• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of geneticsGeneticsCurrent IssueInformation for AuthorsEditorial BoardSubscribeSubmit a Manuscript
Genetics. Apr 2005; 169(4): 1859–1871.
PMCID: PMC1360160

A Systematic High-Throughput Screen of a Yeast Deletion Collection for Mutants Defective in PHO5 Regulation

Abstract

In response to phosphate limitation, Saccharomyces cerevisiae induces transcription of a set of genes important for survival. One of these genes is PHO5, which encodes a secreted acid phosphatase. A phosphate-responsive signal transduction pathway (the PHO pathway) mediates this response through three central components: a cyclin-dependent kinase (CDK), Pho85; a cyclin, Pho80; and a CDK inhibitor (CKI), Pho81. While signaling downstream of the Pho81/Pho80/Pho85 complex to PHO5 expression has been well characterized, little is known about factors acting upstream of these components. To identify missing factors involved in the PHO pathway, we carried out a high-throughput, quantitative enzymatic screen of a yeast deletion collection, searching for novel mutants defective in expression of PHO5. As a result of this study, we have identified at least nine genes that were previously not known to regulate PHO5 expression. The functional diversity of these genes suggests that the PHO pathway is networked with other important cellular signaling pathways. Among these genes, ADK1 and ADO1, encoding an adenylate kinase and an adenosine kinase, respectively, negatively regulate PHO5 expression and appear to function upstream of PHO81.

THE ability to respond appropriately to environmental changes is essential for cell survival. Microorganisms respond to nutrient limitation by regulating the expression of genes important for survival. Inorganic phosphate is an essential nutrient for the synthesis of many cellular components such as nucleic acids, phospholipids, and phospho-metabolites. The budding yeast Saccharomyces cerevisiae responds to changes in extracellular inorganic phosphate concentration by regulating the phosphate-responsive signaling pathway (the PHO pathway) (Lenburg and O'Shea 1996). As a result of PHO pathway signaling, many PHO genes are repressed under high-phosphate conditions and induced under no-phosphate conditions. One of these genes is PHO5, which encodes a secreted acid phosphatase.

The PHO pathway was originally described by Oshima and colleagues (Oshima 1982). PHO2, PHO4, and PHO81 are positive regulators of PHO5; deletion of these genes results in an inability to induce PHO5 upon phosphate starvation (Toh-e et al. 1973). PHO80, PHO85, and PHO84 are required for the repression of PHO5, and loss-of-function mutations in these genes result in constitutive expression of PHO5, even under high-phosphate conditions (Ueda et al. 1975).

Significant progress has been made in understanding the molecular mechanism of the PHO signaling pathway. Central to the PHO pathway is a cyclin/cyclin-dependent kinase (CDK) complex, Pho80/Pho85 (Toh-e and Shimauchi 1986; Uesono et al. 1987; Madden et al. 1988; Toh-e et al. 1988; Kaffman et al. 1994), whose activity is regulated in response to external phosphate concentrations. Pho81 (Coche et al. 1990; Schneider et al. 1994; Ogawa et al. 1995), a CDK inhibitor, binds to Pho80/Pho85 when cells are grown under both high- and no-phosphate conditions. However, it appears that Pho81 inhibits the kinase only during phosphate starvation (Schneider et al. 1994). The inhibition is mediated by a novel CDK inhibitor motif (Huang et al. 2001). The Pho81/Pho80/Pho85 complex regulates the activity of Pho4 (Kaffman et al. 1994; Komeili and O'Shea 1999), a transcription factor required for PHO5 expression. When yeast cells are grown in high-phosphate medium, Pho4 is phosphorylated by Pho80/Pho85. Phosphorylated Pho4 is localized predominantly to the cytoplasm and PHO5 transcription is repressed (O'Neill et al. 1996). When yeast cells are grown in medium devoid of phosphate, the kinase activity of Pho80/Pho85 is inhibited by Pho81. Thus, Pho4 is unphosphorylated and localized to the nucleus, where it activates PHO5 transcription.

Even though the molecular mechanism of signaling from the Pho81/Pho80/Pho85 complex to PHO5 expression has been well characterized, the phosphate sensor is still not known, and the signaling process between the sensor and the kinase complex is not understood. PHO84, encoding a phosphate-starvation-inducible high-affinity H+/PO4 symporter, is suggested to function upstream of PHO81, because loss-of-function mutations in PHO81 are epistatic to mutations in PHO84 (Bun-Ya et al. 1991). Cells of the pho84Δ strain express PHO5 constitutively, suggesting that Pho84 might be the phosphate sensor. However, overexpression of unrelated phosphate transporters or a glycerophosphoinositol transporter in the pho84Δ strain suppresses its constitutive phenotype, suggesting that Pho84 is not required for sensing phosphate (Wykoff and O'Shea 2001). Although traditional genetic studies have been conducted to search for a phosphate sensor and other factors that signal upstream of PHO81 in the PHO pathway, the results were limited by the fact that the assays were not quantitative or systematic (Ueda et al. 1975; Lau et al. 1998).

To identify missing factors in the PHO pathway and to better understand signal transduction for PHO5 regulation, we developed a high-throughput, quantitative acid-phosphatase liquid assay to screen the yeast deletion collection (Winzeler et al. 1999), searching for novel mutants that are defective in PHO5 regulation. We wished to identify signaling components that function upstream of PHO81 and are required for the regulation of PHO5. As a result of this study, we have identified and confirmed at least nine functionally diverse genes that were previously not known to regulate PHO5 expression. When each of these genes was deleted, the resulting yeast strains displayed either constitutive PHO5 expression under high-phosphate conditions or reduced PHO5 expression upon phosphate starvation. Analysis of these mutants suggests that ADK1 and ADO1 (Konrad 1988; Lecoq et al. 2001), encoding an adenylate kinase and an adenosine kinase, respectively, negatively regulate PHO5 expression and appear to function upstream of PHO81.

MATERIALS AND METHODS

Yeast strains and growth conditions:

All yeast deletion strains used in this study had the entire open reading frame of each gene deleted as indicated. For the initial screen and characterization, all strains except kcs1Δ were taken from the yeast deletion collection (MATa haploid complete set) (Winzeler et al. 1999). The kcs1Δ strain from the yeast deletion collection used for this study appeared to contain an additional mutation(s), which resulted in a constitutive phenotype similar to a pho80Δ strain (data not shown). KCS1 is known to have only modest effects on Pho5 expression when it is deleted (Steger et al. 2003). For this work, we replaced the kcs1Δ strain from the yeast deletion collection with a kcs1Δ strain (EY1258) that exhibited the same phenotype as previously published (Steger et al. 2003). For the independently generated deletion strains, all genes were inactivated using a PCR-based deletion protocol that deleted the entire open reading frame (Kitada et al. 1995). Deletion of the appropriate genes was confirmed by PCR and, in some cases, by phenotypic analysis. For disruptions that were initially performed on diploid strains (EY519, EY1406, EY1407, EY1509, EY1510, and EY1600; Table 1), the resulting heterozygous diploids were sporulated, and the tetrads were dissected to isolate haploids with the desired gene deletions (Guthrie and Fink 1991). Yeast strains used in this study are listed in Table 1. All strains were grown in standard yeast medium as described (Guthrie and Fink 1991). No-phosphate medium was prepared as described (Lau et al. 1998).

TABLE 1
Strains ofS. cerevisiae used in this study

High-throughput liquid acid-phosphatase assay:

Yeast cultures were incubated at 30° in a HiGro incubator shaker (Gene Machines), centrifugations were performed at 3000 rpm in a Beckman GS-6KR centrifuge (Beckman, Fullerton, CA), high-throughput liquid assays were conducted in a 96-well format at room temperature using a Biomek FX 96-channel pipetting robot (Beckman), and the OD600 or OD420 was measured in a Spectra Max 340 plate reader (Molecular Devices, Menlo Park, CA). Each 96-well plate of the yeast deletion collection (Winzeler et al. 1999) was thawed and inoculated onto a YEPD plate using a 96-pin tool and then incubated for 2 days. The strains were then inoculated into a 96-well plate containing 600 μl of SD complete medium in each well using a 96-pin tool and grown overnight to saturation. Cells grown to saturation and cells grown at midlog phase showed similar Pho5 activity. Each culture was spun down and washed two times with 600 μl of no-phosphate medium and resuspended in 200 μl of no-phosphate medium. Fifty microliters of cell suspension was then reinoculated into 850 μl of no-phosphate medium and incubated at 30°. Every 120 min after transfer to no-phosphate medium, 200 μl of each culture was withdrawn and the OD600 was measured. To start the assay, 50 μl of each culture was added to 200 μl of p-nitrophenylphosphate (5.62 mg/ml in 0.1 m sodium acetate, pH 4.2), mixed, and incubated at room temperature for 15 min. The reaction was stopped by the addition of 200 μl of ice-cold 10% trichloroacetic acid. A total of 200 μl of the reaction mixture was then withdrawn, added to 200 μl of saturated sodium carbonate solution, mixed, and spun for 10 min at 3000 rpm. Finally, 200 μl of the supernatant was removed and the OD420 was measured. The units of phosphate activity were expressed as OD420/OD600 × 1000. The mutant candidates identified from the initial screen were reorganized into a new set of 96-well plates and a secondary liquid assay was performed in the same manner.

PHO5 mRNA analysis:

Yeast strains were grown to an OD600 of 0.4–1.0 in SD complete medium in 96-well plates as for the high-throughput liquid acid-phosphatase assay described above. A 1-ml culture of each strain was then transferred into a microcentrifuge tube, and total RNA was extracted by standard acid phenol treatment (Guthrie and Fink 1991), reverse transcribed using Stratascript reverse transcriptase following the manufacturer's instructions (Stratagene, La Jolla, CA), and treated with RNaseA. As a control, each sample was additionally mock treated (without the reverse transcriptase). For each reverse transcription (RT) reaction, a PHO5 RT primer (5′-TTGTCTCAATAGACTGGCGTTGTAA) and an ACT1 RT primer (5′-TGGTGAACGATAGATGGACCA) were added into the same reaction. Appropriately diluted RT products were then analyzed by quantitative PCR in real time using an Opticon continuous fluorescence detection system (MJ Research, Watertown, MA) and primers amplifying PHO5 (nucleotides 630–760) or ACT1 (nucleotides 715–816). For each strain, the relative level of PHO5 mRNA was normalized to that of ACT1. For most of the samples, the mock signals were <1% of the actual signals.

Fluorescence microscopy:

All microscopy experiments were performed as described (Huang et al. 2001).

RESULTS

Isolation of mutants defective in Pho5 induction in response to phosphate starvation from a systematic screen of the yeast deletion collection:

We developed a systematic high-throughput quantitative acid-phosphatase liquid assay (see materials and methods) to screen the yeast deletion collection and searched for novel mutants that were defective in PHO5 regulation. The yeast deletion collection consists of 4848 MATa haploid strains, each lacking a single nonessential gene (Winzeler et al. 1999). In contrast to previous genetic studies (Ueda et al. 1975; Lau et al. 1998), our high-throughput liquid phosphatase assay is a highly sensitive and quantitative method to examine Pho5 expression in vivo. Furthermore, the gene mutated in each strain in the yeast deletion collection is known and facilitates rapid analysis. In our initial screen, cells were transferred to no-phosphate conditions, sampled, and assayed for Pho5 phosphatase activity every 120 min for 360 min. This enabled us to identify mutants with kinetic defects in PHO5 induction. We used a kinetic assay because deletion of some genes involved in PHO5 regulation results in kinetic defects in PHO5 expression, but no defects in induction measured after overnight growth in no-phosphate medium (Barbaric et al. 2001; Steger et al. 2003).

During the initial screen, each 96-well plate from the yeast deletion collection was cultured and assayed for Pho5 acid-phosphatase activity upon phosphate starvation as described in the materials and methods. The data from each plate were then graphed and analyzed (Figure 1A). Upon phosphate starvation, a range of induced Pho5 expression in the population was observed. The mean (μ) and standard deviation (σ) were calculated for the entire population at each time point. To identify potential mutant candidates, we then assigned a control range (μ ± σ) (Figure 1). Strains that had a lower Pho5 activity than the control range at three of four time points were called uninducible mutant candidates, and strains that had a higher Pho5 activity than the control range were called hyperinducible mutant candidates. As expected, known mutants in the PHO pathway were located outside of this control range (e.g., pho80Δ and pho4Δ) (Toh-e et al. 1973). A histogram of the 240-min data points of the entire library is shown in Figure 1B. The histograms of other time points have similar distribution patterns (data not shown). To define a manageable number of mutant candidates for the scope of this study, we selected 100 uninducible candidates with Pho5 activity that is at least one standard deviation from the mean, and 240 hyperinducible candidates with Pho5 activity that is at least two standard deviations from the mean.

Figure 1.
A systematic screen of a yeast deletion collection for mutants defective in PHO5 regulation. (A) A time course of acid-phosphatase activity in strains from a typical plate (no. 37) from the yeast deletion collection (Winzeler et al. 1999) in response ...

To confirm the Pho5 expression phenotypes of these 340 candidates, we reorganized them into four new 96-well plates, each of which also contained 12 wild-type (WT) controls from the same library. These 12 controls were obtained by simple random sampling to cover the control range (Figure 1) and do not have a known PHO phenotype. In each plate, we also included an additional six known mutants that cover the spectrum of PHO phenotypes: pho80Δ, pho85Δ, pho81Δ, pho4Δ, arg82Δ, and snf6Δ (Toh-e et al. 1973; Ueda et al. 1975; Neef and Kladde 2003; Steger et al. 2003). These four plates were cultured, assayed, and analyzed in the same manner as the initial screen. The profiles of the 12 wild-type controls from six independent experiments are consistently superimposed on the control range described above (Figure 1C). Of these 340 candidates, 240 showed reproducible defects in Pho5 induction: 160 were hyperinducible and 80 were uninducible. For the follow-up studies, we selected the 62 most statistically significant hyperinducible mutant candidates (>μ + 5σ) and the 28 most statistically significant uninducible mutant candidates (<μ − 2σ).

Initial characterization of the mutant candidates:

The 90 most statistically significant mutant candidates (62 hyperinducible and 28 uninducible) were isolated because they displayed altered Pho5 induction under no-phosphate conditions as described above. To distinguish the hyperinducible candidates that have constitutive Pho5 expression (constitutive mutants) under both high- and no-phosphate conditions from candidates that have hyperinducible Pho5 expression only under no-phosphate conditions, we cultured and assayed the 62 hyperinducible candidates for Pho5 expression in high-phosphate medium at the 300-min time point. The average of three independent measurements is shown in Figure 2A. Of the 62 hyperinducible candidates, 48 exhibited significantly elevated Pho5 activity under high-phosphate conditions as compared to the average value of the 12 wild-type controls (>μ + 2σ). Since these 48 candidates exhibited significantly constitutive Pho5 expression in high-phosphate medium, we referred to them as “constitutive mutant candidates.”

Figure 2.
Initial characterization of the constitutive mutant candidates. (A) Phosphatase activity of the 62 hyperinducible mutant candidates grown in high-phosphate medium. Forty-eight candidates (constitutive mutants, solid columns) exhibited significantly elevated ...

The 28 uninducible mutant candidates (<μ − 2σ) were also cultured in no-phosphate media, assayed, and analyzed in the same manner as in the initial screen. The average of three independent measurements at a 360-min time point is shown in Figure 3A. As expected, these 28 uninducible candidates showed significantly reduced Pho5 activity as compared to the controls (<μ − 2σ).

Figure 3.
Initial characterization of the uninducible mutant candidates. (A) Phosphatase activity of the 28 uninducible mutant candidates grown in no-phosphate medium at the 360-min time point. All 28 uninducible candidates exhibited significantly reduced Pho5 ...

PHO5 mRNA analysis of the strongest mutant candidates:

To distinguish the mutants that affect Pho5 regulation at the level of transcription from the mutants that affect other processes, we performed PHO5 mRNA analysis on the most statistically significant mutant candidates (48 hyperinducible and 28 uninducible).

If a constitutive mutant has defects in PHO5 transcription, the PHO5 mRNA level should be elevated compared to the wild-type strain. If a constitutive mutant instead has defects only in Pho5 protein production or secretion, the PHO5 mRNA level will be similar to the wild-type strain. To differentiate between these types of mutants, we quantitated PHO5 mRNA levels (normalized to ACT1) in the 48 constitutive mutants under high-phosphate conditions at the 300-min time point using reverse transcription-quantitative polymerase chain reaction (RT-QPCR) (materials and methods). We found that the relative PHO5 mRNA levels in 44 of the 48 constitutive mutant candidates are elevated at least twofold over the average of the 12 wild-type controls (Figure 2B). The ssn8Δ, hof1Δ, rps6AΔ, and cin8Δ strains exhibited no significant differences from the controls, suggesting that they might affect Pho5 regulation through effects on process(es) other than transcription. The pho85Δ, pho80Δ, and reg1Δ strains from the yeast deletion collection have the most elevated levels of PHO5 mRNA (>60-fold more than the controls). Among the rest of mutants, the adk1Δ and ado1Δ strains have the most elevated levels of PHO5 mRNA (>30-fold more than the controls).

Similarly, if an uninducible mutant has defects in PHO5 transcription, the PHO5 mRNA level should be reduced compared to the wild-type strain under no-phosphate conditions. The PHO5 mRNA levels of the 28 uninducible candidates under no-phosphate conditions (at the 360-min time point) were analyzed in the same manner as the constitutive candidates described above. We found that the relative PHO5 mRNA levels in 16 of the 28 uninducible mutant candidates are at least twofold less than the average in the 12 wild-type controls (Figure 3B). The pho4Δ, pho81Δ, and pho2Δ strains from the yeast deletion collection had the lowest levels of PHO5 mRNA (>100-fold less than the controls). Among the rest of uninducible candidates, the ino4Δ, spt7Δ, and snf2Δ strains had the most significant reduction in the levels of PHO5 mRNA (5- to 15-fold less than the controls).

For several of the uninducible and constitutive mutants we observed discrepancies between Pho5 activity levels and PHO5 mRNA levels. As suggested above, this lack of correlation may reflect a role for these gene products in Pho5 regulation downstream of transcription. However, in some cases, the lack of correlation may result from global effects of a mutation on gene expression. For example, mutations that affect both ACT1 and PHO5 mRNA levels will appear as if they do not affect PHO5 mRNA because of the way in which we have normalized the data. This is particularly a concern for some of the uninducible mutants with defects in known transcriptional components.

Confirmation of the mutant candidates:

Before characterizing the mutant candidates further, it was essential to confirm that the phenotype of each candidate was the result of the deletion of the gene as indicated in the collection. It is possible that a deletion strain from the yeast deletion collection used for this study might contain additional mutation(s), which could be responsible for the phenotype observed. This was the case for the kcs1Δ strain from the yeast deletion collection used in our study (materials and methods). For the scope of this study, we chose to independently regenerate strains corresponding to the strongest constitutive mutant candidates, which exhibited Pho5 activity at least fourfold higher than that of the controls under high-phosphate conditions (pho85Δ, pho80Δ, reg1Δ, adk1Δ, ado1Δ, ykl169cΔ, and mot2Δ; Figure 2A) along with the kcs1Δ strain as an additional control. We also reconstructed the strongest uninducible mutant candidates that exhibited a Pho5 activity at least fourfold less than that of the controls under no-phosphate conditions (pho81Δ, pho4Δ, snf2Δ, pho2Δ, spt7Δ, ada3Δ, fur4Δ, gcn5Δ, ada2Δ, ino4Δ, vps24Δ, swi3Δ, alt1Δ, and snf6Δ; Figure 3A). These 22 deletion strains were constructed from a starting strain that has the same genetic background as the yeast deletion collection, containing PHO4-YFP integrated at the PHO4 locus (EY1580; Table 1). These reconstructed deletion strains were cultured, assayed for Pho5 activity, and analyzed in the same manner as described above. If the original phenotype of each candidate were the sole result of the deletion of the gene as indicated in the deletion library, these newly constructed mutant strains should exhibit a similar PHO phenotype. Among the seven strongest constitutive mutant strains, five showed profiles similar to the original phenotypes (pho85Δ, pho80Δ, adk1Δ, ado1Δ, and mot2Δ, Figures 2A and and4A).4A). The newly constructed reg1Δ strain exhibited a modest phenotype similar to kcs1Δ, and the new ykl169cΔ strain appeared to behave like the wild-type strain. Among the 14 top uninducible mutant strains, 13 showed profiles similar to the original phenotypes (pho81Δ, pho4Δ, snf2Δ, pho2Δ, spt7Δ, ada3Δ, gcn5Δ, ada2Δ, ino4Δ, vps24Δ, swi3Δ, alt1Δ, and snf6Δ strains; Figures 3A and and4B).4B). The newly constructed fur4Δ strain appeared to behave like the wild-type strain. The time course analysis of Pho5 induction in these 14 uninducible mutants also yielded the same result (data not shown).

Figure 4.
Confirmation of the strongest mutant candidates. (A) Phosphatase activity in the seven reconstructed constitutive mutant candidates grown in high-phosphate medium. These seven constitutive candidates initially exhibited the highest levels of Pho5 activity, ...

Pho4-YFP localization in the confirmed mutants:

Among these 20 confirmed genes whose deletion affects Pho5 regulation, 9 were previously not known to be involved in regulating PHO5 expression: ADK1, ADO1, MOT2, REG1, ADA3 INO4 SWI3, VPS24, and ALT1 (Saccharomyces Genome Database, SGD, http://www.yeastgenome.org). To distinguish genes involved in the signaling process upstream of the Pho80/Pho85 complex from those that affect other aspects of PHO5 regulation (e.g., transcriptional repression), we monitored localization of a Pho4-YFP fusion protein. Pho4 localization in different mutant backgrounds reflects the activity of Pho80/Pho85 and indicates whether the gene acts upstream or downstream of the kinase complex in the PHO pathway. In the reg1Δ, mot2Δ, and kcs1Δ constitutive mutant strains, we found that Pho4-YFP was localized to the cytoplasm under high-phosphate conditions, suggesting that REG1, MOT2, and KCS1 may act downstream of the kinase complex (Figure 5A). In contrast, we found that Pho4-YFP was localized to the nucleus in the ado1Δ and adk1Δ constitutive mutant strains under high-phosphate conditions, suggesting that ADK1 and ADO1 may act upstream of the kinase complex in the PHO pathway (Figure 5A). In all of the uninducible mutant strains (e.g., pho2Δ, snf2Δ, ada3Δ, and gcn5Δ; data not shown for the others) except pho81Δ, we found that Pho4-YFP was localized to the nucleus under no-phosphate conditions, suggesting that these genes may act downstream of the kinase complex (Figure 5B).

Figure 5.
Pho4-YFP localization studies in the strongest mutants. (A) Localization of Pho4-YFP in the strongest constitutive mutant strains grown in high-phosphate medium. The following strains were used: WT (EY1580), pho85Δ (EY1591), pho80Δ (EY1592), ...

PHO81 dependence of ADO1 and ADK1:

To confirm our conclusions from the Pho4-YFP localization study, we analyzed the epistatic relationship of PHO81 to ADO1 and ADK1. We reasoned that a pho81 mutant should be epistatic to mutants defective in the signaling process upstream of the kinase complex. In these mutant strains, deletion of the PHO81 gene should result in cytoplasmic localization of Pho4 and an uninducible Pho5 expression phenotype, whereas the mutants that affect other aspects of PHO5 regulation (e.g., transcriptional repression) will be epistatic to the pho81Δ mutant. Double mutants of pho81Δ ado1Δ and pho81Δ adk1Δ were generated and examined for PHO5 expression by acid-phosphatase plate assay (Figure 6). Both ADO1 and ADK1 showed a PHO81 dependence for Pho5 expression. As expected, in these double-mutant backgrounds Pho4-GFP was localized to the cytoplasm (data not shown). We conclude that ADK1 and ADO1 act upstream of PHO81 in the PHO pathway.

Figure 6.
PHO81 dependence of ADO1 and ADK1. (A) A pho81Δ ado1Δ (EY1509) double mutant was generated and examined for Pho5 expression by acid-phosphatase plate assay. A wild-type strain (K699 MATα) induced Pho5 expression upon phosphate ...

DISCUSSION

In an effort to better understand the signaling process in the PHO pathway, we have conducted a high-throughput and systematic enzymatic screen for mutants that are defective in PHO5 regulation. We wished to identify genes that function upstream of PHO81 and are required for PHO5 repression. Our study identified and confirmed 20 genes that appear to be involved in PHO5 regulation. Among these genes, 7 result in a constitutive PHO phenotype, and 13 result an uninducible PHO phenotype when deleted (Figure 4). Of these 20 genes, 9 were previously not known to be involved in PHO5 regulation (repression—ADK1, ADO1, MOT2, REG1; induction—ADA3, INO4, SWI3, VPS24, ALT1).

Among the constitutive mutants, the pho80Δ and pho85Δ strains showed the most elevated levels of Pho5 phosphatase activity and PHO5 mRNA under high-phosphate conditions (Figures 2 and and4A),4A), consistent with their central role in the PHO pathway. Complete loss of the kinase activity (Pho80/Pho85) results in full activation of the transcription factor Pho4, which then leads to full expression of PHO5. Interestingly, the adk1Δ and ado1Δ strains had the most elevated levels of Pho5 phosphatase activity and PHO5 mRNA under high-phosphate conditions among the rest of the constitutive mutants (Figures 2 and and4A),4A), suggesting that these two genes might play important roles in PHO5 repression. Consistent with this hypothesis, ado1Δ and adk1Δ are the only strains that had nuclear Pho4-YFP localization similar to pho80Δ and pho85Δ strains (Figure 5A) and that exhibited phenotypes that were PHO81 dependent, as seen in the epistatic analysis (Figure 6). These results suggest that ADK1 and ADO1 may act upstream of PHO81 in the PHO pathway (Figure 7).

Figure 7.
The PHO pathway might be networked with other cellular signaling pathways that are important for cell survival. Genes required for efficient PHO5 induction under no-phosphate conditions are boxed on the left next to the known PHO pathway, whereas genes ...

Adk1 and Ado1 play important roles in the regulation of adenosine nucleotides: ADK1 encodes an adenylate kinase (Konrad 1988; Abele and Schulz 1995), which catalyzes the interconversion of nucleotides between AMP and ADP; ADO1 encodes an adenosine kinase (Lecoq et al. 2001), which catalyzes the salvage synthesis of adenine monophosphate from adenosine and ATP. Little is known about these nucleotide kinases in yeast. Yeast cells carrying a disrupted ADK1 locus showed a significant decrease in the level of nucleoside triphosphates (Konrad 1988). The physiological role of Ado1 is suggested to be to recycle adenosine produced by the methyl cycle (Lecoq et al. 2001). Why would ADK1 and ADO1, which are involved in adenosine nucleotide regulation, act upon the PHO pathway? Since inorganic phosphate is essential for nucleotide synthesis, it is possible that nucleotide regulation is connected to the PHO signaling pathway to coordinate these two processes under different nutrient and growth conditions. When inorganic phosphate or nucleotide levels are high, repression of PHO5 expression may conserve energy. Adk1 and Ado1 might repress the inhibitory activity of the CKI, Pho81, leading to activation of the kinase, Pho80/Pho85, which then inactivates the transcription factor Pho4 required for PHO5 expression. On the other hand, if inorganic phosphate or nucleotide levels are low, repression of Pho81 by Adk1 and Ado1 might be relieved. PHO5 expression might then be induced, generating more inorganic phosphate, which would lead to an increase of nucleotide synthesis. Consistent with this hypothesis that Adk1 and Ako1 may regulate PHO5 expression in response to phosphate conditions, it has been shown that Ado1 activity is dependent on the presence of inorganic phosphate and other ions (Maj et al. 2000, 2002).

How might Adk1 and Ado1 repress Pho81? The Pho81/Pho80/Pho85 complex localizes to the nucleus where the regulation of the kinase complex is expected to take place (Kaffman et al. 1998; Huang et al. 2001). Both Adk1 and Ado1 were found in the cytoplasm and nucleus (Huh et al. 2003). Adk1 was also detected in a Pho85-associated complex, which included Pho81, in a high-throughput analysis of protein-protein interaction (Ho et al. 2002). One possible model is that Adk1 inhibits Pho81 by physical contact that is regulated by intracellular inorganic phosphate concentrations. When the concentration of intracellular inorganic phosphate is high, the physical interaction of Adk1 with Pho81 may repress Pho81's inhibitory activity. When intracellular inorganic phosphate is low, the interaction between Adk1 and Pho81 may change (e.g., via a protein conformational switch), leading to the activation of Pho81. Dissecting the interactions between Adk1 and the kinase complex might help us to understand the regulation of Pho81 and how the PHO signaling process is connected to nucleotide regulation.

In addition to the known PHO genes, the four novel genes (ADK1, ADO1, MOT2, and REG1) and one known gene (KCS1) required for PHO5 repression under high-phosphate conditions can be classified into four different subclasses (Figure 7, right). The first subclass plays an important role in regulation of the adenosine nucleotides: ADO1 and ADK1. Our study suggests that they are the only two new genes that might act upstream of PHO81 in the PHO pathway. The second subclass consists of MOT2, encoding a global transcriptional regulator. Mot2 has effects on the expression of many genes involved in diverse pathways (Cade and Errede 1994; Irie et al. 1994; Leberer et al. 1994; Lenssen et al. 2002). We observed that PHO5 mRNA levels were elevated in the mot2Δ strain (Figure 2B), suggesting that the transcription of PHO5 might be repressed by Mot2. Consistent with our findings, it has been shown that deletion of MOT2 caused increased transcription of another PHO gene, PHO84 (Irie et al. 1994). Our Pho4-YFP localization study suggests that MOT2 acts downstream of the kinase complex, Pho81/Pho80/Pho85 (Figure 5A). Mot2 may act directly at the PHO5 promoter or via a more indirect mechanism. It is also interesting to note that the deletion of MOT2 resulted in an increase in Pho5 activity similar to the deletions of ADO1 or ADK1 (Figures 2A and and4A),4A), whereas the PHO5 mRNA level in the mot2Δ strain is significantly less than that in the ado1Δ or adk1Δ strains (Figure 2B). This suggests that Mot2 might regulate Pho5 expression at another level in addition to transcription. It is also possible that the MOT2 deletion strain exhibits pleiotropic transcriptional defects that affect both PHO5 and ACT1, the gene we used for normalization of mRNA levels.

The third subclass of genes involved in the PHO5 repression consists of genes involved in glucose repression and includes REG1 (Gancedo 1998), which has been shown to negatively regulate transcription of glucose-repressive genes such as ADH2 (Dombek et al. 1999). Since the PHO pathway is one of the fundamental metabolic regulatory pathways in yeast, it is possible that the glucose repression pathway and the PHO pathway are coordinately regulated. Consistent with this hypothesis, it has been shown that phosphate and glucose cooperate to activate the protein kinase A pathway (Giots et al. 2003).

The fourth subclass of genes required for PHO5 repression appears to be involved in the stress response: KCS1 is required for the synthesis of inositol pyrophosphates, which are essential for vacuole biogenesis and the cell's response to certain environmental stresses (Saiardi et al. 2000; Dubois et al. 2002). KCS1 is shown to be required for repression of PHO genes by phosphate (El Alami et al. 2003; Steger et al. 2003). By networking the PHO signaling pathway and these general survival processes, yeast might benefit by a better response to environmental changes.

Among the 13 confirmed uninducible mutants, the pho81Δ, pho4Δ, and pho2Δ strains showed no significant PHO5 expression upon phosphate starvation (Figures 3 and and4B),4B), consistent with their critical roles in the PHO pathway. In addition to the known PHO genes, our study also identified five novel genes (ADA3, INO4, SWI3, VPS24, and ALT1) and five known genes (ADA2, GCN5, SPT7, SNF2, and SNF6) that are involved in Pho5 induction upon phosphate starvation. All these genes appeared to act downstream of PHO4 since Pho4-YFP is localized to the nucleus in these deletion mutants under no-phosphate conditions (Figure 5B). These genes involved in efficient Pho5 induction can be classified into four different subclasses (Figure 7, left). The first subclass consists of genes that encode for the components of the SAGA histone acetyltransferase (HAT) chromatin-remodeling complex: ADA2, ADA3, GCN5, and SPT7 (Figures 3 and and4B)4B) (Roth et al. 2001; Narlikar et al. 2002). Gcn5 and Spt7 have been shown to be required for the transcriptional regulation of PHO5 (Gregory et al. 1998; Kuo et al. 1998; Nishimura et al. 1999; Barbaric et al. 2001, 2003; Neef and Kladde 2003). Ada2 and Ada3 are the SAGA components that regulate the HAT activity of Gcn5. As expected, Ada2 is required for chromatin remodeling at the PHO5 promoter by regulating Gcn5 (Barbaric et al. 2003). Consistent with these findings, our study also confirmed the roles of Gcn5, Spt7, and Ada2 in PHO5 regulation. Furthermore, we found that Ada3 was necessary for efficient PHO5 induction (Figures 3 and and4B).4B). Strains lacking Ada3 and Gcn5 have similar profiles in which PHO5 mRNA expression was delayed compared to wild type (data not shown; Figure 3B) (Barbaric et al. 2001), suggesting similar roles for these two SAGA components in PHO5 regulation.

The second subclass of genes required for efficient Pho5 induction upon phosphate starvation consists of the components of the SWI/SNF ATP-dependent chromatin-remodeling complex: SNF2 (SWI2), SWI3, and SNF6 (Peterson and Tamkun 1995; Smith et al. 2003). The SWI/SNF complex has been shown to be required for efficient remodeling of PHO5 promoter chromatin structure: PHO5 chromatin remodeling appears to be defective in the snf6Δ strain (Steger et al. 2003), while induction of the Pho5 acid-phosphatase activity requires Snf2 (Swi2) (Neef and Kladde 2003). Consistent with these findings, our study confirmed the roles of Snf2 and Snf6 in Pho5 regulation and also showed that Swi3 is involved in PHO5 induction, similar to Snf6 (Figures 3 and and4B4B).

The third subclass of genes required for efficient Pho5 induction includes INO4, which encodes a transcription factor required for derepression of inositol-choline-regulated genes involved in phospholipid synthesis, such as INO1 (Ambroziak and Henry 1994). We found that Ino4 was also required for efficient PHO5 induction in response to phosphate starvation. Since phosphate is required for the synthesis of inositol phosphates and phospholipid synthesis, it is possible that the PHO pathway might be linked to these pathways. Ino4 may act directly on the PHO5 promoter, similar to its regulation of INO1 promoter, or via a more indirect mechanism.

The last subclass of genes required for efficient PHO5 induction appears to be involved in alanine metabolism and protein secretion: ALT1 encodes a putative alanine transaminase (Tatusov et al. 2000) and VPS24 encodes a component for a vesicle-mediated transport system involved in protein secretion (Babst et al. 1998). These genes are not likely to be involved in signaling and are more likely to regulate PHO5 expression through the steps of protein synthesis and secretion. Alternatively, they may act more indirectly to modulate Pho5 expression.

As a result of this study, we were able to identify and confirm 20 genes that appear to be involved in PHO5 regulation. Since we chose to screen the yeast deletion collection, any essential gene involved in the PHO5 regulation would not be identified. It is also important to note that, for the scope of this study, we characterized only the 20 genes that showed the strongest phenotype when deleted. There are still many interesting candidates with weaker phenotypes that can be validated in future studies. For example, these candidates include known genes, such as ARG82 encoding an inositol polyphosphate kinase, which is required for chromatin remodeling of the PHO5 promoter (Steger et al. 2003) and PHO23, encoding a probable component of the Rpd3 histone deacetylase complex, which is involved in transcriptional regulation of PHO5 (Lau et al. 1998). We did not isolate 2 of the known genes that cause significant constitutive expression of PHO5:PHO84 encoding a phosphate starvation-inducible high-affinity H+/PO4 symporter, and PHO86, encoding a protein specifically required for packaging of Pho84 (Lau et al. 2000). This is most likely due to the fact that pho84Δ and pho86Δ strains can quickly accumulate suppressor mutations (our unpublished results). It was reported by others that ASF1 is required for the transcription of PHO5 (Adkins et al. 2004). However, we did not observe any significant defect in PHO5 expression (no more than 1.5-fold less than the wild type throughout an 8-hr time course) in the asf1Δ strain from the yeast deletion collection or in the asf1Δ strains that we generated in various strain backgrounds (data not shown).

In summary, this systematic high-throughput study enabled us to identify 20 genes that are involved in PHO5 regulation. We were able not only to confirm the roles of 11 known genes in the PHO pathway, but also to isolate 9 novel genes previously unknown to be involved in PHO5 regulation. Among these genes, ADK1 and ADO1 might act upstream of PHO81 in the PHO pathway, suggesting that the PHO signaling process is connected to nucleotide regulation. The functional diversity of these 9 genes suggests that the PHO pathway might be networked with other important cellular signaling pathways that are important for cell survival (Figure 7). This might enable the yeast cells to better respond to extracellular nutrient changes and conserve energy for survival. A comprehensive analysis of the complete set of constitutive and uninducible mutants will yield important information for understanding the signaling network that allows yeast cells to respond appropriately to environmental changes.

Acknowledgments

We thank members of the O'Shea lab for critical reading of the manuscript and Ledion Bitincka for help with the statistical analysis of the data. This work was supported by grants from the National Institutes of Health (GM51377), the David and Lucile Packard Foundation, and the Howard Hughes Medical Institute (E.K.O.).

References

  • Abele, U., and G. E. Schulz, 1995. High-resolution structures of adenylate kinase from yeast ligated with inhibitor Ap5A, showing the pathway of phosphoryl transfer. Protein Sci. 4: 1262–1271. [PMC free article] [PubMed]
  • Adkins, M. W., S. R. Howar and J. K. Tyler, 2004. Chromatin disassembly mediated by the histone chaperone Asf1 is essential for transcriptional activation of the yeast PHO5 and PHO8 genes. Mol. Cell 14: 657–666. [PubMed]
  • Ambroziak, J., and S. A. Henry, 1994. INO2 and INO4 gene products, positive regulators of phospholipid biosynthesis in Saccharomyces cerevisiae, form a complex that binds to the INO1 promoter. J. Biol. Chem. 269: 15344–15349. [PubMed]
  • Babst, M., B. Wendland, E. J. Estepa and S. D. Emr, 1998. The Vps4p AAA ATPase regulates membrane association of a Vps protein complex required for normal endosome function. EMBO J. 17: 2982–2993. [PMC free article] [PubMed]
  • Barbaric, S., J. Walker, A. Schmid, J. Q. Svejstrup and W. Horz, 2001. Increasing the rate of chromatin remodeling and gene activation—a novel role for the histone acetyltransferase Gcn5. EMBO J. 20: 4944–4951. [PMC free article] [PubMed]
  • Barbaric, S., H. Reinke and W. Horz, 2003. Multiple mechanistically distinct functions of SAGA at the PHO5 promoter. Mol. Cell. Biol. 23: 3468–3476. [PMC free article] [PubMed]
  • Bun-Ya, M., M. Nishimura, S. Harashima and Y. Oshima, 1991. The PHO84 gene of Saccharomyces cerevisiae encodes an inorganic phosphate transporter. Mol. Cell. Biol. 11: 3229–3238. [PMC free article] [PubMed]
  • Cade, R. M., and B. Errede, 1994. MOT2 encodes a negative regulator of gene expression that affects basal expression of pheromone-responsive genes in Saccharomyces cerevisiae. Mol. Cell. Biol. 14: 3139–3149. [PMC free article] [PubMed]
  • Coche, T., D. Prozzi, M. Legrain, F. Hilger and J. Vandenhaute, 1990. Nucleotide sequence of the PHO81 gene involved in the regulation of the repressible acid phosphatase gene in Saccharomyces cerevisiae. Nucleic Acids Res. 18: 2176. [PMC free article] [PubMed]
  • Dombek, K. M., V. Voronkova, A. Raney and E. T. Young, 1999. Functional analysis of the yeast Glc7-binding protein Reg1 identifies a protein phosphatase type 1-binding motif as essential for repression of ADH2 expression. Mol. Cell. Biol. 19: 6029–6040. [PMC free article] [PubMed]
  • Dubois, E., B. Scherens, F. Vierendeels, M. M. Ho, F. Messenguy et al., 2002. In Saccharomyces cerevisiae, the inositol polyphosphate kinase activity of Kcs1p is required for resistance to salt stress, cell wall integrity, and vacuolar morphogenesis. J. Biol. Chem. 277: 23755–23763. [PubMed]
  • El Alami, M., F. Messenguy, B. Scherens and E. Dubois, 2003. Arg82p is a bifunctional protein whose inositol polyphosphate kinase activity is essential for nitrogen and PHO gene expression but not for Mcm1p chaperoning in yeast. Mol. Microbiol. 49: 457–468. [PubMed]
  • Gancedo, J. M., 1998. Yeast carbon catabolite repression. Microbiol. Mol. Biol. Rev. 62: 334–361. [PMC free article] [PubMed]
  • Giots, F., M. C. Donaton and J. M. Thevelein, 2003. Inorganic phosphate is sensed by specific phosphate carriers and acts in concert with glucose as a nutrient signal for activation of the protein kinase A pathway in the yeast Saccharomyces cerevisiae. Mol. Microbiol. 47: 1163–1181. [PubMed]
  • Gregory, P. D., A. Schmid, M. Zavari, L. Lui, S. L. Berger et al., 1998. Absence of Gcn5 HAT activity defines a novel state in the opening of chromatin at the PHO5 promoter in yeast. Mol. Cell 1: 495–505. [PubMed]
  • Guthrie, C., and G. R. Fink, 1991 Guide to Yeast Genetics and Molecular Biology. Academic Press, San Diego.
  • Ho, Y., A. Gruhler, A. Heilbut, G. D. Bader, L. Moore et al., 2002. Systematic identification of protein complexes in Saccharomyces cerevisiae by mass spectrometry. Nature 415: 180–183. [PubMed]
  • Huang, S., D. A. Jeffery, M. D. Anthony and E. K. O'Shea, 2001. Functional analysis of the cyclin-dependent kinase inhibitor Pho81 identifies a novel inhibitory domain. Mol. Cell. Biol. 21: 6695–6705. [PMC free article] [PubMed]
  • Huh, W. K., J. V. Falvo, L. C. Gerke, A. S. Carroll, R. W. Howson et al., 2003. Global analysis of protein localization in budding yeast. Nature 425: 686–691. [PubMed]
  • Irie, K., K. Yamaguchi, K. Kawase and K. Matsumoto, 1994. The yeast MOT2 gene encodes a putative zinc finger protein that serves as a global negative regulator affecting expression of several categories of genes, including mating-pheromone-responsive genes. Mol. Cell. Biol. 14: 3150–3157. [PMC free article] [PubMed]
  • Kaffman, A., I. Herskowitz, R. Tjian and E. K. O'Shea, 1994. Phosphorylation of the transcription factor PHO4 by a cyclin-CDK complex, PHO80–PHO85. Science 263: 1153–1156. [PubMed]
  • Kaffman, A., N. M. Rank, E. M. O'Neill, L. S. Huang and E. K. O'Shea, 1998. The receptor Msn5 exports the phosphorylated transcription factor Pho4 out of the nucleus. Nature 396: 482–486. [PubMed]
  • Kitada, K., E. Yamaguchi and M. Arisawa, 1995. Cloning of the Candida glabrata TRP1 and HIS3 genes, and construction of their disruptant strains by sequential integrative transformation. Gene 165: 203–206. [PubMed]
  • Komeili, A., and E. K. O'Shea, 1999. Roles of phosphorylation sites in regulating activity of the transcription factor Pho4. Science 284: 977–980. [PubMed]
  • Konrad, M., 1988. Analysis and in vivo disruption of the gene coding for adenylate kinase (ADK1) in the yeast Saccharomyces cerevisiae. J. Biol. Chem. 263: 19468–19474. [PubMed]
  • Kuo, M. H., J. Zhou, P. Jambeck, M. E. Churchill and C. D. Allis, 1998. Histone acetyltransferase activity of yeast Gcn5p is required for the activation of target genes in vivo. Genes Dev. 12: 627–639. [PMC free article] [PubMed]
  • Lau, W. T., R. W. Howson, P. Malkus, R. Schekman and E. K. O'Shea, 2000. Pho86p, an endoplasmic reticulum (ER) resident protein in Saccharomyces cerevisiae, is required for ER exit of the high-affinity phosphate transporter Pho84p. Proc. Natl. Acad. Sci. USA 97: 1107–1112. [PMC free article] [PubMed]
  • Lau, W. W., K. R. Schneider and E. K. O'Shea, 1998. A genetic study of signaling processes for repression of PHO5 transcription in Saccharomyces cerevisiae. Genetics 150: 1349–1359. [PMC free article] [PubMed]
  • Leberer, E., D. Dignard, D. Harcus, M. Whiteway and D. Y. Thomas, 1994. Molecular characterization of SIG1, a Saccharomyces cerevisiae gene involved in negative regulation of G-protein-mediated signal transduction. EMBO J. 13: 3050–3064. [PMC free article] [PubMed]
  • Lecoq, K., I. Belloc, C. Desgranges and B. Daignan-Fornier, 2001. Role of adenosine kinase in Saccharomyces cerevisiae: identification of the ADO1 gene and study of the mutant phenotypes. Yeast 18: 335–342. [PubMed]
  • Lenburg, M. E., and E. K. O'Shea, 1996. Signaling phosphate starvation. Trends Biochem. Sci. 21: 383–387. [PubMed]
  • Lenssen, E., U. Oberholzer, J. Labarre, C. De Virgilio and M. A. Collart, 2002. Saccharomyces cerevisiae Ccr4-not complex contributes to the control of Msn2p-dependent transcription by the Ras/cAMP pathway. Mol. Microbiol. 43: 1023–1037. [PubMed]
  • Madden, S. L., C. L. Creasy, V. Srinivas, W. Fawcett and L. W. Bergman, 1988. Structure and expression of the PHO80 gene of Saccharomyces cerevisiae. Nucleic Acids Res. 16: 2625–2637. [PMC free article] [PubMed]
  • Maj, M., B. Singh and R. S. Gupta, 2000. The influence of inorganic phosphate on the activity of adenosine kinase. Biochim. Biophys. Acta 1476: 33–42. [PubMed]
  • Maj, M. C., B. Singh and R. S. Gupta, 2002. Pentavalent ions dependency is a conserved property of adenosine kinase from diverse sources: identification of a novel motif implicated in phosphate and magnesium ion binding and substrate inhibition. Biochemistry 41: 4059–4069. [PubMed]
  • Narlikar, G. J., H. Y. Fan and R. E. Kingston, 2002. Cooperation between complexes that regulate chromatin structure and transcription. Cell 108: 475–487. [PubMed]
  • Neef, D. W., and M. P. Kladde, 2003. Polyphosphate loss promotes SNF/SWI- and Gcn5-dependent mitotic induction of PHO5. Mol. Cell. Biol. 23: 3788–3797. [PMC free article] [PubMed]
  • Nishimura, K., K. Yasumura, K. Igarashi, S. Harashima and Y. Kakinuma, 1999. Transcription of some PHO genes in Saccharomyces cerevisiae is regulated by spt7p. Yeast 15: 1711–1717. [PubMed]
  • Ogawa, N., K.-I. Noguchi, H. Sawai, Y. Yamashita, C. Yompakdee et al., 1995. Functional domains of Pho81p, an inhibitor of Pho85p protein kinase, in the transduction pathway of Pi signals in Saccharomyces cerevisiae. Mol. Cell. Biol. 15: 997–1004. [PMC free article] [PubMed]
  • O'Neill, E. M., A. Kaffman, E. R. Jolly and E. K. O'Shea, 1996. Regulation of PHO4 nuclear localization by the PHO80–PHO85 cyclin-CDK complex. Science 271: 209–212. [PubMed]
  • Oshima, Y., 1982 Regulatory circuits for gene expression: the metabolism of galactose and phosphate, pp. 159–180 in The Molecular Biology of the Yeast Saccharomyces: Metabolism and Gene Expression, edited by J. N. Strathern, E. W. Jones and J. R. Broach. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
  • Peterson, C. L., and J. W. Tamkun, 1995. The SWI-SNF complex: A chromatin remodeling machine? Trends Biochem. Sci. 20: 143–146. [PubMed]
  • Roth, S. Y., J. M. Denu and C. D. Allis, 2001. Histone acetyltransferases. Annu. Rev. Biochem. 70: 81–120. [PubMed]
  • Saiardi, A., J. J. Caffrey, S. H. Snyder and S. B. Shears, 2000. The inositol hexakisphosphate kinase family. Catalytic flexibility and function in yeast vacuole biogenesis. J. Biol. Chem. 275: 24686–24692. [PubMed]
  • Schneider, K. R., R. L. Smith and E. K. O'Shea, 1994. Phosphate-regulated inactivation of the kinase PHO80–PHO85 by the CDK inhibitor PHO81. Science 266: 122–126. [PubMed]
  • Smith, C. L., R. Horowitz-Scherer, J. F. Flanagan, C. L. Woodcock and C. L. Peterson, 2003. Structural analysis of the yeast SWI/SNF chromatin remodeling complex. Nat. Struct. Biol. 10: 141–145. [PubMed]
  • Steger, D. J., E. S. Haswell, A. L. Miller, S. R. Wente and E. K. O'Shea, 2003. Regulation of chromatin remodeling by inositol polyphosphates. Science 299: 114–116. [PMC free article] [PubMed]
  • Tatusov, R. L., M. Y. Galperin, D. A. Natale and E. V. Koonin, 2000. The COG database: a tool for genome-scale analysis of protein functions and evolution. Nucleic Acids Res. 28: 33–36. [PMC free article] [PubMed]
  • Toh-e, A., and T. Shimauchi, 1986. Cloning and sequencing of the PHO80 gene and CEN15 of Saccharomyces cerevisiae. Yeast 2: 129–139. [PubMed]
  • Toh-e, A., Y. Ueda, S.-I. Kakimoto and Y. Oshima, 1973. Isolation and characterization of acid phosphatase mutants in Saccharomyces cerevisiae. J. Bacteriol. 113: 727–738. [PMC free article] [PubMed]
  • Toh-e, A., K. Tanaka, Y. Uesono and R. B. Wickner, 1988. PHO85, a negative regulator of the PHO system, is a homolog of the protein kinase gene, CDC28, of Saccharomyces cerevisiae. Mol. Gen. Genet. 214: 162–164. [PubMed]
  • Ueda, Y., A. Toh-e and Y. Oshima, 1975. Isolation and characterization of recessive, constitutive mutations for repressible acid phosphatase synthesis in Saccharomyces cerevisiae. J. Bacteriol. 122: 911–922. [PMC free article] [PubMed]
  • Uesono, Y., K. Tanaka and A. Toh-e, 1987. Negative regulators of the PHO system in Saccharomyces cerevisiae: isolation and structural characterization of PHO85. Nucleic Acids Res. 15: 10299–10309. [PMC free article] [PubMed]
  • Winzeler, E. A., D. D. Shoemaker, A. Astromoff, H. Liang, K. Anderson et al., 1999. Functional characterization of the S. cerevisiae genome by gene deletion and parallel analysis. Science 285: 901–906. [PubMed]
  • Wykoff, D. D., and E. K. O'Shea, 2001. Phosphate transport and sensing in Saccharomyces cerevisiae. Genetics 159: 1491–1499. [PMC free article] [PubMed]

Articles from Genetics are provided here courtesy of Genetics Society of America
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...