Compensatory membrane expression of the V-ATPase B2 subunit isoform in renal medullary intercalated cells of B1-deficient mice

Teodor G. Păunescu, Leileata M. Russo, Nicolas Da Silva, Jana Kovacikova, Nilufar Mohebbi, Alfred N. Van Hoek, Mary McKee, Carsten A. Wagner, Sylvie Breton, Dennis Brown


Mice deficient in the ATP6V1B1 (“B1”) subunit of the vacuolar proton-pumping ATPase (V-ATPase) maintain body acid-base homeostasis under normal conditions, but not when exposed to an acid load. Here, compensatory mechanisms involving the alternate ATP6V1B2 (“B2”) isoform were examined to explain the persistence of baseline pH regulation in these animals. By immunocytochemistry, the mean pixel intensity of apical B2 immunostaining in medullary A intercalated cells (A-ICs) was twofold greater in B1−/− mice than in B1+/+ animals, and B2 was colocalized with other V-ATPase subunits. No significant upregulation of B2 mRNA or protein expression was detected in B1−/− mice compared with wild-type controls. We conclude that increased apical B2 staining is due to relocalization of B2-containing V-ATPase complexes from the cytosol to the plasma membrane. Recycling of B2-containing holoenzymes between these domains was confirmed by the intracellular accumulation of B1-deficient V-ATPases in response to the microtubule-disrupting drug colchicine. V-ATPase membrane expression is further supported by the presence of “rod-shaped” intramembranous particles seen by freeze fracture microscopy in apical membranes of normal and B1-deficient A-ICs. Intracellular pH recovery assays show that significant (28–40% of normal) V-ATPase function is preserved in medullary ICs from B1−/− mice. We conclude that the activity of apical B2-containing V-ATPase holoenzymes in A-ICs is sufficient to maintain baseline acid-base homeostasis in B1-deficient mice. However, our results show no increase in cell surface V-ATPase activity in response to metabolic acidosis in ICs from these animals, consistent with their inability to appropriately acidify their urine under these conditions.

  • proton pump
  • immunofluorescence
  • pH homeostasis
  • urinary acidification
  • Atp6v1b1−/− mice

the main mediator of intracellular organelle acidification in eukaryotic cells and of proton (H+) secretion along the distal renal nephron is the ubiquitous vacuolar proton-pumping ATPase (vacuolar, or V-type, H+-ATPase, or V-ATPase). The V-ATPase is a complex enzyme, consisting of two large sectors or domains (V0, the transmembrane domain involved in H+ translocation, and V1, the cytosolic domain, responsible for hydrolyzing ATP), which together contain at least 13 distinct subunits (8, 11, 27, 36, 64). A number of V-ATPase subunits are known to be encoded by different genes and are expressed in mammalian tissues as multiple distinct isoforms. Certain subunit isoforms exhibit a remarkable specificity in terms of tissue and/or cell type expression, and in some cases even with respect to their subcellular localization (31, 33, 46, 52, 54, 59, 60).

The V1 cytosolic sector, which constitutes the catalytic domain of the enzyme, is composed of eight subunits, including three copies of the 70-kDa “A” subunit and three copies of the 56-kDa “B” subunit, both believed to be involved in ATP binding (28, 68). This latter V-ATPase subunit occurs as two highly homologous isoforms (sharing 83% identity in their amino acid sequences in the mouse) encoded, respectively, by two different genes, ATP6V1B1 (or “B1”, initially referred to as the “kidney” isoform), encoded by Atp6v1b1, and ATP6V1B2 (or “B2”, originally described as the “brain” isoform), encoded by Atp6v1b2 (7, 9, 26, 38, 58). Mutations in two of the 22 genes currently known to encode for V-ATPase subunits were found to cause autosomal recessive forms of distal renal tubular acidosis (dRTA, or type I RTA), also often associated with sensorineural hearing loss. These two genes are Atp6v1b1 and Atp6v0a4, which encodes the transmembrane “a” subunit isoform ATP6V0A4, or “a4” (34, 55, 57). dRTA is a genetic disease characterized by impaired H+ secretion by the distal renal nephron, leading to metabolic acidosis, hypokalemia, nephrocalcinosis, bone disease, and growth retardation (6).

To investigate the physiological mechanisms of dRTA, B1−/− mice (deficient in the B1 subunit of the V-ATPase) were engineered as a mouse model for the study of this human disease (25). Interestingly, unlike human dRTA patients (35), B1−/− mice fed a normal diet were found to develop normally, without any phenotype of metabolic acidosis, nephrocalcinosis, or inner ear and hearing abnormalities (24, 25). We also reported that the apical membrane expression of the B2 isoform was significantly augmented in proton-secreting collecting duct (CD) A-type intercalated cells (A-ICs) of B1−/− mice, especially in the inner medulla and in the inner stripe of the outer medulla (25). Taken together, these data led to the hypothesis that, under these conditions, B2-containing V-ATPases could be involved in H+ transport across the apical plasma membrane and, thus, compensate for the absence of the B1 isoform. However, when B1−/− mice were challenged with an acid load (by administering NH4Cl in their drinking water), they developed a severe systemic acidosis and their urine pH was found to be inappropriately alkaline. Functional data showing no concanamycin-inhibitable, i.e., no V-ATPase-mediated, H+ extrusion from cortical collecting duct (CCD) ICs of B1−/− mice also suggested that V-ATPases containing the B2 subunit isoform may be capable of compensating for the absence of B1 under some but not all circumstances (25).

Consequently, we set out to investigate these apparently contradictory findings comprehensively and to elucidate whether (and, if so, how) apical V-ATPase function in maintaining body acid-base homeostasis can be preserved in mice devoid of the V-ATPase B1 subunit isoform. In particular, we aimed to determine whether the presumed compensatory mechanism involves the alternate 56-kDa B2 isoform. Given the absence of V-ATPase-mediated H+ secretion found in CCD ICs of B1−/− mice, this study was focused on V-ATPase B2 subunit expression and subcellular distribution, and V-ATPase function in A-ICs of medullary CDs in mice deficient in the V-ATPase B1 subunit isoform.



Three affinity-purified polyclonal antibodies were used. They were raised against synthetic peptides corresponding to the 10 carboxy-terminal amino acids of ATP6V1A (the V-ATPase 70-kDa A subunit), ATP6V1B1 (the V-ATPase 56-kDa B1 subunit isoform), and ATP6V1B2 (the B2 isoform), respectively, as previously described (30, 45). The rabbit anti-A and anti-B1 polyclonal antibodies were characterized previously (13, 30, 46). The affinity-purified chicken anti-B2 antibody was raised against the same sequence as the previously described rabbit anti-B2 antibody (45) and was found to have the same specificity as this antibody (23).

To identify CD A-ICs, we used an anti-AE1 anion exchanger affinity-purified rabbit polyclonal antibody (2, 10), kindly provided by Dr. Seth Alper (Harvard Medical School and Beth Israel Deaconess Medical Center).

The following affinity-purified secondary antibodies were used: indocarbocyanine (Cy3)-conjugated donkey anti-chicken IgY (H+L) (Jackson ImmunoResearch Laboratories, West Grove, PA) at a final concentration of 1.5 μg/ml and FITC-conjugated goat anti-rabbit IgG (H+L) (Jackson ImmunoResearch Laboratories), at a final concentration of 25 μg/ml.

Tissue preparation.

Adult male mice (30–35 g), wild-type (C57BL6, Jackson Laboratory, Bar Harbor, ME) and B1-deficient (B1−/−), were housed under standard conditions and maintained on a standard diet. Generation and breeding of the original B1−/− founders have been described elsewhere (25). All animals were genotyped by PCR as described previously (25). Where indicated, mice were challenged with an acid load by administering 1.5% (280 mM) NH4Cl/1% sucrose in their drinking water for 24 h. All animal studies were approved by the Massachusetts General Hospital Subcommittee on Research Animal Care, in accordance with the National Institutes of Health, Department of Agriculture, and AAALAC requirements, or by the local Swiss Veterinary Authority (Veterinäramt, Zurich, Switzerland), in accordance with Swiss Animal Welfare Laws.

For immunofluorescence experiments, mice were anesthetized using pentobarbital sodium (50 mg/kg body wt ip, Nembutal, Abbott Laboratories, Abbott Park, IL) and perfused through the left cardiac ventricle with PBS (0.9% NaCl in 10 mM phosphate buffer, pH 7.4), followed by paraformaldehyde-lysine-periodate fixative (PLP; 4% paraformaldehyde, 75 mM lysine-HCl, 10 mM sodium periodate, and 0.15 M sucrose, in 37.5 mM sodium phosphate), as previously described (45). Both kidneys were dissected, sliced, and further fixed by immersion in PLP for 4 h at room temperature and subsequently overnight at 4°C, then rinsed extensively in PBS, and stored at 4°C in PBS containing 0.02% sodium azide until use. For immunoblotting and total RNA extraction, kidneys were harvested from anesthetized mice, snap frozen in liquid nitrogen, and stored at −80°C until use.

Immunofluorescence and confocal microscopy.

PLP-fixed kidney slices prepared as described above were cryoprotected in PBS containing 0.9 M sucrose overnight at 4°C and then embedded in Tissue-Tek OCT compound 4583 (Sakura Finetek USA, Torrance, CA), mounted on a specimen disk, and frozen at −20°C. Sections (4 μm) were cut on a Leica CM3050 S cryostat (Leica Microsystems, Bannockburn, IL), collected onto Superfrost Plus precleaned charged microscope slides (Fisher Scientific, Pittsburgh, PA), air-dried, and stored at 4°C until use.

Sections were rehydrated for 3 × 5 min in PBS and treated with 1% (wt/vol) SDS for 4 min for retrieval of antigenic sites (19). After being washed for 3 × 5 min in PBS and incubated for 10 min in 1% (wt/vol) BSA in PBS with 0.02% sodium azide to reduce nonspecific staining, the sections were incubated for 90 min with the primary antibody diluted in Dako antibody diluent (Dako, Carpinteria, CA) at room temperature, as described previously (45). After 3 × 5-min PBS washes, the secondary antibody was applied for 1 h at room temperature and the slides were then rinsed again in PBS for 3 × 5 min. For dual immunostaining, the above protocol was first carried out using the anti-B2 antibody and the corresponding anti-chicken secondary antibody, and then repeated for the second primary antibody as appropriate, followed by the respective anti-rabbit secondary antibody. Slides were mounted in Vectashield medium (Vector Laboratories, Burlingame, CA) for microscopy and image acquisition.

Digital images were acquired as described previously (45) by using a Nikon Eclipse 800 epifluorescence microscope (Nikon Instruments, Melville, NY) equipped with an Orca 100 CCD camera (Hamamatsu, Bridgewater, NJ). Confocal laser-scanning microscopy imaging was performed on a Radiance 2000 confocal microscopy system (Carl Zeiss MicroImaging, Thornwood, NY) using LaserSharp 2000 version 4.1 software. Epifluorescence and confocal images were analyzed using IPLab version 3.2.4 image-processing software (Scanalytics, Fairfax, VA) and then imported into and printed from Adobe Photoshop version 6.0 image-editing software (Adobe Systems, San Jose, CA).

Quantification of mean pixel intensity of apical B2 immunostaining.

Sections from four B1+/+ and four B1−/− mice were immunostained concurrently and under identical conditions, and all digital images were acquired using the same exposure parameters, including no processing, same exposure time (38 ms), and same image size. This exposure time was selected to ensure maximal intensity while preventing signal saturation. Basolateral staining for the AE1 anion exchanger was used to identify CD A-ICs. Cells for which the nucleus, the apical membrane, and/or the lumen of the tubule were not clearly visible due to the section cut were excluded from the analysis. The segmentation function in the IPLab software was used to select the areas corresponding to the B2-associated fluorescence in the apical region of the A-ICs. The mean pixel intensity (MPI) of every such area was measured, and these data were imported into Microsoft Excel version 10 (Microsoft, Redmond, WA) for further statistical analysis. The final calculation included an average of 83 inner medullary collecting duct (IMCD) A-ICs per wild-type mouse and 99 IMCD A-ICs per B1-deficient mouse, with no fewer than 50 cells being taken into account for any given animal. Summary data are expressed for each group as means ± SD.

Immunogold electron microscopy.

Small pieces of PLP-fixed mouse kidney (<1 mm3) from the inner stripe (IS) of the outer medulla (OM) were cryoprotected in PBS containing 2.3 M sucrose. Ultrathin cryosections were cut on a Leica EM FCS (Leica Microsystems) at −80°C and collected onto formvar-coated gold grids. Sections were incubated on drops of anti-B2 antibody diluted in Dako antibody diluent for 2 h at room temperature. After being rinsed in PBS, the grids were incubated on drops of goat anti-chicken IgG secondary antibody coupled to 10 nm gold particles (Ted Pella, Redding, CA) for 1 h at room temperature. The grids were subsequently rinsed in distilled water, stained on drops of a uranyl acetate/tylose mixture for 10 min on ice, and then collected on loops and allowed to dry, as previously described (45). Sections were examined in a JEM-1011 transmission electron microscope (JEOL, Tokyo, Japan) at 80 kV, and images acquired using an AMT digital imaging system (Advanced Microscopy Techniques, Danvers, MA) were subsequently imported into and printed from Adobe Photoshop.

Freeze-fracture electron microscopy.

Wild-type and B1-deficient mice were anesthetized as described above and perfused through the left ventricle with PBS, followed by 2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer. Kidneys were removed, and the IS of the OM was rapidly dissected and cut into smaller pieces in the fixative. After a further 4-h fixation at room temperature, tissues were rinsed in 0.1 M sodium cacodylate buffer and stored at 4°C until further use. After cryoprotection for at least 1 h in 30% glycerol, tissue pieces were placed on a copper freeze-fracture support and frozen in freon 22 cooled by liquid nitrogen. Freeze-fracture replicas from tissues or cells were produced as previously described (51, 62, 63). After removal from the freeze-fracture device, the replicas were cleaned by immersion for 2 h in concentrated sodium hypochlorite bleach. Replicas were washed for 3 × 5 min with distilled water, collected on copper EM grids, and examined with a JEM-1011 electron microscope. Areas of plasma membranes from 10 A-ICs from B1+/+ and B1−/− mice were photographed at a final magnification of ×100,000. The number of rod-shaped intramembranous particles (IMPs) was counted on digital images and is expressed as the number per square micron of membrane surface area.

Protein extraction and immunoblotting.

Mouse kidney tissues were cut into smaller pieces and disrupted with a Tenbroeck tissue grinder in 3 ml of homogenization buffer (10 mM Tris·HCl pH 7.4, 160 mM NaCl, 1 mM EGTA, 1 mM EDTA, and Complete protease inhibitors from Roche Applied Science, Indianapolis, IN, containing Triton X-100 at a final concentration of 1% and 0.05% Igepal CA-630). Homogenates were centrifuged for 15 min at 16,200 g, 4°C, and the supernatant was collected, aliquoted, and stored at −80°C. The protein concentration was determined with the bicinchoninic acid protein assay (Pierce Biotechnology, Rockford, IL) using albumin as a standard. Sixty micrograms of protein were diluted in Laemmli reducing sample buffer, boiled for 5 min, and loaded onto Tris-glycine polyacrylamide 4–20% gradient gels (Cambrex Bio Science, Rockland, ME). After SDS-PAGE separation, proteins were transferred onto an Immun-Blot polyvinylidene difluoride membrane (Bio-Rad Laboratories, Hercules, CA), and the membrane was blocked and incubated overnight at 4°C with the primary antibody diluted 1:2,000 in Tris-buffered saline containing 2.5% milk. The membrane was subsequently washed and incubated with a secondary antibody conjugated to horseradish peroxidase for 1 h at room temperature as previously described (45). Following four additional washes, antibody binding was detected with the Western Lightning chemiluminescence reagent (PerkinElmer Life Sciences, Boston, MA). For quantitative analysis of protein bands from immunoblotting experiments, digital images of the membranes were acquired using the EpiChemi3 imaging system (UVP, Upland, CA) and analyzed using LabWorks 4.6 software (UVP).

Total RNA extraction, reverse transcription, and PCR.

Total RNA was isolated from kidneys using the RNeasy Midi kit (Qiagen, Valencia, CA) per the manufacturer's specifications. The RNA purification included an on-column removal of genomic DNA contamination using the RNase-free DNase set (Qiagen). The amount of extracted RNA was quantified by spectrometry. Extracted RNA was reverse transcribed (RT) and conventional and quantitative real-time PCR (qRT-PCR) were performed with RT products as templates for up to 40 cycles as previously described (23, 32). qRT-PCR was performed using the SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA) and the 7300 Real Time PCR System (Applied Biosystems). The oligonucleotide primers designed to amplify short sequences of the mouse Atp6v1b2 and Gapd (encoding for GAPDH) were as follows: cgaactgtttatgagactttggacatt (Atp6v1b2 forward primer), ggtgctctgagggattctcttc (Atp6v1b2 reverse primer), tgagcaagagaggccctatcc (Gapd forward primer), and ccctaggcccctcctgttat (Gapd reverse primer) (Sigma-Genosys, The Woodlands, TX). Each qRT-PCR reaction was performed in triplicate. Products were also analyzed by electrophoresis on a 2% agarose gel containing GelStar stain (Cambrex Bio Science). The amplicon sizes are 87 (Atp6v1b2) and 98 bp (Gapd).

Intracellular pH measurements.

Outer medullary collecting ducts (OMCDs) and the initial part of IMCDs were isolated from control or acid-loaded mouse kidneys and transferred onto glass coverslips as described previously (65, 66).

Coverslips were transferred to a thermostatically controlled perfusion chamber (at a flow rate of ∼3 ml/min) maintained at 37°C on a Zeiss Axiovert 200 inverted microscope equipped with a video imaging system (Visitron, Munich, Germany). The isolated tubules were incubated in a HEPES-buffered Ringer solution containing the pH-sensitive dye BCECF-AM (10 μM, Invitrogen-Molecular Probes, Eugene, OR) for 20 min and were washed to remove all non deesterified dye. Intracellular pH (pHi) was monitored by alternately exciting the dye with a 10-μm-diameter spot of light at 495 and 440 nm while monitoring the emission at 532 nm with a video imaging system. Each experiment was calibrated for pHi using the nigericin/high-K+ method, and the obtained ratios were converted to pHi as described previously (61, 65, 66). All experiments were performed in the nominal absence of bicarbonate. The initial solution was a HEPES-buffered Ringer solution (in mM: 125 NaCl, 3 KCl, 1 CaCl2, 1.2 MgSO4, 2 KH2PO4, 32.2 HEPES, pH 7.4). Cells were acidified using the NH4Cl (20 mM) prepulse technique and washed with a Na+-free solution (Na+ was replaced by equimolar concentrations of N-methyl-d-glucamine). The rate of V-ATPase activity was determined as the concanamycin-sensitive pHi alkalinization rate in the absence of Na+. Rates were calculated over the same range of pHi (6.55–6.75) for all cells studied. All chemicals used for pHi measurements were from Sigma-Aldrich (St. Louis, MO) and Calbiochem (Darmstadt, Germany).


Subcellular localization of B2 V-ATPase in renal CD ICs.

The novel anti-ATP6V1B2 (V-ATPase B2 subunit isoform) antibody raised in chicken allowed us to perform for the first time a dual immunostaining experiment for both 56-kDa B isoforms in the mouse kidney. As previously published for mouse and other species by us (39, 45) and others (26, 41, 47), ATP6V1B1 (B1) is expressed in the renal CD at high levels in all ICs. B1 localizes to the apical plasma membrane and subapical domain in A-ICs, and in some cases also assumes, at lower levels, a more diffuse staining pattern. In B-ICs, B1 expression is often localized to the basolateral plasma membrane and, at lower levels, throughout the cytosol, and sometimes even to the apical/subapical domain in a bipolar staining pattern (Fig. 1), as previously described (15, 17). On the other hand, the B2 isoform tends to assume a less polarized localization, especially in the CCDs, although even here, but in particular in the medullary CD A-ICs, the B2 levels appear to be more elevated in the region between the apical membrane and the nucleus, as previously shown (45). Accordingly, this region is characterized by the highest degree of coexpression of the two isoforms, while the basolateral plasma membrane stains predominantly for B1 (in B-ICs). The cytosol appears to contain significantly more B2 than B1, which is to be expected given the well-known B2 presence on the membranes of intracellular organelles (18, 45, 64).

Fig. 1.

Dual immunofluorescence staining for the V-ATPase B1 (green) and B2 (red) subunit isoforms in wild-type mouse kidney cortex (CO; AC), inner stripe (IS) of the outer medulla (DF), and inner medulla (IM; GI). The B1 subunit is present in all collecting duct intercalated cells (ICs) in the cortex (A), IS (D), and IM (F), and also in distal convoluted tubules (DCT; A). B2 is coexpressed with B1 in these cells (B), but B2 is the only 56-kDa subunit isoform detected in proximal tubules (PT), as additionally shown in the merged image (C). B2 is also expressed at lower levels in principal cells of the collecting duct in the cortex, IS (E), and IM (H) in the absence of any detectable B1 expression. In the ICs of the medullary collecting duct, B2 is localized in the apical/subapical membrane domain and also diffusely throughout the cytosol, as illustrated by the higher magnification images of IS (E and F) and IM collecting ducts (H and I). Bars = 10 μm.

In the B1-deficient mouse kidney, and particularly in the OMCD and IMCD A-ICs of B1−/− mice, B2 localization is considerably more polarized (Fig. 2). A-ICs were clearly identified here by double staining with an anti-AE1 antibody, which stains basolateral membranes of this cell subtype, but not B-ICs, nor principal cells of the CD. As previously shown for IMCD A-ICs (25), in numerous OMCD A-ICs B2 also localizes significantly more to the apical plasma membrane, where it exhibits a thin line of bright staining.

Fig. 2.

Immunocytochemical localization of the B2 V-ATPase (red) in the A-type ICs (A-ICs) of the collecting duct in wild-type and B1-deficient (B1−/−) mice. A-ICs were identified by double staining with an anti-AE1 antibody (green). When B1−/− mice are compared with their wild-type counterparts, the B2 immunostaining pattern is seen to shift from a diffuse cytoplasmic staining to a polarized localization that appears as a thin, bright band of apical membrane staining in the outer stripe (OS; A and B) and IS of the outer medulla (C and D) and in the IM (E and F). Bars = 10 μm.

Immunogold electron microscopy was used to support the immunohistochemical data presented above by confirming that the V-ATPase B2 subunit isoform localizes to the apical plasma membrane of B1−/− mouse CD A-ICs. In the representative example shown here of a CD A-IC from the IS of the OM (Fig. 3), immunoelectron microscopy demonstrates that the B2 subunit isoform can be located on the apical membrane, besides being expressed in the subapical region.

Fig. 3.

Immunogold electron microscopy of an inner medullary collecting duct A-IC from B1-deficient mouse kidney using an anti-B2 V-ATPase antibody. Numerous gold particles are seen localized to the apical plasma membrane domain and to the microvilli (arrows), confirming B2 membrane expression as suggested by the immunofluorescence experiments. Bar = 0.2 μm.

To ensure that the B2 isoform localized to the A-IC apical membrane in B1−/− mice is incorporated into functional V-ATPase holoenzymes, we performed dual immunostaining experiments with other V-ATPase subunits, such as ATP6V1A (the 70-kDa A subunit) and ATP6V1E1, the ubiquitous 31-kDa “E1” subunit isoform, formerly known as ATP6E2 (31). Both subunits are part of the cytosolic V1 domain of the enzyme and were chosen because A is the V-ATPase subunit involved in the closest interaction with B (as they form an A3B3 hexamer, a subdomain of V1), while E1 is thought to play an essential role in the V0-V1 assembly (3) and moreover was previously shown to relocate from the cytosolic to the apical membrane domain in response to either chronic acidosis (5) or chronic carbonic anhydrase inhibition by acetazolamide (4, 45). We found B2 V-ATPase to colocalize in a tight apical band in B1−/− mouse medullary CD A-ICs with both A (Fig. 4) and E1 subunits (not shown; see Ref. 25 for the IM).

Fig. 4.

Double immunofluorescence localization of B2 (red) and A (green) subunits of the V-ATPase in the IS of the outer medulla (A and C) and IM (DF) of B1−/− mouse kidney. In both regions, these 2 subunits are coexpressed at the level of the apical plasma membrane, seen clearly as yellow staining in the merged images (C and F). Bars = 10 μm.

Quantitative B2 V-ATPase immunofluorescence in IMCD of wild-type and B1−/− mice.

The B2 immunofluorescence results in wild-type and B1-deficient mouse medullary CDs presented above (Fig. 2) suggest that, in the absence of the 56-kDa B1 isoform, the alternate B2 subunit isoform relocates to a large extent from a subapical and cytosolic location to the apical membrane domain. To test this hypothesis, we performed measurements of the MPI of the B2-associated immunostaining in IMCD A-ICs of B1+/+ and B1−/− mice.

To perform the MPI quantification, IMCD A-ICs were positively identified in both groups of animals by double staining with an anti-AE1 antibody (Fig. 5, A and B). Grayscale images of the B2 immunostaining taken with the same set of exposure parameters (Fig. 5, C and D) were used for the quantification, and the areas corresponding to B2 immunofluorescence in the apical region of these cells were selected using the segmentation function in the IPLab software (Fig. 5, EH). The immunostained areas are relatively large in wild-type animals, frequently covering a substantial part of the region between the apical membrane and the cell nucleus (Fig. 5E), while the apical staining in B1−/− mice is brighter, although restricted to a very narrow band (Fig. 5G).

Fig. 5.

Quantification of mean pixel intensity of B2 V-ATPase-associated immunofluorescence. IM collecting duct A-ICs from wild-type (A) and B1-deficient mice (B) were identified by staining with an anti-AE1 antibody (green), and B2 immunostaining (red) was quantified from the respective grayscale pictures [B1+/+ (C) and B1−/− (D)] acquired with identical exposure parameters. For a typical wild-type A-IC (E) designated in A (arrow), the immunostained area was selected (F) using the segmentation function of IPLab software, and the associated MPI was measured. B1-deficient A-ICs were treated identically; shown here is a cell (G) chosen from B (arrow), which exhibits a much narrower yet brighter immunostained region in the apical membrane domain (H). Bars = 15 μm.

Apical MPI values were determined for an average of 83 cells per B1+/+ mouse and 99 cells per B1−/− mouse. These MPI values were averaged for each of the four animals in every group, and the mean MPI per group was 925 ± 88 for B1+/+ mice and 1,860 ± 127 for B1−/− mice (Fig. 6). Statistical ANOVA and t-test analyses were performed and demonstrated that the differences between the two groups of animals were statistically significant. When the MPI value set from any of the four B1-deficient animals is compared with the MPI values from any of the four wild-type mice, the difference was highly significant (P < 0.001).

Fig. 6.

Mean pixel intensity (MPI) of apical B2 V-ATPase-associated immunofluorescence in 4 wild-type (+/+, black bars) and 4 B1−/− (gray bars) mice is shown here as means ± SD. The dark gray bar represents the average of the B1+/+ group, 925 ± 88 (n = 4). The average MPI of all cells quantified from the 4 wild-type animals was 924. The open bar represents the average of the B1−/− group, 1,860 ± 127 (n = 4), representing an approximately 2-fold increase over the wild-type group. The corresponding average MPI of all A-ICs quantified from B1-deficient mice was 1,868.

As indicated above (Figs. 2 and 5), in the A-ICs of B1−/− mice B2 immunostaining within the cytosolic phase appears to decrease, while the apical plasma membrane staining intensifies significantly. This is confirmed quantitatively not only by the twofold increase seen in MPI but also by a decrease in the mean surface area of the region exhibiting the highest levels of B2 immunostaining. Surface area of the regions used in the above MPI calculations was also measured for every cell using the IPLab segmentation function and was found to decrease, on average, from 344 ± 58 squared pixels (px2) in B1+/+ mice to 133 ± 62 px2 in B1−/− mice (n = 4 for each group) (P = 0.0025).

Freeze-fracture analysis of A-IC membranes.

It has been demonstrated previously that plasma membranes and intracellular vesicles from proton-secreting cells, including renal ICs, contain a highly characteristic class of IMPs known as “rod-shaped particles” (RSPs) (15, 29, 43, 56). Based on their appearance in membranes with high V-ATPase content, they are believed to represent transmembrane domains of the V-ATPase (16). When OMCD A-ICs from B1+/+ mice were examined by this technique, abundant rod-shaped IMPs were found associated with their apical plasma membrane domains, as expected. As shown in Fig. 7, similar IMPs were also found in B1−/− mice, but their density appeared to be reduced. Quantification of freeze-fracture replicas indeed showed a decrease in the rod-shaped IMP content of A-IC membranes from B1-deficient mice compared with normal mice, from 3,737 ± 933 RSPs/μm2 (mean ± SD, n = 5 regions from 3 A-ICs) in B1+/+ mice to 1,950 ± 514 RSPs/μm2 in B1−/− mice (n = 7 regions from 4 A-ICs). This decrease was found to be statistically significant by t-test analysis (P = 0.0016).

Fig. 7.

High-resolution freeze-fracture analysis of outer medullary collecting duct (OMCD) A-ICs from B1+/+ and B1−/− mouse kidney. OMCD A-ICs from B1+/+ mice exhibit abundant rod-shaped intramembranous particles associated with their apical membranes (A), while B1−/− mouse OMCD A-ICs contain similar but fewer such particles (B). To emphasize the difference in rod-shaped particle density, regions indicated by arrows in A and B are shown at a higher magnification in C (wild-type mouse) and D (B1-deficient mouse). Bars = 100 nm.

V-ATPase B2 subunit redistributes to the cytosolic phase in colchicine-treated B1−/− mice.

V-ATPases have been shown to recycle rapidly between cytoplasmic vesicles and the cell membrane in proton-secreting epithelial cells (12, 14, 44, 49, 50). Colchicine, an inhibitor of microtubule polymerization, was previously found to disrupt this recycling (20, 21). CD ICs of both subtypes were shown to respond to colchicine treatment by shifting from a polarized V-ATPase distribution to a dispersed cytosolic staining pattern (20). We consequently examined the effect of colchicine treatment on B2-containing V-ATPases in B1−/− mice. Medullary CD A-ICs from B1-deficient mice injected with colchicine (0.5 mg/100 g body wt ip) 4 h before perfusion fixation showed no significant difference in B2 distribution compared with untreated B1−/− mice (data not shown). However, when the duration of colchicine treatment was increased to 17 h, the B2 V-ATPase staining pattern in most IMCD and ISCD A-ICs shifted drastically from the polarized staining described above (Figs. 2, 4, and 5) to a diffuse pattern, with B2 being localized throughout the cytoplasm (Fig. 8). Some cells continued to display detectable apical membrane staining after colchicine treatment, but even these cells also showed significant subapical and/or cytosolic B2 immunofluorescence. This result is similar to that obtained previously in colchicine-treated kidneys from normal rats (20), thus allowing for a similar conclusion, i.e., that apical targeting of B2-containing V-ATPases in B1−/− mice is affected by microtubule disruption and that the B2 subunit recycles between the plasma membrane and an intracellular vesicular pool.

Fig. 8.

Confocal microscopy showing double immunostaining for AE1 (green, A and D) and B2 V-ATPase (red, B and E) of ICs from colchicine-treated B1-deficient mouse IS of the outer medulla (AC) and IM (DF). Compared with the B2 staining seen ordinarily in medullary A-ICs of untreated B1−/− animals (see Figs. 2, 4, and 5), these animals show, in response to colchicine treatment, significantly less membrane-associated B2 staining and more cytosolic diffuse staining. Bars = 15 μm.

V-ATPase B2 subunit isoform mRNA and protein expression in wild-type and B1−/− mice.

The apical plasma membrane B2 immunostaining increases significantly in B1−/− mice, as indicated by the doubling of the immunofluorescence MPI compared with the wild-type mouse group (Fig. 6). We further investigated whether this increase is due only to a relocalization of B2-containing V-ATPases from the cytosolic compartment to the cell membrane, or additionally to an increase in the levels of B2 mRNA and protein.

Conventional and qRT-PCR analysis performed with RNA isolated from whole kidneys yielded similar Atp6v1b2 mRNA signals in all samples, whether extracted from B1+/+ or B1−/− mouse kidneys (Fig. 9A). Results for Atp6v1b2 mRNA from three different qRT-PCR experiments performed in three animals from each group show no statistically significant upregulation in B2 mRNA expression induced by the lack of the B1 isoform (Fig. 9B).

Fig. 9.

Expression of Atp6v1b2 mRNA in the kidney of wild-type and B1-deficient mice. Total RNA was isolated from B1+/+ and B1−/− mouse kidney and genomic DNA contamination was removed. Conventional and quantitative real-time (qRT)-PCR analysis was performed and products were resolved on a 2% agarose gel. Results show similar Atp6v1b2 mRNA signals in B1+/+ and B1−/− mouse kidneys (A). qRT-PCR analysis for Gapd was performed as a control. Atp6v1b2 mRNA data from 3 different experiments performed on n = 3 animals from each group were normalized to their respective Gapd mRNA values and subsequently to the average of the B1+/+ group. Results are shown here as means ± SE (B). No statistically significant difference was found at the B2 mRNA level between B1-deficient (gray bar) and wild-type mice (black bar).

However, since these findings do not preclude the possibility of a compensatory increase in B2 at the protein level in B1−/− mice, we also performed immunoblotting experiments for B2 in wild-type and B1-deficient mice. We showed previously that immunoblotting of total kidney homogenates revealed, as expected, intense B1 expression in B1+/+ and B1+/− mice, but no signal in B1−/− mice (25). We now report that there is no detectable upregulation of B2 expression in B1−/− mice compared with their B1+/+ (Fig. 10A) and B1+/− counterparts (data not shown). The intensity of the chemiluminescence corresponding to the B2 and actin protein bands from this immunoblotting experiment was quantified, and B2 results were normalized to their respective actin loading controls and subsequently to the mean value of the B1+/+ group. Normalized results show no statistically significant difference between the B1-deficient and the wild-type animals (n = 3 for each group, Fig. 10B). Quantitative immunoblotting was performed three different times, and every experiment yielded similar results. We can thus conclude that the lack of the B1 subunit isoform does not induce an increase in the B2 expression in the kidney at the total mRNA or total protein level, in agreement with results published recently for the epididymis (23). Consequently, the elevation in apical membrane B2 immunostaining may not be due to an increase in the amount of B2 protein.

Fig. 10.

Detection of the V-ATPase B2 subunit in total kidney homogenate by immunoblotting. Sixty micrograms of B1+/+ and B1−/− mouse kidney homogenate were subjected to SDS-PAGE and blotted with a chicken anti-B2 antibody (A). A specific 56-kDa band of similar intensity is seen in B1+/+ and B1−/− animals. Loading control was performed with an anti-actin antibody. The chemiluminescence intensity of the B2 and actin protein bands from A was quantified, and the normalized results (B) show no statistically significant difference between B1−/− (gray bar) and B1+/+ mice (black bar). Values are means ± SE; n = 3 for each group.

V-ATPase activity in medullary CD A-ICs of B1−/− mice.

To assess V-ATPase function in wild-type and B1-deficient mice, we measured the rate of recovery of pHi after acute cellular acidification in A-ICs from IMCDs and OMCDs in both animal groups. For every animal studied, measurements were performed in the presence and absence of concanamycin (100 nM), and the rate of V-ATPase activity was determined as the concanamycin-sensitive pHi alkalinization rate. On average, 140 cells were investigated for each experimental condition (B1+/+ vs. B1−/− mice, OMCD vs. IMCD, untreated control vs. concanamycin-treated).

Summary data show that in OMCD A-ICs of B1+/+ mice the baseline pHi recovery rate was 0.047 ± 0.002 pH units/min (mean ± SE, n = 98 cells) and was reduced by concanamycin to 0.024 ± 0.001 pH units/min (n = 144 A-ICs), corresponding to a fraction of ∼50% of the proton efflux, i.e., a rate of 0.023 ± 0.002 pH units/min, being V-ATPase-dependent (Fig. 11). In comparison, measurements in B1−/− mouse A-ICs (control: n = 119 cells and concanamycin-treated: n = 225 cells) yielded a V-ATPase-mediated pHi alkalinization rate of 0.009 ± 0.001 pH units/min, with a baseline pHi recovery rate of 0.028 ± 0.002 pH units/min, decreasing to 0.019 ± 0.001 pH units/min in the presence of concanamycin. We conclude that a significant fraction (40%) of the V-ATPase activity in the OMCD A-ICs is preserved despite the absence of the B1 subunit.

Fig. 11.

V-ATPase activity measured as concanamycin-inhibitable intracellular pH (pHi) recovery rate in A-ICs from the outer medulla (OM) and IM of B1+/+ and B1−/− mice is shown here as means ± SE. V-ATPase-mediated pHi recovery rate is maintained in the OM of B1-deficient mice at a level corresponding to 40% of the V-ATPase activity in wild-type mice. In IMCD A-ICs of B1−/− mice, V-ATPase function represents ∼28% of the value found in wild-type mice.

Analyzing the measurements for IMCD A-ICs in a similar way reveals that the V-ATPase-mediated pHi recovery rate was 0.019 ± 0.003 pH units/min in the wild-type animals (control: n = 153; concanamycin: n = 132 cells), slightly (but not significantly) lower than in the OMCD. This was reduced further in B1-deficient mice to 0.005 ± 0.002 pH units/min (control: n = 153; concanamycin: n = 100 cells). Therefore, V-ATPase function in the apical membrane of IMCD A-ICs in animals lacking the B1 subunit isoform is decreased to ∼28% of the level found in wild-type mice. Importantly, the pHi recovery rate in the presence of concanamycin was similar in the two animal groups: 0.035 ± 0.002 pH units/min (wild-type) and 0.033 ± 0.002 pH units/min (B1-deficient mice).

Analyses of the four data sets demonstrated that in all cases studied, the concanamycin treatment induced a statistically significant reduction in the H+-extrusion rate from A-ICs (P < 0.05 for B1−/− IMCD, P < 0.0001 for B1−/− OMCD, and P < 0.0001 for B1+/+, both IMCD and OMCD), thus confirming the presence of plasma membrane, concanamycin-sensitive V-ATPase activity in OMCD and IMCD A-ICs from both wild-type and B1-deficient mice.

We also determined the rate of recovery of pHi in A-ICs from OMCDs of acid-loaded B1+/+ vs. B1−/− mice. The oral administration of NH4Cl (280 mM for 24 h) was previously shown to cause a significant reduction in systemic pH in both wild-type and B1-deficient mice (25). In acid-loaded wild-type mice, concanamycin-sensitive V-ATPase activity was 0.029 ± 0.002 pH units/min (control: n = 40; concanamycin: n = 44 cells), representing a 26% increase over the rate found in unchallenged animals. In contrast, B1-deficient mice showed no stimulation of the V-ATPase activity following the acid loading; the V-ATPase-mediated pHi alkalinization rate was 0.005 ± 0.002 pH units/min in this group (control: n = 39; concanamycin: n = 36 cells). Similar to the case of unchallenged animals, there were no statistically significant differences in the pHi recovery rate in the presence of concanamycin following the acid challenge in the two animal groups. Immunofluorescence localization of B2 V-ATPase in acid-loaded B1−/− mice (data not shown) was indistinguishable from the pattern seen in unchallenged animals, in which a tight apical localization of the B2 subunit was already detectable (Figs. 2, 4, and 5).


Mice lacking the V-ATPase 56-kDa ATP6V1B1 (B1) subunit isoform are able to partially compensate for its loss via a mechanism that is sufficient to maintain acid-base balance in various organ systems under baseline conditions (25). In this study, we investigated the role of the alternate 56-kDa ATP6V1B2 (B2) isoform in this compensatory mechanism. B2 generally assumes a diffuse intracellular localization, although in some cells, especially in medullary CD A-ICs, it is more concentrated in the apical region, as shown previously (45). This staining pattern is consistent with the B2 isoform being present on intracellular organelles such as endosomes, lysosomes, and parts of the Golgi/trans-Golgi network, where it plays a “housekeeping” role in acidification processes that are important to a variety of cellular functions (27, 40, 42, 47, 64).

However, the B2 isoform is not entirely intracellular and can be detected on the plasma membrane of various cells, including certain renal epithelial cells, indicating that it contributes to H+ secretion under some conditions and in some tissues. For example, proximal tubule epithelial cells normally express apical membrane-associated V-ATPases containing the B2 isoform (18, 45, 64) and, in osteoclasts, membrane expression of V-ATPases containing the B2 and not the B1 isoform mediates H+ transport across the ruffled membrane to permit bone resorption (37). We previously described abundant apical B2 expression in A-ICs from CDs of rats treated with acetazolamide, a condition in which H+ secretion by ICs appears to be stimulated to compensate for decreased proximal bicarbonate reabsorption that would otherwise lead to severe metabolic acidosis (45).

More recently, we reported that A-ICs from the B1−/− mouse kidney showed a more apically polarized distribution of the B2 subunit (25), and we now demonstrate that a significant portion (28–40%) of wild-type plasma membrane V-ATPase activity is retained in medullary A-ICs from B1-deficient mice. We propose that this level of V-ATPase activity is sufficient to maintain normal acid-base levels in B1−/− mice, but that the reduced proton secretory capacity of A-ICs is limiting upon acid loading of the animals, which then develop severe metabolic acidosis with acidemia (25). The lack of stimulation of V-ATPase activity in outer medullary CD A-ICs of B1−/− mice in response to acid loading reported here provides at least a partial explanation for these findings. Furthermore, it has recently been shown that angiotensin II fails to increase pHi recovery in ICs from B1-deficient mice (48). These data, combined with our present findings, could indicate 1) that B2 incorporation/activity in the plasma membrane of B1-deficient mice is already at maximum levels and cannot be increased in response to further challenges, and/or 2) that increased plasma membrane incorporation of B2-containing holoenzymes in response to physiological stimuli is impaired due to the absence of some as yet unknown trafficking or targeting sequence that might be present in the B1 isoform. Nevertheless, while it is clear that under normal circumstances most of the H+ secretory activity of ICs is mediated by V-ATPases containing the B1 subunit, it is also now apparent that this activity can be supplemented in extreme conditions (such as the acetazolamide treatment of animals) (4, 45) or partially replaced (in B1−/− mice) by V-ATPase holoenzymes that contain B2, as shown in the present study.

V-ATPase insertion into the apical plasma membrane of A-ICs from B1-deficient mice is also suggested by the presence of RSPs on these membranes. RSPs have long been associated with the V-ATPase in a variety of cell types (15, 16, 29, 43, 56), although whether they actually represent V-ATPase transmembrane domains has never been definitively proven. The similar appearance of RSPs in both wild-type and B1-deficient mice indicates that their morphology is not related to the presence of a particular B subunit isoform. This agrees with data from prior studies that revealed RSPs in IC membranes (in which the B1 subunit is predominant) (15, 22, 56) and also in the osteoclast ruffled membrane, in which the B2 subunit is expressed (1). However, the number of RSPs in cells from B1-deficient mice is reduced by 50% compared with wild-type mice. It is possible that this reflects a decrease in the association of transmembrane V-ATPase subunits into these characteristic particles. Indeed, the RSP reduction correlates well with the 60% loss of V-ATPase activity measured in OMCD ICs in the pHi recovery assay.

Our present data partially address the mechanism by which isoform replacement may occur. The absence of significant upregulation in B2 expression in the B1−/− mouse kidney at either the mRNA or the protein level suggests that the increase in apical membrane B2 expression is at least in part due to a redistribution of B2-containing V-ATPases from the cytoplasm to the apical membrane domain. That B2-containing holoenzymes can recycle between these domains is supported by the intracellular accumulation of B1-deficient V-ATPases in response to colchicine, which blocks this recycling pathway. These findings are comparable to previously published results that described the relocation of ATP6V1E1 protein (the ubiquitous V-ATPase 31-kDa E1 subunit isoform) from the cytosol to the plasma membrane in rat A-ICs in response to chronic oral acid loading, which occurred in the absence of significant upregulation in mRNA or protein levels even after 14 days of treatment (5). Another study also reported no increase in B1 subunit protein levels in mouse kidney cortex or medulla after up to 7 days of acid loading (26). The results reported here are also in very good agreement with our recent data showing a comparable increase in the B2 V-ATPase-associated immunostaining in the apical pole of clear cells from the cauda epididymis of B1-deficient mice (23). Similarly, this increase cannot be attributed to an upregulation of the B2 isoform at the mRNA or protein level. One can thus infer that analogous mechanisms may underline the B2 response to the absence of functional B1 V-ATPase in proton-secreting renal ICs and epididymal clear cells.

The pHi recovery data presented here indicate that, unlike in ICs from the CCD that show no mean pHi recovery in B1−/− mice (25), up to 40% of the V-ATPase activity of OMCD A-ICs is preserved in the absence of the B1 subunit isoform. In the inner medulla, pHi recovery of ICs was reduced to 28% of wild-type values, less than in the outer medulla but still potentially significant enough to contribute to urinary H+ secretion. In conclusion, all the evidence points toward a compensatory mechanism in the medullary CD A-ICs of B1−/− mice involving the alternate 56-kDa B2 isoform, which can play a part in transporting protons across the apical plasma membrane and, thus, bring an essential contribution to urinary acidification and to the maintenance of acid-base homeostasis in this animal group under baseline conditions. The reasons why this mechanism does not produce similar results in the ICs of the CCD of B1-deficient mice are currently incompletely understood. While this disparity may be related, at least in part, to the differences in the interstitial environment surrounding the cortical and the medullary CD, further investigations will be required to address this issue.

It should be mentioned that in addition to the V-ATPase (i.e., B2)-mediated pHi recovery, our data indicate that sodium-, bicarbonate-, and concanamycin-independent transport processes may also be involved in the normal and B1−/− mouse CD. However, the activity of concanamycin-independent proton extrusion mechanisms, possibly including the H+-K+-ATPase, was similar in normal and B1-deficient mice, indicating that no detectable compensatory upregulation occurred. The nature of additional non-V-ATPase H+ extrusion mechanisms remains to be investigated in future studies.

It has yet to be determined how the B2 subunit can be incorporated into a functional V-ATPase that is targeted to the plasma membrane, and why this mechanism is apparently not sufficient to compensate in humans with dRTA resulting from B1 subunit mutations. It has been shown that B1 constructs bearing the single amino acid mutations described in dRTA patients do not assemble into functional V-ATPase complexes (67). These mutated B1 subunits impair the plasma membrane trafficking and insertion of the V-ATPase, possibly by competing with wild-type holoenzymes for components of the vesicle-targeting machinery. In contrast, the B1-deficient mouse does not express such an abnormal B1 protein that could alter trafficking of wild-type V-ATPases. In the complete absence of B1, it is likely that B2 is free to incorporate into holoenzymes that are targeted to the plasma membrane based on as yet unidentified targeting sequences in other subunits, possibly the transmembrane a4 subunit (46, 53) with which both B1 and B2 can associate (59). Further studies, including the generation of mice expressing B1 subunits that replicate the human dRTA-inducing mutations will be required to explore these possibilities.


This study was supported by National Institutes of Health (NIH) Grants DK-73266 (to T. G. Păunescu), DK-42956 (to D. Brown), DK-38452 (to D. Brown and S. Breton), and HD-40793 (to S. Breton), by Swiss National Science Foundation Grant 31-109677/1, and the EU 6th Framework project EuReGene 005085 (to C. A. Wagner). L. M. Russo is the recipient of an Advanced Post-doctoral Award from the Juvenile Diabetes Research Foundation International.


  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


View Abstract