OUP user menu

Critical findings on the activation cascade of yeast plasma membrane H+-ATPase

Arnošt Kotyk, Georgios Lapathitis, Jaroslav Horák
DOI: http://dx.doi.org/10.1016/S0378-1097(03)00591-3 175-180 First published online: 1 September 2003


Strains of the yeast Saccharomyces cerevisiae, deficient in either of its two G-proteins, in the Snf3 and Rgt2 sensors, in the Gpr1 receptor and in various hexokinases were tested for their ability to start the activation cascade with a metabolizable monosaccharide that leads eventually to activation of plasma membrane H+-ATPase. The acidification rate after addition of glucose to glucose-grown cells and of galactose to galactose-grown ones, and the rate of ATP hydrolysis by purified plasma membranes in both types of cells were studied. It appears unequivocally that phosphorylation of the monosaccharide is essential for the activation; the role of the Gpa2 protein (possibly in combination with the Gpr1 receptor) is very probable while the two sensors appear to play somewhat ambiguous roles — in the absence of both the activation was actually higher than in the parent strain. The Gpa1 G-protein is not involved in acidification but may function in ATPase activity where, in addition to the phosphorylation step, other factors can play a role. There appear to be alternative pathways leading to the ultimate activation of the H+-ATPase, not necessarily involving G-proteins.

  • Yeast H+-ATPase
  • G-protein
  • Activation cascade
  • Role of sensors and receptors

1 Introduction

The most abundant protein of baker's yeast plasma membrane and one that is indispensable for survival, the H+-adenosinetriphosphatase (EC; TC 3.A.3.3.1) is characterized by its being activated in the presence of a metabolizable monosaccharide, such as d-glucose, d-mannose and d-fructose [1,2], as well as of d-galactose after cultivation on it as the source of carbon [3]. It has been clear for several years that it is protein kinase C that, by phosphorylating some amino acid residues near the C-terminus of the enzyme, releases the ATP-binding site and thus enables it to function properly [47]. The protein kinase itself must be activated — this is achieved by phospholipase C (PLC) or rather one of its reaction products, diacylglycerol [810].

At the beginning of this activation cascade glucose sensors have been postulated (for a fine review see [11]) and it was suggested that phosphorylation of the ‘inducing’ sugar is essential [8] and that fructose-6-phosphate might play a role here [3]. However, unless it was accepted that, like in mammalian tissues, PLC can be activated directly by a signal from, say, receptor tyrosine kinase, there remained a gap between the sensor and PLC. The obvious suggestion was that a G-protein was involved. However, the first attempts to assign a role to one of the G-proteins known to occur in Saccharomyces cerevisiae (Gpa1p and Gpa2p) failed because of lack of sufficiently specific antibodies [12]. Now a serious attempt to identify the role of a G-protein in the cascade was published by Souza et al. [13]. However, several experimental details of their approach, particularly in measuring the acidification, are open to discussion.

Hence, in the present work, the use of a broader range of mutants and measurement of H+-pumping activity as well as of ATP hydrolysis under improved conditions may contribute to the final elucidation of the entire activation cascade, starting with glucose and ending with transport of H+ by the plasma membrane H+-ATPase.

2 Materials and methods

2.1 Yeast strains and their cultivation

All the strains were S. cerevisiae — they were grown at 30°C in 0.67% Yeast-Nitrogen base medium and 0.3% Yeast Extract containing 50 mM d-glucose or 50 mM d-galactose up to the beginning of the stationary phase, i.e. when the A600 value of the suspension did not increase by more than 0.02 during the last hour (the cultivation period on glucose/galactose is shown in parentheses). The characteristics of the strains are shown below.

Strains 1–4 were from Prof. J.M. Thevelein of Leuven (Belgium), strain 5 was from Dr. S.J. Dowell of Stevenage (UK), strains 6 and 7 were from the Euroscarf Collection, strains 8–10 were from Dr. E. Boles of Düsseldorf (Germany) and strains 11–14 from Dr. S. Özcan of St. Louis, MO (USA) — via J.M. Thevelein.

After growth the cells were centrifuged, washed, resuspended in 10 ml distilled water and incubated for 60 min at 30°C to deplete their endogenous reserves.

2.2 Measurement of acidification

The suspension after the 60-min incubation was centrifuged and the pellet was transferred to 9 ml of 0.1 mM triethanolamine-phthalic acid buffer of pH 6.0 to ensure a fairly stable initial value for measurement. Incubation was done at 30°C, using a Cole-Parmer pH/mV/temp meter, the suspension density being between 2.5 and 6 mg dry mass per ml, depending on the strain used.

After suspending the pellet the pH generally dropped to 5.0–5.5; after 5 min either d-glucose or d-galactose was added to a 50 mM concentration and pH recording continued for a further 15 min. The values used for rate computation were those between 1 and 3 min after sugar addition (A) and, to check their reliability, between 0 and 15 min (B). The ratio between A and B was regularly at 1.04 for the glucose variant and at 1.17 for the galactose one. This procedure was adopted in contrast to [13] where in fact three different processes were measured together [14,15] (Fig. 1).

Figure 1

Acidification traces (shown as external pH) of a wild strain of S. cerevisiae following different protocols of adding cells, 50 mM glucose and 20 mM KCl to 0.1 mM triethanolamine-phthalic acid buffer of pH 6.0. It is clear that the glucose-initiated ATPase activation is best estimated in the first setup (shown by a heavy line on the acidification trace). The inset shows the beginning of acidification after adding three different sugars — again measured as in the first setup.

2.3 H+-ATPase activity estimation

Yeast cells were grown as above and, after 1 h incubation in water, were incubated for 10 min either with 50 mM glucose or with 50 mM galactose or, for the control, simply in distilled water. Plasma membranes were prepared basically according to Serrano [16] in a slight modification [12] and the ATP-hydrolyzing activity was determined according to Goffeau and Dufour [17] but 10 mM KNO3 was added to the reaction mixture to block any trace of vacuolar H+-ATPase that might be present in the preparation.

2.4 Protein content

Proteins were determined according to Lowry.

2.5 Reagents

All reagents were obtained from Sigma-Aldrich (Prague, Czech Republic), except Yeast Nitrogen Base and Yeast Extract which were from Difco (USA).

3 Results and discussion

3.1 Acidification

In the first set of experiments glucose or galactose was added to the yeast suspension and acidification was followed. The rate between minute 1 and 3 is shown in Table 1.

View this table:
Table 1

Rates of acidification by various yeast strains (see Section 2.1) after addition of a metabolized sugar, expressed in nmol H+ per min per mg dry mass; all values are means of triplicate measurements, the range of results never exceeding 13%

StrainAfter glucose%After galactose%
1 wild-type2.961001.34100
2 deficient in Gpa2p1.78600.8664
3 deficient in Hxk1p and Glk1p1.98672.18163
4 deficient in Hxk1p, Hxk2p, Glk1pno acidification0.9470
5 deficient in Gpa1p3.611222.18163
6 wild-type4.901005.13100
7 deficient in Gpr1p3.04621.3927
8 wild-type5.931005.25100
9 deficient in Snf3p2.19374.2581
10 deficient in Rgt2p4.15704.1579
11 wild-type8.421003.80100
12 deficient in Snf3p7.50892.8575
13 deficient in Rgt2p6.37761.9652
14 deficient in Snf3, Rgt2p18.102153.0881

It was shown before (e.g. [3,18]) that the acidification following addition of glucose has several components, that caused by H+-ATPase often playing a minor role. Moreover, back-titration experiments (not presented here) showed that after galactose addition the consumption of alkaline titrant is about one-third of that after glucose addition. This tallies with the respiratory quotients of a wild-type strain of yeast where the ratio of CO2 produced to O2 consumed was between 1.8 and 2.0 with glucose but only 1.2 with galactose (unpublished results from this laboratory). It resembles the situation with glucose–maltose–maltotriose where the respiratory quotients were 2.1–1.5–1.0 [19]. It may be seen that the acidification ratios were substantially lower in yeast grown on galactose and supplied with galactose before pH measurement (Table 1), the grand average being 5.55 nmol H+ per min per mg dry mass after glucose and only 3.88 after galactose; the ratio is thus 1.43.

The following conclusions can be drawn from the results:

  1. Deletion of the G-protein Gpa2 causes a marked reduction in the acidification rate (like in [13]).

  2. Impossibility to phosphorylate the added sugar leads to a complete loss of acidification capacity. Since galactose can be phosphorylated in the Hxk1,Glk1-deficient mutant there is no decrease in galactose-grown cells; in fact, for unknown reasons, a large increase is observed.

  3. Deficiency in the Gpa1p had a rather positive effect both on glucose-grown and galactose-grown cells.

  4. Deficiency in the Gpr1 receptor causes a marked decrease of acidification rate after glucose (somewhat greater than in [13]– 38 vs. 20%) but a very pronounced one after galactose (73%).

  5. Deficiency in the Snf3 sensor causes little change in acidification in one of the mutants (11%) but 63% in the other mutant. After galactose, the effect is small with both mutants.

  6. Deficiency in the Rgt2 sensor inhibits the rate after glucose by 27% on average of the two mutants (in sharp contrast with [13]) and decreases it substantially after galactose.

  7. Surprisingly, the double mutant with deficiency in both sensor proteins shows a much higher acidification rate after glucose than any of the single mutants. After galactose the effect is not so pronounced although, even here, the double mutant shows a higher activity than any of the single ones.

  8. Deficiency in the Gpa1 Gα protein causes an increase in the rate of acidification.

3.2 ATP hydrolysis

Due to the possible multiplicity of acidification sources, particularly after addition of glucose, it appeared more promising to determine the ATP-hydrolyzing activity of the plasma membrane H+-ATPase. (To check for the possible interference on the part of Na+-ATPase also known to occur in yeast, a mutant deficient in this enzyme (ena) designated G19 from a parent W330 strain, obtained from Dr. A. R. Navarro of Madrid, was examined. Actually the ATPase activity in the mutant was nearly 50% higher than in the wild-type strain.)

Table 2 shows the results — the following conclusions can be drawn:

View this table:
Table 2

Rates of ATP hydrolysis by plasma membrane preparations from different yeast strains; all values were obtained in quadruplicate and are expressed in µmol orthophosphate per min per mg protein

Strain — grown on glucosePreincubated in waterPreincubated in glucoseIncrease after glucose
1 wild-type0.7011001.0351000.341
2 deficient in Gpa2p0.581831.004970.423
3 deficient in Hxk1p and Glk1p0.652931.1071070.455
4 deficient in Hxk1p, Hxk2p, Glk1p
5 deficient in Gpa1p0.650940.828800.178
6 wild-type0.4381000.8081000.370
7 deficient in Gpr1p0.5911350.9701200.379
8 wild-type0.3521000.7611000.409
9 deficient in Snf3p0.222630.738970.516
10 deficient in Rgt2p0.275780.632830.357
11 wild-type0.7141000.9851000.271
12 deficient in Snf3p0.467650.735750.268
13 deficient in Rgt2p0.429600.636650.207
14 deficient in Snf3p, Rgt2p0.7341030.979990.245
Strain — grown on galactosePreincubated in waterPreincubated in galactoseIncrease after galactose
1 wild-type1.0681001.2571000.189
2 deficient in Gpa2p0.609570.830660.221
3 deficient in Hxk1p and Glk1p0.854801.3701090.516
4 deficient in Hxk1p, Hxk2p, Glk1p0.427401.182940.755
5 deficient in Gpa1p1.004941.031820.027
6 wild-type0.4581000.6251000.167
7 deficient in Gpr1p0.6641451.0311650.367
8 wild-type0.3681000.5731000.205
9 deficient in Snf3p0.320870.463870.143
10 deficient in Rgt2p0.275860.556970.281
11 wild-type0.5661000.6691000.103
12 deficient in Snf3p0.388690.564840.176
13 deficient in Rgt2p0.435 770.579870.144
14 deficient in Snf3p, Rgt2p0.6851210.7991190.114
  • The range of results never exceeded 11%.

  1. Like with the acidification rates, there are substantial differences between the various wild-type strains.

  2. There was a marked decrease in ATPase activity in the Gpa2-deficient strain after galactose but a rather slight one after glucose. However, the increase following preincubation with galactose in this strain was in fact higher than in the wild-type.

  3. The effects of the hexokinase deficiencies are clear in that glucose phosphorylation is essential for ATPase activation but otherwise the increases or decreases of activity are difficult to interpret.

  4. Deficiency in the Gpa1 protein has no significant effect on ATP hydrolysis. However, there is virtually no increase following preincubation with galactose, compared with the wild-type (some 15%).

  5. In contrast with the acidification results a defect in the GPR1 gene causes a significant increase in ATPase activity (much like in [13]).

  6. Deficiency in either the Snf3 or the Rgt2 receptor gene is reflected in a not particularly marked decrease in ATPase activity in any of the four mutants tested, either after preincubation with glucose or with galactose (3–35%). However, most surprisingly, like with acidification, there is no decrease of activity in the double mutant and, in fact, a 20% increase after galactose. There is no relevant result in [13].

The above results support most but not all of the conclusions reached earlier [3,13], viz.

  1. Glucose, fructose, mannose and galactose (after growth on it) must be phosphorylated inside the cell. This is achieved by the appropriate hexokinase(s) in the case of Glc, Fru and Man, and by the inducible galactokinase in the case of Gal. The interval elapsing between addition of the sugar and the start of acidification (cf. Fig. 1), which is about 10 s for Fru, 15 s for Glc and Man and 40 s for Gal (see [3]), suggests that fructose-6-phosphate might represent the first signal molecule of the cascade, there being a single step to Fru-6-P from fructose, two steps from glucose or mannose, and four steps from galactose.

  2. This or a similar signal molecule then interacts with the Gpa2 protein, one of the two G-proteins known to be present in the S. cerevisiae genome; it is the one that is involved in adenylyl cyclase activation [20] and in pseudohyphal growth [21].

  3. This protein then, possibly in conjunction with the Gpr1 receptor [22], interacts with PLC and activates it.

  4. PLC then reacts with phosphatidylinositol 4,5-bisphosphate and splits it to inositol 1,4,5-trisphosphate and diacylglycerol. This compound then activates protein kinase C, the ultimate activator of the plasma membrane H+-ATPase.

  5. Participation of either the Snf3 or the Rgt2 receptors remains ambiguous. The effect of the Snf3 mutation on acidification ranged from 11 to 63% after glucose and from 21 to 25% after galactose. That of the Rgt2 mutation lay between 24 and 30% after glucose and between 21 and 48% after galactose. Moreover, the positive effect of the double mutation suggests that an alternative pathway is activated here toward the G-protein.

  6. What remains puzzling is the fact that the basal ATPase activity of all the strains studied is quite appreciable (cf. [1,2,23]), practically always higher after growth on galactose than after growth on glucose. The percent increase upon incubation with glucose is in fact often greater in the mutants than in the wild-type, the only exception being the gpa1 mutant. The increase in ATPase activity after incubation with galactose is generally smaller than after glucose (see Table 2). If these values were to serve as informative, then the Gpa1 protein would qualify as the G-protein involved in the H+-ATPase activation cascade.

  7. Another disquieting observation is the fact that in none of the cases where a mutation is involved the effect is greater than about 40%. How is the remaining 60% activity of the ATPase achieved then if one of the putative links in the cascade is missing? Are there perhaps alternative pathways? What are they? The only thing that remains certain is the necessity of sugar phosphorylation to start the cascade.

Before definitive conclusions can be drawn several queries must be dealt with. In the acidification process, the H+ produced by the H+-ATPase reaction must be clearly differentiated from those arising, e.g. during metabolism, particularly that of glucose [16]. Here, to be sure, the galactose-oriented experiments lead to more conclusive results than the glucose-oriented ones. Another aspect that is generally neglected is the fact that ATP-hydrolyzing activity as estimated here is not necessarily the same as the H+-extruding one. It was shown clearly that substantial ATPase activity can be measured in yeasts where no glucose-induced acidification takes place (Rhodotorula gracilis [24]) and that even in S. cerevisiae inhibitors can act very differently on the two processes [25].


The authors appreciate the skilled technical assistance of Ms. Pavla Herynková of the Institute of Physiology, CzAcadSci. The work was supported by Grant no. 204/02/1240 of the Grant Agency of the Czech Republic and by Institutional Research Concept no. AVOZ 5011922.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
View Abstract