YEAST

VOL. 8:

1077-1087 (1992)

Analysis of Glucose Repression in Saccharomyces cerevisiae by Pulsing Glucose to a Galactose-Limited Continuous Culture LAURENS N. SIERKSTRA, NICO P. NOUWEN, JOHN M. A. VERBAKEL* AND C. T H E 0 VERRIPS* Department of Molecular Cell Biology, University of Utrecht, Padualaan 8,3584 CH Utrecht, The Netherlands * Unilever Research Laboratorium Vlaardingen, Olivier van Noortlaan 120,3133 AT Vlaardingen, The Netherlands

Received 16 December 1991; accepted 5 June 1992

In this study, glucose repression in Saccharomyces cerevisiae was analysed under defined physiological conditions, at both the molecular and physiological levels, by pulsing glucose to a galactose-limited continuous culture. During this pulse of glucose, the galactose feed was kept constant. Directly after the glucose pulse, carbon dioxide production increased while oxygen consumption remained constant, demonstrating that the surplus of glucose had been consumed by means of fermentation. The direct accumulation of galactose in the medium after the glucose pulse indicated that the consumption of galactose had been stopped instantaneously. Galactose uptake experiments revealed that the galactose transporter was still present but apparently was incapable of galactose uptake, which could be due to inhibition of the galactose transporter by glucose. The total concentration of CAMP increased from 5 nmol ggl at t = 0 to 25 nmol g-I at t = 1.5 min. After 2 min the concentration of CAMP gradually decreased again to the normal level. Within 2 min after the addition of glucose, the transcription of the G A L genes and SUC2was inhibited. In addition, the transcription of the HXKl gene, encoding hexokinase isoenzyme 1, was also inhibited, which demonstrates that the HXKI gene is regulated at the transcriptional level comparable with invertase. KEY WORDS -Yeast; glucose repression; continuous

culture; transcriptional regulation.

INTRODUCTION In Saccharomyces cerevisiae, galactose is taken up by the galactose transporter (encoded by GALZ) and subsequently converted into glucose-6-phosphate by the action of the GAL1 (galactokinase), GAL7 (galactose- 1-phosphate uridyl transferase), GAL10 (uridine diphosphoglucose-4-epimerase)and GAL5 (phosphoglucomutase) gene products. These GAL genes are induced in medium containing galactose and the transcription is repressed when the medium contains glucose, except for GALS, which is thought to be constitutive. The mechanism of galactose induction has been studied extensively and involves the GAL3, GAL4 and GAL80 gene products (Johnston, 1987). In this system the GAL3 gene product, which shows protein sequence similarity with galactokinases, is responsible for the synthesis of a still unknown inducer when galactose is present (Bajwa et al., 1988; Bhat and Hopper, 1991). The inducer prevents GAL80 from inhibiting the transcriptional activator GAL4, which subsequently activates the 0749-503X/92/12 1077-1 1 S 10.50 0 1992 by John Wiley & Sons Ltd

transcription of the GAL genes (Johnston, 1987). In addition, phosphorylation has been shown to be important for post-translational regulation of GAL4 (Mylin et al., 1989). The mechanism by which glucose represses the transcription of the GAL genes is poorly understood. Recently Nehlin et al. (1991) have shown that a gal80 deletion mutant shows only a 5.8-fold derepression of the GALl mRNA as compared with the fully derepressed, induced level, indicating that GAL80 is only partially involved in the repression of the GAL genes. The subsequent deletion of MZG1, a transcriptional regulator (Nehlin and Ronne, 1990), resulted in a 50-fold derepression of GALl expression. Glucose not only regulates transcription of genes but addition of glucose to non-repressed cells also causes the rapid activation, e.g. trehalase (van der Plaat, 1974; Ortiz et al., 1983; Uno et al., 1983), or inactivation of enzymes, e.g. fructose- 1,6biphosphatase (Gancedo and Gancedo, 1971; Holzer, 1976). For the transcriptional inactivation of ADH2 by the regulator ADRl and for the

1078 regulation of fructose- 1,6-biphosphatase it has been shown that this is due to phosphorylation by the CAMP-dependent protein kinase (Cherry et al., 1989; Rittenhouse et al., 1987). The signal for glucose excess could in this case be transmitted by the RAS/cAMP pathway, although this pathway seems to be a general signal transduction pathway (Broach, 1991). Most data obtained for the mechanism which regulates glucose repression are from mutant studies in shake flasks under non-defined physiological conditions. Our approach to study glucose repression is to grow a commercial bakers' yeast strain, without any growth deficiencies, under very well-defined conditions in continuous cultures. In this study, we investigated the changes at both the molecular and physiological levels after a glucose pulse to a galactose-limited continuous culture, by analysing activities and mRNA levels of key enzymes in carbon metabolism, and concentrations of glycolytic intermediates, cAMP and ATP. MATERIALS AND METHODS Strain and growth conditions

Commercial bakers' yeast strain SU32 was grown on synthetic medium containing 7.63 g 1-' NH,Cl, 2.81 g 1-' KH,PO,, 0.59 g 1-' MgSO, . 7H,O, 10 ml I-' trace elements (a 100 x concentrate containing 5.5 g I - ' CaCl, . 2H,O, 3.75 g 1-' FeSO, . 7H,O, 1.4 g 1-' MnSO, . H,O, 2.2 g 1-' ZnSO, . 7H,O, 0.4g 1-' CuSO, . 5H,O, 0.45 g 1-' CoCl, .6H,O, 0.26 g 1-' Na,MoO, . 2H,O, 0.4 g 1-' H,Bo,; 0.26 g 1-' KI and 30 g 1-' NaEDTA) and 1.5 ml 1vitamin solution (a 1000 x concentrate containing 0.05g 1-' biotin, 5 g 1-' thiamin, 47g I-' myoinositol, 1.2g 1-' pyridoxin and 23g 1-' panthotenic acid) per 18 g 1-' galactose in the feed. S. cerevisiae SU32 was grown at 30°C in a fermenter with a 2 1 working volume connected to an Applikon AD11020 controller unit (Applikon). The pH was automatically controlled at 5-0 by the addition of ~ M - N H , O H The . airflow was 21/min and the dissolved oxygen tension was kept above 20% by regulating the stirrer speed. Carbon dioxide production, oxygen consumption and ethanol formation were measured on line (approximately every 3 min) by connecting the headspace of the fermenter to a VG gas analysis mass spectrometer MM8-80 (VG instruments). A glucose pulse to the continuous culture was given at a dilution rate of 0.2 h-' by the addition of

L. N. SIERKSTRA ET AL.

glucose to the fermenter by means of a syringe to an end concentration of 80 mM. Preparation of samples

For the preparation of cell-free extracts, used in enzyme assays, 1.5ml of culture liquid was transferred from the fermenter to an Eppendorf tube and put on ice. The cells were pelleted for 30 s by centrifugation and resuspended in 1 ml 1OmMpotassium phosphate buffer (pH 7.5) containing 2 mM-EDTA. The cells were again pelleted and resuspended in 1 ml 100 mwpotassium phosphate buffer (pH 7.5) containing 2 mM-MgC1, and 2 mMdithiothreitol. Cell-free extracts were prepared by adding an equivalent volume of glass beads to the cells. To obtain lysis, the suspension was shaken four times at maximum speed on a vortex mixer for 30s with 1-min intervals on ice. Lysed cells were separated from the glass beads, after which the suspension was centrifuged to remove cell debris. The clear supernatant was quickly frozen in liquid nitrogen and kept at -80°C until it was used for enzyme assays. The protein concentration of cellfree extracts was determined by the method of Bradford with the Biorad dye-reagent using bovine serum albumin (Sigma) as a standard. Protein concentrations were always between 1-6 mg/ml extract, The procedure for the mRNA extraction from yeast cells was as described by Sierkstra et al. (199 1). Samples for determination of glycolytic intermediates, ATP and cAMP were taken by quickly transferring 3 ml ofculture liquid from the fermenter to a tube in liquid nitrogen (within 10 s). These were stored at -80°C. The frozen culture liquid was quickly heated to 80°C in boiling ethanol for 1 min to inactivate enzymes as quickly as possible. The ethanol/culture liquid mixture was removed by vacuum drying. The resulting pellet was dissolved in 1 ml H,O and stored at -80°C until assayed. All enzymes were inactivated by this extraction procedure. Sampling of culture liquid for the determination of extracellular metabolites (acetate, acetaldehyde, glycerol, pyruvate, ethanol and the residual glucose concentration) was performed as described by Postma et al. (1988). For the determination of dry weight, 20 ml culture liquid was collected from the fermenter's outlet. The cells were pelleted, washed with distilled water and transferred to pre-weighed glass vials, in which they

1079

ANALYSIS OF GLUCOSE REPRESSION IN S. CEREVISIAE

Table 1. Oligonucleotidesused for the detection of mRNA mRNA detected HXK2 PGII PDCI HXKI suc2 CALI GAL2 GAL7 GAL10 ACT1 H2A

Oligonucleotidesequence

SAATTGGCTCAGAGATACCTTGTGGGAA3' SACTTCGGTTTCGTTAGTGGACAAAGCAGCG3' SAGACATTCTGTGGAAAACAGTGAAGTCACC3'

STAAGGTGTCCTTGGTGTTTAGCAATTC3' STGGGTCAGTGTTGAAGAAAGTTT'GCAAGGC3' SAGCTAAAGCAACGGCACAAATGAATGCGGC3'

STTGTTCTCCTCAACTGCCAT3' S'GAGCCTAACGGCAGCATAATCATTGGGGAA3'

SGGTGGCGCCTATATAAGCACTATCAGGATT3'

TTGTCTTGGTCTACCGACGATAGATGGGAAG3' TGACTGGAGCACCAGAACCAATTCTTTGGGC3'

were dried overnight in a 120°C stove. After weighing, the amount of dry weight per litre culture liquid (gl- I ) was calculated. Measurement of metabolites, A T P and cAMP Extracellular metabolites were measured by means of high-pressure liquid chromatography analysis and/or enzymatically on the Cobas Mira autoanalyser (Hoffmann Laroche). All other metabolites were measured spectrophotometrically, essentially as described by Bergmeyer (1974) on a Philips UVjVIS spectrophotometer series PU8700. Measurement of total cAMP was done with the Amersham cAMP kit according to the instruction manual. Galactose uptake assay Cells were harvested from the fermenter, immediately put on ice and used for the galactose uptake assay within 15 min. The galactose uptake assay was performed as described by Postma et al. (1988) with the following modifications: uniformly labelled ['4C]galactose with a specific activity of 1 x 10" Bq mmol-' was used for every incubation; incubation times were 6 s (longer incubation times led to non-linearity of the assay); the results given are the mean of four experiments and the radioactivity was measured in a Packard 1900CA liquid scintillation analyser. Enzyme assays Enzyme activities were measured under V,,, conditions essentially as described by Bergmeyer

(1974). Reaction velocities were proportional to the amount of enzyme added. Activities are in U mg-' total cellular soluble protein, in which one unit is defined as the conversion of 1 pmol of substrate in 1 min at 30°C at pH 7.0. All enzyme assays were performed on the Cobas Mira autoanalyser of Hoffmann Laroche. Enzyme activities are related to the protein content of the cell extracts, which was measured by the method of Bradford, using the Biorad reagent, with bovine serum albumin (BSA) as a standard, on a Philips UV/VIS spectrophotometer series PU8700. Labelling of fragments and oligonucleotides

For the detection of mRNA of HXKI (hexokinase isoenzyme l), HXK2 (hexokinase isoenzyme 2), SUC2 (invertase), PGZl (phosphoglucoisomerase), PDCl (pyruvate decarboxylase), GAL2, GAL10, GAL1, GAL7, actin ( A C T ) and histon 2A (H2A), the oligonucleotides shown in Table 1 were used. Oligonucleotides were labelled by incubating 25 pmol of oligonucleotide with 50 pCi of [32P]ATP and 1 unit of polynucleotide kinase (Amersham). Northern blot analysis

The RNA samples (containing 3 pg of total RNA) were separated on a denaturating formamide/ formaldehyde gel. The RNA was blotted on Hybond paper using the vacugene system of Pharmacia or by capillary blotting. The RNA was crosslinked to hybond by exposure to UV-light. The procedure for

1080 the hybridization using oligonucleotides was as follows: blots were prehybridized for at least 2 h in hybridization mix (50 mM-Tris-HC1, pH 7.5, 10mM-EDTA, 1 M-NaCl, 0.1% SDS, 0.1% Napyrophosphate, 0.2% Ficoll, 0.2% BSA, 0.2% PVP and a denatured mixture of 0.1 mg/ml singlestranded DNA and 0.01 mg/ml poly-rA) at 42°C. Hybridization was performed overnight at 42°C in hybridization mix with the labelled oligonucleotide. The filters were washed once for 20 rnin with 5 x SSC, followed by once with 2 x SSC and once with 1 x SSC, all at 42°C. The filters were thereafter exposed to X-ray films. As an internal control for the amount of mRNA blotted, the actin or histon mRNA was used by means of double hybridizations. The amount of specific mRNA was measured by means ofdensitometry on an LKB Ultroscan XL. RESULTS Physiological parameters

L. N. SIERKSTRA ETAL.

COP mmol . h-' 0,mmol . h-' 760 300

RQ,ethanol 91-'

9...

.. ....

350

150

0

0

10

I

I

1

I

30

50

70

90 t (min)

Figure 1 . Carbon dioxide production (- . -), oxygen conS. cerevisiae SU32 growing at a dilution rate of sumption (---), respiratory quotient (RQ) (-) and ethanol 0.2h-I had a specific oxygen consumption of production (. . .0.. .) by S. cerevisiae SU32 after the glucose 5.1 mmol 0, g-' per h, a specific carbon dioxide pulse. production of 5.4mmol CO, g-' per h and consequently a respiratory quotient (RQ) of 1.06. At t = O a pulse of glucose was given to the Galactose and glucose concentrations; galactose uptake experiments galactose-limited continuous culture. In Figure 1 the carbon dioxide production, oxygen consumpAt a dilution rate of 0.2 h-I in a galactose-limited tion, RQ and ethanol concentration as a function of continuous culture, the residual galactose concenthe pulse time are shown. The uptake of glucose by tration was 0.2 g 1-' (Figure 2). The glucose pulse the yeast immediately resulted in an increase in raised the concentration of glucose in the medium the carbon dioxide production by the culture from to 15g I-'. The glucose concentration decreased 75 mmol h-I to 260 mmol h-' and in ethanol pro- immediately (within 1 min) after the pulse. At 30 rnin duction. Also, acetate and pyruvate formation and after the pulse, the concentration had decreased an increase in the glycerol concentration occurred, to 8 g 1-I and after 60 rnin the glucose was nearly while no acetaldehyde was observed in the culture consumed (0.4g 1-I). This decrease in glucose liquid during the pulse (results not shown). The concentration equals the maximum glucose conoxygen consumption by the culture was not sumption by S. cerevisiae SU32 under conditions of increased upon addition of glucose. The fast rise in glucose excess (V,,, = +_ 3.4 g glucose g- I per h). carbon dioxide production during the first 10 min The galactose concentration in the medium was followed by a slight increase in carbon dioxide increased to 1.6g 1-' at 30 rnin after the pulse. production until 5&60 rnin after the pulse. In this Between 3 W O rnin after the glucose pulse, the time interval, only a small increase in the specific yeast resumed galactose uptake, while there was still glucose present. At 60 rnin after the pulse, the oxygen consumption was observed. After 50-60 min, the glucose was nearly con- galactose concentration had decreased to 0.9 g I-'. Galactose uptake experiments, with 10 mM of sumed, which could be seen by a decrease in the carbon dioxide production. Subsequently, the yeast labelled galactose as a substrate, were performed at started consuming both galactose and ethanol as t = 0 min and 30 rnin after the pulse, to investigate if carbon source: the ethanol having been produced the yeast was still capable of galactose uptake. The during the pulse. Therefore the RQ decreased from V,,, of the galactose uptake s stem was decreased from 0.24 mmol galactose g-' per min at t = 0 to 3-5 at t = 50 rnin to 0-86at t = 120 min.

1081

ANALYSIS OF GLUCOSE REPRESSION IN S. CEREVISIAE galactose gi - 1

glucose gi - 1

- 1.6

10

- 1.2

-

-

0.8

- 0.4 6-0 60

30

120 t (min)

90

Figure 2. Residual glucose ( 0 )and galactose concentrations (0)in g 1-' at time intervals after the glucose pulse to S. cerevisiue growing at a dilution rate of 0.2 h-' in a galactoselimited continuous culture.

0

4 II

I

I

1

I

5

10

15

20

'0 t (min)

Figure 3. Concentrations of the intracellular metabolites glucose-6phosphate (G6p; O),fructose-6-phosphate (F6p; B),fructose- 1,6-diphosphate (F1,6dp; 0)and glucose-I-phosphate (Glp; 0) in pmol g-I.

0.1 mmol galactose g-' per rnin at t = 30 min. This demonstrates that 30 rnin after the glucose pulse, the galactose uptake system was still present. Concentrations of CAMP, ATP andglucose metabolites

After the pulse, the respective concentrations of glucose-6-phosphate and fructose-6-phosphate increased from 1.9k0.3 pmol g- at t =0 min to a maximum of 5-6kO.6pmol g-' at t = 1 rnin and

'

from 0.32 k0.07pmol g-' at t = 0 min to a maximum of 1-15+0.07pmol g-' at t = 1.5 min (Figure 3). The concentrations of these metabolites rapidly decreased 2 rnin after the glucose pulse to an intermediate level between the t = O and t = 1-5 min values. The concentrations of ATP (Figure 4) and glucose- 1-phosphate (Figure 3) remained constant between 0 and 20 min after the pulse at 2-8-3.5 pmol g-' and 2.2-3.2 pmol g-' respectively. Fructose1,6-biphosphate (Figure 3) increased from 2.5 0.14pmol g-' a t t=O to a maximum of

*

1082

L. N.SIERKSTRA ETAL.

cAMP nmol g-'

0

ATP pmol g - '

I

I

I

5

10

15

t (min) Figure 4. Concentrations of total cAMP (both intra- and extracellular) in nmol g-I (0)and the concentrationof ATP in pmol g - ' (0).

12-7+2.2pmol g-' at t = 1.5 min. No decrease of the fructose- 1,6-biphosphate concentration was observed during the first 20 rnin after the pulse. The total cAMP concentration (sum of intracellular and extracellular CAMP) increased from 5.3 f 0 . 5 nmol g-' at t = O rnin to a maximum of 2 5 9 f 2.2 nmol g-' at t = 1 rnin (Figure 4). After 2-3 min the concentration of cAMP gradually decreased again to the normal level. Northern analysis mRNA samples were prepared at different times after the glucose pulse and Northern analyses were performed to investigate transcriptional regulation (Figure 5). The amount of specific mRNA was measured by densitometry with actin or histon as an internal control for the total amount of mRNA on the gel. The transcription of the GAL2, GALIO, GAL7, G A L ] , HXKI and SUC2 genes was inhibited 1.5-2 min after the glucose pulse. Assuming that no new mRNA was produced, the half-life times of the GAL2 and HXKI mRNA are between 3&45 s. The GALIO and SUC2 mRNA have a half-life time of 1 min, while GAL7 and GAL1 have a half-life time of about 2 min. The amount of PGIl mRNA remained relatively constant, while the HXK2 mRNA first remained constant but increased four-fold at 60 rnin after the pulse. The PDCI mRNA started to increase between 5-7.5 min after the pulse, and at 60 rnin the PDCI mRNA had increased 5.4-fold. On all Northern blots, a decrease in the amount of actin mRNA, and a more pronounced decrease in

the amount of histon mRNA, was observed at t = 30 min and t=60 after the pulse, as compared to the 3 pg of total RNA loaded on the gel. Enzyme activities

Crude extracts were prepared and the activities of the above-mentioned enzymes were measured at time intervals after the glucose pulse to investigate if specific inactivation/degradation of enzymes had occurred. No significant changes in enzyme activities could be detected after the glucose pulse. DISCUSSION Upon the addition of glucose to the galactoselimited culture, an increase in carbon dioxide and the production of ethanol were observed, while the oxygen consumption by the yeast remained constant. This leads to the conclusion that the yeast is not capable of instantly increasing the capacity to consume glucose by means of oxidative phosphorylation, although normally S . cerevisiue SU32 can consume 6.5-6.8mmol oxygen g-' per h, as opposed to the 5.0mmol oxygen g-' per h at the moment of pulsing (Sierkstra et al., submitted for publication). Following the initial fast rise in CO, production, a slow increase was observed between 1G60 rnin after the pulse, while the oxygen consumption remained almost constant. The absence of an increase in oxygen consumption demonstrates that newly synthesized cells are only fermenting glucose into

ANALYSIS OF GLUCOSE REPRESSION IN S. CEREVISIAE

ethanol and carbon dioxide. This was confirmed by the growth yield of 0.15 g g-' during the pulse, which is typical for fermentative growth. It has been shown by van Urk (1989) that upon the addition of glucose to a glucose-limited continuous culture of S. cerevisiae, the increase in biomass reflects the increase in p (growth rate). This increase in biomass is not due to the synthesis of reserve or stressinduced carbohydrates like glycogen and trehalose, which occurs predominantly in Crabtree-negative yeasts (van Urk, 1989). Immediately after the glucose pulse, the galactose concentration in the medium increased at a rate approximately equal to that of the feed. This indicates that the consumption of galactose by the yeast is instantly stopped upon the addition of glucose. It has been shown that inactivation of the galactose permease is mediated by the CAMPdependent protein kinase, either by direct phosphorylation of the galactose permease or by an indirect mechanism (Ramos and Cirillo, 1989). This could be the reason for the fact that galactose consumption is stopped in our experiments. To test this hypothesis, we investigated whether the yeast was still capable of galactose uptake after the glucose pulse by means of a galactose uptake experiment 30 rnin after the pulse. This revealed that the galactose uptake system was still present but that the V,,, of the uptake system had decreased. Therefore we can conclude that the inactivation of the galactose uptake system by a mechanism involving phosphorylation by the CAMP-dependent protein kinase is probably not the only reason for the fact that the galactose consumption is completely stopped in our experiment. Another reason can be the allosteric inhibition of the galactose permease by glucose or a glucose metabolite. However, we cannot exclude the possibility that galactose metabolism is inhibited at a level other than galactose uptake or that dephosphorylation of the galactose permease occurs during the uptake assays. Upon addition of glucose to the galactose-limited continuous culture, the total concentration of cAMP (intra- and extracellular) increased from 5 nmol gg' to 26 nmol g-' within 1 min. While others have always observed a sharp decrease in the cAMP level after 30-60 s (e.g. Beullens et al., 1988), the level of cAMP in our experiment decreased gradually 2-3 min after the pulse. Additional experiments have revealed that this slow and gradual decrease in the total cAMP concentration is due to the fact that secretion of cAMP into the medium occurs upon the addition of glucose

1083 (unpublished results). Currently we are investigating the mechanism and relevance of the cAMP secretion. Northern analysis of glucose-repressible genes revealed that these mRNAs have very low half-life times. The lowest half-life time, of about 3 W 5 s, was observed for H X K l and GALZ. The actin and histon mRNA also decreased 30-60 rnin after the pulse, as compared to the 3pg of total RNA loaded on the gel. This, however, does not interfere with the measured half-life times because they are based on the interval between t = O to t = 10 rnin after the pulse. The relative decrease of actin and histon mRNA is most likely due to the increased synthesis of growth-specific RNAs (like ribosomal RNAs), which would decrease the relative amount of the actin transcript in the total amount of RNA. At 2 rnin after the glucose pulse, the transcription of the galactose-utilizing enzymes, GAL2, CALI, GAL7 and GALIO, and of SUC2 and HXKl is stopped. The inhibition of transcription is very rapid (within 2 min) and the same pattern is observed for both GAL genes and SUC2 as well as HXKI. The fact that gal80 deletion strains only have a five-fold derepression of the GAL1 mRNA (Nehlin et al., 1991) indicates that glucose repression of the GAL genes is independent of the induction pathway. Our results suggest that a specific and rapid mechanism exists which is capable of controlling the level of glucose-regulated mRNAs. Recently evidence has been obtained that phosphorylation of the transcriptional activator ADRl regulates the mRNA level of ADH2 (Cherry et al., 1989) and that GAL4 is a transcriptional activator for which the phosphorylation state also seems to be important for activity (Mylin et al., 1989). These recent results, together with the short time course in which transcriptional inhibition occurs, which rules out newly synthesized proteins as repressors, make it likely that the transcriptional inhibition is mediated by the phosphorylation of transcription factors. A candidate for phosphorylation is the MIGl protein, a recently isolated transcriptional regulator (Nehlin and Ronne, 1990). MIGl binding sites have now been identified in the sequence of a number of glucose-repressible genes, e.g. SUC2, the GAL genes, GAL4 (Nehlin and Ronne, 1990; Nehlin et al., 1991) and fructose-1,6biphosphatase (FBP; Mercado et al., 1991). In addition, our results demonstrate that HXKl is a glucose-repressible gene, as suggested by Bisson and Fraenkel(l983), whose transcription is regulated in

3

5

25

10

20 30 60

100 150 36 14 14

7

0

0

0

1 1.5

0

2

GAL 10 ACT

0

1

1.5

100

90

110 90

0

1

2

3

5

75

0 11 10

20

60

30

PDC ACT

95 140 190 205 300 475 540

1.5

2

3

5

15

10

20

30

60

100 100 100

50

15

0

0

0

0

0

25

GAL2 ACT

0

1

1.5 2

3

5 75 10 20 30 60

GAL 7 ACT 100 75 140 114 55 59 31 13 0

0

0

Figure 5a. Figure 5. Northern analysis ofmRNA isolated at different time intervals after the glucose pulse. This figure shows the HXKZ, H X K I , PGI, G A L I , GAL2, GALIO, SUC2 and PDC mRNA with actin mRNA as an internal control. For the GAL7 mRNA, histon 2A2 was used as an internal control. Beneath each lane, the amount of specific mRNA, corrected for the internal control, is given as a percentage of the initial value (at t =O).

1

8

1.5

2

3

5

25

10

20

30

60

80

43

13

4

0

0

0

21

3

5

75

10

SUC 2 ACT 100 114 100

0

1

1.5

2

20 30 60

GAL7 HISTON 100 105 130 145 95 45

0

1

1.5

2

30 15

5 25 10

3

0

0 50

20 30 60

HXKI ACT

100 96 95 37 24 0 0

0

0

0

0

nd

25

10

20

30 60

25 10 20 30 60

PG I ACT 100 46 38 38 46 69 0

1

1.5

2

3

5

HXK2 AC T 100 75 100 100 125 100 50 75 100 150 400

1086

L. N. SIERKSTRA ETAL.

Table 2. Comparison between the sequence in the H X K l promoter (Kopetzki et al., 1985) and the consensus sequence for the binding of the MIGl protein according to Griggs and Johnston (1991). The sequence in the HXKl promoter is complementary and in the opposite orientation as compared to the consensus given by Nehlin et al. (1991) C C C C R S AWWWW

Consensus according to Nehlin et al. (1991)

C T C C CGGA T T T T

Sequence in HXKl promoter (-348 to -339)

1 l I : : l : : : :

111:111I:I

T C C C C RGA T T N T

Consensus according to Griggs and Johnston (1991)

coordination with invertase. A sequence homology search revealed that the H X K l promoter, in agreement with other glucose-repressible genes, contains a sequence which closely resembles the consensus sequence (Nehlin et af., 1991; Griggs and Johnston, 1991) for the binding of the MIGl protein (Table 2).

REFERENCES Bajwa, W., Torchia, T. E. and Hopper, J. E. (1988). Yeast regulatory gene G A L 3 Carbon regulations; UAS,, elements in common with G A L l , GALZ, GAL7, GALlO, GAL80 and M E L l ; Encoded protein strikingly similar to yeast and E. coli galactokinases. Mol. Cell Biol. 8,3439-3447. Bhat, P. J. and Hopper, J. E. (1991). The mechanism of inducer formation ingal3 mutants of the yeast galactose system is independent of normal galactose metabolism and mitochondria1 respiratory function. Genetics 128, 233-239. Bergmeyer, H. U. (1974). Methods of Enzymatic Analysis. Verlag Chemie, Weinheim. Beullens, M., Mbonyi, K., Geerts, L., Gladines, D., Detremerie, K., Jans, A. W. H. and Thevelien, J. M. (1988). Studies on the mechanism of the glucoseinduced CAMP signal in glycolysis- and glucose repression-mutants of the yeast S . cerevisiae. Eur. J. Biochem. 172,227-23 1. Bisson, L. F. and Fraenkel, D. G. (1983). Involvement of kinases in glucose and fructose uptake by Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 80, 1730-1734. Broach, J. R. (1991). RAS genes in Saccharomyces cerevisiae: signal transduction in search of a pathway. TIG 7,28-32. Cherry, J. R., Johnson, T. R., Dollard, C., Shuster, J. R. and Denis, C . L. (1989). CAMP-dependent protein kinase phosphorylates and inactivates the yeast transcriptional activator ADRl . Cell 56,409-419. Gancedo, J. M. and Gancedo, C. (1971). Fructose-1,6diphosphatase, phosphofructokinase and glucose6-phosphate dehydrogenase from fermenting and non-fermenting yeast. Arch. Microbiol. 76, 132-1 38.

Griggs, D. W. and Johnston, M. (1991). Regulated expression of the GAL4 activator gene in yeast provides a sensitive switch for glucose repression. Proc. Natl. Acad. Sci. USA 88,8597-8601. Holzer, H. (1976). Catabolite inactivation in yeast. TIBS 1, 178-181. Johnston, M. (1987). A model fungal gene regulatory mechanism: the G A L genes ofSaccharomyces cerevisiae. Microbiol. Rev. 51,458-476. Kopetzki, E., Entian, K. D. and Mecke, D. (1985). Complete nucleotide sequence of the hexokinase PI gene (HXK1) of Saccharomyces cerevisiae. Gene 39, 95-102. Mercado, J. J., Vincent, 0. and Gancedo, J. M. (1991). Regions in the promoter of the yeast FBPl gene implicated in catabolite repression may bind the product of the regulatory gene M I G l . FEBS 291,97-100. Mylin, L. M., Bhat, J. P. and Hopper, J. E. (1989). Regulated phosphorylation and dephosphorylation of GALA, a transcriptional activator. Genes Dev. 3, 1157-1 165. Nehlin, J. 0. and Ronne, H. (1990). Yeast MIGl repressor is related to the mammalian early growth response and Wilms’ tumour finger proteins. EMBO 9, 2891-2898. Nehlin, J. O., Carlsberg, M. and Ronne, H., (1991). Control of yeast G A L genes by MIGl repressor: a transcriptional cascade in the glucose response. EMBO 10,3373-3377. Ortiz, C. H., Maia, J. C. C., Tenan, M. N., Braz-Padrao, G. R., Mattoon, J. R. and Panek, A. D. (1983). Regulation of yeast trehalase by a monocyclic, cyclic AMP-dependent phosphorylation-dephosphorylation cascade system. J . Bacteriol. 153,644-65 1. van der Plaat, J. B. (1974). Cyclic 3’,5’-adenosine monophosphate stimulates trehalose degradation in bakers’ yeast. Biochem. Biophys. Res. Comm. 56,580-587. Postma, E., Scheffers, W. A. and van Dijken, J. P. (1988). Adaption of the kinetics of glucose transport to environmental conditions in the yeast Candida utilis CBS 621: a continuous culture study. J . Gen. Microbiol. 134,1109-1 116. Ramos, J. and Cirillo, V. P. (1989). Role of cyclic-AMPdependent protein kinase in catabolite inactivation of

ANALYSIS OF GLUCOSE REPRESSION IN S. CEREVISIAE

the glucose and galactose transporters in Saccharomyces cerevisiae. J. Bacteriol. 171,3545-3548. Rittenhouse, J., Moberley, L. and Marcus, F. (1987). Phosphorylation in vivo of yeast fructose-] ,6-biphosphatase at the CAMP-dependent site. J. Biol. Chem. 26, 101 14-101 19. Sierkstra, L. N., Verbakel, J. M. A. and Verrips, C. T. (1991). Optimisation of a host/vector system for heterologous gene expression by Hansenulapolymorpha. Curr. Genet. 19,81-87.

1087 Smith, M. E., Dickinson, J. R. and Wheals, A. E. (1990). Intraceilular and extracellular levels of CAMP during the cell cycle of S. cerevisiae. Yeast 6,53-60. Uno, I., Matsumoto, K., Adachi, K. and Ishikawa, T. (1983). Genetic and biochemical evidence that trehalase is a substrate of CAMP-dependent protein kinase in yeast. J. Biol. Chem. 258, 10867-10872. van Urk, H. (1989). Transient responses of yeasts to glucose excess. Thesis. Technical University, Delft.