.

It is concluded that the complex allosteric regulation typical of the normal thyroid phosphofructokinase is still present in the enzyme isolated from the thyroid tumor tissue. The latter, however, is more loosely controlled by its physiological effecters, such as ATP, citrate, and cyclic AMP.

In normal tissues the regulation of the initial sequence of reactions of the glycolytic pathway is dependent on the control exerted by phosphofructokinase (ATP:D-fructose-6-P phospho- transferase, EC 2.7.1.11), an enzyme which catalyzes a unidi- rectional step and has a rate-limiting function (1, 2). In addition, normal mammalian phosphofructokinases exhibit complex allosteric properties: they are inhibited by citrate and by high concentrations of one of their substrates, ATP; this inhibition is relieved by the second substrate, fructose-g-p, or by 5’-AMP, cyclic AMP, fructose-1,6-P,, inorganic phosphate, or ammonium ions (3, 4). These latter allosteric properties led Blangy et al. (5) to propose that phosphofructokinase partici- pated in an important glycolytic control mechanism.

The present report compares the kinetic properties of a normal phosphofructokinase which has a rate-limiting function in the glycolytic pathway, with the kinetic properties of a phosphofructokinase lacking this function, in an effort to provide insight into the regulatory properties of the enzyme. In a previous study of a transplantable rat tumor (line l-8 in

*At present, Visiting Scientist in the Clinical Endocrinology Branch, National Institute of Arthritis, Metabolism, and Digestive Diseases, National Institutes of Health, Bethesda, Maryland 20014.

Wollman’s classification) (6), we showed that the regulation of the rate-limiting step catalyzed by phosphofructokinase was modified in the thyroid tumor with respect to the normal thyroid extracts. Our evidence for this was the concurrent existence of a slight enhancement of phosphofructokinase activity, a strong enhancement of lactate and pyruvate produc- tion .from fructose-6-P to levels obtained from fructose-1,6-P*, and an apparent lack of inhibition of lactate production from glucose-6-P by high concentrations of ATP. Since the slight increase in phosphofructokinase activity did not seem to be consistent with the absence of the pacemaker function by phosphofructokinase in the thyroid tumor, it seemed that this alteration might also be related to a modification of the allosteric properties of the thyroid tumor enzyme. This paper thus describes the purification and the molecular and kinetic properties of phosphofructokinase from normal and tumor rat thyroid cells.

MATERIALS AND METHODS

Microgranular DEAE-cellulose (DE52) was purchased from What- man, Sephadex G-200 and Sepharose 4B from Pharmacia, dextran blue 2000 from Serva, acrylamide and NJ’-methylenebisacrylamide from Eastman Kodak, ammonium sulfate (enzymic grade) from BDH, sucrose (ribonuclease-free) from Serva, and bovine serum albumin

6244

by guest on February 25, 2020http://w

ww

.jbc.org/D

ownloaded from

Phosphofructokinase Regulation in a Thyroid Tumor 6245

from Sigma. The nucleotides, substrates of the glycolytic pathway, auxiliary enzymes, and rabbit muscle pyruvate kinase were from Boehringer. All other chemicals were of reagent grade.

[‘*51]Thyroglobulin from rat thyroid (19 S, molecular weight 660,000) was a gift from Professor G. Salvatore (Istituto di Patologia Generale, II FacoltB di Medicina, Universita di Napoli, Italy). Rat muscle phosphofructokinase was purified using the procedure of Kemp and Forest (7).

The thyroid tumor line employed throughout these studies was one of a series of thyroid tumors developed in Fischer rats by Wollman and designated as line l-8 (8). The tumor was kindly supplied by Dr. S. H. Wollman (National Cancer Institute, National Institutes of Health, Bethesda, Md.). All of the tumors used were derived from a single tumor which was carried by subcutaneous implantation in Fischer rats and had a growth rate of about 60 days. Normal thyroid glands were obtained from Fischer male rats from Charles River Breeding Labora- tories, Inc., Wilmington, Mass.

Assay of Phosphofructokinase Activity

Phosphofructokinase activity was determined by measuring the rate of fructose-1,6-P* formed at pH 8.2 and at 22” in 1 ml of medium using conditions described by Kemp (9), including both 6 mM MgCl, and 5 rnM ammonium sulfate. One unit of enzyme is defined as that amount of phosphofructokinase which converts 1 rmol of fructose-6-P to fructose-1,6-P, in 1 min under the above conditions and in the presence of 1 mM fructose-6-P and 1 mM ATP. NADH oxidase activity was as- sayed in the same reaction mixture but in the absence of fructose-6-P or ATP and was subtracted from the phosphofructokinase activity.

In all experiments comparing the kinetic properties of normal and tumor thyroid phosphofructokinase, an amount of enzyme was added in each assay that would give a velocity of 0.01 rmol of fructose-1,6-P? formed per min per ml of the standard assay mixture; this is defined as V, (9). For each set of kinetic studies, normal and thyroid tumor enzymes were purified and analyzed at the same time. Experiments at pH 7.0 were in the same buffer except that fructose&P, ATP, MgCL, and ammonium sulfate concentrations were varied as described in the legends. Experiments at pH 7.2 (citrate inhibition) were in the presence of 35 mM triethanolamine buffer, pH 7.2, 0.1 mM dithio- threitol, 0.2 rnM NADH, 1.0 rnM ATP, aldolase (0.2 unit/ml), triose- phosphate isomerase (0.1 unit/ml), glycerol-l-P dehydrogenase (0.1 unit/ml), fructose-&P, MgCl*, and other additions stated in the legends to the figures.

All auxiliary enzymes were extensively dialyzed before use to remove ammonium sulfate. The reaction was initiated by addition of phospho- fructokinase; however, in the kinetic experiments at low pH, fruc- tose-6-P was added last, after an equilibration period of 3 min. Potential inhibitors, where used, were added after the addition of ATP. Rates were determined 4 to 6 min after starting the reaction by addition of fructose-6-P, unless otherwise stated. For the experiments which were run in the absence of ammonium sulfate or in the presence of 0.3 to 3.0 mM concentrations of this salt, normal and tumor thyrdid enzymes preparations from Step 3 of the purification procedure were applied to a 5 to 20% sucrose density gradient. The peak of activity was collected after 10 h of centrifugation and used for kinetic experiments with or without previous dialysis.

Protein determination was performed by the Lowry method (10) with crystalline serum albumin as standard or by ultraviolet absorb- ance.

Sucrose Density Gradient Studies

Ultracentrifugation in a sucrose gradient was performed according to the method of Martin and Ames (11). The sucrose gradient (5 to 20% or 10 to 40%) contained 50 mM Tris/phosphate buffer, pH 7.7, 1 mM MgCl,, 0.1 mM EDTA, 2.5 mM mercaptoethanol, and 2.5 rnM ATP. In each run 0.6 unit of enzyme as used. Centrifugation was carried out at 24,000 rpm at 5” for periods of 10 h (5 to 20% sucrose gradient) or 38 h (10 to 40% sucrose gradient) in a Beckman Spinco model L2-75-B preparative ultracentrifuge with an SW 27 rotor. [‘251]Thyroglobulin (19 S) purified from rat thyroid glands (12, 13) and 10 S rabbit muscle pyruvate kinase were used as reference standards in all runs. After centrifugation, fractions of 40 drops were collected by puncturing the bottom of the tubes. The labeled peak of 19 S [l*SI]thyroglobulin was measured using a Nuclear Chicago counter. Pyruvate kinase activity was assayed according to Valentine and Tanaka (14). The sedimenta- tion coefficient of normal and tumor thyroid phosphofructokinase was estimated by comparing their sedimentation rate with that of the markers (11).

Separation of Isozymes of Phosphofructokinase by Chromatography on DEAE-Cellulose Column

It is well known that isozymes of phosphofructokinase derived from rabbit tissues elute differently on DEAE-cellulose columns. Thus, rabbit muscle, brain, and liver phosphofructokinase are eluted by 0.03. 0.15, and 0.5 M ammonium sulfate, respectively (15, 16), and rat muscle and liver phosphofructokinase are eluted by 0.028 and between 0.2 and 0.6 M ammonium sulfate, respectively.’ To evaluate the isozymes of the rat thyroid and the rat thyroid tumor, 2 to 3 units of phosphofructokinase in 0.05 M Tris/phosphate buffer, pH 7.8, contain- ing 0.2 rnM ATP, 5 mM mercaptoethanol, and 0.2 rnM EDTA were applied to a column of DEAE-cellulose (0.9 x 13 cm) equilibrated with the same buffer (17). The enzymes were eluted with a linear gradient of ammonium sulfate 0 to 0.15 M in 0.05 M Tris/phosphate, pH 7.8, 0.2 mM

ATP, 5 mM mercaptoethanol, and 0.2 mM EDTA. The gradients were formed by mixing 100 ml each of the low salt and high salt buffer. Fractions of 1 ml were collected and analyzed for both phosphofruc- tokinase activity and conductivity.

Electrophoresis of Phosphofructokinase

Electrophoresis was performed at 4” on B-inch cellulose acetate strips (Gelman Instrument Co.) in a medium containing 0.05 M glycylglycine, 5 mM ammonium sulfate, 0.1 mM EDTA, 5 mM mercap- toethanol, and 1 rnM ATP, pH 8.9 (9). The voltage and duration of electrophoresis were varied to obtain the best migration. Following electrophoresis, the enzymatic activity was detected by placing the cellulose acetate strips face down on a thin layer agar gel containing 0.3% Ionagar No. 2, 20 mM Tris/HCl, pH 8.3, 10 mM sodium arsenate, 2 mM EDTA, 1 mM fructose-6-P. 1 mM ATP, 4 mM MgCl,, 1 mM NAD’, 0.4 mg/ml of nitroblue tetrazolium chloride, 0.06 mg/ml of diaphorase, 0.12 mg/ml of glyceraldehyde 3-P dehydrogenase, 0.1 mg/ml of aldolase, and 0.01 mg/ml of triosephosphate isomerase (Me- dium 1). The final pH was 8.3. Fructose-6-P was omitted from the gels used as controls (18).

Polyacrylamide disc gel electrophoresis was performed at 4” in 5% gels (0.5 x 12 cm) with 0.08 M Tris/HCI buffer, pH 8.6, containing 0.1 mM ATP, 0.1 rnhg EDTA, and 2.5 mM mercaptoethanol. Following electrophoresis, the gels were stained for either protein or enzyme ac- tivity. Protein staining was with 1% Amido black in 7% acetic acid. Enzymatic activity on the gels was detected by incubating them in Medium 1 in the absence of Ionagar at 37”. Densitometer tracings of the gels were obtained at 550 in using a Gilford recording spectro- photometer.

Preparation and Purification of Normal and Tumor Rat Thyroid Phosphofructokinase

Step I: Extraction-Immediately after death of the animals by exsanguination, the thyroids and the tumors were rapidly excised and placed in an ice-cold medium, pH 7.5, containing 0.03 M phosphate buffer, 0.2 mM EDTA, and 5 mM mercaptoethanol (referred to as Buffer A). All operations were performed at O-4” unless otherwise noted. Aliquots of the tissues were minced with precooled scissors, washed, and homogenized in a glass-Teflon homogenizer with 1.5 volumes of ice-cold Buffer A. The homogenate was centrifuged at 3,000 x g for 10 min and then at 22,000 x g for 60 min. The supernatant was filtered through cheesecloth and centrifuged at 105,000 x g for 90 min in a Spinco model L-75-B ultracentrifuge with a 40 or 60 Ti rotor, and the supernate was saved.

Step 2: Chromatography on Dextran Blue 2000 Coupled to Poly- acrylamide Gel-Chromatography on dextran blue 2000 coupled to polyacrylamide gel was needed to separate phosphofructokinase from thyroglobulin, which is the main protein component of the thyroid tissue. Several small columns (0.9 x 15 cm) were packed to a height of 6 cm with dextran blue polyacrylamide gel prepared according to BGhme et al. (19) and exhaustively washed with distilled water prior to equilibration with Buffer A; the 105,000 x g supernatant was applied to the columns; and the columns were washed with the same buffer until the 280 nm absorbance of the effluent dropped below 0.1. Phosphofructokinase was then eluted by 5 rnM ATP in Buffer A, and the active fractions were collected. In some instances a batch prepara- tion of blue dextran polyacrylamide gel was used.

Step 3: DEAE-cellulose Chromatography-For this step usually two to five preparations from Step 2 were combined. The eluate from Step 2 was applied to a DEAE-cellulose column previously equilibrated with 50 mM Tris/phosphate containing 0.2 mM EDTA, 5 mM mercaptoetha-

’ M. F. Meldolesi, unpublished results.

by guest on February 25, 2020http://w

ww

.jbc.org/D

ownloaded from

6246 Phosphofructokinase Regulation in a Thyroid Tumor

nol, and 0.1 rn~ ATP, and adjusted to a final pH of 7.8 (referred to as Buffer B). The size of the column corresponded approximately to 1 ml of DEAE-cellulose per mg of protein. The column was then washed with Buffer B, and the enzyme was eluted with 0.1 M ammonium sulfate in Buffer B (referred to as Buffer Cl. The peak tubes containing phosphofructokinase activity were pooled and used for Step 4 or applied to a sucrose gradient after concentration by using Minicon B-15 concentrators (Amicon, Boston, Mass.). Both normal and tumor thyroid phosphofructokinases were completely eluted by 0.1 M ammo- nium sulfate. Stepwise elution of the same column with increasing concentrations of ammonium sulfate from 0.1 to 0.8 M did not elute any other detectable peak of activity.

Step 4: Gel Filtration Chromatography-Fractions from Step 3 were applied to a Sepharose 4B column (0.9 x 30 cm) previously equili- brated with Buffer C. The flow rate was adjusted to 2.5 ml/h. The active fractions were pooled and concentrated if necessary before use for kinetic studies.

The results of a typical preparation of the normal and tumor enzymes are presented in Table I. The specific activities of the nearly purified enzymes were almost the same and approximated those reported by other authors for purified mammalian phosphofructoki- nases.

RESULTS

Physical Properties of Normal and Tumor Phosphofruc- tokinase-Electrophoresis of the 105,000 x R supernatant on cellulose acetate strips showed a single band of activity with similar mobilities for both the normal and tumor phosphofruc- tokinase (Fig. 1).

Polyacrylamide gel electrophoresis of the purified prepara- tions showed a main protein peak which penetrated the gels for a few millimeters only and which was superimposable with enzymatic activity. When normal and tumor thyroid phospho- fructokinase were separately applied to a Sepharose 4B column equilibrated with Buffer D, they both emerged as a single peak, although the tumor phosphofructokinase peak was slightly broader than that of the normal thyroid phosphofructokinase (data not shown).

On DEAE-cellulose columns which separate the isozymes of phosphofructokinase, both the normal and tumor enzymes behaved as muscle isozymes, although slight differences in their elution properties were noted. Thus, the normal and tumor thyroid phosphofructokinases eluted at 0.04 and 0.064 M ammonium sulfate, respectively.

When a 0.6 unit of normal thyroid phosphofructokinase was centrifuged on a sucrose gradient, a single relatively narrow

peak was found with an estimated sedimentation constant of 12.2 + 0.16. When 0.6 unit of tumor phosphofructokinase was centrifuged on a sucrose gradient, a broader peak was obtained with an estimated sedimentation constant of 14.4 S + 0.17 (Fig. 2).

Kinetic Properties of Normal and Tumor Phosphoftuctoki- nuses-The activity of the normal and tumor enzymes was assayed at different pH values. Maximal enzyme activity was between pH 7.5 and 9.0, and only minor differences were observed between the two enzymes (Fig. 3).

At pH 8.2 and in the presence of 5 mM ammonium sulfate, both normal and tumor phosphofructokinase exhibit first order kinetics with respect to ATP and fructose-6-P (4). The Michae- lis-Menten constants for both substrates were calculated from the Lineweaver-Burk plots of the kinetic data. In the presence

of 0.5 mM ATP, the K, value for fructose-6-P for both enzymes was 2.6 x 1O-5 M. In the presence of 0.5 mM fructose-6-P, the K, values for ATP for the normal and thyroid tumor enzymes were 5.3 x 10m5 and 1.2 x lo-’ M, respectively. This difference was constant in various preparations of the enzymes.

Below pH 7.4 (and in the absence of ammonium sulfate), mammalian phosphofructokinases show cooperative kinetics (4); thus, when the velocity was determined at pH 7.2 at a given level of ATP and with varying levels of fructose-6-P, a sigmoid response curve was obtained for both the normal and tumor enzymes (Fig. 4). The V,,, was very low, and the Hill plots of log [u/V,,, ~ u] uersus log fructose-6-P concentration gave straight lines from which could be calculated both the interaction coefficient and the concentration of substrate required to give half-maximal velocity. The Hill coefficient was n = 4.25 for the normal and n = 2.65 for the tumor thyroid phosphofructokinase; the K,., for fructose-6-P was 0.3981 and 0.5821 mM, respectively.

When the velocity was determined at pH 7.0 at a given level of fructose-6-P by varying the concentration of ATP, it became evident that the normal thyroid enzyme was more strongly inhibited by ATP than the tumor thyroid phosphofructokinase (Fig. 5). The concentration of ATP required to reduce the velocity of two-thirds that of the optimum were 2.0 and 3.7 mM for the normal and tumor phosphofructokinases, respectively. These results account for the fact that in the presence of 1 mM ATP, the K,,, for fructose-6-P of phosphofructokinase in

TABLE I

Purification of phosphofructokinase from rat thyroids and from rat thyroid tumor

Tissue step Procedure Total Total Specific protein activitv activitv Yield Purification

A. Rat thyroid w units unitslmg % -fold

Crude extract (105,000 x g 292 26.4 0.090 100 1 supernatant)

Polyacrylamide gel/blue 4.96 18.0 3.63 68 40 dextran

DEAE-cellulose 1.64 18.1 11.04 68 123 Sepharose 4B 0.103 13.9 134.95 53 1,499

B. Rat thyroid tumor

Crude extract (105,000 x g 405 42.5 0.105 100 1 supernatant)

Polyacrylamide gel/blue 9.14 31.9 3.49 75 33 dextran

DEAE-cellulose 2.45 32.1 13.10 75 125 Sepharose 4B 0.180 25.3 140.55 59 1,338

by guest on February 25, 2020http://w

ww

.jbc.org/D

ownloaded from

Phosphofructokinase Regulation in a Thyroid Tumor 6247

FIG. 1. Cellulose acetate electrophoresis of phosphofructokinase in extracts of the normal and tumor thyroid tissue. Electrophoresis was performed at 4’ and at 250 V for 160 min in 50 mM glycylglycine, 5 rnM ammonium sulfate, 0.1 mM EDTA, 5 mM mercaptoethanol, and 1.0 mM

ATP, pH 8.9. Right, normal thyroid enzyme; middle, thyroid tumor enzyme; left, a mixture of both. The origin is at the top; migration is toward the anode. Enzyme activity was detected on agar plates as described under “Materials and Methods.”

19 S THYROGLOBULINJ

I FRACTION NUMBER Bottom

I TOP

FIG. 2. Sucrose gradient centrifugation of normal and tumor phos- phofructokinase. Sucrose density gradients of 10 to 40% were used as described under “Materials and Methods.” Normal (04) or tumor (0-j thyroid phosphofructokinase (0.6 unit) was cen- trifuged for 38 h at 24,000 rpm at 5” with an SW 27 rotor. Ten units of rabbit muscle pyruvate kinase and about 12,000 cpm of 19 S [“Y]thyroglobulin were centrifuged as markers in the same experi- ment.

tumor cells is shifted to the left in Fig. 4 compared to the en- zyme in normal cells.

Cyclic AMP activated both enzymes; in fact, if one calcu- lates the per cent inhibition by a given concentration of ATP per mM, compared to that at the V,,,, the inhibition appears to be the same, plus or minus the cyclic nucleotide (Fig. 5).

.E 5 E 3 0.05 4 a

6

FIG. 3. Effect of pH on the activity of phosphofructokinase purified from normal rat thyroid (0, A) and purified from the thyroid tumor tissue (0, A). Activity was measured in the presence of 0.2 mM fructose-6-P, 1.0 mM ATP, 1.5 mM MgCI,, 0.1 mM dithiothreitol, 0.2 rnM EDTA, 5 mM ammonium sulfate, and 50 mM buffer and is expressed as absorbance change at 340 nm/min. Buffers used were glycylglycine buffer (O---O, O--O) and glycine (A-A, A-A).

0.

< -0

*r

ok

FRUCTOSE-6-P ImMl

FIG. 4. Plot of velocity versus fructose-6-P concentration in the absence of ammonium sulfate. Standard assay conditions at pH 7.2 were used with 1.5 mM MgCl, and 1.0 mM ATP. An amount of enzyme was added to each assay that would give a velocity of 0.01 rmol of fructose-l, 6-P* formed per min per ml of the standard assay mixture at pH 8.2 in the presence of 1 mM ATP and 1 mM fructose-6-P. This is defined as V,. Normal thyroid (O--O) and thyroid tumor (04) phosphofructokinase.

Although citrate was not a powerful inhibitor of either the normal or tumor phosphofructokinase (Fig. 6), the tumor enzyme was about 7 times less sensitive to the citrate inhibi- tion than the normal thyroid enzyme. The degree of inhibition of both phosphofructokinases by citrate was slightly modified by changing the enzymatic concentration, but the difference between the degree of inhibition of the two enzymes was quite evident at all enzyme concentrations tested. For comparison, muscle phosphofructokinase was inhibited by 10 times lower citrate concentration than the thyroid enzyme.

Analogous citrate inhibition data were obtained either with purified preparations or with crude extracts of normal thyroid, thyroid tumor, and the rat muscle phosphofructokinase, i.e. there was no desensitization to ATP and citrate regulation during the purification. Analogous citrate inhibition data were also obtained using extraction and homogenization mixtures different from those described in the purification paragraph, such as Tris-phosphate buffer, pH 7.8, containing 0.2 mM

EDTA, 5 mM mercaptoethanol, and 0.5 mM ATP, or Tris/HCl buffer, pH 8, containing 0.25 M sucrose, 0.2 mM EDTA, 0.1 mM

by guest on February 25, 2020http://w

ww

.jbc.org/D

ownloaded from

6248 Phosphofructokinase Regulation in a Thyroid Tumor

dithiothreitol, and 0.1 mM ATP; these conditions are the conditions used to protect regulatory properties maximally (9, 20).

Since the experiments reported in Fig. 6 were run in the presence of a fixed and relatively low MgClz concentration, and it is well known that citrate chelates Mg2+ strongly, the inhibition of normal and tumor thyroid phosphofructokinase by citrate was also measured in the presence of various concentrations of MgCl, (Fig. 7). The inhibition of normal thyroid phosphofructokinase by 8 mM citrate was counteracted by increasing the concentration of MgCl, up to 25 mM;

however, even at 10 mM MgCl,, citrate inhibition was quite evident. In the same experimental conditions the inhibition of tumor phosphofructokinase by 8 mM citrate was still less pro- nounced than that of the normal enzyme.

Since fructose-6-P as well as cyclic AMP cooperatively de- crease citrate inhibition of some mammalian phosphofructo- kinases (4, 21), the inhibition of normal and tumor phospho- fructokinase was studied as a function of citrate concentration in the presence of three different concentrations of fructose- 6-P. The data shown in Table II are expressed as the ratio be- tween enzymatic activity in the absence (VJ and that in the presence (VJ of the inhibitor. The higher the fructose-6-P con-

00 1.0 20 3.0 0 10 20

ATPhM,

FIG. 5. Effect of ATP concentration on phosphofructokinase activity in the presence (0) as well as in the absence (0) of 0.1 mM cyclic AMP. Standard assay conditions at pH 7.0 were used with 0.15 mM fructose-6-P, a MgCL:ATP ratio of 2:1, and 5 mM ammonium sulfate. V, is defined in the legend to Fig. 4. A, normal thyroid phosphofruc- tokinase; B, tumor thyroid phosphofructokinase.

centration, the higher was the difference between the degree of citrate inhibition between normal and tumor thyroid enzyme. Moreover, in the presence of 0.3 mM fructose-6-P, cyclic AMP was able to relieve citrate inhibition of thyroid phospho- fructokinase but was almost ineffective on the tumor enzyme. At lower fructose-6-P concentration, however, the reversal of citrate inhibition by cyclic AMP seemed to be almost equally effective on both enzymes (Table II).

All the above results, except those of Fig. 4, were obtained in the presence of 5 mM (NH,),SO,. Under these experimental conditions (which are close to those described by Kemp (9)), the effect of fructose-6-P on ATP and citrate inhibition was modified, since ammonium sulfate is an allosteric activator of phosphofructokinase (22) and is able to abolish the cooperativ- ity of erythrocyte enzyme (23, 24). As shown in Fig. 8, velocity was determined at two given levels of (NH,),SO, by varying the levels of fructose-6-P. In the presence of 0.3 mM (NH,),SO,, the V,,, of both enzymes increased with respect to that observed in Fig. 4 in the absence of (NH,),SO,, but the Hill coefficient calculated from the data shown in Fig. 8 dropped to n = 2.78 and rz = 1.72 for the normal (Fig. 8A) and tumor (Fig. 8B) phosphofructokinase. In the presence of 3.0 mM

WHd2S04, the V,,, was even more increased, and the Hill

10 A s

FIG. 7. Effect of MgC12 on phosphofructokinase inhibition by citrate in the presence of 1 mM ATP and 0.3 mM fructose-6-P. Standard assay conditions at pH 7.2 included 5 mM ammonium sulfate. A, normal thyroid phosphofructokinase in the absence (O---O) and in the presence (0-O) of 8 mM citrate. B, thyroid tumor phosphofructoki- nase in the absence (O--O) and in the presence (02) of 8 mM citrate. V, is defined in the legend to Fig. 4.

0.8 I I

0.6 - -

0 I I 0.1 1 .o 10 100

CITRATE (mM1

FIG. 6. Effect of citrate concentration on normal (0) and tumor (0) phospho- fructokinase activity. The standard assay mixture at pH 7.2 was used, i.e. 0.3 mM fructose-6-P, 1.0 mM ATP, 1.5 mM MgCl,, and 5 mM ammonium sul- fate. V, is defined in the legend to Fig. 4.

by guest on February 25, 2020http://w

ww

.jbc.org/D

ownloaded from

Phosphofructokinase Regulation in a Thyroid Tumor 6249

TABLE II

Effect of fructose-6-Pconcentration on citrate inhibition and reversal of citrate inhibition by cyclic AMPof nearlypurifiedphosphofructokinase from normal and tumor thyroid phosphofructokinase

The results are calculated as the ratio of control activity at pH 7.2 Rates were determined 2 to 4 min after starting the reaction by adding (VJ to the activity in the presence of citrate (V,). Assay conditions the enzymatic preparation. Cyclic AMP (0.6 IIN) was added 41% min were the same as in Fig. 7. An amount of enzyme was added that gave after starting the reaction in the presence or in the absence of citrate, an absorbance change at 340 nm/min of 0.130 at pH 8.2 in the stan- and the data are from 5.30 to 7.0 min after starting the reaction. dard assay conditions described under “Materials and Methods.”

Phosphofructokinase activity (V./V,)

Tissue Fructose-6-P Citrate inhibition Citrate (mM)

Reversal of citrate inhibition by 0.6 ITIM cyclic AMP addition

Citrate (mhr)

0 0.5 2.0 8.0 0 0.5 2.0 8.0

mhf

A. Rat thyroid 0.30 1.00 1.10 1.50 6.04 0.98 0.97 0.98 1.0 0.15 1.00 1.23 1.82 6.68 0.94 0.98 1.18 1.3 0.075 1.00 1.32 2.10 7.50 0.93 1.0 1.24 1.8

B. Rat thyroid 0.30 1.00 1 .oo 1.20 2.05 0.98 0.99 1.04 1.8 tumor 0.15 1.00 1.10 1.75 4.20 0.96 1.05 1.44 2.1

0.075 1.00 1.17 2.05 7.90 0.95 1.03 1.24 2.2

0.3

0.2

0.1

? - 0 > 0.6

0.4

0.2

c

I I I / I

3 x 10e4M Ammonium Sulfate

A B

3 x 10-S M Ammomum Sulfate C 0 .

Dp 0.5 1 .o 1.5 2.0 0 0.5 1 .o 1.5 2.c I

FRUCTOSE-6-P ImM)

FIG. 8. Plot of velocity versus fructose-6-P concentration in the presence of 0.3 mM ammonium sulfate (A and B) or of 3.0 mM ammonium sulfate (C and D). Thyroid (A and C) and tumor (Band II) phosphofructokinase activity was assayed in the absence (04) or in the presence of 0.8 rnM citrate (0-O) or 4.0 rnM citrate (A.--A). V, is defined in the legend to Fig. 4. Experimental conditions were identical to those of Fig. 4.

plots of such data gave a curve which was characterized by two n values: in the presence of low concentrations of fructose-6-P there was a strong effect of (NH,),SO, on the cooperativity of both the normal (Fig. 8C) and tumor (Fig. 8D) enzyme so that the Hill coefficient was decreased to about n = 1.2.

It is well known that phosphofructokinase shows complex allosteric properties, the degree of complexity being related to the extent of evolutionary development (3-5, 25-28). Although many effecters are active on phosphofructokinases isolated from most mammalian sources, their effectiveness diverges widely according to the presence of different phosphofructoki- nase isozymes in the various tissues. Such isozymes show different electrophoretic mobilities, different affinity to DEAE-cellulose, and different immunological and kinetic properties (9, 15-18, 20, 21, 28-30). By two different criteria, DEAE-cellulose affinity and precipitation with an antiserum to muscle phosphofructokinase, the thyroid enzyme behaves like a muscle A type isozyme (17). Despite this physical relation to a muscle enzyme, it has kinetic properties more closely resembling those of the rabbit brain isozyme (16). Since an antiserum to muscle A type isozyme also cross-reacts strongly with the erythrocyte B type isozyme (17) and also with the brain phosphofructokinase (15, 16), the exact nature of the thyroid isozyme is not clear.

In the presence of 3.0 ammonium sulfate, citrate inhibition The kinetic differences observed between normal and tumor was less effective on both the normal (Fig. 8C) and tumor (Fig. thyroid phosphofructokinase seem to be not related to clearcut 8U) phosphofructokinase. At low fructose-6-P concentration differences in their isozyme structure. There are some very and in the presence of 4.0 mM citrate, the Hill coefficient was slight but constant differences of affinity to DEAE-cellulose

nearly equal to 1.0, but it increased with increasing concentra- between normal and tumor thyroid phosphofructokinase. tions of fructose-6-P for both enzymes. A higher amount of These differences are very low as compared to those described

fructose-6-P was still necessary to decrease to 50% the citrate inhibition of the normal as opposed to the tumor phosphofruc- tokinase (Fig. 8, C and D).

DISCUSSION

The present report shows that normal and tumor thyroid phosphofructokinases have different kinetic and molecular properties. The kinetic properties indicate that the tumor enzyme is less sensitive to ATP and citrate inhibition as well as to the reversal of citrate inhibition by cyclic AMP than the normal thyroid phosphofructokinase (Figs. 5 to 8 and Table II). The relative insensitivity of the thyroid tumor enzyme to allosteric effecters such as ATP and citrate and the greater sensitivity to the effects of fructose-6-P may be related to the lack of control on the glycolytic pathway previously shown in thyroid tumor l-8 (6).

by guest on February 25, 2020http://w

ww

.jbc.org/D

ownloaded from

6250 Phosphofructokinase Regulation in a Thyroid Tumor

between muscle, brain, and liver isozymes, but are similar to those existing between tetrameric hybrids formed by both muscle and brain subunits (15, 16). Therefore, even if the differences in the kinetic data are not supported by a conspicu- ous shift in the distribution of isozymes, like that described in some fast-growing hepatic tumors (18), it may be possible, as shown for pyruvate kinase (31) and for other nonenzymatic proteins (32), that a so-called fetal subunit or a subunit of phosphofructokinase codified by the thyroid cell genome but normally not synthesized, may be produced by the transformed thyroid cells. I f these fetal subunits were to become assembled in a pure state or were to associate with the normal thyroid phosphofructokinase subunits, it might result in the appear- ance of a fully active phosphofructokinase with kinetic prop- erties different from those of normal thyroid phosphofructoki- nase but closely resembling it in its DEAE-affinity and elec- trophoretic properties.

The thyroid tumor phosphofructokinase shows a greater degree of size and/or shape heterogeneity than the normal thyroid enzyme, as seen by sucrose density gradient centrifuga- tion (Fig. 2). It is well known that the sedimentation coefficient of various mammalian phosphofructokinases depends on the enzyme concentration (21, 33-34), on the pH (35), and on in- hibitor or activator concentration (34, 36). Since ATP reduces the sedimentation coefficient of several mammalian phospho- fructokinases, it is not clear whether or not the slight increase in sedimentation coefficient and the greater heterogeneity of thyroid tumor phosphofructokinase relative to the normal thyroid enzyme reflects a slightly different degree of aggrega- tion of the tumor 1-8 enzyme itself or a lower sensitivity to ATP modulation, as shown by the kinetic data.

Desensitization to modulating effecters such as citrate and ATP of some mammalian phosphofructokinases has been described by photooxidation, by slight tryptic digestion, or during purification (37). In this report, we have shown that the kinetic properties of crude extracts and purified preparations of normal and tumor thyroid phosphofructokinase are approxi- mately the same. Moreover, the extraction procedure at O-4” has been very carefully accomplished and was identical for the various phosphofructokinase preparations. It is known that phosphofructokinase activity may be modified by interaction with fructose-diphosphatase or aldolase, both of which en- hance the inhibition of phosphofructokinase by ATP, by citrate, or by glycerate-3-P, by detaching and metabolizing fructose-1,6-Pz from phosphofructokinase (38). The latter is pro- tected against thermal inactivation (39) or inactivation by fructose-diphosphatase (40) by one or more peptide-stabilizing factor(s) which are present in the cell sap but have no direct effect on phosphofructokinase inhibition by ATP (40, 41). Since the molar ratio fructose-diphosphatase/phosphofructoki- nase for obtaining 50% inhibition varies from 100 to 400 (38), it is unlikely that such high amounts of aldolase or fructose- diphosphatase could be present as contaminants in the nearly purified phosphofructokinase preparations. Therefore, the 7 times differences observed in citrate inhibition between normal and tumor phosphofructokinase does not seem to be dependent on the presence of fructose-diphosphatase.

In conclusion, the origin of the differences observed between normal and tumor thyroid phosphofructokinases may be re- lated (a) to a different proportion of the isoenzymes or of hybrids formed by a mixture of different subunits, according to a recent functional concept of the role of minor isoenzymes of phosphofructokinase in various tissues (15, 16, 20, 30); (b) to a different modulation of the tumor enzyme by allosteric effec-

1.

2. 3. 4. 5. 6.

I. 8. 9.

10.

11.

12.

13.

14.

15. 16. 17.

18.

19.

20.

21

22

23

24

25 26 27

28

29.

30. 31.

32.

33.

REFERENCES

Scrutton, M. C., and Utter, M. F. (1968) Annu. Reu. Biochem. 37, 249-302

Meldolesi, M. F. (1971) Eur. J. Biochem. 22, 27-30 Atkinson, D. E. (1966) Annu. Reu. Biochem. 35, 85-124 Mansour, T. E. (19’72) Current Topics Cell. Regul. 5, 2-46 Blangy, D., But, H., and Monod, J. (1968) J. Mol. Biol. 31,13-35 Meldolesi, M. F., and Macchia, V. (1972) Cancer Res. 32,

2793-2798 Kemp, R. G., and Forest, P. B. (1968) Biochemistry 7,2596-2603 Wollman, S. H. (1963) Recent Prog. Hormone Res. 19, 579-618 Kemp, R. G. (1971) J. Biol. Chem. 246, 245-252 Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J.

(1951) J. Biol. Chem. 193, 265-275 Martin, R. G., and Ames, B. N. (1961) J. Biol. Chem. 236,

1372-1379 Salvatore, G., Salvatore, M., Cahnmann, H. J., and Robbins, J.

(1964) J. Biol. Chem. 239, 3267-3274 Schneider, A. B., and Edelhoch, H. (1971) J. Biol. Chem. 246,

6592-6598 Valentine, W. N., and Tanaka, K. R. (1966) Methods Enzymol. 9,

468-473 Tsai, M. Y., and Kemp, R. G. (1973) J. Biol. Chem. 248,785-792 Tsai. M. Y.. and Kemu. R. G. (1974) J. Biol. Chem. 249,6590-6596 Layzer, R. B., and Conway, G. M. (1970) Biochem. Biophys. Res.

Comma. 40, 1259-1265 Kurata, N., Matsushima, T., and Sugimura, T. (1972) Biochem.

Biophys. Res. Commun. 48, 473-479 Bihme, H. J., Kopperschlager, G., Schulz, J., and Hofmann, E.

(1972) J. Chromatogr. 69, 209-214 Dunaway, G. A., and Weber, G. (1974) Arch. Biochem. Biophys.

162, 620-628 Layzer, R. B., Rowland, L. P., and Bank, W. J. (1969) J. Biol.

Chem. 244, 3823-3831 Lowry, 0. H., and Passonneau, J. V. (1966) J. Biol. Chem. 241,

2268-2279 Akkerman, J. W. N., Gorter, G., Sixma, J. J., and Staal, G. E. J.

(1974) Biochim. Bioohys. Acta 370, 1022112 Kiihn, B., Jacobasch,’ G:, Gerth, C. C., and Rappaport, S. M.

(1974) Eur. J. Biochem. 43, 437-442 Uyeda, K., and Kurooka, S. (1970) J. Biol. Chem. 245,3315-3324 Walker. P. R.. and Bailey, E. (1969) Biochem. J. 111, 365-369 Beta., A., and Moore, c. (1967) Arch. Biochem. Biophys. 120,

268-273 Lindell, T. J., and Stellwagen, E. (1968) J. Biol. Chem. 243,

9077912 Tanaka, T., Imamura, K., Ann, T., and Taniuchi, K. (1972)

GANN Monogr. Cancer Res. 13, 219-234 Kwiatkowska, I. (1973) Biochim. Biophys. Acta 321, 475-483 Irving, M. G., and Williams, J. F. (1976) Biochem. Biophys. Res.

Commun. 68, 284-291 Groudine, M., and Weintraub, H. (1975) Proc. N&l. Acad. Sci.

U. S. A. 72, 4464-4468 Mansour, T. E., Wakid, N., and Sprouse, H. M. (1966) J. Biol.

Chem. 241, 1512-1521

tors which may be related to a modification of the tumor thyroid enzyme structure with respect to the normal; and (c) to an effect produced by a co-purified unknown molecule interfer- ing with the allosteric interactions of the normal or tumor thyroid phosphofructokinase. The consequences, however, of the changes in sensitivity to the allosteric effecters, whatever is the cause, seem to be crucial for the regulatory functions of phosphofructokinase in the tumor thyroid tissue with respect to the normal thyroid glycolytic pathway.

Aclznoluledgments-We are indebted to Dr. Harold Edel- hoch and to Dr. Leonard D. Kohn of the National Institutes of Health, Bethesda, Maryland, for their continuous advice and encouragement throughout this work. We are very grateful to Mrs. Helen Jenerick (Laboratory of Biochemical Pharmacol- ogy, National Institute of Arthritis, Metabolism and Digestive Diseases) for the aid in the preparation of this manuscript.

by guest on February 25, 2020http://w

ww

.jbc.org/D

ownloaded from

Phosphofructokinase Regulation in a Thyroid Tumor 6251

34. Parmeggiani, A., Luft, J. H., Love, D. S., and Krebs, E. G. (1966) 39. Dunaway, G. A., and Segal, H. L. (1974) Biochem. Biophys. Res. J. Bid. Chem. 241, 4625-4637 Commun. 56, 689-696

35. Bock, P. E., and Frieden, C. (1974) Biochemistry 13, 4191-4199 40. Sankaran, L., Proffitt, R. T., Pogell, B. M., Dunaway, G. A., and 36. Hill, D. E., and Hammes, C. G. (1975) Biochemistry 14, 203-213 Segal, H. L. (1975) Biochem. Biophys. Res. Commun. 67, 37. Setlow, B., and Mansour, T. E. (1970) J. Biol. Chem. 245, 220-227

5524-5533 41. Dunaway, G. A., and Segal, H. L. (1976) J. Biol. Chem. 251, 38. Uyeda, K., and Luby, L. J. (1974) J. Biol. Chem. 249, 4562-4570 2323-2329

by guest on February 25, 2020http://w

ww

.jbc.org/D

ownloaded from

M F Meldolesi, V Macchia and P Laccetticells.

Differences in phosphofructokinase regulation in normal and tumor rat thyroid

1976, 251:6244-6251.J. Biol. Chem. 

  http://www.jbc.org/content/251/20/6244Access 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/251/20/6244.full.html#ref-list-1

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

by guest on February 25, 2020http://w

ww

.jbc.org/D

ownloaded from