Eur. J. Biochem. 154, 141 - 145 (1986) 0 FEBS 1986

Effect of benzoate on the metabolism of fructose 2,6-bisphosphate in yeast Jean FRANCOIS, Emile VAN SCHAFTINGEN and Henri-Gkry HERS

Laboratoire de Chimie Physiologique, Universitk Catholique de Louvain; and International Institute of Cellular and Molecular Pathology, Brussels

(Received September 4/0ctober 15, 1985) - EJB 85 1000

When benzoate (2 mM, pH 3.5) was added together with glucose (0.1 M) to a suspension of Saccharomyces cerevisiae in the stationary phase, it caused a relative increase in the concentration of glucose 6-phosphate and fructose 6-phosphate and a decrease in the concentration of fructose 1,6-bisphosphate. These effects are in confirmation of similar observations made by Krebs et al. [Biochem. J. 214, 657-663 (1983)l and are indicative of an inhibition of 6-phosphofructo-I-kinase. Benzoate also caused an about fourfold relative decrease in the concentration of fructose 2,6-bisphosphate, an increase in that of cyclic AMP with no change in that of ATP. It also greatly decreased the activation of 6-phosphofructo-2-kinase, but not that of trehalase, both of which normally occur upon addition of glucose to a yeast suspension. When added 10 min after glucose, benzoate caused a rapid (within 2 - 3 min) decrease in fructose 2,6-bisphosphate concentration and in 6-phosphofructo-2-kinase activity. In the presence of benzoate, there was also a parallel decrease in the concentration of fructose 2,6- bisphosphate and in the rate of ethanol production when the external pH was dropped from 5.0 to 2.5, with minimal change in the concentration of ATP. Purified 6-phosphofructo-2-kinase was inhibited by benzoate and also by an acid pH. Experiments with cell-free extracts did not provide an explanation for the rapid disappearance of fructose-2,6-bisphosphate or the inactivation of 6-phosphofructo-2-kinase in yeast upon addition of benzoate.

Benzoate has been extensively used as a food preservative because it inhibits the growth of fungi and bacteria. The mechanism of its inhibitory action remains badly understood. However, it has been clearly established that both its penetra- tion into cells and its inhibitory action increase with acidifica- tion of the medium, and are essentially proportional to the concentration of the undissociated acid. Nonetheless, it re- mains unclear whether the acid must enter the cell to exert its activity or if the inhibition of growth is due to the interaction of the acid form of the drug with some reaction connected with the cell membrane (reviewed in [l , 21). Recently, Krebs et al. [3] measured the distribution of benzoate between extra- cellular and intracellular water in suspensions of Saccharo- myces cerevisiae and calculated that at concentrations of 2 - 10 mM, benzoate causes a shift of the intracellular pH by more than 1. They also observed that benzoate causes an inhibition of glucose fermentation, accompanied by an accumulation of hexose monophosphates and a decrease in intermediates beyond PFK 1, suggesting that this enzyme was inhibited. This inhibition was proposed to be a direct effect of pH since PFK 1 was shown to be more sensitive to a lowering of pH than hexokinase. Subsequent inhibition of

Correspondence to H.-G. Hers, Laboratoire de Chimie Physio- logique de 1’Universitt Catholique de Louvain, UCL 75.39, Avenue Hippocrate 75, B-1200 Bruxelles, Belgium

Abbreviations. Fru(2,6)P2, fructose 2,6-bisphosphate; Fru6P, fructose 6-phosphate; Glc6P, glucose 6-phosphate; Fru(1,6)P2, fructose 1,6-bisphosphate; PFK 1, 6-phosphofructo-1-kinase; PFK 2,6-phosphofructo-2-kinase.

Enzymes. (IUB Recommendations 1984). 6-Phosphofructo-l- kinase (EC 2.7.1.11); 6-phosphofructo-2-kinase (EC 2.7.1.105); trehalase (EC 3.2.1.28); fructose 1,6-bisphosphatase (EC 3.1.3.11); fructose 2,6-bisphosphatase (EC 3.1.3.46).

glycolysis was assumed to cause a fall in ATP concentration and thus to restrict growth.

Another explanation for the cross-over in the concentra- tion of glycolytic metabolites observed by Krebs et al. [3] would be a decreased concentration of Fru(2,6)P2 caused either by benzoate itself or by the decrease in pH. In yeast, PFK 1 is indeed greatly stimulated [4] and fructose 1,6-bis- phosphatase is inhibited [5], by Fru(2,6)P2. This phosphoric ester is formed from Fru6P and ATP by PFK 2 and the latter enzyme is activated by phosphorylation catalyzed by cyclic- AMP-dependent protein kinase [6]. The stimulation of gly- colysis by glucose in yeast is therefore currently explained by the sequential formation of cyclic AMP by adenylate cyclase, phosphorylation of PFK 2 by cyclic-AMP-dependent protein kinase, formation of Fru(2,6)P2 and stimulation of PFK 1 [6].

The purpose of the present work was to investigate the possibility that the presence of benzoate in the external medi- um would affect the concentration of Fru(2,6)P2 in yeast and consequently, inhibit glycolysis.

MATERIALS AND METHODS

Materials

Cyclic [3H]AMP assay kit was purchased from Amersham (UK). Fru(2,6)P2 [7] and [2-32P]Fru(2,6)P, [S] were prepared as described. Yeast extract and bactopeptone were from Difco (Detroit, MI, USA), glucose and potassium sorbate from Fluka (Buchs, Switzerland). Sodium benzoate, sodium sali- cylate, and all other chemicals were from Merck (Darmstadt, FRG). Auxiliary enzymes and biochemicals were from Boehringer (Mannheim, FRG). PFK 1, purified to homogeneity from baker’s yeast [9, lo], was kindly provided

142

3.0 Hexose-6-P [Benzoate! (mM1 Fru(2,6)P2 by Dr E. Hofmann (Leipzig, GDR). PFK 2 was purified and activated by phosphorylation as described previously [6] except that bovine heart type I1 cyclic-AMP-dependent pro- tein kinase (Sigma) and 2 pM cyclic AMP were used instead -- of the catalytic subunit. m

- 2.0. Fru(l,6)PP - 1.0 Yeast strain and growth conditions - L

W

' - 0 -

CAMP Saccharomyces cerevisiae X2180 was grown with aeration 1.0 - at 30°C in a medium containing 1% yeast extract, 2% bactopeptone and 2% glucose. Cells were harvested at the 5 o

-

- e - 1.5

stationary phase by centrifugation, washed once with water 2.0 ~ ATP - 1.0 and once with a 50 mM tartaric acid/NaH2P04 buffer (pH

05

2

0

1.0 - 2 3.5) and resuspended in this buffer at a concentration- of approximatively 20 mg wet weight/ml. In the experiment re- ported in Fig. 3, the cells were resuspended in the same buffer previously adjusted to the indicated pH values. Incubations were performed at 25 "C.

Preparation of extracts

Cells were collected and extracts were prepared as de- scribed in [6]. However, 6% trichloroacetic acid was used instead of 15% perchloric acid for the preparation of the extracts used for the determination of acid-stable metabolites. This solution was evaporated to dryness by incubation at 70°C under a stream of nitrogen. The residue was dissolved in 1 ml of a solution containing 100 mM Tris/HCl, pH 7.4 and 1 mM EDTA.

Analytical procedure

PFK 1 was measured as in [3] but in the absence of Glc6P, AMP, ADP and citrate. Unless otherwise indicated, PFK 2 and trehalase were measured as in [6]. The hydrolysis of Fru(2,6)P2 in a crude extract of yeast was measured by the production of [32P]Pi from [2-32P]Fru(2,6)P2 in a mixture containing 50 mM Mes at pH 5 -6.5 or Hepes at pH 7 - 7.5, 10 mM MgC12, 1 mM dithiothreitol and 10 pM [2-32P]- Fru(2,6)P2. All enzymic assays were performed at 30°C. One unit is the amount of enzyme that catalyzes the conversion of 1 pmol substrate in 1 min under the conditions of the assay. The concentration of protein was determined by the method of Bradford [ l l ] using bovine immunoglobulin G as a stan- dard.

Fru(2,6)Pz was measured in alkaline extracts as described in [6]. Glc6P and Fm6P [12], Fru(l,6)Pz [13], ATP [I41 and ethanol [15] were assayed by published procedures. Cyclic AMP was measured with the Amersham assay kit as described in [16]. Concentrations are expressed as mol/g wet weight.

RESULTS

Effect of benzoate on the concentration of hexose phosphates, ATP and cyclic AMP and on the rate of glycolysis in yeast incubated in the presence of glucose

We report in Figs 1 and 2 the results of experiments in which the effects of benzoate added to yeast simultaneously with (Fig. l), or 10 min after (Fig. 2), glucose were investi- gated. These experiments confirm that glucose alone causes a large increase in the concentrations of hexosed-P, Fru(1 ,6)P2, Fru(2,6)P2 and cyclic AMP [6, 171 and that benzoate in the presence of glucose causes a further rise in hexose-6-P and a relative decrease in Fru(1,6)P2 [3], as would be expected from

I . I t o 30 0 3 10 30

Time after addition of glucose (mini

Fig. 1 . ESfect of glucose added with or without benzoate on the concentration of hexose-6-P, Fru(l ,6)P2, ATP, Fru(2,6)P2 and cyclic A M P in S . cerevisiae. The cell suspension was incubated at 25°C for 30 min. Glucose (0.1 M) was then added with or without sodium benzoate (2 mM) at the indicated final concentration

Fru(2,6)P2 I

" " 0 3 10 20 LO 0 3 10 20 LO

Time after addition of glucose lminl

Fig. 2. Effect of benzoate added after glucose on the concentration of hexose-6-P, Fru(1,6)Pz, Fru(2.6)Pz and cyclic AMP in S . cerevisiae. Same procedure as in Fig. 1 except that sodium benzoate was added 10 min after glucose

a decreased activity of PFK 1. Under the conditions used (2 mM benzoate, pH 3 . 9 , ATP was little affected. The new facts revealed by these experiments are that the rise in Fru(2,6)Pz caused by glucose was greatly reduced in the pres- ence of benzoate and that the addition of benzoate 10 min after that of glucose, when the concentration of Fru(2,6)P2 was elevated, caused a remarkable disappearance of this bis- phosphoric ester. Like glucose and independently of it, benzoate caused a rise in the concentration of cyclic AMP (see also Fig. 6) .

When higher concentrations of benzoate were used (10 mM at pH 3.5, 40 mM at pH 5), the metabolic situation was dominated by a rapid decrease in ATP concentration (experiments not shown), an effect already reported by Krebs et al. [3] and which we considered too complex to be analyzed in detail.

143

0 , lclosed symbols1

' 0 ATP

-

L E I 0 .c

10

2.5 30 3.5 1.0 1.5 5.0 PH

Fig. 3. Effect of external p H on the rate of ethanol production and on the concentration of Fru(2,6)Pz, hexose-6-P, Fru(l ,6)P2 and ATP in S. cerevisiae incubated in the presence of glucose with or without benzoate. The cell suspension was incubated at 25°C for 30 min at the pH indicated. Glucose was then added at a final concentration of 0.1 M without (open symbols) or with (closed symbols) 2 mM sodium benzoate. The cells were collected 3 min later for the measurement of phosphate esters. Ethanol was measured in the medium after 60 min

PFK 2

100 - [Benzoate1 ImM)

25 -

v ( Trehalase

l o o t

6

0 ' ; ; Ib 20 3c Time after addition of glucose ( m i d

Fig. 4. Effect of glucose added with or without benzoate on the activity of PFK 2 and trehatase in S. cerevisiae. Details as in Fig. 1

The effect of external pH on some of the changes described above and on the rate of glycolysis was also investigated (Fig. 3). In the absence of benzoate, acidification of the medi- um down to pH 3 had little effect on any of these parameters. In the presence of 2 m M benzoate, there was a parallel decrease in the concentration of Fru(2,6)P2 and in the rate of glycolysis when the pH was shifted from 5 to 3.5; there was also a rise in hexose-6-P and a decrease in Fru(1,6)P2. These

f 100-

h 80- F

C (Y

0

.- e

5 2 60- r > - ._ .- I :: L O -

:: 2 0 -

N

Y

Benzoate I

' ' 0 10

Time after addition of Glucose lmin)

Fig. 5. Effect of benzoate added after glucose on the activity of PFK 2 and trehalase in S. cerevisiae. Details as in Fig. 2

Fru(2,6) P, ?

. Trehalase - '$ L O l b PFK 2

N 20 Y L L P

O 0 3 10 20 LO Time after addition of benzoate (minl

Fig. 6 . Effect of benzoute and of a subsequent addition of glucose on the concentration of Fru(2,6)P2 and cyclic A M P and the activities oj PFK 2 and trehalase in S . cerevisiae. Same procedure as in Fig. 1 , except that glucose was added 10 min after benzoate

changes were not or little increased when the external pH was further dropped to 2.5. Remarkably there were minimal changes in ATP concentration, despite the reduced rate of ethanol formation.

Effects similar to those described in Fig. 1 were also obtained when 2mM benzoate was replaced by 1 mM salicylate or 2 mM sorbate, two other antifungal agents, acting similarly to benzoate [I, 21.

Effect of benzoate on the activation of PFK 2 by glucose

In the experiments described in the preceding section, the activities of PFK 2 and of trehalase were also measured. These two enzymes have indeed in common the property of being activated through phosphorylation by cyclic-AMP-dependent protein kinase [6] (and references therein). The data shown in Figs 4 - 6 confirm the parallel activation o f the two enzymes upon addition of glucose to a yeast suspension. Furthermore,

1 44

they reveal an inhibitory effect of benzoate on the activation of PFK 2 and this in deep contrast with the lack of effect on the activation of trehalase. The inhbitory effect of benzoate on PFK 2 activation was more pronounced when benzoate was added simultaneously with glucose than when it was added 10 min before glucose (compare Fig. 4 with Fig. 6). When benzoate was added 10min after glucose, at a time when PFK 2 had been activated, it caused a nearly complete inactivation of the enzyme within about 3 min (Fig. 5).

When benzoate was added alone (Fig. 6), i.e. in the ab- sence of glucose, it caused a rise in cyclic AMP concentration and a large activation of trehalase (analogous to the activation caused by glucose). However, there was only a small and transient activation of PFK 2. The subsequent addition of glucose caused an additional increase of cyclic AMP concentration and trehalase activity, and a smaller rise in PFK 2 activity.

Experiments with a cell-free extract and purified enzymes

Krebs et al. [3] have estimated that at 2 mM benzoate and pH 3.5 (the experimental conditions that we use) the intracellular concentration of benzoate should be close to 21 mM and the pH 5.27. These are therefore the conditions to be taken into consideration when trying to explain at the molecular level the main effects of benzoate reported in pre- ceding sections, i. e. to decrease the concentration of Fru- (2,6)P2 as well as the activity of PFK 2 when the cells are incubated in the presence of glucose.

We have made two types of observations which could explain a decreased formation of Fru(2,6)P2. First, and as illustrated in Fig. 7, the activity of PFK 2 both in its native and activated form was markedly reduced by an acidification of the medium, reaching at pH 5.5 only 10% of the value at pH 7.0. The same figure also confirms a similar effect of pH on the activity of PFK 1, measured in the presence and the absence of Fru(2,6)P2, as previously reported by Krebs et al. [3]. Second, we observed a direct inhibitory effect of benzoate, sorbate and salicylate on PFK 2, an effect that reached 70% at 20 mM benzoate. This effect was observed both with the non-activated (Fig. 8) and activated (not shown) forms of the enzyme.

With the hope of finding an explanation for the rapid disappearance of Fru(2,6)P2 from cells to which benzoate had been added after glucose (see Fig. 2), we incubated a cell-free extract in the presence of 10 pM Fru(2,6)P2 at pH values varying between 6.0 and 7.8 but we could not detect any appreciable decrease in the concentration of the phosphoric ester. We also made similar experiments in the presence of 10 pM [2-32P]Fru(2,6)Pz at pH values ranging between 5.0 and 7.5, with and without 20 mM benzoate and under these conditions, we could maximally detect the formation of 0.2 nmol [32P]Pi min-' (g wet cells)-' at pH 6, a value which is fivefold lower than the rate of Fru(2,6)Pz destruction in intact cells as shown in Fig. 3.

With regard to the effect of benzoate in preventing the activation of PFK 2 by glucose in yeast, we investigated the effect of benzoate and of an acid pH on the activation of both PFK 2 and trehalase in a cell-free extract incubated in the presence of cyclic AMP and ATP-Mg as previously reported [6]. No conclusive information could be drawn from these experiments. Furthermore, no appreciable effect of an acid pH on a potential protein phosphatase, as estimated by the rate of inactivation of PFK 2, could be shown.

r I

" 5.0 5.5 6.0 6.5 7.0 7.5 " PH

Fig. I . Effect of pH on the activities of PFK 1, measured in the presence and the absence of 5 pM Fru(2,6)Pz, and of PFK 2 , before and after activation by cyclic-AMP-dependent protein kinase. A 50 mM Hepes/ 50 mM Mes buffer was used

I 1

!i I \\\-Sorbate I \tf Salicylate I

" 0 5 10 20 50 llnhibitorl h M 1

Fig. 8. Effect of benzoate, sorbate and salicylate on the activity of PFK 2. The activity of purified non-activated yeast PFK 2 was measured as in [6] except that a 50 mM Hepes/SO mM Mes buffer pH 7.5. was used and that the concentrations of Glc6P and Fru6P were 2 mM and 0.5 mM respectively

DISCUSSION

The role of Fru(2,6)P2 in the inactivation of PFK I by benzoate

The benzoate effects of increasing the concentration of hexose-6-P and of decreasing that of Fru(1,6)P2 in yeast incubated in the presence of glucose (Figs 1 and 2) are in full agreement with the previous report by Krebs et al. [3]. In our experiments, these effects were observed under relatively mild conditions (2 mM benzoate, extracellular pH 3.5) which did not affect the intracellular concentration of ATP. They were also observed with salicylate and sorbate, two other antifungal agents. These effects are indicative of an inhibition of PFK 1, which, according to Krebs et al. [3], would not result from a direct interaction of benzoate with the enzyme but from a modification of the intracellular environment. In the light of our results, this modification would not only include a drop in pH, as proposed by Krebs et al. [3], but also a decrease in the concentration of Fru(2,6)P2.

145

Indeed, under all conditions under which benzoate, salicylate or sorbate affected PFK 1 activity (as testified by a change in the ratio of its substrate and products), the concentration of Fru(2,6)P2 was reduced by a factor of about 4, compared to the control. Since Fru(2,6)P2 is known potently to stimulate PFK 1 [4], the activity of which is other- wise too low to account for the glycolytic rates observed in vivo [18, 191, a decrease in its concentration would decrease the activity of PFK 1. This mechanism is further supported by the parallel decrease in Fru(2,6)P2 concentration and in glycolytic rate observed when the medium was acidified in the presence of benzoate (Fig. 3). It is also remarkable that under these conditions, the concentration of ATP was not markedly affected whereas growth is known to be arrested [I, 31. This indicates that the arrest of growth cannot simply be explained by the deficit in energy supply that could result from the inhibition of glycolysis. The proposed explanation does not apply to the effect of benzoate to prevent the growth of bac- teria in which, up to now, Fru(2,6)P2 could not be detected.

The mechanism by which benzoate decreases Fru(2,6)P2 Concentration in yeast

As a rule, the cellular concentration of Fru(2,6)P2 depends on the relative activities of PFK 2 and fructose 2,6-bis- phosphatase; the first of these enzymes forms Fru(2,6)P2 from Fru6P and ATP, the second converts it back to Fru6P [20]. In yeasts, however, only the first of these two enzymes has been clearly identified since we have been unable up to now to detect an enzymatic system which, in a cell-free extract, would catalyze the destruction of Fru(2,6)P2.

The activity of yeast PFK 2 can be affected by several factors such as pH, the concentration of substrates and of various effectors including benzoate, and also phosphoryla- tion by cyclic-AMP-dependent protein kinase. Since the decrease in Fru(2,6)P2 concentration due to the presence of benzoate in the external medium was concomitant with an increase in the concentration of Fru6P with no change in ATP, the lack of activity of PFK 2 cannot be explained by a lack of substrate. One can explain it by the acidification of the intracellular medium, by the inhibitory effect of benzoate itself, as well as by a covalent modification of the enzyme (see next paragraph). These several effects provide together a satisfactory explanation for the reduced rise of Fru(2,6)P2 concentration but do not explain the rapid decrease (1 nmol . min-' . g-') of Fru(2,6)P2 occurring when benzo- ate was added 10 min after glucose. One is therefore forced to assume that yeast contains an as yet unidentified enzyme able to destroy Fru(2,6)P2. This enzyme either is activated by benzoate or is continuously active, compensating then the activity of PFK 2. According to this second hypothesis, the inactivation of PFK 2 would suffice to explain the dis- appearance of Fru(2,6)P2 in the presence of benzoate.

Thr mechanism of PFK 2 inactivation

It has been previously established that the activation of both PFK 2 and trehalase upon addition of glucose to yeast involves successively the formation of cyclic AMP, the

activation of cyclic-AMP-dependent protein kinase and the phosphorylation of the two enzymes by the activated kinase [6] (and references therein). Furthermore, this sequence of events is counteracted by the action of a protein phosphatase, which causes a rapid reversible inactivation of PFK 2 as soon as cyclic-AMP-dependent protein kinase ceases to be active. Since the activation of PFK 2 by glucose in vivo, but not that of trehalase, is greatly diminished by benzoate and since cyclic AMP formation is not decreased but is increased, one is forced to conclude that benzoate interferes specifically with the phosphorylation or the dephosphorylation of PFK 2. This conclusion is reinforced by the fact that, as shown in Fig. 5, benzoate also caused the rapid inactivation of PFK 2, in- dicating that dephosphorylation of the enzyme prevailed over its phosphorylation. The mechanism of this de-equilibrium between the kinase and phosphatase reactions remains to be elucidated.

REFERENCES 1 . Bosund, I. (1962) Adv. Food Res. 11, 331 -353. 2. Sinskey, A. (1980) in Developments in,foodpreservatives (Tilburg,

R. H., ed.) pp. 1 1 1 - 136, Applied Science Publications, Lon- don.

3. Krebs, H. A,, Wiggins, D. , Stubbs, M., Sols, A. & Bedoya, F. (1983) Biochem. J . 214,657-663.

4. Bartrons, R., Van Schaftingen, E., Vissers, S. & Hers, H. G .

5. Van Schaftingen, E. & Hers, H. G. (1982) Proc. Natl Acud. Sci.

6. FranCois, J., Van Schaftingen, E. & Hers, H. G. (1984) Eur. J . Biochem. 145,187-193.

7. Van Schaftingen, E. & Hers, H. G. (1981) Eur. J . Eiochem. 117, 319-323,

8. Van Schaftingen, E., Davies, D. R. & Hers, H. G. (1982) Eur. J . Biochem. 124, 143 - 149.

9. Diezel, W., Bohme, H. J . , Nissler, K., Freyer, R., Heilmann, W., Kopperschlager, G. & Hofmann, E. (1973) Eur. J . Biochem. 38,

10. Kopperschlager, G., Bar, J., Nissler, K. & Hofmann, E. (1977)

11. Bradford, M. (1976) Anal. Eiochem. 72, 248-254. 12. Hohorst, H. J. (1963) in Methods of enzymatic analysis

(Bergmeyer, H. U. ed.) pp. 134-139, Academic Press, New York.

13. Michal, G. & Beutler, H. 0. (1974) in Methods of enzymatic analysis (Bergmeyer, H. U., ed.) 2nd edn, vol. 3, pp. 1314- 1319, Academic Press, New York.

14. Lamprecht, W. & Trautschold, I. (1963) in Methods ofenzymatic analysis (Bergmeyer, H. U., ed.) pp. 543 - 551, Academic Press, New York.

15. Cornell, N. W. & Veech, R. L. (1983) Anal. Biochem. 132,438- 423.

16. Tovey, K. C., Oldham, H. G . & Whelan, J . A. M. (1974) Clin. Chern. Acta 56,221 - 234.

17. Lederer, B., Vissers, S., Van Schaftingen, E. & Hers, H. G. (1981) Biochem. Biuphys. Res. Commun. 103.1281 - 1287.

18. Banuelos, M., Gancedo, C. & Gancedo, J. M. (1977) J. BioI. Chem. 252,6394-6398.

19. Clifton, D. & Fraenkel, D. G. (1983) J . Biol. Chem. 158, 9245 - 9249.

20. Hers, H. G. & Van Schaftingen, E. (1982) Biochem. J . 206, 1 - 12.

(1982) F E B S k t t . 143,137-140.

USA 78,2861 -2863.

479-488.

Eur. J . Biochem. 81, 317-325.