Cell Physiology

V-ATPase B1-subunit promoter drives expression of EGFP in intercalated cells of kidney, clear cells of epididymis and airway cells of lung in transgenic mice

R. Lance Miller, Ping Zhang, Maren Smith, Valerie Beaulieu, Teodor G. Păunescu, Dennis Brown, Sylvie Breton, Raoul D. Nelson


The kidney, epididymis, and lungs are complex organs with considerable epithelial cell heterogeneity. This has limited the characterization of pathophysiological transport processes that are specific for each cell type in these epithelia. The purpose of the present study was to develop new tools to study cell-specific gene and protein expression in such complex tissues and organs. We report the production of a transgenic mouse that expresses enhanced green fluorescent protein (EGFP) in a subset of epithelial cells that express the B1 subunit of vacuolar H+-ATPase (V-ATPase) and are actively involved in proton transport. A 6.5-kb portion of the V-ATPase B1 promoter was used to drive expression of EGFP. In two founders, quantitative real-time RT-PCR demonstrated expression of EGFP in kidney, epididymis, and lung. Immunofluorescence labeling using antibodies against the B1 and E subunits of V-ATPase and against carbonic anhydrase type II (CAII) revealed specific EGFP expression in all renal type A and type B intercalated cells, some renal connecting tubule cells, all epididymal narrow and clear cells, and some nonciliated airway epithelial cells. No EGFP expression was detected in collecting duct principal cells (identified using an anti-AQP2 antibody) or epididymal principal cells (negative for V-ATPase or CAII). This EGFP-expressing mouse model should prove useful in future studies of gene and protein expression and their physiological and/or developmental regulation in distinct cell types that can now be separated using fluorescence-assisted microdissection, fluorescence-activated cell sorting, and laser capture microdissection.

  • collecting duct
  • enhanced green fluorescent protein

the collecting duct of kidney and the excurrent duct of the male reproductive tract are composed of diverse epithelial cell types that are involved in numerous physiological transport processes. This complexity has limited the understanding of the regulation of specific functions performed by these subsets of specialized cells. The purpose of the present study was therefore to develop new tools to facilitate the study of specific epithelial cells in these two tissues. The development of the kidney and the epididymis is closely intertwined. Both of these organs are derived from the Wolffian duct and thus share the same embryological origin (41, 43). This is reflected by the presence of similar epithelial cells that share several transport processes in these tissues. For example, intercalated cells from the kidney collecting duct as well as epididymal narrow and clear cells express high levels of the vacuolar H+-ATPase (V-ATPase) in their plasma membrane and are responsible for active proton transport (8).

The renal collecting duct plays a critical role in the regulation of extracellular volume, osmolality, and pH. It is composed of two structurally and functionally distinct cell types: principal cells and intercalated cells (33). While principal cells are involved in the maintenance of sodium and water balance, intercalated cells play an important role in acid-base homeostasis (1, 7, 34, 44). Two types of intercalated cells have been identified. Type A intercalated cells (A-IC), which express the V-ATPase in their apical pole, are involved in proton secretion. Type B intercalated cells (B-IC), which exhibit apical, bipolar, or basolateral V-ATPase localization, secrete either protons (B-IC with apical V-ATPase) or HCO3 (B-IC with basolateral V-ATPase) (2, 911, 48). Type B intercalated cells also have been identified by their lack of expression of the basolateral Cl/HCO3 exchanger AE1 (1). Other reports have labeled the cells with apical V-ATPase and without AE1 expression as “non-A-non-B cells” (29). In the present study, for simplicity, all V-ATPase-positive and AE1-negative intercalated cells are referred to as B-IC. Genetic (27, 45) and acquired (38) alterations of the V-ATPase are associated with disorders of acid-base balance. Such examples indicate that the proper expression and regulation of the V-ATPase in intercalated cells is essential to the maintenance of acid-base balance. However, the regulation of V-ATPase in intercalated cells in health and disease remains incompletely understood.

The epididymis is the site where spermatozoa undergo their final maturation and are then stored (3, 18, 26, 40). As sperm travel along the epididymis, they are maintained immotile and the enzymes involved in acrosomal initiation are prevented from inducing premature activation. An initial step in sperm activation occurs after ejaculation, when spermatozoa are mixed with prostatic fluid, which is rich in HCO3 and higher in pH than the epididymal fluid. Therefore, acidification of the luminal fluid of the epididymis is a critical factor in the establishment of an optimum environment for sperm maturation and storage. Similar to the kidney, the epididymal epithelium is a complex structure composed of at least three cell types: narrow/clear, principal, and basal cells. Narrow and clear cells express high levels of the V-ATPase in their apical plasma membrane and in subapical vesicles (4, 12, 14). While narrow cells are found in the initial segments of the epididymis, clear cells are present in the caput, corpus, and cauda epididymidis. These cells resemble type A intercalated cells. They are responsible for the bulk of proton secretion and are therefore key players in luminal acidification (4, 12, 14).

The characterization of intercalated cell and narrow/clear cell development and function has been limited considerably by the complexity of the kidney and epididymis and by the cellular heterogeneity of the epithelial tubules that form these organs. The understanding of specific gene and protein expression in these cells would therefore benefit greatly from techniques allowing selective isolation and purification of these cell types from the kidney and epididymis. The purpose of this study was to create a transgenic mouse that expresses enhanced green fluorescent protein (EGFP) specifically in kidney intercalated cells and in epididymal narrow/clear cells to allow direct visualization of these cells, which would facilitate their study in the intact animal and would allow their selective isolation using fluorescence-assisted techniques.

Because both intercalated and narrow/clear cells express high levels of the V-ATPase, an EGFP transgene was designed using the promoter of one of its subunits. The V-ATPase is a multisubunit complex composed of two domains: the V1 and V0 domains (46, 49). The V1 domain (catalytic domain) is involved in the hydrolysis of ATP and is composed of eight different subunits (A–H). The V0 domain (membrane domain) is responsible for proton translocation across the plasma membrane and consists of five different subunits. Interestingly, while many V-ATPase subunits are expressed in several cell types, some of these subunits, including the B subunit, have more than one isoform with cell-specific expression patterns. For example, while the B2 isoform is present in a variety of tissues, the B1 isoform is expressed much more abundantly in a limited number of cell types, including kidney intercalated cells (36), connecting tubule cells (17), and epididymal narrow/clear cells (5). Therefore, in the present study, the promoter of the ATP6V1B1 gene encoding the V-ATPase B1 subunit was used to drive cell-specific expression of an EGFP transgene. Characterization of these transgenic mice using quantitative RT-PCR demonstrated specific expression in the kidney, the male reproductive tract, and the lung. Immunofluorescence microscopy demonstrated specific expression in intercalated cells of the collecting duct and connecting tubule, connecting tubule cells, and narrow/clear cells in the epididymis. In addition, nonciliated airway cells also expressed EGFP and the B1 V-ATPase subunit.


Cloning of the human V-ATPase B1-subunit gene.

An arrayed human bacterial artificial chromosome library was screened for the human V-ATPase B1 subunit (or ATP6V1B1 gene) using PCR (Genome Systems, St. Louis, MO). Overlapping genomic fragments were subcloned and spliced together to generate a 6.5-kb fragment containing the 5′-flanking region of the gene and were sequenced to verify identity with the ATP6V1B1 gene (GenBank accession no. NT_22184) in the Human Genome Project.

Construction of the human V-ATPase B1-subunit promoter EGFP transgene.

The EGFP coding region from the pEGFP-1 vector (Clontech, La Jolla, CA) was ligated into the vector containing 6.5 bp of the 5′-flanking region of the ATP6V1B1 gene. A polyadenylation signal from the SV40 virus early region with an added ClaI site was ligated into a site downstream from the EGFP coding region. This product was designated as pB1EGFPpA. Each plasmid intermediate and the final plasmid product were analyzed by performing PCR and restriction mapping to confirm the proposed structure. All ligation junctions were sequenced to confirm that the construct had the desired structure. The key elements are the 6.5-kb B1 promoter, 0.7-kb EGFP coding region, and 0.3-kb SV40 early region polyadenylation signal with unique ClaI sites upstream from the promoter and downstream from the polyadenylation sequence.

Generation and breeding of transgenic mice.

The B1-EGFP transgene was separated from the vector by agarose gel purification after digestion of pB1EGFPpA with ClaI. The DNA was isolated by electroelution, concentrated using an Elutip-D column (Schleicher & Schuell, Keene, NH), and resuspended in injection buffer consisting of 10 mM Tris, pH 7.4, and 0.1 mM EDTA. Transgenic mice were created by the University of Utah transgenic mouse core facility using standard procedures (21). Briefly, the pronucleus of single-cell C57BL6 × CBA F1 embryos was microinjected with the purified transgene, and the resulting embryos were implanted into pseudopregnant females. The resulting pups were genotyped by performing PCR of tail DNA. The transgenic founders were each bred with C57BL/6 × CBA F1 mice. The resulting F1 and F2 animals were analyzed for expression of the transgene. All procedures were performed according to protocols approved by the University of Utah Institutional Animal Care and Use Committee.

PCR genotyping.

DNA was isolated from a tail biopsy specimen. Transgene was detected by performing PCR amplification of 100 ng of mouse genomic DNA using 25-μl reactions containing 0.05 U/μl Taq DNA polymerase (Invitrogen, Carlsbad, CA), 200 μM each of dATP, dCTP, dGTP, and dTTP, 1.5 mM MgCl2, 50 mM KCl, 10 mM Tris·HCl, pH 8.4, 0.4 μM forward primer, and 0.4 μM reverse primer. The thermocycling programs for genotyping consisted of a cycle at 94°C for 60 s followed by 30–35 cycles of 94°C for 20 s, 57–60°C for 20 s, and 72°C for 50 s, and a final extension at 72°C for 5 min (Table 1). The cycle number and annealing temperature were primer set dependent (Table 1). The cycle number was limited to nonsaturating conditions. Nontransgenic tail DNA was spiked with transgene DNA to simulate 1–100 copies per cell equivalent of transgene DNA. The oligonucleotide primers B1EGFPF and B1EGFPR were used to detect the B1-EGFP transgene (Table 1). As a control for DNA integrity, the oligonucleotide primers AQP2GF and AQP2GR (Table 1) were used to detect the endogenous AQP2 gene (GenBank accession no. NT_039350).

View this table:
Table 1.

Primers and conditions for PCR genotyping and RT-PCR analysis of gene expression

RT-PCR analysis of transgene and endogenous gene expression.

Organ RNA was isolated using TRIzol reagent (Invitrogen) followed by RNeasy minicolumns with on-column DNase I treatment (Qiagen, Valencia, CA). RNA was reverse transcribed using Superscript II (Invitrogen) with oligo(dT) (1218) according to the manufacturer's recommendations. cDNA was then amplified by performing real-time PCR using SYBR Green (Molecular Probes, Eugene, OR) detection according to the following protocol (16, 31, 35, 50). Each PCR reaction contained the following final concentrations: 1× buffer (20 mM Tris·HCl, pH 8.3, 50 mM KCl, and 3 mM MgCl2), 0.3 μM forward primer, 0.3 μM reverse primer, 1× additive reagent (0.2 mg/ml bovine serum albumin, 150 mM trehalose, and 0.2% Tween-20), 0.25× SYBR Green (Molecular Probes, Eugene, OR), 1.5 U Platinum Taq DNA Polymerase (Invitrogen), 200 μM each of dATP, dCTP, dGTP, and dTTP, 1 μl DNA, and H2O to bring the final volume to 25 μl. PCR amplification was performed in the following steps using the Smart Cycler (Cepheid, Sunnyvale, CA): 1) 95°C for 2 min, 2) 95°C for 12 s, 3) 57–62°C for 15 s, 4) 72°C for 20 s, and 5) fluorescent detection at 86°C for 6 s. Steps 2–5 were repeated for 40 to 45 cycles. At the end of the last cycle, the temperature was increased 0.2°C/s from 60 to 95°C to produce a melt curve. LCMB1F and LCMB1R were used to amplify the mouse 58-kDa B1 subunit of the V-ATPase (ATP6V1B1 gene or B1 subunit) (GenBank accession no. NM_134157). LCEGFPF and LCEGFPR were used to amplify EGFP (GenBank accession no. U55761). GAPDHF and GAPDHR were used to amplify glyceraldehyde-3-phosphate dehydrogenase (GAPDH; GenBank accession no. NM_008084). All PCR products were cloned into pT-Adv (Clontech) or pGEM-T (Promega) for DNA sequencing. The authenticity of the products was confirmed by performing DNA sequencing of at least five cDNA clones in TA plasmid vectors. Fluorescence data were collected and analyzed using Smart Cycler software (Cepheid). Standard curves were generated by using serial dilutions of a standard cDNA from a single preparation of transgenic kidney RNA derived from a previous transgenic mouse line expressing EGFP in kidney (51). PCR reactions were considered valid if the amplification was linear, i.e., R2 ≥ 0.98 with ≥97% efficiency. Homogeneity of the products was confirmed using melting analysis.

RNA was then prepared for three offspring from each founder, and reverse transcription reactions were performed in the presence and absence of reverse transcriptase to verify that transgene expression resulted from amplification of cDNA rather than from residual genomic DNA (data not shown). Real-time PCR was performed with a panel of mouse organ cDNA in parallel with dilutions of reference cDNA as a standard curve. Expression was calculated from the standard curves and then expressed in arbitrary units of EGFP or the B1 subunit relative to GAPDH.

Fluorescence microscopy.

Transgenic mouse tissues were prepared according to the following procedures. Transgenic mice were anesthetized by administering inhaled halothane and fixed using cardiac perfusion with 2% paraformaldehyde (PFA) in PBS (10 mM sodium phosphate buffer containing 0.9% NaCl, pH 7.4; 2% PFA) in PBS at room temperature. The kidneys, male reproductive tract, and lungs were removed and fixed by immersion in 2% PFA in PBS for 2 h at room temperature and overnight at 4°C. The tissues were processed using two methods.

Transgenic kidneys were embedded in 3% agarose in PBS and cut into 150- to 200-μm sections with an oscillating microtome (OTS-3000; Electron Microscopy Sciences, Fort Washington, PA). Sections were viewed using fluorescence confocal microscopy (MRC 1024 confocal system; Bio-Rad, Hercules, CA). Images were captured digitally using Lasersharp version 3.2 software (Bio-Rad). A z-series of 11 sections was collected every 1, 1.5, 3, 5, or 10 μm with a ×2.5 lens objective. A montage for each z-series was digitally created using Lasersharp version 3.2 software.

Transgenic kidneys, epididymis, and lung were sectioned with a cryomicrotome and immunostained according to the following procedure (13). Mice were fixed using cardiac perfusion with PLP fixative containing 4% PFA, 10 mM sodium periodate, 75 mM lysine, and 5% sucrose. The tissues were cryoprotected by immersion in 30% sucrose in PBS for 4 h, mounted in Tissue-Tek (Miles, Elkhart, IN), and frozen at −30°C in a Reichert Frigocut cryostat (Reichert Jung, Derry, NH). Sections were cut at 4 μm and placed onto Superfrost/Plus microscope slides (Fisher Scientific, Pittsburgh, PA). Sections were hydrated for 5 min in PBS and treated for 4 min with 1% sodium dodecyl sulfate (SDS) in PBS. After the sections were washed three times with PBS for 5 min each, nonspecific staining was blocked in a solution of 1% BSA in PBS for 15 min. The sections were incubated at room temperature with an affinity-purified rabbit antibody to AQP2 (47), a polyclonal rabbit antiserum to the V-ATPase B1 subunit (5, 36), or an affinity-purified chicken antibody to the V-ATPase E subunit (6). These primary antibodies were detected using goat anti-rabbit or anti-chicken IgG conjugated to CY3. Some kidney sections were double-immunostained for the V-ATPase B1 subunit and for calbindin 28K using a mouse anti-calbindin antibody (Calbindin D-28K clone CB-955; Sigma). In these sections, the V-ATPase was detected using goat anti-rabbit AMCA (blue), and calbindin was detected with donkey anti-mouse CY3. Some epididymal sections were stained using a polyclonal rabbit anti-rat CAII antibodies (provided by William S. Sly, Dept. of Biochemistry and Molecular Biology, St. Louis University School of Medicine, St. Louis, MO), followed by goat anti-rabbit CY3. Lung sections were stained with a monoclonal anti-α-tubulin antibody (T5168; Sigma) and detected with donkey anti-mouse AMCA or donkey anti-mouse CY3. Some lung sections were stained with 4,6-diamidino-2-phenylindole (DAPI) stain for nuclei. The slides were mounted in Vectashield (Vector Laboratories, Burlingame, CA) diluted 1:1 in Tris·HCl buffer (1.5 M, pH 8.9). Images were captured with a Hamamatsu Orca digital camera (Bridgewater, NJ) mounted on a Nikon Eclipse 800 microscope. Pseudocolor images were merged using IPLab Spectrum software (Scanalytics, Fairfax, VA). Cy3 appears red, AMCA and DAPI appear blue, and EGFP is green.

DNA sequencing.

Plasmids containing the RT-PCR products, transgene, and transgene intermediates were sequenced by the University of Utah DNA Sequencing Facility. They use the dye primer system for universal primers and the dye terminator system for other primers. The products were analyzed using the ABI Prism 377 or 3700 DNA analyzers (Applied Biosystems, Foster City, CA). The results were analyzed using Sequencher (Gene Codes, Ann Arbor, MI) and Omiga (Oxford Molecular).


Design and construction of B1-EGFP transgene.

The V-ATPase B1-subunit gene (6.5 kb, ATP6V1B1 gene; GenBank accession no. NM_039350) 5′-flanking region was used to drive expression of an EGFP expression cassette (Clontech) (Fig. 1). The transgene includes the 5′-flanking region of the ATP6V1B1 gene extending to but excluding the endogenous translational start codon. The EGFP cassette includes the EGFP coding region, with its own translational start site, and an SV40 polyadenylation signal. Oligonucleotide primers were designed to anneal to the promoter and EGFP coding region to perform PCR genotyping. Primers were also designed to anneal within the EGFP coding region to perform RT-PCR analysis. Note that this EGFP expression cassette has been used successfully to express EGFP in principal cells of the renal collecting duct using the AQP2 promoter (51). The transgene minus the vector backbone is referred to as B1-EGFP.

Fig. 1.

The human (h)B1-enhanced green fluorescent protein (EGFP) transgene. The human vacuolar H+-ATPase (V-ATPase) gene is shown at top. The hB1-GFP transgene is shown at bottom. The human V-ATPase 5′-flanking region (6.5 kb) was linked to an EGFP expression cassette including the SV40 early region polyadenylation signal. The transcriptional initiation sites are indicated by arrows. Animals were genotyped using PCR genotyping with oligonucleotide primers at the positions shown. RNA expression was determined by performing RT-PCR using the oligonucleotide primers at the position shown. This figure is schematic and not to scale.

Creation, breeding, and genotyping of transgenic mice.

The B1-EGFP transgene was used to create transgenic mice. Pronuclear microinjections and embryo implantation was performed by the University of Utah Transgenic Mouse Core Facility using standard methods (20). Approximately 50 live pups were analyzed for the presence of the B1-EGFP transgene by performing PCR analysis of tail DNA. Oligonucleotide primers and conditions of PCR are shown in Table 1. The integrity of tail DNA was verified using PCR with primers specific for the endogenous mouse aquaporin type 2 (AQP2) gene (GenBank accession no. NT_039621). Six founders containing the B1-EGFP transgene were identified (Fig. 2). Transgene levels in the founders were estimated by comparing the intensity of PCR products to that observed for nontransgenic mouse DNA spiked with 0, 1, 10, and 100 copies per cell equivalent of the transgene. Founders F1, F3, F16, and F43 appear to have at least 100 copies per cell equivalent of the transgene, while founders F39 and F40 have ∼10 copies. All founders were crossed with wild-type C57BL/6 × CBA F1 mice and hemizygous offspring analyzed for expression of the transgene. Offspring derived from the F1 and F16 founders are referred to as lines 1 and 3, respectively. They demonstrated kidney, epididymis, and lung expression of the B1-EGFP transgene and continued to do so in the F1, F2, and F3 generations. One additional line exhibited expression of the B1-EGFP transgene in a pattern that was consistent with collecting duct expression, but the expression was incomplete and therefore was not fully characterized.

Fig. 2.

PCR genotyping of mouse tail DNA showed that the hB1-EGFP transgene was present in six founder mice. Transgenic tail DNA (left) and normal mouse DNA spiked with 0, 1, 10, and 100 copies per cell equivalent of the purified transgene DNA (right) were amplified by PCR using primers specific for the hB1-EGFP transgene (top) and the endogenous murine aquaporin type 2 gene (AQP2; bottom).

RT-PCR analysis for transgene expression.

Real-time RT-PCR assays were developed for determination of transgene and endogenous gene expression using SYBR Green for detection (35). PCR primers were designed to amplify cDNA for EGFP, V-ATPase B1 subunit (B1 subunit), and GAPDH (Table 1).

Qualitative real-time RT-PCR analysis of kidney and male reproductive tract revealed expression of EGFP in three different lines of transgenic mice. Quantitative RT-PCR analysis was performed on organ panels prepared from three animals each for lines 1 and 3. A limited panel of epithelial and nonepithelial organs was chosen. Note that the male reproductive tract preparation included a combination of epididymis and vas deferens. Representative organ panels for lines 1 and 3 are shown in Fig. 3, A and B. EGFP is expressed in the kidney, male reproductive tract, and lung, where the B1 subunit also is expressed. The expression of EGFP and the B1 subunit is below the threshold for detection in the other organs. EGFP is expressed at high levels in the kidney, male reproductive tract, and lung in both transgenic lines of mice. This expression parallels that of the endogenous B1-subunit gene. These results indicate that 6.5 kb of the B1-subunit promoter are sufficient to drive selective expression in kidney, male reproductive tract, and lung.

Fig. 3.

Quantitative RT-PCR analysis of organs from hB1-EGFP transgenic mice. A: line 1; B: line 3. V-ATPase B1 subunit and EGFP transgene expression are represented relative to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene expression. Kidney, lung, and male reproductive tract (male reprod) were all positive for the B1 subunit and EGFP, while heart, brain, muscle, and liver were all negative. The threshold for detection of B1 and EGFP expression is 125 ng of RNA equivalents or <0.001 ratio of B1 to GAPDH.

Confocal fluorescence microscopy of vibratome sections of kidney.

Confocal fluorescence microscopy was performed on vibratome sections (200 μm thickness) of fixed whole kidneys to screen all founder lines for expression of the B1-EGFP transgene at 4–6 wk of age. Kidneys from at least three animals per line were examined. The transgene was expressed in a pattern that was consistent with collecting duct expression in three lines of mice. Lines 1 and 3 (Fig. 4, A and B, respectively) exhibited EGFP expression in the majority of collecting ducts and therefore were complete. An additional line exhibited expression in a small percentage of collecting ducts and therefore was incomplete. This line was not further characterized. The pattern of fluorescence in lines 1 and 3 suggests that EGFP is expressed in the collecting duct in the outer portion of the inner medulla, the entire outer medulla, and the cortex (Fig. 4, A and B). This is the pattern of expression that has been described for B1-subunit expression (36).

Fig. 4.

Fluorescent confocal microscopic image showing a representative sagittal section of kidney from transgenic mice. Transgenic kidney sections (200 μm) were cut using a vibratome. A: line 1; B: line 3. Photographs of the entire kidney were obtained using a ×2.5 lens objective and merged to create a continuous image. The pattern of fluorescence is consistent with collecting duct expression.

Immunofluorescence microscopy of 4-μm cryosections of kidney.

Immunofluorescence microscopy was performed on thin cryosections of transgenic mouse kidney to determine whether EGFP was expressed selectively in intercalated cells of the collecting duct. Cryosections were immunolabeled using antibodies against the V-ATPase B1 or E subunits, CAII, calbindin, or AQP2. Because the pattern of EGFP expression was identical in transgenic lines 1 and 3, the results are shown for line 1 only (see Figs. 58).

Fig. 5.

Cortical collecting duct showing expression of EGFP in intercalated cells. A: EGFP (green); B: V-ATPase B1 subunit immunostaining (red); C: merged image showing EGFP (green) and V-ATPase (red). Intercalated cells, positive for the V-ATPase, express variable levels of EGFP. Arrows indicate intercalated cells expressing high levels of EGFP. Arrowheads indicate intercalated cells with moderate or low EGFP expression and basolateral V-ATPase expression, identifying them as type B intercalated cells (B-IC). Principal cells negative for the V-ATPase do not express EGFP. Bars, 10 μm.

Within the cortical collecting duct, EGFP is expressed exclusively in intercalated cells, which express the V-ATPase B1 subunit (Fig. 5). It is interesting to note that type B intercalated cells with basolateral V-ATPase staining often show lower EGFP fluorescence intensity compared with type A intercalated cells. Expression of V-ATPase is very low in principal cells (42), and these cells do not express detectable EGFP.

Connecting segments, identified by their calbindin-positive staining (2, 19), also show EGFP expression (Fig. 6). In this segment, most cells express the B1 subunit of the V-ATPase. Connecting tubule (CNT) cells, which express calbindin, show lower expression of both the V-ATPase B1 subunit and EGFP compared with intercalated cells, which are negative for calbindin but express high levels of the V-ATPase B1 subunit and EGFP. Thus the level of EGFP expression parallels the level of V-ATPase B1-subunit expression.

Fig. 6.

Two connecting segment profiles expressing EGFP (A and E), double-immunostained for calbindin (B and F), and the V-ATPase B1 subunit (C and G). D and H: Merged images showing EGFP (green), calbindin (red), and V-ATPase (blue). In this tubule segment, most cells express both EGFP and the B1 subunit of the V-ATPase. Intercalated cells, negative for calbindin, show higher levels of EGFP expression (arrows) than connecting tubule (CNT) cells, identified by their positive calbindin immunoreactivity. Bars, 10 μm.

Collecting ducts in the inner stripe of the outer medulla show high expression of EGFP in V-ATPase-positive intercalated cells exclusively (Fig. 7). In this segment, all intercalated cells are type A. Principal cells, positive for AQP2, do not express EGFP (data not shown). Identical results were obtained in collecting ducts from the outer stripe of the outer medulla (data not shown). Finally, collecting ducts from the initial region of the inner medulla showed expression in intercalated cells exclusively, identified by their negative staining for AQP2 (Fig. 8). Principal cells, positive for AQP2, do not express EGFP. These findings show that the V-ATPase B1-subunit promoter is sufficient to drive expression of EGFP selectively in intercalated cells within the medullary and cortical collecting duct and in both intercalated and CNT cells in the connecting tubule.

Fig. 7.

Collecting duct from the inner stripe of the outer medulla showing expression of EGFP in intercalated cells exclusively. A: EGFP; B: V-ATPase B1-subunit immunostaining; C: merged image showing EGFP (green) and V-ATPase (red). In this kidney region, all intercalated cells, positive for the V-ATPase, express high levels of EGFP (arrows). Principal cells, negative for the V-ATPase, do not express EGFP. Bars, 10 μm.

Fig. 8.

Collecting duct from the initial region of the inner medulla showing expression of EGFP in intercalated cells exclusively. A: EGFP; B: AQP2 immunostaining; C: merged image showing EGFP (green) and AQP2 (red). In this kidney region, all intercalated cells, negative for AQP2, express high levels of EGFP (arrows). Principal cells, positive for AQP2, do not express EGFP. Bars, 10 μm.

Immunofluorescence microscopy of the epididymis.

Cryosections of all regions of the epididymis from at least three transgenic mice from each line of mice were immunolabeled for V-ATPase. The results for lines 1 and 3 were identical. In the initial segments of the epididymis, EGFP is expressed in a subset of cells (Fig. 9). These cells are identified as narrow cells by their positive apical staining for V-ATPase (Fig. 9, AC). However, a few narrow cells, positive for EGFP, do not exhibit any V-ATPase staining, because their apical pole, where the V-ATPase is located, is not present in the section (Fig. 9, AC, arrowheads). Staining of the sections for CAII, a cytosolic marker of narrow cells, demonstrated that all narrow cells express EGFP (Fig. 9, DF). No EGFP is detected in principal cells, which are negative for either V-ATPase or CAII.

Fig. 9.

Initial segments of the epididymis showing expression of EGFP in narrow cells exclusively. A and D: high EGFP expression in a subpopulation of epithelial cells. These cells were identified as narrow cells by their positive immunoreactivity for V-ATPase (B) and carbonic anhydrase type II (CAII) (E). C and F: merged images showing EGFP (green) and V-ATPase (red) (C) and EGFP (green) and CAII (red) (F). Arrows indicate EGFP-positive cells cut through their entire apical-to-basal axis; all cells positive for V-ATPase (B) or CAII (E) also show EGFP expression (C and F, respectively). In some cases, only the basal region of narrow cells is visible in the section. While these cells are positive for cytosolic EGFP (A and D; arrowheads), V-ATPase staining is not present in this region of the cells (B, arrowhead). In contrast, cytosolic CAII is detectable in the basal region of these cells (E, arrowhead). Bars, 20 μm.

EGFP is also expressed in clear cells of the caput, corpus (data not shown), and cauda epididymidis (Fig. 10). In Fig. 10, AC, clear cells identified by their apical V-ATPase staining, show high expression of EGFP. Again, the apical region of a few clear cells was not present in the section; therefore, these cells showed EGFP expression with low V-ATPase staining (Fig. 10, AC, arrowhead). These EGFP-positive cells were further identified as clear cells by their CAII cytosolic staining (Fig. 10, EF).

Fig. 10.

Proximal cauda epididymis showing expression of EGFP in clear cells exclusively. A and D: high EGFP expression in a subpopulation of epithelial cells. These cells were identified as clear cells by their positive immunoreactivity for V-ATPase (B) and CAII (E). C and F: merged images showing EGFP (green) and V-ATPase (red) (C) and EGFP (green) and CAII (red) (F). Arrows indicate EGFP positive cells cut through their entire apical-to-basal axis: all cells positive for V-ATPase (B and C) or CAII (E and F) show EGFP expression. Arrowheads indicate EGFP-positive cells that are visible only at their basal pole. These cells do not show V-ATPase staining, which is mainly located in the apical pole (B and C), but they are positive for CAII, a cytosolic marker (E and F). In the proximal cauda of mouse epididymis, some tubules contain rows of adjacent clear cells (AC). Bars, 20 μm.

Quantification was performed to determine the efficiency of the EGFP transgene expression. Examination of 710 cells distributed in 50 tubules revealed that EGFP is expressed in all narrow and clear cells, which are identified by their V-ATPase or CAII expression. These results indicate the high efficiency of the V-ATPase B1 promoter in driving expression of the EGFP transgene.

Immunofluorescence microscopy of lung airway epithelia.

The localization of EGFP in the lung was examined. As shown in Fig. 11, EGFP was expressed in airway epithelial cells. Double-labeling for tubulin showed that EGFP is abundant in nonciliated cells, negative for tubulin, and not detectable in ciliated cells. The B1 subunit of the V-ATPase was then localized in epithelial cells of lung airways, and its distribution was compared with the EGFP expression pattern. Some EGFP-expressing airway epithelial cells were brightly stained for the B1 subunit. Other EGFP-positive cells showed less intense B1 staining. Furthermore, apical tubulin staining was absent from EGFP-expressing cells (Fig. 12D). We conclude, therefore, that only nonciliated cells express EGFP and that the level of V-ATPase B1-subunit expression in these nonciliated cells is variable but is often much greater than in the adjacent ciliated cells.

Fig. 11.

Conducting airway of lung showing expression EGFP in nonciliated cells that do not express tubulin. AC: cross section of bronchus at low magnification. A: EGFP expression (green). B: tubulin immunostain (red). C: superposition of A and B. Nuclei are stained with 4,6-diamidino-2-phenylindole (DAPI). D and E: airway epithelium at high magnification. D: EGFP expression in airway epithelial cells. E: tubulin immunostaining shows ciliated cells (red). DAPI stain was used for nuclei (blue).

Fig. 12.

Conducting airway of lung showing expression of EGFP in nonciliated cells that sometimes express the V-ATPase B1 subunit. A: subset of airway cells showing a low level of V-ATPase B1-subunit immunostaining in apical region of cell (red). B: subset of airway cells showing EGFP expression (green). C: ciliated cells are identified by immunostain for tubulin (blue). D: arrows indicate EGFP-expressing cells (green) that also express V-ATPase B1 subunit (red), but these cells are identified as nonciliated by the lack of immunostaining for tubulin. Bar, 20 um.


In the present study, we took advantage of the specificity of the V-ATPase B1-subunit expression in specialized cells of the kidney and epididymis to produce a transgenic mouse that expresses EGFP selectively in these cells. Examination of the ATP6V1B1 gene promoter in human, rat, and mouse revealed a high degree of conservation within 6.5 kb of the 5′-flanking region. This suggested an important regulatory function of this region. Therefore, 6.5 kb of the human V-ATPase B1 subunit promoter that contained this conserved noncoding region was used to drive expression of EGFP in mice. High levels of EGFP were observed in all cells of the kidney and epididymis that express the V-ATPase B1 subunit, and no EGFP was detected in cells that do not contain this subunit. In addition, EGFP was also expressed in nonciliated airway epithelial cells in which the B1 subunit is also located. These results demonstrate the high efficiency of this promoter in driving EGFP expression in a cell-specific manner.

The study of subsets of specialized cells, including kidney intercalated cells and epididymal narrow/clear cells, has thus far been limited by the complexity of the organs in which they are found. The tubules that compose these organs contain various cell types, which all present specific characteristics. Our B1-EGFP transgenic mouse should prove useful for a variety of fluorescence microscopy-assisted techniques aimed at isolating intercalated cells and clear cells as well as collecting ducts. For example, we are now in a position to isolate these cells using fluorescence-activated cell sorting (FACS) of tissue digests either to culture them in vitro or to examine and characterize their gene and protein expression patterns. Also, it is now possible to microdissect specifically collecting ducts from collagenase-treated kidneys using fluorescence microscopy. Alternatively, laser capture microdissection, complemented with fluorescence microscopy, can now be used to isolate intercalated or clear cells from kidney or epididymis sections. With any one of these cell isolation methods, gene expression is determined using quantitative RT-PCR, microarray analysis, or immunoblotting under a variety of physiological and pathophysiological conditions. In this way, interference due to the contribution of neighboring cells is considerably (if not completely) reduced and the collecting duct, intercalated cell, and clear cell-specific gene expression profiles can be determined more precisely.

EGFP can also be used as a vital marker for the appearance of kidney intercalated cells and epididymal clear cells during development. Intercalated cells and epididymal clear cells appear progressively during pre- and postnatal development, respectively (5, 28, 30). By examining EGFP fluorescence, the onset of expression is easily determined and intercalated or clear cells can be isolated at each particular stage of development for examination of their gene and protein expression profiles.

These novel mice also are extremely useful for characterizing the regulation of the V-ATPase B1-subunit expression itself. EGFP is a vital marker that could be used as a reporter gene for expression of this subunit. Comparison of the proximal 10 kb of human and mouse V-ATPase B1 subunit promoters reveals that two regions of conservation are present in the promoter: a short proximal segment and a longer distal segment. On the basis of the current understanding of comparative genomics, these regions probably represent regulatory elements that may participate in the cell type-specific expression of the ATP6V1B1 gene (32). Clearly, further studies are required to elucidate the cell-specific regulatory elements and the transcription factors involved in the regulation of the V-ATPase B1-subunit expression in kidney intercalated cells and epididymal narrow and clear cells. Mating of these mice with other genetically modified mice that have the potential to alter development will allow for rapid and easy screening of the factors involved in the establishment of the intercalated cell and narrow/clear cell phenotype and/or in the expression of the V-ATPase B1-subunit or other genes in these cells under physiological and pathophysiological conditions.

A final point is that the expression of EGFP in nonciliated cells of the airway was an unexpected finding. A review of the literature indicates that the pH of airway surface liquid is relatively acidic (25). Although glandular epithelium secretes HCO3-rich fluid, the surface epithelium acidifies airway surface liquid (24). Furthermore, the acidification is inhibited by bafilomycin A1, suggesting involvement of vacuolar H+-ATPase (24). Airway pH may affect ciliary beat frequency (15), mucous rheology (22), and smooth muscle tone (39). Thus airway pH may be important for normal airway physiology. In addition, abnormal airway surface layer pH also may be important in diseases such as asthma (23) and cystic fibrosis (37). Further studies of V-ATPase are required to understand its role in lung physiology and pathophysiology.

In summary, the B1-EGFP-transgenic mice clearly exhibit specific cell expression of EGFP in the kidney and epididymis as well as in lung airway cells. Our studies demonstrate that cell-specific expression in these organs is conferred by 6.5 kb of the V-ATPase B1-subunit promoter. These mice represent a powerful new model for the study of the factors responsible for the establishment, maintenance, and modulation of these cell phenotypes.


This study was funded by National Institutes of Health Grants HD-40793 (to S. Breton and R. D. Nelson), DK-42956 (to D. Brown), and DK-53990 (to R. D. Nelson).


  • 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