Site-directed Mutagenesis of the 100-kDa Subunit (Vph1p) of the Yeast Vacuolar (H (cid:49) )-ATPase*

Vacuolar (H (cid:49) )-ATPases (V-ATPases) are multisubunit complexes responsible for acidification of intracellular compartments in eukaryotic cells. V-ATPases possess a subunit of approximate molecular mass 100 kDa of un- known function that is composed of an amino-terminal hydrophilic domain and a carboxyl-terminal hydropho- bic domain. To test whether the 100-kDa subunit plays a role in proton transport, site-directed mutagenesis of the VPH1 gene, which is one of two genes that encodes this subunit in yeast, has been carried out in a strain lacking both endogenous genes. Ten charged and twelve polar residues located in the seven putative transmem- brane helices in the COOH-terminal domain of the mol-ecule were individually changed, and the effects on pro- ton transport, ATPase activity, and assembly of the yeast V-ATPase were measured. Two mutations (R735L and Q634L) in transmembrane helix 6 and at the border of transmembrane helix 5, respectively, showed greatly reduced levels of the 100-kDa subunit in the vacuolar membrane, suggesting that these mutations affected sta- bility of the 100-kDa subunit. Two mutations, D425N and K538A, in transmembrane helix 1 and at the border of transmembrane helix 3, respectively, showed reduced assembly of the V-ATPase, with the D425N mutation also reducing the activity of V-ATPase complexes that did assemble. Two mutations, H743A and K593A, in trans- membrane helix

The vacuolar (H ϩ )-ATPases (V-ATPases) 1 are a family of proton pumps responsible for acidification of intracellular compartments in eukaryotic cells (for reviews see Refs. [1][2][3][4][5][6][7][8][9]. Among the compartments acidified by V-ATPases are clathrincoated vesicles, endosomes, lysosomes, secretory vesicles, such as synaptic vesicles and chromaffin granules, and the central vacuoles of plants, Neurospora and yeast. Acidification of these compartments, in turn, plays an important role in such processes as receptor-mediated endocytosis, intracellular membrane traffic, protein processing and degradation, and coupled transport of small molecules. In yeast, acidification of the central vacuole is important both to maintain the activity of degradative enzymes and to drive uptake of solutes such as Ca 2ϩ and amino acids (5).
Although the V-ATPases are homologous to the F-ATPases (28 -31), both in overall structure (10,27) and in sequence homology of several of the subunits (11-13, 25, 32-38), no obvious structural homolog exists for the 100-kDa subunit in the F-ATPases. The 100-kDa subunit of the V-ATPase in yeast is encoded by two homologous genes, VPH1 (21) and STV1 (22). VPH1 encodes a 95-kDa protein, which possesses a hydrophilic amino-terminal domain of approximately 45 kDa and a carboxyl-terminal hydrophobic domain of approximately 50 kDa containing 6 -7 putative transmembrane helices (21). STV1 encodes a 102-kDa protein that shares the same domain arrangement and is 54% identical in amino acid sequence with the product of the VPH1 gene (Vph1p) (22). Disruption of the VPH1 gene leads to somewhat reduced growth at neutral pH relative to acidic pH (21) and disruption of STV1 has no obvious phenotypic consequences (22). By contrast, disruption of both VPH1 and STV1 leads to the typical Vma Ϫ phenotype, including the inability to grow at neutral pH and hypersensitivity to Ca 2ϩ (22). These results suggest that the VPH1 and STV1 gene products are at least partially able to substitute for each other. The reason that yeast possess two genes encoding the 100-kDa subunit is uncertain, but the results available thus far suggests that V-ATPases possessing Vph1p and Stv1p may be targeted to different intracellular membranes (22). In order to determine whether the 100-kDa subunit plays a direct role in proton transport by the V-ATPase complex, site-directed mutagenesis of the VPH1 gene product has been carried out in a strain lacking endogenous Vph1p and Stv1p.

EXPERIMENTAL PROCEDURES
Materials and Strains-Zymolyase 100T was obtained from Seikagaku America, Inc. [ 35 S]Trans-label was purchased from ICN. Bafilomycin A 1 was a kind gift from Dr. Karlheinz Altendorf (University of Osnabruck). Leupeptin was from Boehringer Mannheim. 9-Amino-6chloro-2-methoxyacridine was from Molecular Probes, Inc. ATP, phenylmethylsulfonylfluoride, and most other chemicals were purchased from Sigma.
Transformation-Yeast cells were transformed with the wild type plasmid (pRS316-VPH1), mutants or the vector pRS316 alone (as a negative control) using the lithium acetate procedure (41) and were selected on Ura Ϫ plates as described previously (42). The mutants were tested for growth on pH 7.5 or pH 5.5 YPD plates buffered with 50 mM KH 2 PO 4 and 50 mM succinic acid (43).
Isolation of Vacuolar Membrane Vesicles-Vacuolar membrane vesicles were isolated using a modification of the protocol described by Kim et al. (44). Yeast were grown overnight at 30°C to 1 ϫ 10 7 cells/ml in 1 liter of selective medium. Cells were pelleted, washed once with water, and resuspended in 50 ml of 10 mM dithiothreitol, 100 mM Tris-HCl, pH 9.4. After incubation at 30°C for 15 min, cells were pelleted again, resuspended in 50 ml of YEPD medium containing 0.7 M sorbitol, 2 mM dithiothreitol, and 100 mM Tris-Mes, pH 7.5, and 5 mg of Zymolyase 100T and incubated at 30°C with gentle shaking for 90 min. The resulting spheroplasts were washed twice with ice-cold 1.2 M sorbitol, and pelleted at 3500 ϫ g for 10 min at 4°C. The pellet was resuspended in 40 ml of homogenization buffer (10% glycerol, 1.5% polyvinylpyrrolidone (M r 40,000), 0.25 mM MgCl 2 , 2 mg/ml bovine serum albumin, 50 mM Tris-ascorbate, pH 7.5, 1 mM phenylmethylsulfonyl fluoride, 1 mg/ml leupeptin), transferred to a Dounce glass homogenizer and subjected to 20 strokes with a tight fitting pestle. The homogenate was centrifuged at 3500 ϫ g for 15 min at 4°C, and the supernatant was transferred to a Ti 45 centrifuge tube and spun for 35 min at 100,000 ϫ g at 4°C. The pellets were resuspend in 8 ml of overlay medium, which contained 1.1 M glycerol, 2 mM dithiothreitol, 0.25 mM MgCl 2 , 2 mg/ml bovine serum albumin, and 5 mM Tris-Mes, pH 7.6, and homogenized by 10 strokes in a Dounce glass homogenizer using a tight fitting pestle. The homogenate was overlaid onto a one step 30-ml 10 -30% sucrose gradient and centrifuged for 2 h at 100,000 ϫ g in an SW-28 rotor. Material at the 10 -30% interface was collected, diluted 10-fold with overlay medium, and centrifuged at 100,000 ϫ g for 35 min at 4°C. The pellets were resuspend in 0.5-1 ml of overlay medium, quick frozen with liquid nitrogen, and stored at Ϫ80°C until used.
Biochemical Characterization-ATPase activity was measured using a coupled spectrophotometric assay (45) with the modification of using 0.35 mM of NADH instead of 0.5 mM NADH. ATP-dependent proton transport was measured in transport buffer (25 mM Mes/Tris, pH 7.2, 5 mM MgCl 2 , and 25 mM KCl) using the fluorescence probe amino-6chloro-2-methoxyacridine as described previously (46) in the presence or the absence of 10 nM bafilomycin A 1 . Protein concentration was measured using the Lowry method (47). SDS-polyacrylamide gel electrophoresis was carried out as described by Laemmli (48). Silver staining was performed using the method of Oakley et al. (49). Western blots were probed with mouse monoclonal antibodies 8B1-F3 against the 69-kDa subunit, (from Molecular Probes, Inc.) or 10D7 against the 100-kDa subunit, (a generous gift from Dr. P.Kane), followed by horseradish peroxidase-conjugated secondary antibody (Bio-Rad). Blots were developed using a chemiluminescent detection method obtained from KPL. Quantitations were done using an IS-1000 Digital imaging system (Alpha Innotech Corporation).
Immunoprecipitations were carried out as described (50), with the following modifications. Cells were grown overnight in supplemented minimal medium lacking methionine and then converted to spheroplasts by incubation for 20 min with 0.5 unit of zymolase/10 7 cells in SD-Met, 1.2 M sorbitol, 50 mM Tris-Mes, pH 7.5. Aliquots containing 5 ϫ 10 6 spheroplasts were then incubated with [ 35 S]Trans-label (50 Ci) for 60 min at 30°C. Spheroplasts were then pelleted, lysed in phosphatebuffered saline with C 12 E 9 and immunoprecipitated (50) using 7.5 g of purified 8B1-F3 antibody and protein A-Sepharose followed by SDSpolyacrylamide gel electrophoresis on 12% acrylamide gels and autoradiography as described (50).

RESULTS
It has previously been shown that deletion of genes encoding subunits of the yeast V-ATPase leads to a conditional lethal phenotype such that strains carrying such deletions are unable to grow at neutral pH but are able to grow at acidic pH (51). We have employed a strain in which both the VPH1 and STV1 genes encoding the 100-kDa subunit of the V-ATPase have been disrupted, leading to the typical Vma Ϫ phenotype (22). Expression of the VPH1 gene on a CEN plasmid in this strain leads to growth at pH 7.5. Twenty-two individual site-directed mutations were introduced into the VPH1 gene and the mutant proteins expressed in the double knock out strain. The residues selected for mutation all correspond to polar or charged residues located within the seven putative transmembrane helices in the COOH-terminal half of the 100-kDa subunit. These residues were selected in an effort to identify buried polar or charged amino acids that might directly contribute to the proton conduction pathway in the V-ATPase complex. Polar residues were changed to nonpolar residues of similar size (for example substituting alanine for serine or phenylalanine for tyrosine), whereas charged residues were replaced either by similar polar amino acids (i.e. glutamine for glutamic acid) or by nonpolar amino acids (usually alanine). It is anticipated that replacement of a polar or charged amino acid that directly participates in proton movement across the membrane will disrupt ATP-dependent proton transport. Of these 22 mutations, only three showed greatly reduced growth at neutral pH (E789Q, D425N, and R735L) (data not shown), with the latter two having an identical growth phenotype to the double knock out strain. Because as little as 20% of the wild type V-ATPase activity is sufficient to rescue the growth phenotype of a Vma Ϫ strain, 2 it was necessary to measure the vacuolar proton pumping and V-ATPase activity in each mutant strain to assess the effect of the mutation on V-ATPase activity. Fig. 1 shows the ATP-dependent proton transport and the bafilomycin-sensitive ATPase activity for isolated vacuoles purified from a ⌬vph1⌬stv1 strain expressing the pRS316 plasmid alone, pRS316 containing the wild type VPH1 gene, or the VPH1 gene bearing the indicated point mutations. Bafilomycin A 1 has been shown to be a specific inhibitor of the V-ATPases (52). As expected on the basis of the growth phenotype, three mutations (D425N, R735L, and E789Q) showed 20% or less of the wild type proton transport and V-ATPase activity, with both D425N and R735L having virtually no detectable activity. Four additional mutations (K538A, K593A, Q634L, and H743A) showed between 20 and 70% of wild type activity, with Q634L having the lowest activity of this group. The remaining mutants all displayed 70% or greater of wild type activity. None of the mutations resulted in significant uncoupling of proton transport from ATP hydrolysis (i.e. a greater loss of proton transport than ATPase activity), although one mutation (Q634L) did decrease ATPase activity to a somewhat greater extent than proton transport, suggesting a possible increase in coupling efficiency.
The most obvious ways in which mutations in the 100-kDa subunit might lead to a decrease in V-ATPase activity is if these mutations resulted in a 100-kDa subunit that was unstable and rapidly degraded or that was unable to correctly assemble with the remaining subunits of the V-ATPase complex. To assess these possibilities, Western blot analysis was performed on isolated vacuoles using antibodies directed against Vph1p and the A subunit of the V-ATPase. It has previously been shown that disruption of the VPH1 gene leads to the inability of the V 1 domain (including the A subunit) to assemble onto the vacuolar membrane (21). Moreover, the absence of any of the V 1 subunits (with the exception of the 54-kDa subunit (14)) leads to loss of assembly of the entire V 1 domain onto the vacuolar membrane (53,54). Thus the pres-ence of the 100-kDa and A subunits on the vacuolar membrane provides a reasonable measure of V-ATPase assembly. Fig. 2 shows a Western blot carried out on vacuolar membranes from the wild type, deletion, and mutant strains using antibodies directed against either Vph1p or the A subunit, whereas Fig. 3 shows the results of quantitative analysis of the Western blot as well as the activity data on each mutant for comparison. As can be seen, two mutants (R735L and Q634L) showed greatly reduced levels of the 100-kDa subunit in the vacuolar membrane as well as greatly reduced levels of A subunit. These mutations thus appear to affect stability of the 100-kDa subunit. Two other mutations (D425N and K538A) show significantly reduced levels of A subunit associated with the vacuolar membrane with only slightly reduced levels of the 100-kDa subunit. These mutations thus appear to have an effect on assembly of the V-ATPase complex. It should be noted from Fig. 3 that the D425N mutant, while still having detectable levels of assembly as assessed by the presence of the A subunit, is completely devoid of proton transport and ATPase activity, suggesting that any V-ATPase that does assemble is inactive. This result suggests that Asp 425 may play a role in both assembly and activity of the V-ATPase.
The remaining mutants have near normal levels of the 100-kDa and A subunits. In particular, the E789Q mutant shows wild type levels of both subunits while possessing less than 20% of the wild type levels of proton transport and ATPase activity. Similarly, examination of Fig. 3 1. Effect of Vph1p mutations on bafilomycin-sensitive ATPase activity and ATP-dependent proton transport in purified vacuolar membrane vesicles. ATPase activity and ATP-dependent proton transport were measured on aliquots of purified vacuolar membrane vesicles containing 5 g of protein as described under "Experimental Procedures." Activities are expressed relative to the ⌬vph1⌬stv1 strain expressing the pRS316 plasmid containing the wild type VPH1 gene (defined as 100%). The specific ATPase activity of these vacuolar membrane vesicles was 3.3 mol ATP/min/mg protein at saturating ATP and 37°C, with approximately 80% of the ATPase activity inhibitable by 10 nM bafilomycin. All of the ATP-dependent proton transport in vacuolar membrane vesicles isolated from cells expressing the wild type Vph1p was inhibitable by 10 nM bafilomycin. No ATP-dependent proton transport or bafilomycin sensitive-ATPase activity was observed in vacuolar membrane vesicles isolated form cells transformed with the vector alone. Each bar represents the average of two or three determinations made on two or three independent vacuolar membrane preparations, with the error corresponding to the standard deviation. WT, wild type.

Vph1p Mutagenesis
proton transport and ATPase activity as predicted on the basis of the amount of assembled V-ATPase. To test whether, for these mutations, the observed decrease in activity might be due to loss of one of the other V-ATPase subunits from the complex, the following experiment was performed. Cells were converted to spheroplasts and metabolically labeled with [ 35 S]Trans-label for 60 min at 30°C followed by cell lysis, detergent solubilization, and immunoprecipitation using the anti-A subunit antibody 8B1-F3 and protein A-Sepharose. As can be seen in Fig. 4, the complete complement of V 1 and V 0 subunits are immunoprecipitated from cells expressing wild type Vph1p, whereas only the V 1 subunits are immunoprecipitated in the vector control. Of the three mutants listed above, H743A and E789Q showed wild type patterns of immunoprecipitation, whereas K593A showed the normal pattern of subunits but with somewhat reduced levels of the V 0 subunits, as predicted on the basis of the data shown in Fig. 3. Thus, all three of these mutations appear to impair activity rather than assembly of the V-ATPase complex. It should be noted, however, that the absence of one of the smaller V-ATPase subunits (i.e. the VMA7 or VMA10 gene products) might not be detectable by this method and that no test for "correct" assembly of the V-ATPase complex has been performed.
To further characterize these mutants, we have determined the K m for ATP and V max values as well as the pH optima for the wild type and for each of the mutants affecting activity as well as for Q634L. As can be seen from the data in Table I, no change greater than 40% was observed in K m values for any of the mutations tested, although the V max values are in good agreement with the activity data shown in Fig. 1. These data indicate that the observed decrease in activity in these mutants is not due to a greatly diminished affinity for ATP. Interestingly, two of the mutations did result in a very significant shift in the pH optima of the enzyme. Thus, whereas the optimum pH for the wild type V-ATPase was approximately 7.2, that for H743A was 8.2, whereas that for E789Q was 9.7. In addition, the activity at the pH optimum of these mutants was approximately 75% (for H743A) and 30% (for E789Q) of the activity at the optimum pH of the wild type.
An important question with regard to the V-ATPases is the mechanism by which they translocate protons. For the bovine coated vesicle V-ATPase, the V 0 domain, which is responsible for proton translocation (58), is composed of four subunits of approximate molecular masses 100, 38, 19, and 17 kDa (subunit c) (57) that are present in a stoichiometry of 100 1 38 1 19 1 c 6 (27). Of these subunits, only the 17-kDa c subunit has been shown to directly participate in proton translocation by virtue of its reaction with dicyclohexylcarbodiimide (59). By analogy with the F-ATPase c subunit (60), dicyclohexylcarbodiimide reacts with a buried carboxyl group located near the middle of the last of four putative transmembrane helices. Although a low level of proton translocation has been reported for the isolated, reconstituted c subunit (61), optimal levels of dicyclohexylcarbodiimide-inhibitable proton transport are only observed when the complete complement of V 0 subunits are reassembled prior to reconstitution (58). The basis for difference in proton conduction properties of the native and reassembled V 0 domains remains uncertain.
In the case of the F-ATPases, the F 0 domain of Escherichia coli is composed of three subunits of molecular masses 30 (a), 17 (b), and 8 kDa (c) that are present in a stoichiometry of a 1 b 2 c 10 -12 (62). Reassembly studies indicate that for F 0 , all three subunits are required to form a functional proton channel (63,64). Mutational analysis has demonstrated that the buried carboxyl group located in the second of two transmembrane helices of the c subunit is critical for proton translocation (60). Moreover, genetic analysis of the a subunit indicates the presence of several buried polar and charged residues, particularly in the last two transmembrane helices, which also play a critical role in proton transport (65,66). Thus, substitutions at Ser 206 , Arg 210 , and Glu 219 in the fifth putative transmembrane helix or His 245 in the sixth putative transmembrane helix of the a subunit significantly impair proton translocation through F 0 . These studies have led to a model in which the proton conduction pathway is composed of several residues of the a subunit, possibly arranged as an amphipathic helix, with the buried carboxylate of the c subunit providing a gate in the conduction pathway (60,65).
By analogy with the F-ATPases, we would predict that the V-ATPases should have some homolog to the a subunit, which would serve a comparably important role in proton translocation. Of the V 0 subunits besides the c subunit (which contains no more buried polar or charged residues than the F 0 c subunit (38,25)), the 36-kDa subunit (Vma6p) contains no putative transmembrane helices (23,67). The sequence of the highly hydrophobic mammalian 19-kDa subunit has not been obtained; however, no yeast counterpart to this subunit has yet been identified. The remaining yeast V 0 subunit, the 13-kDa product of the VMA10 gene (24), resembles the product of the VMA6 gene in possessing no putative transmembrane helices. Thus the only yeast V 0 subunit that has been identified that might serve the role of the a subunit in the V-ATPase complex is the 100-kDa subunit.
FIG. 2. Effect of Vph1p mutations on stability of the 100-kDa subunit and association of the 69-kDa A subunit with the vacuolar membrane. Vacuolar membrane vesicles (5 g) isolated from the ⌬vph1⌬stv1 strain expressing the wild type VPH1 gene, the VPH1 gene bearing the indicated mutations, or the vector alone were subjected to SDS-polyacrylamide gel electrophoresis on a 12% acrylamide gel followed by transfer to nitrocellulose and Western blot analysis using the monoclonal antibody 10D7 against Vph1p or 8B1-F3 against the 69-kDa A subunit as described under "Experimental Procedures." WT, wild type.

Vph1p Mutagenesis
The 100-kDa subunit is composed of an NH 2 -terminal hydrophilic domain of 45 kDa and a COOH-terminal hydrophobic domain of 55 kDa containing 6 -7 putative transmembrane helices (21,22,68). In addition to the two yeast 100-kDa subunit genes (VPH1 and STV1), three additional cDNAs from mouse, rat, and bovine have been cloned (68 -70), with 40 -95% amino acid identity observed between pairs of these sequences (22). Based upon hydropathy analysis (21), a tentative model for the folding of the 100-kDa subunit in the membrane is shown in Fig. 5. The amino-terminal hydrophilic domain has been placed on the lumenal side of the membrane based upon two observations. First, labeling of the coated vesicle 100-kDa subunit by membrane impermeant reagents is only observed after detergent permeabilization of the membrane (27), suggesting that the large hydrophilic domain is sequestered within the lumen of the coated vesicle. Second, proteolysis of the 100-kDa subunit by trypsin in intact coated vesicles results in cleavage at a site between transmembrane helices five and six. 3 Because coated vesicles are oriented with the cytoplasmic surface exposed, this places the loop between H5 and H6 on the cytoplasmic side of the membrane. Tracing of the polypeptide back to the amino terminus places the amino-terminal domain on the lumenal side of the membrane. There are other data, however, that make this assignment tentative. Thus protease treatment of intact yeast vacuoles leads to disappearance on Western blots of any band recognized by a polyclonal antiserum raised against a peptide located in the amino-terminal domain of Vph1p, suggesting that this epitope is exposed on the cyto-plasmic side of the vacuole. 4 Further work is thus necessary to resolve the actual orientation of the 100-kDa subunit in the membrane.
Like the F 0 a subunit, the 100-kDa subunit possesses multiple polar and charged residues located within the putative transmembrane helices in the hydrophobic COOH-terminal domain. To test whether the 100-kDa subunit might be playing an analogous role to the a subunit in proton translocation, we carried out site-directed mutagenesis of 22 such polar and charged residues in this domain. The residues selected for mutagenesis (shown circled in Fig. 5) are all conserved between the available 100-kDa sequences from yeast, mouse, rat, and bovine.
Interestingly, most of the mutations tested did not show any impairment of proton translocation. These residues are presumably not individually critical for proton translocation by the V-ATPase, although it is possible that replacement of several of these residues together might impair proton transport. Mutations at Arg 735 and Gln 634 in H6 and the border of H5, respectively, led to the nearly complete absence of the 100-kDa subunit in the vacuolar membrane, suggesting that the proteins containing these mutations were either unstable and rapidly degraded or else mistargeted to some other intracellular membrane. Given the results that suggest that the 100-kDa subunit possesses targeting information in yeast (22), the latter possibility should not be ruled out. However, the fact that for most integral membrane proteins in yeast targeting to the vacuole represents the default pathway (71) makes it more 3 I. Adachi and M. Forgac, unpublished data. 4 M. Manolson, unpublished observations.

FIG. 3. Comparison of effects of mutations in Vph1p on bafilomycin-sensitive ATPase activity, ATP-dependent proton transport, and the presence of A subunit on isolated vacuolar membrane vesicles.
Bafilomycin-sensitive ATPase activity and ATP-dependent proton transport were measured on vacuolar membrane vesicles (5 g of protein) as described in the legend to Fig. 1. Each bar represents the average of two determinations made on a single vacuolar preparation. An aliquot of the same preparation containing 5 g of protein was analyzed by Western blot using the antibody 8B1-F3 against the 69-kDa A subunit as described in the legend to Fig. 2, and the resultant blot was quantitated using an IS-1000 Digital Imaging System from Alpha Innotech Corporation. WT, wild type.

Vph1p Mutagenesis
likely that these mutations have affected stability of the 100-kDa subunit. Mutations at Asp 425 and Lys 538 in H1 and at the border of H3, on the other hand, did not prevent folding and targeting of the 100-kDa subunit to the vacuolar membrane but did interfere with proper assembly of the V-ATPase complex as demonstrated by the reduced level of the peripheral A subunit on the vacuolar membrane. Because these four mutations prevented the appearance of a V-ATPase complex in the vacuolar membrane, it is not possible to determine the role of the corresponding residues in proton transport.
The mutations of greatest interest are those that inhibited proton transport and ATPase activity but did not have obvious effects on assembly or stability of the V-ATPase complex. The three residues that fell into this last category are Lys 593 at the border of H4, His 743 in H6, and Glu 789 in H7. For both K593A and H743A, proton transport and ATPase activity were only 50% of that predicted on the basis of the amount of assembly observed, whereas for E789Q, the V-ATPase complex was vir- FIG. 4. Effect of Vph1p mutations on V-ATPase assembly. Yeast cells (the ⌬vph1⌬stv1 strain) expressing the wild type VPH1 gene, the indicated mutations or the vector alone were grown overnight in methionine-free medium followed by conversion to spheroplasts and incubation with [ 35 S]Trans-label (50 Ci/5 ϫ 10 6 spheroplasts) for 60 min at 30°C. Spheroplasts were then pelleted and lysed in phosphate-buffered saline with C 12 E 9 , and the V-ATPase immunoprecipitated using the monoclonal antibody 8B1-F3 directed against the 69-kDa A subunit and protein A-Sepharose followed by SDS-polyacrylamide gel electrophoresis on a 12% acrylamide gel and autoradiography as described under "Experimental Procedures." The positions of the V-ATPase subunits are indicated and were confirmed by comparison with the migration of 14 C-labeled molecular mass standards. WT, wild type. The orientation of the 100-kDa subunit in the membrane is based upon labeling and proteolysis data obtained for the 100-kDa subunit of the bovine coated vesicle V-ATPase (see text). In particular, the position at which trypsin cleaves the bovine 100-kDa subunit from the cytoplasmic side of the membrane is indicated. Residues that were mutated in this study are circled. Two mutations (R735L and Q634L) affected stability of the 100-kDa subunit, two mutations (D425N and K538A) affected assembly of the V-ATPase complex, whereas three mutations (H743A, K593A, and E789Q) affected proton transport and ATPase activity of the V-ATPase complex. K593A, Q634L, H743A, and E789Q pH optima were determined by measurement of bafilomycin-sensitive ATPase activity in purified vacuolar membranes (5 g of protein) over the pH range of 5.5-10.5 in the assay mixture. K m for ATP and V max values were determined for bafilomycin-sensitive ATPase activity on purified vacuolar membranes (5 g of protein) over a range of ATP concentrations of 0.5-10 mM. MgCl 2 was varied such that the total Mg 2ϩ concentration was in all cases 1.0 mM higher than the ATP concentration. This ensured that most of the ATP was present as the MgATP complex but avoided the possibility of inhibition of activity due to high concentrations of free Mg 2ϩ . The V max values are reported relative to the wild type, which had a specific activity of 3.3 mol ATP/min/mg protein. The values shown are the average of two independent determinations (at five ATP concentrations each) on a single vacuolar membrane preparation for each mutant, with the errors corresponding to the average deviation from the mean.

Vph1p Mutagenesis
tually completely devoid of activity. In addition, both the H743A and E789Q mutations resulted in a significant change in the pH optimum of the enzyme. While it is difficult to assign a precise interpretation to these findings, they may reflect an alteration in the environment of residues whose protonation state is important for transport to occur. Interestingly, as with the F 0 a subunit, positively and negatively charged residues in the last two putative transmembrane helices appear to be important for activity, possibly serving to line a polar channel necessary to allow protons to gain access to the buried carboxyl group of the c subunit. These results suggest that the 100-kDa subunit may serve an analogous role to the a subunit in the V-ATPase complex. It is important to recognize, however, that further work will be required to determine whether the residues identified play a direct role in proton translocation or whether they serve some other function in the V-ATPase complex, such as coupling of proton transport to ATP hydrolysis. We have previously presented data that suggest that the 100-kDa subunit may also possess the binding site for the specific V-ATPase inhibitor bafilomycin A 1 (58). Thus, reconstituted V 0 or isolated 100-kDa subunit are both able to protect the intact V-ATPase from inhibition by bafilomycin. When the 100-kDa mutants constructed in the present study were tested for their sensitivity to 1 nM bafilomycin (a subsaturating concentration that inhibits 40 -50% of the V-ATPase activity in yeast vacuoles), no significant differences in bafilomycin sensitivity were observed. Obviously mutations that resulted in complete loss of activity (such as D425N and R735L) could not be tested for bafilomycin sensitivity, but none of the remaining residues appears to be critical for bafilomycin binding to the V-ATPase. It is possible that the bafilomycin binding site might reside on the soluble amino-terminal domain or that hydrophobic rather than hydrophilic residues in the integral domain are important in bafilomycin binding. Further studies will be necessary to resolve this question.