Expression and Function of the Mouse V-ATPase d Subunit Isoforms*

We have identified a cDNA encoding a novel isoform of the mouse V-ATPase d subunit (d2). The protein encoded is 350 amino acids in length and shows 42 and 67% identity to the yeast d subunit (Vma6p) and the mouse d1 isoform, respectively. Reverse transcriptase-PCR analysis using isoform-specific primers demonstrate that d2 is expressed mainly in kidney and at lower levels in heart, spleen, skeletal muscle, and testis. Although d1 and d2 show similar levels of sequence homology to Vma6p, only the d1 isoform can complement the phenotype of a yeast strain in which VMA6 has been disrupted when cells are grown at 30 °C. The d2 isoform, however, can complement the vma6 (cid:1) phenotype when cells are grown at 25 °C. Moreover, partial assembly of the V-ATPase complex on the vacuolar membrane can be detected under these conditions, although assembly is sig-nificantly lower than that observed for the strain expressing Vma6p. This reduced assembly is also re-flected in a reduced level of concanamycin-sensitive ATPase activity and proton transport in isolated vacuoles. Comparison of the kinetic properties of V-ATPase complexes containing Vma6p and d1 demonstrate that although the K m for ATP hydrolysis is similar (0.26 and 0.31 m M , respectively), the coupling ratio (proton trans-port/ATP hydrolysis) is (cid:1) 3–6-fold higher for d1-contain-ing complexes than for Vma6p-containing complexes. These results suggest that subunit d may play a role in coupling of proton transport and ATP hydrolysis. was measured using a coupled spectrophotometric assay in the presence or absence of 1 (cid:2) M concana- mycin, as previously described (22). ATP-dependent proton transport was measured in transport buffer (25 m M Mes-Tris (pH 7.2) 5 m M MgCl 2 ) using the fluorescence probe 9-amino-6-chloro-2-methoxyacri- dine in the presence or absence of 1 (cid:2) M concanamycin, as previously described (22). SDS-polyacrylamide gel electrophoresis was carried out as described by Laemmli (32).

The vacuolar (H ϩ )-ATPase (or V-ATPase) 1 functions as an ATP-dependent proton pump to acidify intracellular compartments in eukaryotic cells. The V-ATPases are present in a variety of intracellular compartments, including clathrincoated vesicles, endosomes, lysosomes, Golgi-derived vesicles, chromaffin granules, synaptic vesicles, and the central vacuoles of yeast, Neurospora, and plants (1)(2)(3)(4)(5)(6)(7)(8). Vacuolar acidification plays an important role in many cellular processes, including receptor-mediated endocytosis, intracellular targeting, protein processing and degradation, and coupled transport. In certain mammalian cells, V-ATPases also function in the plasma membrane to transport protons from the cytoplasm to the extracellular environment (9 -13). In osteoclasts, plasma membrane V-ATPases play a role in bone resorption (11), whereas in intercalated cells in the kidney they function in renal acidification (9). V-ATPases in the plasma membrane of tumor cells have also been implicated in metastasis (13).
The V-ATPases from fungi, plants, and animals are structurally very similar and are composed of two functional domains, V 1 and V 0 (1)(2)(3)(4)(5)(6)(7)(8). The V 1 domain is a peripheral complex with molecular mass of 640 kDa composed of eight different subunits of molecular mass 70 -14 kDa (subunits A-H) that is responsible for ATP hydrolysis. The V 0 domain is a 260-kDa integral complex composed of five subunits of molecular masses 100 -17 kDa (subunits a, d, c, cЈ, and cЉ) that is responsible for proton translocation. In yeast cells, all subunits are encoded by single genes with the exception of subunit a, which is encoded by two genes, VPH1 and STV1 (14,15).
Information about the function of four of the five V 0 subunits has been obtained by site-directed and random mutagenesis. Thus, each of the three proteolipid subunits (c, cЈ, and cЉ) have been shown to contain a single, buried glutamate residue that is essential for proton translocation (16,17). In addition, the a subunit contains a number of buried charged residues that influence proton transport (18 -20), as well as a single buried arginine residue in transmembrane segment seven, which is required for proton translocation (20). Subunit a has also been shown to play a role in intracellular targeting of the V-ATPase (15,21) and in controlling assembly with V 1 , coupling of proton transport and ATP hydrolysis, and reversible dissociation of the V 1 and V 0 domains in response to glucose depletion (21,22).
Despite this information regarding most of the V 0 subunits, virtually nothing is known about the function of subunit d. The gene encoding subunit d was initially cloned from bovine adrenal medulla (23) and subsequently from yeast as the VMA6 gene, where its disruption was shown to lead to a typical Vmaphenotype (24). Sequence analysis revealed the presence of no obvious transmembrane segments (23,24), despite the fact that it remains firmly attached to the V 0 domain upon dissociation of V 1 using chaotropic agents (25) or, in vivo, upon glucose withdrawal (26). This suggests that subunit d is bound to the V 0 domain through protein-protein interactions rather than by integration into the membrane. Mild proteolysis results in rapid cleavage of subunit d in intact clathrin-coated vesicles, suggesting that it is exposed on the cytoplasmic side of the membrane (27).
In addition to yeast and bovine, the gene encoding subunit d has also been cloned and sequenced from humans and plants (28,29). Like numerous other V-ATPase subunits in higher eukaryotes, subunit d appears to be encoded by multiple genes (28). In this paper, we have identified a novel isoform of subunit d in mouse, investigated its pattern of expression, and begun to probe the function of this subunit by heterologous expression in yeast.

EXPERIMENTAL PROCEDURES
Materials-Escherichia coli culture media was purchased from Difco Laboratories. Restriction endonucleases, T4 DNA ligase, and other molecular biology reagents were from Invitrogen, Promega, and New England Biolabs. Most other chemicals were purchased from Sigma.
Isolation of cDNA Clones-The cDNA encoding the mouse d subunit isoform d2 was identified in the expression sequence tag data base and expressed sequence tag clone uj32h10 was obtained from the American Tissue Culture Collection (Manassas, VA). The cDNA encoding the mouse d1 isoform (uj10a07) was also obtained from the American Tissue Culture Collection.
Test of the Ability of the Mouse d Subunit Isoforms to Complement the Phenotype of the Yeast VMA6 Deletion Strain-Yeast cells lacking the functional endogenous VMA6 gene were made from YPH500 (MAT␣ ura3-52 lys2-801 ambe r ade2-101 ochre trp1-⌬63 his3-⌬200 leu2-⌬1) by replacing the coding region of VMA6 with the TRP1 gene. The yeast VMA6 gene was isolated from the YPH500 genomic DNA by PCR using the following primers: forward, ctccagcatacatattaagttgca; reverse, gtccaggttaggcgaggatattact. The amplified fragments were cloned into the TOPO-pCR2.1 vector (Invitrogen). The VMA6 gene was excised from the resulting plasmid at XbaI and EcoRI sites and integrated into the XbaI and EcoRI sites of the pRS413 vector for expression. The mouse d subunit isoform cDNAs were inserted just downstream of the GPD gene promoter in the pRS413 vector (ATCC). Cells were transformed with plasmids pRS413, pRS413-VMA6, pRS413-GPD-m d1, and pRS413-GPD-m d2 using the lithium acetate method and selected on SD histidine minus plates. Growth phenotypes of the transformants were assessed on YPD plates buffered with 50 mM KH 2 PO 4 /succinic acid to either pH 7.5 or 5.5.
RT-PCR-RT-PCR was performed using cDNA prepared from different mouse tissues or different developmental stages of mouse embryos and the following subunit-specific primers: m d1 forward AGTCTAGT-TGCGTGCGGGCAGAT; m d1 reverse, ATAAGTTATGAAGTCCAGGA-AGCTGGCGA; m d2 forward, TTCAGTTGCTATCCAGGACTCGGA; m d2 reverse, GCATGTCATGTAGGTGAGAAATGTGCTCA.

Identification of Transcription Initiation Sites for d Subunit Isoform
Genes-To determine the transcription initiation site for each d subunit isoform gene, 5Ј-rapid amplification of cDNA ends was performed using the First-Choice rapid amplification of cDNA ends kit (Ambion). Reactions were performed using the manufacturer's recommended protocol, and mouse kidney poly(A) RNA was purchased from Ambion (Palo Alto, CA). Amplified fragments were cloned into the TOPO-pCR2.1 vector (Invitrogen) and sequenced. Primers specific for the genes encoding the d subunit isoforms (m d1Rv and m d2 Rv2) that were used for amplification are indicated above.
Isolation of Vacuole-enriched Membranes-Subcellular fractionation of organelles from yeast cells expressing Vma6p, the mouse d1 isoform, or the mouse d2 isoform was performed by differential centrifugation of cell lysates as described previously (30,31). Yeast cells were cultured in SD-HIS medium to an A 600 of 1 at 30 or 25°C. Cells (3 ϫ 10 9 ) were converted to spheroplasts by treatment with 150 g/ml Zymolase 100T in 50 mM potassium phosphate (pH 7.5), 1.2 M sorbitol, 1 mM MgCl 2 followed by lysis in lysis buffer (50 mM Tris-HCl (pH 7.5), 0.2 M sorbitol, 1 mM EDTA) and sedimentation for 5 min at 500 ϫ g to remove unbroken cells. The supernatant was centrifuged for 10 min at 13,000 ϫ g to generate the vacuole-enriched membrane fraction, and the supernatant (S13) was centrifuged at 100,000 ϫ g for 30 min, and the supernatant of this high speed sedimentation was used as the cytosol fraction. The vacuole-enriched fraction was resuspended in sample buffer (8 M urea, 5% SDS, 40 mM Tris-HCl (pH 7.4) 0.1 mM EDTA, 0.4 mg/ml bromphenol blue, 10% 2-mercaptoethanol). Proteins in the cytosol fraction were precipitated by the addition of trichloroacetic acid to a final concentration of 5%, collected by centrifugation for 5 min at 13,000 ϫ g, washed twice, and resuspended in sample buffer. The proteins were separated by SDS-PAGE on 12.5% acrylamide gels, transferred to nitrocellulose filters, and probed with the antibodies indicated in the figure legends. Blots were developed using appropriate secondary antibodies conjugated to horseradish peroxidase and the chemiluminescence detection kit from Kirkegaard and Perry Laboratories.
Vacuole Membrane Isolation-Vacuole membrane vesicles were isolated as described previously (22) except that all incubations carried out prior to treatment with Zymolyase were performed at 25°C. The yeast vma6⌬ strain transformed with pRS413-VMA6, pRS413-TEF-d1, and pRS413-TEF-d2 were grown to an A 600 of 0.8 at 25°C. Cells were pelleted, washed once with water, and resuspended and incubated in alkaline buffer (10 mM dithiothreitol, 100 mM Tris-HCl, pH 9.4) at 25°C for 15 min. Cells were pelleted, resuspended in YEPD medium containing 0.7 M sorbitol, 2 mM dithiothreitol, 100 mM Mes-Tris (pH 7.5), and 50 g/ml Zymolase 100T, and incubated at 30°C with gentle shaking for 60 min. Spheroplasts were washed with YEPD medium containing 0.7 M sorbitol, 2 mM dithiothreitol, osmotically lysed by incubation in lysis buffer (50 mM Tris-HCl (pH 7.5) 0.2 M sorbitol, 1 mM EDTA), and the vacuoles were isolated by flotation on two consecutive Ficoll gradients as previously described (22). Protein concentrations were measured by the BCA protein assay (Pierce).
Detection of V-ATPase Subunits Present on Isolated Vacuolar Membranes-Vacuolar membranes were isolated from the yeast vma6⌬ strain transformed with pRS413-VMA6, pRS413-TEF-d1, or pRS413-TEF-d2 and were subjected to SDS-PAGE followed by Western blot analysis using the mouse monoclonal antibodies 10D7 against Vph1p, 8B1-F3 against Vma1p, and 13D11-B2 against Vma2p (Molecular Probes, Inc.). Blots were also probed with rabbit polyclonal antibodies against Vma6p, Vma7p, Vma8p, Vma10p, and Vma13p (all generously provided by Dr. Tom Stevens), Vma4p (a generous gift of Dr. Dan Klionsky), and Vma5p (a generous gift of Dr. Patricia Kane). Following removal of unbound primary antibodies by washing, blots were incubated with horseradish peroxidase-conjugated secondary antibodies (Bio-Rad) and developed using a chemiluminescence detection method obtained from Kirkegaard and Perry Laboratories.
Other Methods-Protein concentrations were determined by the BCA protein assay (Pierce). ATPase activity was measured using a coupled spectrophotometric assay in the presence or absence of 1 M concanamycin, as previously described (22). ATP-dependent proton transport was measured in transport buffer (25 mM Mes-Tris (pH 7.2) 5 mM MgCl 2 ) using the fluorescence probe 9-amino-6-chloro-2-methoxyacridine in the presence or absence of 1 M concanamycin, as previously described (22). SDS-polyacrylamide gel electrophoresis was carried out as described by Laemmli (32).

Identification of the cDNA Encoding the Mouse V-ATPase d2
Subunit Isoform-The cDNA sequence encoding the d subunit of the V-ATPase was first reported in bovine (23), and was subsequently identified in other species (24,(33)(34)(35)(36)(37). Genomic sequencing has recently demonstrated the existence of a second isoform of subunit d, d2, which has been identified in humans (28) and plants (29). A search of the mouse expressed sequence tag and genome data bases revealed several clones encoding the second isoform of subunit d in mouse. One of these clones (uj32h10) was completely sequenced and the deduced amino acid sequence was aligned with the other d subunit isoform, d1, and the yeast d subunit, Vma6p (Fig. 1). The protein encoded contains 350 amino acid residues and has a predicted molecular mass of 40,459. The gene encoding the mouse d1 isoform is located on chromosome 4, whereas the gene encoding d2 is located on chromosome 8. Based on the genomic sequence, the genes encoding the d1 and d2 isoforms contain seven introns, with the intron/exon borders located at the same sites (Fig. 1). The overall identity (and similarity) to the mouse d1, human d1 (33), human d2 (28) 2). These results suggest that clone uj32h10 encodes the mouse d2 subunit isoform of the V-ATPase. Comparison of the amino acid sequences between species reveals that isoforms of the V-ATPase d subunit may have arisen more than once during the course of evolution (Fig. 2).
Tissue Distribution of the Mouse d Subunit Isoforms-To investigate the tissue distribution of mRNA for the two d subunit isoforms, RT-PCR was performed using isoform-specific primers and RNA isolated from eight mouse tissues and several developmental stages. As can be seen in Fig. 3, the d1 isoform is ubiquitously expressed in all tissues tested, whereas the d2 isoform is expressed predominantly in kidney, although some expression is also detected in lung, testis, skeletal muscle, and heart, in agreement with previous observations (23). With respect to developmental stage, the d1 isoform is expressed throughout early development, whereas the d2 isoform is expressed heavily at day 7 followed by a disappearance and a gradual recovery to original levels by day 17. Development of the mouse kidney begins after embryonic day 14. These results suggest that expression of the d2 isoform from day 15 is correlated with kidney development. Expression of the d2 isoform at day 7 suggests that this isoform may also serve some other function in early mouse development.
The 5Ј-Untranslated Region of the Mouse and Human d Subunit Isoforms-To address the basis for the tissue-specific expression of the d2 isoform, the nucleotide sequence of 5Јupstream regions of the mouse and human genes were compared. The 5Ј-upstream sequence of each gene was identified from human and mouse genome sequence data bases reported at NCBI. Harr-plot analysis indicates that although the 5Јupstream regions of the genes encoding the mouse d1 and d2 isoforms have no obvious homology (data not shown), both of the d1 and d2 5Ј-upstream regions show significant sequence homology between mouse and human (Fig. 4). The transcription initiation sites were identified using 5Ј-rapid amplification of cDNA ends, primers specific for each gene and mouse kidney cDNA as the template and are shown in Fig. 4, a and b, by closed circles. As can be seen, for the gene encoding the d1 these sites fall within one of the two regions of high sequence conservation, whereas for the gene encoding d2, transcription starts just downstream of the 3Ј most region of homology.
Complementation of the Yeast vma6⌬ Phenotype by the Mouse d Subunit Isoforms at 30°C-To analyze the function of the mouse d subunit isoforms, we expressed each of these isoforms in a yeast strain (vma6⌬) in which the endogenous VMA6 gene has been disrupted. Despite the very similar level of sequence identity between the two mouse proteins and  Vma6p (45 and 42% for d1 and d2, respectively), only d1 was able to complement the growth defect of this strain when assayed at pH 7.5 and 30°C (Fig. 5a). There are several possible reasons to account for the inability of the d2 isoform to complement the Vma-phenotype of the vma6⌬ strain, including loss of activity of the V-ATPase and failure to assemble a complex. As an initial assessment of the assembly competence of the heterologously expressed proteins, Vma6p, the mouse d1 isoform or the mouse d2 isoform were expressed in the vma6⌬ strain, the cells were homogenized and the whole cell lysate was fractionated by differential sedimentation to give a cytosolic fraction and a vacuole-enriched fraction, as previously described (30,31). This method was employed rather than the conventional vacuole isolation procedure because it is gentler and therefore less likely to disrupt V-ATPase complexes. These fractions were then separated by SDS-PAGE and Western blot analysis was performed using antibodies against Vma1p (a marker for the V 1 domain) and Vph1p (a marker for the V 0 domain). As can be seen in Fig. 5b, in cells expressing the d2 isoform no assembly of subunit A (Vma1p) onto vacuoles occurs. By contrast, for cells expressing either Vma6p or the d1 isoform, assembly of subunit A onto vacuolar membranes is observed. Vph1p is observed in both the whole cell lysate and on vacuolar membranes of cells expressing Vma6p but is not observed in cells lacking subunit d or in cells expressing the d2 isoform. This is consistent with previous observations indicating that a stable V 0 domain is not formed in the absence of subunit d (24) and indicates that the d2 isoform does not allow for the formation of a stable V 0 domain. For the d1 isoform, a low level of Vph1p is observed associated with vacuoles, suggesting that the V 0 domain assembled with d1 is of lower stability than for Vma6p. Nevertheless, the d1-containing V 0 domain is sufficiently stable to give near normal assembly of subunit A onto the vacuolar membrane. It should be noted that while the antibody against Vma6p does recognize the d1 isoform, no evidence for cross-reactivity with the d2 isoform has been obtained.
Complementation of the Yeast vma6⌬ Phenotype by the Mouse d Subunit Isoforms at 25°C-To further test the ability of the mouse d subunit isoforms to complement the loss of the yeast Vma6p, the above complementation experiments were repeated, but at 25°C instead of 30°C. As shown in Fig. 6a, at this lower temperature both d1 and d2 were able to complement the loss of Vma6p. To examine the assembly properties of the V-ATPase containing the d1 and d2 isoforms, cell fractionation and Western blot analysis was performed on cells grown at the lower temperature. As can be seen in Fig. 6b, assembly of subunit A is now observed on vacuolar membranes for both d1 and d2, although to a lower degree for d2 than d1. Interestingly, for the d1 isoform, a significant amount of d1 is observed in the cytosol fraction. This may be because of the presence of an excess of d1 over the amount necessary to form V 0 domains or to a partial dissociation of d1 from the other yeast V 0 subunits.
Enzymatic Properties of V-ATPase Complexes Containing Different d Subunit Isoforms-Because V-ATPase complexes containing as little as 20 -25% of wild type activity are able to complement the growth phenotype of the Vma-strain (38,39), it is necessary to examine the enzymatic properties of the V-ATPase from yeast expressing different forms of the d subunit. Vacuolar membranes were isolated from the vma6⌬ strain expressing either Vma6p, the mouse d1 isoform or the mouse d2 isoform after growth at 25°C. Concanamycin-sensitive ATPase activity was then measured over a range of ATP concentrations from 0.05 to 2.0 mM using a coupled spectrophotometric assay and concanamycin-sensitive, ATP-dependent proton transport was measured from the initial rate of fluorescence quenching using the fluorescence dye 9-amino-6-chloro-2-methoxyacridine. The tightness of coupling of proton transport and ATP hydrolysis was also estimated by calculating the ratio of fluorescence quenching to ATPase activity. As can be seen from the results in Table I, the d1 isoform gave a V-ATPase complex with a K m for ATP that was very similar to that observed for the complex containing Vma6p (250 -300 M), but displayed an ATPase activity only 3-4% of wild type. The proton transport obtained with the d1 isoform, however, was 22% of that obtained with Vma6p, when measured at the same protein concentration. This gave a coupling ratio for the d1-

FIG. 4. 5-Upstream regions and transcription initiation sites of the genes encoding the mouse d1 and d2 subunit isoforms.
Nucleotide sequence of the 5Ј-upstream regions of the genes encoding the mouse d1 (a) and d2 (b) subunits (accession numbers NW_000349 and NW_000204). Regions highly homologous between mouse and human (d1, NT_010478 and d2, NT_008117) sequences are indicated by boxes. Transcription initiation sites, which were identified by 5Ј-rapid amplification of cDNA ends analysis using mouse kidney RNA, are indicated by closed circles. The number of circles indicate the number of 5Ј-ends obtained by this analysis.
containing vacuoles of 80.4, ϳ6-fold higher than for the wild type vacuoles (13.4). If the wild type vacuoles were diluted 21-fold to give approximately the same level of ATPase activity as was observed for the d1 mutant, a coupling ratio of 24.6 was obtained, corresponding to a 3-fold difference in coupling ratio between the wild type and d1 mutant. These results suggest that the V-ATPase complex containing the mouse d1 isoform was more tightly coupled than the wild type complex. Whereas ATPase activity obtained with the d2 isoform was below the detection level, some concanamycin-sensitive proton transport was still observed.
Assembly of the V-ATPase Complexes Containing the Mouse d1 or d2 Isoforms-To further evaluate the assembly status of V-ATPase complexes containing the mouse d1 and d2 isoforms, vacuoles were isolated from cells grown at 25°C, and Western blotting was performed using antibodies against subunits A, B, C, D, E, F, G, and H in the V 1 domain and subunit a in the V 0 domain. As can be seen in Fig. 7, the d1 isoform showed reduced levels of V-ATPase subunits on vacuolar membranes relative to Vma6p. For the d2 isoform, there was an even greater reduction in subunit levels. These results suggest that although the V-ATPase complexes containing the d2 isoform are sufficiently active to complement the Vma-phenotype of the vma6⌬ strain and sufficiently stable to show partial assembly using the fractionation procedure employed in Fig. 6, they are not sufficiently stable to survive the conventional vacuole isolation procedure used to prepare vacuolar membranes for the experiment shown in Fig. 7.   FIG. 5. Expression of the mouse d1 isoform but not the mouse d2 isoform complements the Vma-phenotype of the vma6⌬ strain at 30°C. a, yeast cells lacking the VMA6 gene (vma6⌬) were transformed with pRS413 (Ϫ), pRS413-VMA6 (Vma6), pRS413-GPD-d1 (m d1), or pRS413-GPD-d2 (m d2). The growth phenotype of the transformants was tested on YPD plates buffered with 50 mM KH 2 PO 4 /succinic acid to either pH 5.5 or 7.5 at 30°C. b, distribution of V-ATPase subunits in vma6⌬ yeast transformed with pRS413 (Ϫ), pRS413-VMA6 (Vma6), pRS413-GPD-d1 (d1), or pRS413-GPD-d2 (d2). Whole cell lysates (Lys) were fractionated into a vacuole-enriched fraction (Vac) and a cytosol fraction (Cyt) as described under "Experimental Procedures." Following SDS-PAGE, Western blotting was performed using antibodies against subunit A (Vma1p), subunit a (Vph1p), and subunit d (Vma6p), as described. The anti-Vma6p antibody showed cross-reaction with the d1 but not the d2.

DISCUSSION
Of the five V 0 subunits, the least characterized in terms of function is subunit d. Thus all three proteolipid subunits (c, cЈ, and cЉ) as well as subunit a have been shown to participate in proton translocation (16 -20). Subunit a has also been shown to function in intracellular targeting and in control of coupling efficiency and in vivo dissociation (15,21,22). Although disruption of the gene encoding subunit d in yeast (VMA6) prevents assembly of both the intact complex and the V 0 domain (24), no specific function has yet been assigned to this subunit. Subunit d does not appear to be embedded in the lipid bilayer (23,24), but is instead exposed on the cytoplasmic side of the membrane (27), where it can be released from the membrane by treatment at alkaline pH or with urea (24). It may, therefore, contribute to the cytoplasmic mass observed in electron microscopic images of the V 0 domain (40).
Multiple isoforms have been identified for several of the V-ATPase subunits in higher eukaryotes, including subunits B, C, E, G, and a (11,28,39,(41)(42)(43)(44)(45)(46)(47). Recently, two isoforms of subunit d have been identified in humans and plants (28,29). We have isolated a cDNA encoding a second isoform of subunit d in mouse (d2), and have demonstrated that it is expressed predominantly in kidney, although expression is also detected in lung, testis, skeletal muscle, heart, and spleen (Fig. 3). This expression pattern is similar to that observed for two other subunit isoforms, namely B1 and a4, which have been suggested to be "kidney specific" (41,42,47,48). 2 Comparison of the 5Ј-upstream regions of the mouse and human genes encoding the d1 and d2 isoforms reveals the presence of several isoform-specific regions that are conserved between species (Fig. 4). These regions may play a role in tissue-specific expression of the corresponding genes (49). Consistent with this idea, the sites of transcription initiation are located in or near these conserved regions. Previously, the 5Ј-upstream region of the genes encoding subunits c, cЉ, and B2 were analyzed and possible promoter sequences or cis-elements were reported (50,51). However, no homology is observed in the 5Ј-upstream regions of these genes when compared with those encoding the d subunit isoforms (not shown).
To try to gain insight into the function of subunit d, each mouse d subunit isoform was expressed in a yeast strain disrupted in VMA6. Interestingly, the d1 isoform is able to complement the Vma-phenotype in this strain at both 25 and 30°C, whereas the d2 isoform complemented the phenotype only at the lower temperature. Western blot analysis of vacuole-enriched membranes detected no assembly of V-ATPase complexes in the strain expressing d2 at 30°C (Fig. 5), indicating that either d2 is not stable or the V-ATPase complexes containing d2 are not stable under these conditions. By contrast, assembly of d2-containing complexes on vacuole-enriched membranes is detected at 25°C, although at lower levels than for cells expressing either Vma6p or d1 (Fig. 6). Moreover, d2-containing complexes display greatly reduced stability, even at 25°C, as indicated by the very low levels of V-ATPase subunits present on purified vacuolar membranes (Fig. 7). Assembly of d1-containing complexes is observed at both 25 and 30°C, although the level of V-ATPase subunits on purified vacuolar membranes is lower for cells expressing d1 than for Vma6p-expressing cells (Fig. 7). These results indicate that both mouse isoforms are able to replace the yeast d subunit in assembly of V-ATPase complexes, although the d1 isoform is more effective in this capacity. A temperature-dependent difference in the ability of E subunit isoforms from mouse to replace the yeast VMA4 gene has also been reported (45).
Consistent with the above results, V-ATPase complexes containing the d1 isoform showed much lower levels of concanamycin-sensitive ATPase activity than Vma6p-containing complexes, and virtually no V-ATPase activity was detectable on purified vacuolar membranes from d2-expressing cells. It was therefore not possible to characterize the activity properties of V-ATPase complexes containing d2. Interestingly, V-ATPase complexes containing the d1 isoform showed 3-6-fold greater coupling of proton transport to ATP hydrolysis than Vma6pcontaining complexes (Table I). These results suggest that the yeast V-ATPase complex is not optimally coupled under normal conditions, in agreement with previous results (52). In that study, a mutation was identified in the "non-homologous" region of the yeast A subunit that resulted in a 4-fold increase in coupling efficiency relative to the wild type enzyme. Mutations altering coupling of proton transport to ATP hydrolysis have also been identified in a number of other V-ATPase subunits, including subunits D and a (53,20), and a 5-fold difference in coupling ratio has been reported for V-ATPase complexes containing different a subunit isoforms (22). These results suggest that a number of V-ATPase subunits in both the V 1 and V 0 domains act in coordinate fashion to control coupling of proton transport and ATP hydrolysis, and that subunit d may also function in this capacity.
While this manuscript was in preparation, another paper describing the identification of a second isoform of subunit d in mouse appeared (54). Our results are in agreement with this paper in demonstrating high expression of d2 in kidney and in showing that only the d1 isoform was able to complement the Vma-phenotype of the vma6⌬ strain at 30°C. We have extended these results in the current article by demonstrating a temperature-dependent ability of the d2 isoform to complement the Vma-phenotype of the vma6⌬ strain and by characterizing the assembly and activity properties of the hybrid yeast complexes containing the mouse d subunit isoforms.