Protein phosphorylation is estimated to affect 30% of the proteome and is a major regulatory mechanism that controls many basic cellular processes. Until recently, our biochemical understanding of protein phosphorylation on a global scale has been extremely limited; only one half of the yeast kinases have known in vivo substrates and the phosphorylating kinase is known for less than 160 phosphoproteins. Here we describe, with the use of proteome chip technology, the in vitro substrates recognized by most yeast protein kinases: we identified over 4,000 phosphorylation events involving 1,325 different proteins. These substrates represent a broad spectrum of different biochemical functions and cellular roles. Distinct sets of substrates were recognized by each protein kinase, including closely related kinases of the protein kinase A family and four cyclin-dependent kinases that vary only in their cyclin subunits. Although many substrates reside in the same cellular compartment or belong to the same functional category as their phosphorylating kinase, many others do not, indicating possible new roles for several kinases. Furthermore, integration of the phosphorylation results with protein-protein interaction and transcription factor binding data revealed novel regulatory modules. Our phosphorylation results have been assembled into a first-generation phosphorylation map for yeast. Because many yeast proteins and pathways are conserved, these results will provide insights into the mechanisms and roles of protein phosphorylation in many eukaryotes.

Elm1p, Pak1p, and Tos3p are upstream kinases for the SNF1 complex that have partially redundant functions.

The Std1 protein modulates the expression of glucose-regulated genes, but its exact molecular role in this process is unclear. A two-hybrid screen for Std1-interacting proteins identified the hydrophilic C-terminal domains of the glucose sensors, Snf3 and Rgt2. The homologue of Std1, Mth1, behaves differently from Std1 in this assay by interacting with Snf3 but not Rgt2. Genetic interactions between STD1, MTH1, SNF3, and RGT2 suggest that the glucose signaling is mediated, at least in part, through interactions of the products of these four genes. Mutations in MTH1 can suppress the raffinose growth defect of a snf3 mutant as well as the glucose fermentation defect present in cells lacking both glucose sensors (snf3 rgt2). Genetic suppression by mutations in MTH1 is likely to be due to the increased and unregulated expression of hexose transporter genes. In media lacking glucose or with low levels of glucose, the hexose transporter genes are subject to repression by a mechanism that requires the Std1 and Mth1 proteins. An additional mechanism for glucose sensing must exist since a strain lacking all four genes (snf3 rgt2 std1 mth1) is still able to regulate SUC2 gene expression in response to changes in glucose concentration. Finally, studies with green fluorescent protein fusions indicate that Std1 is localized to the cell periphery and the cell nucleus, supporting the idea that it may transduce signals from the plasma membrane to the nucleus.

cThe glucose analog 2-deoxyglucose (2DG) inhibits the growth of Saccharomyces cerevisiae and human tumor cells, but its modes of action have not been fully elucidated. Yeast cells lacking Snf1 (AMP-activated protein kinase) are hypersensitive to 2DG. Overexpression of either of two low-affinity, high-capacity glucose transporters, Hxt1 and Hxt3, suppresses the 2DG hypersensitivity of snf1⌬ cells. The addition of 2DG or the loss of Snf1 reduces HXT1 and HXT3 expression levels and stimulates transporter endocytosis and degradation in the vacuole. 2DG-stimulated trafficking of Hxt1 and Hxt3 requires Rod1/Art4 and Rog3/ Art7, two members of the ␣-arrestin trafficking adaptor family. Mutations in ROD1 and ROG3 that block binding to the ubiquitin ligase Rsp5 eliminate Rod1-and Rog3-mediated trafficking of Hxt1 and Hxt3. Genetic analysis suggests that Snf1 negatively regulates both Rod1 and Rog3, but via different mechanisms. Snf1 activated by 2DG phosphorylates Rod1 but fails to phosphorylate other known targets, such as the transcriptional repressor Mig1. We propose a novel mechanism for 2DG-induced toxicity whereby 2DG stimulates the modification of ␣-arrestins, which promote glucose transporter internalization and degradation, causing glucose starvation even when cells are in a glucose-rich environment.C ells sense and respond to changes in the nutrient supply to ensure optimal cell growth and survival. To achieve this adaptation, cell-signaling cues dictate compensatory alterations in the transcriptome and proteome (1-5). The addition of the glucose analog 2-deoxyglucose (2DG) to cells causes a glucose starvation-like response, inhibiting growth and reducing viability even in the presence of abundant glucose (6, 7). 2DG is taken up and converted to 2-deoxyglucose-6-phosphate (2DG-6P) (8, 9); however, the absence of a hydroxyl group on C-2 prevents the further catabolism of 2DG-6P by phosphoglucose isomerase. Accumulation of 2DG-6P may result in product inhibition of hexokinase, thereby inhibiting glycolysis (10). In Saccharomyces cerevisiae, 2DG reportedly inhibits the biosynthesis of both cell wall polysaccharide and glycoprotein, causing cells to become osmotically fragile (11,12). Whether these are the only means by which 2DG short-circuits normal glucose utilization remains unclear.Addressing this question is of significant clinical importance, because 2DG is a potent inhibitor of cancer cell proliferation. 2DG impedes cancer progression in animal models and continues to be assessed as an anticancer therapeutic (13-16). 2DG selectively inhibits cancerous cells as a result of a key metabolic shift that distinguishes many malignant cells from the surrounding normal tissues. Many tumor cells shunt glucose through the glycolytic pathway and use lactic acid fermentation to generate ATP, a phenomenon first recognized by Otto Warburg (17,18). In spite of the widespread use of 2DG, the mechanism by which it inhibits cell growth remains controversial; it has been reported to generate a dead-end metabolite in 2DG-6P tha...

The Snf1 kinase and its mammalian homolog, the AMP-activated protein kinase, are heterotrimeric enzymes composed of a catalytic a-subunit, a regulatory g-subunit and a b-subunit that mediates heterotrimer formation. Saccharomyces cerevisiae encodes three b-subunit genes, SIP1, SIP2 and GAL83. Earlier studies suggested that these subunits may not be required for Snf1 kinase function. We show here that complete and precise deletion of all three b-subunit genes inactivates the Snf1 kinase. The sip1D sip2D gal83D strain is unable to derepress invertase, grows poorly on alternative carbon sources and fails to direct the phosphorylation of the Mig1 and Sip4 proteins in vivo. The SIP1 sip2D gal83D strain manifests a subset of Snf phenotypes (Raf + , Gly ± ) observed in the snf1D 10 strain (Raf ± , Gly ± ), suggesting that individual b-subunits direct the Snf1 kinase to a subset of its targets in vivo. Indeed, deletion of individual b-subunit genes causes distinct differences in the induction and phosphorylation of Sip4, strongly suggesting that the b-subunits play an important role in substrate de®nition.

SUMMARY The SNF1 protein kinase complex plays an essential role in regulating gene expression in response to the level of extracellular glucose in budding yeast. SNF1 shares structural and functional similarities with mammalian AMP-activated protein kinase. Both kinases are activated by phosphorylation on a threonine residue within the activation loop segment of the catalytic subunit. Here we show that ADP is the long-sought metabolite that activates SNF1 in response to glucose limitation by protecting the enzyme against dephosphorylation by Glc7, its physiologically relevant protein phosphatase. We also show that the regulatory subunit of SNF1 has two ADP binding sites. The tighter site binds AMP, ADP, and ATP competitively with NADH, whereas the weaker site does not bind NADH, but is responsible for mediating the protective effect of ADP on dephosphorylation. Mutagenesis experiments suggest that the general mechanism by which ADP protects against dephosphorylation is strongly conserved between SNF1 and AMPK.

Phosphorylation of the Saccharomyces cerevisiae Snf1 kinase activation loop is determined by the integration of two reaction rates: the rate of phosphorylation by upstream kinases and the rate of dephosphorylation by Glc7. The activities of the Snf1-activating kinases do not appear to be glucose-regulated, since immune complex kinase assays with each of the three Snf1-activating kinases show similar levels of activity when prepared from cells grown in either high or low glucose. In contrast, the dephosphorylation of the Snf1 activation loop was strongly regulated by glucose. When de novo phosphorylation of Snf1 was inhibited, phosphorylation of the Snf1 activation loop was found to be stable in low glucose but rapidly lost upon the addition of glucose. A greater than 10-fold difference in the rates of Snf1 activation loop dephosphorylation was detected. However, the activity of the Glc7-Reg1 phosphatase may not itself be directly regulated by glucose, since the Glc7-Reg1 enzyme was active in low glucose toward another substrate, the transcription factor Mig1. Glucose-mediated regulation of Snf1 activation loop dephosphorylation is controlled by changes in the ability of the Snf1 activation loop to act as a substrate for Glc7.

The yeast Snf1 kinase and its metazoan orthologues, the AMP-activated protein kinases, are activated in response to nutrient limitation. Activation requires the phosphorylation of a conserved threonine residue in the activation loop of the catalytic subunit. A phosphopeptide antibody was generated that specifically recognizes Snf1 protein that is phosphorylated in its activation loop on threonine 210. Using this reagent, we show that phosphorylation of threonine 210 correlates with Snf1 activity, since it is detected in cells subjected to glucose limitation but not in cells grown in abundant glucose. A Snf1 mutant completely lacking kinase activity was phosphorylated normally on threonine 210 in glucosestarved cells, eliminating the possibility that the threonine 210 modification is due to an autophosphorylation event. Cells lacking the Reg1 protein, a regulatory subunit for the Glc7 phosphatase, showed constitutive phosphorylation of Snf1 threonine 210. Exposure of cells to high concentrations of sodium chloride also induced phosphorylation of Snf1. Interestingly, Mig1, a downstream target of Snf1 kinase, is phosphorylated in glucose-stressed but not sodium-stressed cells. Finally, cells lacking the ␥ subunit of the Snf1 kinase complex encoded by the SNF4 gene exhibited normal regulation of threonine 210 phosphorylation in response to glucose limitation but are unable to phosphorylate Mig1 efficiently. Our data indicate that activation of the Snf1 kinase complex involves two steps, one that requires a distinct upstream kinase and one that is mediated by the ␥ subunit of the kinase itself.