Purification and partial characterization of different forms of phosphofructokinase in man.

Phosphofructokinase (ATP:D-fructose-6-phosphate 1-phosphotransferase, EC 2.7.1.11) from human muscle, brain, heart and granulocytes has been purified using a two or three step purification procedure. The main step is Blue Dextran-Sepharose 4B chromatography with selective elution of phosphofructokinase by formation of the ternary complex ADP or ATP-fructose-6-P-enzyme. Muscle and heart contain only enzyme subunits with a molecular weight of 85,000. This type of subunit is predominnant in brain, where it co-exists with subunits of about 80,000 daltons. A single type of subunits is found in the granulocytes, with a molecular weight of 80,000. Anti-muscle phosphofructokinase antiserum reacts only with M-type enzyme. Anti-granulocyte enzyme antiserum, absorbed by pure brain phosphofructokinase, exhibits a narrow specificity against the so-called L-type enzyme. Anti-brain antiserum, absorbed by pure muscle phosphofructokinase and partly purified red cell enzyme, exhibits a narrow specificity against a phosphofructokinase form predominant in fibroblasts and present in brain (F-type).     

The genetic system of the L-type pyruvate kinase forms in man. Subunit structure, interrelation and kinetic characteristics of the pyruvate kinase enzymes from erythrocytes and liver

Pyruvate kinase (ATP: pyruvate 2-O-phosphotransferase, EC 2.7.1.40) from human liver and red cells has been purified to homogeneity; its subunit structure and some of its kinetic characteristics have been studied. The influence of a partial proteolysis by trypsin on the subunit structure, the isozymic pattern and the kinetic characteristics of red cell and liver enzyme have been investigated. From the results of this study we may conclude that: 1. Liver (L-type) pyruvate kinase is composed of 4 identical L subunits while the major form of erythrocyte enzyme (PK-R2) is a heterotetramer designated as L2L2', the molecular weight of L' being slightly higher than that of L subunits (63 000 and 58 000 respectively). Pyruvate kinase PK-R1, predominant in the erythroblasts and the young red cells, is composed of four identical L' subunits. 2. A mild tryptic attack is able to transform PK-R1 into PK-R2, then PK-R2 into pyruvate kinase L (PK-L). The same proteolytic treatment transforms the L' subunits into L ones. 3. Consequently L-type pyruvate kinase seems to be initially synthesized in the erythroid precursors as an L4' enzyme secondarily partially proteolysed into L2L2'. In liver a very active proteolytic system would be responsible for the total transformation into L4 pyruvate kinase. 4. L4' enzyme exhibits Michaelis-Menten kinetic behaviour with an apparent Michaelis constant of 3.8 mM whereas L4 enzyme shows both positive and negative homotropic interactions towards phosphoenolpyruvate and has [S] 0.5 of 1.2 mM. The characteristics of L2L2' are roughly intermediate between those of L4' and of L4. Fructose 1,6-biphosphate decreases [S]0.5 for these three pyruvate kinase forms without suppressing the differences in the apparent affinity for phosphoenolpyruvate of these enzymes. 5. L4 pyruvate kinase is more inhibited by Mg-ATP than L4', with L2L2' in the intermediate range. 6. Tryptic treatment of each enzyme form studied transforms its kinetic behaviour into that observed for L4.     

Purification of F4 phosphofructokinase from human platelets and comparison with the other phosphofructokinase forms

The phosphofructokinase (ATP:D-fructose-6-phosphate 1-phosphotransferase, EC 2.7.1.11) tetramers F4, F3L and F2L2 have been separated from human platelets, and purified to homogeneity by affinity chromatography on Dextran Blue-Sepharose 4B. The F subunits have a molecular weight of 85 000, identical to that of the M subunits. By contrast with L-type phosphofructokinase, the F-type enzyme seems to exist predominantly in a tetrameric form and not to aggregate to high molecular weight polymers. Specific activity of pure F4 phosphofructokinase is about 140 IU/mg of protein. Immunologically, it is easy to distinguish all the basic phosphofructokinase forms (i.e. M, L and F types); nevertheless a slight immunological cross-reactivity seems to exist between all these forms.     

Mode of interaction of phosphofructokinase with the erythrocyte membrane

Phosphofructokinase is known to associate with the human erythrocyte membrane both in vitro and in vivo. Such association activates the enzyme in vitro by relieving the allosteric inhibition imposed by ATP (Karadsheh, N.S., and Uyeda, K. (1977) J. Biol. Chem. 252, 7418-7420). We now demonstrate that ADP, ATP, and NADH, all of which are known to bind to the enzyme's adenine nucleotide activation site, are particularly potent in eluting the enzyme from the membrane. In addition, both inside-out red cell membrane vesicles and a 23-kDa fragment containing the amino terminus of the membrane protein, band 3, cause a slow, partial, and reversible inactivation of phosphofructokinase. The dependence of the residual phosphofructokinase activity on phosphofructokinase concentration demonstrates that inactivation occurs through the dissociation of active tetramers to inactive dimers. Dimers of phosphofructokinase associate with the membrane more avidly than tetramers. The kinetics of phosphofructokinase inactivation are consistent with the dissociation of tetramers in solution followed by the binding of dimers to the membrane. There is no indication of an association equilibrium between tetramers and dimers of phosphofructokinase bound to the membrane. Taken together, these results suggest that the amino-terminal segment of band 3 binds to the adenine nucleotide activation site, which is thought to be located in a cleft between the dimeric subunits of phosphofructokinase. As a result, band 3 not only rapidly activates the phosphofructokinase tetramer but also slowly inactivates the enzyme by preferentially binding its dissociated subunits.     

Phosphofructokinase (PFK) isozymes in man

Summary Isozymic heterogeneity of human phosphofructokinase was investigated by means of ATP inhibition, immunoneutralization by antihuman muscle-type and antiliver-type phosphofructokinase antisera, solubility in (NH₄)₂SO₄ solutions, and starch gel and polyacrylamide slab gel electrophoresis. The enzymes studied by these methods were purified from various normal and malignant human adult tissues by chromatography on blue Dextran Sepharose 4 B columns. From the results of these studies we suggest that three basic phosphofructokinase isozymes could exist: muscle-type, fibroblast-type, and liver-type isozymes.Muscle-type isozyme is the single form found in adult muscle, and is involved in the enzymes from heart, brain, red cell, and testis.Fibroblast-type isozyme is found mainly in the placenta, fibroblasts, kidney, and some malignant tissues.Liver-type phosphofructokinase seems to be very definitely the predominant form in mature polymorphonuclear cells, platelets, and liver.Testis and red cell phosphofructokinase enzymes definitely include muscle-type and liver-type subunits, associated in various hybrid forms.With the technical assistance of R. KernempUnité 129 de l'Institut National de la Santé et de la Recherche Médicale, laboratoire associé 85 au Centre National de la Recherche Scientifique     

Phosphorylation of human erythrocyte pyruvate kinase by soluble cyclic-AMP-dependent protein kinases. Comparison with human liver L-type enzyme

Human red cell contain soluble adenosine-3',5'-phosphate-dependent protein kinases, which are able to phosphorylate the L' subunits of erythrocyte pyruvate kinase. Efficiency and maximum level of phosphorylation are very comparable in human liver and red cells. Phosphorylation of red cell pyruvate kinase results in the same kinetic modifications as for liver enzyme, namely a shift towards a 'T' allosteric state characterized by a decreased affinity for phosphoenolpyruvate and increased inhibition by the allosteric inhibitors ATP and alanine. In the course of red cell aging a small amount of partially proteolysed pyruvate kinase, devoid of the phosphorylatable site, appears; it resembles the subtilisin-treated L'4 enzyme and accounts for less than 20% of total pyruvate kinase subunits. Endogenous phosphorylation of pyruvate kinase from erythrocytes incubated in the presence of cyclic nucleotides produces the same kinetic modifications as phosphorylation in partially purified extract; this, however, does not change glucose consumption, lactate production and glycolytic intermediate concentrations of the incubated cells.     

Influence of fructose 2,6-bisphosphate on the aggregation properties of rat liver phosphofructokinase

The influence that fructose 2,6-bisphosphate (Fru-2,6-BP) has on the aggregation properties of rat liver phosphofructokinase has been studied by observing the fluorescence polarization of the enzyme covalently bound to the fluorescent probe pyrenebutyric acid. Fru-2,6-BP dramatically slows the dissociation of the high molecular weight aggregate forms of the enzyme when the enzyme is diluted to 3.2 micrograms/ml (4 X 10(-8) M subunits). Furthermore, Fru-2,6-BP is a strong promoter of reassociation to tetramer and larger forms if the enzyme has been previously allowed to dissociate to the dimer in its absence. Unlike many other positive effectors of liver phosphofructokinase, Fru-2,6-BP is also able to overcome the tendency of MgATP to promote tetramer formation and instead stabilize a very high degree of high molecular weight aggregate formation even in the presence of MgATP. The apparent affinity of liver phosphofructokinase for Fru-2,6-BP was measured by its ability to promote reassociation and compared to that for Fru-1,6-BP. The apparent dissociation constant for Fru-2,6-BP under these conditions is 36 microM, about 40-fold lower than the value of 1.4 mM measured for Fru-1,6-BP. Both ligands demonstrate synergism with the substrate Fru-6-P, which can lower the dissociation constant for Fru-2,6-BP 9-fold to 4 microM and that for Fru-1,6-BP 5-fold to 0.28 mM. These data are interpreted to suggest that influencing the aggregation state of rat liver phosphofructokinase may be one way in which Fru-2,6-BP achieves its effects on the enzyme in vivo.     

Alterations in phosphofructokinase isoenzymes during early human development. Establishment of adult organ-specific patterns.

Human 6-phosphofructokinase (EC 2.7.1.11) exists in tetrameric isoenzymic forms composed of muscle (M), liver (L) and platelet (P) subunits, which are under separate genetic control. In the adult, the proportion of these subunits in different organs reflects the relative activity of glycolysis versus gluconeogenesis. To elucidate the developmental basis for the observed distribution, we investigated the isoenzymic transitions of phosphofructokinase in human foetuses (12-40 weeks' gestation) by using high-resolution chromatography and monoclonal antibodies. We studied skeletal muscle, heart, liver and brain because these organs show very different glycolytic fluxes and isoenzymic patterns in adult individuals. Our results demonstrate that there is no unique 'foetal' form of phosphofructokinase in humans, but all three loci are variably expressed in all foetal organs during early gestation. As development proceeds, muscle and liver isoenzyme patterns show dramatic changes, with disappearance of P and L subunits in muscle and transient reappearance of M and P subunits in liver; in contrast, phosphofructokinase isoenzymes change little in brain and heart. Most changes occur at mid-gestation and near term, and adult isoenzyme patterns are expressed at birth, indicating that organ differentiation is complete. These studies show that phosphofructokinase undergoes changes of isoenzyme patterns similar to, but not identical with, those of other multilocus isoenzyme systems of glycolysis. The observed changes probably reflect changing patterns of gene expression, with repression of some loci and activation of others.     

Hormone-stimulated phosphorylation of liver phosphofructokinase in vivo.

The effect of glucagon on the phosphorylation and the enzymic activity of phosphofructokinase in rat liver in vivo was investigated. Glucagon stimulated the phosphorylation of liver phosphofructokinase approximately 3- to 5-fold and increased cAMP levels 5-fold and blood glucose levels 2-fold over the values obtained for control animals. The specific radioactivity of ATP isolated from liver was the same in both control and hormone-treated animals. During the purification of the 32P-labeled enzyme from both animals, no difference was observed in the total or specific enzyme activities of the enzymes from the various fractions. Thus, phosphofructokinase appears to be phosphorylated in vivo by a cyclic AMP-dependent protein kinase. Although phosphorylation does not affect the maximum catalytic activity of the enzyme, it does render the enzyme significantly more sensitive to ATP inhibition. Thus, at a given concentration of ATP, the phosphorylated phosphofructokinase exhibits considerably lower activity than the unphosphorylated enzyme. The possible relationship between our observations and glucagon-mediated control of glycolysis is discussed.     

Association of phosphofructokinase and aldolase with the membrane of the intact erythrocyte

The binding of phosphofructokinase and aldolase to the membrane of the intact human erythrocyte was assessed by the rapid hemolysis/filtration method of Kliman and Steck (Kliman, H. J., and Steck, T. L. (1980) J. Biol. Chem. 255, 6314-6321). We found that about 50% of the phosphofructokinase was membrane-bound in fresh red cells prior to hemolysis. Binding was not significantly altered by deoxygenation. Approximately 40% of aldolase was membrane-associated in fresh red cells. In outdated, blood-banked red cells, aldolase was 73% membrane-bound while, following metabolic repletion, 40% of the enzyme was membrane-associated. These results support the hypothesis that certain glycolytic enzymes in the red cell are membrane-bound in a rapidly reversible and metabolically sensitive fashion.     

The A2B2 hybrid isozyme of phosphofructokinase.

The hybrid isozyme of phosphofructokinase, A2B2, was formed by incubation of rabbit muscle enzyme. A4, and rabbit liver enzyme, B4, in the presence of sodium citrate at neutral pH. The enzyme composition of the resulting mixture of A2B2 and the homoprotomeric forms was identical to that found in rabbit adipose tissue extracts. Hybrid formation, which apparently proceeds by way of dimers, can be blocked by fructose-1,6-P2, fructose-6-P, and high concentrations of MgATP. The A2B2 isozyme was separated from A4 and B4 by ion exchange chromatography. The kinetic regulatory properties of A2B2 were compared with those of A4, B4, and a 1:1 mixture of A4 and B4. ATP inhibition of A2B2 was intermediate between that observed with A4 and B4 and was clearly not identical to a simple summing of the effects of A and B subunits. Similar comparisons were made using other phosphofructokinase inhibitors, citrate, 2,3-P2-glycerate, and P-creatine. In each case the observed inhibition was intermediate between the observed with A4 and B4. The existence in a number of tissues of phosphofructokinase A2B2 provides added diversity to the regulatory mechanisms of glycolysis.     

The distribution of phosphofructokinase isoenzymes in the liver of camel, rat and rabbit

1. The distribution of phosphofructokinase isoenzymes have been compared among camel, rat and rabbit livers. 2. Only a single phosphofructokinase isoenzyme is present in the camel liver which has shown different physical and regulatory properties from the isoenzymes of rat and rabbit liver. 3. The ammonium sulphate precipitation curves of the camel and rabbit enzymes were monophasic, whereas the rat enzyme was biphasic. 4. Rabbit liver phosphofructokinase was slightly more anodic than the rat enzyme, whereas the camel enzyme was the least anodic as shown by the techniques of DEAE-cellulose chromatography and cellulose acetate electrophoresis. 5. Partially purified camel liver phosphofructokinase showed different regulatory properties from the rabbit and rat isoenzymes as the apparent Km values were 0.58, 0.45 and 0.82 mM respectively.     

Plasmodium berghei: acid-insensitive phosphofructokinase in infected mouse erythrocytes

The rate of glucose utilization by red blood cells infected with Plasmodium berghei was not inhibited by an acidic pH which completely inhibited normal red cell glucose consumption. This insensitivity to acid conditions by P. berghei-parasitized red cells was associated with an electrophoretically separable and kinetically distinct form of the enzyme phosphofructokinase (EC 2.7.1.11) which exhibited a pH response similar to that of whole-cell glucose consumption.     

Sulfhydryl groups of yeast phosphofructokinase-specific localization on beta subunits of fructose 6-phosphate binding sites as demonstrated by a differential chemical labeling study

Yeast phosphofructokinase contains 83 +/- 2 cysteinyl residues/enzyme oligomer. On the basis of their reactivity toward 5,5-dithiobis(2-nitrobenzoic acid), the accessible cysteinyl residues of the native enzyme may be classified into three groups. For titrations performed with N-ethylmaleimide, subdivisional classes of reactivity are evidenced. In each case, the 6 to 8 most reactive cysteines are not protected by fructose 6-phosphate from chemical labeling and do not seem involved in subsequent enzyme inactivation. Differential labeling studies as well as direct protection experiments in the presence of fructose 6-phosphate, indicate that 12 -SH groups/enzyme oligomer (i.e. three -SH groups per binding site) are protected by the allosteric substrate from the chemical modification. Specific labeling by the differential method of the cysteinyl residues protected by fructose 6-phosphate and further separation of the two types of subunits constituting yeast phosphofructokinase, show that the substrate binding sites are localized exclusively on subunits of beta type. Thus, alpha subunits are not implicated directly in the catalytic mechanism of yeast phosphofructokinase reaction.     

Purification and properties of phosphofructokinase 2/fructose 2,6-bisphosphatase from chicken liver and from pigeon muscle

Phosphofructokinase 2 and fructose 2,6-bisphosphatase extracted from either chicken liver or pigeon muscle co-purified up to homogeneity. The two homogeneous proteins were found to be dimers of relative molecular mass (Mr) close to 110,000 with subunits of Mr 54,000 for the chicken liver enzyme and 53,000 for the pigeon muscle enzyme. The latter also contained a minor constituent of Mr 54,000. Incubation of the chicken liver enzyme with the catalytic subunit of cyclic-AMP-dependent protein kinase in the presence of [gamma-32P]ATP resulted in the incorporation of about 0.8 mol phosphate/mol enzyme. Under similar conditions, the pigeon muscle enzyme was phosphorylated to an extent of only 0.05 mol phosphate/mol enzyme and all the incorporated phosphate was found in the minor 54,000-Mr constituent. The maximal activity of the native avian liver phosphofructokinase 2 was little affected by changes of pH between 6 and 10. Its phosphorylation by cyclic-AMP-dependent protein kinase resulted in a more than 90% inactivation at pH values below 7.5 and in no or little change in activity at pH 10. Intermediary values of inactivation were observed at pH values between 8 and 10. Muscle phosphofructokinase 2 had little activity at pH below 7 and was maximally active at pH 10. Its partial phosphorylation resulted in a further 25% decrease of its already low activity measured at pH 7.1 and in a negligible inactivation at pH 8.5. Phosphoenolpyruvate and citrate inhibited phosphofructokinase 2 from both origins non-competitively. The muscle enzyme and the phosphorylated liver enzyme displayed much more affinity for these inhibitors than the native liver enzyme. Fructose 2,6-bisphosphatase from both sources had about the same specific activity but only the chicken liver enzyme was activated about twofold upon incubation with ATP and cyclic-AMP-dependent protein kinase. All enzyme forms were inhibited by fructose 6-phosphate and this inhibition was released by inorganic phosphate and by glycerol 3-phosphate. Both liver and muscle fructose 2,6-bisphosphatases formed a 32P-labeled enzyme intermediate when incubated in the presence of fructose 2,6-[2-32P]bisphosphate.     

Reaction of phosphofructokinase 2/fructose 2,6-bisphosphatase with monoclonal antibodies. A proof of the bifunctionality of the enzyme

Monoclonal antibodies were derived from mice immunized against homogeneous chicken liver phosphofructokinase 2/fructose 2,6-bisphosphatase. Of 112 clones, 30 were found to secrete antibodies that specifically reacted with the antigen in enzyme-linked immunoabsorbant assay (ELISA) while 17, which were ELISA-negative, produced antibodies that affected the enzymic activity of the antigen. Four clones were subcloned and used for an extensive investigation of the reaction of the corresponding antibodies with the supposedly bifunctional enzyme. A definite proof of the bifunctionality of the enzyme was obtained from the two following observations. First, the two activities were similarly retained by the four antibodies that had been coupled to Sepharose. Second, one of the antibodies inhibited both activities with the same efficiency. Furthermore, the antigen-antibody reaction led to the formation of aggregates with an apparent molecular mass of several megadaltons, showing that the two subunits of the antigen reacted with the same antibody and were therefore identical. The four monoclonal antibodies affected the activity of phosphofructokinase 2. This effect was seen as an up to 17-fold activation as well as an up to 85% inhibition. Only one of the four antibodies (antibody 10) had inhibitory effects on fructose 2,6-bisphosphatase, an effect which was in part explained by a decrease in the rate of formation of the intermediary phosphoenzyme. All the effects described above were obtained on both the chicken liver and the pigeon muscle enzymes but with lower doses of antibody in the case of the former enzyme. Antibody 10 was also shown to react with mouse liver phosphofructokinase 2/fructose 2,6-bisphosphatase, and with phosphofructokinase 2 from chicken brain, heart and testis and from frog skeletal muscle and liver. None of the four antibodies cross-reacted with phosphofructokinase 2 from Saccharomyces cerevisiae or from spinach leaves.     

Studies on the oligomeric structure of yeast phosphofructo-kinase by means of cross-linking diimidoesters.

Intracellular cross-linking of yeast phosphofructokinase with a series of diimidoesters of different chain length resulted in the appearance of tetramers as largest cross-linked product of the enzyme subunits. The native enzyme is evidently composed of eight subunits being arranged in two tetramers α₄β₄. In the tetramers the monomers are probably assembled in tetrahedral geometry.     

Mouse (C57BL/KsJ) liver phosphofructokinase. Allosteric kinetics and age-related changes in the genetically diabetic state

The regulatory kinetic properties of phosphofructokinase partially purified from the livers of C57BL/KsJ mice were studied. The fructose 6-phosphate saturation curves were highly pH dependent. At a fixed MgATP concentration (1 mM), allosteric kinetics was observed in the range of pH studied (7.3 to 8.3) and the S0.5 values for fructose 6-phosphate decreased by about 0.2 to 0.3 mM for each 0.1-unit increment in pH. Allosteric effects on the sigmoidal response to fructose 6-phosphate: activation by AMP, NH4+, and glucose 1,6-bisphosphate, inhibition by MgATP2-, and synergistic inhibition between ATP and citrate, were all present at pH 8.0 to 8.2. Comparative kinetic studies with liver phosphofructokinase isolated from both the normal (C57BL/KsJ) and the genetically diabetic (C57BL/KsJ-db) mice of 9 to 10 and 15 to 16 weeks of age showed that the enzyme from the livers of diabetic mice exhibited decreased activity at subsaturating concentrations of fructose 6-phosphate. However, phosphofructokinase isolated from the livers of normal and genetically diabetic mice of 4 to 5 weeks of age showed no difference in kinetic properties. Thus, there appears to be a correlation between the change in properties of liver phosphofructokinase and the expression of hyperglycemia and obesity in the genetically diabetic mice. The decreased activity of liver phosphofructokinase in the older diabetic animals may well be one of the causes of the increased blood glucose levels. The results are also discussed in a general context with regard to the possible role of phosphofructokinase in the regulation of hepatic gluconeogenesis.     

Undifferentiated patterns of key carbohydrate-metabolizing enzymes in injured livers. I. Acute carbon-tetrachloride intoxication of rat.

In acute CCL4 intoxication of rats significantly increased activities of hepatic low-Km hexokinases, glucose-6-phosphate dehydrogenase, phosphofructokinase, aldolase A and pyruvate kinase M2 with concurrently decreased activities of glucokinase, glucose-6-phosphatase, fructose-1,6-diphosphatase, aldolase B and pyruvate kinase L were observed. The resulting enzyme pattern was apparently different from that in dietary induction. Principal component analysis revealed that the degree of enzyme deviation in the injured liver was much greater than that in the regenerating liver after partial hepatectomy and was closer to that in fetal liver or hepatoma tissue.