Fructose 2,6-Bisphosphate and the Control of Glycolysis by Glucocorticoids and by Other Agents in Rat Hepatoma Cells1

[CANCER RESEARCH 45, 4263-4269, September 1985]

Anne M. Loiseau,2 Guy G. Rousseau, and Louis Hue3

Hormone ana Metabolic Research Unit, International Institute of Cellular and Molecular Pathology, and Louvain University Medical School, 75 Avenue Hippocrate, B-1200, Brussels, Belgium

ABSTRACT hepatocytes and HTC cells as models, we have compared sev eral aspects of glycolysis in normal and cancer cells. HTC cells The rate, key enzymes, and several metabolites of glycolysis were chosen because of their high degree of dedifferentiation in rat hepatoma (HTC) cells have been compared to those in rat and high rate of glycolysis (7). Moreover, when studying the hepatocytes. At 5 to 10 mw glucose, lactate release was greater glycolysis of HTC cells, we had found that this process is in HTC cells. This could be explained in part by the absence of stimulated by glucocorticoid hormones, presumably via a recep key gluconeogenic enzymes, by the substitution of glucokinase tor-mediated mechanism (8). We have now examined whether by hexokinase, and by an increase in phosphofructokinase 1 and fructose 2,6-bisphosphate is present in HTC cells and whether pyruvate kinase activity. In addition, fructose 2,6-bisphosphate, their PFK1 is sensitive to this stimulator. We show that this is the most potent stimulator of phosphofructokinase 1, was iden the case and that fructose 2,6-bisphosphate could be involved tified in HTC cells and shown to stimulate phosphofructokinase in the control of glycolysis by glucocorticoids in HTC cells. We 1 partially purified from these cells. have also examined whether agents known to decrease fructose Dexamethasone increased the release of lactate in HTC cells. 2,6-bisphosphate concentration in liver would do so in HTC cells This glucocorticoid increased the concentration of fructose 2,6- and whether this would correlate with an inhibition of glycolysis. bisphosphate and the Vmaxof the enzyme that catalyzes its synthesis, phosphofructokinase 2. The data were consistent with MATERIALS AND METHODS an indirect effect at the gene level, mediated by glucocorticoid receptors. Dexamethasone had no effect on the other rate- Cell Culture. HTC cell lines 4 and 99 are derived from the "wild" type limiting glycolytic enzymes. Several agents (adenosine, dibutyryl originally received from Dr. G. M. Tomkins. The variant cell line 268EC, cyclic adenosine 3':5'-monophosphate, ethanol, antimycin) partially resistant to glucocorticoids, was a gift of Dr. E. B. Thompson. known to decrease fructose 2,6-bisphosphate in hepatocytes All experiments, except those in Table 2, were performed with cell line 4. The cells were grown (9) as suspension cultures in Swim's Medium were without effect on this stimulator in HTC cells. DL-Glyceral- S-77 containing 10% (v/v) newborn calf serum. Where indicated (serum- dehyde inhibited glycolysis in HTC cells and eventually killed them. Although this substance decreased fructose 2,6-bisphos free medium), the latter was replaced by a solution of bovine serum albumin (10 g/liter) in phosphate-buffered saline (0.15 M NaCI:2.5 rriM phate, inhibition of glycolysis through an action at another level KCI:8 rnw Na2HPO4:1.5 rriM KH2PO4, pH 7.45). Glucose concentration in could not be ruled out. the culture medium was 17.5 mM and 18.3 mw in absence and presence of serum, respectively. Incubation of Cells. Two types of incubations were performed. In the INTRODUCTION first one, 2 ml of a suspension containing about 5 x 106 HTC cells/ml of Krebs-Henseleit bicarbonate medium (10) were incubated like hepato Unlike normal cells, many cancer cells maintain a high glycolytic cytes (11 ) at 37Â°Cin a rotatory incubator under an O2:CO2 atmosphere rate under aerobic conditions. The biochemical mechanism of (19:1) for no longer than 1 h, in the presence of glucose at the concen this phenomenon is not fully understood (1-3). The discovery (4) trations indicated. For the measurement of fructose 2,6-bisphosphate, of fructose 2,6-bisphosphate led us to reassess the issue. Fruc 0.5-ml aliquots were taken and immediately frozen in tubes kept in liquid tose 2,6-bisphosphate, a naturally occurring molecule, is the N2; the frozen cells were then extracted in 50 mM NaOH as described most potent stimulator of PFK1," one of the rate-limiting enzymes earlier (12). For the measurement of glucose, lactate, and metabolites, 1 of glycolysis. It is synthetized from fructose 6-phosphate and ml of incubation mixture was deproteinized by 1 ml of 10% (w/v) HCIO4; ATP by PFK2 and hydrolyzed into fructose 6-phosphate and P> the deproteinized supernatants were neutralized by a 3 M KOH:1 M KHCC-3 mixture. by fructose 2,6-bisphosphatase. In liver, fructose 2,6-bisphos In the second type of incubation, performed to study the effect of phate allows glycolysis to proceed even if ATP, an inhibitor of glucocorticoids, the cells in culture medium were agitated in a rotatory PFK1, is present (for a review, see Refs. 5 and 6). Using rat incubator under atmospheric air. Except when mentioned otherwise, the 1This research was supported in part by a grant from the Caisse Generate cells were taken in their exponential phase of growth and resuspended d'Epargne et de Retraite (Belgium) and from the Fonds de la Recherche Scientifique in serum-free medium. Steroids in ethanol or ethanol alone (final concen Medicate (3.4539.81). tration, < 0.01%) was added to the cultures. For the measurement of 2 Fellow of Institut pour l'Encouragement de la Recherche Scientifique dans fructose 2,6-bisphosphate, 10 ml of cells were centrifuged (5000 x g for l'Industrie et l'Agriculture (Belgium). 1 min, 0Â°C),and the cell pellet was kept frozen at -80Â°C until further 3 MaÃ®trede Recherches of the Fonds National de la Recherche Scientifique (Belgium). To whom requests should be addressed, at UCL-ICP 7529, 75 Avenue processed. For the measurement of glucose and lactate, aliquots of the Hippocrate, B-1200 Brussels, Belgium. culture medium were deproteinized as described above. â€¢Theabbreviations used are: PFK1, phosphofructokinase 1 (ATP:o-fructose-6- Enzyme Assays. Frozen pellets of HTC cells (corresponding to about phosphate 1-phosphotransferase, EC 2.7.1.11); PFK2, phosphofructokinase 2 (ATP:o-fructose-6-phosphate 2-phosphotransferase, EC 2.7.1.-); HTC, hepatoma 200 mg of cells) or livers from rats fasted for 24 h were homogenized at 0-4Â°C with an UltraTurrax in 4 volumes of 0.1 M KCI:0.1 mM EDTA:2 tissue culture. Received 10/2/84; revised 1/18/85; accepted 3/25/85. mÂ«MgCbiSO mM 4-(2-hydroxyethyl)-1-piperazineethanolsulfonic acid at

CANCER RESEARCH VOL. 45 SEPTEMBER 1985 4263

pH 7.2. After centrifugation (10,000 x g for 10 min), appropriate dilutions in HTC cells as compared with those in the liver of normal rats of the supernatant were used for enzyme assays. The activities of fasted for 24 h. In agreement with earlier work (7, 24, 25), the hexokinase (EC 2.7.1.1) (13), glucokinase (EC 2.7.1.2) (13), glucose-6- enzyme profile of cancer cells differed quantitatively and quali phosphatase (EC 3.1.3.9) (14), fructose-1,6-bisphosphatase (EC tatively from that of normal cells. The activity of gluconeogenic 3.1.3.11 ) (15), pyruvate kinase (EC 2.7.1.40) (16), lactate dehydrogenase (EC 1.1.1.28) (17), PFK1 (EC 2.7.1.11 ) (18), and PFK2 (EC 2.7.1.-) (19) enzymes was decreased or absent, whereas that of glycolytic were measured under Vâ„¢,conditions. Measurement of tyrosine amino- enzymes was increased. One qualitative change was the replace transferase (EC 2.6.1.5) and alkaline phosphodiesterase I (EC 3.1.4.1) ment of glucokinase by hexokinase. Table 1 also shows the concentration of several metabolites. Hexoses 6-phosphate were was made as described previously (9). For the partial purification of PFK1, frozen pellets of HTC cells and rat more concentrated, and phosphoenolpyruvate was less concen livers were homogenized in 5 volumes of 50 HIM NaF:1 mw dithioerythri- trated in HTC cells than in hepatocytes; no significant difference tol:1 rriM ATP-Mg:50 mw Tris-HCI at pH 8. After centrifugation (10,000 in ATP and fructose 1,6-bisphosphate was observed. Glycogen x g for 10 min), the enzyme present in the supernatant was precipitated was not detectable. by 33% (w/v) ammonium sulfate, collected by centrifugation (15,000 x The rate of lactate release and the concentration of fructose g for 10 min), resuspended in 0.1 M KF:1 HIM dithioerythritol:50 mw 2,6-bisphosphate of HTC cells were measured after incubation potassium phosphate:0.5 mw MgCI2:10% (w/v) glycerol:50 mw Tris-HCI with various concentrations of glucose under aerobic and anaer at pH 7.6, and applied on a column (0.7 x 5 cm) of ATP:agarose obic conditions. The relationship between glucose concentration equilibrated in the latter medium. PFK1 was eluted from the column in the presence of 1 mw ATP-Mg. All the steps were performed at 0-4Â°C. and lactate release was hyperbolic, with an apparent Km for glucose of about 1 rriM (Chart ~\A).In contrast, this relationship The enzyme was purified 100- to 150-fold. The kinetic measurement of PFK1 activity was performed as described previously (20). Analytical Methods. Glucose and glycogen (21), glucose 6-phos- m phate, fructose 6-phosphate, lactate, ATP, and phosphoenolpyruvate ^ 5 were measured enzymically in neutralized HCIO4 extracts (22). Fructose 2,6-bisphosphate was measured in alkaline extracts of HTC cells as N. described previously (12). Protein was determined according to the method of Lowry ef al. (23). "5I Chemicals. All tissue culture components were purchased from Grand Island Biological Co., Paisley, Scotland. Enzymes and biochemical re e agents were either from Sigma Chemical Co., St. Louis, MO, or from Boehringer, Mannheim GmbH, Western Germany. Dexamethasone (Merck, Sharp & Dohme, Rahway, NJ), corticosterone (Steraloids, Wilton, NH), aldosterone (Merck, Darmstadt, Western Germany), progesterone (Steraloids), estradici (Sigma), testosterone (Sigma), dexamethasone mesylate (gift from Dr. M. V. Govindan), RU486 (gift from Roussel UCLAF, Romainville, France), and actinomycin D (Sigma) were obtained as indi cated. 0L 20 50 RESULTS

Glycolysis in HTC Cells. For a given glucose concentration, the overall glycolytic flux is geared by the activity of the rate- u limiting enzymes. Table 1 shows the activities of these enzymes et 2 Table 1 Ih Maximal activities of several key glycolytic and gluconeogenic enzymes and concentration of several metabolitesin HTC cells and in liver of normal rats fasted for 24 h Uver HTC cells ÃŽ Enzymes Enzymeactivity (iÂ¿mol/min/gcells) Hexokinase 3.52 Â±0.40a(6)" Glucokinase 1.8 f ' Phosphofructokinase 1 1.2 Â±0.3(4) 5.02 Â±0.57(6) ÃŽ Pyruvate kinase 45.0 Â±3.1(4) 425 Â±21(6) (O 12.0 NDC(3) Â« Glucose-6-phosphatase CM Fructose-1,6-bisphosphatase 11.3 ND(3) e Metabolites" Metabolite concentration (ii/nol/g cells) +* ATP + 0.17(3) Â±0.26(4) u Glucose 6-phosphate + fructose (3)0.0200.05 Â±0.01 (4)0.0250.32 Â±0.05 6-phosphate I I -MM Fructose 1,6-bisphosphate Â±0.002(3) Â±0.014(3) IO I5 20 50 Phosphoenolpyruvate 0.27 Â±0.02(3) 0.06 Â±0.01(3) Glycogen(glucose equivalents)2.80 <52.63 ND Glucose (nrM) ' Mean Â±SE. Chart 1. Effect of glucose concentration on lactate production (A) and on 6 Numbers in parentheses,number of determinations. fructose 2,6-bisphosphate concentration (8). HTC cells were incubated in the c ND, not detectable. absence (O) or presence (â€¢)of10 MMantimycin.Points, mean for 3 experiments; d Values for liver are from studies with isolated hepatocytes (22, 27). oars, SE.

CANCER RESEARCH VOL. 45 SEPTEMBER 1985 4264

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1985 American Association for Cancer Research. CONTROL OF GLYCOLYSIS IN RAT HEPATOMA CELLS is known to be sigmoidal in hepatocytes from fasted rats (22, fructose 2,6-bisphosphate and AMP, another stimulator of PFK1. 26). This is consistent with the replacement of glucokinase by The concentration of fructose 2,6-bisphosphate required for half- hexokinase. In HTC cells, the release of lactate at physiological maximal stimulation depended on the concentration of other concentrations of glucose (5 to 10 mM) was 1.4 to 1.7 /tmol/min/ effectors. In the presence of physiological concentrations of g cells. This is much greater than in hepatocytes in which lactate substrates and effectors (50 pM AMP, 2.5 mM ATP-Mg, 5 mM PÂ¡, release is barely detectable at glucose concentrations lower than 0.25 mM fructose 6-phosphate), the Ka for fructose 2,6-bisphos 10 mM (22, 26). Chart 1B shows that fructose 2,6-bisphosphate phate was about 3 and 2 /Â¿Mfor the liver and HTC enzyme, is present in HTC cells and that its concentration depends on respectively (Chart 2). Thus, the kinetic properties of PFK1 were the concentration of glucose in the incubation medium, as is the similar in liver and HTC cells. However, the requirement for AMP case in hepatocytes (27). At 5 to 10 mM glucose, the concentra to obtain a stimulation by fructose 2,6-bisphosphate was appar tion of fructose 2,6-bisphosphate was lower in HTC cells (1.5 ently less stringent in HTC than in liver cells (Chart 2). The concentration of fructose 2,6-bisphosphate in HTC cells (see nmol/g cells) than in hepatocytes (5.7 nmol/g cells). Under culture conditions, however, the concentration of fructose 2,6-bisphos above) is within the range at which their PFK1 is sensitive to this phate in HTC cells could be as high as in hepatocytes (see stimulator (Chart 2B). Therefore, in HTC cells, PFK1 is expected to be controlled by endogenous fructose 2,6-bisphosphate. below). Incubation of HTC cells under anaerobic conditions stim Stimulation of Glycolysis by Glucocorticoids. HTC cellswere ulated lactate release but did not modify the concentration of fructose 2,6-bisphosphate (Chart 1) nor the concentration of exposed to a concentration (0.5 MM)of dexamethasone, a sem- isynthetic glucocorticoid agonist, that saturates the glucocorti- ATP (results not shown). In HTC cells, the anaerobic glycolytic coid receptor. As expected, the wild-type lines displayed an flux in Krebs-Henseleit medium (Chart 1^) was comparable to increase in the activity of the tyrosine aminotransferase and that of cells incubated under culture conditions (see Table 2 alkaline phosphodiesterase I, two glucocorticoid-inducible en below). Sensitivity of PFK1 to Fructose 2,6-bisphosphate. The ki zymes (9), and in the glycolytic rate (Table 2). After 24 h in the presence of dexamethasone, lactate concentration in the me netic properties of PFK1 partially purified from liver and HTC dium was increased by 36.7 Â±4.9% (SE) (6 experiments, P < cells were compared. Chart 2 shows the stimulation of PFK1 0.001) in the absence of serum and by 24.7 Â±4.0% (7 experi activity by fructose 2,6-bisphosphate and the synergism between ments, P < 0.001) in the presence of serum. This increase in lactate release is consistent with the finding that, under identical I conditions, the pH of the medium was lower than in control cells (Table 3). It is noteworthy that, over the time course of the experiments, the cells kept dividing at the same rate in the absence or presence of serum, and whether dexamethasone was present or not. In dexamethasone-stimulated cells, lactate release ac 0.5 counted for 74 Â±2% of the glucose consumption (18 experi ments), which is not different from the situation in untreated cells (71 Â±2%, mean for 20 experiments). Increased glycolysis by dexamethasone was independent of stimulation of tyrosine ami X a notransferase and alkaline phosphodiesterase I by this steroid. This is illustrated by the experiments with the variant cell line 268EC. These cells possess normal glucocorticoid receptors, aÂ» Â«Â» 10 but their tyrosine aminotransferase is not inducible (28), and their > alkaline phosphodiesterase I activity is as high as in wild-type cells stimulated by dexamethasone (29). The data in Table 2 B show that, in the variant cell line, stimulation of glycolysis by St dexamethasone persisted when neither of these enzymes was induced by the steroid. incubation of HTC cells for 5 h with 0.5 UM dexamethasone also resulted in an increase in both fructose 2,6-bisphosphate 0.5 and in the activity of PFK2 (Table 3). This is in contrast with the situation in hepatocytes, where dexamethasone does not stim ulate PFK2 activity under conditions when it stimulates tyrosine aminotransferase activity (results not shown). It is noteworthy that PFK2 activity is greater in rat liver (1 to 3 nmol/min/g) than in HTC cells. This could explain the difference in fructose 2,6- 0 L bisphosphate concentration between the 2 tissues. The time course of the effect of dexamethasone on tyrosine aminotransferase activity, lactate release, fructose 2,6-bisphos Fructose 2,6-bisphosphate ( |iM) phate concentration, and PFK2 activity in HTC cells is illustrated Chart 2. Relative activity of PFK1 partially purified from rat liver (A) or HTC cells (B). Effect of fructose 2,6-bisphosphate at different AMP concentrations: none (O); in Charts 3 and 4. Tyrosine aminotransferase activity increased soMÂ«W;100MM(D). after a 2-h lag period (Chart 4), which is consistent with an

CANCER RESEARCH VOL 45 SEPTEMBER 1985 4265

Table 2 Effect of a 24-h glucocorticoid treatment on glycolysis and enzymeactivity in HTCcells Enzyme activity (milliunits/mg protein) Glycolysis(mmol/24 h/g cells) amino- phospho- Dexamethasone transferase diesterase I con produc Cellline4 (0.5 Ã•/M) activity1.9 activity67.6 pH7.69 sumption2.94 tion5.18

+ 26.5 167 7.54 5.05 5.83 4 2.0 67.2 7.60 6.60 9.91 + 55.0 208 7.51 9.94 13.79 99 ++Tyrosine 5.1 27.2 7.69 3.66 5.43 + 113 96.6 7.65 3.93 5.76 260a 268 ECSerum+ 3.3 7.77 3.70 5.28 302aCulture + 3.3Alkaline 7.68Glucose 5.07Lactate 7.88 Differencenot significant based on other experiments.

Table 3 Glycolytic parameters in HTCcells after exposureto dexamethasone ParameterHexokinase"PFK1"Pyruvate cellsfimol/min/g Â±0.5"(6f5.9 Â±0.7(6)5.3 cells^mol/min/g Â±0.6(6)507 Â±0.3(6)445 kinase"PFK2"Fructose cellsnmol/min/g Â±26(6)0.77 Â±57(6)1.24 cellsnmcrt/g Â±0.09(6)5.2 Â±0.07(6)7.4 2,6-bisphosphatea cells Â±0.5(8) Â±0.7(8)" Acidity of culture medium"1' PHrtiMControl4.2 7.44 Â±0.05(9) 7.22 Â±0.06(9)" Lactate in culture medium6Units(imol/min/g 8.55 Â±0.35(7)Dexamethasone4.411.92 Â±0.26(6)"

6 Mean Â±SE. 0 Numbers in parentheses,number of experiments. d Statistically significant difference (P< 0.05) between dexamethasone-treatedand control cells. e Twenty-four h of dexamethasone(0.5 Â»Â»M)treatment. ' Data from Ref 8

30 -H -, I 5 IO

2 4

20 2 - IO

4-1 -I O 01

3 IO I. -

J 0 012345 0 w _l O IO 20 Time of incubation (hr) Time of incubation(hr) Chart 4. Earlytime course of the effect of 0.5 MMdexamethasone(â€¢,A),added Chart 3. Lactate accumulation (O, â€¢),fructose2,6-bisphosphate concentration at time zero, on the activity of tyrosine aminotransferase(A, A) and PFK2 (O, â€¢) in HTC cells. Control cells (O, A) received the vehicle. One microunit of PFK2 is (D, â€¢),andtyrosine aminotransferase activity (A, A) in HTC cells incubated in the amount of enzyme that catalyses the synthesis of 1 pmol of fructose 2,6- absence (open symbols) or presence (closed symbols) of 0.5 Â»Â¿Mdexamethasone added at time zero. Points, mean for 4 to 8 experiments; bars, SE. *, statistically bisphosphate per min under the assay conditions. significant difference (P < 0.05) between dexamethasone-treated and control cultures. cells, the increase in fructose 2,6-bisphosphate was larger (2.3- fold increase in 5 h) and persisted for a longer period of time. indirect effect of the steroid at the gene level (30). The effect on After 48 h, the concentration of fructose 2,6-bisphosphate in lÃ¡clate release was detectable between 5 and 8 h. The concen dexamethasone-treated cells was 3 times higher than in control tration of fructose 2,6-bisphosphate increased in control cells to cells, whether serum was present or not (results not shown). reach a maximal value (1.8-fold increase) at 5 h and returned to This effect of dexamethasone could be accounted for by the basal values thereafter. In this type of experiment, the only increase in PFK2, the activity of which was stimulated by the manipulation of the cells consisted of their resuspension in steroid after a 2-h lag period (Chart 4). After 5 h of treatment, serum-free medium at zero time. Therefore, the change in fruc PFK2 activity was increased by about 60% when compared to tose 2,6-bisphosphate is due to either the replacement of nu untreated cells (Table 3). The change in PFK2 resulted from an trients or the removal of metabolites. In dexamethasone-treated increase in Vmaxwithout change in the apparent Km for fructose

CANCER RESEARCH VOL. 45 SEPTEMBER 1985 4266

2,6-bisphosphate. The modification was probably not due to the by glucocorticoids of tyrosine aminotransferase and of PFK2. presence of a ligand or a small-molecular-weight stimulator, since Incubation of HTC cells with dexamethasone produced no it did persist after filtration through Sephadex G-25 (Chart 5). change in the Vâ„¢Â»ofhexokinase, PFK1, and pyruvate kinase An increase in PFK2 activity was observed in HTC cells (Table 3). Thus, it is likely that the increased glycolysis seen incubated for 5 h with 0.5 ^M dexamethasone or corticosterone under these conditions results from a stimulation by the steroid but not with the same concentration of aldosterone, progester of the cascade PFK2, fructose 2,6-bisphosphate, PFK1. one, estradiol, or testosterone. Thus, stimulation by steroids was Inhibition of Glycolysis by Various Agents. Adenosine, di- specific for glucocorticoids. The half-maximal increase in fructose butyryl cyclic AMP, and ethanol are known to decrease fructose 2,6-bisphosphate and in tyrosine aminotransferase was obtained 2,6-bisphosphate and the glycolytic flux in hepatocytes (6). To with the same concentration (about 30 nM) of dexamethasone. further assess the role of fructose 2,6-bisphosphate in the control The effect of 0.5 tiM dexamethasone on the activity of PFK2 was of glycolysis in HTC cells, the cells were incubated with these inhibited by 10 V.Mof either of the 2 glucocorticoid antagonists, substances. None of them induced a change in fructose 2,6- dexamethasone mesylate and RU486, by 5 nQ of actinomycin D bisphosphate concentration or in lactate release at concentra per ml, and by 5 mw sodium butyrate. All these agents inhibited tions at which they are active in the liver. DL-Glyceraldehyde is tyrosine aminotransferase induction by dexamethasone in the also known to decrease fructose 2,6-bisphosphate in hepato same experiments, in keeping with earlier work (29). Thus, it cytes (6) and to block glycolysis in cancer cells (31). As shown appears that similar mechanisms are involved in the stimulation in Chart 6, there was a dose-dependent inhibition of lactate release by DL-glyceraldehyde which occurred in parallel with a 6 fall in fructose 2,6-bisphosphate. The concentration of glucose 6-phosphate and fructose 6-phosphate was decreased by 90% in the presence of 10 mw DL-glyceraldehyde, indicating that the 5 fall in fructose 2,6-bisphosphate could result from a decrease in its precursor, fructose 6-phosphate, which is the substrate of 4 PFK2. After 24 h of incubation with 10 rriM DL-glyceraldehyde, 95% of the cells were dead (trypan blue exclusion test).

DISCUSSION

The data presented confirm that the glycolytic rate of HTC 2 cells is greater than that of normal liver cells and provide new cÂ» e clues for the interpretation of this phenomenon. Since HTC cells t do not contain glycogen, they were compared to hepatocytes from fasted rats. In such hepatocytes, lactate release from 5 to 10 mw glucose under aerobic conditions is negligible (22, 26), Â£ O whereas in HTC cells, it is close to maximal capacity. This m 0.5 difference can be accounted for, at least in part, by the qualitative ZG and quantitative changes in enzyme activity (Ref 7; this paper) e 3 and possibly in hexose transport (2). The replacement of glucokinase (Km for glucose > 5 mM, sigmoidal saturation curve for Â« e glucose) by hexokinase (Km < 0.1 mM, hyperbolic saturation curve) in HTC cells can explain the more rapid phosphorylation of glucose at low and physiological glucose concentrations. This may also explain the presence of higher concentrations of hex-

2 E 3 o> i 2

0.5 I 1

Fructose 6-phosphate (rnM) J0 Chart 5. Kineticsof PFK2as a function of substrate concentration. InA, enzyme 10 activity was measured in extracts from cells incubated for 5 h in the absence (O) Glyceraldehyde (mM) or presence(â€¢)of45 nu dexamethasone.Points, meanfor 3 experiments;uars.SE. Activity without added substrates is due to endogenousfructose 2,6-bisphosphate. Chart 6. Effect of exposure of intact HTC cells to DL-glyceraldehydefor30 min InB, enzyme activity was measuredafter Sephadex G-25 filtration of extracts from on fructose 2,6-bisphosphateconcentration (O)and on lactate concentration in the cells incubated for 3 h in the absence(O)or presence(â€¢)of0.5 Â»IMdexamethasone. medium (â€¢).

CANCER RESEARCH VOL. 45 SEPTEMBER 1985 4267

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1985 American Association for Cancer Research. CONTROL OF GLYCOLYSIS IN RAT HEPATOMA CELLS oses 6-phosphate in HTC cells as compared to hepatocytes. As PFK2 (5). Whether PFK2 is also a bifunctional enzyme in tissues to the quantitative changes, we find that fructose-1,6-bisphos- other than liver remains an open question. phatase is absent and that the activities of the key glycolytic Another difference between HTC and liver cells lies in the enzymes PFK1 and pyruvate kinase are, respectively, 4- and 10- insensitivity of HTC cells to several agents known to decrease fold greater than in liver, in agreement with earlier work (7). The fructose 2,6-bisphosphate in liver cells. Adenosine and dibutyryl low concentration of phosphoenolpyruvate and of fructose 1,6- cyclic AMP act by stimulating the phosphorylation, hence the bisphosphate as compared to other cancer cells (3) may indicate inactivation, of PFK2. Thus, resistance of HTC cells to these that pyruvate kinase is more active in HTC cells. As suggested agents might result from a defect at the level of cyclic AMP- earlier (32), the large increase in pyruvate kinase could lead to a dependent protein kinase (36) or beyond (e.g., PFK2). A differ competition for ADP between this enzyme and mitochondria, ence in the kinetic properties of PFK2 as compared to the liver and it could thus be a determining factor for the high aerobic enzyme might also explain the resistance to ethanol. Anoxia was glycolysis of tumors. expected to decrease fructose 2,6-bisphosphate as in liver (22). We have shown that fructose 2,6-bisphosphate is present in This was not the case in HTC cells, again emphasizing the HTC cells and that their PFK1, like that of ascites tumor cells difference in the control of glycolysis between the 2 tissues. (33), is sensitive to this stimulator. The data are consistent with Concerning the inhibition of glycolysis by DL-glyceraldehyde, the idea that fructose 2,6-bisphosphate could play a role in the the data do not allow an interpretation to be made only in terms high glycolytic flux of these cells. Indeed, the concentration of of fructose 2,6-bisphosphate. DL-Glyceraldehyde also caused a this modulator in HTC cells is large enough to account for a decrease in hexoses 6-phosphate, the precursors of fructose stimulation of PFK1, the Vmaxof which is already increased as 2,6-bisphosphate. In turn, the decrease in hexoses 6-phosphate compared to liver. In hepatocytes incubated with 5 to 10 mw concentration could result from an inhibition of hexokinase, as glucose, most of the fructose, 2,6-bisphosphate is probably already postulated (31). bound to proteins, and fructose-1,6-bisphosphatase may ac The selective inhibition of PFK2 would offer a powerful tool to count for about 80% of this binding (26, 34). Since this enzyme inhibit glycolysis and could lead to development of new drugs is not detectable in HTC cells (Table 1), one can expect that, for active against cancer cells. the same total fructose 2,6-bisphosphate concentration, the proportion that is free will be greater in these cells than in liver. ACKNOWLEDGMENTS Therefore, the stimulation of PFK1 by fructose 2,6-bisphosphate should be more efficient in HTC cells. Thus, both fructose 2,6- We thank M. A. Gueuning and L. Maisin for help and P. Lahy and T. Lambert bisphosphate and the changes in key glycolytic enzymes appear for typing the manuscript. to be important regulatory factors of glycolysis in cancer cells. A further difference between HTC and normal liver cells with REFERENCES respect to the control of glycolysis is the way in which they 1. Racker, E. A New Look at Mechanisms in Bioenergetics, pp. 153-175. New respond to glucocorticoids. These hormones, which stimulate York: Academic Press, Inc., 1976. gluconeogenesis in liver (35), stimulated glycolysis in HTC cells. 2. Weinhouse, S. The Warburg hypothesis fifty years later. Z. Krebsforsch., 87: Since dexamethasone did not affect the maximal activity of the 115-126,1976. 3. Eigenbrodt, E., and Glossman, H. Glycolysis, one of the keys to cancer? key glycolytic enzymes, the increase in fructose 2,6-bisphos Trends Pharmacol. Sci., 7: 240-245,1980. phate is the only obvious explanation for the stimulation of 4. Van Schaftingen, E., Hue, L., and Hers, H. G. Fructose 2,6-bisphosphate, the glycolysis by glycocorticoids. The rise in fructose 2,6-bisphos probable structure of the glucose- and glucagon-sensitive stimulator of phosphofructokinase. Biochem. J., 792. 897-901, 1980. phate is likely to result from the 2- to 3-fold increase in N/â„¢Â«of 5. Claus, T. H., EI-Maghrabi, M. Râ€žRegen, D. M., Stewart, B., McGrane, M., PFK2, which is seen in HTC cells but not in hepatocytes incu Kountz, P., Nyfeler, F., Pilkis, J., and Pilkis, S. J. The role of fructose 2,6- bisphosphate in the regulation of carbohydrate metabolism. Curr. Top. Cell. bated with dexamethasone. This change could result from either Regul., 23: 57-88, 1984. a covalent modification or an increased steady-state concentra 6. Hers, H. G ., and Hue, L. Gluconeogenesis and related aspects of glycolysis. tion of the enzyme, or both. Stimulation by a small-molecular- Annu. Rev. Biochem., 52: 617-653,1983. 7. Schamhart, D. H. J., Van de Poll, K. W., and Van Wyck, R. Comparative weight metabolite has been ruled out, since the change in Vmax studies of glucose metabolism in HTC, RLC, MH1C1, and Reuber H35 rat of PFK2 induced by dexamethasone persisted after gel filtration hepatoma cells. Cancer Res., 39:1051-1055,1979. of the enzyme. The fact that this change appeared after a 2-h 8. Verhaegen, M., Granner, D. K., Amar-Costesec, A., Edwards, P. M., and Rousseau, G. G. Effects of glucocorticoids on the plasma membrane of rat lag period and was inhibited by actinomycin D and sodium hepatoma cells. In: S. lacobelli, era/, (eds.), Hormones and Cancer, pp. 113- butyrate suggests that this effect of glucocorticoids involves a 124. New York: Raven Press, 1980. 9. Rousseau, G. G., Amar-Costesec. A., Verhaegen, M., and Granner, D. K. stimulation of gene expression. Schamhart ef al. (7) have shown Glucocorticoid hormones increase the activity of plasma membrane alkaline coordinate changes in the pattern of glycolytic and gluconeogenic phosphodiesterase in rat hepatoma cells. Proc. Nati. Acad. Sci. USA, 77: 1005-1009,1980. enzymes in relation to glucose metabolism in hepatoma versus 10. Krebs, H. A., and Henseleit, K. Untersuchungen Ã¼berdie Harnstoffbildung Â¡m liver cells. Our data show that such differences extend to the TierkÃ¶rper. Hoppe-Seyler's Z. Physiol. Chem., 270: 33-66,1932. control of glycolysis and/or gluconeogenesis by glucocorticoids. 11. Hue, L., Feliu, J. E., and Hers, H. F. Control of gluconeogenesis and of enzymes of glycogen metabolism in isolated rat hepatocytes. Biochem. J., It is noteworthy that the effects of dexamethasone described 776:791-797,1978. here could be observed in a chemically defined medium and are 12. Van Schaftingen, E., Lederer, B., Bartrons, R., and Hers, H. G. A kinetic study of pyrophosphate: fructose 6-phosphate phosphotransferase from potato tub therefore independent of serum factors. Our results do not rule ers. Application to a microassay of fructose 2,6-bisphosphate. Eur. J. out an effect of dexamethasone on fructose-2,6-bisphosphatase. Biochem., 729: 191-195,1982. A decreased activity of this enzyme would indeed result in a rise 13. Bontemps, F., Hue, L., and Hers, H. G. Phosphorylation of glucose in isolated in fructose 2,6-bisphosphate concentration. In the liver, fructose- hepatocytes. Biochem. J., 774: 603-611,1978. 14. Hers, H. G. Glycogen storage diseases. Adv. Metab. Disord., 7: 4-44, 1964. 2,6-bisphosphatase activity is borne by the same protein as is 15. Tejwani, G. A., Pedrosa, F. O., Pontremoli, S., and Horecker, B. L. The

CANCER RESEARCH VOL. 45 SEPTEMBER 1985 4268

purification and properties of rat liver fructose 2,6-bisphosphatase. Arch. and fructose 2,6-bisphosphate concentration. Biochem. J., 224: 779-786, Biochem. Biophys., 777: 253-264,1976. 1984. 16. Feliu, J. E., Hue, L., and Hers, H. G. Hormonal control of pyruvate kinase 27. Van Schaftingen, E., Hue, L., and Hers, H. G. Control of the fructose 6- activity and of gluconeogenesis in isolated hepatocytes. Proc. Nati. Acad. Sci. phosphate/fructose 1,6-bisphosphate cycle in isolated hepatocytes by glucose USA, 73: 2762-2766,1976. and glucagon. Biochem. J., 792: 887-895,1980. 17. Vassault, A. LÃ¡clate dehydrogenase. In: H. U. Bergmeyer (ed.), Methods of 28. Thompson, E. B., Granner, D. K., Gelehrter, T., Erickson, J., and Hager, G. L. Enzymatic Analysis, Vol. 3, pp. 118-126. Weinheim, Verlag Chemie, 1983. Unlinked control of multiple glucocorticoid-induced processes in HTC cells. 18. Van Schartingen, E., Hue, L., and Hers, H. G. Study of the fructose 6- Mol. Cell. Endocrinol., 75:135-150, 1979. phosphate/fructose 1,6-bisphosphate cycle in the liver in vivo. Biochem. J., 29. Rousseau, G. G. Control of gene expression by glucocorticoids. Biochem. J., 792:263-271, 1980. 224:1-12, 1984. 19. Bartrons, R., Hue, L., Van Schaftingen, E., and Hers, H. G. Hormonal control 30. Rousseau, G. G. Interaction of steroids with hepatoma cells: molecular mech of fructose 2,6-bisphosphate concentration in isolated rat hepatocytes. anisms of glucocorticoid hormone action. J. Steroid Biochem., 6:75-89,1975. Biochem. J., 274: 829-837, 1983. 31. Lardy, H. A., Wiebelhaus, V. D., and Mann, K. M. The mechanism by which 20. Van Schaftingen, E., Jett, M. F., Hue, L., and Hers, H. G. Control of liver 6- glyceraldehyde inhibits glycolysis. J. Biol. Chem., 787: 325-337, 1950. phosphofructokinase by fructose 2,6-bisphosphate and other effectors. Proc. 32. Gosalvez, M., Perez-Garcia, J., and Weinhouse, S. Competition for ADP Nati. Acad. Sci. USA, 78: 3483-3486,1981. between pyruvate kinase and mitochondrial oxidative phosphorylation as a 21. Hue, L., Bontemps, F., and Hers, H. G. The effect of glucose and of potassium control mechanism in glycolysis. Eur. J. Biochem., 46: 133-140,1974. ions on the interconversion of the two forms of glycogen phosphorylase and 33. Bosca, L., Aragon, J. J., and Sols, A. Specific activation by fructose 2,6- of glycogen synthase in isolated rat liver preparations. Biochem. J., 752: 105- bisphosphate and inhibition by P-enolpyruvate of ascites tumor phosphofruc- 114, 1975. tokinase. Biochem. Biophys. Res. Commun., ÃŽ06:486-491,1982. 22. Hue, L. Role of fructose 2,6-bisphosphate in the stimulation of glycolysis by 34. McGrane, M. M., EI-Maghrabi, M. R., and Pilkis, S. J. The interaction of anoxia in isolated hepatocytes. Biochem. J., 206: 359-365,1982. fructose 2,6-bisphosphate and AMP with rat hepatic fructose 1,6-bisphospha- 23. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. Protein tase. J. Biol. Chem., 258: 10445-10454, 1983. measurement with the Folin phenol reagent. J. Biol. Chem., 793: 265-275, 35. Exton, J. H. Regulation of gluconeogenesis by glucocorticoids. In: J. D. Baxter 1951. and G. G. Rousseau (eds.), Glucocorticoid Hormone Action, pp. 535-546. 24. Weber, G. Enzymology of cancer cells. N. Engl. J. Med., 296: 486-493,1977. Berlin: Springer Verlag, 1979. 25. Weber, G. Enzymology of cancer cells. N. Engl. J. Med., 296: 541-551,1977. 36. Granner, D. K. Absence of high-affinity adenosine 3',5'-monophosphate bind 26. Hue, L., Sobrino, F., and Bosca, L. Difference in glucose sensitivity of liver ing sites from the cytosol of three hepatic-derived cell lines. Arch. Biochem. glycolysis and glycogen synthesis. Relationship between lactate production Biophys., 765: 359-368, 1974.

CANCER RESEARCH VOL. 45 SEPTEMBER 1985 4269

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1985 American Association for Cancer Research. Fructose 2,6-Bisphosphate and the Control of Glycolysis by Glucocorticoids and by Other Agents in Rat Hepatoma Cells

Anne M. Loiseau, Guy G. Rousseau and Louis Hue

Cancer Res 1985;45:4263-4269.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/45/9/4263

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/45/9/4263. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.