Pendrin (encoded by Pds, Slc26a4) is a Cl−/HCO3− exchanger expressed in the apical regions of type B and non-A, non-B intercalated cells of kidney and mediates renal Cl− absorption, particularly when upregulated. Aldosterone increases blood pressure by increasing absorption of both Na+ and Cl− through increased protein abundance and function of Na+ transporters, such as the epithelial Na+ channel (ENaC) and the Na+-Cl− cotransporter (NCC), as well as Cl− transporters, such as pendrin. Because aldosterone analogs do not increase blood pressure in Slc26a4−/− mice, we asked whether Na+ excretion and Na+ transporter protein abundance are altered in kidneys from these mutant mice. Thus wild-type and Slc26a4-null mice were given a NaCl-replete, a NaCl-restricted, or NaCl-replete diet and aldosterone or aldosterone analogs. Abundance of the major renal Na+ transporters was examined with immunoblots and immunohistochemistry. Slc26a4-null mice showed an impaired ability to conserve Na+ during dietary NaCl restriction. Under treatment conditions in which circulating aldosterone is increased, α-, β-, and 85-kDa γ-ENaC subunit protein abundances were reduced 15–35%, whereas abundance of the 70-kDa fragment of γ-ENaC was reduced ∼70% in Slc26a4-null relative to wild-type mice. Moreover, ENaC-dependent changes in transepithelial voltage were much lower in cortical collecting ducts from Slc26a4-null than from wild-type mice. Thus, in kidney, ENaC protein abundance and function are modulated by pendrin or through a pendrin-dependent downstream event. The reduced ENaC protein abundance and function observed in Slc26a4-null mice contribute to their lower blood pressure and reduced ability to conserve Na+ during NaCl restriction.
- intercalated cell
- apical Cl−/HCO3− exchanger
- epithelial sodium channel
the kidney is a major participant in the long-term control of blood pressure (10). Previous studies have focused primarily on how the kidney regulates blood pressure by changing Na+ excretion. Vascular volume depletion activates the renin-angiotensin-aldosterone system, which increases renal Na+ absorption, thereby expanding vascular volume and normalizing blood pressure. The major mechanisms of Na+ absorption in renal epithelia are the proximal tubule Na+/H+ exchanger (NHE3), the kidney-specific Na+-K+-2Cl− cotransporter of the thick ascending limb (NKCC2/BSC1), the thiazide-sensitive Na+-Cl− cotransporter of the distal convoluted tubule (NCC/TSC), the Na+-K+-ATPase, and the epithelial Na+ channel (ENaC) (2). The ubiquitous Na+-K+-2Cl− cotransporter (NKCC1) modulates transepithelial transport of Na+ along the collecting duct, although its physiological significance is unknown (49). After either an inactivating mutation or pharmacological inhibition of NHE3 (36), BSC1/NKCC2 (Bartter syndrome) (29), NCC/TSC (Gitelman syndrome) (29), or the α-, β-, or γ-subunits of ENaC (pseudohypoaldosteronism type 1) (39), hypotension and apparent vascular volume contraction are observed. Conversely, hypertension is associated with activation mutations of Na+ transporters, such as the β- or γ-subunits of ENaC (Liddle syndrome) (29). Thus intake of Na+ is thought to be the ion responsible for the maintenance of vascular volume and the renal regulation of blood pressure.
However, there is considerable evidence that dietary Cl− intake raises blood pressure markedly in humans and in rodent models of salt-sensitive hypertension (19, 20). Uninephrectomized rats given a high-NaCl diet and the aldosterone analog deoxycorticosterone pivalate (DOCP) develop hypertension. However, if dietary Cl− is substituted with HCO3−, hypertension is not observed (20). Similarly, although a high dietary NaCl intake raises blood pressure in people with documented NaCl-sensitive hypertension, equivalent intake of Na+ given as Na3 citrate is without effect (19).
Pendrin, encoded by the Slc26a4 gene, is an Na+-independent, Cl−/HCO3− exchanger expressed in the apical regions of type B and non-A, non-B intercalated cells of kidney (47). During either NaCl restriction or after the administration of DOCP and a high-NaCl diet, we observed lower blood pressure and greater Cl− excretion in mouse models of Pendred syndrome (Slc26a4−/− mice) than in wild-type mice (43, 48). Most likely, pendrin-mediated renal Cl− absorption expands vascular volume, which increases blood pressure. However, pendrin may modulate blood pressure and apparent vascular volume indirectly by changing renal Na+ transporter protein abundance rather than through a direct effect on renal Cl− absorption. Thus we hypothesized that the lower blood pressure observed in Slc26a4-null mice may result from downregulation of one or more renal Na+ transporters. Conversely, genetic disruption of Slc26a4 may upregulate Na+ transporter protein abundance, thereby minimizing the hypotension observed in the absence of pendrin-mediated Cl− absorption. In particular, pendrin may modulate ENaC-mediated Na+ absorption, since these transporters colocalize to the same nephron segments, although they are expressed in different cell types (23, 47).
Thus the purpose of the present study was to determine whether kidneys from Slc26a4-null mice have an impaired ability to conserve Na+ and reduced expression and/or function of specific renal Na+ transporters.
Animals and animal conditioning.
The Institutional Animal Care and Use Committee at Emory University and the Atlanta Veterans Affairs Medical Center approved all animal treatment protocols. Slc26a4−/− mice (∼25 g) developed by Everett et al. (7) were bred in parallel with wild-type mice from the same strain (129S6/SvEv Tac; Taconic Farms, Germantown, NY). Age- and sex-matched, pair-fed Slc26a4−/− and Slc26a4+/+ mice underwent treatment protocols for 7 days unless otherwise specified. In series 1, mice were fed a NaCl-restricted diet (0.13 meq/day NaCl, no. 53881300; Zeigler Brothers, Gardeners, PA) prepared as a gel (43). In series 2, mice ate the same gelled diet as in series 1 but supplemented with NaCl (NaCl-replete diet, 0.8 meq/day NaCl), which gives a daily dietary NaCl intake similar to that provided by the standard rodent diet used in the Emory University Vivarium (LabDiet 5001, PMI Nutritional International, Richmond, IN; 0.40% sodium; 0.65% chloride). In series 3, mice received a single dose of 1.7 mg DOCP by intramuscular injection (Ceiba-Geigy Animal Health) and ate the NaCl-replete, gelled diet given in series 2 (0.8 meq NaCl/day). In series 4, mice received 250 μg·kg body wt−1·day−1 aldosterone or vehicle (0.9% saline with 2.4% DMSO) by continuous infusion through osmotic minipumps (Alzet, Palo Alto, CA) and ate the NaCl-replete gelled diet given in series 1 (0.8 meq/day NaCl). In series 5, mice ate the NaCl-restricted gelled diet of series 1 but supplemented with 1.1 meq/day NaCl or diet with furosemide (100 mg·kg body wt−1·day−1) added to the gel for 5 days. In series 1–5, mice did not receive water other than that present in the gel (∼9 ml H2O/day). In series 6, mice ate the NaCl-restricted diet of series 1 but given as pellets and drank water ad libitum for 6 days. On the 7th day, or the day before death, water was withheld overnight (18 h). In each of these treatment protocols, mice were killed under anesthesia with 1–2% isofluorane in 100% O2 at 1 l/min and kidneys were fixed in situ as described previously (47).
Rabbit polyclonal antibodies, which recognize the α-, β-, and γ-subunits of ENaC, the NHE3, the NCC/TSC, the kidney-specific Na+-K+-Cl− cotransporter type 2 (NKCC2) (2) (kindly provided by Drs. Mark A Knepper, Soren Nielsen, and Lawrence Palmer), and a mouse monoclonal antibody against Na+-K+-ATPase α1-subunit (Upstate Biotechnology, Lake Placid, NY) were employed to detect major renal Na+ transporters. The anti-NKCC1 antibody (α-wCT, a gift of R. James Turner) was raised in rabbits against a 6× His fusion protein corresponding to amino acids 750–1203 of the rat NKCC1 (25). Anti-aquaporin 2 (AQP2) (27)and anti-aquaporin 3 (AQP3) antibodies (LL763) (5) were gifts of Dr. Mark Knepper.
Arterial blood was collected through the abdominal aorta under anesthesia with 1–2% isofluorane in 100% O2. Serum aldosterone was measured by the Cardiovascular Pharmacology Research Laboratory at the University of Iowa with the use of a radioimmunoassay assay (49). Plasma renin concentration was measured as described previously (22). Serum total triiodothyronine, triiodothyronine uptake, and thyroxine were measured at the Boston Medical Center Clinical Chemistry laboratories with a chemiluminescence assay and the Bayer Advia Centaur automated system (Bayer Healthcare, Tarrytown, NY) (1). Urinary corticosterone concentration was measured by the laboratory of Dr. Robert Bonsall (Department of Psychiatry and Behavioral Sciences, Emory University School of Medicine) using a radioimmunoassay (48). Thyroid-stimulating hormone was measured by Anilytics (Gaithersburg, MD) using a radioimmunoassay and a mouse standard supplied by A. F. Parlow.
Immunohistochemistry and morphometry.
In situ fixation of mouse kidneys was performed as described previously (47). For paraffin embedding, tissues were dehydrated in a series of graded ethyl alcohol followed by xylene and embedded in paraffin and then cut into 2-μm sections. Sections were deparaffinized and rehydrated. Endogenous peroxidase was blocked by 0.5% H2O2 in absolute methanol for 30 min at room temperature. To reveal antigens, sections were incubated in 1 mM Tris solution (pH 9.0) supplemented with 0.5 mM EGTA and heated in a microwave oven for 10 min. Nonspecific binding of IgG was prevented by incubating the sections in 50 mM NH4Cl-PBS for 30 min, followed by blocking in PBS supplemented with 1% BSA, 0.05% saponin, and 0.2% gelatin. Sections were incubated overnight at 4°C with primary antibodies diluted in PBS supplemented with 0.1% BSA and 0.3% Triton X-100. After sections were rinsed with PBS supplemented with 0.1% BSA, 0.05% saponin, and 0.2% gelatin, labeling was visualized with horseradish peroxidase-conjugated secondary antibody (1:200, DAKO), followed by incubation with 3,3′-diaminobenzidine (brown color). Sections were washed with distilled water, dehydrated with graded ethanol and xylene, mounted in Eukitt, and examined by light microscopy.
Samples from the same kidneys were processed in Lowicryl K4M (Electron Microscopy Sciences, Ft. Washington, PA) for electron microscopy. Ultrathin sections of renal cortex containing cortical collecting duct (CCD) or initial collecting tubules (iCT) were stained with uranyl acetate and photographed at a primary magnification of ×5,000 with a transmission electron microscopy and recorded on SO63 film. Apical plasma membrane boundary length and cytoplasmic area of randomly selected principal cells from the CCD and iCT were measured by point and intersection counting using the Merz curvilinear test grid and standard morphometric formulas as previously described (47). A minimum of five principal cells per animal were included in the analyses for both collecting duct segments. CCD and iCT were identified as described previously (47).
Semiquantitative immunoblots1 of lysates from kidney, thyroid, and distal colon from Slc26a4−/− and Slc26a4+/+ mice were performed as reported previously (49). Tissue was homogenized in dissecting buffer (0.3 M sucrose, 25 mM imidazole, 1 mM EDTA, pH 7.2, containing 8.5 μM leupeptin and 1 mM PMSF). Protein samples were dissolved in Laemmli buffer and resolved by SDS-PAGE. Equal protein loading was confirmed by running a gel in parallel stained with Coomassie blue dye, as reported previously (42). Protein was electrophoretically transferred onto nitrocellulose membranes and probed with antibodies to the renal Na+ transporters. Immunolabeling was detected with horseradish peroxidase-conjugated goat anti-rabbit secondary antibody (Upstate Biotechnology, Lake Placid, NY) or with goat, anti-mouse IgG conjugated with horseradish peroxidase (Upstate Biotechnology) using an enhanced chemiluminescence system (Amersham Biosciences, Little Chalfont, UK). Band density was quantified with Quantity One Image software (Bio-Rad, Hercules, CA). Relative band densities between groups were compared. Band densities were quantified and normalized to the mean band density of lysates from mice prepared and run in parallel on the same gel from a single treatment condition.
Measurement of blood pressure by telemetry.
Under anesthesia with 1–2% isofluorane in 100% O2, the anterior neck of each mouse was shaved, disinfected with Betadine and 70% alcohol, and covered with sterile surgical drapes. Atropine (0.25 mg/kg) was administered to reduce airway secretions. The carotid artery was isolated, ligated, and retracted with sutures through a ventral midline incision. The transducer catheter was inserted into the carotid near the bifurcation and advanced until its tip is just inside the thoracic aorta. The catheter was secured with sutures after placement was ensured based on standard landmarks and real-time blood pressure waveforms. The transmitter battery was inserted into a subcutaneous pouch formed by blunt dissection along the right flank close and then closed. Blood pressure measurements occurred at least 7 days after surgery, when a stable baseline blood pressure and a stable diurnal variation in blood pressure were established. Blood pressure measurements were collected by the Dataquest A.R.T. system (Transoma Medical) and sterile Data Sciences PA-C10 blood pressure transducers. Mean arterial pressure measured by telemetry produced values similar to those reported previously (28).
Transepithelial voltage (VT) was measured in CCDs perfused in vitro from furosemide-treated mice (series 5) (34). The perfusion pipette was connected to a high-impedance electrometer through an agar bridge saturated with 0.16 M NaCl and a calomel cell, as described previously (46).
When two groups were compared, an unpaired Student's t-test was employed. When three or more groups were compared, ANOVA was used with a Holm-Sidak post hoc test. P < 0.05 indicates statistical significance. Data are means ± SE.
Slc26a4-null mice have a lower blood pressure and an impaired ability to conserve Na+ during NaCl restriction relative to wild-type mice.
Because Slc26a4-null mice have greater apparent vascular volume contraction than wild-type mice during NaCl restriction (48), we asked whether these mutant mice have an impaired renal ability to conserve Na+. Thus 24-h urinary Na+ excretion was measured after mice were switched from a NaCl-replete (series 2) to a NaCl-restricted (series 1) diet (Fig. 1). As shown, during the first day of NaCl restriction, Slc26a4-null mice excreted more Na+ than pair-fed wild-type mice. However, within 7 days of NaCl restriction, mice come into steady state, which results in similar rates of Na+ excretion in wild-type and mutant mice (48). We conclude that kidneys from Slc26a4-null mice have an impaired ability to conserve both Na+ and Cl− (44, 48) during dietary NaCl restriction.
We asked whether the impaired ability of Slc26a4-null mice to conserve urinary Na+ occurs in tandem with reduced blood pressure. Mean arterial pressure was slightly lower in Slc26a4-null than in wild-type mice after either a NaCl-replete or a NaCl-restricted diet (Table 1). 2 Thus Slc26a4-null mice have an impaired ability to conserve Na+ and Cl−, which contributes to the lower blood pressure observed in these animals.
Abundance of major renal Na+ transporters in Slc26a4+/+ and Slc26a4−/− mice.
Because kidneys from Slc26a4-null mice have an impaired ability to conserve Na+ (Fig. 1), we asked whether renal Na+ transporter protein abundance is reduced in these mutant mice. Following the NaCl-replete diet (series 2), wherein circulating aldosterone and plasma renin concentrations are appropriately low (Table 1), NHE3, NKCC1/2, and Na+-K+-ATPase α1-subunit transporter protein abundance is similar in wild-type and Slc26a4-null mice (Fig. 2). Abundances of the α- and β-subunits of ENaC were also similar in mutant and wild-type mice after the NaCl-replete diet, although abundance of the 85-kDa γ-fragment of γ-ENaC was 10–20% lower in mutant than in wild-type mice.
Following NaCl restriction (series 1; Fig. 3), a treatment model in which circulating aldosterone concentration and plasma renin concentration are appropriately increased (Table 1), abundance of each ENaC subunit is lower in mutant than in wild-type mice. Abundances of the α-, β-, and the 85-kDa fragment of γ-ENaC were 25–35% lower in Slc26a4-null than in wild-type mice (series 1, Fig. 3). As expected, a broad band detected by the anti-γ-ENaC antibody appeared at ∼70 kDa after NaCl restriction (23), the abundance of which was ∼65% lower in mutant mice relative to wild-type mice. Thus abundance of ENaC, particularly the γ-subunit, is reduced in kidneys from Slc26a4−/− mice after NaCl restriction. Abundance of the other Na+ transporters was similar in Slc26a4-null and wild-type mice. Mean NCC band density was 50% higher in mutant than in wild-type mice during NaCl restriction, although differences did not reach statistical significance.
Differences in thyroid, adrenal, and hypothalamic function do not explain differences in ENaC protein abundance observed in Slc26a4+/+ and Slc26a4−/− mice.
We asked whether ENaC protein expression is reduced in Slc26a4-null mice due to impaired release of a circulating hormone that regulates ENaC abundance, such as renin (via ANG II) (2), aldosterone (23, 39), thyroid hormone (33), corticosterone (37), or vasopressin (4). Thus endocrine function was evaluated after a low-NaCl diet (Table 1). As shown, circulating levels of these renal, adrenal, and thyroid hormones were not reduced in mutant relative to wild-type mice. To determine whether Slc26a4-null mice have a defect in either the release or the action of vasopressin, urinary concentrating ability was evaluated (series 6). After 18 h of water restriction, urinary osmolality was 2,771 ± 366 mosmol/kgH2O in mutant mice (n = 6) and 2,667 ± 200 mosmol/kgH2O in wild-type mice (n = 6, P = not significant). Thus urinary concentrating ability is similar in Slc26a4-null and wild-type mice, making it unlikely that these mutant mice have a significant defect in either the release or the action of vasopressin. Thus differences in circulating levels of thyroxine, vasopressin, aldosterone, corticosterone, or renin do not explain the low levels of ENaC protein expression observed in kidneys from Slc26a4-null mice.
The aldosterone-induced increment in ENaC abundance and posttranslational processing is blunted in Slc26a4-null mice.
Although differences did not reach statistical significance, we observed a trend toward a lower serum aldosterone level in Slc26a4-null than in wild-type mice with NaCl restriction. Therefore, we could not exclude the possibility that ENaC protein abundance is lower in Slc26a4-null than in wild-type mice from reduced circulating aldosterone concentration. Thus ENaC protein expression was measured in Slc26a4-null and wild-type mice after administration of aldosterone and aldosterone analogs (DOCP; Fig. 4A). After DOCP treatment, ENaC subunit abundance was reduced in mutant mice, similar to observations following low-NaCl intake (Fig. 3). Abundance of the other Na+ transporters was similar in DOCP-treated wild-type and Slc26a4-null mice, with the exception that NKCC2 protein expression was reduced 40% in Slc26a4−/− mice. However, after aldosterone administration (series 4; Fig. 4B), NKCC2 abundance was not reduced in kidneys from pendrin-null relative to wild-type mice.
We explored whether aldosterone-induced changes in ENaC subunit protein expression is impaired in kidneys from Slc26a4-null mice. In wild-type mice, 7 days of aldosterone increased serum aldosterone from 1.84 ± 0.11 (n = 6) to 17.3 ± 1.6 (n = 6) nM3 and increased α-ENaC protein abundance ∼2.5-fold (Fig. 5). However, in Slc26a4-null mice, α-subunit protein abundance rose ∼1.5 fold with aldosterone treatment. Thus the aldosterone-induced increment in α-ENaC protein expression appears slightly blunted in Slc26a4-null mice (Fig. 6).
We asked whether aldosterone-induced posttranslational modification of the γ-ENaC subunit is altered in Slc26a4-null mice, thereby reducing production of the “active” 70-kDa γ-ENaC fragment. With aldosterone treatment, abundance of the 70-kDa γ-ENaC fragment increased fivefold in wild-type mice but increased only ∼2.7-fold in Slc26a4-null mice. Moreover, the ratio of 70-kDa to total γ-ENaC (85 + 70 kDa band density) was 49.8 ± 1.9% (n = 9) in aldosterone-treated wild-type mice but only 41.6 ± 3.3% (n = 7) in aldosterone-treated pendrin knockout mice (P = 0.055). Similarly, during dietary NaCl restriction, in which circulating aldosterone concentration is appropriately increased, the 70-kDa fragment represents 26.7 ± 5.67% of total kidney γ-ENaC expressed in kidneys from wild-type mice but represents only 13.2 ± 2.6% of total γ-subunit expressed in kidneys from Slc26a4-null mice (P < 0.05). Thus aldosterone-induced posttranslational γ-ENaC processing is blunted in Slc26a4-null mice.
We asked whether the reduced ENaC expression observed in pendrin-null mice reflects a generalized reduction in principal cell protein abundance. To answer this question, we quantified principal cell size and the abundance of other principal cell, aldosterone-sensitive proteins in wild-type and Slc26a4−/− mice. As shown (Table 2 and Fig. 6), kidney weight, principal cell cross-sectional area, and boundary length were similar in aldosterone-treated wild-type and pendrin null mice. To determine whether kidneys from pendrin-null mice have a reduced abundance of aldosterone-sensitive proteins other than ENaC, we compared AQP2 and AQP3 (21, 38) abundances in kidneys from aldosterone-treated wild-type and pendrin-null mice. As shown (Fig. 6), AQP2 and AQP3 abundances are similar in kidneys from aldosterone-treated wild type and pendrin-null mice. We conclude that, although ENaC expression is reduced in pendrin-null mice, principal cell number (17) and size are unchanged. Moreover, the abundance of other principal cell aldosterone-sensitive proteins is also maintained in Slc26a4-null mice. Thus genetic disruption of Slc26a4 does not reduce the abundance of all principal cell aldosterone-sensitive proteins.
Amiloride-sensitive VT is low in CCDs from Slc26a4-null mice.
We asked whether ENaC function is altered in Slc26a4-null mice. To answer this question, CCDs were perfused in vitro from wild-type and Slc26a4-null mice given a high-NaCl diet and furosemide, since furosemide increases circulating plasma renin and serum aldosterone (Table 3), thereby increasing ENaC expression and function (15). ANG II (10−8 M) was applied to the bath solution to further increase ENaC-mediated transport (31). Changes in VT, observed after addition of the ENaC inhibitor benzamil hydrochloride (10−6 M) to the perfusate, were taken to reflect ENaC-mediated current.4 As shown, a lumen-negative VT was observed in CCDs from wild-type mice, which fell markedly with the application of benzamil (Fig. 7). In contrast, in CCDs from Slc26a4-null mice, VT was very low and not changed appreciably with benzamil application. We conclude that ENaC-dependent VT is much lower in CCDs from Slc26a4-null than from wild-type mice.
ENaC subunit targeting is maintained within principal cells of Slc26a4−/− and Slc26a4+/+ mice.
We asked whether the low ENaC-dependent VT observed in CCDs from pendrin-null mice results from aberrant apical plasma membrane ENaC trafficking in addition to reduced ENaC subunit protein abundance. Because ENaC is redistributed to the apical plasma membrane after application of aldosterone (23), ENaC immunolabeling was examined in kidneys from Slc26a4-null and wild-type mice following aldosterone treatment (Figs. 8 and 9). Following the NaCl-replete diet, where circulating aldosterone concentration is low, we observed diffuse ENaC labeling throughout the cytoplasm of principal cells in pendrin-null mice, consistent with ENaC localization to intracellular vesicles. After aldosterone administration, ENaC labeling was discrete along the apical aspect of principal cells in these mutant mice, whereas diffuse labeling was absent. These findings show that aldosterone induces redistribution of ENaC from cytoplasmic vesicles to the apical plasma membrane of principal cells in the mutant mice (Figs. 8 and 9), similar to observations in wild-type mice. Thus the subcellular distribution of ENaC responds appropriately to aldosterone administration in Slc26a4-null mice, at least qualitatively.
ENaC abundance in colon and thyroid is similar in wild-type and Slc26a4-null mice.
To determine whether pendrin alters ENaC protein expression through a systemic mechanism, the effect of genetic disruption of pendrin on aldosterone-induced changes in ENaC protein abundance was examined in colon and thyroid. In thyroid, a tissue that highly expresses pendrin (34), α-, β-, and γ-ENaC protein abundance was not reduced with genetic disruption of Slc26a4, either under basal conditions or after the application of aldosterone (Fig. 10). In colon, an aldosterone-sensitive tissue with low levels of pendrin expression (not shown), ENaC abundance was similar in wild-type and pendrin-null mice under both basal conditions and after the administration of aldosterone (Fig. 11). Thus pendrin modulates ENaC abundance in kidney but not in other ENaC-expressing tissues such as colon or thyroid.
The present study demonstrates that pendrin-knockout mice have lower blood pressure than wild-type mice, at least in part because of reduced ENaC expression and function, which reduce their ability to conserve Na+. Pendrin and ENaC localize to the aldosterone-sensitive region of the kidney, i.e., the distal convoluted tubule, the connecting tubule, and the CCD (11, 47). Within these segments, ENaC and pendrin protein abundances change in tandem in many treatment models. For example, ENaC and pendrin protein abundance increase with aldosterone and aldosterone analogs (23, 43), with dietary NaCl restriction (23, 48) and with Cl− restriction alone (8, 16, 32, 44). Moreover, ENaC-mediated Na+ current and apical anion exchange of the type B intercalated cell, which is likely the gene product of Slc26a4, increase with aldosterone and aldosterone analogs (39, 43, 48).
Within the kidney, we observed that ENaC protein expression and function are pendrin dependent. The reduced ENaC abundance observed in these pendrin-null mice likely contributes to their lower blood pressure2 (43). Abundance of all three ENaC subunits is lower in Slc26a4-null than in wild-type mice whether circulating aldosterone is stimulated or suppressed, although differences appear more marked when circulating aldosterone is increased. Most notable was the marked reduction in aldosterone-induced 70 kDa γ-ENaC protein abundance observed in kidneys from Slc26a4-null mice, which occurs from lower γ-ENaC total protein abundance and blunted posttranslational processing.
Low levels of ENaC-mediated Na+ absorption are observed after an NaCl-replete diet because the γ-ENaC subunit is in the immature or “uncleaved” form (39). With aldosterone administration or during dietary NaCl restriction, the γ-subunit undergoes both glycosylation and proteolytic cleavage, which result in a mature ENaC complex, thereby increasing ENaC-mediated Na+ absorption (39). Maturation of γ-ENaC involves furin-dependent cleavage at a single site within the γ-subunit extracellular loop (3). For full subunit activation, a second protease, prostasin, cleaves γ-ENaC distal to the furin cleavage site (3). Because γ-ENaC cleavage events occur near the NH2 terminus (6), whereas the γ-ENaC antibody employed recognizes an epitope near the COOH terminus (amino acids 629–850 of the rat γ-ENaC sequence) (23), both the unprocessed “immature” 85 kDa γ-subunit and the “mature” 70-kDa fragments of γ-ENaC are detected by immunoblot and immunohistochemistry.
The reduced abundance of the 70-kDa fragment of γ-ENaC observed in kidneys from Slc26a4-null mice demonstrates that posttranslational processing of γ-ENaC is pendrin dependent. Modulation of posttranslational processing by pendrin is potentially very physiologically significant because proteolysis of ENaC subunit extracellular domains modulates channel gating. In the absence of proteolytic processing, Na+ channels have markedly reduced activity and have enhanced inhibition by external Na+ (Na+ self-inhibition) (3).
α-ENaC subunit processing involves glycosylation and furin-dependent cleavage at two sites in the extracellular loop (3), which reduces protein size by ∼60 kDa (6, 13). However, the anti-rat α-ENaC antibody employed recognizes the 85-kDa peptide but detects the 30-kDa peptide poorly (6, 23).
Although α-ENaC protein abundance may be rate limiting for ENaC assembly (24, 40), efficient ENaC trafficking out of the endoplasmic reticulum and full ENaC-mediated current require assembly of all three ENaC subunits (α, β, and γ) (9, 14, 40). However, apical plasma membrane ENaC protein abundance increases appropriately in Slc26a4-null mice after aldosterone administration. Thus the reduced ENaC-dependent VT observed in these mutant mice reflects, at least in part, reduced total ENaC protein abundance and altered posttranslational modification rather than markedly aberrant membrane targeting.
How pendrin modulates ENaC protein abundance and function remains to be determined. Because pendrin and ENaC are expressed in different cell types, ENaC and pendrin cannot interact directly through protein-protein interaction. Moreover, intercalated and principal cells probably do not communicate through gap junctions, since intracellular pH differs in these two cell types and since the intercellular spread of Lucifer yellow is not observed under either basal or stimulated conditions (35). Instead, these transporters might cooperate through a paracrine effect or through changes in downstream luminal solute composition. One of these pendrin-dependent effects might modulate 70-kDa γ-ENaC abundance by 1) changing the fraction of γ-ENaC traversing the Golgi complex relative to ENaC bypassing the Golgi (12, 13), 2) modulating γ-ENaC cleavage at the cell surface through changes in protease activity or changes in cell surface residence time (18), or 3) changing the ratio of proteases to protease inhibitors (26, 41, 45).
Although we observed no difference in circulating levels of thyroid, adrenal, or renal hormone concentrations, pendrin might modulate the release of another circulating factor(s) that alters ENaC protein abundance. However, genetic disruption of pendrin did not produce a generalized reduction in ENaC protein abundance. In contrast to kidney, ENaC protein abundance in colon and thyroid is not altered with genetic disruption of Slc26a4. Thus it is unlikely that pendrin alters ENaC through changes in concentration of a circulating hormone.
Na+ transporter protein abundance in the proximal tubule did not change with genetic disruption of Slc26a4. However, protein abundance of the kidney-specific Na+-K+-2Cl− cotransporter NKCC2 was lower in Slc26a4-null than in wild-type mice after administration of DOCP but not after NaCl restriction or the administration of aldosterone. Why NKCC2 is reduced in kidneys from DOCP-treated Slc26a4-null mice requires further study.
We conclude that, within kidney, aldosterone-induced stimulation of ENaC subunit abundance and posttranslational processing of γ-ENaC are blunted in pendrin-knockout mice. Thus ENaC is modulated by pendrin-mediated transport or by a downstream event that follows changes in pendrin-mediated transport.
This study was supported by Project 2 of National Institutes of Health grant PO1 061521 (to S. M. Wall). V. Pech is the recipient of American Heart Association Postdoctoral Fellowship Award 0525384B.
↵1 α, β, and γ-ENaC band densities were nearly a linear function of protein loaded per lane. α-ENaC and γ-ENaC (70 and 85 kDa) band densities as a function of protein loading were slightly supralinear, whereas β-ENaC expression was slightly sublinear (supplemental material for this article is available online at the Am J Physiol Renal Physiol website).
↵2 In previous studies, our laboratory (43, 48) did not observe a difference in systolic blood pressure between Slc26a4-null and wild-type mice after either NaCl-replete (0.8 meq/day NaCl) or NaCl-restricted diet (0.13 meq/day NaCl) when measured by tail cuff. In the present study, Slc26a4-null mice were noted to have a slightly lower blood pressure relative to wild-type mice after either the NaCl-replete diet or after moderate NaCl restriction (0.13 meq/day NaCl). Differences in blood pressure results observed in the present and previous studies likely reflect the increased sensitivity of blood pressure measurements when detected by telemetry vs. tail cuff. Nevertheless, both the present and the previous studies show that differences in blood pressure between wild-type and pendrin-null mice are small after either NaCl-replete diet or after moderate NaCl restriction. However, larger differences in blood pressure between wild-type and pendrin-null mice are observed after treatment models under which circulating aldosterone increases further, such as after the NaCl-deficient diet (48), after NaCl-replete diet and furosemide (30), or after NaCl-replete diet and DOCP (43).
↵4 Based on Ohm's law, current equals voltage divided by resistance. Thus changes in voltage could reflect changes in current and/or changes in resistance. We observed an amiloride-sensitive VT of −3.9 ± 1.3 mV in wild-type mice (n = 4) but −0.54 ± 0.30 mV (n = 4) in pendrin-null mice (P < 0.05). Thus, if differences between wild-type and mutant mice in benzamil-sensitive VT result from differences in resistance, rather than differences in current, resistance must be 8-fold higher in CCDs from mutant relative to wild-type mice, which seems unlikely.
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 © 2007 the American Physiological Society