• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Feb 10, 2004; 101(6): 1572–1577.
Published online Jan 30, 2004. doi:  10.1073/pnas.0305901101
PMCID: PMC341776
Cell Biology

Glucose sensing and signaling in Saccharomyces cerevisiae through the Rgt2 glucose sensor and casein kinase I


The yeast Saccharomyces cerevisiae senses glucose through two transmembrane glucose sensors, Snf3 and Rgt2. Extracellular glucose causes these sensors to generate an intracellular signal that induces expression of HXT genes encoding glucose transporters by inhibiting the function of Rgt1, a transcriptional repressor of HXT genes. We present the following evidence that suggests that the glucose sensors are coupled to the membrane-associated protein kinase casein kinase I (Yck1). (i) Overexpression of Yck1 leads to constitutive HXT1 expression; (ii) Yck1 (or its paralogue Yck2) is required for glucose induction of HXT1 expression; (iii) Yck1 interacts with the Rgt2 glucose sensor; and (iv) attaching the C-terminal cytoplasmic tail of Rgt2 to Yck1 results in a constitutive glucose signal. The likely targets of Yck1 in this signal transduction pathway are Mth1 and Std1, which bind to and regulate function of the Rgt1 transcription factor and bind to the C-terminal cytoplasmic domain of glucose sensors. Potential casein kinase I phosphorylation sites in Mth1 and Std1 are required for normal glucose regulation of HXT1 expression, and Yck1 catalyzes phosphorylation of Mth1 and Std1 in vitro. These results support a model of glucose signaling in which glucose binding to the glucose sensors causes them to activate Yck1 in the cell membrane, which then phosphorylates Mth1 and Std1 bound to the cytoplasmic face of the glucose sensors, triggering their degradation and leading to the derepression of HXT gene expression. Our results add nutrient sensing to the growing list of processes in which casein kinase I is involved.

Nutrient sensing is crucial for all cells. How cells sense glucose, the most abundant monosaccharide on earth and the primary carbon and energy source for most cells, is of great interest because defects in glucose sensing can have serious consequences for organisms. One way the yeast Saccharomyces cerevisiae senses glucose is through a novel receptor-mediated signaling pathway that begins with two glucose sensors in the cell membrane. Extracellular glucose causes these sensors to generate an intracellular signal that induces expression of several HXT genes encoding hexose transporters (1). The glucose signal induces this expression by influencing the function of the Rgt1 transcriptional repressor. In the absence of glucose, Rgt1 is functional and binds to promoters of HXT genes and represses their expression (2, 3). Glucose derepresses HXT gene expression by inhibiting Rgt1 repressor function. This process requires the SCFGrr1 ubiquitin-protein ligase (3, 4), suggesting that it involves protein ubiquitination and degradation. Repression of transcription by Rgt1 also requires the paralogous proteins Mth1 and Std1 (3, 5). Because Mth1 and Std1 interact with both the Rgt1 repressor (5, 6) and the Snf3 and Rgt2 glucose sensors (7, 8), they are good candidates for the components of the signal transduction pathway that are inhibited by the glucose signal. Indeed, it was recently reported that Mth1 is degraded when glucose is added to cells by SCFGrr1 (3).

It is not known how Snf3 and Rgt2 generate a signal in response to glucose or how that signal is transduced to Rgt1, Mth1, and Std1. Snf3 and Rgt2 are probably transmembrane receptors for extracellular glucose, because they are similar to hexose transporters but are unable to transport glucose (9). They differ from bona fide glucose transporters in possessing unusually long C-terminal tails that are predicted to reside in the cytoplasm (10). These tails are sufficient for glucose signaling, because attaching them to a glucose transporter confers on it glucose signaling ability (9), and anchoring the tails by themselves in the cell membrane leads to a constitutive glucose signal (11).

Our current view of this signaling pathway is that the binding of glucose to the glucose sensors induces a conformational change in them that triggers the degradation of Mth1 and Std1. We present evidence that suggests that this process occurs by activation of casein kinase I (Yck), a protein kinase involved in diverse cellular processes that resides in the cell membrane. Our results suggest a model in which glucose-activated Yck catalyzes phosphorylation of Mth1 and Std1 bound to the tails of the glucose sensors, triggering their recognition by the SCFGrr1 ubiquitin-protein ligase, which marks them for degradation by the proteasome and leads to inactivation of the Rgt1 repressor and derepression of HXT gene expression.

Materials and Methods

Yeast Strains and Growth. S. cerevisiae cells were cultivated as described (12). Yeast strains used are YM4128 (MATα his3 leu2 lys2 trp1 tyr1 ura3), YM6107 (rgt2Δsnf3Δ) (9), YM6212 (MATa his3 leu2 met15 ura3 mth1Δ::KanMX2 std1Δ::KanMX2), LRB341 (YCK1 YCK2), LRB264 (yck1Δ), LRB343 (yck2Δ), LRB362 (yck1Δyck2ts) (13), L40 (14), and A2085 (15).

Plasmids. DNA fragments used in plasmid constructions were amplified by PCR with the Expand High Fidelity PCR System (Roche Molecular Biochemicals) and cloned by gap repair in yeast (16). Plasmids containing the HXT1 promoter fused to lacZ are pBM2636 and pBM3212 (17). All constructs in pBM2974 (9) were expressed from the ADH1 promoter (PADH1). Plasmid constructions are as follows: pBM4523, URA3 in pBM2636 (18) was replaced by KanMX from pUG6 (19); pBM4524, GFP from pQBI63 (Wako Pure Chemical, Osaka) and the C-terminal tail of Rgt2 (amino acids 546–end) were fused and cloned into pBM2974; pBM4525, the lipidation signal from Ras2 (GSGGCCIIS) was attached to Rgt2 just before its stop codon in pBM4524; pBM4526, Val-404 of Rgt2 in pBM3272 (9) was changed to Ile; pBM4527, hemagglutinin (HA) epitope (YPYDVPDYASL) was attached to the C terminus of Hxt1 in pBM3362 (9); pBM4528, the HA epitope was attached to the C terminus of Rgt2 in pBM3333 (9); pBM4529, the transmembrane domain of Hxt1 (amino acids 1–546) and the Rgt2 tail (amino acids 579–end) were fused and cloned into pBM2974; pBM4530, Phe-371 of Hxt1 on pBM4529 was changed to Ile; pBM4531, Rgt2 (amino acids 1–578) and Hxt1 (amino acids 547–end) were fused and cloned into pBM2974; pBM4532, Val-404 of Rgt2 on pBM4531 was changed to Ile; pBM4533 and pBM4534, PADH1 of pBM4530 or pBM4531 was replaced by the RGT2 promoter (from -795 nt to -1 nt before ATG), then the PRGT2-Hxt1-Rgt2tail or PRGT2-Rgt2Δtail fragment was cloned into pRS316 (20); pBM4535, YCK1 (from ATG to the stop codon) was cloned into pBM2974; pBM4536, 7× His Protein A from pYM10 (21) was fused at the C terminus of Yck1 just before the palmitoylation signal (Cys–Cys) on pBM4535; pBM4537, Lys-98 of Yck1 on pBM4536 was changed to Arg; pBM4538, Yck1 (amino acids 1–536) and Rgt2tail (amino acids 584 to the stop codon) were fused and cloned into pBM2974; pBM4539, Asp-211 of Yck1 on pBM4538 was changed to Ala; pBM4540, STD1 (-830 nt from ATG to +156 nt after the stop codon) was cloned into pRS313 (20); pBM4541, Ser-129, -132, -135, -136, -137, -140, -141, -144, and -147 of Std1 on pBM4540 were changed to Ala; pBM4542, MTH1 (-850 nt from ATG to +130 nt after the stop codon) was cloned into pRS313; pBM4561, Ser-118, -121, -125, -126, -129, -130, -133, and -136 of Mth1 on pBM4542 were changed into Thr; pBM4544 and pBM4545, 9xmyc epitope (EQKLISEED) from pYM6 (21) was attached to the C terminus of Std1 on pBM4540 or pBM4541; pBM4560 and pBM4562, 9xmyc was attached to the C terminus of Mth1 on pBM4542 or pBM4561; pBM4546 and pBM4547, RGT2 or HXT1 (ATG through the stop codons) was cloned into p415-Cub-PLV (PCYC1-Cub-PLV-TCYC1, cen-ARS, LEU2) (gift from I. Stagljar), so that Cub-PLV was fused to the C terminus of Rgt2; pBM4548, Rgt2 “signaling box” (amino acids 665–696) of pBM4546 was replaced by HA epitope; pBM4549, pBM4563, pBM4550, pBM4551, pBM4552, or pBM4553, MTH1, MTH18ST, STD1, STD19SA, YCK1, or yck1K98R (from ATG to the stop codon) was cloned into pNubG-HA-X (PCYC1-NubG-HA-TCYC1, cen-ARS, TRP1) (gift from I. Stagljar) so that NubG was fused to the N-terminal of each protein; pBM4554 (pGEX-STD1), pBM4555 (pGEX-STD19SA), or pBM4556 (pGEX-MTH1), STD1, STD19SA, or MTH1 (from ATG to the stop codon) was cloned into pGEX-5X-2 (Amersham Pharmacia).

β-Galactosidase (β-Gal) Assays. Yeast cells harboring the indicated plasmids were cultivated overnight in minimal medium containing 2% galactose. The cells were transferred to the same fresh medium containing 2% galactose or 4% glucose. After 5 h of cultivation, β-gal activity of the cell extract was measured with the Yeast β-Galactosidase Assay Kit (Pierce) according to the supplier's protocol.

Detection of Mth1 and Std1. Yeast cells harboring plasmids were cultivated overnight in rich medium containing 2% galactose. The cells were transferred to the same fresh medium containing 2% galactose or 4% glucose. At the indicated time, cell extracts were prepared as described (22). Western blot analysis was performed with mouse anti-myc antibody 9E10 (Berkeley Antibody, Richmond, CA) and horseradish peroxidase-conjugated rabbit anti-mouse antibodies (Santa Cruz Biotechnology) and visualized with the Supersignal West Pico chemiluminescent substrate (Pierce).

Split-Ubiquitin Method. L40 cells harboring Cub-PLV fusion plasmid and NubG plasmid were cultivated overnight in minimal medium containing 2% galactose or 4% glucose, and β-gal activity of the cells was determined.

Protein Kinase Assays. Each GST-fusion protein expressed in Escherichia coli was used still attached to beads and affinity-purified with glutathione-Sepharose 4B (Amersham Pharmacia) as described in the supplier's protocol. We purified 7× His-tagged Yck1 and Yck1K98R from yeast A2085 cells harboring pBM4536 or pBM4537 cultivated in yeast extract/peptone/dextrose after the cells were broken with glass beads in buffer A [50 mM Tris·Cl, pH 6.8/100 mM NaCl/0.5% Triton X-100/1 mM DTT/1 mM PMSF/Complete protease inhibitor mixture (Roche Molecular Biochemicals)]. Cell debris was removed by centrifugation, and 7× His-tagged proteins were purified with Ni-nitrilotriacetic acid agarose (Qiagen, Valencia, CA) and then eluted with buffer A containing 250 mM imidazole.

Soluble and membrane protein fractions were prepared from cells cultivated in yeast extract/peptone/dextrose and broken by glass beads in buffer A without Triton X-100. Cell debris was removed by brief centrifugation, and then the cell extract was centrifuged at 10,000 × g for 20 min. The supernatant was taken for the soluble protein fraction; the membrane pellet was resuspended in buffer A. Equal amounts of protein were used in the protein kinase assays.

Protein kinase assays were started by mixing 20 μl of purified GST-fusion protein attached to beads, 0.5 μCi of [γ-32P]ATP (1 Ci = 37 GBq), 100 μM ATP, 10 mM MgCl2, and protein kinase sample in 50 μl of buffer A. After 30 min incubation at 28°C, beads were washed with buffer A containing 0.5 M NaCl and then washed with buffer A alone. Proteins bound to the beads (GST-fusion proteins) were eluted by boiling and separated by SDS/PAGE. After Coomassie brilliant blue staining, phosphorylated proteins were visualized with a PhosphorImager (Molecular Dynamics).


The Rgt2 Transmembrane Domain Generates the Glucose Signal. Previous reports suggested that the C-terminal tails of the glucose sensors are necessary and sufficient for glucose signaling (9, 11). Indeed, when RGT2 is expressed at its normal low level, its tail is required for signaling (Table 1, line 5), and Hxt1 carrying the Rgt2 tail causes constitutive HXT1 expression if it is expressed at high levels (Table 1, line 8) but not if it is expressed at normal levels (Table 1, line 3). We were thus surprised to discover that Rgt2 without its tail can be a functional glucose sensor if it is overexpressed (Table 1, line 11). This seems to be authentic glucose signaling because Hxt1 does not have this ability (Table 1, line 7), and this signaling is abolished by a rgt2 mutation (V404I; Table 1, line 12) that affects an amino acid critical for signaling (Table 1, line 4). [We believe this residue is required for glucose binding, because the orthologous residue of Hxt1 (F371) is necessary for its ability to transport glucose (data not shown)]. These results lead us to conclude that the sensor tails are not required for its ability to signal. We imagine that the role of the sensor tails is to recruit Mth1 and Std1 (7, 8) to the vicinity of a protein that modulates their activity and that is activated by the transmembrane domain of the glucose-bound sensors. We suggest that enough Mth1 and Std1 happen to reside near the (overexpressed) tailless sensor for them to be inhibited by the unknown protein that is coupled to the sensors.

Table 1.
HXT1 expression in strains expressing chimeric sensors

We suspected the protein coupled to the glucose sensors is attached to the membrane, because anchoring the tail of the Rgt2 glucose sensor to the membrane, either by giving it a farnesylation signal (Table 1, line 14) or by attaching it to Hxt1 (Table 1, line 8), generates a constitutive glucose signal [but only if it is overexpressed (Table 1, line 3); the higher levels of HXT1 expression caused by glucose in cells with the Hxt1 carrying the Rgt2 tail or the membrane-tethered Rgt2 tail is probably due to activation by glucose of the ADH1 promoter (23) that drives their expression]. That this signal is constitutive is supported by the fact that the F371I mutation in HXT1 that likely impairs its ability to bind glucose does not reduce signaling (Table 1, line 9). We suggest that high levels of the glucose sensor tail attached to the membrane brings enough Mth1 and Std1 near to the unknown protein for it to trigger the event that inhibits their function. These results suggested to us that the transmembrane domain of the glucose sensors activate a membrane protein that initiates inhibition of Mth1 and Std1.

Casein Kinase I Is Required for Glucose Signaling by Rgt2. Glucose stimulates mth1 degradation, a reaction that requires the SCFGrr1 ubiquitin protein ligase (3). Because recruitment of substrates by SCFGrr1 (e.g., Cln2) requires them to be phosphorylated (24), we further surmised that the unknown protein is a protein kinase that catalyzes phosphorylation of Mth1 and Std1. Accordingly, we screened for protein kinases that induce HXT1 expression when overexpressed by using an array of nearly all protein kinase-encoding genes fused to the GAL1 promoter (25). Because HXT1 is normally not expressed in cells growing on galactose, each protein kinase could be expressed at high levels on media containing galactose and tested for its ability to induce expression of a HXT1-lacZ reporter gene. Thirty strains, each overexpressing a different protein kinase, exhibit detectable HXT1 expression when growing on galactose (Fig. 1A). The 13 protein kinases that induce the strongest HXT1 expression are the best candidates for the component of the signal transduction pathway we are seeking, but none of the 12 nonessential protein kinases in this group are required for HXT1 expression (data not shown), suggesting that the protein kinase we are looking for is either essential or redundant.

Fig. 1.
Casein kinase I (Yck) is involved in the glucose induction pathway. (A) Screening protein kinases for a role in the glucose induction pathway. After being mated with YM4218 harboring pHXT1-lacZ (pBM4523), 119 yeast strains expressing GST-protein kinases ...

This combination of characteristics left casein kinase I (Yck1) (Fig. 1 A, lane 6, row E), which is essential for cell viability and is encoded by two genes (YCK1 and YCK2) (26), as the only candidate. Casein kinase I is an attractive candidate, because it is palmitoylated and therefore associates with the cell membrane (27) and because it is also required for the expression of a gene in Kluyveromyces lactis encoding a low-affinity glucose transporter (28). We found that HXT1 expression in a yck1Δyck2ts strain is not inducible by glucose, even in cells grown at a temperature semipermissive for yck2ts (Fig. 1B). In addition, glucose-triggered degradation of Mth1 and Std1 was not observed in the yck1Δyck2ts strain (Fig. 1C). Thus, membrane-bound casein kinase I seems to be a component of the glucose induction pathway upstream of Mth1 and Std1.

Casein Kinase I Interacts with Rgt2. If Yck is activated by the glucose sensors, it should bind to them. To test for interaction between Rgt2 and Yck1, we used the split-ubiquitin system (14) because it is compatible with membrane proteins. Rgt2 appears to interact with catalytically inactive Yck1 [because of a mutation (K98>R) in the ATP-binding site of Yck1] but not with active Yck1 or Hxt1 (Table 2). Thus, Yck1 appears to interact with Rgt2, and the catalytic activity of Yck1 inhibits this interaction.

Table 2.
Split-ubiquitin system interaction between RGT2 and YCK1

A Yck1-Rgt2tail Chimera Can Mimic the Glucose Signal. Tethering the (overexpressed) C-terminal tails of the glucose sensors by themselves to the cell membrane causes constitutive expression of HXT genes (ref. 11 and Table 1, line 14). We speculated that they bring Mth1 and Std1 near Yck that happens to be in their vicinity in the plasma membrane. Overexpressing the Rgt2 tail in the cytoplasm is unable to induce HXT1 expression (Table 1, line 13), probably because Yck is not in the cytoplasm. Likewise, overexpression of Yck1 in the cytoplasm (achieved by deleting the two Cys residues at its C terminus that are required for its palmitoylation) does not lead to induction of HXT1 expression (Table 1, line 17), probably because it has no means of recruiting Mth1 and Std1. However, attaching the Rgt2 tail to cytoplasmic Yck1 causes constitutive HXT1 expression (Table 1, line 15), presumably because it is able to recruit its substrates Mth1 and Std1. This HXT1 expression depends on the catalytic activity of Yck1 (Table 1, line 16).

Potential Casein Kinase I Phosphorylation Sites in Mth1 and Std1 Are Essential for Their Degradation and Derepression of HXT1. If Yck1 is indeed involved in the glucose signaling mechanism, its likely substrates are Mth1 and Std1. Phosphorylation of these proteins could target them to SCFGrr1 for their ubiquitination and subsequent degradation, as is the case for other targets of SCFGrr1 (24). To search for potential casein kinase I phosphorylation sites in Mth1 and Std1, we aligned them with their orthologues from other Saccharomyces yeast species (29) (Fig. 2A). There is a conserved cluster of serine residues in the middle of Mth1 and Std1 (under the solid line in Fig. 2 A) that matches the consensus target sequence for casein kinase I (SXXS, bold letters in Fig. 2 A) (30). We changed these serines in the conserved cluster and measured the ability of the altered proteins to support repression of HXT1 expression. Nine serines in Std1 were converted to alanines (Std19SA), and eight serines in Mth1 were converted to threonines (Mth18ST) [because their conversion to alanine destroyed function of Mth1 (data not shown)]. Native Mth1 and Std1 mediate 7- to 10-fold repression of HXT1 expression in the presence of galactose; addition of glucose to cells relieves this repression (Fig. 2B). Mth18ST and Std19SA missing their potential casein kinase I phosphorylation sites are functional, given that they promote normal repression of HXT1 in the presence of galactose, but this repression is not relieved by glucose (Fig. 2B). Furthermore, glucose does not induce degradation of Mth18ST and Std19SA (Fig. 2C, lanes 5, 6, 11, and 12) as it does wild-type Mth1 and Std1 (Fig. 2C, lanes 2, 3, 8, and 9). These results are consistent with the idea that glucose stimulates casein kinase I to catalyze phosphorylation of Mth1 and Std1, which leads to their degradation.

Fig. 2.
(A) clustalw alignment of Mth1 and Std1 orthologues from S. cerevisiae (Sce), Schwanniomyces castellii (Sca), and Saccharomyces kluyveri (Skl)(S. cerevisiae Mth1 amino acids 104–150 and its orthologous regions are shown) (29). The underlined regions ...

The defect in Mth18ST and Std19SA regulation is not caused by their inability to bind to the Rgt2 tail, because they interact normally with Rgt2, as detected by the split-ubiquitin method (Table 3). This interaction depends on the Rgt2 signaling box. The degree of this interaction is reduced in the presence of glucose, probably because of glucose-induced degradation of Mth1 and Std1. Indeed, Mth18ST and Std19SA exhibit increased interaction with Rgt2 in glucose-grown cells, probably because they are not degraded.

Table 3.
Split-ubiquitin system interaction between RGT2 and MTH1 or STD1

Casein Kinase I Phosphorylates Mth1 and Std1 in Vitro. To test whether casein kinase I can phosphorylate Mth1 and Std1, we used Mth1 and Std1 (purified from E. coli as GST-fusion proteins) as substrates in a protein phosphorylation assay of Yck1 (fused to 7× His-Protein A and affinity-purified from yeast cells). Yck1 catalyzed phosphorylation of Std1 (Fig. 3A, lane 3) and Mth1 (Fig. 3B, lane 3). Inactivation of Yck1 catalytic activity (Yck1-K98R) or removal of its phosphorylation sites from Std1 (Std19SA) prevented phosphorylation of Std1 (Fig. 3A, lane 4, and Fig. 3B, lane 2), suggesting that the cluster of serines is the major site of phosphorylation of Std1 by Yck1. Changing the equivalent serines of Mth1 (8S>T) did not significantly reduce its phosphorylation in vitro catalyzed by Yck1 (data not shown), probably because Mth1 has many more potential Yck1 phosphorylation sites than does Std1. However, the additional serines in Mth1 must not be in vivo substrates of Yck1, because the eight serines altered in Mth18ST are clearly necessary for glucose inhibition of Mth1 function (Fig. 2 B and C, compare Mth1 to Mth18ST). Because inactivation of Yck1-K98R did not reduce Mth1 phosphorylation (perhaps another kinase(s) copurified with Yck1 or contaminated the preparation; data not shown), we tested the Mth1 phosphorylation activity of crude cell lysate fractions. We found that the membrane fraction gave maximal Mth1 phosphorylation activity (Fig. 3C, lanes 3 and 4) and that this activity is substantially reduced in the extract of a yckts strain (Fig. 3C, lane 4). These results suggest that Mth1 and Std1 are substrates of Yck1.

Fig. 3.
(A and B) Protein kinase activity of affinity-purified Yck1-7× His-ProtA (Yck1-WT) or Yck1K98R-7× His-ProtA (Yck1-K98R) toward GST or GSTStd1 was measured. Approximately equal amounts of Yck1 and Yck1-K98R were used in these experiments ...


Yck1 As a Component of the Glucose Signaling Pathway. We have described several experimental results that suggest that the Rgt2 glucose sensor (and presumably also the Snf3 glucose sensor) signals glucose availability through casein kinase I (Yck1 and Yck2). First, overexpression of Yck1 generates a constitutive glucose signal that induces HXT1 expression (Fig. 1 A). Second, YCK1 (or its orthologue YCK2) is required for glucose induction of HXT expression (Fig. 1B) and the glucose-triggered degradation of Mth1 and Std1 (Fig. 1C). Third, Yck1 interacts with Rgt2 in vivo (Table 2). Fourth, a Yck1-Rgt2tail chimera expressed in the cytoplasm mimics the glucose signal (Table 1, line 15). Fifth, serine residues in Mth1 and Std1 that match the consensus sequence for phosphorylation by casein kinase I (SXXS) are required for normal regulation of HXT1 expression by glucose (Fig. 2B) and for degradation of Mth1 and Std1 (Fig. 2C). Finally, Yck1 phosphorylates Mth1 and Std1 in vitro (Fig. 3). These results, together with the knowledge that Mth1 and Std1 bind to the C-terminal cytoplasmic tails of the glucose sensors (7, 8), lead us to propose that glucose binding to Rgt2 (and Snf3) activates Yck1 (and Yck2), which then catalyzes phosphorylation of Mth1 and Std1 bound to the tails of the glucose sensors (Fig. 4). Because glucose-induced degradation of Mth1 and Std1 requires the SCFGrr1 ubiquitin-protein ligase (ref. 3 and our unpublished observation), their phosphorylation would allow them to be recognized by SCFGrr1 for ubiquitination, which targets them for degradation. Indeed, all known proteins that are targets of F-box proteins, like Grr1, must be phosphorylated to interact with this ubiquitin-protein ligase (31).

Fig. 4.
Current model for transduction of the glucose signal from Snf3 and Rgt2 to Mth1 and Std1. Binding of glucose (Glu) to the transmembrane domain of the glucose sensors (Rgt2/Snf3) is postulated to change their conformation and activate membrane-bound casein ...

One well known function of casein kinase I is that it regulates the endocytosis of membrane proteins (13, 32). Although we cannot rule out the possibility that glucose-induced endocytosis of the glucose sensors stimulated by Yck triggers the degradation of Mth1 and Std1, we do not favor this possibility, because internalization of Rgt2 (visualized by using an Rgt2-GFP chimeric protein) is not stimulated by glucose (data not shown), and glucose signaling is normal in mutants that are defective in endocytosis, such as end3 or end4 (our unpublished results).

Signal Generation Mechanism of the Glucose Sensors. The fact that the transmembrane domain of Rgt2 is sufficient for glucose signaling (provided it is expressed at a high level; Table 1, line 11) suggests that the C-terminal tail is not necessary for, but rather enhances, signaling. We believe it does this by bringing Mth1 and Std1 close to Yck. In this view, high levels of the tailless sensor can generate a glucose signal by activating Yck1, which we imagine phosphorylates any Mth1 and Std1 that happens to be in the vicinity. The cytoplasmic Yck1-Rgt2tail chimera mimics a constitutive signal (provided it is expressed at a high level), probably because the tail domain attracts Mth1 and Std1 to the vicinity of Yck1.

Binding of glucose to the glucose sensors could stimulate Yck1 to phosphorylate Mth1 and Std1 simply by stimulating recruitment of Yck1, thereby bringing it into the vicinity of its substrates that are bound to the sensor tail. Alternatively, the glucose-bound sensors could play an active role in stimulating the catalytic activity of Yck1. We currently favor the latter possibility, because glucose is not required for interaction of Yck1 with Rgt2 (Table 2). It is also possible that recruitment of Mth1 and Std1 to the tails of the glucose sensors is stimulated by glucose. We do not favor this idea, because glucose generates a signal through a tailless glucose sensor (Table 1, line 11), suggesting that some event other than recruitment of Mth1 and Std1 is activated by glucose.

Std1 seems to be less sensitive to degradation than Mth1. Mth1 cannot be detected 30 min after addition of glucose to cells (Fig. 2C, lane 2), but Std1 protein is still observed after 1 h when the blot is overexposed to film (data not shown). Indeed, Mth1, but not Std1, degradation was observed in a recent study (3). We believe this difference is due to the different regulation of MTH1 and STD1 expression. STD1 expression is induced by glucose [via the Rgt2/Snf3-Rgt1 pathway (33)], which would be expected to counteract Std1 degradation. However, glucose increases STD1 expression at the same time that it induces its degradation. MTH1 expression, on the other hand, is repressed by glucose [via the Snf1-Mig1 pathway, (33)], which would be expected to reinforce Mth1 degradation, but glucose reduces MTH1 expression at the same time that it stimulates its degradation. The relative resistance of Std1 to degradation may account for the different roles in regulation of HXT expression that are apparent for Mth1 and Std1 (8).

Do Other Organisms Have Glucose Sensors Similar to Rgt2 and Snf3? We previously thought that the glucose sensors are yeast-specific, because no glucose transporter-like protein with a long C-terminal tail is apparent in the genome sequences of other organisms. However, our finding that the glucose sensor tail is not required for its function raises the possibility that similar glucose sensors exist in other organisms. Proteins similar to the yeast glucose sensors, but without the long C-terminal tails, have been identified in some fungal species [e.g., Neurospora crassa Rco3 (34), Uromyces fabae Hxt1 (35), and Amanita muscaria MstI (GenBank accession no. CAB06078)]. The mammalian GLUT2 glucose transporter seems to be involved in transducing a glucose signal that affects gene expression, a function in which its internal cytoplasmic loop has been implicated (36, 37). Also, the mammalian GLUT1 glucose transporter was reported to activate ERK protein kinase by means of its short C-terminal cytoplasmic domain (38).

Casein Kinase I Couples Extracellular Signals to Degradation of Transcriptional Regulators. Casein kinase I is in a family of highly conserved, ubiquitously expressed protein kinases found in all eukaryotic organisms. It has recently become apparent that it regulates diverse processes, including Wnt signaling, circadian rhythms, nuclear import of proteins, and Alzheimer's disease progression (reviewed in ref. 39). Our results add nutrient sensing to the list of processes in which casein kinase I is involved. It seems that casein kinase I regulates Wnt signaling like it regulates glucose signaling in yeast: by triggering the degradation of a protein that regulates transcription (β-catenin) in response to a signal generated by a transmembrane receptor (frizzled) via an E3 ubiquitin ligase that includes an F-box protein (SCFβ-TRCP) (40). We expect that casein kinase I will turn up in other signal transduction pathways.


We thank Dr. Igor Stagljar (University of Zürich, Irchel, Switzerland) for providing reagents for the split-ubiquitin system and advice on their use, Heng Zhu and Mike Snyder (Yale University, New Haven, CT) for generously providing their array of yeast protein kinases fused to GST, and our colleague John Majors for suggestions on the manuscript. This work was supported by National Institutes of Health Grant R01-32540 and funds provided to support the James S. McDonnell Laboratory of Molecular Genetics.


This paper was submitted directly (Track II) to the PNAS office.

Abbreviation: HA, hemagglutinin.


1. Ozcan, S. & Johnston, M. (1999) Microbiol. Mol. Biol. Rev. 63, 554-569. [PMC free article] [PubMed]
2. Kim, J. H., Polish, J. & Johnston, M. (2003) Mol. Cell. Biol. 23, 5208-5216. [PMC free article] [PubMed]
3. Flick, K. M., Spielewoy, N., Kalashnikova, T. I., Guaderrama, M., Zhu, Q., Chang, H. C. & Wittenberg, C. (2003) Mol. Biol. Cell 14, 3230-3241. [PMC free article] [PubMed]
4. Li, F. N. & Johnston, M. (1997) EMBO J. 16, 5629-5638. [PMC free article] [PubMed]
5. Lakshmanan, J., Mosley, A. L. & Ozcan, S. (2003) Curr. Genet. 44, 19-25. [PubMed]
6. Tomas-Cobos, L. & Sanz, P. (2002) Biochem. J. 368, 657-663. [PMC free article] [PubMed]
7. Lafuente, M. J., Gancedo, C., Jauniaux, J. C. & Gancedo, J. M. (2000) Mol. Microbiol. 35, 161-172. [PubMed]
8. Schmidt, M. C., McCartney, R. R., Zhang, X., Tillman, T. S., Solimeo, H., Wolfl, S., Almonte, C. & Watkins, S. C. (1999) Mol. Cell. Biol. 19, 4561-4571. [PMC free article] [PubMed]
9. Ozcan, S., Dover, J. & Johnston, M. (1998) EMBO J. 17, 2566-2573. [PMC free article] [PubMed]
10. Marshall-Carlson, L., Celenza, J. L., Laurent, B. C. & Carlson, M. (1990) Mol. Cell. Biol. 10, 1105-1115. [PMC free article] [PubMed]
11. Dlugai, S., Hippler, S., Wieczorke, R. & Boles, E. (2001) FEBS Lett. 505, 389-392. [PubMed]
12. Adams, A., Gottschling, D. E., Kaiser, C. A. & Stearns, T. (1997) Methods in Yeast Genetics (Cold Spring Harbor Lab. Press, New York).
13. Panek, H. R., Stepp, J. D., Engle, H. M., Marks, K. M., Tan, P. K., Lemmon, S. K. & Robinson, L. C. (1997) EMBO J. 16, 4194-4204. [PMC free article] [PubMed]
14. Stagljar, I., Korostensky, C., Johnsson, N. & te Heesen, S. (1998) Proc. Natl. Acad. Sci. USA 95, 5187-5192. [PMC free article] [PubMed]
15. Moriya, H., Shimizu-Yoshida, Y., Omori, A., Iwashita, S., Katoh, M. & Sakai, A. (2001) Genes Dev. 15, 1217-1228. [PMC free article] [PubMed]
16. Oldenburg, K. R., Vo, K. T., Michaelis, S. & Paddon, C. (1997) Nucleic Acids Res. 25, 451-452. [PMC free article] [PubMed]
17. Ozcan, S., Dover, J., Rosenwald, A. G., Wolfl, S. & Johnston, M. (1996) Proc. Natl. Acad. Sci. USA 93, 12428-12432. [PMC free article] [PubMed]
18. Ozcan, S. & Johnston, M. (1995) Mol. Cell. Biol. 15, 1564-1572. [PMC free article] [PubMed]
19. Guldener, U., Heck, S., Fielder, T., Beinhauer, J. & Hegemann, J. H. (1996) Nucleic Acids Res. 24, 2519-2524. [PMC free article] [PubMed]
20. Sikorski, R. S. & Hieter, P. (1989) Genetics 122, 19-27. [PMC free article] [PubMed]
21. Knop, M., Siegers, K., Pereira, G., Zachariae, W., Winsor, B., Nasmyth, K. & Schiebel, E. (1999) Yeast 15, 963-972. [PubMed]
22. Kushnirov, V. V. (2000) Yeast 16, 857-860. [PubMed]
23. Denis, C. L., Ferguson, J. & Young, E. T. (1983) J. Biol. Chem. 258, 1165-1171. [PubMed]
24. Hsiung, Y. G., Chang, H. C., Pellequer, J. L., La Valle, R., Lanker, S. & Wittenberg, C. (2001) Mol. Cell. Biol. 21, 2506-2520. [PMC free article] [PubMed]
25. Zhu, H., Klemic, J. F., Chang, S., Bertone, P., Casamayor, A., Klemic, K. G., Smith, D., Gerstein, M., Reed, M. A. & Snyder, M. (2000) Nat. Genet. 26, 283-289. [PubMed]
26. Robinson, L. C., Hubbard, E. J., Graves, P. R., DePaoli-Roach, A. A., Roach, P. J., Kung, C., Haas, D. W., Hagedorn, C. H., Goebl, M., Culbertson, M. R., et al. (1992) Proc. Natl. Acad. Sci. USA 89, 28-32. [PMC free article] [PubMed]
27. Babu, P., Bryan, J. D., Panek, H. R., Jordan, S. L., Forbrich, B. M., Kelley, S. C., Colvin, R. T. & Robinson, L. C. (2002) J. Cell Sci. 115, 4957-4968. [PubMed]
28. Blaisonneau, J., Fukuhara, H. & Wesolowski-Louvel, M. (1997) Mol. Gen. Genet. 253, 469-477. [PubMed]
29. Cliften, P., Sudarsanam, P., Desikan, A., Fulton, L., Fulton, B., Majors, J., Waterston, R., Cohen, B. A. & Johnston, M. (2003) Science 301, 71-76. [PubMed]
30. Kennelly, P. J. & Krebs, E. G. (1991) J. Biol. Chem. 266, 15555-15558. [PubMed]
31. Craig, K. L. & Tyers, M. (1999) Prog. Biophys. Mol. Biol. 72, 299-328. [PubMed]
32. Marchal, C., Dupre, S. & Urban-Grimal, D. (2002) J. Cell Sci. 115, 217-226. [PubMed]
33. Kaniak, A., Xue, Z., Macool, D., Kim, J. H. & Johnston, M. (2004) Eukaryotic Cell, in press.
34. Madi, L., McBride, S. A., Bailey, L. A. & Ebbole, D. J. (1997) Genetics 146, 499-508. [PMC free article] [PubMed]
35. Voegele, R. T., Struck, C., Hahn, M. & Mendgen, K. (2001) Proc. Natl. Acad. Sci. USA 98, 8133-8138. [PMC free article] [PubMed]
36. Guillemain, G., Munoz-Alonso, M. J., Cassany, A., Loizeau, M., Faussat, A. M., Burnol, A. F. & Leturque, A. (2002) Biochem. J. 364, 201-209. [PMC free article] [PubMed]
37. Guillemain, G., Loizeau, M., Pincon-Raymond, M., Girard, J. & Leturque, A. (2000) J. Cell Sci. 113, 841-847. [PubMed]
38. Bandyopadhyay, G., Sajan, M. P., Kanoh, Y., Standaert, M. L., Burke, T. R., Jr., Quon, M. J., Reed, B. C., Dikic, I., Noel, L. E., Newgard, C. B. & Farese, R. (2000) J. Biol. Chem. 275, 40817-40826. [PubMed]
39. Vielhaber, E. & Virshup, D. M. (2001) IUBMB Life 51, 73-78. [PubMed]
40. Polakis, P. (2002) Curr. Biol. 12, R499-R501. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Compound
    PubChem Compound links
  • Gene
    Gene links
  • GEO Profiles
    GEO Profiles
    Related GEO records
  • HomoloGene
    HomoloGene links
  • MedGen
    Related information in MedGen
  • Pathways + GO
    Pathways + GO
    Pathways, annotations and biological systems (BioSystems) that cite the current article.
  • Protein
    Published protein sequences
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem Substance links
  • Taxonomy
    Related taxonomy entry
  • Taxonomy Tree
    Taxonomy Tree