We sought to assess whether the distal convoluted tubule (DCT) segment of the rabbit nephron expresses a functional epithelial sodium channel. First, the transepithelial voltage (V te, lumen vs. bath) was measured in isolated perfused DCT segments (assessed separately in the upstream half and the downstream half of the DCT). V tewas zero and not affected by amiloride or barium in the upstream DCT.V te was sometimes negative in the downstream DCT and depolarized by amiloride and hyperpolarized by barium, suggesting inclusion of connecting tubule (CNT) cells. To determine expression of epithelial sodium channel (ENaC) mRNA subunits by the upstream DCT, rabbit α-, β-, and γ-ENaC cDNA fragments were cloned and primers were selected for single-nephron RT-PCR analysis. Although α-ENaC was expressed by the DCT, β- and γ-ENaC were not detected in the DCT. In contrast, the CNT, CCD, and outer medullary collecting duct (OMCD) expressed all three subunits. Nedd4 was also not detected in the DCT but was expressed by the CNT, CCD, and OMCD. When upstream DCT fragments were grown to confluent monolayers in primary culture, the epithelia exhibited negative voltages and high transepithelial resistances and expressed mRNA for all three ENaC subunits as well as for Nedd4. The absence of a negative voltage and failure to detect transcript for β- and γ-ENaC and Nedd4 in the native rabbit DCT suggest that the sodium channel is not a significant pathway for sodium absorption by this segment. The phenotype conversion observed when DCT cells are grown in culture does not rule out the possibility that there may be conditions in which the DCT in the intact kidney expresses sodium channel activity. The results are consistent with the notion that DCT sodium transport is predominantly, if not exclusively, electroneutral.
- epithelial sodium channel
- cell culture
the presence of amiloride-sensitive sodium channels in the mammalian renal distal tubule1 and the cortical collecting duct (CCD) is well established. Although it is clear that the late2 portion of the distal tubule exhibits both amiloride-sensitive net sodium absorption and a lumen-negative transepithelial voltage (V te) (4, 11, 19), the expression of sodium channel activity in the early3 portion of the distal tubule is controversial. Data from at least two groups (5, 8,29) that assessed the function of the early distal tubule [predominantly distal convoluted tubule (DCT) epithelium] in the intact rat kidney suggest that electrogenic sodium transport is not the primary mechanism for sodium absorption. These studies demonstrated the primary site of action of the thiazide class of diuretics is in the early distal segment, whereas the late distal segment is the primary site of action of the potassium-sparing diuretic amiloride. Unfortunately, the high degree of cellular heterogeneity and the short length of the distal tubule do not permit ruling out an effect of amiloride in DCT cells nor ruling out an effect of thiazides in initial collecting tubule (ICT) cells.
Studies using a cell line derived from immunodissected mouse DCT+thick ascending limb (TAL) cells (21) and having features of DCT cells have demonstrated amiloride-sensitive transport in these MDCT cells in culture (10). This flux cannot be localized to the apical or the basolateral membrane, however, because MDCT cells do not form tight junctions and do not develop aV te or resistance (R te). These studies do not establish whether an amiloride-sensitive sodium channel is a feature of the DCT of the intact kidney.
The cloning of the epithelial sodium channel (ENaC) in the rat (2) has enabled investigators to probe for mRNA expression in kidney and to perform immunolocalization studies. Duc and co-workers (7) concluded from studies using both in situ hybridization methods and immunocytochemistry that all three subunits of ENaC, α-, β-, and γ-, are expressed along the rat distal nephron extending from the DCT to the outer medullary collecting duct (OMCD). Ciampollillo and co-workers (3) used in situ PCR methods to localize α-ENaC to the region of the nephron extending from the medullary TAL (mTAL; including the DCT) to the inner medullary collecting duct (IMCD) (3). Positive identification of the DCT in histological sections undergoing these procedures is often not possible, however, and additional markers to co-localize DCT cells were not used in these studies.
In vitro studies of the isolated and perfused DCT of the rabbit kidney report luminal negative voltages (15, 24, 25, 31) and amiloride effects (24, 25). We found that the luminal negative V te was low; however, it varied over a wide range. It is possible that this variability is caused by inadvertent inclusion of CNT cells in the perfused structure.
ENaC regulation by a cell-expressed, developmentally downregulated neuronal precursor (Nedd4) (26, 27) has been identified as an important mechanism by which the level of sodium channel expression in the apical membrane is controlled. In an immunohistochemical study (28), Nedd4 expression was detected primarily in the distal portions of the nephron but was notably absent from the DCT segment. Thus it is possible that both ENaC and Nedd4 are not expressed in the DCT.
The purpose of the present study is to examine whether the DCT segment of the rabbit nephron expresses the functional sodium channel. To do this we assessed amiloride-sensitive V te in isolated perfused DCT subsegments and determined α-, β-, and γ-ENaC mRNA expression by using the single-dissected-nephron RT-PCR technique. Second, to determine whether ENaC expression correlates with Nedd4 expression, we determined the distribution of Nedd4 mRNA along the nephron. Third, we determined expression of ENaC and Nedd4 mRNA by DCT cells grown in primary culture and compared this with native tissue.
Dissection of nephron segments.
New Zealand White rabbits (1–2.5 kg) were anesthetized with ketamine (50 mg/kg im) and pentobarbital sodium (50 mg/kg iv), kidneys removed via a flank incision and chilled in ice-cold dissecting solution (pH 7.4; in mM: 135 NaCl, 1 Na2HPO4, 1.5 MgSO4, 2 CaCl2, 5 KCl, 5 glucose, 5 HEPES). The rabbit was euthanized by overdose with pentobarbital sodium. RNAse-free media, beakers, and instruments were used when tubules were to be used for RT-PCR experiments. Transverse slices (1 mm) of the kidney were transferred to a dissection dish that was placed on the stage of a dissecting microscope (transmitted light, model SZH, Olympus) fitted with a water-chilled chamber. Sections were transilluminated, and tubules were dissected by hand at ×60 magnification by using sharpened forceps.4 Guidelines for dissection in the heterogeneous distal portion of the nephron were taken from Morel (16).
The DCT was identified by locating and dissecting the structure consisting of a glomerulus with associated upstream TAL and downstream DCT and CNT segments. The procedure for isolating and harvesting the DCT portion is described as follows. The nephron is cut within the DCT, downstream of the transition from TAL to DCT. This transition usually occurs close to the glomerulus, and the abrupt change in tubule diameter is not visible (when the transition is visible, the DCT is cut downstream to this site). Within the DCT, the nephron generally makes a hairpin turn and returns to contact its own glomerulus. Before the nephron makes contact with the afferent arteriole (6), it undergoes a visible transition from DCT to CNT (DCT length is ∼0.5 mm from TAL transition to CNT transition). To harvest the DCT for RT-PCR and primary culture experiments, the tubule is cut at the hairpin turn, yielding ∼300 μm of DCT tissue/dissected nephron. This fragment of DCT (see Perfusion of isolated DCT segments in vitro) does not contain TAL cells or CNT cells.
Proximal convoluted tubules (PCT) cortical thick ascending limbs (cTAL) and cortical collecting ducts (CCD) were dissected from the rays in the cortex. Proximal straight tubules (P3), mTAL, and OMCD were dissected from the outer medulla. Glomeruli (G) and connecting tubules (CNT) were dissected from the labyrinth and arcades. IMCD were dissected from the inner medulla. For these segments it was usually possible to obtain single nephron fragments at least 1 mm in length. To minimize contamination of specific nephron samples with other cells, small bundles of tissue were separated first from kidney slices and then transferred to a new dish for final dissection in fresh medium.
Perfusion of isolated DCT segments in vitro.
For in vitro perfusion of the DCT, the entire hairpin structure (seeDissection of nephron segments) is utilized, and the segment harvested extends from the TAL-DCT transition to the DCT-CNT transition (visual criteria used to exclude TAL and CNT cells). Subsequently, cutting once at the top of the hairpin turn yields two fragments, the “upstream” DCT and the “downstream” DCT.
Individual tubules are then transferred to a bath chamber (Hampel) mounted on the stage of an inverted microscope (Zeiss IM35). The microscope is fixed to a vibration isolation table (Micro-G). The bath contains control perfusion solution (continuously flowing), with its temperature maintained at 37°C. Each end of the tubule is aspirated into a glass tubule-holding pipette fitted into a tubule perfusion system (Luigs and Neumann) according to the method of Burg (1) and modified by Greger and Hampel (12). A third pipette concentric to one of the holding pipettes is now advanced into the lumen of the tubule (perfusion end) and permits perfusion of the lumen. Fluid is collected continuously at the other end (collection end). An outermost pipette with a dielectric liquid (Sylgard) in the tip is advanced over the tubule-holding pipette and the first part of the tubule to make an electrical seal and prevent short-circuiting theV te. V te is measured at the perfusion end via the perfusion pipette that is connected to an electrometer (WPI, KS-700) via a salt bridge. The perfusion pipette is fitted with a fluid exchange pipette, permitting multiple reloading with different solutions during one experiment and exposure of the apical membrane to more than two solutions. Fluid changes in the lumen are accomplished by switching a hydrostatic pressure head from one channel to anther via a pneumatically activated three-way valve (Altex).
Solutions bathing the luminal and basolateral aspects of the tubules were identical under control conditions (in mM: 150 Na, 4 K, 141 Cl, 1.5 Ca, 1 Mg, 1.5 lactate, 5 acetate, 5 glucose, 5 PIPES). Luminal perfusates containing amiloride (10−5 M) or barium (1 mM) were prepared from this control solution.
Cloning of rabbit ENaC cDNA fragments (α-, β-, and γ-).
Total RNA was isolated from 1-g samples of kidney by using TRIzol reagent (GIBCO-BRL). Poly-A RNA was isolated subsequently from total RNA by using oligo(dT)-cellulose (Boehringer Mannheim). One microgram polyA-RNA was primed with oligo(dT), and first-strand cDNA was synthesized by using Superscript II reverse transcriptase (GIBCO-BRL) in a final volume of 20 μl.
Degenerate primers for each subunit were selected from regions of amino acid identity between rat (CAA49905, CAA54904, CAA54905) and human (CAA60632, CAA60633) or bovine (AAB48988) sequences. The primer sequences were GG(TCAG)ATGATGTATTGGCA(AG)TT (sense) and TT(AG) TA(AG)TA(AG)CA(AG)TA(AGCT)CCCCA (antisense) for α-ENaC; AT(TCA)TT(TC)AA(TC)TGGGG(TCAG)ATGAC (sense) and GTCAT(TC)TT(AG)TA(TC)TG(ATCG)GT(AG)TC (antisense) for β-ENaC; and CA(TC)TA(TC)ATGAA(TC)AT(TCA)ATGGC (sense) and TACATCCA(AG)TT(ACTG)GG(AG)TG(TC)TG (antisense) for γ-ENaC. AmpliTAQ DNA polymerase (PerkinElmer) was used to amplify cDNA fragments from 1 μl cDNA by PCR. All samples were heated to 94°C for 3 min. Forty cycles of PCR were performed as follows: denature at 94°C for 1 min.; anneal at 50–55°C for 1 min.; extend at 72°C for 1 min. Cycling (PerkinElmer DNA Thermal Cycler 480 or Hybaid Omigene) was performed by using block control of temperature and 0.5-ml thin wall microfuge tubes.
cDNA fragments from PCR reactions were separated electrophoretically on a 1% agarose gel (NuSieve), and the band of expected size (α-ENaC, 1,080 bp; β-ENaC, 547 bp; γ-ENaC, 601 bp) was cut out. The fragment was simultaneously blunt-ended with T4-DNA polymerase, and a 5′ phosphate group was attached by using T4-polynucleotide kinase, then purified using glass beads (Gene-clean, GIBCO-BRL). The purified cDNA was ligated into pBluescript (Stratagene) previously cut at theEco RV site, and 5′ phosphate groups were removed by using calf intestinal phosphatase. Competent Escherichia colicells (DH5α) were subsequently transformed with the vector, and white colonies were selected and grown in ampicillin-containing Luria-Bertani medium for extraction of the plasmid (QIAGEN maxi-prep). The insert was cut out of the plasmid by using EcoRI and HindIII and analyzed on an agarose gel to confirm the size of the cloned cDNA fragment.
Clones were submitted for sequencing in both directions by the Keck Foundation at Yale University.
Cloning of rabbit Nedd4 cDNA fragment.
RNA preparation from rabbit kidney and cDNA synthesis were carried out as described above in the cloning of ENaC cDNA fragments. Exact-match primers were selected from regions of nucleotide identity between mouse (D85414) and rat (U50842) sequences. The sense primer was GATGAAAATCGTTTGACAAGAGATGATTTC; the antisense primer was GAAGTTTCATTTCAAATTTGTTTGG. PCR was performed as above except that XL DNA polymerase (PerkinElmer) was used, and extension was at 68°C for 2 min.
Northern blot of kidney cortex, outer medulla, and inner medulla.
Cortex, outer medulla, and inner medulla were dissected from 1- to 2-mm transverse slices of rabbit kidney, and polyA-RNA was extracted as described above. Ten micrograms polyA-RNA were loaded per lane and transferred to a nylon membrane. Blots were prehybridized for at least 4 h in Church Gilbert [0.5 M sodium phosphate (pH 7.2), 1 mM EDTA, 7% SDS, 1% bovine serum albumin, 100 μg/ml salmon sperm DNA] at 68°C and then hybridized (106 cpm/ml, where cpm is counts/min) overnight by using random prime-labeled probes. Blots were rinsed once and then washed for 30 min in 2× standard sodium citrate (SSC; composition: 0.15M NaCl, 0.15 sodium citrate, pH 7) and 0.5% SDS at room temperature. Final wash at 68°C was done for 30 min in 0.5× SSC and 0.1 SDS, with periodic monitoring by a Geiger counter for low background counts. Blots were exposed to film from several hours up to several days.
Three blots were prepared and probed separately with α-, β-, and γ-ENaC probes, followed by a second hybridization of each blot with a glyceraldehyde 3-phosphate dehydrogenase (GAPDH) probe to control for loading.
Single-nephron RT-PCR analysis.
For single-nephron RT/PCR experiments, a total of 4 mm of tubule (2–12 nephron fragments; see Dissection of nephron segments) was transferred directly into a 0.5-ml microfuge tube containing 10 μl of lysis solution (2% Triton X-100, 5 mM dithiothreitol, 2.2 U/μl RNAsin, diethylpyrocarbonate-treated water). Tubule fragments were transferred by adsorbing to glass beads (0.5 mm diameter) held by forceps. All samples were frozen at −70°C until the time of the experiment.
The single-nephron RT-PCR technique (17) was modified from the methods described by us previously (9, 32). Briefly, total RNA was first extracted from individual nephron segment samples (RNAqueous, Ambion) after addition of 20 μg E. coliribosomal RNA (Boehringer). All RNA was precipitated with sodium acetate in 70% ethanol. The pellet was resuspended in 10 μl of diethylpyrocarbonate-treated water, and the entire amount was used as a template for cDNA synthesis (Superscript II, GIBCO) using oligo(dT) primers in a final volume of 20 μl. In each experiment, up to 10 nephron segments were tested. In addition, two control tubes were added: a reagent blank consisted of 10 μl of lysis solution, but no kidney tissue, and all other kit reagents; an RT-PCR control to determine the quality of the kit reagents consisted of 50 ng of rabbit kidney total RNA. A volume of 2–4 μl of cDNA was used for each subsequent PCR reaction.
In each experiment, all tubes were assayed for multiple genes. One set of cDNA aliquots was used to determine whether any cDNA was synthesized (positive control): primers were used to amplify a segment of the gene responsible for the oculocerebral-renal syndrome of Lowe (OCRL) (9). This gene product is expressed in all nephron segments examined from the glomerulus to the collecting duct (9) and serves as a positive control for cDNA synthesis during the RT step.
The rabbit-specific primer pairs selected from the cloned cDNA fragments (see above) were sense, GACCTGGACAGCATCACCCAGCAGACG, and antisense, GCAGCGGGATGAAGTCATTCTGCTCTG, for α-ENaC; sense, GGAACCGAATTTGGCCTGAAGCTGAT, and antisense, GACTCCTTGCACATGCGGATGCAC, for β-ENaC; sense, CTATTCCGCCGAGGAGCTGCTG, and antisense, GCTGGTAGTTGCAGTAGTTGGCTGCG, for γ-ENaC; and sense, ATATCCTGGGAAGGACCTACTATGTAAACC, and antisense, TGGTTTTGGTGTTGTGGTCAATAAAGAACG, for for Nedd4.
In separate pilot experiments for each ENaC or Nedd4 primer pair, tubes containing tubules or 50 ng total RNA were used to test for PCR amplification from the genomic template by omitting the reverse transcription step and performing PCR.
Samples were electrophoresed on a 1% gel, photographed to record ethidium bromide staining, and transferred to a nylon membrane (Gene Screen Plus, DuPont). Nylon membranes were Southern blotted by using an internal 32P-labeled RNA probe (Maxiscript, Ambion). Membranes were hybridized in Church Gilbert solution as described above for Northern blots and were exposed to film from several hours up to several days.
Primary culture of DCT explants.
DCT cells were grown on a transparent collagen-coated, permeable membrane mounted on a plastic cylinder (Transwell-col, Costar, 6-mm-diameter cell culture insert). This cell culture insert was placed in a 24-well tissue culture plate with 1 ml of culture medium (RPMI-1640, 5% fetal bovine serum, 2 g/l NaHCO3, 10 mM HEPES buffer, 100 U penicillin g/ml, 0.1 mg/ml streptomycin, 5 μg/ml transferrin, 5 μg/ml insulin, 50 nM dexamethasone, 5 pM triiodothyronine) in the bath, and 150 μl of medium inside the insert. The plates were preequilibrated at 37°C with 5% CO2.
Dissected DCT segments were temporarily transferred to a fresh dissecting dish on ice containing culture medium. About five to eight DCT segments were deposited onto each cell culture insert by using a micropipettor. Plates were left undisturbed for 2–3 days, then observed daily for tubule attachment and cellular outgrowth. Cells were fed about three times per week. On reaching visual confluence (10–15 days), V te and resistance (R te) were determined (EVOM, WPI). An insert without cells was used as a blank to correct for baselineV te and R te. Values forV te (mV) are apical vs. bath;R te is in Ω/cm2. When stable values for V te and R tewere attained, samples of apical and basal medium were taken 24 h after feeding for analysis of Na and K concentrations by flame photometry (IL 443, Instrumentation Laboratory).
RT-PCR experiments to determine ENaC and Nedd4 mRNA expression were performed on confluent DCT cells. RNA was extracted as in the single-nephron experiments described above by using the RNAqueous kit by Ambion. First, culture medium was removed and lysis-binding solution was added directly onto the cells growing on the insert membrane. After pipetting up and down several times, this lysate was transferred to a 0.5-ml microfuge tube, and samples were subsequently processed as described above for tubules.
In vitro perfusion of DCT subsegments.
The portion of DCT extending from the junction with the cTAL to the junction with the CNT was dissected, and all cTAL and CNT were cut away by using visual criteria (see methods). The DCT was cut into upstream and downstream subsegments. Figure1 shows that theV te of all upstream DCT segments was zero when they were perfused with control solution. Addition of either amiloride or barium to the luminal aspect of the upstream DCT had no effect on the V te. When the downstream DCT was studied, some tubules exhibited a lumen negative V te. In each case, a lumen negative V te was abolished with luminal amiloride or increased with application of luminal barium. The effects of both agents were rapid and readily reversible.
Cloning of rabbit ENaC subunits permitted selection of primers for the single-nephron RT-PCR experiments described below. All primers used in the present study are listed in methods.
Table 1 shows the percent similarity of the cloned rabbit ENaC subunit amino acid sequence to rat and human forms. The alignment was performed by using MegAlign (DNA Star, Clustal method, gap penalty of 10 for multiple alignments, and PAM250 residue weight table).
Cloning of rabbit Nedd4 permitted selection of primers for the single-nephron RT-PCR experiments described below. All primers used in the present study are listed in methods. The percent similarities of rabbit Nedd4 cDNA fragment to human, rat, and mouse sequence are 88.8, 78.4, and 80.6, respectively. The accession number for the rabbit Nedd4 sequence is AF229024.
Northen blot of ENaC in cortex, outer medulla, and inner medulla.
Figure 2 shows three blots of rabbit kidney cortex, outer medulla, and inner medulla polyA-RNA. Theleft side of the top panel shows one blot probed with α-ENaC. A single prominent band of ∼3.7 kb was detected in cortex, outer medulla, and inner medulla. The amount of message was similar in each of the three regions. The same blot was washed and subsequently probed with GAPDH. A band of ∼1.6 kb was detected for GAPDH and was of similar intensity in cortex, outer medulla, and inner medulla, suggesting equal loading of the lanes. Faint higher bands in GAPDH panels represent residual ENaC signal that did not wash off. Themiddle panel shows that β-ENaC is expressed in cortex and outer medulla but not in inner medulla. The single band detected runs at ∼3.1 kb and was more intense in cortex than in outer medulla. Probing with GAPDH confirmed even loading of the lanes. Thebottom panel shows that γ-ENaC is expressed in cortex and outer medulla but not inner medulla. Two bands were detected at ∼4.0 and 2.8 kb. The larger transcript was more prominent than the lower transcript, and the cortex expresses both at a higher level than outer medulla. Probing with GAPDH confirmed even loading of the lanes with polyA-RNA.
Figure 3 shows an ethidium bromide-stained gel from a single experiment employing primers for the OCRL gene. Raw data for the IMCD are not shown on this figure because of physical constraints of the gel box. The IMCD was tested in separate experiments and was also positive. A band of the expected size is visible for each of the nephron segments tested (including the IMCD; data not shown). This constitutes the positive control for cDNA synthesis (see methods) in each sample tube.
Expression of α-, β-, and γ-ENaC was tested subsequently in the same cDNA samples. The top panel of Fig. 4 presents one ethidium bromide-stained gel and the Southern blot from an experiment in which expression of α-ENaC mRNA was tested. The reagent control was negative, and the total RNA sample was positive. Consistent α-ENaC expression was detected in the mTAL, cTAL, DCT, CNT, CCD, OMCD, and also in the IMCD (IMCD data not shown in gel; see Table2).
Figure 4 also presents (middlepanel) one ethidium bromide-stained gel and the Southern blot from an experiment in which expression of β-ENaC was tested. The reagent control was negative, and the total RNA sample was positive. β-ENaC expression was detected consistently in the CNT, CCD, OMCD, and glomerulus. No consistent signal for β-ENaC could be detected in the DCT.
The bottom panel of Fig. 4 presents one ethidium bromide-stained gel and the Southern blot from an experiment in which expression of γ-ENaC was tested. The reagent control was negative, and the total RNA sample was positive. γ-ENaC expression was detected consistently in the CNT, CCD, and OMCD. No consistent signal for γ-ENaC could be detected in the DCT. One of four samples tested positive in the glomerulus, and one of five tested positive in the IMCD (IMCD data not shown in gel; see Table 2).
Table 2 summarizes the results from four series of experiments. The DCT appears to express α-ENaC but little if any β- or γ-ENaC whereas the CNT, CCD, and OMCD consistently express all three subunits.
Data for Nedd4 expression along the nephron are shown in Fig.5 and Table3. The reagent control was negative (lane 2), and the total RNA sample was positive (not shown on this gel). The ethidium bromide-stained gel also depicts visible bands for the CNT, CCD, glomerulus, and the IMCD. The Southern blot reveals additional bands for the P3, cTAL, and OMCD. No positive signal was detected for the DCT.
The summarized data in Table 3 show that a consistent signal was observed in the CNT, CCD, OMCD, IMCD, and glomerulus. Some but not all samples from the proximal tubule and ascending limb tested positive, whereas no Nedd4 signal was detected in the DCT.
Primary DCT culture.
“Upstream” DCT explants were grown until visually confluent, and subsequently the V te andR te were monitored daily until stable values were recorded for 2–3 consecutive days (maximum total culture time of explants is 21 days). Lumen negative V te andR te of confluent DCT monolayers were −13.7 ± 2.28 mV (n = 33) and 932 ± 86.3 Ω/cm2 (n = 33), respectively. Luminal sodium concentration was lowered (mean decrease 27.2 ± 2.66 mM,n = 33), reflecting net sodium absorption, and luminal potassium concentration was increased (mean increase 8.2 ± 1.22 mM, n = 33), reflecting net potassium secretion over a 24 h period.
To determine expression of ENaC mRNA in cultured DCT segments, monolayers were harvested, and RNA was extracted for RT-PCR studies. Figure 6 shows ethidium bromide-stained gels from experiments testing OCRL, and α-, β-, and γ-ENaC expression. OCRL was detected in each of the four samples, confirming that cDNA was synthesized. Visible bands were detected for each ENaC subunit in each monolayer sample.
Figure 7 shows an ethidium bromide-stained gel from an RT-PCR experiment designed to test for expression of Nedd4 from the same cultured monolayers. Each DCT monolayer expressed Nedd4.
The final processing of glomerular filtrate by distal nephron segments is critical in the regulation of renal sodium excretion. More than 85% of sodium filtered is absorbed before tubule fluid emerges from the loop of Henle and enters the distal tubule. Regulated absorption of sodium by the DCT, CNT, CCD, OMCD, and IMCD is responsible for providing overall sodium homeostasis by the kidney and results in the excretion of only ∼1% of the filtered amount. The DCT segment of the mammalian kidney plays an important role in this process; however, the precise mechanisms contributing to sodium absorption by the DCT have not been elucidated fully. In this study we sought to investigate whether amiloride-sensitive epithelial sodium channels contribute to sodium transport by the DCT segment of the rabbit kidney.
The present data suggest that, in the native DCT of the rabbit, sodium absorption does not likely occur via epithelial sodium channels. In support of this finding, when DCT are perfused in vitro, the voltage is zero and not affected by amiloride. Also, mRNA for the β- and γ-ENaC subunits was not detectable in the native DCT segment. In addition, Nedd4, which has been associated with ENaC trafficking from the apical membrane, does not appear to be expressed in the native DCT. In contrast to native tissue, DCT segments grown in primary culture exhibited a different phenotype. Thus monolayers generated from cultured DCT explants developed a lumen negative voltage, expressed all three ENaC subunits, and in addition expressed Nedd4. Taken together, these findings are consistent with amiloride-sensitive sodium transport occurring primarily in the connecting tubule and collecting duct system but not in the rabbit DCT. Furthermore, it appears that under the present culture conditions, DCT cells dedifferentiate and do not provide a suitable model for the DCT.
In vitro perfusion of DCT segments.
The rabbit was chosen for performance of these studies because the transitions from TAL to DCT and DCT to CNT occur abruptly, and the DCT can be dissected readily (see methods). In the rat kidney, a second type of DCT cell has been identified that appears at mid-DCT (18), which would decrease the likelihood of obtaining a homogeneous segment for this type of perfusion study.
In a previous preliminary study (31), we had observed that the voltage of perfused rabbit DCT segments was lumen negative. However, although most of the values were low or near zero, there was a large degree of scatter in the data. Also, because the DCT segment of the nephron is short (∼0.5 mm in length), tubules were cut as close to the transitions as possible to maximize segment lengths for perfusion. In the present study we sought to reexamine theV te of the true DCT.
Upstream DCT segments (see methods) uniformly had voltages near zero (control; C in Fig. 1) and lacked response to luminal amiloride and barium. In contrast, some downstream DCT segments exhibited a barium- and amiloride-sensitive lumen negative voltage. It is possible that, when the nephron was cut at the DCT-CNT junction, in some instances CNT cells were included in the perfused structure. It may be that these cells account for the barium- and amiloride-sensitive lumen negative voltage noted in some of the downstream DCT segments. The effects can be accounted for by the presence of apical sodium channels and potassium channels in some of the cells being perfused. We conclude that rigorous criteria must be applied during dissection to separate DCT and CNT cells. Furthermore, we believe that some CNT cells may have been included in a number of the DCT segments perfused in our previous study (31). The present data suggest that the native rabbit DCT likely does not express significant amiloride-sensitive sodium channel activity. In all of the subsequent experiments performed in the present study, when the DCT was investigated, the upstream DCT was used exclusively.
The present results also provide a basis for the interpretation of previous V te data obtained by using the isolated perfused tubule technique. It is possible that, in those studies (15, 24, 25, 31), inadvertent inclusion of CNT cells in the perfused structure could have contributed to the lumen negativeV te that was observed.
ENaC Northern analysis.
Northern analysis of rabbit kidney revealed expression of all three ENaC subunits. α- and β-ENaC were expressed as single prominent bands whereas two transcripts for γ-ENaC were detected. Although the transcript size for α-ENaC is similar to that of the rat (3.7 kb), transcript sizes for β- and γ-subunits differ somewhat from the results reported in the rat (2). In rabbit, the β-subunit is larger (3.1 vs. 2.2 kb), and two γ-subunit transcripts (2.8 and 4 kb) are expressed instead of one (3.2 kb).
The distribution and level of expression of ENaC subunits vary within the kidney. In the rabbit kidney cortex, robust expression of all three ENaC subunits was found. In the outer medulla, similar to the cortex, all three subunits were also expressed. However, in contrast to the cortex, the level of expression of β- and γ-subunits was relatively lower in the outer medulla. Furthermore, the larger γ-ENaC transcript is preferentially expressed in the outer medulla, and only traces of the lower transcript were detected. In the inner medulla, in contrast to the cortex and outer medulla, only α-ENaC was expressed. The level of transcript was similar to that observed in cortex and outer medulla. β- and γ-ENaC mRNA was not detected in the inner medulla.
Thus it appears that in the adult rabbit kidney, a functional sodium channel composed of all three subunits may not be expressed in the inner medulla. These data are also in agreement with the results obtained in the single-nephron RT-PCR experiments described below.
Our data in the rabbit inner medulla differ qualitatively from previous data obtained in rats. In one study (20), expression of all three ENaC subunits was detected in the inner medulla by RT-PCR whereas in another study (7) none of the subunits was detected, either by in situ hybridization or by immunocytochemistry. A third study (35) found that, during the first month of life, expression of all three subunits decreased gradually, with the β- and γ-subunits being barely detectable in the inner medulla at the end of this time period. Species differences and dissection or experimental techniques may have contributed to some of these apparently divergent results. However, in the present study, the results from Northern analysis and the single-nephron RT-PCR studies in the rabbit are internally consistent, suggesting at least consistent dissection methods.
ENaC and Nedd4 mRNA expression by tubule segments.
ENaC was expressed predominantly by distal nephron segments; however, the distribution was not uniform for the individual ENaC subunits. α-ENaC was detected in all nephron segments downstream of the thin ascending limb of Henle's loop. The only segments where α-ENaC could not be detected were the proximal tubule and the glomerulus. β- and γ-ENaC were expressed consistently only in the CNT, CCD, and OMCD. On the basis of previous studies both in vivo and in vitro (4, 5, 8,11, 19, 29), amiloride-sensitive sodium transport and lumen negative voltages are found in these segments of the nephron. Thus the present data are consistent with a fully functional sodium channel present in these nephron segments and confirm that all three subunits are expressed.
It is not clear from these studies why the thick ascending limb cells, the DCT, and the IMCD preferentially express only the α-subunit and the glomerulus expresses preferentially only the β-subunit. In the mouse, the thick ascending limb and the IMCD also have been found to express α-ENaC, although β- and γ-ENaC were not tested (3). The single-nephron data in the present study (Fig. 3) concur with the data from Northern analysis in Fig. 2, which suggest that the IMCD expresses α- but not β- or γ-ENaC. Further studies are needed to address the role of single expressed ENaC subunits in some segments.
Native rabbit DCT segments consistently expressed α-ENaC. This is in sharp contrast to the results of experiments testing for the presence of β- or γ-ENaC subunits in the DCT. These two subunits were not readily detected. They were detected only once in the DCT as a very faint band on an overexposed Southern blot (Table 2). Although it is possible that α-ENaC alone can form sodium-transporting channels, it is unlikely that the activity of this channel is very high because it requires at least one of the other two subunits to function effectively (2). Expressed amiloride-sensitive currents in oocytes that are formed from all three subunits are two to three orders of magnitude greater than currents expressed when α-ENaC alone is expressed. The lack of expression of all three subunits in the DCT of the rabbit, however, is in agreement with the results from the in vitro perfusion experiments discussed above. This nephron segment, when isolated and perfused in vitro, does not exhibit anyV te or amiloride effects (Fig. 1). Although the data do not provide an explanation as to why α-ENaC is expressed in the DCT, they lend support to the view that all three subunits are required for a functional sodium channel to be expressed. These experiments, therefore, provide additional support for the body of evidence obtained primarily in the rat in vivo that suggests the DCT is not the primary site for the absorption of sodium via sodium channels.
Further support for low or absent β-ENaC expression in the DCT comes from a recent study of the developing rat kidney (23). Schmitt and co-workers (13) showed that the onset of ENaC expression occurred in the DCT2 portion of the distal tubule, together with the onset of expression of sodium-calcium exchange. In another recent study in the rabbit, Loffing and co-workers (13) showed that expression of the γ-ENaC subunit begins at the transition from DCT to CNT and is not expressed by the DCT. These data are entirely consistent with our present findings. However, in another study, Duc and co-workers (7) concluded that the DCT of the rat kidney expressed α-, β-, and γ-ENaC at both the mRNA level and at the protein level. Our present results are at odds with this second study, albeit that our data were obtained in the rabbit. It is nevertheless possible that some of the profiles identified in the rat kidney sections as DCT segments were not exclusively DCTs because additional specific markers to colocalize the DCT were not used (7).
Nedd4 expression was detected consistently in nephron segments downstream of the DCT, although message was also detected in segments upstream of the DCT (Table 3). The DCT was the only segment in which Nedd4 message was not detected (0 of 4 samples). Results from a previous study (28) in which Nedd4 was immunolocalized in kidney agree, generally, with our present results. Faint variable staining was observed throughout the proximal portions of the nephron and the inner medullary collecting ducts, whereas intense staining was detected along the CCD and OMCD. The DCT did not stain for Nedd4. However, Staub et al. (28) did not detect staining in the TAL or in the glomeruli whereas in the rabbit we did detect its presence by RT-PCR in these segments.
Taken together, the results in intact rabbit kidney tubules are in agreement with previous physiological data obtained primarily in the rat in vivo. Little if any amiloride-sensitive sodium transport likely occurs in the rabbit DCT.
DCT grown in culture.
Our results in the DCT grown in vitro under cell culture conditions contrast starkly with the results obtained in the native tubules. Primary cultures of DCT fragments express α-, β-, and γ-ENaC mRNA as well as Nedd4 mRNA (Figs. 4 and 7). In addition, the monolayers exhibit a lumen negative voltage, absorb sodium, and secrete potassium. All of these features are typical of more distal nephron segments, including the connecting tubule and CCD.
The present in vitro culture conditions do not support the expression of the features typical of DCT cells. In a previous study using similar solutions and culture methods, we were unable to detect a significant transepithelial chloride gradient, thiazide effects on the sodium, or chloride gradients in cultured DCT or CNT fragments (30). It appears that our present culture conditions primarily support expression of the amiloride-sensitive ENaC by DCT cells. DCT cells in primary culture have been studied previously by a number of groups (14, 21, 30). It is remarkable that each of these studies detected expression of an amiloride-sensitive sodium channel when DCT cells were isolated from mice or rabbits and studied in vitro.
Studies using a cell line (MDCT cells) derived from immunodissected mouse DCT+TAL cells (21) and having features of DCT, have, in addition, demonstrated the presence of amiloride-sensitive uptake (10). Unfortunately, this sodium uptake cannot be localized to the apical or the basolateral membrane, because MDCT cells do not form tight junctions and do not develop aV te or R te. These studies, as well as the studies cited above, demonstrate that an amiloride-sensitive sodium channel is a feature of the DCT in culture; however, taken together with the results of the present study, they also suggest this may be an artifact of the experimental method.
Further studies are required to assess the effect of different substrates and hormones in cell culture media on expression of β- and γ-ENaC and the thiazide-sensitive sodium-chloride cotransporter by the DCT in culture. Clearly, removal of the DCT from its native environment and growth in vitro cause a transition of the phenotype from DCT-like to CCD-like: β- and γ-ENaC and Nedd4 expression are upregulated, and thiazide-sensitive sodium-chloride cotransporter expression is downregulated.
In summary, a reexamination of the voltage in the rabbit DCT perfused in vitro shows that the V te is zero and not lumen negative. The lumen negative voltage reported previously may have been caused by inadvertent inclusion of CNT cells in the perfused structure. We also found that the DCT expressed message only for α- and not β- or γ-ENaC and that Nedd4 was not expressed. This is in contrast to the CNT or CCD, which expressed all three ENaC subunits and also Nedd4. In vitro culture of the DCT dramatically changes the phenotype from that observed in native tissue and does not provide a suitable model for study of the DCT under present culture conditions. Together, these data suggest that an active sodium channel may not be operating in the rabbit DCT, and we conclude that net sodium absorption by the mammalian DCT is virtually exclusively electroneutral.
↵1 The distal tubule is defined as that portion of the nephron extending from the macula densa to the confluence of the nephron with another nephron to form the collecting duct. The distal tubule is a cytologically heterogeneous nephron segment and contains a portion of the thick ascending limb (TAL), the distal convoluted tubule (DCT), the connecting tubule (CNT) and the initial collecting tubule (ICT).
↵2 The term “late” refers to the final 50% of the distal tubule. The late distal tubule is composed primarily of ICT and a portion of the CNT.
↵3 The term “early” refers to the initial 50% of the distal tubule. The early distal tubule is composed primarily of DCT and a portion of the CNT.
↵4 We have also used a method of collagenase digestion (22) to aid in dissection of rabbit kidney nephron segments. Results from experiments using this method do not differ from results using hand-dissected segments. Collagenase treatment can only be utilized for molecular experiments and not for in vitro microperfusion studies.
Address for reprint requests and other correspondence: H. Velázquez, VA Connecticut Healthcare System, Research Office 151, West Haven, CT 06516 (E-mail:).
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 © 2001 the American Physiological Society