Inhibition of Glycolysis by Amino Acids in Ascites

Vol. 268, No. 11, Issue of April 15, pp. 7809-7817,1993 Printed in U.S.A.

THEJOURNALOF BIOLOGICAL CHEMISTRY 0 1993 by The American Society for Biochemistry and Molecular Biology, Inc

Inhibition of Glycolysis by Amino Acidsin Ascites Tumor Cells SPECIFICITYANDMECHANISM* (Received for publication, July 23, 1992)

Fernando Gonzhlez-Mateos,Maria-Esther Gomez, Leticia Garcia-SalgueroS, Valentina Sanchez, and Juan J. Aragon8 From the Departamento de Bioquimica de la Universidad Autonoma and Instituto de Inuestigmiones Biomedicas del Consejo Superior de Inuestigaciones Cientificas, Facultad de Medicina de la Universidad Autonoma, 28029 Madrid, Spain

paragine is another aminoacid for whichmalignant cells show a marked preference, and this is a basis for the antitumor activity of asparaginase (15, 16), which depletes circulating asparagine in animals and cancer patients (17,18). Nevertheless, the significance of the great demand for amino acids shown by proliferating cells is still poorly understood (1, 10, 14,19). It has been suggested that high rates of glycolysis and glutamine degradation arerequired for high sensitivity of the pathways to specific regulators (2). However, other authors (1)have questioned a simultaneous flux through both processes as essential for cell proliferation on the basisof glucose limitation experiments and the use of mutants with defects in theiroxidative metabolism. Glutaminehasbeenreportedto decrease the levels of fructose-2,6-P2 in HeLa cells and chick embryo fibroblasts (20), producing a mild diminution of the glycolytic flux only in the latter typeof cells. Addition of glutamine to L929 cell cultures withglucose and insulin has been shown to stimulate lactate production (21) without change in glucose oxidation, butthemechanismhasnot beenclarified. Therefore, we reinvestigated the possibility of a significant interaction between the metabolism of glutamine and asparagine and the cells. We have control of glucose consumptionincancer examined whether these aminoacids affect the concentration of fructose-2,6-P2 and the operationof glycolysis, as well as their influence on other signals considered important for the regulation of this pathway. Ascites tumor cells were chosen since it is well known that they show high aerobic glycolysis Many tumors and rapidly dividing cells in general exhibit (11, 22) and a great dependence on glutamine (11-14) and a characteristically enhanced aerobic glycolysis (1, 2). The asparagine (18,23). Wefound that these amino acids strongly inhibit theglycolytic rate and that this inhibition is due to an mechanism and significance of this observation remain unclear despite numerous explanations that have been advanced intense decrease in theflux through the phosphofructokinase toaccountforit (1-9). Thistype of cellalso exhibits a reaction. Elucidation of the specificity and mechanismof this remarkably high utilization of glutamine, which serves both phenomenon is also reported. as a precursor in the synthesis of nucleotides, proteins, and other compounds and as a major energy substrate (1, 2, 10). MATERIALSANDMETHODS In Ehrlich ascites cells, glutamine is oxidized 20 times faster Chemicals and Enzymes-Phosphoric esters, nucleotides, PEG,’ than glucose (11) with production of glutamate, aspartate, phenylmethylsulfonyl fluoride, leupeptin, cycloheximide, and auxiland COz (12). The pathways of glutamine metabolism have iary enzymes were purchased from Sigma. Radiochemicals were obalso been investigated in isolated mitochondria (13, 14). As- tained from Amersham International. Amino acids were also from

The effect of glutamine and asparagine on glucose metabolism has been studied in ascites tumor cells. Either of these amino acids decreased the glycolytic flux about 80%.Half-maximal effects were obtained with 0.14 mM glutamine and 0.087 mM asparagine. Among the 20 L-amino acids, onlyglutamate produced a similar effect. Glutamine and asparagine caused a 70% increase of hexose monophosphates and a large decrease of fructose-1,6-Pz and triose phosphates, evidencing a strong inhibition of the phosphofructokinase (EC 2.7.11) reaction. Analysis of the levels of various phosphofructokinase effectors revealed that fructose-2,6-Pzand AMP decreased &fold, phosphoenolpyruvate, citrate,and ATP increased 4-, 3-, and 1.8fold, respectively, and that there was no change in ADP, Pi, and intracellular pH. Assay of phosphofructokinase at concentrations of substrates and effectors determined to bein the cells showed that the low activity of this enzyme could beaccounted for by the change in the concentration of effectors, the major mechanism being the change in adenine nucleotides. The decrease in fructose-2,6-P2 contributed very little to the inhibition of phosphofructokinase activity. The effects of amino acids were prevented by amino-oxyacetate, suggesting that transamination was an obligatory step for these changes.

* This work wassupported by Grant PB89-0168 from the Direction

Sigma except for aspartate, glutamine, and proline, as well as other chemicals, that were obtained from Merck. Ehrlich Ascites Cells-A hypertriploid strain of Ehrlich-Lettri. ascites carcinoma cells was maintained by weekly inoculation into the abdominal cavity of 2-month-old male Swiss mice. The animals received standard Panlab food (Barcelona, Spain) and water ad libitum. Cells were harvested 7-9 days after inoculation, suspended in 38.5 mM NaCl, centrifuged a t 1,000 X g for 5 min at room

General de Investigacion Cientifica y Tbnica. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. $ On leave of absence from: Dept. Bioquimica y Biologia Molecular, Fac. Ciencias, Univ. Granada, 18001 Granada, Spain. 8 To whom correspondence should be addressed Dept. de BioquimiThe abbreviations used are: PEG, polyethylene glycol; TES, N ca, Facultad de Medicina de la Universidad Autonoma, Arzobispo tris(hydroxymethyl)-2-aminoethanesulfonic acid; PIPES, piperazineMorcillo 4, 28029Madrid, Spain. Tel.: 34-1-3975333; Fax: 34-1- N,N”bis(2-ethanesulfonic acid); P-enolpyruvate; phosphoenolpyru3150075. vate.

’

7809

7810

Inhibition of Glycolysis Acids by Amino

temperature, twice washed free of erythrocytes in 115 mM NaCl, 5 mM potassium phosphate, pH 7.4, and resuspended to 25 mg/ml (wet weight) in buffer A (50 mM TES, 50 mM PIPES, 85 mM NaCl, 6 mM KCl, pH 7.0). Five to seven animals were used for each experiment. Incubations-The cell suspension was preincubated for 90 min under aerobic conditions a t 37 "C to deplete the intracellular pool of endogenous metabolites and to place cells coming from different animals in a similar situation. At the end of the preincubation, cells were collected by centrifugation. A gram of wet weight under these conditions, considered throughout this work as the reference value, is equivalent to (0.50 k 0.02) X lo8 cells. The cells were suspended to 40 mg/ml in buffer A and distributed in aliquots. Additions indicated in each experiment were made and the cells incubated for various periods of time under aerobic conditions, a t 37 "C, with gentle shaking. More than 90% of the cells remained viable after incubation for 180 min, as estimated by trypan blue staining. Incubations were terminated by addition of 8% (v/v) perchloric acid in 40% (v/v) ethanol unless otherwise indicated. Metabolite Measurements-For the measurement of fructose-2,6Pz, samples of 1 ml of cell incubations were centrifuged at 1,000 X g for 30 s. The cell pellets were deproteinized at 80 "C for 10 min with 0.4 ml of hot (80 "C) 0.1 M NaOH. After homogenization and centrifugation at 25,000 X g for 10 min, the supernatant was assayed for fructose-2,6-Pz as a stimulator of potato pyrophosphate:fructose-6phosphate phosphotransferase (24). This stimulation was suppressed by treatment of the extract at 25 "C at pH 1-2 for 10 min. For the measurement of glycolytic intermediates, citrate, ATP, ADP, and inorganic phosphate, samples of cell incubations were homogenized in 1volume of ice-cold 8% (v/v) perchloric acid in 40%(v/v) ethanol. After centrifugation at 25,000 X g for 10 min, the supernatant was neutralized to pH 7.0 with a solution containing 2 M KOH and 0.5 M triethanolamine, the mixture was allowed to standon ice for 15 min, and precipitated Kclo, was removed by centrifugation. Metabolites were measured by standard enzymatic techniques (25, 26) in a Shimadzu model UV3000dual wavelength spectrophotometer, set toread absorbance at 340 nm minus absorbance at 400 nm. For the measurement of AMP, samples of 5 ml of cell incubations were centrifuged at 1,000 X g for 30 s, and the cell pellets were deproteinized with 1 ml of ice-cold 4% perchloric acid in 20% ethanol as above. AMP was determined as described (27). The ATPused in the assay was purified by chromatography on Dowex-1 chloride (X2) to remove nucleotide impurities. Inorganic phosphate was measured using an enzymatic method similar to that reported by Guynn et al. (28), except that hexokinase was not included in theassay. The internal water volume was determined by using [3H]H20 and ["Clinulin following a procedure similar to that of Rottenberg (29). The intracellular water thus obtained was 56 2%. A wet weight/ dry weight ratio of 4.6 k 0.1 was obtained by weighing and drying a small sample of pelleted cells in an oven at 104 "C. These values were used to calculate the intracellular concentration of metabolites and intracellular pH. To make the calculations it was assumed that the metabolites are evenly distributed throughoutthe intracellular water. Evaluation of Glycolysis-The glycolytic flux was evaluated by the amount of lactate produced, the release of labeled anions (mainly lactate and pyruvate) from [U-'4C]glucose into the incubation medium, and the consumption of [U-"C]glucose. Samples of cells incubatedin the presence of 5 mM [U-'4C]glucose (0.2 pCi/ml) were centrifuged at 10,000 X g for 30 s. Portions (0.2 ml) of the supernatants were diluted &fold with 0.1 M formic acid and applied on Dowex-1 formate (X8) columns (0.9 X 1.6 cm). Residual [U-"C] glucose was washed out with 3 ml of 0.1 M formic acid. '"-Labeled anions were eluted with 4 ml of 0.4 M ammonium formate. Radioactivity was measured in both column fractions. [U-"C]Glucose consumption was calculated as thedifference between the amount added and thatrecovered in the medium after incubation. Enzyme Preparations-Cells washed free of erythrocytes were resuspended in 5 volumes of buffer B (50 mM HEPES, 100 mM KC1,5 mM MgCl,, pH 7.4), mixed with 10 times their weight of glass beads (0.5-mm diameter), and vigorously shaken by vortexing for three 1min periods at 1-min intervals. During the intervals, the tubes were kept on ice. The extracts were then centrifuged at 31,000 X g at 4 "C for 30 min. Phosphofructokinase activity was partially purified by ammonium sulfate fractionation of the supernatant at30-43% saturation. The precipitate was collected, dissolved in buffer B (0.5 ml/g of cells) that contained 20% (v/v) glycerol, and filtered through a column of medium Sephadex G-25 (0.5 X 22 cm) equilibrated and eluted with buffer B, containing 20% (v/v) glycerol. The resulting preparation had a specific activity of 0.8 unit/mg of protein measured

*

with 2 mM fructose-6-P and 1 mM MgATP, pH 7.2. For the prepaextracts were made as described ration of 6-phosphofructo-2-kinase, above except that the buffer employed was 50 mM HEPES, 100 mM KCl, 20 mM KF, 0.5 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride, 2.5 pg/ml leupeptin, pH 7.4. 6-Phosphofructo-2-kinase activity was partially purified from the 31,000 X g supernatant by fractionation with PEG. The precipitated material between 6 and 15% (w/v) was dissolved in 10 mM HEPES, 5 mM EGTA, 5 mM EDTA, pH 7.0, containing 20% (v/v) glycerol (0.5 ml/g of cells) and filtered through a column of medium Sephadex G-25 (0.5 X 22 cm) equilibrated and eluted with the same buffer. This preparation had a specific activity of 22 microunits/mg of protein measured with 2 mM fructose-6-P, 2.8 mM MgATP, and 7 mM Pi, pH 7.2. When 6phosphofructo-2-kinase and fructose-2,6-bisphosphataseactivities were measured in extracts of cells treated with selected amino acids, extracts were made and filtered on Sephadex G-25 as described in the preparation of 6-phosphofructo-2-kinase. Enzyme Assays-Phosphofructokinase activity was measured in an assay mixture that unless otherwise indicated contained 50 mM HEPES, 0.1 M KCl, 5 mM MgCl,, pH 7.2, 0.15 mM NADH, 2.8 mM MgATP, 1.2 units of aldolase, 10 units of triosephosphate isomerase, 1 unit of glycerol-3-P dehydrogenase, and 10 p1 of the enzyme preparation in atotal volume of 1ml. After 4 min, the reaction was started by adding 0.16 mM fructose-6-P and was followed by measuring the absorbance change a t 340 nm minus 400 nm in the dual wavelength spectrophotometer at 37 "C. Glucose-6-P was always added to fructose-6-P in the proportion 3:l. Auxiliary enzymes were desalted by centrifugation and dialysis for several hours against 10 mM HEPES, pH 7.0, 20% (v/v) glycerol. 6-Phosphofructo-2-kinase activity was measured by following the production of fructose-2,6-Pz. Unless otherwise indicated, the assay mixture contained 50 mM HEPES, 0.1 M KCl, 10 mM MgCl,, pH 7.0, 1mM Pi, 1 mM fructose-6-P, 1 mM MgATP, and 60 pl of the enzyme preparation in a total volume of 400 pl. After incubation for 10 min a t 37 "C, the reaction was terminated by addition of 1 volume of 0.1 M NaOH followedby heating at 80 "C. After cooling on ice, the reaction mixture was centrifuged for 10 min at 15,000 X g and the supernatant assayed for fructose-2,6-Pz as described above. Under these conditions the production of fructose-2,6-Pz was linear for at least 20 min. Glucose-6-P was always added to fructose-6-P in the proportion 3:l. Fructose-2,6-bisphosphataseactivity was measured by following the disappearance of fructose-2,6-Pz.The assay mixture contained 50 mM HEPES, 0.1 M KCl, 5 mM MgClZ,pH 7.0, 10 p M fructose-2,6-Pz, and 150 plof the enzyme preparation in a total volume of 500 pl. This mixture was incubated at 37 "C, and samples were removed at time intervals, mixed with 1volume of 0.1 M NaOH, heated at 80 "C, and the remaining fructose-2,6-P2 was measured as above. Under these conditions the disappearance of fructose-2,6-Pz was linear for a t least 5 h. One unit of activity is defined as the amount that catalyzes the conversion of 1 pmol of substrate/min under these conditions. Other Assays-Intracellular pH was measured in samples of cell incubations by determining the distribution of [7-"C]benzoic acid essentially as described by Doppler et al. (30). Protein was determined by the method of Bradford (31). RESULTS

Effect of Certain Amino Acids onFructose-2,6-Pz ContentFructose-2,6-P2 is the most potent effector of glycolysis because of the strong allosteric activation exerted on phosphofructokinase (32). Therefore, the first stepin our investigation was to ascertain whether glutamine and asparagine modified the levels of fructose-2,6-Pz. The content of this compound increased 27 times after incubation of tumor cells with 5 mM glucose for 20 min and then remained nearly constant (Fig. 1).Addition of 1mM of either glutamine or asparagine reduced the contentof fructose-2,6-Pzby 70-80% after incubation for 180 min. The effect of the amino acid wasapparent only after 20 min of incubation, when the levels of fructose-2,6-P2 reached a maximum. A similar effect was observed when both amino acids were added together or when each of them was added after preincubation of tumor cells with glucose for 40 min (data not shown). As seen in Fig. 1, glutamate exhibited

Inhibition of Glycolysis Acids Amino by

7811

mM for isoleucine, 0.17 mM for glutamine, 0.35 mM for glutamate, and >1.0 mM for aspartate (data not shown). The concentrations of these amino acids in mouse plasma on the 7th day after ascites tumorinoculation are 0.1 mM for proline, 0.03 mM for asparagine, Co.09 mM for isoleucine, 0.26 mM for glutamine, 0.02 mM for glutamate, and 0.04 mM for aspartate, as reported by Marquez et al. (33). Therefore, this effect was obtained with physiological concentrations of most of these compounds. Evidence for Decreased Glycolysis in the Presence of Certain Amino Acids-Further studies were carried out to investigate the influence of glutamine and asparagine on glucose metabolism. Proline was also included in these studies to gain some insight into the significance of the change observed in fructose-2,6-P2, since this was the most potent amino acid in reducing its levels without apparent modification in the content of hexose monophosphates. Tumor cells incubated with 5 mM glucose produced lactate at a rate of 0.7 & 0.1 pmol/min/g cells, which is within the I I I 1 I I range of values reported by others (11,34).Neither glutamine, I asparagine, nor proline, used at a concentration of 1 mM, significantly modified the time course of lactate production I I during incubation up to 180 min in the presence of sugar (data notshown). However, the observation that hexose monophosphates accumulated when glutamine or asparagine were added to theincubation medium (Table I) suggested a change in the glycolytic rate. Accordingly, we measured the release of labeled anions (mainly lactate and pyruvate) from [U-"C] glucose into the medium to asses if lactate derived from the amino acid may compensate for a decrease in lactate production from glucose. Fig. 2A shows that glutamine and asparagine diminished the production of labeled anions by about 45% after incubation for 180 min. The glycolytic flux progressively slowedafter 60 min of incubation and was reduced by about 80% between 120 and 180 min (0.17 and 0.10 pmol " of lactate plus pyruvate/min/g cells in the presence of glutaX) 60 9 0 1 2 o I S o ~ mine and asparagine, respectively, uersm 0.72 pmol/min/g cells in the control with glucose alone). The decrease in lactate Time (mid production was paralleled by a practically stoichiometric reFIG. 1. Effect of selected amino acids on content of fructose2,6-Pa of ascites tumor cells as a function of time. Cells were duction of [U-'4C]glucose consumption under similar condiincubated with 5 mM glucose without further addition (control con- tions (Fig. 2B). A similar diminution in glycolysis was obditions) or plus 1 mM amino acid as indicated. Each point shows the served when the inhibitory amino acid was added after preinmean f S.E. of three to five separate experiments. Differences be- cubation of tumor cells with glucose for 60 min (data not tween incubations with glucose and glucose plus amino acid that are shown). Fig.2, A and B also shows that, in contrast to statistically significant are indicated by * ( p < 0.05),** ( p < 0.01). Only significance values corresponding to incubations for more than glutamine and asparagine, glucose metabolism was affected 30 min are shown to simplify the figure. When nearby points had the much less by proline. same significance value this is shown only once. The effect of glutamine and asparagine was concentration-

a pattern similar to that of glutamine, while aspartate had a much smaller effect. Among the 20 L-amino acids, only isoleucine and proline produced an effect on the levels of fructose-2,6-P2similar to that of asparagine, glutamate, andglutamine (Table I). NH,Cl was without effect, either alone or in thepresence of aspartate. None of the amino acids that elicited a change in the content of fructose-2,6-P2 modified the production of total lactate by tumor cells after incubation for 180 min. Addition of either asparagine, glutamate, or glutamine increased the content of hexose monophosphates by about 50%,suggesting a decrease in the flux through the phosphofructokinase reaction. However, no significant change in hexose monophosphates occurred with aspartate, isoleucine, and proline. The decrease in fructose-2,6-P2 was dose dependent, and the concentrations of amino acids required for half-maximal effect were 0.05 mM for proline, 0.07 mM for asparagine, 0.11

" 1

'1

TABLEI Effect of amino acids on fructose-2,6-Pzand hexose monophosphates content and lactate production in ascites tumor cells Cells were incubated for 180 min with 5 mM glucose and 1 mMof the indicated additions. Data from the addition of amino acids that had no significant effect on fructose-2,6-P2 content are presented together as "all other amino acids." Results show the mean f S.E., with the number of separate experiments in parenthesis. Significant differences between incubations with glucose alone and glucose plus addition are indicated by * ( p < 0.05) or ** ( p < 0.01). Addition

Fructose-2,6-P, Fructose-6-P nmol/g cells

None Asparagine Aspartate NH&I Aspartate + NH&l Glutamate Glutamine isoleucine Proline All other amino acids

2.6 f 0.2 (7) 120.3 0.6 f 0.1 (3)** 2.0 & 0.3 (9) 132.0 3.0f 0.4 (5) 2.0 f 0.1 (6)* 0.7 f 0.1 (4)** 0.9 f 0.2 (3)** 137.9 0.9 -r 0.2126.2 (3)** 0.7 f 0.1 (5)** 2.7 ? 0.1 (3-6)

+

Lactate

glucose-6-P mmol/g cells

f 6.4 (3)

120.3 f 3.6 (3) 0.26 f 1.3 (3) 0.18 132.0 f 1.3 (3) 0.27 f 0.6 (3) 0.25 f 2.0 ( 3 ) 126.2 f 2.0(3)0.19

0.17f 0.02 (6) f 0.02 (6)** f 0.01 (3) f 0.04(6)* k 0.04(3)* 0.21 f 0.01 (3) f 0.03(3)

Inhibition of Glycolysis Acids by Amino

7812 1

I

I

I

1

Control +glutamine A + asporagine 0

T

1

Fructose-1.6-P2 0.08

3

O u t

Y

Time (mid FIG. 2. Effect of glutamine, asparagine, and proline on the release of"C-labeled anions ( A )and [U-"C]glucose consumption ( B )by ascites tumor cells 88 a function of time. For details see legend for Fig. 1.

dependent. Half-maximal decrease in the rate of release of labeled anions, evaluated between 120 and 180 min incubations, was obtained with 0.14 mM glutamine and 0.087 mM asparagine (data not shown), This effect is relevant to the situation in vivo because these values are in the order of the physiological concentrations of these amino acids (33). Glutamate (1 mM) also decreased the release of labeled anions from glucose, after incubation of cells for 180 min to anextent similar to that of glutamine or asparagine, whereas all other amino acids used at the same concentration were without effect (data not shown). The influence of glutamine on glycolysis was not modified by 0.2 mg/ml cycloheximide (data not shown). Changes in key glycolytic intermediates during the inhibition of the pathway are shown in Fig. 3. Incubation of tumor cells with glucose caused an increase in the contents of glucose-6-P plus fructose-6-P, fructose-1,6-Pz, and triose phosphates. Accumulation of these metabolites has also been observed by other authors upon incubation of ascites tumor cells with sugar for shorter periods of time (34,36). However, addition of either glutamine or asparagine further increased the content of hexose monophosphates to about 70% over the control values, whereas the contents of fructose-1,6-Pz and triose phosphates drastically diminished under the same conditions. This effect was apparent after incubation for 60 min and remained over the next 120 min. Hence, a crossover point was produced at the step catalyzed by phosphofructokinase evidencing a strongdecrease in theflux through this reaction. Incubation in the presence of proline changed the contentsof these intermediates inthe same direction but toa much lesser extent, in agreement with the lower effect promoted by this amino acid on the operation of glycolysis. The levels of both

".

"

I

I

60

120

1

0.

Time (mid FIG.3. Effect of glutamine, asparagine, and proline on content of selected glycolyticintermediates of ascites tumor cells as a function of time. Glucose-6-P and fructose-6-P were measured separately, but theirvalues were added to simplify the figure. TrwseP indicates glyceraldehyde 3-phosphate plus dihydroxyacetone phosphate. Dashed lines have been drawn between the points obtained at zero time and 60 min, as thechanges in the content of some of these intermediates in ascites cells are known to be not linear during the early minutes of incubation (34-36). Each point shows the mean & S.E.of three to six separate experiments. Incubation conditions and other details were as described in the legend for Fig. 1.

Inhibition of Glycolysis by Amino Acids

7813

P-enolpyruvateandpyruvateincreasedafter 60-180 min when glutamine or asparagine were added to the incubation medium; thesechangestoo were greaterthanthechange produced by proline. The 4-fold increase in P-enolpyruvate over the control values may reflect a n inhibition of pyruvate kinase relatedto thelarge decrease in the contentof fructosel,6-Pz, a potent allosteric activator of this enzyme (37). This is supported by the intracellular concentrations of fructose(29 and 330 p ~respectively, , after l,6-P2 and P-enolpyruvate incubation with glutamine or asparagine, as calculated from Fig. 3), that agree with the K,,, value of 25 p M for fructosel,6-Pz at 250 p~ P-enolpyruvate reported for pyruvate kinase from ascites tumor (37). This hypothesis could also explain the lower increase in P-enolpyruvate observed after incubaas thisamino acid produced a smaller tionwithproline, decrease in fructose-1,6-P2. The pyruvate that accumulated about 12-fold above the control, after incubations with glutamine or asparagine for 60-180 min, must therefore have originated from the aminoacid. The Mechanism of PhosphofructokinaseInhibition-One may think that the marked inhibition of phosphofructokinase and glycolysis promoted by glutamine and asparagine was due i€ to the concomitant decrease in the contentof fructose-2,6-P2. However, proline affected the operation of glycolysis very FIG. 4. Effect of fructose-2,6-Pzon the activity of partially little in spiteof being the most potent aminoacid in decreas- purified ascites tumor phosphofructokinase at physiological ing fructose-2,6-P2, andisoleucine, which also greatly dimin- concentrations of substrates and effectors. The assay contained ished the levels of fructose-2,6-P2, had no effect on the gly- 0.16 mM fructose-6-P, 2.8 mM MgATP, 7 mM Pi, 0.05 mM AMP, 0.5 colytic flux (see above). Furthermore, as shown in Fig. 4, the mM NH,+, 0.92 mM magnesium citrate, and 0.34 mM P-enolpyruvate. These concentrations corresponded to the intracellular concentramaximal change found in the concentration of fructose-2,6- tions determined after incubation of cells in the presence of 1 mM Pz, from 6 p M after 180 min of incubation of cells with 5 mM glutamine for 120 min and were calculated from data shown in Figs. glucose to about 2 p~ after incubation in the presence of 3 and 5. The Pi concentration was determined under similar conditions. The NH,+ concentration was taken from Ref. 23. The assay sugar plus 1 mM glutamine or asparagine, produced only a 10% decrease in phosphofructokinase activity assayed under was carried out at pH 7.2, as this was the intracellular pH value conditions near to those prevailing in vivo when glycolysis measured under theseconditions. 10% PEG was added to thereaction mixture to increase the local enzyme concentration for abetter was most strongly inhibited,i.e. upon incubation of cells with simulation of the situation in vivo, since phosphofructokinase from either of these amino acids from 120 to 180 min.2 Neither ascites tumor and several other sources is known to exhibit concenglutamine, asparagine, nor proline (1mM) had any effect on tration-dependent activity (38). Arrows indicate the intracellular concentration of fructose-2,6-Pzdetermined after incubation of cells the activity of isolated ascites tumor phosphofructokinase, and this activity was not changed in gel-filtered extracts of without (6 p M ) and with 1 mM glutamine (2 p M ) for180 min, as calculated from data shown in Fig. 1. cells incubatedwithany of theseaminoacids(datanot shown). observed increase of this metabolitemay contribute very little We measured the levels of other phosphofructokinase efto the enzyme inhibition, since its highest intracellular confectors, including P-enolpyruvate,Pi,theintracellularH+, citrate, and adenine nucleotides in ascitescells incubated with centration in the presenceof amino acid, 0.33 mM, was below the aminoacids to find out if any of these mightbe responsible thereported Kivalue, 0.7 mM (42).Thecontent of Pi, a for the observed inhibition of glycolysis. Whatever the signal positive effector, decreased from 3.3f 0.1 pmol/g at zero time to 1.1 f 0.1 pmollg after 180 min of incubation with 5 mM might be, it should be generated during incubation with either glutamine or asparagine but toa much lesser extent or nota t glucose, but this change was not altered significantly upon all with proline. Fructose-1,B-P2 and NH,' were not consid- addition of any of the three amino acids to the incubation medium (data not shown). The calculated intracellular pH ered since the activity of the ascites tumor isozyme is not sensitive to the former (42) and the latter,also a n activator, remained constant at about pH 7.2 after incubations up to is known to accumulate upon incubation of these cells with 180 min either in the absence or the presence of any amino glutamine or asparagine(23).P-enolpyruvate is known to acid (data not shown). Fig. 5 shows that citrate, a negative inhibit ascites tumor phosphofructokinase (42). However, the effector, increased two to three times after incubation with amino acid, but no significant difference was observedbeThe proportion of free fructose-2,6-Pz available for phosphofruc- tween the three amino acids. Fig. 5 also shows the changes tokinase in ascites cells is unknown. Nevertheless, it is likely to be found in adenine nucleotides. Incubation of cells with glucose substantially higher than that reported in liver by Hue et al. (39), since fructose-1,6-bisphosphataseactivity, the protein that accounts resulted in a 40% decrease in the content of ATP after 120 for more than 80% of the total binding capacity of this tissue for min, no significant change of ADP, and a decrease of about fructose-2,6-P2,is two orders of magnitude lower in ascites cells, 0.23 60% AMP, the levels of total adenine nucleotides being dif 0.01 unit/g according to Gumaa and McLean (34) versus 15 units/ minished. A rapid and eventuallyreversible decrease of ATP g in liver (39 and references therein). Moreover, the activity of 6- and total adenine nucleotidesupon addition of glucose to ph0sphofructo-2-kinase/fructose-2,6-bisphosphatase, which contrib- starved ascites tumor cells has been observed by others (34, utes to the binding of this compound to a much smaller extent (39), is also lower in ascites cells, we measured 0.27 f 0.05 milliunit/g for 36, 43). When glutamine, or asparagine, were added to the the kinase and 1.7 f 0.2 milliunits/g for the bisphosphatase in these incubation medium, thecontent of ATP was maintained cells, versus 3.2 milliunits/g (40) and 10 milliunits/g (41), respectively, considerably higher, over 80% after 120 min of incubation, reported in liver. AMP levels further decreased to about 24% of the control

7814

Inhibition of Glycolysis by Amino Acids

04

03 02 0.1

'F !

0

I

I

I

-

= t

'3i 0.50

ATP

0.25

tion of glutamine (1827 k 47 M-') or asparagine (1805 k 63 M").3 This is consistent with the observed inhibition of phosphofructokinase. An increase of ATP from 1.4 to 2.8 mM, the intracellular concentrations calculated after 120 min of incubation without and with the inhibitory amino acid, respectively, resulted in a 30% decrease in the activity of ascites tumor phosphofructokinase assayed at the concentrations of fructose-6-P and relevant effectors determined after incubation with glutamine or asparagine, as described in the legend for Fig. 4 and in the presence of 2 PM fructose-2,6-Pz. By the same token, the decrease of AMP from 0.15 to 0.04 mM, observed under the same conditions caused a 50% decrease of enzyme activity. Hence, the individual change in ATP and AMP concentration may elicit a modification of phosphofructokinase activity greater than that promoted by other effectors, such as fructose-2,6-P2 or P-enolpyruvate, although not enough to fully account for the decrease of the glycolytic flux found in uiuo. This is not surprising, since phosphofructokinase is well known to be a multimodulated enzyme the net activity of which results from the balance between its various regulatory signals (38, 45, 46). Therefore, we assessed the joint effect of metabolites on the enzyme activity under conditions comparable to those of tumor cells during the observed inhibition of glycolysis. As shown in Table 11, the reaction rate at concentrations of substrates and effectors determined after incubation of cells for 120 min in the presence of glutamine was only 29% of that measured under the conditions of the control. Thus, the change in metabolite levels may account for practically 90% of the inhibition of glycolysis found upon incubation of cells with glutamine or asparagine since the glycolytic flux decreased about 80% under these conditions. In contrast, under the conditions of incubation in the presence of proline the enzyme activity was as much as 65% of the control, in agreement with the lower effect produced by this amino acid on the operation of glycolysis. Studies on the Mechanismof the Decrease of Fructose-2,6Amino Acids-6-Phosphofructo-2-kinase and

P2 Content by

ATP+ADP+AMP 0.25 I

60

I 120

I80

Time ( m i d FIG. 5. Effect of glutamine, asparagine, and proline on content of citrate and adenine nucleotides of ascites tumor cells as a function of time. Each point shows the mean f S.E. of three to six separate experiments. Incubation conditions and other details were as described in the legends for Figs. 1 and 3.

values, ADP was unaffected, and theadenine nucleotide pool increased. In contrast, ATP content scarcely changed in the presence of proline, as compared with the control, and the decrease of AMP was substantially smaller. The calculated [ATP]/[ADP] x [Pi] ratio or cytosolic phosphorylation potential(44) did not significantly change after 120 min of incubation with proline (1004 k 23 K ' ) versus the control (924 k 73 M-'), whereas it increased about 2-fold after addi-

fructose-2,6-bisphosphatase, the enzymatic activities involved in the synthesis and degradation of fructose-2,6-Pz, respectively, have been identified andpartially characterized in ascites tumor cells (8).No significant change in these activities was observed when they were measured in gel-filtered extracts of cells incubated from 60 to 180 min with any of the amino acids that induced a fall in the content of fructose-2,6P2 (data not shown). This suggests that the decrease of this compound was not mediated by a stable modification of 6phosphofructo-2-kinase/fructose-2,6-bisphosphatase but by the change in the intracellular concentration of some of the metabolites that regulate these activities. In this case, the changes responsible should be apparent upon incubation of cells with either glutamine, asparagine, or proline, since the three of them promoted a similar decrease in fructose-2,6-P2 content (Table I).4 This condition was not met by several of the potential regulators already determined during the incubations, such as Pi, fructose-6-P, P-enolpyruvate:, or ATP. If the free cytosolic [ADP] is assumed to be about 20-fold lower than measured [ADP] content, as suggested by Veech et al. (44) in mitochondrial-containing tissues, then the calculated values for the cytosolic phosphorylation potential would be 20 times higher. 'Neither glutamine, asparagine, nor proline (1mM) had any effect on 6-phosphofructo-2-kinase and fructose-2,6-bisphosphataseactivities from ascites tumor cells (data not shown). The increase of P-enolpyruvate detected after 120 min of incubation of cells in the presence of proline (Fig. 3) did not significantly modify 6-phosphofructo-2-kinaseactivity assayed at theintracellular concentrations of fructose-6-P, ATP, Pi, and citrate determined under those conditions (data not shown).

7815

Inhibition of Glycolysis by Amino Acids TABLEI1

Activity of partially purified ascites tumor phosphofructokinase at concentrations of substrates and effectors present in ascites tumorcells under various incubations conditions The concentrations of substrates and effectors were calculated from the data shown in Figs. 3 and 5 after incubations of ascites tumor cells for 120 min in the presence of glucose without and with either glutamine or proline. The reaction mixture also contained 7 mM Pi, 0.5 mM NH:, and 10% PEG. The assay was carried out at pH 7.2. Results show the mean k S.E. of three determinations. Conc. of substrates and effectors Activity Condition of incubation AMP Fructose-2,6-P2 P-enolpyruvate citrate Fructose-6-P ATP mM

5 mM glucose +1mM glutamine +1mM proline

0.10 0.16 0.12

1.4 2.8 1.6

0.15 0.04 0.085

0.07 0.34 0.12

0.006 0.92 0.002 0.87 0.002

0.40

rnrnol/rnin/rng

%

0.147 f 0.008 0.043 f 0.005 0.096 2 0.006

100 29 65

r

However, citrate, an inhibitor of 6-phosphofructo-2-kinase 0 Controls from liver (47, 48), muscle (47), and yeast (49), increased in a similar way after incubation of cells with any oneof the three “C-Anions aminoacids (Fig. 5).Wefoundthatcitrateinhibitedthe ascites tumor enzyme with a Ki value of 0.45 mM, its effect being t o decrease the V,, without significantly affecting the K,,, (data not shown). This Kivalue was within the range of the intracellular concentration of citrate in ascites tumor cells. An increase in the concentration of citrate from 0.4 to 0.9 mM, the values obtained after 120-min incubations without Fructose-2.6-P2 and with amino acid, respectively, produced a 40% decrease activity measured at physiologiin 6-phosphofructo-2-kinase cal concentrations of substrates and Pi. Furthermore, this activity decreased 51-55% when determined at the concentrations of fructose-6-P, ATP, Pi, and citrate calculated after 120 min of incubation of cells in the presenceof either of the threeaminoacidscomparedwiththecontrolexperiment. Fructose-Ib-P, Therefore, the changes in the levels of citrate may play an important role in accounting for the observed diminution of fructose-2,6-Pz. The Effect of Inhibiting Transamination on the Metabolic Changes Induced by Amino Acids-The phenomena described above were specifically produced by certainamino acids, which therefore must exhibit some common metabolic feature. There is no structural relationbetween the R groups of all of Citrate Triose-P 1 . . I them. However, transamination with a-ketoglutarate is involved in their degradation. This reaction is an intermediate step in the utilization of asparagine, aspartate, and isoleucine. Glutamate, a final productof the catabolismof glutamine and proline, is fed into the citric acid cycle to generate ATP via glutamate dehydrogenase or transamination. To investigate the role of transamination, we studied the effect of aminooxyacetate, a potent inhibitor of transaminases in general 3 (50), on relevant metabolic changes induced by amino acids in ascites cells (Fig. 6). Amino-oxyacetate (1 mM) prevented the effects of glutamine, asparagine, or proline, as shown by the decrease in therelease of 14C-labeledanions andof glucose consumption, fructose-1,6-P2, triose phosphates, and fructoseFIG. 6. Effect of amino-oxyacetate on the release of ‘“C2,6-P2 and the accumulation of hexose monophosphates, ATP, labeled anions, the consumption of [U-14C]glucose,and the and citrate. T

1 8

DISCUSSION

Glutamineandasparaginecausedan 80% inhibition of glycolysis in ascites tumor cells which was specific and dosedependent. This interaction was mediated by a strong inhibition of phosphofructokinase activity (Fig. 3), which can be accounted for by changes in various allosteric effectors (Table 11). The major change responsible for this phenomenon is the increase of ATP and the decrease of AMP produced in the presence of glutamine or asparagine (Fig. 5). These amino acids aswell as several others brought about a marked fallof fructose-2,6-Pz content (Fig. 1 and Table I), which, although

content of selected metabolites in the presence of either glutamine, asparagine, or proline in ascites tumor cells. Cells were incubated for 180 min with (black bars)or without (white bars) 1 mM amino-oxyacetate in the presence of 5 mM glucose and 1 mM amino acid as indicated.Results show the mean f S.E. of three to six separate experiments. Differences between incubations with glucose and glucose plus any other addition that are statistically significant are indicated by * ( p < 0.05), ** ( p < 0.01).

specific (Table I) and dose-dependent, was of less significance, since it modified very little the activity of phosphofructokinase assayed under physiological conditions (Fig. 4). On the other hand, some of the amino acids that decreased fructose2,6-Pz had littleor no effect on the rate of glycolysis. Inhibi-

7816

Inhibition of Glycolysis byAmino Acids

tion of phosphofructokinase cangive rise to a sequential series effects provoked byglutamine and asparagine, since they were of secondary feedback mechanisms thatreinforce the decrease suppressedwhenthisreaction was inhibited with aminoin glycolysis, mainly an inhibition of hexokinase by the in- oxyacetate (Fig. 6).Two otherconsequences follow from these crease of glucose-6-P (9, 35), and of pyruvate kinase through observations. First, they evidence that glutamate catabolism the diminution of fructose-1,6-P2 (37). The high levels of in ascites tumor cells basically proceeds via transamination, ATP, also an inhibitor of pyruvate kinase from ascites cells as the increase of ATP content detected in the presence of (37), induced by glutamine and asparagine,may contribute to either of these amino acidswas blocked by amino-oxyacetate. the drop in the activity of this enzyme as well. Hence, the This agrees with data of Moreadith and Lehninger (13) and inhibitory amino acids tend to reestablish the Pasteureffect, Matsuno (19) reporting that this compound totally inhibited the oxidation of glutamate and glutamine by mitochondria i.e. the inhibition of glycolysis by respiration, and our data suggest that they do so by triggering the inhibition of phos- isolated from ascites tumorcells and chicken hepatomacells, phofructokinase through a change in adenine nucleotides. If respectively. Krebs and Bellamy (60) suggested a minor role these results are operativein uiuo, previous studies indicating for glutamate dehydrogenase in the conversion of glutamate being the preferred route a high rate of glycolysis in ascites tumor cells, carried out in into a-ketoglutarate, transamination the absence of amino acids (34) or upon incubationfor shorter for this process. Secondly, these experiments show that the regulatory signals involved in the metabolic changes induced periods of time ( l l ) , should be re-evaluated. The role of phosphofructokinase as the step controlling theby amino acids are dependent on the operation of the citric flux through glycolysis in ascites tumor cells has been ques- acid cycle, which flux decreased upon inhibiting transaminationed (51) on the basis of published data showing an accu- tion. It should be recalled that only 1.5% of glucose appears mulation of thereactionproduct,fructose-1,6-P2,in cells to be utilized by the citric acid cycle in ascites tumor cells incubated with glucose as thesole energy source. The authors (11).Therefore, these findings further support the importance related the capacityof the tumor tometabolize large amounts of thechangesinATPandcitratetocontributetothe of glucose to thelack of regulation at thephosphofructokinase observed inhibition of glycolysis and the fall of fructose-2,6suggest that thereason why level. This view is not supportedby our results. Fructose-1,6- Pz,respectively. It is tempting to P2accumulated in the control experiment, pointinga to rate- other amino acidswere without effect on glucose metabolism use as energysources by tumor cells. limiting step further down the pathway under these condi- may rest on their limited In conclusion, the results presented here clearly show that tions. However, this metabolite pattern was completely reversed when cellswere incubated with glucose plus glutamine the glycolytic function in ascites tumor cells is dependent on or asparagine, which in fact resembles a more physiological glutamine and asparagine utilization,hence, allowing a coorsituation. The further increaseof glucose-6-P and fructose-6- dinated operation for these two important aspects of their P and the large decrease of fructose-1,6-Pz and triose phos- energetic metabolism. It would be of interest to determine cells or phates induced by the amino acids (Fig. 3) demonstrate the whether this interaction is restricted to ascites tumor it also occurs in other types of cancer and rapidly dividing strong limitation to the glycolytic flux imposed by the phoscells. phofructokinase reaction. It is noteworthy that the metabolic changes induced by Acknowledgment-We thank Dr. C. Gancedo for critical reading of asparagine were remarkably similar to those promoted by the manuscript. glutamine, indicating that the formeris as efficiently metabolized as the latter. Both amino acids contributed to maintain REFERENCES high levels of adenylates, in agreement with data reported by 1. McKeeban, W. L. (1982) Cell. Biol. Int. Rep. 6 , 635-650 2. Newsholme. E. A,. Crabtree., B... and Ardawi. M. S. M. (1985) . . Biosci. ReD. others (11, 52, 53) showing that the oxidation of glutamine 6, 393-400 by these tumor cells results in a significant increase of ATP 3. Warburg, 0.(1926) Science 123,309-314 4. Weinhouse, S. (1972) Cancer Res. 10,2007-2016 and the total adenine nucleotide pool. This is apparently 5. Weber, M. J. (1973) J. Biol. Chem. 248,2978-2983 caused by an activation of the synthesis de novo of purine 6. Racker, E. (1976) J. Cell. Physiol. 89,697-700 7. Weber, G. (1977) N . Engl. J. Med. 296,541-551 nucleotides. T o our knowledge, an effect of asparagine on 8. Bosca, L., Aragon, J. J., and Sols, A. (1985) Curr. Top. Cell. Reg. 27, 411glucose metabolism has not been described previously. With 418 9. Nakashima, R. A,, Paggi, M. G., Arora, K. K., and Pedersen, P. L. (1990) respect to glutamine, other workers haveshowneffects of in Integration of Mitochondrial Function (Lemasters, J. J., Hackenbrock, various kinds. In addition to reports on L929 cells (21) and C. R., Thurman, R. G., andWesterhoff, H. V., eds) pp. 405-411, Plenum Publishing Company, New York chick embryo fibroblasts (20) mentioned above, Ardawi and 10. Krebs, H. A. (1980) in Glutamine: Metabolism, Enzymology and Regulation (Mora, J., andPalacios, R., eds)pp. 319-329, Academic Press, New York Newsholme (54) described a stimulation of lymphocytes gly11. Lazo, P. A. (1981) Eur. J. Biochem. 117,19-25 colysis due, in part, to an activationof phosphofructokinase. 12. Coles, N. W., and Johnstone, R. M. (1962) Biochem.,J. 83,284-291 13. Moreadith, R. W., and Lehninger, A. L. (1984) J. E d . Chern. 2 5 9 , 6215In isolated hepatocytes,glutamineandotheraminoacids 6221 stimulate glycogen synthesis and lipogenesis and inhibit ke- 14. Kovacevic , Z., Brkljac, O., and Bajin, K. (1991) Biochern. J. 273,271-275 togenesis (55) but do not seem to affect levels the of fructose- 15. Lavietes, B. B., Regan, D. H., and Demopoulos, H. B. (1974) Proc. Natl. Acad. Sci. U.S. A. 71,3993-3997 2,6-P2 (56, 57) and lactate production fromglucose (56). 16. Broome, J. D., and Schwartz, J. H. (1967) Biochirn. Biophys. Acta 1 3 8 , 637-639 effect of LKedryna et al. (58) reportedaninhibitory 17. Capizzi, R. L., and Cheng, Y.-C. (1981) in Enzymes as Drugs (Holcenberg, cysteine on glycolysis after 30- and 60-min incubations of J. S., and Roberts, J., eds) pp. 1-24, John Wiley and Sons, New York Kien, C. L., and Holcenberg, J. S. (1981) Cancer Res. 41,2051-2055 ascites tumor cells, which was localized at the step catalyzed 18. 19. Matsuno, T. (1987) Int. J. Biochem. 19,303-307 by pyruvate kinase. We did notobserve a significant effect of 20. Mojena, M., Bosca, L., and Hue, L. (1985) Biochem. J. 232,521-527 21. Lanks, K. W. (1986) J. Cell Physiol. 126,319-321 this amino acid, although our observations were made after 22. Wu, R., and Racker, E. (1959) J. Biol. Chem. 2 3 4 , 1029-1035 incubations for 180 min (Table I). Methionine has alsobeen 23. Carrascosa,J. M., Martinez, P., and NGiiez de Castro, I. (1984) Cancer Res. 44,3831-3835 shown to inhibit lactate production in ascites tumor cells 24. Van Schaftingen, E., Lederer, B., Bartrons, R., andHers, H.-G. (1982) Eur. grown in tissue culture(59). However, this effect was observed J. Biochem. 1 2 9 , 191-195 H. U.,Bergmeyer, J., and Grapl, M. (1984) Methods of Enzywith 10 mM methionine, which is 10 times higher than the 25. Bergmeyer, matic Analysis, Vol. VI, pp. 191-588, Verlag Chemie, Weinheim concentration used by us and well above its plasma levels of 26. Bergmeyer, H. U.,Bergmeyer, J., and Gra,91, M. (1985) Methods of Enzymatic Analysis, Vol. VII, pp. 3-370, VHC Verlagsgellschaft, Weinheim 20-40 PM in tumor-bearingmice (33). 27. Jaworek, D., and Welsch, J. (1985)in Methods of EnzymaticAnalysis (Bergmeyer, H. U., Bergmeyer, J., and Grassl, M., e&) Vol. VII, pp. 365Transamination was required as a necessary step for the '

Inhibition of Glycolysis by 370,VHC Verlagsgellschaft,Weinheim 28. Guynn, R. W., Veloso, D., and Veech, R. L.(1972)Anal Biochem. 46,277-

285 29. Rottenber H. (1979)Methods Enzyml 66,547-569 30. Dop ler, Jaggi, R., and Groner, B. (1987)Gene ( A m t . )64,147-153 31. Braaford, M.M. (1976)Anal Biochem. 72,248-254 32. Van Schaftingen, E. (1987)Adu. Enzymol. 69,315-395 33. Mirquez, J., Shnchez-Jimbnez, F., Medina, M. A,, Quesada, A. R., and Nuiiez de Castro, I. (1989)Arch. Biochem. Biophys. 288, 667-675 34. Gum=, K. A., and McLean, P. (1969)Biochem. J. 116,1009-1029 35. Kosow, D. P., and Rose, I. A. (1972)Biochem. Bwphys. Res. Commun. 48, 376-383 36. Glaser, G., Giloh, H., Kasir, J., Gross, M., and Mager, J. (1980)Bwchem. J. 192,793-800 37. Feliu, J. E.,and Sols, A. (1976)Mol. CeU. Biochem. 13,3144 38. A r a g h , J. J., and Sols, A. (1991)FASEB J . 6 , 2945-2950 39. Hue, L., Sobrino, F., and Bosca, L. (1984)Biochem. J. 224, 779-789 40. Sakakivara, R., Kitajima, S., and Uyeda,K.(1984)J. BbL Chem. 269,4146 41. Van Schaftingen, E.,Davies, D. R., and Hers,H.-G. (1982)Eur. J . Biochem. 124.143-149 , ". 42. Bo%& L., Aragbn, J. J., and Sols, A. (1982) Biochem. Biophys.Res. Commun. 108,486-491 43. Chapman, A. G., Miller, A. L., and Atkinaon, D. E. (1976)Cancer Res. 38, 1144-1150

#.,

~~

Amino Acids

7817

44. Veech. R. L.. Lawson I, J. W. R., Cornell, N. W., and Krebs, H. A. (1979)J. Bioi. CGh.284T6538-6547 45. Tejwani, G. A. (1978)Trends Biochem. Sci. 3,30-33 46. Uyeda,K. (1979)Adu. Enzymol. Relat. Areas Mol. Bwl. 48, 193-244 47. Laloux, M., Van Schaftingen, E., Frangois, J., and Hers, H.G. (1985)Eur. 89 J. Biochem. 148.155-f5 , ". 48. V& SchaftingeGE., and Hers, H.G. (1986)Eur. J. Biochem. 159, 359365 49. Franpois, J., Van Schaftingen, E., and Hers, H.-G. (1988)Eur. J . Biochem. 171.599-608 - -,- - - - - 50. Soling, H.-D., Kleineke, J., Willms, B., Janson, G.,and Kuhn, A. (1973) Eur. J. Bwchem. 37,233-243 51. Bosca, L., and Corredor, C. (1984)Trends Biochem. Sci. 9,372-373 52. Kovacevic, Z., Brkljac, O., and Jerance, D. (1991)Biochem. J. 273, 277^" ~

~

~~~

~~~

53. Yushok, W. D. (1971)J. BWL Chem. 246,1607-1617 54. Ardawi, M. S. M., and Newsholme, E.A. (1983)Biochem. J. 212 835-642 55. Baquet, A., Lavoinne, A. and Hue, L. (1991)Biochem. J. 273 53-62 56. Lavoinne, A., Baquet, A.: and Hue L. (1987)Biochem. J. 248' 429-437 57. Uyeda,K.,Furuya, E.,Richards, Cl S., and Yokoyama, M. (1962)Mol. Cell. Biochem. 48,97-120 58. Kedryna, T.,Guminska, M., and Marcbut, E. (1983)Bhchim. Bwphys. Acta 763,64-71 59. Boerner, P., and Racker, E. (1985)Proc. Natl. Acad. Sci. U. S. A. 82,67506754 60. Krebs, H. A., and Bellamy, D. (1960)Biochem. J. 76,523-529