Genetics, Vol. 160, 1423-1437, April 2002, Copyright © 2002
Protein Phosphatase Type 1 Regulates Ion Homeostasis in Saccharomyces cerevisiae
Tara Williams-Harta,
Xiaolin Wua, and
Kelly Tatchella
a Department of Biochemistry and Molecular Biology, Louisiana State University Health Sciences Center, Shreveport, Louisiana 71130
Corresponding author:
Kelly Tatchell, Louisiana State University Health Sciences Center, 1501 Kings Highway, Shreveport, LA 71130., ktatch{at}lsuhsc.edu (E-mail)
Communicating editor: M. JOHNSTON
 | ABSTRACT |
|---|
Protein phosphatase type 1 (PP1) is encoded by the essential gene GLC7 in Saccharomyces cerevisiae. glc7-109 (K259A, R260A) has a dominant, hyperglycogen defect and a recessive, ion and drug sensitivity. Surprisingly, the hyperglycogen phenotype is partially retained in null mutants of GAC1, GIP2, and PIG1, which encode potential glycogen-targeting subunits of Glc7. The R260A substitution in GLC7 is responsible for the dominant and recessive traits of glc7-109. Another mutation at this residue, glc7-R260P, confers only salt sensitivity, indicating that the glycogen and salt traits of glc7-109 are due to defects in distinct physiological pathways. The glc7-109 mutant is sensitive to cations, aminoglycosides, and alkaline pH and exhibits increased rates of L-leucine and 3,3'-dihexyloxacarbocyanine iodide uptake, but it is resistant to molar concentrations of sorbitol or KCl, indicating that it has normal osmoregulation. KCl suppresses the ion and drug sensitivities of the glc7-109 mutant. The CsCl sensitivity of this mutant is suppressed by recessive mutations in PMA1, which encodes the essential plasma membrane H+ATPase. Together, these results indicate that Glc7 regulates ion homeostasis by controlling ion transport and/or plasma membrane potential, a new role for Glc7 in budding yeast.
PROTEIN phosphatase type 1 (PP1) is a well-characterized serine/threonine protein phosphatase (SHENOLIKAR and NAIRN 1991 
; SHENOLIKAR 1994 ) with roles in a wide range of physiological processes. Multiple isoforms of mammalian PP1 have been implicated in cell cycle regulation (reviewed by SHENOLIKAR 1994 ), facilitation of CREB gene transcription, the activation of glycogen synthesis and lipid metabolism (reviewed by SHENOLIKAR 1994 ; VILLAFRANCA et al. 1996 ), neurotransmission (BROWNING et al. 1985 ; MULKEY et al. 1994 ; ALLEN et al. 1997 ; GREENGARD et al. 1998 ; MACMILLAN et al. 1999 ; SMITH et al. 1999 ), the control of Ca2+ transport in cardiac sarcoplasmic reticulum (reviewed by SHENOLIKAR and NAIRN 1991 ), and the regulation of translational initiation (HE et al. 1998 ; NOVOA et al. 2001 ). It is widely accepted that substrate specificity is dictated by targeting subunits that facilitate the association between the catalytic subunit of PP1 (PP1c) and specific substrates (EGLOFF et al. 1997 ).
In Saccharomyces cerevisiae, the single, essential gene GLC7 encodes PP1c and as in the case of mammalian PP1, Glc7 has been implicated in a wide range of processes based primarily on the diversity of GLC7 mutant phenotypes. GLC7 mutants have been described with defects in meiosis and sporulation (TU et al. 1996 ; BAKER et al. 1997 ; RAMASWAMY et al. 1998 ; BAILIS and ROEDER 2000 ), mitosis (HISAMOTO et al. 1995 ; MACKELVIE et al. 1995 ; BAKER et al. 1997 ; BLOECHER and TATCHELL 1999 ; SASSOON et al. 1999 ), glucose repression (TU and CARLSON 1995 ; BAKER et al. 1997 ), glycogen metabolism (FENG et al. 1991 ; CANNON et al. 1994 ; BAKER et al. 1997 ; RAMASWAMY et al. 1998 ), vacuole fusion (PETERS et al. 1999 ) and cell wall integrity (ANDREWS and STARK 2000 ). Although mammalian PP1 has been linked to ion transport, there has been no previous work characterizing a role for Glc7 as a regulator of ion homeostasis in S. cerevisiae.
The maintenance of ion homeostasis is an essential and highly regulated process involving numerous transport systems. These transport systems are involved in regulating cell volume, maintaining the intracellular pH and ionic composition, extracting and concentrating environmental metabolic fuels, extruding toxic substances, and generating essential ionic gradients (reviewed by ANDRE 1995 ; VAN DER REST et al. 1995 ). The maintenance of these transport systems in fungi and plants is driven by the plasma membrane H+ATPase, which is encoded by the essential gene, PMA1 in S. cerevisiae. Pma1 protein is an H+ pump that is responsible for the regulation of intracellular pH, ion transport, and membrane electrochemical potential (SERRANO 1983A ; CHANG and SLAYMAN 1991 ; SERRANO et al. 1991 , SERRANO et al. 1992 ). Evidence suggests that Pma1 is both transcriptionally and post-translationally regulated. Transcription of Pma1 is influenced by various transcription factors that are stimulated by glucose metabolism, ammonium starvation, cell growth, and cell wall integrity (CAPIEAUX et al. 1989 ; RAO and SLAYMAN 1993 ; GARCIA-ARRANZ et al. 1994 ; OZER et al. 1998 ). However, the mRNA and protein levels of Pma1 do not change significantly during activation by glucose, suggesting that in this case post-translational modulation of ATPase activity may be the primary mechanism for Pma1 regulation (ERASO et al. 1987 ).
The activity of Pma1 changes in response to extracellular glucose (PORTILLO et al. 1991 ; GARCIA-ARRANZ et al. 1994 ; VAN DER REST et al. 1995 ) and Pma1 is highly phosphorylated on multiple serine and threonine residues in vivo (KOLAROV et al. 1988 ; BERTORELLO et al. 1991 ; CHANG and SLAYMAN 1991 ; SERRANO et al. 1991 ). Two-dimensional phosphopeptide gel analysis of Pma1 in temperature-sensitive secretion (sec) mutants has revealed that phosphorylation of plasma membrane-specific sites is associated with increased ATPase activity during growth on glucose (HOLCOMB et al. 1988 ; CHANG and SLAYMAN 1991 ). Upon glucose starvation, dephosphorylation occurs concomitantly with a decrease in enzymatic activity and both are rapidly reversed by the addition of glucose (CHANG and SLAYMAN 1991 ; ESTRADA et al. 1996 ). These results suggest that reversible, site-specific phosphorylation adjusts the ATPase activity and protein stability (HASPER et al. 1999 ; GONG and CHANG 2001 ) in response to glucose.
This work was motivated by the observation that glc7-109, one of 20 charged to alanine-scanning alleles, uniquely confers a NaCl-sensitive phenotype (BAKER et al. 1997 ). Defects in ion homeostasis have been previously noted for null mutants in genes encoding the phosphatases calcineurin/PP2B (NAKAMURA et al. 1993 ; BREUDER et al. 1994 ; MENDOZA et al. 1994 ; FARCASANU et al. 1995 ) and Ppz1 (POSAS et al. 1995 ) but not for mutants in GLC7. We show here that glc7-109 mutants are sensitive to cations, alkaline pH, and aminoglycoside antibiotics. These defects can be partially ameliorated by exogenous K+ or mutations in PMA1. The mutant exhibits an increased rate of uptake of the fluorescent dye DiOC6(3) and the amino acid L-leucine. Together, these results suggest that Glc7 plays a role in the cellular response to salt stress, possibly by regulating ion transport or membrane potential.
| MATERIALS AND METHODS |
|---|
Yeast strains and media:
The yeast strains used in this study are listed in Table 1. All strains used in this study except KT1849 are congenic to strain KT1112 (MATa ura3-52 leu2 his3; STUART et al. 1994 ). The trk1::KanMX and gip2::KanMX (Research Genetics, Huntsville, AL) and cnb1::LEU2 (CYERT and THORNER 1992 ) mutations were introduced into the KT1112 background by seven serial backcrosses. All growth phenotypes shown in this study required media that contained Difco (Detroit) peptone, yeast extract, and agar. In this study, media made from constituents manufactured by Angus Buffers and Biochemicals impaired the growth of glc7-109 and suppressor mutants on salt and antibiotic media. Yeast strains were routinely incubated at either 24° or 30° in YPD (1% yeast extract, 2% Bacto-peptone, 2% dextrose) or synthetic media (0.67% yeast nitrogen base, 2% glucose, and the appropriate 0.01–0.05% amino acids). Plate media contained 2% agar.
Assays for ion tolerance were performed by liquid growth assays or spot dilution tests on agar media. Liquid growth assays were performed by diluting stationary phase cultures to 4 x 107 cells/ml with media described below. Cell concentrations were determined by measuring optical density at 600 nm (OD600) and/or by microscopic cell counting using a standard hemacytometer. Cultures were incubated in the specified media for 1–3 hr. Spot dilution tests were done by spotting three serial 10-fold dilutions of 1 x 108 cells/ml of each strain on YPD plates, synthetic media, or YPD supplemented with various concentrations of salts and drugs. Growth of strains at pH 3, 5, and 8.5 was tested in liquid YPD medium that was adjusted with 10 N KOH or 1 M HCl and buffered with 100 mM potassium phosphate or 50 mM succinate, respectively. Medium containing 0.2 mM KCl consists of a synthetic medium containing 0.5% ammonium sulfate, 2% glucose, 8 mM ammonium phosphate, 17 mM NaCl, 2 mM MgSO4, 0.2 mM CaCl2, 0.2 mM KCl, plus amino acids, vitamins, and trace elements as described (GUTHRIE and FINK 1991 ). 3,3'-Dihexyloxacarbocyanine iodide [DiOC6(3)] fluorescence was assayed by flow cytometery as described (MADRID et al. 1998 ). To assay strains in 0.4 M KCl, strains were incubated overnight in YPD and diluted into YPD containing 0.4 M KCl and incubated for 4 hr at 30° prior to flow cytometry analysis.
General methods:
Escherichia coli strains DH5
F', XL1-Blue, and HB101 were used for cloning and propagation of plasmids. Diploid cells were induced to sporulate on YPA (2% potassium acetate, 2% peptone, 1% yeast extract) agar media. Yeast transformations were performed by the lithium acetate method as described (GIETZ et al. 1992 ). Qualitative glycogen accumulation measurements were performed as described (CHESTER 1968 ). Quantitative glycogen accumulation measurements were performed as described (WU and TATCHELL 2001 ). Isolation of plasmid yeast DNA and genomic DNA was performed by a variation of the "smash and grab" protocols by Hoffman (HOFFMAN and WINSTON 1987 ) as described by AUSUBEL et al. 1989 . DNA sequence was determined using the T7 Sequenase Version 2.0 kit (Amersham, Life Sciences) according to the manufacturer's instructions.
Plasmid construction:
Plasmids used in this study are listed in Table 2. Plasmid p1407-1 is a genomic library clone in the CEN4 URA3 vector, YCp50, that contains PMA1 and a 460-bp open reading frame, YGL007. To analyze the previously described alanine-scanning allele, glc7-109 (BAKER et al. 1997 ), genomic integration of the glc7-109 (K259A, R260A) mutation was performed as described (VENTURI et al. 2000 ) using plasmid pTW2. Plasmid pXZ03 (a gift from James Haber) contains an HA epitope-tagged PMA1 gene. To create the K259A, R260A, and R260P mutations in GLC7, site-directed mutagenesis was carried out as described (WU and TATCHELL 2001 ) using unique oligonucleotide primers. Plasmid pWu361 was constructed using oligonucleotide primer K259A+ (5'-GGTTATGAATTCTTTAGTGCCAGACAATTGGTGACA-3') and the complementary primer K259A- (5'-TGTCACCAATTGTCTCGGACTAAAGAATTCATAACC-3'). Plasmid pWu363 was constructed using oligonucleotide primer R260A+ (5'-GAATTCTTTAGTAAAGCCCAATTGGTGACACTT-3') and the complementary primer R260A- (5'-AAGTGTCACCAATTGGGCTTTACTAAAGAATTC-3'). Plasmid pWu365 was constructed using oligonucleotide primer R260P+ (5'-GAATTCTTTAGTAAACCGCAATTGGTGACACTT-3') and the complementary primer R260P- (5'-AAGTGTCACCAATTGCGGTTTACTAAAGAATTC-3'). The underlined nucleotide residues represent the differences between the complementary strands of R260A and R260P oligonucleotide primers.
Isolation and characterization of glc7-109 revertants:
Three independent cultures of glc7-109 strains (KT1596 and TW25) were grown in YPD at 24° to 3 x 107 cells/ml, plated at 2 x 104 cells/ml on YPD plates containing 0.1 M CsCl, and incubated for 2 days at 24°. This concentration of CsCl completely inhibits the growth of the glc7-109 mutant. Six of 105 Cs+-tolerant revertants were characterized in detail. TW30 (pma1-s1 glc7-109), TW31 (pma1-s2 glc7-109), and TW33 (pma1-s3 glc7-109) were isolated from the first culture, TW38 (pma1-s4 glc7-109) was isolated from the second culture, and TW43 (pma1-s5glc7-109) and TW45 (pma1-s6 glc7-109) were isolated from the third culture. To assay for dominance, each revertant was mated to a glc7-109 strain and the resulting diploids were assayed for resistance to CsCl and glycogen accumulation. To determine if the reversion events were extragenic, each revertant was mated to a GLC7 strain, the resultant diploids were sporulated, and tetrad analysis was performed to assay for recovery of the glc7-109 phenotype. Linkage and complementation between the suppressors was determined by performing tetrad analysis on glc7-109 diploid strains heteroallelic for two different suppressors. On the basis of complementation and genetic linkage analysis, the six suppressors were assigned to one complementation group.
Cloning of the suppressor locus:
A pma1-s2 glc7-109 mutant strain (TW31) was transformed with a yeast genomic library in the CEN URA3 vector YCp50, and transformants were screened for growth at 37° and failure to grow on synthetic media containing 0.1 M CsCl. One transformant that was complemented for both traits was obtained from approximately 66,000 transformants. A plasmid (p1407-1) was recovered from this transformant and verified to be responsible for the complementation upon retransformation into TW31. p1407-1 was able to complement the cold sensitivity of TW60 (pma1-s1 glc7-109), TW38 (pma1-s4 glc7-109), and TW78 (pma1-s6 glc7-109). The sequence of p1407-1 was determined at the junctions between the vector and genomic insert using T7 Sequenase Version 2.0 (Amersham, Arlington Heights, IL) and primers that annealed to the YCp50 vector flanking the insert. The PMA1 gene in p1407-1 was solely responsible for complementation of the revertant phenotype, as shown by the ability of plasmid pXZ03 (a gift from James Haber), which contains only the PMA1 gene, to complement the conditional phenotypes and restore CsCl sensitivity to TW61 (pma1-s1 glc7-109), TW64 (pma1-s2 glc7-109), and TW71 (pma1-s4 glc7-109).
Genetic analysis of pma1 and calcineurin strains:
Each of the suppressor strains TW61 (pma1-s1 glc7-109), TW64 (pma1-s2 glc7-109), TW68 (pma1-s3 glc7-109), TW71 (pma1-s4 glc7-109), TW75 (pma1-s5 glc7-109), and TW79 (pma1-s6 glc7-109) were mated to strain TW82 (cnb1::LEU2) or TW83 (cnb1::LEU2). The resultant diploid strains were sporulated and the haploid segregants were assayed for conditional growth defects and growth on leu-, salt, and drug agar media described above. The mutants containing each suppressor (pma1) and calcineurin (cnb1::LEU2) were characterized.
l-Leucine uptake assays:
L-Leucine uptake assays were performed as previously described (KOTLIAR et al. 1994 ; NORBECK and BLOMBERG 1998 ) with some modifications. Briefly, the L-leucine uptake assays were performed with wild-type and glc7-109 haploid strains grown to an OD600 of 
0.5 (5 x 106 cells/ml) at 30° in synthetic media supplemented with 0.01% leucine, histidine, and uracil. A solution of L-14C-radiolabeled leucine (11.2 GBq/mmol; CEB67, Amersham) at 304 mCi/mmol was added to the cell culture to a final concentration of 0.05 mM. At 1, 2, 3, 4, and 5 min, a 750-µl sample of cells containing L-14C-leucine was withdrawn and mixed with 750 µl ice-cold, nonlabeled 0.1 M L-leucine solution to stop uptake. The cells were immediately filtered and washed with ice-cold 0.1 M L-leucine solution as described. The L-14C-radiolabeled leucine measurements represent the average of triplicate experiments (NORBECK and BLOMBERG 1998 ).
Biochemical methods:
To assay calcineurin-dependent responsive element (CDRE)-dependent gene expression, strains were transformed with pAMS366 (STATHOPOULOS-GERONTIDES et al. 1999 ) and exponentially growing transformants (1–5 x 107 cells/ml) were assayed for ß-galactosidase activity as described (MILLER 1972 ; KAISER et al. 1994 ). Yeast protein extracts were prepared as described (STUART et al. 1994 ) from cells at a density of 1 x 108 cells/ml and precipitated with 10% trichloroacetic acid as described (DUFOUR and GOFFEAU 1978 ). Pma1p levels were determined from total protein extracts as described (WITHEE et al. 1998 ) using affinity-purified rabbit polyclonal antibody generously provided by Carolyn Slayman (NAKAMOTO et al. 1991 ). The signal was visualized with enhanced chemiluminesence and a horseradish peroxidase-conjugated secondary antibody (Amersham). Pma1p levels were quantitatively measured using NIH Image with phosphoglycerate kinase as a loading control to compare Pma1p in 5, 10, and 15 µg total protein extracts from wild-type and glc7-109 cells. Protein concentrations were determined as described (BRADFORD 1976 ) with the Bio-Rad protein assay reagent (Bio-Rad Laboratories, Hercules, CA) using bovine serum albumin as standard. Membrane fractions were separated on 10% SDS-polyacrylamide gels and immunoblotted as described (AUSUBEL et al. 1989 ). Total membrane fractions were isolated and VO4-sensitive ATPase activity was assayed at pH 5.7 with 5 mM ATP as described (SERRANO 1983B ). To determine relative membrane potential in our strains, we performed flow cytometry analysis using a FACS Calibur (Becton Dickinson) as described (MADRID et al. 1998 ).
| RESULTS |
|---|
glc7-109 mutants have a recessive, salt sensitive and a dominant, hyperglycogen phenotype:
A single GLC7 mutant (glc7-109) from a group of 20 alanine-scanning mutants was uniquely sensitive to high concentrations of NaCl (BAKER et al. 1997 ). It also accumulated higher than normal levels of glycogen, as observed previously for several other alleles (BAKER et al. 1997 ; RAMASWAMY et al. 1998 ). In crosses with strains containing the chromosomal glc7-109, sensitivity to a variety of monovalent and divalent ions and two aminoglycoside antibiotics segregated with the hyperglycogen phenotype, indicating that the single mutation was responsible for both traits. In these crosses we noted that the salt and drug sensitivities were recessive to GLC7 but the glycogen hyperaccumulation trait was clearly dominant. As shown in Fig 1A, a diploid strain heterozygous for glc7-109 is as resistant to CsCl and paromomycin sulfate as the wild-type GLC7 strain but accumulates glycogen levels intermediate between the glc7-109 homozygous strain and the wild type, indicating that the hyperglycogen phenotype of glc7-109 is semidominant.
View larger version (97K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1.
The salt sensitivity and glycogen hyperaccumulation traits of the glc7-109 mutant are separable. (A) Strains were incubated on YPD medium at 24° for 2 days and then replica plated onto agar containing YPD, YPD supplemented with 1 mg/ml of paromomycin sulfate (Paro), 0.1 M CsCl, and synthetic complete medium stained with iodine (I2) as described in MATERIALS AND METHODS. The glycogen [Gly] accumulation of all strains was quantitated as described in MATERIALS AND METHODS. The strains used are wild-type haploid and diploid, KT1937 and KT1943, respectively; GLC7/glc7-109, KT1944; glc7-109/glc7-109, KT1945; K259A; and glc7-109, KT1935. (B) Strains were incubated on YPD medium at 24° for 2 days and then replica plated onto YPD, YPD supplemented with 1 mg/ml of paromomycin sulfate, 0.1 M CsCl, or 0.9 M NaCl, and YPD stained with iodine as described in MATERIALS AND METHODS. The strains used are GLC7, KT1937; glc7-109 (K259A R260A), WU322; glc7-K259A, WU341; glc7-R260A, WU318; and glc7-R260P, WU320.
|
|
To test the possibility that the dominant glycogen and recessive salt-sensitive phenotype of glc7-109 are independently caused by the two missense mutations in glc7-109 (K259A and R260A), we constructed mutants that contained the single missense mutations. As shown in Fig 1B, glc7-R260A confers a phenotype similar to that of glc7-109 while glc7-K259A confers a phenotype very similar to that of the wild type. Another missense mutant with proline substituted for Arg260 retains the salt and paromomycin sulfate sensitivities of R260A, but has lost the hyperglycogen phenotype. In fact, glycogen levels are reduced below that of the wild type in the glc7-R260P mutant (Fig 1B). The fact that the glc7-R260P strain retains only the salt-sensitive trait of glc7-109 supports the hypothesis that the salt-sensitive and glycogen traits are physiologically unrelated.
glc7-109 partially bypasses the requirement for the Glc7-regulatory subunits for glycogen accumulation:
The Glc7 holoenzyme that dephosphorylates and activates glycogen synthase is thought to contain Gac1, a targeting subunit that binds directly to both Glc7 and glycogen synthase (WU et al. 2001 ). Gac1 is rate limiting for glycogen accumulation (FRANCOIS et al. 1992 ; STUART et al. 1994 ) and mutations in GLC7 or GAC1 that prevent the association cause a reduction in glycogen levels (STUART et al. 1994 ; WU et al. 2001 ). Gac1 most likely makes multiple contacts with Glc7, as judged by the locations of amino acid substitutions in Glc7 that exhibit reduced interactions with Gac1 (BAKER et al. 1997 ; RAMASWAMY et al. 1998 ; WU et al. 2001 ). One of these contacts is a hydrophobic channel that is thought to interact with a conserved VXF-motif found in many PP1- and Glc7-targeting subunits (EGLOFF et al. 1997 ). Arg260, the residue altered in Glc7-109 that is responsible for the hyperglycogen phenotype, is near F256, a residue in the hydrophobic channel that makes contact with the phenylalanine residue of the VXF-motif (EGLOFF et al. 1997 ). A Glc7 F256A mutant has defects in binding to multiple targeting subunits (WU and TATCHELL 2001 ), supporting the hypothesis that this residue has a role in binding to VXF-containing proteins. The locations of R260 and the hydrophobic channel are highlighted in the crystal structure of PP1 (Fig 2). On the basis of this structure, we propose that R260 might partially occlude the binding of Gac1; a substitution of this residue for alanine may increase the affinity for Gac1, resulting in an increase in the concentration of the Gac1-Glc7 holoenzyme and increased glycogen levels. Consistent with this hypothesis, we detected an increased interaction between Glc7-109 and Gac1 using the two-hybrid assay (D. L. FREDERICK and K. TATCHELL, unpublished data). If such an occlusion model is responsible for the hyperglycogen phenotype of the glc7-109 mutant, then the gac1 null mutation should be epistatic to glc7-109. Contrary to expectations, glc7-109 gac1::LEU2 mutants accumulate high levels of glycogen (Fig 3). Thus, Gac1 binding is not responsible for the glc7-109 phenotype.
View larger version (84K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3.
glc7-109 and glycogen-targeting subunits. Serial dilutions of strains with the designated genotypes were incubated for 24 hr at 30° on YPD, YPD containing 0.1 M CsCl, and YPD stained with iodine. Only the patches with the lowest serial dilution were imaged on YPD and YPD + CsCl but both the lowest and second lowest dilutions were imaged for the iodine-treated samples. The strains are ascosporal clones from a cross between KT1939 (MATa gac1::LEU2 pig1::URA3) and TW321 (MAT gip2::KanMX glc7-109).
|
|
The yeast genome contains three additional genes whose products are similar to Gac1. PIG1, PIG2, and GIP2 were identified in a two-hybrid screen for Gsy2-interaction proteins (CHENG et al. 1997 ; WU et al. 2001 ) while Gip2 was also identified independently as a Glc7p-interacting protein (TU et al. 1996 ). Although null mutations in these genes were reported to have little effect on glycogen levels, we tested whether the products of several of these might be necessary for the hyperglycogen phenotype of the glc7-109 mutant. Glycogen levels were elevated in glc7-109 strains containing deletions of GAC1, PIG1, and GIP2 until stationary phase, when glycogen levels dropped below that of the wild type. As shown in Fig 3, the higher serial dilution of the gac1 pig1 gip2 glc7-109 strain stained more with iodine vapor than did the lower serial dilution, presumably reflecting a later stage in the growth phase. These results indicate that glc7-109 partially bypasses the need for Gac1, Pig1, and Gip2 for glycogen accumulation, which may have important implications for the role of Glc7 in glycogen metabolism (see DISCUSSION).
Glc7 and calcineurin act through different pathways to affect ion homeostasis:
Yeast strains lacking the ser/thr protein phosphatase calcineurin/PP2B are sensitive to high concentrations of Na+, Li+, and Mn2+ (NAKAMURA et al. 1993 ; BREUDER et al. 1994 ; MENDOZA et al. 1994 ; FARCASANU et al. 1995 ), due to their failure to induce expression of a number of membrane ATPases. The sensitivity of calcineurin mutants to Mn2+ is at least partly caused by reduced transcription of PMR1 (CUNNINGHAM and FINK 1996 ), encoding a Golgi P-type ATPase (ANTEBI and FINK 1992 ); sensitivity to Na+ and Li+ is caused by reduced transcription of PMR2/ENA1 (MENDOZA et al. 1994 ), the gene encoding the major plasma membrane Na+ATPase (RUDOLPH et al. 1989 ). Calcineurin mutants are also defective in induction of Pmc1, a Ca2+ATPase, which causes them to be resistant to media containing 200 mM CaCl2 (CUNNINGHAM and FINK 1994 ). These transcriptional defects are due to the failure of calcineurin mutants to activate the Crz1/Tcn1 transcription factor (MATHEOS et al. 1997 ; STATHOPOULOS and CYERT 1997 ).
The apparent similarity between the calcineurin and glc7-109 phenotypes prompted us to directly compare the two mutants. We used a cnb1::LEU2 null mutant for the comparison. CNB1 encodes a conserved regulatory subunit of calcineurin (KUNO et al. 1991 ; CYERT and THORNER 1992 ) that is necessary for phosphatase activity. The cnb1::LEU2 and glc7-109 mutants have qualitative and quantitative differences in ion sensitivity. The glc7-109 mutant is generally more sensitive to mono- and divalent ions. As shown in Fig 4, the glc7-109 mutant fails to grow on YPD medium containing 25 mM CsCl, whereas the cnb1::LEU2 strain exhibits no defect at this or higher concentrations tested (50 and 100 mM). glc7-109 and cnb1::LEU2 mutants are equally sensitive to 5 mM MnCl2, but the glc7-109 mutant is more sensitive to Li+ and Na+ (data not shown). Furthermore, the glc7-109 strain does not exhibit the calcium resistance observed for calcineurin mutants (Fig 4). Differences were also noted in the temperature dependence of Li+ sensitivity of glc7-109 and cnb1::LEU2 strains. The glc7-109 mutant exhibits greater sensitivity to LiCl at 24° than at 30°, but no such temperature differential was noted for the cnb1::LEU2 strain. An additional difference was the sensitivity to aminoglycoside antibiotics. The cnb1::LEU2 mutant does not exhibit growth defects on media containing 50 µg/ml hygromycin B (Fig 4) and 1 mg/ml paromomycin sulfate (data not shown) in contrast to the hypersensitivity of glc7-109 mutants. An indication that Glc7 and calcineurin act in separate pathways is the finding that the glc7-109 cnb1::LEU2 double mutant is much more sensitive to ions than the single mutants. If Glc7 and calcineurin act in the same pathway to affect ion homeostasis, then we would predict an epistatic relationship between the two mutants. Instead, we found that the ion defects conferred by the two mutations are additive or synergistic; the glc7-109 cnb1 mutant is more sensitive to Mn2+, Li+, and Cs+ than is either single mutant (Fig 4). Calcineurin exerts many of its physiological effects through the regulation of the transcription factor Tcn1/Crz1 (MATHEOS et al. 1997 ; STATHOPOULOS and CYERT 1997 ; STATHOPOULOS-GERONTIDES et al. 1999 ). As shown in Table 3, the cnb1::LEU2 strain fails to induce expression of the CRZ1-dependent Lac-Z gene. The levels of ß-galactosidase activity in the glc7-109 strain are 35–70% of those found in the wild type but they are still orders of magnitude higher than those found in the cnb1::LEU2 strain. These data indicate that the substrates of calcineurin and Glc7 that are responsible for their ion sensitivities are likely to be different.
View larger version (22K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4.
The phenotypes of glc7-109 and calcineurin-deficient (cnb1::LEU2) mutants. Serial dilutions of these strains were incubated on the indicated agar media for 2–4 days at 24°. The strains that were used are wild type (KT1112), glc7-109 (TW25), cnb1::LEU2 (TW82), and glc7-109 cnb1::LEU2 (TW90).
|
|
The glc7-109 strain exhibits increased rates of l-leucine uptake:
The glc7-109 mutant is hypersensitive to the aminoglycoside antibiotics paromomycin sulfate and hygromycin B. Since the mechanisms of action of these two antibiotics are different, it is likely that the hypersensitivity of glc7-109 strains to these drugs is caused by defects in uptake or export. Although we have not been able to distinguish between these two possibilities, we note that glc7-109 strains grow more rapidly than the wild type on synthetic media containing low concentrations of amino acids (data not shown). To confirm that this difference is due to a change in amino acid uptake we measured L-leucine uptake in glc7-109 mutants The rate of L-leucine uptake in the glc7-109 mutant was approximately twofold higher than that in the wild type (Table 3).
Many defects due to glc7-109 are K+ remedial:
To rule out the possibility that the sensitivity of glc7-109 mutants to high concentrations of ions was due to osmotic effects, we tested the growth properties of glc7-109 strains on YPD media containing 1 M KCl and 1 M sorbitol. While no growth defects were observed on media with only KCl or sorbitol, KCl partially relieved the salt and drug sensitivities of glc7-109 strains. The presence of 200 mM KCl in the medium reduces the sensitivity of glc7-109 to Cs+, Li+, hygromycin B, and paromomycin sulfate (Fig 5), and to Mn2+ and Na+ (data not shown). The K+ remedial phenotype caused by glc7-109 led us to examine its effect on growth on low concentrations of KCl, because mutants defective in the potassium transporters Trk1 and Trk2 are hypersensitive to hygromycin B and toxic cations (MULET et al. 1999 ). Growth of yeast cells on media containing low concentrations of KCl (0.2 mM) requires the activity of the high-affinity K+ transporter encoded by TRK1 (GABER et al. 1988 ). We found that a trk1 strain (TW272) failed to grow on synthetic medium containing 0.2 mM KCl whereas glc7-109 and wild-type strains grew equally well on this medium. Therefore, the ion sensitivities of the glc7-109 mutant are unlikely to be caused by a defect in high-affinity K+ transport.
View larger version (72K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5.
KCl suppresses the salt and drug sensitivities of the glc7-109 mutant. Serial dilutions of wild type (KT1112) and glc7-109 (TW25) strains were incubated for 2–4 days at 24° on media containing YPD, YPD containing 1 M KCl, 0.05 M CsCl, 0.05 M CsCl + 0.2 M KCl, 0.1 M LiCl, 0.1 M LiCl + 0.2 M KCl, 50 µg/ml hygromycin B, 50 µg/ml hygromycin B + 0.2 M KCl, 1 mg/ml paromomycin sulfate, and 1 mg/ml paromomycin sulfate + 0.2 M KCl.
|
|
Suppressors of the Cs+-sensitive phenotype of glc7-109:
We characterized six spontaneous glc7-109 revertants that grew on YPD medium containing 0.1 M CsCl, as described in MATERIALS AND METHODS. The mutations responsible for the reversion events were recessive, as shown by CsCl sensitivity of diploid strains heterozygous for suppressor mutations and homozygous for glc7-109. To determine if the mutations responsible for suppression were extragenic to GLC7, tetrad analysis was performed on diploid strains created by mating each of the six revertants to a wild-type strain. The salt-sensitive phenotype of glc7-109 was observed in approximately one-fourth of the spore clones from each cross, indicating that the mutation responsible for the reversion was unlinked to GLC7. To determine if the suppressors were in the same complementation group, diploid strains were constructed that were homozygous for glc7-109 and heterozygous for two different suppressors. All diploids were resistant to 0.1 M CsCl, indicating that all suppressors were in the same complementation group. Furthermore, tetrad analysis of meiotic progeny of these diploid strains revealed that all spore clones retained the Cs+-resistant phenotype of the parents, indicating that the suppressor loci are tightly linked.
Although these suppressors are genetically linked, they confer diverse phenotypes. While all six revertants are able to grow on media containing 0.1 M CsCl or 0.9 M NaCl, only pma1-s1, pma1-s3, and pma1-s4 suppress the Li+ defect of glc7-109 (Fig 6). In contrast, pma1-s1 and pma1-s4 strains actually grow better than the wild-type strain on 0.1 M LiCl (Fig 6). No revertant strains except those containing pma1-s2 exhibit the glc7-109 growth defect on 10 mM MnCl2 (Fig 6). Some of these revertants also have conditional growth defects in the glc7-109 genetic background. pma1-s2 and pma1-s5 strains grow more slowly than the wild-type strain at 37° while pma1-s1, pma1-s4, and pma1-s6 strains grow more slowly than the wild type at 15° (Fig 6). These six suppressor mutants were also characterized in a wild-type GLC7 background. pma1-s6 strains grow poorly on 0.1 M CsCl and 0.1 M LiCl, whereas pma1-s2 grows poorly on LiCl (Fig 7). In contrast, the pma1-s1, pma1-s3, pma1-s4, and pma1-s5 strains grow faster than the wild-type strain on LiCl (Fig 7). Most of the conditional growth defects observed in a glc7-109 background were also apparent in a GLC7+ background. For example, pma1-s1 and pma1-s4 strains grow more slowly than the wild type at 15° (Fig 7) and pma1-s2 strains grow more slowly than the wild type at 37° (data not shown).
View larger version (35K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6.
pma1 mutations suppress the Cs+ sensitivity conferred by glc7-109. Serial dilutions of pma1 mutants in a glc7-109 background were plated onto YPD medium, YPD media supplemented with the designated salts, and synthetic complete (SC) medium. The plates were incubated from 2 to 4 days. The strains that were used are wild type (KT1112), glc7-109 (TW25), pma1-s1 glc7-109 (TW30), pma1-s2 glc7-109 (TW31), pma1-s3 glc7-109, (TW33), pma1-s4 glc7-109 (TW38), pma1-s5 glc7-109 (TW43), and pma1-s6 glc7-109 (TW45).
|
|
View larger version (51K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7.
The phenotype of pma1 mutants in a GLC7 background. Serial dilutions of pma1 mutants in a GLC7+ background were plated onto YPD medium, YPD media supplemented with the designated salts, and synthetic complete (SC) medium. The plates were incubated for 2–4 days. The strains that were used are wild type (KT1112), glc7-109 (TW25), pma1-s1 (TW62), pma1-s2 (TW65), pma1-s3 (TW69), pma1-s4 (TW72), pma1-s5 (TW76), and pma1-s6 (TW80).
|
|
The glc7-109 suppressors are allelic to PMA1, encoding the essential plasma membrane H+ATPase:
The suppressor locus was cloned by complementation of the temperature sensitivity of strain TW64 (pma1-s2 glc7-109) as described in MATERIALS AND METHODS. The complementing plasmid clone (p1407-1) also complemented the cold-sensitive defects of strains TW30 (pma1-s1 glc7-109) and TW38 (pma1-s4 glc7-109). These transformants grew well at 37° and 15°, but failed to grow on 0.1 M CsCl media. Sequence analysis revealed that the genomic insert in p1407-1 contained PMA1 and an adjacent 378-bp open reading frame, YGL007. A plasmid (pXZ03) containing PMA1 but lacking YGL007 was also able to suppress the temperature sensitivity and Cs+ resistance of TW64 (pma1-s2 glc7-109), indicating that PMA1 is responsible for the complementation. To confirm that the suppressor mutations are alleles of PMA1, we mapped the distance between pma1-s4 and leu1, which lies 1 cM centromere proximal to PMA1. Tetrad analysis of a cross between KT1850 (MAT ura3-52 his3 glc7-109 pma1-s4) and KT1849 (MATa ura3-52 his3 leu1) revealed that the distance between leu1 and pma1-s4 is 1.1 cM (P:NP:T = 43:0:1), identical to the previously determined map distance (P:NP:T = 869:0:22; MCCUSKER and HABER 1988 ). Together, our linkage and complementation data indicate that all six glc7-109 suppressors are alleles of PMA1.
PMA1 mutants have been isolated that suppress the salt sensitivities of calcineurin mutants (NASS et al. 1997 ; WITHEE et al. 1998 ). To determine if our glc7-109 suppressors also suppress these defects, we crossed strains containing each of our suppressors to a cnb1 strain and assayed the meiotic products for salt sensitivity. Three of our suppressors fail to suppress the ion sensitivities of cnb1, but the other three partially suppress the Na+ and/or Li+ defects of the cnb1 strain (Table 4). The two cold-sensitive mutations (pma1-s1 and pma1-s4) that most strongly suppress glc7-109 also showed the greatest suppression of cnb1::LEU2.
Many pma1 mutants are sensitive to low pH (MCCUSKER et al. 1987 ; WITHEE et al. 1998 ). We compared the growth of the glc7-109 mutant and a cold-sensitive pma1 mutant strain, TW71 (pma1-s4 glc7-109), in liquid YPD media at pH 3, 5, and 8.5. At pH 5, the growth of the wild-type, glc7-109, and pma1-s4 glc7-109 strains were all comparable (data not shown). As expected, the suppressor strain TW71 (pma1-s4 glc7-109) grows slowly at acidic pH but unexpectedly, the glc7-109 strain grows slowly at alkaline pH (Fig 8).
View larger version (14K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8.
The growth of the glc7-109 mutant and the pma1-s4 glc7-109 mutant in YPD medium adjusted to pH 3 and 8.5. Log-phase cultures of each strain were incubated in liquid YPD media at pH 3 and 8.5 as described in MATERIALS AND METHODS. The growth of each strain was monitored by optical density at 660 nm. The strains that were used are wild type (KT1112), glc7-109 (TW25), and pma1-s4 glc7-109 (TW71).
|
|
The ability of pma1 mutations to suppress the salt-sensitive phenotype of glc7-109 led us to assay Pma1 levels and activity in glc7-109 and pma1-s4 mutants. Pma1 levels in the glc7-109 strain were 90% that of the wild type, as assayed by immunoblot analysis. Vanadate-sensitive ATPase activities, a measure of Pma1 activity, were comparable in the wild-type and glc7-109 strains (Table 3), but that of the pma1-s4 mutant is 61% that of the wild type (data not shown). Indirect immunofluorescence revealed no obvious differences between glc7-109 and wild-type strains in Pma1 localization (data not shown). Thus, we find no compelling evidence that glc7-109 directly influences the activity, abundance, or distribution of Pma1.
glc7-109 cells may be hyperpolarized:
The potassium-remedial sensitivity of glc7-109 mutants to toxic cations, high pH, and aminoglycoside antibiotics and the suppression of these defects by mutations in PMA1 are consistent with the hypothesis that the plasma membrane potential of the glc7-109 mutant is hyperpolarized. A conventional determination of membrane potential using electrophysiological methods is not possible in yeast, but relative membrane potentials can be assessed using fluorescent dyes. Although such methods suffer from complications caused by the contribution of the mitochondria to the uptake process, MADRID et al. 1998 have shown that cellular fluorescence of the cyanine dye DiOC6(3) can be determined under conditions in which mitochondria do not interfere. We have followed their procedure closely by growing strains to low density in glucose-containing medium and assaying DiOC6(3) fluorescence levels by flow cytometry. As shown in Table 5, DiOC6(3) fluorescence in the glc7-109 strain (KT1596) was 150–200% that of the wild-type strain. In contrast, DiOC6(3) fluorescence in a pma1-s4 strain, TW72, was only 30–40% that of wild type and that of the pma1-s4 glc7-109 strain (TW71) was only 50–60% of wild type. As a control, we assayed DiOC6(3) fluorescence in a wild-type strain grown in synthetic media containing a low concentration of K+, which is known to hyperpolarize the plasma membrane (MADRID et al. 1998 ). DiOC6(3) fluorescence in wild-type cells under these conditions was 450–650% of the wild-type cells grown in YPD medium (data not shown). The results of these assays suggest that many of the defects of glc7-109 strains are caused either by a hyperpolarized plasma membrane or by the activation of a nonspecific ion conductance.
| DISCUSSION |
|---|
Like many glc7 mutations, glc7-109 is very pleiotropic. The glc7-109 mutant hyperaccumulates glycogen and is sensitive to a wide range of cations, high pH, and aminoglycoside antibiotics. The analysis of the single missense mutations that make up glc7-109 indicated that the R260A substitution is responsible for all the traits conferred by glc7-109. It is likely that the hyperglycogen trait and the ion and drug hypersensitivity traits reflect the influence of Glc7 on at least two separate physiological processes. This is because the glycogen phenotype of glc7-109 is dominant, whereas the other traits are recessive, and because glc7-R260P confers the ion-related defects but not the hyperglycogen phenotype. Yeast PP1 has been associated with glycogen metabolism for years; its acronym was derived from the glycogen deficiency of glc7-1 (CANNON et al. 1994 ). However, glc7-109 is the first allele of glc7 that to our knowledge has been linked to ion homeostasis.
glc7-109 and glycogen:
In yeast, as in mammals, PP1 is thought to dephosphorylate and activate glycogen synthase. PP1 activity toward glycogen synthase is thought to be regulated by targeting subunits that tether PP1c to glycogen synthase. Abundant evidence indicates that Gac1 plays this role in yeast. gac1 null mutants accumulate low levels of glycogen and glycogen synthase in these strains remains in a phosphorylated, inactive form (FRANCOIS et al. 1992 ). The product of glc7-1, the original glycogen-deficient allele of GLC7, fails to interact with Gac1 (STUART et al. 1994 ). Gac1 contains separate domains required for binding Glc7 and glycogen synthase, both of which are required for functional activity (WU et al. 2001 ). Mutations in the glycogen synthase kinase, Pho85/Pcl10, suppress the glycogen deficiency of glc7-1 and gac1 strains (HUANG et al. 1996 , HUANG et al. 1998 ; TIMBLIN et al. 1996 ). Furthermore, mutations in the major isoform of glycogen synthase, Gsy2, that cannot be phosphorylated also suppress the glycogen deficiencies of glc7-1 and gac1 strains (HARDY and ROACH 1993 ; ANDERSON and TATCHELL 2001 ). Together, these data suggest that Gac1 acts as a molecular scaffold to tether Glc7 and Gsy2. We were therefore surprised to discover that the hyperglycogen trait of glc7-109 is epistatic to a gac1 null mutation. Glycogen levels remain high in strains disrupted for GAC1 and two GAC1-related genes, PIG1 and GIP2.
One explanation for these results is that additional glycogen-targeting subunits can substitute for the loss of Gac1, Pig1, and Gip2. Pig2 is a candidate for such a redundant subunit. It is most similar in sequence to Gip2 and although loss-of-function mutations in PIG2 have not been found to alter glycogen levels, it is possible that Glc7-109 could recruit Pig2 in the absence of other targeting subunits. Another possibility is that the high levels of glycogen in glc7-109 strains are partially due to changes in the activity of enzymes other than glycogen synthase. We have not assayed activity levels of glycogen synthase in our strains and we cannot rule out the possibility that the degradative pathway is altered in glc7-109 strains. It is worth noting that Pho85, the kinase that phosphorylates glycogen synthase, also regulates the activity of glycogen phosphorylase through cyclins Pcl6 and Pcl7 (WANG et al. 2001 ). Furthermore, a large-scale analysis of protein complexes has found that Gip2 and Glc7 associate with glycogen phosphorylase (HO et al. 2002 ), suggesting that Glc7, possibly through one or more of its targeting subunits, may regulate glycogen phosphorylase as well as glycogen synthase. A third possibility is that the Glc7-109 mutant protein no longer requires glycogen-targeting subunits to increase its Vmax toward glycogen synthase, assuming that this is a role of glycogen-targeting subunits in yeast. There is direct biochemical evidence for this effect with a PP1-targeting subunit in smooth muscle (TANAKA et al. 1998 ) and some genetic evidence for such a role in yeast. Small fragments of Gac1 that contain only the Glc7-binding domain partially complement the glycogen deficiency of a gac1 null mutant when expressed at high levels (WU et al. 2001 ). Clearly, further genetic and biochemical studies are necessary to sort out the complex roles of Glc7 in glycogen metabolism.
Glc7 and ion homeostasis:
The pleiotropic salt- and drug-sensitive phenotype conferred by glc7-109 can be accounted for by at least two broad mechanisms. One possible explanation is that the glc7-109 mutant has a hyperpolarized plasma membrane potential. In Neurospora crassa, in which membrane voltage measurements can be performed (SLAYMAN 1965A , SLAYMAN 1965B ), the plasma membrane H+ATPase is largely responsible for maintaining the membrane voltage (SLAYMAN et al. 1973 ; GRADMANN et al. 1978 ). In S. cerevisiae, mutants in the gene encoding the plasma membrane H+ATPase, PMA1, which are inferred to have a reduced membrane potential (PERLIN et al. 1988 ; SETO-YOUNG and PERLIN 1991 ), have defects that are in general opposite to that of glc7-109 mutants, exhibiting increased resistance to drugs and toxic cations (MCCUSKER et al. 1987 ). In contrast, yeast cells with defects in the plasma membrane K+ transporters encoded by TRK1 and TRK2 have been inferred to have an increased membrane potential (MADRID et al. 1998 ) and have a phenotype that is similar, but not identical, to that of the glc7-109 mutant (VIDAL et al. 1990 ; MADRID et al. 1998 ; MULET et al. 1999 ). Direct measurements of membrane potential in S. cerevisiae are not possible for technical reasons and indirect determinations using fluorescent dyes are subject to artifacts (BALLARIN-DENTI et al. 1994 ). Nevertheless, our analysis of DiOC6(3) fluorescence is consistent with hyperpolarization of the plasma membrane in glc7-109 cells. The isolation of mutations in PMA1 as suppressors of the CsCl sensitivity of glc7-109 mutants further supports this hypothesis.
How could Glc7 act to influence membrane potential? The two most obvious possibilities are via Pma1 and the potassium transporters, Trk1 and Trk2. Pma1 is essential for maintenance of the plasma membrane potential and its activity is regulated in response to a range of external factors. Pma1 is phosphorylated at multiple sites and at least two classes of protein kinases may regulate its activity (KOLAROV et al. 1988 ; BERTORELLO et al. 1991 ; SERRANO et al. 1991 ). The protein kinase Ptk2 is a positive regulator of Pma1 and appears to act through the activating phosphorylation site at Ser-899 (GOOSSENS et al. 2000 ). In contrast, the Yck1 and Yck2 kinases may act as negative regulators of Pma1 activity (ESTRADA et al. 1996 ). Thus, Pma1 is a likely candidate to be regulated by Glc7. However, we have not found a direct involvement of Glc7 in Pma1 regulation. Pma1 levels and Pma1-dependent H+ATPase activities are similar in glc7-109 and GLC7 strains, but we cannot exclude the possibility that glc7-109 exerts a subtle effect on Pma1 activity or that Glc7 regulates Pma1 in response to specific environmental stresses.
The potassium transporters Trk1 and Trk2 are also candidates for Glc7 regulation. Like glc7-109, trk1 and trk1 trk2 double mutants exhibit high levels of DiOC6(3) fluorescence (MADRID et al. 1998 ) and share traits in common with glc7-109, including sensitivity to toxic cations and hygromycin B (MULET et al. 1999 ). However, unlike mutants lacking the potassium transporters, the glc7-109 mutant exhibits no obvious growth defect on media containing low concentrations of potassium. glc7-109 also confers sensitivity to alkaline pH as opposed to the acidic pH sensitivity of trk1 mutants. Surprisingly, K+ transport and regulation of membrane potential may be independent properties of Trk1 and Trk2. MADRID et al. 1998 have found that trk1 trk2 mutants expressing a K+ transporter (HAK1) from Schwanniomyces occidentalis that is unrelated in structure to Trk1 or Trk2 exhibit normal levels of K+ transport but retain the hyperpolarization of the trk1 trk2 mutant. In contrast, expression of a plant K+ transporter related to Trk1 suppresses both the K+ transport and membrane potential defects of trk1 trk2 mutants. Using patch clamping on yeast spheroplasts, BIHLER et al. 1999 have observed an inward, H+-dependent current that is dependent upon Trk2. Surprisingly, this current is independent of K+ and is also independent of Trk1. A possible scenario arising from these studies is that the K+ transporters may have an auxiliary function unrelated to K+ transport, which strongly regulates membrane potential. Two related protein kinases, Hal4 and Hal5, may modulate Trk1 and Trk2 activity (MULET et al. 1999 ). Yeast cells lacking Hal4 and Hal5 have a phenotype similar to that of trk1 trk2 strains, including hypersensitivity to toxic cations and hygromycin B and a deficiency in K+ uptake. In contrast, increased expression of Hal4 or Hal5 results in resistance to Li+ and Na+ (MULET et al. 1999 ). These results suggest that Hal4 and Hal5 are positive regulators of Trk1 and Trk2. However, if Glc7 regulates K+ transport, it probably does not do so by simply opposing the kinase activity of Hal4 and Hal5 since glc7-109 confers a similar phenotype to that of a hal4 hal5 null mutant.
An alternative explanation to hyperpolarization for the pleiotropic drug- and ion-hypersensitive phenotype of the glc7-109 mutant is that glc7-109 results in the activation of a nonspecific uptake mechanism or conductance. In this scenario, a reduction of Pma1 activity or the application of exogenous K+ could suppress the drug and ion toxicity of the glc7-109 mutation through indirect effects. In this regard, we note that BIHLER et al. 1998 have identified a nonspecific cation channel in S. cerevisiae by patch-clamping yeast spheroplasts. This conductance, termed nonselective cation channel 1 (NSC1), carries a large inward current and has low cation specificity. NSC1 is independent of Trk1, Trk2, and the potassium channel Duk1/Tok1, but the gene or genes encoding this channel have not been identified. It is possible that NSC1 remains open in glc7-109 mutants, thus causing an increased toxicity to cations. Of course, we cannot exclude the possibility that Glc7 has multiple targets that influence ion homeostasis. Many GLC7 mutants are pleiotropic and it is not unreasonable to propose that multiple substrates are affected in glc7-109 strains. Further genetic analysis may allow us to identify these targets.
| ACKNOWLEDGMENTS |
|---|
We thank Carolyn Slayman and Ken Allen for providing antibody to Pma1 and for helpful advice concerning Pma1 activity assays. We acknowledge Martha Cyert, James Haber, and Peter Roach for kindly providing strains and plasmids. We thank Andrew Bloecher, Lucy Robinson, Heather Panek, Clifford Slayman, and Guglielmo M. Venturi for helpful discussions. We gratefully acknowledge Clifford Slayman and Lucy Robinson for critically reading this manuscript. This work was funded by National Institutes of Health research grant GM-47789.
Manuscript received September 18, 2001; Accepted for publication January 25, 2002.
| LITERATURE CITED |
|---|
ALLEN, P. B., C. C. OUIMET, and P. GREENGARD, 1997 Spinophilin, a novel protein phosphatase 1 binding protein localized to dendritic spines. Proc. Natl. Acad. Sci. USA 94:9956-9961[Abstract/Free Full Text].
ANDERSON, C. and K. TATCHELL, 2001 Hyperactive glycogen synthase mutants of Saccharomyces cerevisiae suppress the glc7-1 protein phosphatase mutant. J. Bacteriol. 183:821-829[Abstract/Free Full Text].
ANDRE, B., 1995 An overview of membrane transport proteins in Saccharomyces cerevisiae.. Yeast 11:1575-1611[Medline].
ANDREWS, P. D. and M. J. STARK, 2000 Type 1 protein phosphatase is required for maintenance of cell wall integrity, morphogenesis and cell cycle progression in Saccharomyces cerevisiae.. J. Cell Sci. 113:507-520[Abstract].
ANTEBI, A. and G. R. FINK, 1992 The yeast Ca(2+)-ATPase homologue, PMR1, is required for normal Golgi function and localizes in a novel Golgi-like distribution. Mol. Biol. Cell 3:633-654[Abstract].
AUSUBEL, F. M., R. BRENT, R. E. KINGSTON, D. D. MOORE, J. G. SEIDMAN et al., 1989 Current Protocols of Molecular Biology. John Wiley & Sons, New York.
BAILIS, J. M. and G. S. ROEDER, 2000 Pachytene exit controlled by reversal of Mek1-dependent phosphorylation. Cell 101:211-221[Medline].
BAKER, S. H., D. L. FREDERICK, A. BLOECHER, and K. TATCHELL, 1997 Alanine scanning mutagenesis of protein phosphatase type 1 in the yeast Saccharomyces cerevisiae.. Genetics 145:615-626[Abstract].
BALLARIN-DENTI, A., C. L. SLAYMAN, and H. KURODA, 1994 Small lipid-soluble cations are not membrane voltage probes for Neurospora or Saccharomyces.. Biochim. Biophys. Acta 1190:43-56[Medline].
BERTORELLO, A. M., A. APERIA, S. I. WALAAS, A. C. NAIRN, and P. GREENGARD, 1991 Phosphorylation of the catalytic subunit of Na+,K(+)-ATPase inhibits the activity of the enzyme. Proc. Natl. Acad. Sci. USA 88:11359-11362[Abstract/Free Full Text].
BIHLER, H., C. L. SLAYMAN, and A. BERTL, 1998 NSC1: a novel high-current inward rectifier for cations in the plasma membrane of Saccharomyces cerevisiae.. FEBS Lett. 432:59-64[Medline].
BIHLER, H., R. F. GABER, C. L. SLAYMAN, and A. BERTL, 1999 The presumed potassium carrier Trk2p in Saccharomyces cerevisiae determines an H+-dependent, K+-independent current. FEBS Lett. 447:115-120[Medline].
B
TOP
ABSTRACT
MATERIALS AND METHODS
