In this work, we describe the hexokinase 2 (Hxk2) signalling pathway within the yeast cell. Hxk2 and Mig1 are the two major factors of glucose repression in Saccharomyces cerevisiae. The functions of both proteins have been extensively studied but there is no information about possible interactions among them in the repression pathway. Our results demonstrate that Hxk2 interacts directly with Mig1 in vivo and in vitro and that the ten amino acids motif between K6 and M15 is required for their interaction. This interaction has been detected at the DNA level both in vivo by chromatin immunoprecipitation experiments and in vitro using purified proteins and a DNA fragment containing the MIG1 site of the SUC2 promoter. This demonstrates that the interaction is of physiological relevance. Our findings show that the main role of Hxk2 in the glucose signalling pathway is the interaction with Mig1 to generate a repressor complex located in the nucleus of S. cerevisiae.
The ability of the yeast Saccharomyces cerevisiae to grow in different media is due to its capacity to sense and respond to changes in the availability of nutrients. Among sugars, glucose is probably the major signalling nutrient for S. cerevisiae as well as the carbon and energy source used preferentially. S. cerevisiae is able to detect the glucose levels of extracellular medium and to generate intracellular signals that result in an adequate cellular response to variations in the glucose medium composition. Many of these responses involve alterations in gene expression. The majority of these alterations occur at the level of mRNA transcription by the repression or activation of the transcription of many genes that encode enzymes implicated in carbon metabolism [1–3].
Present knowledge concerning factors required to transmit the glucose signal from the environment to the cell nucleus is mainly derived from studies with mutants affected in different regulatory circuits governed by glucose . In the last few years important advances in this field have been made [4–6]. These studies have characterized, in different detail, several signal transduction pathways that allow the yeast to perceive the level of glucose in the medium and to initiate the appropriate metabolic response (Figure 1).
Overview of Snf3/Rgt2, Gpr1 and Mig1/Hxk2 signalling pathways
S. cerevisiae has membrane proteins that act as glucose receptors. Glucose binds to these receptors and generates an intracellular signal. In the Rgt2/Snf3 pathway, these two proteins act as glucose receptors. The Rgt2 and Snf3 proteins resemble hexose transporters in structure but have long cytoplasmic tails that are required for signal transduction . Glucose binding to these transmembrane proteins initiates signals that activate a pathway that allows hexose transporter gene expression by repressing Rgt1 function .
An additional pathway that involves transcriptional changes in response to glucose is the stimulation of adenylyl cyclase and the increase in intracellular cyclic AMP. This pathway includes a G-protein coupled receptor (Gpr1) and two G proteins Gpa1 and 2, necessary for the glucose-specific increase in cAMP [9,10]. Finally, glucose activation of adenylyl cyclase leads to activation of the cAMP-dependent protein kinase A (PKA). Upon activation of PKA by cAMP the Rap1 transcription factor activity is increased inducing expression of genes encoding ribosomal proteins and proteins required for glycolysis. In addition, PKA inactivates other transcription factors, such as Msn2 and Msn4, down-regulating the expression of STRE-controlled genes .
Once inside the cell, glucose activates another pathway involved in the repression of genes not needed during growth on glucose. In this pathway, both the Mig1 and the hexokinase 2 (Hxk2) proteins are necessary to generate the glucose repression signal. Thus, in hxk2 mutants repression of several genes by glucose is no longer operative even in the presence of the functional MIG1 gene. The mechanism by which Mig1 operates in this system is quite well known [12–14]. However, the mechanism by which Hxk2 participates in the glucose signalling pathway, is almost unknown until now, in particular, the position of this factor in the signalling cascade and the interactions with the other factors of the cascade. Here we will briefly describe the Hxk2 signalling pathway in the S. cerevisiae cell, a protein that in addition to its classical metabolic role plays an important function in glucose signalling.
Hxk2 interacts with the Mig1 protein and can enter the nucleus
Hxk2, Mig1 and Med8 form part of the same protein complex that interacts with a regulatory element of the SUC2 promoter located between −451 and −416 bp [15–20]. However, an interaction between Hxk2 and Mig1 proteins has not been demonstrated. To test this possibility, we have performed yeast two-hybrid assays, immunoprecipitation and glutathione S-transferase (GST)-pull down experiments.
Our results demonstrate that about 15% of the total Hxk2 protein is localized in the nucleus and that this localization is required for glucose repression of several genes . A possible explanation of why in a wild-type strain a limited fraction of total Hxk2, about 15%, is located in the nucleus and that this amount was not modified when the cell content of Hxk2 was increased (for example, by expressing the HXK2 gene contained in a high copy number plasmid), could be that Hxk2 is sequestered in the nucleus by direct interaction with Mig1 and thus, the amount of nuclear Hxk2 depends on the amount of Mig1 present in the nucleus. We have also found that Hxk2 interacts directly with Mig1 in vivo and in vitro and that the ten amino acid motif between K6 and M15, previously identified as essential for Hxk2 translocation to the nucleus, is actually required for the Hxk2–Mig1 interaction .
Hxk2 binds to a SUC2–Mig1 complex both in vivo and in vitro
To determine the biological relevance of the interaction between Hxk2 and Mig1, we have investigated whether these two proteins co-localize in DNA–protein complexes in vivo by chromatin immunoprecipitation experiments and in vitro by gel-retardation assays.
Chromatin immunoprecipitation experiments have shown that the Mig1 and Hxk2 proteins are actively recruited to a DNA fragment of the SUC2 promoter (−451/−416) containing the Mig1 binding site in response to high-glucose conditions (Figure 2).
Both Mig1 and Hxk2 are bound to the SUC2 promoter under high-glucose conditions as detected by chromatin immunoprecipitation analysis
To demonstrate that the interaction is produced directly between both proteins without the concurrence of other auxiliary proteins, we have performed an in vitro gel mobility shift analysis by using purified proteins from Escherichia coli expressed genes. We have studied the different mobility of the complex formed with Mig1 and a DNA fragment carrying the Mig1 binding site of the SUC2 promoter and the complex formed with Mig1, Hxk2 and the same DNA fragment. The different mobility of these complexes suggests that Hxk2 interacts directly with Mig1. An additional control experiment using an anti-Hxk2 antibody confirms this finding since this antibody inhibits the formation of the Hxk2–Mig1 complex .
These results allow us to conclude that the Hxk2 can bind directly to a complex formed by the transcriptional repressor Mig1 and the MIG1 element of the SUC2 promoter.
The nuclear localization of Hxk2 is regulated by glucose and is Mig1-dependent
To analyse the factors that regulate the subcellular localization of Hxk2, green fluorescent protein (GFP) was fused to the C-terminal end of the Hxk2 protein. The fusion protein was expressed in several yeast strains from the HXK2 promoter. Fluorescent microscopy experiments demonstrated that in glucose-growing cells, Hxk2 is predominantly localized in the cytoplasm although a small fraction is located in the nucleus in good agreement with previous results [16,21]. In cells grown overnight in 3% ethanol, conditions under which repression does not occur, Hxk2 is found only in the cytoplasm, apparently excluded from the nucleus. This result suggests that the cell modulates the regulatory function of Hxk2 by controlling the transport of the protein between the nucleus and the cytoplasm.
To determine whether Mig1 participates in the regulation of Hxk2 localization, we examined it in a mig1 mutant strain. In glucose-grown cells, Hxk2 was found only in the cytoplasm. This suggests that either Mig1 retains Hxk2 in the nucleus or Mig1 is necessary for the translocation of the Hxk2 protein to the nucleus. Since it is known that glucose causes a rapid translocation of Mig1 to the nucleus and that the removal of the sugar results in cytoplasm localization , it could be thought that the amount of Hxk2 located in the nucleus under glucose-grown conditions could be dependent on the Mig1 level. To check this, we increased the amount of Mig1 in the cell by introducing a high copy number plasmid containing the MIG1 gene under control of its own promoter. Under this condition we examined the Hxk2 localization during glucose growth. Interestingly, Hxk2 was intensely enriched in the nucleus .
These results suggest that the cell modulates the Hxk2 regulatory function by controlling the amount of the protein localized in the nucleus and that Hxk2 is sequestered in the nucleus through interaction with Mig1. However, these experiments do not clarify if the Mig1–Hxk2 complex is a carrier necessary for Mig1 translocation to the nucleus.
The subcellular localization of Mig1 is regulated by Snf1 and is not Hxk2-dependent
In order to investigate this possibility GFP was fused to the C-terminal end of the Mig1 protein and the subcellular localization of Mig1 was determined under different metabolic conditions and in several mutant strains. As has been reported previously, Mig1 has a nuclear location in cells growing in glucose medium and a cytosolic localization in cells growing in ethanol medium . However, in an hxk2 mutant cell, Mig1 always has a cytosolic localization. This result could mean that the formation of the heterodimeric complex Hxk2–Mig1 is necessary to carry on the Mig1 translocation to the nucleus.
However, on the basis of other previously reported results [23–25], it is necessary to take into account that the activity of the Snf1 kinase complex is regulated by the Rgt1–Glc7 phosphatase complex and by Hxk2. In the absence of any one of these proteins a conformational change is produced in the Snf1 complex that activates its protein kinase function. Thus, when Snf1 is activated by mutations in the HXK2 gene, it was expected that the Mig1 protein would be phosphorylated and translocated to the cytosol. Interestingly, in a snf1/hxk2 double mutant, the Mig1 protein is always located in the nucleus, indicating that the presence of the Hxk2 protein is not necessary for a nuclear localization of Mig1.
Our results are in agreement with the model shown in Figure 3, explaining the involvement of Hxk2 in the glucose signalling pathway. At high glucose concentrations, Mig1 is dephosphorylated by the Glc7 phosphatase complex and imported to the nucleus. At the same time Hxk2 enters the nucleus and there it is sequestered by interacting with Mig1. The amount of Hxk2 present in the nucleus is limited by the Mig1 levels of the cell. Once access to the nucleus has been gained the heterodimeric Hxk2–Mig1 complex binds to the Mig1 target gene promoters and recruits the general co-repressor complex Cyc8–Tup1 to repress the transcription of genes not required for growth in glucose.
Involvement of Mig1 in the subcellular localization of Hxk2
Upon glucose depletion, Snf1 protein kinase is activated and phosphorylates Mig1. Phosphorylation induces Mig1 nuclear export, sequestering the protein in the cytoplasm together with Hxk2. Genes needed for growth on other carbon sources different from glucose are thus derepressed. An HXK2 mutation also activates Snf1 complex, and therefore Mig1 is phosphorylated and translocated to the cytosol, signalling derepression of Mig1 target genes. Thus in the absence of Hxk2, the Snf1 protein kinase complex is active even in the presence of high levels of glucose in the medium. Our results suggest that one major function of Hxk2 may be to inhibit Snf1 protein kinase function by blocking Mig1 phosphorylation at the nuclear level.
Nutrient Sensing through the Plasma Membrane of Eukaryotic Cells: Focused Meeting held at the Royal Agricultural College, Cirencester, 25–29 September 2004. Organized by S. Shirazi-Beechey (Liverpool, U.K.), J. Thevelein and F. Stolz (Leuven, Belgium). Edited by S. Shirazi-Beechey. Sponsored by Flanders Interuniversity Institute for Biotechnology and Nestlé U.K. Ltd.
We are very grateful to J.M. Gancedo and C. Gancedo for fruitful discussions. This work was supported by Grant PB97-1213-C0202 from the Dirección General de Enseñanza Superior e Investigación Científica (DGESIC, Spain).