A Genetic Study of Signaling Processes for Repression of PHO5 Transcription in Saccharomyces cerevisiae

Corresponding author: Erin K. O'Shea, Department of Biochemistry and Biophysics, University of California, 513 Parnassus Ave., San Francisco, CA 94143-0448., oshea{at}biochem.ucsf.edu (E-mail).

Communicating editor: M. CARLSON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

In the yeast Saccharomyces cerevisiae, transcription of a secreted acid phosphatase, PHO5, is repressed in response to high concentrations of extracellular inorganic phosphate. To investigate the signal transduction pathway leading to transcriptional regulation of PHO5, we carried out a genetic selection for mutants that express PHO5 constitutively. We then screened for mutants whose phenotypes are also dependent on the function of PHO81, which encodes an inhibitor of the Pho80p-Pho85p cyclin/cyclin-dependent kinase complex. These mutations are therefore likely to impair upstream functions in the signaling pathway, and they define five complementation groups. Mutations were found in a gene encoding a plasma membrane ATPase (PMA1), in genes required for the in vivo function of the phosphate transport system (PHO84 and PHO86), in a gene involved in the fatty acid synthesis pathway (ACC1), and in a novel, nonessential gene (PHO23). These mutants can be classified into two groups: pho84, pho86, and pma1 are defective in high-affinity phosphate uptake, whereas acc1 and pho23 are not, indicating that the two groups of mutations cause constitutive expression of PHO5 by distinct mechanisms. Our observations suggest that these gene products affect different aspects of the signal transduction pathway for PHO5 repression.

ALL cells must respond appropriately to changes in their environment. When microorganisms are limited for nutrients, they respond by regulating expression of genes important for survival. In the yeast Saccharomyces cerevisiae, transcription of a gene for a secreted acid phosphatase, PHO5, is regulated in response to changes in the extracellular concentration of inorganic phosphate. PHO5 transcription is repressed in high-phosphate medium and derepressed in low-phosphate medium.

A genetic pathway for PHO5 regulation has been established 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 ). Another set of genes, including PHO80, PHO85, and PHO84, are required for PHO5 repression, and loss-of-function mutations in these genes result in constitutive expression of PHO5, even in high-phosphate medium (Pho^c phenotype; UEDA et al. 1975 ).

In recent years, progress has been made in elucidating the molecular mechanism of a signal transduction pathway leading to the transcriptional regulation of PHO5 (LENBURG and O'SHEA 1996 ). PHO4 encodes a transcription factor that is required for PHO5 expression (TOH-E et al. 1973 ). When yeast cells are grown in high-phosphate medium, Pho4p is phosphorylated by the Pho80p-Pho85p cyclin/cyclin-dependent kinase (CDK) complex (KAFFMAN et al. 1994 ). Phosphorylated Pho4p is localized predominantly to the cytoplasm (O'NEILL et al. 1996 ), and PHO5 transcription is repressed. When yeast cells are starved for phosphate, the activity of the Pho80p-Pho85p cyclin-CDK complex is inhibited by the CDK inhibitor (CKI) Pho81p (SCHNEIDER et al. 1994 ; OGAWA et al. 1995 ). Under these conditions, Pho4p is unphosphorylated and localized to the nucleus, where it activates PHO5 transcription.

In high-phosphate medium, repression of PHO5 transcription requires inactivation of the CKI activity of Pho81p. However, it is still not known where the signal to repress PHO5 transcription is generated, nor how the signal results in activation of the kinase activity of the Pho80p-Pho85p cyclin-CDK complex. Loss-of-function mutations in PHO81 are epistatic to mutations in PHO84, suggesting that PHO84 functions upstream of PHO81. Biochemical data are consistent with this model—Pho84p is a transmembrane protein required in vivo for high-affinity phosphate uptake (TAMAI et al. 1985 ; BUN-YA et al. 1991 ), and recombinant Pho84p expressed in Escherichia coli is capable of transporting phosphate in vitro when assembled into synthetic phospholipid vesicles (BERHE et al. 1995 ). Whether Pho84p is directly involved in the signaling that leads to PHO5 repression is currently unclear.

In an effort to understand the signaling process for PHO5 repression, we designed a genetic selection using a PHO5-HIS3 reporter to identify genes that function upstream of PHO81 and are required for PHO5 repression. Our strategy was to isolate mutants that exhibit a Pho^c phenotype that is also PHO81 dependent. Our goal was to identify components of the signal transduction pathway required to prevent inhibition of the Pho80p-Pho85p complex by Pho81p in high-phosphate medium. Loss-of-function mutations in these genes should cause accumulation of unphosphorylated Pho4p in the nucleus in high-phosphate medium as a result of inhibition of Pho80p-Pho85p kinase activity by Pho81p. We have determined that mutations in five genes (PHO84, PHO86, PMA1, ACC1, and PHO23) result in constitutive PHO5 expression in a PHO81-dependent manner. Mutations in PMA1, ACC1, and PHO23 have not been reported to confer a Pho^c phenotype. Analysis of these mutants indicates that these genes are likely to affect different aspects of the signaling pathway. Possible mechanisms for the action of these gene products in the phosphate repression signaling pathway are discussed.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Yeast strains, plasmids, media, and genetic methods:
Yeast strains used in this study are described in Table 1. Our yeast strains are derived from K699 or have been crossed into the K699 genetic background (SCHWOB and NASMYTH 1993 ). Plasmids are listed in Table 2. Standard yeast media are as described (AUSUBEL et al. 1993 ). Low-phosphate medium is phosphate-depleted medium (O'CONNELL and BAKER 1992 ). No-phosphate medium is synthetic medium consisting of yeast nitrogen base lacking inorganic phosphate. Yeast nitrogen base lacking inorganic phosphate was made with components described in the Difco manual, except that potassium phosphate was substituted with the same amount of potassium chloride. Uptake medium consists of no-phosphate medium containing 10 mM potassium acetate to buffer it to pH 4.2 or 10 mM potassium citrate to buffer it to pH 3.0. Media supplemented with fatty acids were prepared by adding 1% Tween 40 (v/v) with 0.05% (w/v) palmitic acid. Crossing, sporulation, and tetrad analysis were done by standard genetic methods (SHERMAN et al. 1978 ). Yeast cells were transformed by the lithium acetate method (ITO et al. 1983 ) or by electroporation (BECKER and GUARENTE 1991 ).

Mutagenesis and isolation of novel, PHO81-dependent Pho^c mutants:
The PHO5-HIS3 reporter fusion was constructed as follows. The PHO5 promoter was amplified using the polymerase chain reaction (PCR) from plasmid pMH313 (HAN and GRUNSTEIN 1988 ) with the following two primers: 5'-GCGGAATTCGATCCGAAAGTTGCA-3' and 5'-TTAAGCTCTAATGGTTACCCTAGGGCG-3'. The amplified product contains nucleotides -550 to -1 of the PHO5 promoter, which includes the Pho4p- and Pho2p-binding sites, as well as sites for three of four positioned nucleosomes that undergo transitions in low-phosphate medium (ALMER and HORZ 1986 ). This PCR fragment was cut with EcoRI and BamHI and subcloned into pRS314 (SIKORSKI and HIETER 1989 ) to create pRS314-PHO5. The HIS3 open reading frame and 3'-untranslated region, a total of 1100 bp, were amplified using PCR from plasmid pJJ217 (JONES and PRAKASH 1990 ) with the primers 5'-GCGGGATCCACAGAGCAGAAAGCC-3' and 5'-TCCCCGCGGATCACCACAACTAAC-3'. This amplified product was cut with BamHI and SacII and cloned into pRS314-PHO5 to create plasmid pPHO5-HIS3. Finally, nucleotides -460 to -110 of the 5'-untranslated region of HIS3 were amplified from plasmid pJJ217 by PCR with the following primers: 5'-GCGGGGCCCCTGCACGGTCCTGTT-3' and 5'-GCGGAATTCGAGTCATCCGCTAGG-3'. This amplified product was cut with ApaI and EcoRI and cloned into plasmid pPHO5-HIS3 to create plasmid pPHO5-HIS3int. To integrate this fusion construct into the yeast genome, plasmid pPHO5-HIS3int was cut with ApaI and SacII, and the liberated 1.9-kb piece was gel purified and transformed into a pho80 strain (EY0486). The net result of the targeted integration at the HIS3 locus was to delete most of the HIS3 promoter and replace it with 525 bp of the PHO5 promoter. After selection for stable transformants on SD-His, individual colonies were tested for the integration of the construct at the HIS3 locus by colony PCR using primers 5'-TTAAGCTCTAATGGTTACCCTAGGGCG-3' and 5'-TTTAGCTTCTCGACGTGGGCCTTT-3'. A 950-bp product was produced from cells that had undergone the integration event, but not from cells that contained random insertions. The resulting strain was crossed to a wild-type strain (EY0091). The heterozygous diploid was sporulated to generate the starting strain for the selection, EY0100.

A logarithmically growing culture of strain EY0100 in YPD was washed and resuspended in distilled water, mutagenized with UV to 33% survival, and plated onto SD -His with the addition of 2 mM 3-amino-1,2,4-triazole. These plates were incubated for 4 days at 25° in the dark. Colonies were picked and patched onto YPD plates. The Pho^c mutants were identified by assaying the histidine prototrophs for constitutive PHO5 expression using acid phosphatase plate assays (TOH-E et al. 1973 ) on rich medium. Recessive mutants were isolated by mating the Pho^c mutants to a wild-type strain (EY0183) and assaying the resulting diploid strains for constitutive PHO5 expression. Mutations in genes that were previously known to be required for PHO5 repression (i.e., PHO80, PHO85, or PHO84) were identified by crossing the recessive mutants to a pho80 pho85 double mutant (EY0179) or a pho84 mutant (EY0211) and analyzing the resulting diploids for the Pho^c phenotype using acid phosphatase plate assays. Complementation tests among the remaining non-pho80, non-pho85, and non-pho84 recessive Pho^c mutants were performed by mating them to each other after backcrossing. PHO5 expression was examined in these diploids by acid phosphatase plate assays.

To test whether these novel recessive mutants are PHO81 dependent, we crossed one mutant from each group to EY0150 (pho81::TRP1). The resulting diploid cells were sporulated, and the double-mutant haploid cells were identified from the nonparental ditype tetrads by examining PHO5 expression in low-phosphate medium.

Plasmid cloning:
Each novel recessive PHO81-dependent Pho^c mutant was backcrossed at least three times to a wild-type strain. The Pho^c phenotype segregated 2:2 for each class, suggesting that the mutation resides in a single gene. The wild-type versions of the genes containing the novel PHO81-dependent Pho^c mutations were cloned by screening a centromere-based (YCp50) yeast genomic library (a gift from A. Murray and K. Hardwick) for plasmids that were able to complement the Pho^c mutant phenotype in acid phosphatase plate assays. Because we found that the lithium acetate transformation method was somewhat mutagenic and that our PHO81-dependent Pho^c mutants picked up suppressors or reverted at a fairly high rate, we transformed the DNA library primarily by electroporation. Transformants were streaked to isolate single colonies, which were patched and tested for plasmid dependence on 5-FOA plates. Plasmids were then isolated from these transformants and retransformed into the original mutant strains to test complementation.

Phosphate uptake assays:
Phosphate uptake assays were performed with a modification of a previously described procedure (TAMAI et al. 1985 ). Phosphate uptake was measured on whole cells. Yeast strains grown to log phase in synthetic high-phosphate medium were inoculated into no-phosphate medium at an optical density at 600 nm (OD₆₀₀) of 0.1–0.2 and grown for 4–6 hr at 30°. Cells were washed and resuspended in uptake medium at pH 4.5 or pH 3.0 to a final OD₆₀₀ of ~0.5–0.8, and they were incubated at 30° for 20 min before uptake experiments were performed. The final external inorganic phosphate concentration was adjusted to 10 µM, and an appropriate amount of H₃³²PO₄ (Dupont NEN Research Products) was added as a radioactive label. The labeled substrate was added to prewarmed cells, and samples were withdrawn at different time intervals. Samples were filtered immediately through 0.8-µm nitrocellulose membrane filters (Millipore, Bedford, MA) and washed twice with 5 ml of 100 mM KH₂PO₄. The amount of phosphate taken up by the yeast cells was determined by scintillation counting.

Quantitative studies of PHO5-GFP expression:
Log-phase yeast strains harboring pPHO5-GFP (EB0180) were inoculated into SD -Ura or SD low-phosphate -Ura medium. The pPHO5-GFP plasmid was made by fusing the PHO5 promoter (nucleotides -550 to -1 upstream of the ATG) in-frame to the N terminus of green fluorescent protein (GFP; CHALFIE et al. 1994 ) cloned into the URA3-marked vector pRS316 (SIKORSKI and HIETER 1989 ). Emission at 510 nm was measured after excitation at 395 nm of whole cells using a fluorometer (Photon Technology International, Inc., Monmouth Junction, NJ). The final emission was normalized to the cell density as measured by OD₆₀₀. Background fluorescence was determined by using wild-type cells transformed with pRS316.

Pho4-GFP localization:
pACPHO4-GFP (EB0347) consists of the PHO4 promoter (nucleotides -323 to -1 upstream of the ATG) and the entire PHO4 coding region fused in frame to the N terminus of a GFP mutant (S65T; HEIM et al. 1995 ) cloned into the URA3-marked vector pRS316. pACPHO4-GFP was used to transform different yeast strains. Using a BX60 microscope (Olympus Corp., Lake Success, NY), localization of the fusion protein was examined by direct fluorescence of live yeast grown in SD -Ura.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

A genetic selection to isolate mutants that express PHO5 constitutively:
To identify novel factors that signal PHO5 repression, we designed a selection to isolate mutants that express PHO5 constitutively (Figure 1). Our starting strain for this selection was his3::PHO5-HIS3 (EY0100), in which the promoter of the HIS3 gene was replaced by the PHO5 promoter (see MATERIALS AND METHODS). We demonstrated that HIS3 gene expression is subject to the same regulatory control as the PHO5 gene in three ways. First, in high-phosphate medium, this strain was a histidine auxotroph, whereas in low-phosphate medium, it was a histidine prototroph. Second, in high-phosphate medium, a his3::PHO5-HIS3 pho80 strain was a histidine prototroph. Third, a his3::PHO5-HIS3 pho2 strain was a histidine auxotroph (data not shown). Thus, the ability to grow in the absence of histidine reflects PHO5 expression in the strain his3::PHO5-HIS3.

View larger version (27K): In this window In a new window Download PPT slide   Figure 1. Selection strategy for novel PHO81-dependent Phoc mutants. The starting strain for the selection was his3::pPHO5-HIS3 (EY0100), in which the promoter of the HIS3 gene was replaced by the PHO5 promoter; it is a histidine auxotroph in high-phosphate media. The mutant screen included two parts. In the first step, histidine prototrophs were selected on phosphate-rich media lacking histidine. In the second step, these mutants were screened for PHO81-dependence.

To carry out the selection for Pho^c mutants, we plated approximately one million live, mutagenized yeast cells on SD -His plates containing aminotriazole. A total of 420 colonies of heterogenous sizes grew, and 167 of 420 conferred unambiguously the constitutive PHO5 expression phenotype. The Pho^c phenotype was scored by the acid phosphatase assay in which a chromagenic phosphatase substrate, -naphthylphosphate, was applied in an agar overlay; yeast strains expressing PHO5 turn red by these assays (TOH-E et al. 1973 ). We determined that 81 mutants contained dominant Pho^c mutations. Among the remaining 86 recessive mutants, we recovered 24 alleles of either PHO80 or PHO85, 44 alleles of PHO84, and 18 alleles of mutations in genes not previously identified in the screen of UEDA et al. 1975 . These 18 novel Pho^c mutants fell into eight complementation groups. Mutants in four of these complementation groups also had a slow growth phenotype at 16°. Genetic analysis has revealed that the mutation that causes the cold-sensitive (cs) phenotype is tightly linked to the Pho^c mutation in each case.

Four novel recessive Pho^c mutant groups are PHO81 dependent:
To distinguish genes involved in the signaling process from those that affect other aspects of PHO5 regulation (e.g., transcriptional repression), we analyzed the epistatic relationship between pho81 and the novel Pho^c mutant classes. We reasoned that a pho81 mutant should be epistatic to mutants defective in the signaling process (see also DISCUSSION); in these mutant strains, deletion of the PHO81 gene should result in Pho4p being localized to the cytoplasm and an uninducible PHO5 expression phenotype. Double mutants of pho81 and novel Pho^c mutants were generated (see MATERIALS AND METHODS) and examined for PHO5 expression and Pho4-GFP localization. Four complementation groups showed a PHO81 dependence for PHO5 expression and Pho4-GFP localization. Each of the four groups contains only one mutant allele, namely wk29, wk153, wk196, and wk383. Mutants wk29, wk153, and wk196 also have a slow growth phenotype at 16°.

We measured PHO5 expression in these four PHO81-dependent mutants in high-phosphate medium in two ways. We first used acid phosphatase plate assays (TOH et al. 1973 ) and the color intensity of different Pho^c mutant strains as indicators of the level of PHO5 activity (Figure 2, A and C). Then, to better quantitate PHO5 expression, we measured PHO5 promoter activity by using a pPHO5-GFP construct in which the PHO5 promoter was fused to the GFP (EB0180; Figure 2, B and D). The strength of the PHO5 promoter was measured by GFP fluorescence at 510 nm (see MATERIALS AND METHODS).

View larger version (259K): In this window In a new window Download PPT slide   Figure 2. Analysis of PHO5 expression in novel, PHO81-dependent Phoc mutants in the presence (A and B) and absence (C and D) of PHO81. (A and C) Endogenous PHO5 expression was determined by acid phosphatase plate assays in high- (left column) and low- (right column) phosphate synthetic media. (B and D) Quantitation of the expression of PHO5 in high- and low-phosphate media, respectively. Expression of GFP driven by the PHO5 promoter in various strains was determined by GFP emission at 510 nm. Data are reported as the mean values of three independent experiments, and the error bars indicate the standard deviation.

Results from these two methodologies show strong correlation; mutants with a stronger Pho^c phenotype (darker color) have a higher fluorescent emission at 510 nm. Figure 2, A and B, demonstrates that the four novel PHO81-dependent mutants express PHO5 constitutively, but to different degrees. Among the four PHO81-dependent complementation groups harboring pPHO5-GFP, wk153, which confers the strongest Pho^c phenotype, emits a GFP fluorescence at 510 nm that is approximately five- to eightfold above that of a wild-type strain. wk196 and wk29 are of an intermediate Pho^c phenotype and emit GFP fluorescence approximately threefold above wild type, whereas wk383, which confers the weakest Pho^c phenotype, emits GFP fluorescence approximately twofold above wild type (Figure 2B).

To examine PHO5 expression in double mutants of pho81 and the novel recessive Pho^c mutants, we performed acid phosphatase plate assays and GFP fluorescence assays in low-phosphate medium, as shown in Figure 2, C and D, respectively. Only the results for the PHO81-dependent mutants are shown. A pho80 pho81 double-mutant strain was included as a control for PHO5 expression in low-phosphate medium because PHO80 is a gene known to function downstream of PHO81. Similarly, a pho84 pho81 double-mutant strain was also included as a control because pho81 is epistatic to pho84. The four mutant groups show a phenotype similar to that of pho84 pho81, indicating that pho81 is epistatic to these novel Pho^c mutants, and that they are likely to define genes that function upstream of PHO81. The four complementation groups whose Pho^c phenotype is independent of PHO81 were not analyzed further in this study.

We next used the localization of Pho4-GFP in high-phosphate medium to monitor the PHO81 dependence of our Pho^c mutants. The localization of Pho4p has been characterized previously (O'NEILL et al. 1996 ). In both wild-type and pho81 strains, Pho4-GFP is localized to the cytoplasm in high-phosphate medium (Figure 3). In mutants that express PHO5 constitutively, such as pho84, Pho4-GFP is localized exclusively to the nucleus. However, in the double mutant pho84 pho81, Pho4-GFP is again localized to the cytoplasm because pho81 is epistatic to pho84. In contrast, in a pho80 pho81 strain, Pho4-GFP is localized to the nucleus because pho80 is epistatic to pho81. Thus, the epistatic relationship between the Pho^c mutants and pho81 can be studied easily using this assay. In the mutant wk153, Pho4-GFP is localized exclusively to the nucleus, while in wk29, Pho4-GFP is localized both to the nucleus and to the cytoplasm. It should be noted that this result correlates with the observation that wk153 expresses more GFP when harboring the pPHO5-GFP construct and has higher acid phosphatase activity in high-phosphate medium than wk29; this is consistent with the model in which the concentration of nuclear Pho4p modulates the level of PHO5 expression. In contrast, in the double mutants wk153 pho81 and wk29 pho81, Pho4-GFP is predominantly cytoplasmic, indicating that pho81 is epistatic to wk153 and wk29. Thus, the Pho4-GFP localization is consistent with both the results from acid phosphatase plate assays for PHO5 expression and the PHO5-GFP fluorescence assays, and it confirms that pho81 is epistatic to wk153 and wk29. The Pho4-GFP localization analysis in wk196 and wk383 is very similar to that of wk29 (data not shown).

View larger version (72K): In this window In a new window Download PPT slide   Figure 3. Studies of the subcellular localization of Pho4-GFP in mutant strains. Direct fluorescence was used to determine the localization of Pho4-GFP in live yeast cells grown in phosphate-rich media. wk29 and wk153 were renamed acc1-29 and pho86-153, respectively.

The novel PHO81-dependent Pho^c mutations represent alleles of PHO86, PMA1, ACC1, and PHO23:
During our mutant characterization, YOMPAKDEE et al. 1996 reported the isolation of a novel mutant that expresses PHO5 constitutively, pho86-1, on the basis of a screen for arsenate-resistant mutants. Pho86p is a membrane protein with two putative transmembrane domains (YOMPAKDEE et al. 1996 ) and has no homology to other proteins. On the basis of the following observations, we conclude that wk153 is an allele of PHO86. First, wk153 failed to complement the Pho^c phenotype of a pho86::TRP1 strain (EY0404). Second, after the diploid resulting from the cross of wk153 and EY0404 was sporulated, the mutation that caused the 16° slow growth phenotype segregated away from Trp⁺ spores, indicating that this mutation in wk153 was tightly linked to the PHO86 locus. Third, a low-copy-plasmid expressing PHO86 (a gift from Nobuo Ogawa) was able to complement both the PHO5 constitutive phenotype and the 16° slow growth phenotype of wk153. We therefore referred to wk153 as pho86-153.

To clone the remaining PHO81-dependent genes, we determined which genes could complement the PHO5 constitutive expression phenotype of each mutant by screening a low-copy yeast genomic library (see MATERIALS AND METHODS). Both the Pho^c and the slow growth phenotype at 16° of the wk196 mutant were complemented by two plasmids. The overlapping region of the insert fragments in these two plasmids contains two intact ORFs, PMA1 and YGL007w. Subsequent subcloning localized the complementing region to a HindIII-XhoI genomic fragment (EB0658), that contains only one complete ORF, PMA1, which encodes a p-type plasma membrane ATPase (SERRANO et al. 1986 ).

To confirm that the mutation responsible for the Pho^c phenotype in wk196 was tightly linked to the PMA1 locus, we subcloned the PstI/XhoI fragment of the PMA1 ORF into a LEU2-marked integrating vector, pRS305 (SIKORSKI and HIETER 1989 ). This plasmid (EB0660) was linearized with HindIII, a unique site within the PMA1 open reading frame, and was targeted to the genomic site via homologous recombination in a wild-type strain (EY0183). The integration of this vector at the PMA1 locus was confirmed by Southern blotting (data not shown). The resulting strain (EY0469) was crossed to a wk196 strain, and the diploid was sporulated. All 31 tetrads examined were of the parental ditype (i.e., the mutation that caused the Pho^c phenotype segregated away from the Leu⁺ spores), indicating that this mutation in wk196 was tightly linked to the PMA1 locus. On the basis of our complementation and linkage analyses, wk196 was named pma1-196.

One plasmid (EB0657) complementing the Pho^c phenotype of the wk29 mutant strain was isolated, and it contained only one intact ORF, ACC1, which encodes acetyl-CoA carboxylase. This plasmid also rescues the slow-growth phenotype of wk29 at 16°. To demonstrate that ACC1 corresponded to the wk29 locus, a HindIII fragment that contained part of the ACC1 gene was subcloned into pRS306, a URA3-marked integrating vector. The plasmid (EB0661) was linearized with NarI and targeted to the genomic site by homologous recombination. Southern blotting confirmed the integration of the vector at the ACC1 locus (data not shown). The resulting strain (EY0468) was crossed to the wk29 strain, and the diploid was sporulated. All 28 tetrads examined were of the parental ditype, demonstrating linkage of ACC1 to wk29. We then referred to wk29 as acc1-29.

The mutant phenotype of wk383 was complemented by three identical plasmids isolated from independent clones. Further subcloning localized the complementing region to a PstI-SpeI genomic fragment, which was subcloned into the PstI and SpeI sites of pRS314. The resulting plasmid (EB0659) contains only one ORF, YNL097c, which we named PHO23, and wk383 was named pho23-1. The wk383 mutation was shown to be linked to the RAS2 gene that is adjacent to YNL097c by mating wk383 to a ras2::LEU2 strain (a gift from J. Whistler and J. Rine) and analyzing the resulting tetrads.

To confirm that the YNL097c ORF corresponds to the wk383 locus, we constructed a disruption vector. An EcoRI-SphI fragment containing the HIS3 gene (from pJJ215, EB0098) was inserted into the NcoI-SphI fragment of EB659, with the EcoRI and NcoI ends blunted by treatment with Klenow. The TATA box and approximately half of the YNL097c ORF, including the ATG start codon, were deleted in the resulting plasmid (EB0662). EB0662 was then cut with EcoRI and NotI and transformed into a diploid wild-type cell so that one of the endogenous YNL097c copies in the yeast genome was replaced by one-step gene replacement (ROTHSTEIN 1983 ). His⁺ transformants were selected and sporulated. All His⁺ haploid cells also exhibited a Pho^c phenotype in >20 tetrads analyzed. The disruption of YNL097c was confirmed by Southern blotting (data not shown). The resulting strain has no growth defect on either YPD or synthetic media with high or low phosphate at a variety of temperatures, indicating that PHO23 is nonessential for cell viability. The Pho^c phenotype of the pho23::HIS3 strain is similar to that of pho23-1.

PHO23 encodes a putative 330-amino-acid polypeptide, and the hydropathy profile of this protein suggests that it has no transmembrane domains. It has weak homology with two hypothetical yeast proteins, Yhr090p and Ymr075p, according to a search of the yeast genome database using the BLAST program (ALTSCHUL et al. 1990 ). Moreover, Pho23p is similar to a human protein p33ING1, a tumor suppressor functioning in the p53-signaling pathway (GARKAVTSEV et al. 1998 ). Interestingly, Pho23p shares significant similarity with p33ING1 at the C terminus, with 61% identity over 52 residues using the BLAST program.

Enhanced Pho^c phenotype in double mutants between acc1, pma1, and pho23:
To test whether the ACC1, PMA1, and PHO23 genes function in a common pathway in PHO5 regulation, we examined the constitutive PHO5 expression phenotype in the double mutants acc1-29 pma1-196 (EY0464), acc1-29 pho23 (EY0465), and pma1-196 pho23 (EY0466). We examined the expression of PHO5 in high-phosphate medium using acid phosphatase plate assays. Each of the double mutants displayed a stronger Pho^c phenotype than either parent strain (data not shown). Normally, the synergy in phenotypes in a double mutant resulting from two total loss-of-function mutations implies that the two genes do not function in a linear pathway. However, in our case, the interpretation is complicated because ACC1 and PMA1 are essential genes, and we isolated only partial loss-of-function alleles of these two genes. Thus, the enhanced Pho^c phenotype in these double mutants is either because these novel Pho^c genes do not function in a linear pathway signaling PHO5 repression, or because acc1-29 or pma1-196 are only partial loss-of-function alleles.

Phosphate uptake analysis defines two classes of mutants:
Because PHO84 encodes a high-affinity phosphate transporter and PHO86 is implicated to function in the phosphate transport system (YOMPAKDEE et al. 1996 ), we performed phosphate uptake assays to test whether the remaining genes that we identified are also involved in the high-affinity phosphate transport system. Figure 4 documents the results of phosphate uptake experiments. The pho84, pho86, and pma1 strains have a clear defect in phosphate uptake. This result is consistent with the function of Pma1p as a proton pump, as well as the fact that pH is able to affect phosphate uptake kinetics in yeast (NIEUWENHUIS and BORST-PAUWELS 1984 ; TAMAI et al. 1985 ). In contrast, acc1-29 and pho23-1 are able to take up phosphate as well as a wild-type strain. Moreover, the acc1-29 pho23 (EY0465) double mutant does not have a phosphate uptake defect (data not shown), confirming that neither ACC1 nor PHO23 are involved in high-affinity phosphate transport.

View larger version (17K): In this window In a new window Download PPT slide   Figure 4. Analysis of phosphate uptake in different mutants. Phosphate uptake was measured as described in MATERIALS AND METHODS. Uptake rates (nmol min-1 ml-1 OD600-1) were determined at a phosphate concentration of 10 µM. , EY0183 (wild-type strain), 2.17 ± 0.073; +, EY0452 (acc1-29), 1.88 ± 0.044; , EY0454 (pma1-196), 0.11 ± 0.008; x, EY0455 (pho23-1), 1.64 ± 0.016; , EY0453 (pho86-153), 0.03 ± 0.05; , EY0211 (pho84), not distinguishable from background noise.

The phosphate uptake defect of pma1-196 can be rescued by lowering the pH:
To determine which aspects of Pma1p function are required for constitutive PHO5 expression, we obtained several pma1 mutant strains from other labs and assayed them for PHO5 expression. The strains we obtained have mutations in residues required for ATPase activity [A608T (VAN DYCK et al. 1990 ), H285Q (WACH et al. 1996 ), and A135V (NA et al. 1993 )]; for membrane targeting [P434A and G789S (CHANG and FINK 1995 )], and for proton pumping activity [I183A (WANG et al. 1996 ) and S368F (PERLIN et al. 1989 )]. Only one of the mutants expressed PHO5 constitutively; I183A conferred a weak Pho^c phenotype. This mutant has decreased proton pumping ability, although its ATPase activity is comparable to that of the wild type (WANG et al. 1996 ), raising the possibility that an impaired proton pump causes poor phosphate uptake, which in turn results in constitutive PHO5 expression by an unknown mechanism.

To determine if the proton pumping defect in pma1-196 is indeed the reason for poor phosphate uptake, we examined if the phosphate uptake defect of pma1-196 could be rescued by providing a more acidic extracellular environment to restore the proton gradient across the plasma membrane. We measured phosphate uptake in a wild-type strain and pma1-196 at pH 4.5 and pH 3.0, as shown in Figure 5. At pH 3.0, pma1-196 could take up phosphate at about the same rate as a wild-type strain, whereas at pH 4.5, it had a phosphate uptake defect (Figure 4 and Figure 5). We noted that the wild-type strain could take up phosphate at a higher rate at pH 4.5 than at pH 3.0, consistent with the observation that the optimum pH for the high-affinity phosphate transport system is between 4 and 5 (NIEUWENHUIS and BORST-PAUWELS 1984 ; TAMAI et al. 1985 ). We were not able to investigate whether lowering the pH would rescue the Pho^c phenotype in pma1-196 because pma1-196 would not grow in low pH medium. These data suggest that the proton pumping defect in pma1-196 is the reason for its poor uptake of phosphate and the resulting Pho^c phenotype.

View larger version (39K): In this window In a new window Download PPT slide   Figure 5. The phosphate uptake defect of pma1-196 is rescued by lowering the pH. Phosphate uptake was measured as described in MATERIALS AND METHODS. Uptake rates were determined at a phosphate concentration of 10 µM. Data are reported as mean values of three independent experiments, and the error bars represent the standard deviation.

The PHO5 constitutive expression phenotype of acc1-29 can be rescued by supplementation of fatty acids:
The Pho^c phenotype of acc1 indicates a possible connection between fatty acid synthesis and PHO5 regulation. Recently, several acc1 alleles that confer novel phenotypes have been isolated, leading to the proposal that Acc1p performs a function in addition to its role as an acetyl-CoA carboxylase (SCHNEITER and KOHLWEIN 1997 ). To determine if the effect of the acc1 mutation on PHO5 regulation was allele specific, we examined PHO5 expression in various acc1 mutants in high-phosphate medium, including acc1-2150 [a temperature-sensitive fatty acid auxotroph (MISHINA et al. 1980 )], acc1-7-1 [defective in mRNA export (SCHNEITER et al. 1996 )], and acc1cs [synthetic lethal with hpr1 (GUERRA and KLEIN 1995 )]. Constitutive PHO5 expression was observed in all acc1 alleles tested using acid phosphatase plate assays, indicating that the Pho^c phenotype is not allele specific, but that it can be attributed to a general defect in acc1 mutants. To investigate if this general defect is caused by the inefficient synthesis of long chain fatty acids, we supplemented the medium with palmitic acid and examined PHO5 expression in acc1-29. The Pho^c phenotype of acc1-29 is rescued by the supplementation of palmitic acid (Figure 6), suggesting that inefficient fatty acid synthesis causes constitutive PHO5 expression in acc1 mutants.

View larger version (42K): In this window In a new window Download PPT slide   Figure 6. The PHO5 constitutive phenotype of acc1-29 can be rescued by supplementation with palmitic acid (PA). PHO5 expression was determined by acid phosphatase plate assays in the absence (left column) and presence (right column) of PA.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Isolation of mutants defective in signaling PHO5 repression:
In this study, we have described a genetic selection for mutants that express PHO5 constitutively in high-phosphate medium. Because we are interested in how the high-phosphate signal is generated and transduced, we have focused on mutants whose phenotype is dependent on the function of the CDK inhibitor Pho81p. Repression of the phosphate response pathway by a high concentration of extracellular phosphate is a complex process, consisting of successful detection and generation of the phosphate signal, proper activation of the kinase activity of the Pho80p-Pho85p cyclin-CDK complex, appropriate localization of Pho4p to the cytoplasm, and correct establishment of transcriptional repression (e.g., repressive chromatin structure). Many mutations could interfere with these processes and cause inappropriate expression of PHO5 in high-phosphate medium. Molecules that are involved in transducing a high-phosphate signal should be specifically required to prevent inhibition of the Pho80p-Pho85p complex by Pho81p; the pho81 mutant will be epistatic to mutations in these signaling molecules. We isolated five mutant classes whose phenotypes are PHO81 dependent.

Two of the five mutant classes were previously known to confer a Pho^c phenotype: PHO84 encodes a high-affinity phosphate transporter, and PHO86 is required in vivo for high-affinity phosphate uptake (YOMPAKDEE et al. 1996 ). It is unclear if PHO86 is a positive regulator of PHO84 or if it is directly involved in signal transduction.

We isolated additional mutations in genes previously unknown to be involved in PHO5 regulation. PMA1 encodes a plasma membrane ATPase, which is required for the in vivo function of several transporters, including the high-affinity phosphate transport system (VALLEJO and SERRANO 1989 ; KOTYK 1994 ). PMA1 is likely to be a positive regulator of PHO84 because Pma1p functions as a proton pump and Pho84p is a proton-phosphate symporter (BORST-PAUWELS 1981 ).

Two other classes of PHO81-dependent Pho^c mutants include ACC1 and a new gene, PHO23. Mutations in these two genes do not cause high-affinity phosphate uptake defects, suggesting that they cause the Pho^c phenotype by mechanisms distinct from those of PHO86 and PMA1. ACC1 encodes an acetyl-CoA carboxylase, which catalyzes the rate-limiting step in the de novo synthesis of fatty acids. Mutant alleles of ACC1 were initially isolated as fatty acid auxotrophs. The ACC1 gene is essential for cell viability and is evolutionarily conserved. Its role in the phosphate metabolism pathway is unclear (see next section). Null alleles of PHO23 result in only a partial Pho^c phenotype, which argues against Pho23p being a direct inhibitor of Pho81p function because hyperactive alleles of PHO81 confer a much more severe Pho^c phenotype (data not shown). Alternatively, Pho23p may play a role in inhibiting Pho81p if there exist additional genes, possibly the two homologues, with redundant function. In summary, these genes are likely to affect different aspects of the transduction of the phosphate signal.

Constitutive PHO5 expression in acc1 mutants:
In our screen for constitutive PHO5 expression mutants, we isolated an allele of ACC1. We have shown that the Pho^c phenotype is not allele specific. Moreover, the Pho^c defect can be rescued by supplementing exogenous palmitic acid (Figure 6). These results suggest that the defect in PHO5 repression in an acc1 mutant is caused by the inefficient synthesis of fatty acids, and that there may exist crosstalk between fatty acid biosynthesis and phosphate metabolism pathways. What is the mechanism of action of ACC1 in the signal transduction pathway for PHO5 repression? If ACC1 functions upstream of PHO81, it is possible that some metabolite(s) of fatty acids may serve as a second messenger to signal PHO5 repression. In acc1 mutants, this metabolite would be absent, resulting in constitutive expression of PHO5. Another possibility is that a protein involved in PHO5 regulation requires specific fatty acylation for its in vivo function. The intracellular level of the required fatty acid in acc1 mutants would be so low that fatty acylation is impaired, leading to the constitutive expression of PHO5.

Recently, some acc1 mutants that have defects in nuclear morphology have been isolated (SCHNEITER and KOHLWEIN 1997 ), leading to the speculation that ACC1 is directly involved in maintaining membrane structure. acc1-7-1 was isolated as a mutant defective in mRNA export, and further analysis revealed a defect in the nuclear envelope of the mutant yeast strain (SCHNEITER et al. 1996 ). Whether the membrane structure of organelles in acc1-29 is defective has not been investigated. The fact that the high-affinity phosphate transport system is intact in the acc1-29 mutant suggests that there is no general defect in plasma membrane structure in this mutant strain.

Is the high-affinity transport system directly involved in signaling?
Because PHO84, PHO86, and PMA1 are required for PHO5 repression, one intriguing possibility is that the high-affinity phosphate transport system is directly involved in signaling repression of PHO5 transcription. There are two models that could explain the constitutive PHO5 expression in pho84 or pho86 mutants. It is possible that in addition to their phosphate transport functions, Pho84p and Pho86p may also act as sensor proteins for the levels of extracellular phosphate and transduce a repression signal when phosphate levels are high. Alternatively, the phosphate transport function of the high-affinity phosphate uptake system may be only indirectly involved in the signaling process. In pho84 or pho86 mutants, defects in phosphate uptake might result in a low level of intracellular phosphate (or some metabolite) that serves as a messenger; a lack of this messenger may cause a defect in the production of a repression signal.

The first model, in which the phosphate transporter is also the phosphate sensor, has a precedent in bacteria. The Pho regulon of E. coli is also regulated by extracellular phosphate levels; its transcription is repressed when phosphate levels are high and is induced when phosphate levels are low (WANNER 1996 ). In the E. coli system, the phosphate signal is transduced by a two-component regulator complex consisting of phoR and phoB. Repression of the Pho regulon also requires an intact Pst phosphate transport system and a protein called PhoU. PhoR senses phosphate starvation signals and activates phoB, which is a transcription factor required for transcription of the Pho regulon (WANNER 1996 ). Mutations in the Pst transport complex that separate the transport and repression functions have been isolated, suggesting that the phosphate transporter is able to directly sense changes in the extracellular concentration of phosphate (COX et al. 1988 ).

In yeast, there are also examples of transporters serving as sensors. It has been proposed that some glucose transporters act as receptors for sensing glucose levels. Dominant gain-of-function mutations in two yeast genes, SNF3 and RGT2, which encode glucose transporters, have been isolated (OZCAN et al. 1996 ). These mutants cause induction of the expression of the glucose-regulated gene HXT2 in the absence of glucose, suggesting that the glucose signal is transmitted into the cell by glucose transporters that also act as glucose receptors that sense extracellular glucose levels (OZCAN et al. 1996 ). Interestingly, both Snf3p and Rgt2p belong to the same family of 12-transmembrane domain transporters as Pho84p (BUN-YA et al. 1991 ). Recently, an ammonium permease, Mep2p, was also proposed to be involved in a signal transduction pathway that regulates pseudohyphal growth in response to ammonium starvation (LORENZ and HEITMAN 1998 ).

There are also examples of intracellular metabolites that serve as signals to control gene expression. For example, intracellular concentrations of glutamine are able to regulate gene expression. In yeast, the expression of GLN1, which encodes a glutamine synthetase, is inactivated by an increase in the intracellular glutamine concentration (MAGASANIK 1991 ). Similarly, the expression of glnA, the structural gene for bacterial glutamine synthetase, is also inactivated by high levels of intracellular glutamine. This process depends on the regulation of another two-component complex with NR_II as the sensor and NR_I as the effector (PARKINSON 1993 ).

Previous studies on the yeast phosphate transport system suggest that in addition to the high-affinity uptake system, there exists a constitutive low-affinity phosphate transport system with a proposed K_m of 1 mM (NIEUWENHUIS and BORST-PAUWELS 1984 ; TAMAI et al. 1985 ). We supplied the medium with external phosphate 100 mM to saturate the low-affinity transport system, and we examined PHO5 expression in pho84 and pho86 strains. The Pho^c phenotype of pho84 and pho86 strains cannot be suppressed at phosphate levels 100-fold above the K_m of the low-affinity phosphate uptake system, suggesting that these mutants may be defective in transducing a repression signal downstream rather than simply in phosphate uptake. However, whether the high-affinity phosphate uptake system can directly sense changes in extracellular phosphate levels remains to be determined.

	ACKNOWLEDGMENTS

We thank D. Perlin, A. Chang, A. Goffeau, R. Schneiter, S. Kohlwein, A. Tartakoff, E. Schweizer, H. Klein, N. Ogawa, J. Whistler, and J. Rine for their generous gifts of yeast strains. We thank A. Murray and K. Hardwick for the YCp50 genomic library, N. Ogawa for the PHO86 low-copy plasmid, and J. Li for the LYS2 disruption vector. We also thank A. Yi for making the pPHO5-GFP construct. We thank the members of the O'Shea lab for their technical assistance, advice, and encouragement. We are especially grateful to M. Lenburg, L. Huang, F. Banuett, and M. Christman for their critical comments on the manuscript. This work was supported by grant GM-51377 from the National Institutes of Health (E.K.O.) and a Howard Hughes Medical Institute predoctoral fellowship (W.-T.W.L.).

Manuscript received May 1, 1998; Accepted for publication August 7, 1998.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ALMER, A. and W. HORZ, 1986  Nuclease hypersensitive regions with adjacent positioned nucleosomes mark the gene boundaries of the PHO5/PHO3 locus in yeast. EMBO J. 5:2681-2687[Medline].

ALTSCHUL, S. F., W. GISH, W. MILLER, E. W. MYERS, and D. J. LIPMAN, 1990  Basic local alignment search tool. J. Mol. Biol. 215:403-410[Medline].

AUSUBEL, F., R. BRENT, R. KINGSTON, D. MOORE, J. SEIDMAN et al., 1993 Current Protocols in Molecular Biology. John Wiley & Sons, Inc., New York.

BECKER, D. M. and L. GUARENTE, 1991  High-efficiency transformation of yeast by electroporation. Methods Enzymol. 194:182-187[Medline].

BERHE, A., U. FRIESTEDT, and B. PERSSON, 1995  Expression and purification of the high-affinity phosphate transporter of Saccharomyces cerevisiae.. Eur. J. Biochem. 227:566-572[Medline].

BORST-PAUWELS, G. W., 1981  Ion transport in yeast. Biochim. Biophys. Acta 650:88-127[Medline].

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[Abstract/Free Full Text].

CHALFIE, M., Y. TU, G. EUSKIRCHEN, W. W. WARD, and D. C. PRASHER, 1994  Green fluorescent protein as a marker for gene expression. Science 263:802-805[Abstract/Free Full Text].

CHANG, A. and G. R. FINK, 1995  Targeting of the yeast plasma membrane [H⁺]ATPase: a novel gene AST1 prevents mislocalization of mutant ATPase to the vacuole. J. Cell Biol. 128:39-49[Abstract/Free Full Text].

COX, G. B., D. WEBB, Z. J. GODOVAC, and H. ROSENBERG, 1988  Arg-220 of the PstA protein is required for phosphate transport through the phosphate-specific transport system in Escherichia coli but not for alkaline phosphatase repression. J. Bacteriol. 170:2283-2286[Abstract/Free Full Text].

GARKAVTSEV, I., I. A. GRIGORIAN, V. S. OSSOVSKAYA, M. V. CHERNOV, and P. M. CHUMAKOV et al., 1998  The candidate tumour suppressor p33ING1 cooperates with p53 in cell growth control. Nature 391:295-298[Medline].

GUERRA, C. E. and H. L. KLEIN, 1995  Mapping of the ACC1/FAS3 gene to the right arm of chromosome XIV of Saccharomyces cerevisiae.. Yeast 11:697-700[Medline].

HAN, M. and M. GRUNSTEIN, 1988  Nucleosome loss activates yeast downstream promoters in vivo.. Cell 55:1137-1145[Medline].

HEIM, R., A. B. CUBITT, and R. Y. TSIEN, 1995  Improved green fluorescence [letter]. Nature 373:663-664[Medline].

ITO, H., Y. FUKUDA, K. MURATA, and A. KIMURA, 1983  Transformation of intact yeast cells treated with alkali cations. J. Bacteriol. 153:163-168[Abstract/Free Full Text].

JONES, J. S. and L. PRAKASH, 1990  Yeast Saccharomyces cerevisiae selectable markers in pUC18 polylinkers. Yeast 6:363-366[Medline].

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[Abstract/Free Full Text].

KOTYK, A., 1994  Dependence of the kinetics of secondary active transports in yeast on H⁺-ATPase acidification. J. Membr. Biol. 138:29-35[Medline].

LENBURG, M. E. and E. K. O'SHEA, 1996  Signaling phosphate starvation. Trends Biochem. Sci. 21:383-387[Medline].

LORENZ, M. C. and J. HEITMAN, 1998  The MEP2 ammonium permease regulates pseudohyphal differentiation in Saccharomyces cerevisiae.. EMBO J. 17:1236-1247[Medline].

MAGASANIK, B., 1991 Regulation of nitrogen utilization, pp. 283–317 in The Molecular and Cellular Biology of the Yeast Saccharomyces, edited by E. W. JONES, J. R. BROACH and J. PRINGLE. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

MISHINA, M., R. ROGGENKAMP, and E. SCHWEIZER, 1980  Yeast mutants defective in acetyl-coenzyme A carboxylase and biotin: apocarboxylase ligase. Eur. J. Biochem. 111:79-87[Medline].

NA, S., D. S. PERLIN, Y. D. SETO, G. WANG, and J. E. HABER, 1993  Characterization of yeast plasma membrane H(+)-ATPase mutant pma1-A135V and its revertants. J. Biol. Chem. 268:11792-11797[Abstract/Free Full Text].

NIEUWENHUIS, B. J. and G. W. BORST-PAUWELS, 1984  Derepression of the high-affinity phosphate uptake in the yeast Saccharomyces cerevisiae.. Biochim. Biophys. Acta 770:40-46[Medline].

O'CONNELL, K. F. and R. E. BAKER, 1992  Possible cross-regulation of phosphate and sulfate metabolism in Saccharomyces cerevisiae.. Genetics 132:63-73[Abstract].

O'NEILL, E., A. KAFFMAN, E. JOLLY, and E. O'SHEA, 1996  Regulation of PHO4 nuclearlocalization by the PHO80-PHO85 cyclin-CDK complex. Science 271:209-212[Abstract].

OGAWA, N., K.-I. NOGUCHI, H. SAWAI, Y. YAMASHITA, and 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[Abstract].

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.

OZCAN, S., J. DOVER, A. G. ROSENWALD, S. WOLFL, and M. JOHNSTON, 1996  Two glucose transporters in Saccharomyces cerevisiae are glucose sensors that generate a signal for induction of gene expression. Proc. Natl. Acad. Sci. USA 93:12428-12432[Abstract/Free Full Text].

PARKINSON, J. S., 1993  Signal transduction schemes of bacteria. Cell 73:857-871[Medline].

PERLIN, D. S., S. L. HARRIS, Y. D. SETO, and J. E. HABER, 1989  Defective H(+)-ATPase of hygromycin B-resistant pma1 mutants from Saccharomyces cerevisiae. J. Biol. Chem. 264:21857-21864[Abstract/Free Full Text].

ROTHSTEIN, R. J., 1983  One-step gene disruption in yeast. Methods Enzymol. 101:202-211[Medline].

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[Abstract/Free Full Text].

SCHNEITER, R., M. HITOMI, A. S. IVESSA, E. V. FASCH, and S. D. KOHLWEIN et al., 1996  A yeast acetyl coenzyme A carboxylase mutant links very-long-chain fatty acid synthesis to the structure and function of the nuclear membrane-pore complex. Mol. Cell. Biol. 16:7161-7172[Abstract].

SCHNEITER, R. and S. D. KOHLWEIN, 1997  Organelle structure, function, and inheritance in yeast: a role for fatty acid synthesis? Cell 88:431-434[Medline].

SCHWOB, E. and K. NASMYTH, 1993  CLB5 and CLB6, a new pair of B cyclins involved in DNA replication in Saccharomyces cerevisiae.. Genes Dev. 7:1160-1175[Abstract/Free Full Text].

SERRANO, R., M. C. KIELLAND-BRANDT, and G. R. FINK, 1986  Yeast plasma membrane ATPase is essential for growth and has homology with (Na⁺ + K⁺), K⁺- and Ca²⁺-ATPases. Nature 319:689-693[Medline].

SHERMAN, F., G. FINK and C. LAWRENCE, 1978 Methods in Yeast Genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

SIKORSKI, R. S. and P. HIETER, 1989  A system of shuttle vectors and yeast host strains designed for efficient manipulation of DNA in Saccharomyces cerevisiae.. Genetics 122:19-27[Abstract/Free Full Text].

TAMAI, Y., A. TOH-E, and Y. OSHIMA, 1985  Regulation of inorganic phosphate transport systems in Saccharomyces cerevisiae.. J. Bacteriol. 164:964-968[Abstract/Free Full Text].

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[Abstract/Free Full Text].

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[Abstract/Free Full Text].

VALLEJO, C. G. and R. SERRANO, 1989  Physiology of mutants with reduced expression of plasma membrane H⁺-ATPase. Yeast 5:307-319[Medline].

VAN DYCK, L., J. H. PETRETSKI, H. WOLOSKER, J. G. RODRIGUES, and A. SCHLESSER et al., 1990  Molecular and biochemical characterization of the Dio-9-resistant pma1-1 mutation of the H(+)-ATPase from Saccharomyces cerevisiae.. Eur. J. Biochem. 194:785-790[Medline].

WACH, A., P. SUPPLY, J. P. DUFOUR, and A. GOFFEAU, 1996  Amino acid replacements at seven different histidines in the yeast plasma membrane H(+)-ATPase reveal critical positions at His285 and His701. Biochemistry 35:883-890[Medline].

WANG, G., M. J. TAMAS, M. J. HALL, A. A. PASCUAL, and D. S. PERLIN, 1996  Probing conserved regions of the cytoplasmic LOOP1 segment linking transmembrane segments 2 and 3 of the Saccharomyces cerevisiae plasma membrane H+-ATPase. J. Biol. Chem. 271:25438-25445[Abstract/Free Full Text].

WANNER, B. L., 1996 Phosphorus assimilation and control of the phosphate regulon, pp. 1357–1381 in Escherichia coli and Salmonella: Cellular and Molecular Biology, edited by F. C. NEIDHARDT. American Society for Microbiology Press, Washington, D.C.

YOMPAKDEE, C., M. BUN-YA, K. SHIKATA, N. OGAWA, and S. HARASHIMA et al., 1996  A putative new membrane protein, Pho86p, in the inorganic phosphate uptake system of Saccharomyces cerevisiae.. Gene 171:41-47[Medline].

This article has been cited by other articles:

				P. Catarecha, M. D. Segura, J. M. Franco-Zorrilla, B. Garcia-Ponce, M. Lanza, R. Solano, J. Paz-Ares, and A. Leyva A Mutant of the Arabidopsis Phosphate Transporter PHT1;1 Displays Enhanced Arsenic Accumulation PLANT CELL, March 1, 2007; 19(3): 1123 - 1133. [Abstract] [Full Text] [PDF]

				S. Wongwisansri and P. J. Laybourn Disruption of Histone Deacetylase Gene RPD3 Accelerates PHO5 Activation Kinetics through Inappropriate Pho84p Recycling Eukaryot. Cell, August 1, 2005; 4(8): 1387 - 1395. [Abstract] [Full Text] [PDF]

M. R. Thomas and E. K. O'Shea An intracellular phosphate buffer filters transient fluctuations in extracellular phosphate levels