The renal late distal convoluted tubules and connecting tubules are sites for the fine regulation of Na+ and Ca2+ reabsorption. The role of these segments in Na+ and K+ homeostasis is possibly underestimated, as the tubules are technically difficult to isolate in sufficient numbers and purity for functional analysis. To overcome these difficulties, we have developed a transgenic mouse model expressing enhanced green fluorescent protein in late distal convoluted tubules and connecting tubules. Enhanced green fluorescent protein expression was driven by the promoter for the transient receptor potential subfamily V, member 5. Confocal fluorescence microscopy allowed detection of enhanced green fluorescent protein in living, isolated late distal convoluted tubules and connecting tubules and in the initial cortical collecting ducts. Enhanced green fluorescent protein expression was validated by double- and triple-fluorescence immunolabeling with specific tubule markers. Freshly isolated late distal convoluted tubules and connecting tubules increased their intracellular Ca2+ levels in response to the V2 receptor-specific agonist deamino-Cys,d-Arg8-vasopressin (2 × 10−10 M) after 1 min of superfusion. In addition, both late distal convoluted tubules and connecting tubules displayed a concentration-dependent intracellular Ca2+ response to 1α,25-dihydroxyvitamin D3 (range 10−10 to 10−8 M). This suggests that 1α,25-dihydroxyvitamin D3 can act through a nongenomic signaling pathway in these tubules. In conclusion, the transgenic mouse model, expressing enhanced green fluorescent protein, is suitable for rapid isolation of viable late distal convoluted tubules, connecting tubules, and initial cortical collecting ducts and provides an ideal tool for a more exhaustive functional characterization of these segments.
- transgenic mice
- distal nephron
- collecting duct
- acute signaling
the distal tubule of the mammalian kidney reabsorbs 5–10% of the filtered Na+ load under normal conditions and participates in K+ secretion (13). Furthermore, it plays a central role in systemic Ca2+ and Mg2+ homeostasis and contributes to net acid secretion (26). All of these actions are tightly regulated by various hormones and local mediators (33), each acting on one or more tubular segments. The complex architecture of mouse distal tubules has been clarified by previous studies. In brief, the connecting tubule (CNT), containing both CNT cells and intercalated cells, arises gradually from the late portion of the distal convoluted tubule (DCT2), which contains DCT cells and a few intercalated cells. After fusion of two or more CNTs, the gradual occurrence of principal cells, and the disappearance of CNT cells, indicates the transition from CNT to the initial cortical collecting duct (iCCD), which emerges into the cortical collecting duct (CCD) in the medullary array (10, 23, 24).
Acute effects of single hormones or local mediators on the various renal transport pathways are difficult to assess in intact animals, as 1) various hormones act on the same transport processes; 2) hormone levels are often interdependent; 3) individual hormones act on multiple, interconnected pathways; and 4) the observed effects are difficult to ascribe to specific renal tubular segments. The understanding of how individual hormones act is essential for a full understanding of the regulation of transport pathways in the DCT2 and CNT. The majority of functional information on the DCT2 and CNT has been obtained from micropuncture studies in rats and rabbits (reviewed in Refs. 28 and 32) and in a recent study on genetically modified mice (31). A number of studies have been performed using isolated, perfused tubules from rabbit, rat, and mouse DCT2 and CNT (1, 4, 19, 39). Despite this functional information, the precise physiological and pathological roles of the segments are still being debated (25, 32, 34), as both DCT2 and CNT are difficult to isolate in sufficient numbers and purity for many experimental applications (27).
The present study was undertaken to overcome some of the technical difficulties in isolating viable DCT2s and CNTs and thereby provide a means of more exhaustive characterization of these nephron segments. We describe the development of a transgenic mouse model expressing enhanced green fluorescent protein (EGFP) driven by the promoter of the transient receptor potential subfamily V, member 5 gene (TRPV5), which is expressed in the DCT2 and CNT (17, 23). Based on the emitted fluorescence, DCT2, CNT, as well as iCCD, were easily and quickly isolated for functional studies. The approach constitutes a novel platform with which to study acute regulatory effects by single hormones on signaling and transport mechanisms in the DCT2 and CNT.
MATERIALS AND METHODS
Construction of mouse TRPV5 promoter-EGFP transgene and generation of transgenic mice.
A 3.6-kb fragment of the mouse TRPV5 promoter, deduced from the Ensemble genome browser (gene ID ENSMUSG00000036899), was cloned into the pAcGFP1-N1 vector (Clontech) using a bacterial artificial chromosome (BAC) clone (ResGen, Invitrogen) as a template. The PCR mixture was 2× Native plus pfu master mix (Stratagene), 200 ng BAC clone, and 3.0 μM mouse promoter TRPV5 (mpTRPV5) sense and antisense primers (MWG-Biotech) (see Table 2) in a total volume of 20 μl. The thermal cycling conditions were 1 cycle at 94°C for 5 min followed by 34 cycles of 94°C for 30 s, 68°C for 1 min, and 72°C for 4 min. A poly-A tail was added during an additional cycle at 72°C for 30 min. Amplification products were isolated from a 1% agarose gel stained with ethidium bromide by QIAquick gel extraction (Qiagen).
SalI and AgeI restriction sites (New England BioLabs) were added to the TRPV5 promoter fragment and cloned into the pAcGFP1-N1 vector. The PCR conditions in a total of 20 μl were 2× HotStart Taq master mix (Qiagen), 1 μg TRPV5 promoter DNA, and 1 μM of each modified promoter mTRPV5 sense and antisense primers (MWG-Biotech) (see Table 2), containing restriction sites at the 5′-end. To ensure restriction enzyme binding, an ATTATT sequence was added. The cycling program was as described for TRPV5 promoter amplification, except for an initial denaturing cycle at 94°C for 15 min. The PCR product was purified according to the QIAQuick PCR purification kit manual.
The TRPV5 promoter sequence containing restriction sites was inserted in a 3:1 ratio (according to the TOPO TA cloning instruction manual, Invitrogen) in the pAcGFP1-N1 vector and transformed into TOP10 chemically competent Escherichia coli (Invitrogen) using 50 μg/ml kanamycin as a selection agent.
Colonies were sequenced (MWG-Biotech), and the TRPV5 promoter-EGFP transgene was isolated from the vector backbone using the SalI and AflII restriction enzymes (New England BioLabs). To avoid vector contamination, the transgene was further digested with the NaeI restriction enzyme (New England BioLabs). TRPV5 promoter-EGFP transgenic mice were generated by pronuclear microinjection into C57BL/6 single-cell embryos (18). EGFP-positive founder mice (F0) were mated with C57BL/6 mice, resulting in EGFP-expressing transgenic mice and nontransgenic littermates. These heterozygous mice were used throughout the study. All procedures conformed with Danish animal welfare regulations. The generation of TRPV5 promoter-driven, EGFP-expressing mice was reported to the national authorities through the institutional veterinarian. The authors are licensed to breed the mice and conduct the described experiments by the Danish Ministry of Justice.
Measurements of urine Ca2+, Na+, K+, pH, and osmolality.
Osmolality was determined using a Wide-Range Osmometer 3W2 (Advanced Instruments), and pH was determined by a microcombination pH electrode (microelectrodes) with two-point calibration. Na+ was determined by flame photometry (FLM3, Radiometer) and by a microion monosodium electrode and a microreference electrode (Microelectrodes) using five-point standard curves, respectively. Voltages were recorded using a Radiometer Analytical PHM 240 pH/ion meter. Urinary Ca2+ and K+ levels were determined by flame photometry (FLM3, Radiometer).
Protein extraction and immunoblotting.
Both EGFP-positive and EGFP-negative mice were anesthetized by isoflurane inhalation and euthanized before dissection of various tissues for immunoblotting. The various tissues were homogenized (Ultra-Turrax T8 homogenizer, IKA Labortechnik, Staufen, Germany) in ice-cold dissection buffer (in mM: 300 sucrose, 25 imidazole, 1 EDTA, pH 7.4) containing the protease inhibitors leupeptin (8.5 μM) and phenylmethylsulfonyl fluoride (1 mM). To remove cellular debris and nuclei, the homogenates were centrifuged at 4,000 g for 15 min at 4°C. The total protein concentration was measured (BCA protein assay reagent kit, Pierce, Rockford, IL). Supernatants were dissolved in Tris buffer (final: 3% SDS, 8.7% glycerol, pH 6.8, bromophenol blue, and 30 mg/ml dithiothreitol) for 15 min at 65°C. SDS-PAGE was performed with 12.5% polyacrylamide gels (Criterion gels, Bio-Rad) at 100 V for 70 min. Proteins were electrotransferred onto Hybond-ECL nitrocellulose membranes (Amersham Biosciences) for 60 min at 100 V. Membranes were blocked in 5% nonfat dry milk in PBS-T (in mM: 281 Na+, 100 Cl−, 21 H2PO4−, 80 HPO42−, and 0.1% Tween 20, pH 7.5) for 1 h at room temperature and incubated overnight at 4°C with primary antibodies. The sites of antibody-antigen reaction were visualized with enhanced chemiluminescence system (ECL Plus Western Blotting detection system, GE Lifesciences) and exposed to photographic film (Hyperfilm ECL, GE Lifesciences).
Tissue fixation and immunohistochemistry.
EGFP-positive and EGFP-negative mice were anesthetized by isoflurane inhalation and perfusion fixed with 3% paraformaldehyde in phosphate buffer (PBS; in mM: 167 Na+, 2.8 H2PO4−, 7.2 HPO42−, pH 7.4). The tissues were dehydrated in graded ethanol, incubated overnight in xylene, embedded in paraffin wax, and cut in 2-μm-thick sections using a rotary microtome (Leica). The sections were dewaxed with xylene and rehydrated with graded ethanol. Endogenous peroxidase was blocked by 0.5% H2O2 in absolute methanol for 10 min. Antigens were retrieved by boiling for 10 mM Tris with 0.5 mM EGTA, pH 9, for 10 min. After cooling, the sections were quenched with 50 mM NH4Cl in PBS for 30 min and blocked in PBS with 1% BSA, 0.2% gelatin, and 0.05% saponin. The sections were incubated overnight at 4°C with primary antibodies diluted in PBS with 0.1% BSA and 0.3% Triton X-100. Upon washing, the sections were incubated with fluorophore-conjugated secondary antibodies in PBS with BSA and Triton X-100. After washing, sections were mounted with a coverslip in Glycergel Antifade Medium (Dako) and inspected on a Leica DMIRE2 inverted microscope with a TC5 SPZ confocal unit using a ×64/1.32 numerical aperture HCX Pl Apo objective. The immunofluorescence images were merged with differential interference contrast images to reveal the relationship between the tissue structures and the fluorescence labeling. Digital images were processed and analyzed using Image-Pro Plus software (Media Cybernetics, Silver Spring, MD). For low-resolution images, slides were scanned on a LiCor Odyssey scanner with 21-μm resolution.
Immunogold electron microscopy.
Tissue blocks prepared from mouse kidneys were cryoprotected with 2.3 M sucrose and rapidly frozen in liquid nitrogen. The samples were freeze substituted by sequential equilibration over 3 days in methanol containing 0.5% uranyl acetate at temperatures raised gradually from −80 to −70°C, rinsed in pure methanol for 24 h while the temperature was increased from −70 to −45°C, and infiltrated with Lowicryl HM20 and methanol 1:2, 1:1, 2:1, and then pure Lowicryl HM20 before ultraviolet polymerization for 2 days at −45°C and 2 days at 0°C. Ultrathin sections on nickel grids were target retrieved with saturated NaOH in absolute ethanol (for 2–3 s), rinsed, and blocked for 15 min with 50 mM glycine, 0.1% skim milk powder, and 20 mM NaN3 in PBS. Sections were rinsed and incubated overnight at 4°C with primary antibody diluted in 0.1% skim milk in PBS. Upon rinsing, the sections were incubated for 1 h at room temperature with colloidal gold-conjugated secondary antibody (BioCell Research Laboratories) in 0.1% skim milk powder, 0.06% polyethylene glycol, and 0.1% fish gelatin in PBS. The sections were washed and counterstained with uranyl acetate and lead citrate before examination in an FEI Morgagni 268D electron microscope.
Primary antibodies were goat anti-GFP (ab6673), rabbit anti-GFP (ab290) (both from Abcam), mouse anti-calbindin-D28K (Research Diagnostics), rabbit anti-AQP2 (7661) (12), rabbit anti-V1-ATPase B1 subunit (7659) (7), chicken anti-AQP2 (3), rabbit anti-NCC (TSC357) (20), and rabbit anti-TRPV5 (ECAC1AP, Alpha Diagnostics). For immunoblotting, horseradish peroxidase-conjugated anti-rabbit was used as a secondary antibody (Dako). Fluorescent secondary antibodies were donkey anti-goat Alexa 488, 546, and 633; donkey anti-rabbit Alexa 488, 555, 647, and 680; donkey anti-mouse Alexa 555 and 800; goat anti-rabbit Alexa 633; and goat anti-chicken Alexa 488 (Invitrogen). Topro3 was used as a nuclear marker (Invitrogen). Anti-rabbit IgG conjugated to 10-nm colloidal gold particles (BioCell Research Laboratories) was used as a secondary antibody in immunogold electron microscopy.
Renal tubule isolation.
Mice were anesthetized by isoflurane inhalation and perfused through the left ventricle with 1 mg/ml collagenase type-IV (PAN Biotech) and 1 mg/ml pronase (Roche Diagnostics) in 10 ml isolation buffer (in mM: 150 Na+, 3.6 K+, 1.0 Mg2+, 1.3 Ca2+, 140 Cl−, 0.4 H2PO4−, 1.6 HPO42−, 1.0 SO42−, 10 acetate−, 1.3 gluconate−, 1.0 α-ketoglutarate, and 5 glycine) containing 48 mg/l trypsin inhibitor and 25 mg/l DNase I at pH 7.4 and 37°C. Kidneys were quickly removed, and the renal cortex was transferred into 2 ml of isolation buffer containing 1 mg/ml collagenase type-II (PAN Biotech) and 1 mg/ml pronase and swirled at 850 rpm with a 3-mm mixing stroke at 37°C (Eppendorf Thermomixer). After 10 min, a 1-ml tubule suspension was transferred into 1 ml of ice-cold isolation buffer containing albumin (0.5 mg/ml) and kept on ice. One milliliter of isolation buffer (37°C) was added to the remaining tubule suspension, and the enzymatic reaction was continued for 5 min at 850 rpm and 37°C. This procedure was repeated three times, resulting in four tubule fractions. Isolated and sedimented tubules were resuspended in ice-cold HEPES-buffered salt solution (HBS; in mM: 145.0 Na+, 3.6 K+, 1.8 Ca2+, 0.8 Mg2+, 138.6 Cl−, 0.8 SO42−, 0.4 H2PO4−, 1.6 HPO42−, 10.0 HEPES, 5.6 glucose, pH 7.4) and kept on ice until use. The tubule preparations were validated using a dissection microscope (Leica MZ125) equipped with green emission filters (∼535/50-nm band-pass, Chromas) and a blue LED as a light source (IMAC IDBA-C27/34B). We readily identified the fluorescent segments in EGFP-expressing mice and subdivided these into DCT2, CNT, and iCCD tubules morphologically using differential interference contrast. In wild-type mice, identification was based only on morphology. The isolated DCT2 and CNT segments appear much thinner than the proximal tubules and lack the brush border. DCT2 and CNT appear to contain two cell types and are wider and more curved than the thick ascending limb (TAL). iCCD appear thin and cobblestone-like compared with DCT2 and CNT, and without basal stripes. Furthermore, the wall thickness-to-lumen diameter ratio is much lower in iCCD than in CNT. Branching sites were sometimes observed in CNTs and iCCDs. In a few examples, the TAL, DCT, CNT, and iCCD were still connected due to the mild isolation.
Measurement of intracellular Ca2+ levels.
Five hundred microliters of tubule suspension was transferred to Cell-Tak (BD Biosciences) adhesive-coated coverslips and allowed to attach for 30 min at 37°C. Coverslips were mounted in a RC-21BR closed perfusion chamber (358-μl perfusion chamber with 1-ml/min inlet giving linear flow rate of 0.8 mm/s, Harvard Apparatus) on the stage of an inverted microscope (T-2000, 60×/1.4 numerical aperture oil immersion Plan Apo VC objective, Nikon) situated in a heated dark box. Images of tubular EGFP fluorescence and differential interference contrast were acquired. Tubular cells were then loaded for 10 min at 37°C with the fluorescent Ca2+ indicator fluo 4-AM (10 μM, final concentration, Invitrogen) in HBS with 5 mM probenecid (Sigma-Aldrich) to retain the probe inside the cells. The cells were illuminated with a 495-nm light from a Polychrome V monochromator (Till Photonics). The 510- to 535-nm light emission from the fluo 4-loaded cells was collected by a 12-bit cooled CCD monochrome camera (QImaging, Retiga EXi). QED InVivo imaging software (Media Cybernetics) was used to control wavelength, light exposure time (40 ms), frequency (1 image pair each 4 s), and binning (348 × 260-pixel images), as well as the data collection from user-defined regions of interest (individual cells). All experiments were run at pH 7.4 and 37°C and started with HBS superfusion for at least 2 min. Our inclusion criteria were cells capable of 1) taking up the fluo 4-AM dye, 2) cleaving the membrane-permeable form of the dye (fluo 4-AM) to the fluorescent fluo 4, and 3) retaining the dye without leak for the duration of the observation period. Cellular Ca2+ levels were recorded during HBS superfusion with 10−10, 10−9, or 10−8 M 1α,25 dihydroxyvitamin D3 [1α,25(OH)2D3,], 2 × 10−10 M deamino-Cys,d-Arg8-vasopressin (dDAVP), or 2 × 10−6 M ionomycin (all from Sigma-Aldrich) in the presence or absence of 1.8 mM Ca2+ and 1 mM EGTA, as indicated on traces (see Fig. 5B). Based on our immunohistochemical analysis of kidney sections, ∼32% of the cells in the EGFP-positive tubules were EGFP and H+-ATPase double negative and therefore useful for the Ca2+ recording. In the functional experiments, 27.8 ± 2.7% of the EGFP-negative cells reacted to dDAVP and 1α,25(OH)2D3, confirming acceptable viability. In wild-type mice, ∼63% of the cells were H+-ATPase negative by immunostaining and 52.3 ± 3.3% of the cells reacted to the drugs in functional experiments. Data from 3–10 individual cells per recording were pooled for each experiment. Each animal was regarded as one experiment. The cellular fluorescence readings were normalized to the initial fluorescence of the individual cell, and the fluorescence peaks were identified and quantified using IgorPro software (Wavemetrics). Only cells responding to ionomycin in the presence and absence of extracellular Ca2+ were included in the analysis.
Data were analyzed using a Mann-Whitney nonparametric test (urine analysis) and paired t-test (Ca2+ recording) using Graph Pad InStat3 software. The level of statistical significance was chosen as P < 0.05.
Generation of TRPV5 promoter-EGFP transgenic mice.
TRPV5 promoter-EGFP transgenic mice were generated from pronuclear injection of a linearized TRPV5 promoter-EGFP transgene (Fig. 1A). Founder mice (F0) transmitted the transgene to their offspring, reflecting the random integration of the TRPV5-EGFP transgene into the genome. Seven EGFP-positive F0 generation mice were mated with C57BL/6 mice, and the heterozygous F1 generation was characterized in this study.
Tissue profile of EGFP expression.
Figure 1B shows an immunoblot of multiple tissues with a specific anti-GFP antibody. EGFP-expressing transgenic mice displayed strong EGFP protein levels in the kidney, intermediate levels in the cerebellum, and low levels in the cecum. EGFP expression was not observed in nontransgenic littermates. Figure 1C shows a low-magnification scanning of anti-EGFP (green) and anti-calbindin (red) immunofluorescence in the kidney of an EGFP-expressing transgenic mouse (left) and a nontransgenic littermate (right). The EGFP-expressing transgenic mouse displayed partial overlap of EGFP and calbindin-D28k in cortical segments. EGFP extended further into the outer medulla.
Analysis of EGFP expression in isolated living renal tubules.
Figure 2A shows a schematic representation of the renal tubular system with an indication (green line) of the known TRPV5 expression sites, i.e., DCT2 and CNT (17, 23). The confocal micrograph in Fig. 2B shows EGFP fluorescence readily detected from freshly isolated DCT2 and CNT of the EGFP-expressing transgenic mouse. A subset of CCDs also produced EGFP fluorescence (Fig. 2C). Some EGFP-positive tubules displayed partial EGFP expression among the cells. The number of non-EGFP-expressing cells in the EGFP-positive tubules and the fluorescence intensity varied among the different EGFP-expressing transgenic mouse lines. Seven mouse lines with different EGFP expression levels and fluorescence intensities were chosen for immunohistochemical analysis (see below). Two mouse lines, a high-level and a low-level EGFP-expressing mouse line, were chosen for functional analysis (see below). No other renal or nonrenal structures displayed detectable EGFP fluorescence (not shown).
Ca2+, Na+, K+, pH, and osmolality levels in urine from EGFP-expressing transgenic mice did not differ from nontransgenic littermates [not significant, n = 3 (EGFP) and n = 4 (nontransgenic) for Na+, pH, and osmolality; n = 5 (EGFP) and n = 3 (nontransgenic) for Ca2+, K+, and Na+ by flame photometry] (Table 1).
Immunohistochemical analysis of renal tubular EGFP expression.
Double labeling of kidney sections of EGFP-expressing transgenic mice, using anti-EGFP and anti-TRPV5 antibodies, displayed a pronounced cellular colocalization of TRPV5 and EGFP in DCT2 and CNT (Fig. 2D). In superficial CCDs, high-level EGFP-expressing transgenic mice displayed more EGFP-positive cells than TRPV5-positive cells (Fig. 2E). In addition, EGFP-positive cells were observed in a few deeper CCDs (Fig. 2F).
The exact EGFP expression sites were determined by triple labeling of kidney sections using anti-EGFP antibodies and various tubule markers. The Na+-Cl− cotransporter NCC is expressed selectively in the DCT (6), whereas the Ca2+ binding protein calbindin-D28K is weakly expressed in DCT1, with increasing intensity toward the DCT2 and decreasing expression in the CNT (23). EGFP colocalized with NCC in high-calbindin-D28K-abundant DCT areas (DCT2) (Fig. 3A). Colocalization of calbindin-D28K and EGFP extended into non-NCC-expressing segments (CNT).
In the renal cortex, aquaporin-2 (AQP2) is expressed in CNT cells and CCD principal cells (14). EGFP and AQP2 colocalized in some segments (Fig. 3B). In addition, we observed a portion of the EGFP-positive nephron that expressed neither AQP2 nor NCC. As was observed by TRPV5 and EGFP double labeling, AQP2, NCC, and EGFP triple labeling revealed EGFP expression in some cortical collecting ducts (Fig. 3C). Furthermore, we observed EGFP and AQP2 colocalization in only a fraction of the AQP2-positive cells, indicating that EGFP is not expressed in all principal cells.
Throughout DCT2, CNT, CCD, the outer medullary collecting duct (OMCD), and initial inner medullary collecting duct (IMCD), intercalated cells express the vacuolar proton pump (H+-ATPase) (5). Despite the fact that all cortical H+-ATPase-positive tubules expressed EGFP, we did not observe cellular colocalization of EGFP and H+-ATPase in either DCT2, CNT (Fig. 3D), or CCD (Fig. 3E). Thus EGFP expression is confined to DCT2 cells, CNT cells, and iCCD principal cells. EGFP expression in high-level EGFP-expressing transgenic mice extended into the CCD in the medullary arrays.
Immunogold electron microscopic localization of renal tubular EGFP expression.
A high-magnification and -resolution technique confirmed cytosolic EGFP expression in DCT2 and CNT cells in EGFP-expressing transgenic mice but not in nontransgenic littermates. Figure 4A shows the junction of two neighboring cells in CNT from an EGFP-expressing transgenic mouse. The top cell is labeled with gold particles (indicated by arrows). The bottom cell is EGFP negative. This finding confirms the occurrence of EGFP-negative nonintercalated cells. Labeling was not observed in distal tubules of nontransgenic mice (Fig. 4B). In iCCD, we observed EGFP-positive principal cells (indicated by arrows) (Fig. 4C), whereas intercalated cells were EGFP negative (Fig. 4D). No other tubules displayed EGFP immunoreactivity.
Imaging of intracellular Ca2+ levels in freshly isolated tubules.
DCT2, CNT, or iCCD tubules were identified using a combination of differential interference contrast (Fig. 5A, top) and EGFP fluorescence (Fig. 5A, middle). Tubules were loaded with fluo 4-AM (Fig. 5A, bottom) and superfused with various HBS at 37°C as indicated above the traces (Fig. 5B). Non-EGFP-expressing cells from high-level EGFP-expressing transgenic mice responded to superfusion with dDAVP (2 × 10−10 M) by a transient increase in the intracellular Ca2+ concentration ([Ca2+]i) (P < 0.05, n = 4). A similar response was observed in low-level EGFP-expressing cells of transgenic mice (data included in mean values) (Fig. 5C). In these cases, the EGFP fluorescence signal (i.e., the difference in minimal fluorescence level in Ca2+-free/ionomycin solution of EGFP-expressing and -nonexpressing cells) was subtracted from the total fluorescence before normalization of the fluo 4 signal. Only EGFP-negative cells were analyzed in high-level EGFP-expressing mice, as the EGFP fluorescence exceeded the fluo 4 signal. There was no statistically significant difference in amplitudes between tubule cells from EGFP-expressing transgenic mice and nontransgenic littermates (P = 0.93, n = 4).
TRPV5, localized along the DCT2 and CNT, is functionally regulated by 1α,25(OH)2D3 (16). However, to our knowledge, no studies have reported transient 1α,25(OH)2D3-induced Ca2+ signals in DCT2 and CNT. A physiological concentration of 1α,25(OH)2D3 (10−9 M) induced a [Ca2+]i signal in dDAVP-responsive DCT2 and CNT cells from both EGFP-expressing and non-EGFP-expressing mice (P < 0.05, n = 4, Fig. 5C). Amplitudes between tubule cells from EGFP-expressing transgenic mice and nontransgenic mice showed no statistically significant difference (P = 0.66, n = 4). Figure 5C summarizes the mean amplitudes of the Ca2+ signals relative to baseline fluorescence levels in 10 tubules (DCT2, CNT, and iCCD) from four EGFP-expressing transgenic mice and four nontransgenic mice. The [Ca2+]i response was dependent on the 1α,25(OH)2D3 concentration in the range from 10−10 to 10−8 M (P < 0.05, n = 5, Fig. 5D). The [Ca2+]i level during each of these recordings was verified by clamping the [Ca2+]i to [Ca2+]e using the Ca2+ ionophore ionomycin (2 × 10−6 M). High [Ca2+]i was obtained using normal-Ca2+ HBS (1.8 × 10−3 M, saturating the dye), whereas the minimal fluo 4 signal was obtained during superfusion with a Ca2+-free HBS containing EGTA. The fluorescence signal of EGFP-positive cells from high-level EGFP-expressing transgenic mice did not change with either dDAVP, 1α,25(OH)2D3 or ionomycin superfusion or with fluo 4 loading (not shown).
This study reports a new transgenic mouse model enabling rapid isolation of DCT2, CNT, and iCCD based on TRPV5 promoter-driven EGFP expression in these tubules. Cellular viability of EGFP in DCT2, CNT, and iCCD was demonstrated after tubule isolation using a modified enzymatic protocol. We show for the first time that dDAVP and 1α,25(OH)2D3 induce [Ca2+]i signals in mouse DCT2 and CNT cells.
Immunoblotting and detailed immunohistochemical analysis showed that the EGFP-expressing transgenic mice expressed TRPV5 protein in tissues as previously reported (17, 23). As expected, the TRPV5 labeling in DCT2 and CNT coincided with pronounced EGFP expression in the EGFP-expressing transgenic mice. In the iCCD, however, EGFP expression was observed in TRPV5-positive as well as TRPV5-negative cells. In high-level EGFP-expressing mice, EGFP expression extended farther into the medullary array CCD. The TRPV5 expression in iCCD has been shown previously (23), whereas the reason for the apparent “overflow” of EGFP expression in TRPV5-negative cells remains unknown. However, we suggest three explanations for the apparent EGFP overflow: 1) lack of one or more native TRPV5 promoter suppressors, 2) expression of previously undetectable levels of TRPV5 in CCD, or 3) distal tubular luminal EGFP secretion and subsequent apical uptake by the collecting ducts. However, the lack of EGFP fluorescence in urinary samples from both low-level and high-level EGFP-expressing transgenic mice (not shown) renders the last suggestion very unlikely. Moreover, the measurements of renal function with respect to urinary Ca2+, Na+, and K+ concentrations, pH, and osmolality did not differ from the nontransgenic littermates. Taken together, it appears that EGFP expression is restricted to DCT2 cells, CNT cells, and iCCD principal cells in the EGFP-expressing transgenic mice and that EGFP-expressing tubules are suitable for physiological studies of these tubular segments.
In addition, we observed a lack of detectable AQP2 labeling in the initial portion of the CNT (iCNT). It has previously been reported that rats, in the absence of exogenous vasopressin, lack detectable AQP2 labeling in the iCNT. However, chronic treatment with vasopressin induced AQP2 expression throughout the rat CNT including this presumable initial portion (8). Therefore, it is likely that the NCC- and AQP2-negative (but calbindin-D28k-positive) distal tubules represent iCNT in the mouse.
The challenge of isolating tubules for physiological experiments was overcome by the application of a modified enzymatic isolation protocol. The first indications of good cellular viability were their capability to 1) take up the fluo 4-AM dye, 2) activate the fluo 4 by action of intracellular esterases, and 3) retain the fluo 4 effectively during the experiments. The fluo 4 dye has a similar excitation and emission spectra as EGFP (15). A disadvantage of using fluo 4 instead of the widely used near-UV indicator fura 2 is that fluo 4 becomes undetectable in high-level EGFP-expressing cells. However, our objective in the present study was mainly to study the EGFP-negative cells of the EGFP-positive tubule segments to avoid concerns regarding the physiological relevance of cells expressing EGFP in general. Compared with fura 2, fluo 4 displays a greater sensitivity to normal cytosolic Ca2+ changes. Furthermore, the applied low intensity at 488-nm excitation is less damaging to the cells than the UV excitation used with fura 2. In conclusion, the approach allows 1) identification of the EGFP-expressing tubules of interest, 2) subsequent loading of the fluo 4-AM, and 3) recording of [Ca2+]i without changing emission filters or excitation wavelength.
To demonstrate the value of the combined approach of TRPV5-promoter-driven EGFP-expressing transgenic mice and the tubule isolation protocol, tubules were superfused with the V2 vasopressin receptor-specific agonist dDAVP and the active 1α,25(OH)2D3. AVP is a key hormone in the regulation of Na+, Ca2+, and H2O reabsorption along the renal distal nephron and collecting duct (9, 30, 40). It is well known that vasopressin induces an increase in rat collecting duct [Ca2+]i levels (36, 37). A similar effect has been reported in isolated rabbit CCD (2), as well as in the cultured rabbit CNT (40). AVP also induce [Ca2+]i increases in mouse CCD (29), but to our knowledge there are no reports of vasopressin affecting [Ca2+]i in mouse DCT2 or CNT. We find a significant increase in [Ca2+]i within 2 min of superfusion by dDAVP in freshly isolated mouse DCT2, CNT, and iCCD. The Ca2+ signaling from both low-level and high-level EGFP-expressing tubules was comparable to similar tubules from nontransgenic mice, indicating that both high-level and low-level EGFP-expressing cells may be relevant in functional studies.
1α,25(OH)2D3 is classically described to act via modification of gene transcription (35). However, 1α,25(OH)2D3 induces an acute Ca2+ signaling in rabbit proximal tubules as well as chicken intestine (11, 38). To our knowledge, acute effects of 1α,25(OH)2D3 have not been reported for the distal nephron. It is well known that calbindins have a tightly buffering effect (22) and that Ca2+ signaling is unaffected by the reabsorptive activity (21). Therefore, Ca2+ reabsorption in CNT occurs without significantly affecting the cytosolic free Ca2+ level. Furthermore, transcellular Ca2+ movement is unlikely to be significant in our experimental set-up due to the lack of perfusion of the tubular lumen. We demonstrate that [Ca2+]i levels in EGFP-positive tubule cells increase significantly within 2 min of superfusion with 1α,25(OH)2D3. Furthermore, we demonstrate a 1α,25(OH)2D3 concentration dependency of the [Ca2+]i level in DCT2 and CNT, validating the acute 1α,25(OH)2D3 response. These findings indicate a nongenomic effect of 1α,25(OH)2D3 in DCT2 and CNT. Although these observations are potentially of great interest, the functional relevance with respect to acute regulation of tubular ion transport remains to be elucidated.
The straightforward identification of viable fluorescent tubules enables a variety of future studies of the DCT2, CNT, and iCCD. Tubules are easily identified in a standard dissection microscope with minor modifications, as described. DCT2 and CNT can also be isolated by fluorescence-assisted microdissection for studies of perfused tubules. For large-scale applications, tubules can be sorted using a biosorter as described by Miller and colleagues (27). TRPV5-EGFP transgenic mice appear to be phenotypically normal. Therefore, they can be mated with any genetically modified mouse line of relevance to distal nephron function and dysfunction, allowing easy analysis of a nephron-specific phenotype. In conclusion, TRPV5-promoter-driven EGFP was expressed in DCT2, CNT, and iCCD. Tubules containing fluorescent cells were isolated enzymatically and responded acutely to dDAVP and 1α,25(OH)2D3 treatment by [Ca2+]i signaling. Thus we have generated an EGFP-expressing transgenic mouse model that will be used for numerous future functional studies on DCT2 and CNT.
Support for this study was obtained from the Danish Medical Research Council, Karen Elise Jensens Fond, Lundbeckfonden, and Novo Nordisk Fonden. M. V. Hofmeister is supported by the Faculty of Health Sciences, University of Aarhus. R. A Fenton is supported by a Marie Curie Intra-European Fellowship. The Water and Salt Research Centre at the University of Aarhus is established and supported by the Danish National Research Foundation (Danmarks Grundforskningsfond).
The authors thank Inger Merete S. Paulsen, Christian V. Westberg, Zhila Nikrozi, and Helle Høyer for expert technical assistance. The laboratory of Ernst-Martin Füechtbauer assisted in generating the transgene mouse line. Tove Lindahl Andersen and Torben Clausen are thanked for assistance regarding flame photometry. Nina Himmerkus and Markus Bleich are thanked for sharing a method for tubule isolation and Helle A. Praetorius for fruitful discussions and constructive comments on the manuscript.
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.
- Copyright © 2009 the American Physiological Society