Characterization of an Allele-Nonspecific Intragenic Suppressor in the Yeast Plasma Membrane H⁺-ATPase Gene (PMA1)

Corresponding author: Francisco Portillo, Instituto de Investigaciones Biomédicas, C/Arturo Duperier, 4, E-28029 Madrid, Spain., fportillo{at}biomed.iib.uam.es (E-mail).

Communicating editor: M. CARLSON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We have analyzed the ability of A165V, V169I/D170N, and P536L mutations to suppress pma1 dominant lethal alleles and found that the P536L mutation is able to suppress the dominant lethality of the pma1-R271T, -D378N, -D378E, and -K474R mutant alleles. Genetic and biochemical analyses of site-directed mutants at Pro-536 suggest that this amino acid may not be essential for function but is important for biogenesis of the ATPase. Proteins encoded by dominant lethal pma1 alleles are retained in the endoplasmic reticulum, thus interfering with transport of wild-type Pma1. Immunofluorescence studies of yeast conditionally expressing revertant alleles show that the mutant enzymes are correctly located at the plasma membrane and do not disturb targeting of the wild-type enzyme. We propose that changes in Pro-536 may influence the folding of the protein encoded by a dominant negative allele so that it is no longer recognized and retained as a misfolded protein by the endoplasmic reticulum.

THE yeast plasma membrane H⁺-ATPase from Saccharomyces cerevisiae is an electrogenic proton pump essential for nutrient uptake and intracellular pH regulation (see review by SERRANO 1991 ). The ATPase belongs to the P-type, ion-translocating ATPase family, which forms an acyl-phosphate intermediate during catalysis and is sensitive to vanadate (PEDERSEN and CARAFOLI 1987 ).

The genes encoding members of the P-type ATPase family from bacterial, fungal, plant, and animal cells have been cloned and sequenced, in each case allowing the primary structure of the protein to be deduced. Comparison of the P-ATPases' amino acid sequences showed that they share several regions of homology and a common topology (SERRANO 1989 ; GOFFEAU and GREEN 1990 ; WACH et al. 1992 ). Site-directed mutagenesis and intragenic suppressor analysis have been used to probe structure-function relationships of the yeast enzyme. Site-directed mutagenesis has allowed the identification of a series of mutations that cause cell growth arrest even when wild-type Pma1 protein is coexpressed (HARRIS et al. 1994 ; PORTILLO 1997 ). Expression of these dominant lethal proteins leads to the accumulation of mutant and wild-type proteins in subplasma membrane structures probably derived from the endoplasmic reticulum (ER; HARRIS et al. 1994 ). From intragenic suppression analysis of selected pma1 alleles, three mutations (A165V, V169I/D170N, and P536L) were identified that exhibited characteristics of allele-nonspecific suppressors, as they were able to suppress mutations located either at the carboxyl terminus (ERASO and PORTILLO 1994 ) or at the ATP-binding domain of the enzyme (MALDONADO and PORTILLO 1995 ). Similar allele-nonspecific suppressors have been isolated in other systems and appear to affect protein stability (SHORTLE and LIN 1985 ; THOMAS et al. 1991 ). In this report, we show that the P536L mutation is able to suppress dominant lethal pma1 mutations located in three different regions of the ATPase. Further analysis of site-directed mutants at Pro-536 suggests that this residue is important for biogenesis of the enzyme and that suppression of dominant lethality is probably caused by compensatory structural alterations that alleviate retention of the mutant protein at the ER.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Yeast strains and growth conditions:
Yeast strain XZ611 is GAL⁺ MATa ura3 leu2 trp1 (HARRIS et al. 1994 ), and it was used to determine the dominant lethality of pma1 genes and in immunofluorescence experiments. Strain SY4 is GAL⁺ MATa ura3 leu2 his4 sec6 pma1::YIpGAL-PMA1 and was used for expression of the ATPase in secretory vesicles (NAKAMOTO et al. 1991 ). Synthetic media with 2% dextrose (SD), 2% lactic acid (SL), or 2% galactose (SG) and the appropriate requirements were used (ROSE et al. 1990 ). Yeast cells were transformed using the lithium acetate procedure (ITO et al. 1983 ). To test the dominant lethality of the pma1 alleles, transformants were transferred to SL medium and, after growth, suspended in water at a cell density of 2 x 10⁷ cell/ml, and 5 µl was dropped on SG.

Construction of pma1 alleles:
The pma1 genes containing Pro-536 mutations were produced by site-directed mutagenesis. The Amersham Kit for oligonucleotide-directed mutagenesis (Amersham Corp., Arlington Heights, IL) was used to introduce the mutations into a 2.2-kb XbaI fragment of the PMA1 gene previously subcloned into M13mp19. After mutagenesis, the entire XbaI fragment was sequenced to verify that only the nucleotide changes introduced by the mutagenic oligonucleotides were obtained. The XbaI fragments containing the mutation were moved into the full-length PMA1 gene in plasmid pRS427 (VALLEJO and SERRANO 1989 ). Plasmid pRS427 is a derivative of the URA3 single-copy plasmid YCp50 (ROSE and BROACH 1991 ) and containts the PMA1 gene joined to the GAL1 promoter. The pma1 alleles in Table 1 and Table 3 were constructed by placing the pma1 genes containing dominant lethal mutations under the control of the galactose-inducible promoter in plasmid pRS427, as described in PORTILLO 1997 . The Pro-536 mutations were subsequently combined in cis with the different dominant lethal mutations by exchange of a 2.2-kb XbaI fragment containing the allele-nonspecific mutations with the corresponding fragment in the dominant lethal GAL1::pma1 gene construction. All mutant constructions were verified by sequencing. The hemagglutinin (HA)-tagged pma1 alleles used in this study were generated by exchange of a 3.4-kb BstEII-HindIII fragment containing the desired mutations with the corresponding fragment in plasmid pFP302. The plasmid pFP302 is identical to pRS427, except that it carries an HA-tagged wild-type PMA1 under the control of the GAL1 promoter. The nine–amino acid HA epitope was introduced after the second amino acid of the ATPase by site-directed mutagenesis (PORTILLO 1997 ). All mutant constructions were verified by sequencing.

When coexpression of dominant lethal and wild-type proteins was required, cells were transformed with a pRS427 derivative containing different dominant lethal mutations and plasmid pFP239. The plasmid pFP239 is a derivative of the LEU2 single-copy plasmid pSB32 (ROSE and BROACH 1991 ) and contains an HA-tagged wild-type PMA1 under the control of the GAL1 promoter.

Replacement of the P536N and P536L mutant alleles by the chromosomal wild-type PMA1:
The ATPase gene fragments containing the Pro-536 substitutions were subcloned by exchange of a 2.2-kb XbaI fragment in plasmid pFP36. Plasmid pFP36 is a derivative of pUC18 and contains a PMA1 gene in which a URA3 marker was introduced into a synthetic XhoI linker located at the intergenic region between PMA1 and LEU1 (SERRANO et al. 1986 ). Strain FPY183 was transformed with the HindIII pma1::URA3 fragments containing the Pro-536 mutations. Strain FPY183, derived from autodiploid W303 (PORTILLO 1997 ), is heterozygous for a LEU2-marked deletion of PMA1 in which the 3.37-kb EcoRV-SalI PMA1–coding region was replaced with a 2.3-kb LEU2 fragment. After transformation, Ura⁺ Leu^- diploids were selected and tetrads were dissected. The diploid pma1-P536N::URA3/PMA1 yielded only two Ura^- viable spores, and the pma1-P536L::URA3/PMA1 rendered four viable spores. In each case, two independent diploids were dissected and segregation of 15 tetrads of each were studied. After growth and analysis of the Ura⁺ segregants containing the P536L mutation, genomic DNA-containing portions of the pma1-P536L gene were amplified by polymerase chain reaction and sequenced to ensure that only the P536L mutation was present.

Fluorescence microscopy:
Strain XZ611, transformed with different pma1 alleles, was selected in SD medium. To induce the GAL1::pma1 gene expression, cells were cultured in SL medium overnight at 30°, collected, resuspended in SG medium, and cultured for 5 hr at 30°. Cells were fixed and stained for immunofluorescence as described (PRINGLE et al. 1991 ). Localization of HA-tagged Pma1 was determined by immunofluorescence using a rhodamine-conjugated anti-HA antibody (Boehringer Mannheim, Mannheim, Germany).

Expression of the ATPase in secretory vesicles:
The 2.6-kb BamHI-HindIII fragments containing the Pro-536 substitutions were ligated into plasmid pPMA1.2 (NAKAMOTO et al. 1991 ) that had been digested with BamHI and HindIII. After the fragments were moved into plasmid pPMA1.2, a 3.77-kb SacI-HindIII fragment containing the entire ATPase coding sequence was cloned into vector YCp2HSE (NAKAMOTO et al. 1991 ), placing the mutant alleles under the control of a heat-shock-inducible promoter. The resulting plasmids were transformed into strain SY4 (NAKAMOTO et al. 1991 ). This strain has the chromosomal PMA1 gene under the control of the GAL1 promoter and also carries the sec6-4 mutation that blocks fusion of secretory vesicles with the plasma membrane. For expression studies, transformed SY4 were grown to midexponential phase (OD₆₆₀~0.5) on SG and shifted to SD for 3 hr to turn off transcription of chromosomal PMA1 and then shifted to 37° for 2 hr to turn on expression of the plasmid-borne gene and block fusion of secretory vesicles with plasma membrane. Secretory vesicles were then isolated by differential centrifugation as described (NAKAMOTO et al. 1991 ).

Biochemical methods:
Plasma membranes were purified from glucose-metabolizing yeast by differential and sucrose gradient centrifugation (SERRANO 1983 ). Immunoquantification of the ATPase was performed as described (RAO and SLAYMAN 1993 ). ATPase activity was assayed at pH 6.5 with 5mM ATP (SERRANO 1988 ). H⁺ pumping into secretory vesicles was monitored by fluorescence quenching of acridine orange (NAKAMOTO et al. 1991 ). Proton efflux from the cells was measured after starvation at 4° and glucose addition at 30° (SERRANO 1980 ). Protein concentration was determined by the Bradford method (BRADFORD 1976 ) with the Bio-Rad Protein Assay Reagent (Bio-Rad Laboratories, Hercules, CA) and bovine IgG as standard. The monoclonal antibodies Mab12 against the yeast ATPase (SERRANO et al. 1993 ) and 12CA5 against the HA epitope (WILSON et al. 1984 ) were used in immunoquantification and immunoblot. Plasma membrane proteins were separated by SDS-PAGE on 8% acrylamide using the Laemmli system (LAEMMLI 1970 ). Western blot with second antibody conjugated to alkaline phosphatase (Bio-Rad) was as described (BLAKE et al. 1984 ).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Suppression spectrum of allele-nonspecific suppressors:
The A165V, V169I/D170N, and P536L mutations were selected previously by suppression of the otherwise unrelated S911A/T912A (ERASO and PORTILLO 1994 ) and K474R (MALDONADO and PORTILLO 1995 ) mutations. The fact that the K474R mutation is a dominant lethal allele (PORTILLO 1997 ) prompted us to examine the effect of the A165V, V169I/D170N, and P536L mutations on the dominant lethality of several other pma1 mutations located in different regions of the enzyme. For this, we created novel combinations of dominant lethal and suppressor mutations in cis, and each of the new pma1 alleles was placed under the control of the GAL1 promoter (see MATERIALS AND METHODS). Plasmids carrying the GAL1::pma1 alleles were used to transform a wild-type strain. Under these conditions, yeast cells carrying a dominant lethal allele arrest growth when they are transferred from glucose- to galactose-containing medium. The results of this analysis for dominant lethality of the different in vitro-generated alleles are presented in Table 1. Only P536L reversion could suppress, in addition to K474R, several other dominant lethal mutations tested.

Genetic and biochemical analyses of Pro-536 substitution:
To examine the effect of amino acid replacement on suppression ability, we created by site-directed mutagenesis new pma1 alleles in which Pro-536 was replaced with residues of different size and charge (Pro-536 A, V, I, N, G, T, H, K, D). Before the above-mentioned analysis, the phenotype of the pma1 alleles generated was determined. We first examined whether or not the new mutations were dominant lethal. For this, the mutant ATPase genes were placed under control of the GAL1 promoter and transformed into the appropriate wild-type yeast to test for growth in galactose-containing medium. The results of this test for dominant lethality of the different site-directed mutations are summarized in Figure 1. Substitution of Pro-536 by either small uncharged (Gly) or charge residues (His, Lys, and Asp) renders a dominant lethal allele.

View larger version (72K): In this window In a new window Download PPT slide   Figure 1. Spot test for growth of yeast strains expressing dominant lethal alleles of PMA1. Strains are recombinant, carrying the PMA1 wild-type allele or the indicated pma1 allele under the control of the galactose-inducible promoter. The test was performed as indicated in MATERIALS AND METHODS. Plates were photographed after 36 hr at 30°.

The expression of dominant lethal proteins blocks their transport to the plasma membrane, and the proteins are accumulated into cytoplasmic structures called "Bip bodies" (SUPPLY et al. 1993 ; HARRIS et al. 1994 ). The expression of dominant negative protein also blocks the transport of wild-type Pma1 and leads to its colocalization in the same structures (HARRIS et al. 1994 ). To date, all dominant lethal PMA1 alleles identified exhibit the same phenotype (HARRIS et al. 1994 ; PORTILLO 1997 ). To test whether the dominant lethal mutations detected here also interefered with wild-type Pma1 targeting to the plasma membrane, we studied the transport of wild-type ATPase to the cell surface when dominant negative alleles are also expressed. To this end, we introduced the HA epitope in the wild-type Pma1 protein, and both the HA-tagged and dominant lethal allele were coexpressed under the control of the GAL1 promoter in the same cell. The location of the wild-type HA-tagged protein was studied in galactose-cultured cells by immunofluorescent staining of wild-type Pma1 with the anti-HA monoclonal antibody (Figure 2). When HA-tagged wild-type protein was induced simultaneously with the dominant lethal protein, the anti-HA antibody decorated a set of cytoplasmic structures similar to those observed for other dominant negative Pma1 proteins (HARRIS et al. 1994 ; PORTILLO 1997 ). When only the HA-tagged dominant negative proteins were expressed under the control of the GAL1 promoter, the same staining of cytoplasmic dots was observed (data not shown). These results suggest that dominant lethality of pma1-P536G, -P536H, -P536K, and -P536D is caused by the interference of the mutant protein with the transport of wild-type Pma1 protein.

View larger version (51K): In this window In a new window Download PPT slide   Figure 2. Accumulation of wild-type HA-tagged Pma1 in cells expressing dominant lethal Pro-536 mutations. Transformants carried the HA-tagged PMA1 gene and either the wild-type or the indicated dominant lethal allele under the control of the GAL1 promoter. Expression of the GAL::pma1 genes was induced with galactose for 5 hr. Panels show staining of the nucleus (DAPI) or immunofluorescence using a rhodamine-conjugated anti-HA antibody (anti-HA Ab).

The secretion of the nondominant lethal P536A, P536V, P536L, P536I, P536N, and P536T proteins was also studied. We introduced the HA epitope in the mutant proteins, and HA-tagged proteins were expressed under the control of the GAL1 promoter. We examined the location of each of these HA-tagged mutant proteins by immunofluorescent staining after galactose induction. The anti-HA antibody decorated the cell periphery in all the cases (data not shown), except in cells expressing the HA-tagged P536N protein, in which it accumulated in subplasma membrane structures reminiscent of those observed for dominant negative proteins (Figure 3). In contrast to dominant lethal proteins, the P536N protein does not block transport of wild-type Pma1 protein to the cell surface (data not shown). This may indicate that the P536N mutation renders a recessive lethal pma1 allele, which was confirmed by analyzing the segregation of the heterozygous diploid pma1-P536N::URA3/PMA1 (see MATERIALS AND METHODS). After sporulation and tetrad dissection of two independent diploids, only two viable spores/tetrad were obtained; both were Ura^- (15 tetrads studied in each of the diploids), thus confirming that pma1-P536N is a recessive lethal allele.

View larger version (50K): In this window In a new window Download PPT slide   Figure 3. Accumulation of the recessive lethal HA-tagged P536N protein in cytoplasmic structures. The transformant carried either the HA-tagged PMA1 or HA-tagged P536N genes under the control of the GAL1 promoter. After galactose induction, cells were stained for the nucleus (DAPI) or for the HA-tagged protein (anti-HA).

Pro-536 is located within the sequence motif ⁵³⁴DPPR, which is fully conserved among all eukaryotic, cation-transporting ATPases (SERRANO 1989 ; GOFFEAU and GREEN 1990 ; WACH et al. 1992 ) and is believed to form part of the ATP-binding site (RAO et al. 1989 ; SERRANO 1991 ). This prompted us to further characterize the effect of Pro-536 substitution on enzyme function. The P536A, P536V, P536L, P536I, or P536T mutant genes were placed under the control of a heat-shock-inducible promoter in the plasmid YCp2HSE (NAKAMOTO et al. 1991 ). Plasmids YCp2HSE containing the mutant genes were transformed into strain SY4 (NAKAMOTO et al. 1991 ). This strain has the chromosomal PMA1 gene under control of the GAL1 promoter and also carries the temperature-sensitive sec6-4 mutation that blocks fusion of secretory vesicles with the plasma membrane. Thus, when cells were incubated in galactose medium at 23°, the chromosomal wild-type ATPase was produced, and when the cells were transferred to glucose medium at 37°, the expression of the wild-type ATPase was repressed and the mutant ATPase was expressed. Because strain SY4 carries a temperature-sensitive sec6-4 mutation, the shift to 37° also led to the accumulation of the newly synthesized mutant ATPase in secretory vesicles. Secretory vesicles were purified by differential centrifugation for analysis of the newly synthesized mutant ATPase (NAKAMOTO et al. 1991 ). Table 2 summarizes the behavior of the mutant enzymes. In all the mutants, expression levels were significantly higher than that of the wild-type protein and, similarly, all had relatively high ATPase activity and H⁺-pumping rates. After correction for the amount of Pma1, values ranged from 60% hydrolysis/53% pumping in P536A to 114% hydrolysis/93% pumping in P536I. Other kinetic properties assayed, K_m for ATP and K_i for orthovanadate, appeared to be essentially wild type. Thus, it appears that Pro-536 is not indispensable for enzyme activity and that the main defect of these Pro-536 substitutions seems to be hyperaccumulation of enzymes in the secretory vesicles.

To determine whether or not the Pro-536 mutations also led to hyperaccumulation of the mutant Pma1 protein at the plasma membrane, we analyzed the transport of wild-type and mutant enzymes to the cell surface. Cells expressing HA-tagged wild-type or mutant proteins under the control of the GAL1 promoter were shifted from glucose- to galactose-containing medium, and plasma membranes were purified at different induction times and analyzed by Western blot using anti-HA antibody. An example of this analysis is shown in Figure 4 for wild-type and P536L Pma1 protein. The newly synthesized HA-tagged P536L Pma1 protein is accumulated at the plasma membrane at levels higher than those of the wild type. Considering that the wild-type ATPase is a long-lived and metabolically stable protein with a half-life of >20 hr (BENITO et al. 1991 ), the observed hyperaccumulation of newly synthesized P536L protein at shorter times cannot be explained by changes in the stability of the mutant protein. Thus, it seems that Pro-536 may play an important role in the biogenesis of the plasma membrane ATPase.

View larger version (21K): In this window In a new window Download PPT slide   Figure 4. Accumulation of wild-type and P536L proteins at the plasma membrane. Strains are recombinants carrying either the HA-tagged PMA1 (HA · Pma1) or HA-tagged P536L (HA · P536L) gene. Cells were grown in SL medium and transferred to SG to induce the expression of the GAL::pma1 genes. At the times indicated, samples were taken and the plasma membrane was purified. Purified plasma membrane was analyzed by immunoblot using monoclonal antibody 12CA5 against the HA epitope (anti-HA) and Mab12 against Pma1 (anti-Pma1). The upper blot was intentionally developed longer than the lower one to detect expressed HA-tagged Pma1 at shorter times.

Suppression spectrum of Pro-536 substitution:
After phenotypic characterization of the mutant carrying the Pro-536 mutation, the pma1-P536A, -P536V, -P536I, and -P536T alleles were analyzed for competence to suppress the R271T, D378N, D378E, and K474R dominant lethal mutations. We combined the Pro-536 substitutions with the dominant negative mutations in cis (see MATERIALS AND METHODS). The new pma1 alleles were placed under control of the GAL1 promoter and tested for dominant lethality. The result of this intragenic suppression analysis is shown in Table 3. The dominant lethality of the R271T and D378N mutations was suppressed by any of the Pro-536 substitutions tested, and none of the Pro-536 mutations suppressed the lethality of the pma1-D378E allele. K474R was suppressed by the P536V and P536T, but not by the P536A mutations, which suggests that the bulkiness of the amino acid side chain is important for this intragenic suppression.

Effect of Pro-536 substitutions on localization of dominant lethal Pma1:
To understand how Pro-536 mutations suppress the dominant lethal alleles, it is necessary to determine whether or not revertant mutant enzymes are still accumulated at Bip bodies. We examined the transport of HA-tagged revertant enzymes. Cells expressing HA-tagged wild-type, dominant lethal, or revertant proteins under the control of the GAL1 promoter were shifted from glucose- to galactose-containing medium, and the location of the HA-tagged proteins was examined by immunofluorescent staining after galactose induction. Figure 5 shows, as an example, the results obtained with the revertants generated from the P536L mutation: R271T/P536L, D378N/P536L, D378E/P536L, and K474R/P536L. In contrast to the R271T, D378N, D378T, and K474R dominant negative proteins, the revertant Pma1 proteins are localized at the plasma membrane. This indicates that P536L abolishes retention of the proteins encoded by the dominant lethal mutations at the ER and leads to transport of the revertant enzymes to the cell surface.

View larger version (76K): In this window In a new window Download PPT slide   Figure 5. Effect of P536L mutation on localization of dominant lethal Pma1 proteins. Transformants carried the indicated HA-tagged pma1 allele under the control of the GAL1 promoter. After galactose induction of the GAL1::pma1 genes, cells were stained for immunofluorescence with a rhodamine-conjugated anti-HA antibody.

Phenotype of yeast expressing the pma1-P536L allele:
The amount of Pma1 in the plasma membrane appears to be tightly regulated, as overexpression of the ATPase gene on a multicopy plasmid in yeast yields only a slight increase in the amount of Pma1 in the plasma membrane (ERASO et al. 1987 ). The fact that P536L mutant enzyme expression significantly increased the amount of mutant Pma1 at the plasma membrane led us to explore the phenotype of a yeast strain expressing only the pma1-P536L allele. Two heterozygous diploids, pma1-P536L::URA3/PMA1 (see MATERIALS AND METHODS), were sporulated and tetrads were dissected. In all tetrads studied, four viable spores/tetrad were obtained. Three Ura⁺ haploid segregants containing the pma1-P536L allele were selected to evaluate the physiological consequences of the P536L mutant protein expression. The results of this study are summarized in Table 4. Yeast expressing the pma1-P536L allele showed a twofold increase in the amount of Pma1 compared with the wild-type strain. As a consequence, the in vitro ATPase activity assayed in purified plasma membrane was also higher than that of the wild-type strain. Yeast cells actively extrude protons into the medium, an activity caused by plasma membrane H⁺-ATPase, which reflects the in vivo activity of the enzyme (SERRANO 1980 ; CID et al. 1987 ) and correlates with both the in vitro ATPase activity and the specific growth rate (PORTILLO and SERRANO 1989 ; VALLEJO and SERRANO 1989 ). It was therefore expected that cells exhibiting high in vitro ATPase activity would also show high H⁺-pumping activity relative to the wild-type cells. The fact that cells expressing the P536L mutant enzyme exhibited lower H⁺-pumping activity relative to the wild-type strain suggests that not all of the P536L mutant enzyme accumulated in the plasma membrane is physiologically active or that the mutant enzyme is only partially active. This is reinforced by the fact that the specific growth rate of the mutant cell decreased relative to that of the wild type.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Role of Pro-536 in ATPase function:
Pro-536 is fully conserved among all known members of the P-type ATPase family (SERRANO 1989 ; GOFFEAU and GREEN 1990 ; WACH et al. 1992 ). The P536A, P536V, P536L, P536I, and P536T mutations resulted in active enzymes that exhibited kinetic parameters similar to those of the wild-type enzyme; therefore, at least in these cases, Pro-536 does not appear to be essential for enzyme function. The most remarkable phenotype of these enzymes appears to be a hyperaccumulation at the cell surface. In contrast, enzymes bearing P536G, P536H, P536K, and P536D mutations were defective in targeting to the plasma membrane, as they are retained at the ER. These results suggest that the residue at position 536 may play an important role in enzyme biogenesis. In wild-type yeast cells, newly synthesized Pma1 is delivered to the plasma membrane via the secretory pathway (HOLCOMB et al. 1988 ; CHANG and SLAYMAN 1991 ). Several lines of evidence suggest that transport of the protein to the plasma membrane is under a tight control that mediates the retention of either overexpressed wild-type (SUPPLY et al. 1993 ) or aberrant (HARRIS et al. 1994 ; PORTILLO 1997 ) Pma1 proteins at the ER. Although little is known about both the mechanisms that mediate the retention at the ER and the molecular determinants involved in the targeting to the plasma membrane of the ATPase, the finding that P536A, P536V, P536L, P536I, and P536T mutant proteins are hyperaccumulated at the cell surface raises the possibility that these mutants bypass some of the tight control on transport of the protein, thus improving the transport of the mutant enyzme to the cell surface.

We found that yeast cells expressing only the pma1-P536L mutant allele exhibited a significantly decreased in vivo proton transport rate relative to both activity and abundance of ATPase in purified plasma membrane. This may indicate that not all of the accumulated mutant ATPase is physiologically active. Several hypotheses can be considered to explain this striking phenotype. One hypothesis could be that expression of P536L protein was detrimental to yeast growth. If expression of P536L enzyme were toxic, there would be strong selective pressure during growth for the accumulation of second-site mutations able to suppress the P536L mutation. This problem would not affect the P536L protein expressed in secretory vesicles, as the mutant protein is transiently expressed in this case. Nevertheless, we rescued and sequenced the pma1 allele after yeast growth, and only the P536L mutation was present. However, extragenic mutations cannot be completely discarded. The possibility that P536L mutant protein expression led to saturation of the secretory pathway, resulting in proliferation of internal membranes as a mechanism to control the amount of Pma1 at the plasma membrane (SUPPLY et al. 1993 ; DE KERCHOVE D'EXAERDE et al. 1996 ), is discarded based on the following arguments: P536L protein was accumulated during yeast growth in a membrane fraction that copurified with the plasma membrane, and immunofluorescence staining of yeast cells expressing the HA-tagged P536L mutant protein with the anti-HA antibody indicated that the mutant protein was localized exclusively at the cell surface. An alternative hypothesis, which could account for cell surface localization of the mutant enzyme and yet low in vivo proton transport, could be that the P536L mutant protein remains trapped in subplasma membrane structures derived from the plasma membrane. Based on the finding that the MOP2/END4 gene, which is required for receptor-mediated and fluid-phase endocytosis (RATHS et al. 1993 ), is also important to control the abundance of Pma1 at the plasma membrane (NA et al. 1995 ), it seems reasonable to think that such subplasma membrane structures could be derived from the plasma membrane by endocytosis. This hypothesis raises the possibility that yeast cells could also control the amount of ATPase at the plasma membrane by regulating its endocytic recycling. Nevertheless, the present results do not allow us to dismiss an alternative model in which hyperaccumulation of ATPase at the plasma membrane triggers a posttranscriptional mechanism that down-regulates the ATPase. This type of down-regulation of ATPase has already been observed and is mediated by the plasma membrane–associated protein kinase YCK1 (ESTRADA et al. 1996 ). Data allowing discrimination between the various possibilities might be provided by analyzing the expression and activity of P536L protein in mutants altered in endocytosis or YCK1.

Mechanism of suppression of dominant lethal mutations by the P536L mutation:
We report here that P536L mutation suppresses the dominant lethality of the pma1-R271T, -D378N, -D378E, and -K474R mutant alleles. These dominant lethal pma1 alleles are located in different regions of the ATPase: Arg-271 in the stalk region, Asp-378 in the phosphorylation domain, and Arg-474 in the ATP-binding domain (RAO et al. 1989 ; SERRANO 1991 ). Intragenic suppression analysis of selected pma1 mutations has suggested a conformational coupling between the stalk region and both the ATP-binding and -phosphorylation domains (NA et al. 1993 ; HARRIS et al. 1994 ; MALDONADO and PORTILLO 1995 ). This structural coupling could explain the suppression pattern of the P536L mutation. Some specificity is observed for suppression of the Asp-378 substitution, as the P536L mutation suppresses the dominant lethality caused by conservative changes (D378 N, E), but not that induced by a nonconservative substitution (D378 T) (Table 1). Some steric requirements are also observed for suppression of the K474R and D378E mutations. In this case, the bulkiness of the amino acid side chain at position 536 appears to be important for suppression (Table 3). Although dominant lethal proteins are accumulated into cytoplasmic structures (HARRIS et al. 1994 ), we showed that the P536L mutation suppresses the mislocalization defect caused by the dominant lethal mutations, leading to transport of the revertant enzymes to the cell surface. There are several mechanisms by which the P536L mutation might correct the mislocalization of R271T, D378N, D378E, and K474R mutant proteins. The easiest interpretation is that revertant proteins are localized more efficiently at the plasma membrane, merely because of the improved transport caused by the P536L mutation. This explanation, however, although feasible, does not account for the observed specificity of the intragenic suppressor mutation because not every dominant lethal mutation studied in this work is suppressed by Pro-536 mutations, pointing to a steric requirement for suppression. Suppressor mutations able to correct the mislocalization of mutated proteins have been also found in the human cystic fibrosis transmembrane conductance regulator (CFTR; TEEM et al. 1993 ). CFTR functions as a chloride ion channel (ANDERSON et al. 1991 ; BEAR et al. 1992 ) located at the apical membrane of epithelial cells (DENNING et al. 1992A ). In CFTR, the deletion of a single amino acid, phenylalanine, at position 508 (F508), alters the conformation of CFTR (THOMAS et al. 1992 ; THOMAS and PEDERSEN 1993 ) and produces two defects: it disrupts the normal processing of the protein in such a way that the F508 mutant protein is retained in the ER (CHENG et al. 1990 ; DENNING et al. 1992B ; LUKACS et al. 1994 ), and it alters chloride transport (DALEMANS et al. 1991 ). TEEM et al. 1993 isolated intragenic revertants of F508 that partially suppressed both the mislocalization and transport defects caused by the deletion of F508, and they interpreted this finding to mean that suppressor mutations act as compensatory mutations that allow proper folding and normal function of mutant CFTR. Although we have no direct evidence that dominant lethal mutations cause misfolding and misfunction of ATPase, both the pattern and specificity of the suppression observed for P536L mutation, together with the fact that suppressor mutations correct the mislocation caused by the dominant lethal mutations, suggest that a mechanism similar to that proposed for intragenic suppressors of F508 could be responsible for suppression of pma1-R271T, -D378N, -D378E, and -K474R dominant lethal alleles. Our genetic data cannot define the position of these residues within the tertiary structure of the ATPase, but they indicate that these residues could interact with Pro-536. Such interaction might contribute to the correct folding of the ATPase that is needed for proper localization of the enzyme. If dominant lethal mutations disturb such interactions, Pro-536 suppressor mutations could act as compensatory mutations by restoring those interactions, thus allowing proper folding and transport to the plasma membrane. Conformational analysis of polypeptides containing the dominant lethal and/or the Pro-536 suppressor mutations could provide data supporting this hypothesis.

	ACKNOWLEDGMENTS

We are indebted to CAROLYN SLAYMAN for SY4, pPMA1.2, and YCp2HSE-PMA1; JIM HABER for XZ611; JORGE PÉREZ-MARTIN for providing the anti-HA antibody and RAMÓN SERRANO for providing the anti-Pma1 antibody; MARÍA J. MAZÓN and PILAR ERASO for critical reading of the manuscript; and ANTONIO FERNÁNDEZ for help with the art work. ANA M. MALDONADO and NATALIA DE LA FUENTE were recipients of fellowships from the Gobierno Vasco and the Ministerio de Educación y Ciencia, respectively. This study was supported by Spanish Grant DGICYT-PB94-0096.

Manuscript received January 14, 1998; Accepted for publication May 21, 1998.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ANDERSON, M. P., R. J. GREGORY, S. THOMPSON, D. W. SOUZA, and S. PAUL et al., 1991  Demonstration that CFTR is a chloride channel by alteration of its annion selectivity. Science 253:202-205[Abstract/Free Full Text].

BEAR, C. E., C. H. LI, N. KARTNER, R. J. BRIDGES, and T. J. JENSEN et al., 1992  Purification and functional reconstitution of the cystic fibrosis transmembrane conductance regulator (CFTR). Cell 68:809-818[Medline].

BENITO, B., E. MORENO, and R. LAGUNAS, 1991  Half-life of the plasma membrane ATPase and its activating system in resting yeast cells. Biochim. Biophys. Acta 1063:265-268[Medline].

BRADFORD, M. M., 1976  A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248-254[Medline].

BLAKE, M. S., K. H. JOHNSTON, G. J. RUSSELL-JONES, and E. C. GOTSCHLICH, 1984  A rapid, sensitive method for detection of alkaline phosphatase anti-antibody on Western blots. Anal. Biochem. 136:175-179[Medline].

CID, A., R. PERONA, and R. SERRANO, 1987  Replacement of the promoter of the yeast plasma membrane ATPase gene by a galactose-dependent promoter and its physiological consequences. Curr. Genet. 12:105-110[Medline].

CHANG, A. and C. SLAYMAN, 1991  Maturation of the yeast plasma membrane [H⁺]ATPase involves phosphorylation during intracellular transport. J. Cell Biol. 115:289-295[Abstract/Free Full Text].

CHENG, S. H., R. J. GREGORY, J. MARSHALL, S. PAUL, and D. W. SOUZA et al., 1990  Defective intracellular transport and processing of CFTR is the molecular basis of most cystic fibrosis. Cell 63:827-834[Medline].

DALEMANS, W., P. BARBRY, G. CHAMPIGNY, S. JALLAT, and K. DOTT et al., 1991  Altered chloride ion channel kinetics associated with the delta F508 cystic fibrosis mutation. Nature 354:526-528[Medline].

DE KERCHOVE D'EXAERDE, A., P. SUPPLY, and A. GOFFEAU, 1996  Subcellular traffic of the plasma membrane H⁺-ATPase in Saccharomyces cerevisiae. Yeast 12:907-916[Medline].

DENNING, G. M., L. S. OSTEDGAARD, S. H. CHENG, A. E. SMITH, and M. J. WELSH, 1992a  Localization of cystic fibrosis transmembrane conductance regulator in primary cultures of cystic fibrosis airway epithelia. J. Clin. Invest. 89:339-349.

DENNING, G. M., L. S. OSTEDGAARD, and M. J. WELSH, 1992b  Abnormal localization of cystic fibrosis transmembrane conductance regulator in primary cultures of cystic fibrosis airway epithelia. J. Cell Biol. 118:551-559[Abstract/Free Full Text].

ERASO, P. and F. PORTILLO, 1994  Molecular mechanism of regulation of yeast plasma membrane H⁺-ATPase by glucose. Interaction between domain and identification of new regulatory sites. J. Biol. Chem. 269:10393-10399[Abstract/Free Full Text].

ERASO, P., A. CID, and R. SERRANO, 1987  Tight control of yeast plasma membrane ATPase during changes in growth and gene dosage. FEBS Lett. 224:193-197[Medline].

ESTRADA, E., P. AGOSTINI, J. R. VANDENHEEDE, J. GORIS, and W. MERLEVEDE et al., 1996  Phosphorylation of yeast plasma membrane H⁺-ATPase by casein kinase I. J. Biol. Chem. 271:32064-32072[Abstract/Free Full Text].

GOFFEAU, A., and N. M. GREEN, 1990 The H⁺-ATPase from fungal plasma membrane, pp. 155–169 in Monovalent Cations in Biological Systems, edited by C. A. PASTERNAK. CRC Press, Boca Raton, FL.

HARRIS, S. L., S. NA, X. ZHU, D. SETO-YOUNG, and D. S. PERLIN et al., 1994  Dominant lethal mutations in the plasma membrane H⁺-ATPase gene of Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 91:10531-10535[Abstract/Free Full Text].

HOLCOMB, C. L., W. J. HANSEN, T. ETCHEVERRY, and R. SCHEKMAN, 1988  Secretory vesicles externalize the major plasma membrane ATPase in yeast. J. Cell Biol. 106:641-648[Abstract/Free Full Text].

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

LAEMMLI, U. K., 1970  Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685[Medline].

LUKACS, G., A. MOHAMED, N. KARTNER, X.-B. CHANG, and J. R. RIORDAN et al., 1994  Conformational maturation of CFTR but not its mutant counterpart (F508) occurs in the endoplasmic reticulum and requires ATP. EMBO J. 13:6076-6086[Medline].

MALDONADO, A. M. and F. PORTILLO, 1995  Genetic analysis of the fluorescein isothiocyanate binding site of the yeast plasma membrane H⁺-ATPase. J. Biol. Chem. 270:8655-8659[Abstract/Free Full Text].

NA, S., D. S. PERLIN, D. SETO-YOUNG, 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].

NA, S., M. HINCAPIE, J. H. MCCUSKER, and J. E. HABER, 1995  MOP2 (SLA2) affects the abundance of the plasma membrane H⁺-ATPase of Saccharomyces cerevisiae. J. Biol. Chem. 270:6815-6823[Abstract/Free Full Text].

NAKAMOTO, R. K., R. RAO, and C. W. SLAYMAN, 1991  Expression of the yeast plasma membrane [H⁺]ATPase in secretory vesicles. A new strategy for directed mutagenesis. J. Biol. Chem. 266:7940-7949[Abstract/Free Full Text].

PEDERSEN, P. L. and E. CARAFOLI, 1987  Ion motive ATPases. I. Ubiquity, properties and significance to cell function. Trends Biochem. Sci. 12:146-150.

PORTILLO, F., 1997  Characterization of dominant lethal mutations in the yeast plasma membrane H⁺-ATPase gene. FEBS Lett. 402:136-140[Medline].

PORTILLO, F. and R. SERRANO, 1989  Growth control strength and active site of yeast plasma membrane ATPase studied by site-directed mutagenesis. Eur. J. Biochem. 186:501-507[Medline].

PRINGLE, J. R., A. E. M. ADAMS, D. G. DRUBIN, and B. K. HAARER, 1991  Immunofluorescence methods for yeast. Methods Enzymol. 194:565-602[Medline].

RAO, R. and C. W. SLAYMAN, 1993  Mutagenesis of conserved residues in the phosphorylation domain of the yeast plasma membrane H⁺-ATPase. Effects on structure and function. J. Biol. Chem. 268:6708-6713[Abstract/Free Full Text].

RAO, R., R. K. NAKAMOTO and C. W. SLAYMAN, 1989 The nucleotide-binding site of the plasma membrane H⁺-ATPase of Neurospora crassa: a comparison with other P-type ATPases, pp. 35–43 in Ion Transport, edited by D. KEELING and C. BENHAM. Academic Press, New York.

RATHS, S., J. ROHER, F. CRAUSAZ, and H. RIEZMAN, 1993  end3 and end4: two mutants defective in receptor-mediated and fluid-phase endocytosis in Saccharomyces cerevisiae. J. Cell. Biol. 120:55-65[Abstract/Free Full Text].

ROSE, M. D. and J. R. BROACH, 1991  Cloning genes by complementation in yeast. Methods Enzymol. 194:195-230[Medline].

ROSE, M. D., F. WINSTON and P. HIETER, 1990 Methods in Yeast Genetics. A Laboratory Course Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

SERRANO, R., 1980  Effects of ATPase inhibitors on the proton pump of respiratory-deficient yeast. Eur. J. Biochem. 105:419-424[Medline].

SERRANO, R., 1983  In vivo glucose activation of the yeast plasma membrane ATPase. FEBS Lett. 56:11-14.

SERRANO, R., 1988  H⁺-ATPase from plasma membrane of Saccharomyces cerevisiae and Avena sativa roots: purification and reconstitution. Methods Enzymol. 157:533-534[Medline].

SERRANO, R., 1989  Structure and function of plasma membrane ATPase. Annu. Rev. Plant Physiol. Plant Mol. Biol. 40:61-94.

SERRANO, R., 1991 Transport across yeast vacuolar and plasma membrane, pp. 523–585 in The Molecular and Cellular Biology of Yeast Saccharomyces. Genome Dynamics, Protein Synthesis and Energetics, edited by J. R. BROACH, J. R. PRINGLE and E. W. JONES. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

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⁺-ATPase. Nature 319:689-693[Medline].

SERRANO, R., B. C. MONK, J. M. VILLALBA, C. MONTESINOS, and E. W. WEILER, 1993  Epitope mapping and accessibility of immunodominant regions of yeast plasma membrane H⁺-ATPase. Eur. J. Biochem. 212:737-744[Medline].

SHORTLE, D. and B. LIN, 1985  Genetic analysis of staphylococcal nuclease: identification of three intragenic "global" suppressors of nuclease minus mutations. Genetics 110:539-555[Abstract/Free Full Text].

SUPPLY, P., A. WACH, D. THINES-SEMPOUX, and A. GOFFEAU, 1993  Proliferation of intracellular structures upon overexpression of the PMA2 ATPase in Saccharomyces cerevisiae. J. Biol. Chem. 268:19744-19752[Abstract/Free Full Text].

TEEM, J. L., H. A. BERGER, L. S. OSTEDGAARD, D. P. RICH, and L.-C. TSUI et al., 1993  Identification of revertants for the cystic fibrosis F508 mutation using STE6-CFTR chimeras in yeast. Cell 73:335-346[Medline].

THOMAS, P. J. and P. L. PEDERSEN, 1993  Effect of the F508 mutation on structure, function and folding of the first nucleotide-binding domain of CFTR. J. Bioenerg. Biomembr. 25:11-19[Medline].

THOMAS, P. J., P. SHENBAGAMURTHI, J. SONDEK, J. M. HULLIHEN, and P. L. PEDERSEN, 1992  The cystic fibrosis transmembrane conductance regulator: effects of the most common cystic fibrosis-causing mutation on the secondary structure and stability of a synthetic peptide. J. Biol. Chem. 267:5727-5730[Abstract/Free Full Text].

THOMAS, W., R. M. SPELL, M. E. MING, and C. HOLM, 1991  Genetic analysis of the gyrase A–like domain of DNA topoisomerase II of Saccharomyces cerevisiae. Genetics 128:703-716[Abstract].

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

WACH, A., A. SCHLESSER, and A. GOFFEAU, 1992  An alignment of 17 deduced protein sequences from plant, fungi and ciliate H⁺-ATPase genes. J. Bioenerg. Biomembr. 24:309-317[Medline].

WILSON, I. A., H. L. NIMAN, R. A. HOUGHTEN, A. R. CHERENSON, and M. L. CONNOLLY et al., 1984  The structure of an antigenic determinant in a protein. Cell 37:767-778[Medline].