Pendrin (Slc26a4) localizes to type B and non-A, non-B intercalated cells in the distal convoluted tubule, the connecting tubule, and the cortical collecting duct (CCD), where it mediates apical Cl−/HCO3− exchange. The purpose of this study was to determine whether angiotensin II increases transepithelial net chloride transport, JCl, in mouse CCD through a pendrin-dependent mechanism. JCl and transepithelial voltage, VT, were measured in CCDs perfused in vitro from wild-type and Slc26a4 null mice ingesting a NaCl-replete diet or a NaCl-replete diet and furosemide. In CCDs from wild-type mice ingesting a NaCl-replete diet, VT and JCl were not different from zero either in the presence or absence of angiotensin II (10−8 M) in the bath. Thus further experiments employed mice given the high-NaCl diet and furosemide to upregulate renal pendrin expression. CCDs from furosemide-treated wild-type mice had a lumen-negative VT and absorbed Cl−. With angiotensin II in the bath, Cl− absorption doubled although VT did not become more lumen negative. In contrast, in CCDs from furosemide-treated Slc26a4 null mice, Cl− secretion and a VT of ∼0 were observed, neither of which changed with angiotensin II application. Inhibiting ENaC with benzamil abolished VT although JCl fell only ∼50%. Thus substantial Cl− absorption is observed in the absence of an electromotive force. Attenuating apical anion exchange with the peritubular application of the H+-ATPase inhibitor bafilomycin abolished benzamil-insensitive Cl− absorption. In conclusion, angiotensin II increases transcellular Cl− absorption in the CCD through a pendrin- and H+-ATPase-dependent process.
- transepithelial voltage
- knockout mice
- intercalated cell
within the cortical collecting duct (CCD), transepithelial transport of Cl− occurs through both paracellular and transcellular transport (29, 30). The driving force for paracellular transport-mediated Cl− absorption is the lumen-negative transepithelial voltage, VT, generated by ENaC-mediated Na+ absorption (29, 30). In rabbit CCD, only ∼20% of net Cl− absorption occurs through paracellular transport, while the majority of net Cl− absorption occurs through transepithelial transport (30, 33). However, the molecular mechanism responsible for this transepithelial transport is poorly understood. With the recent explosion in genetically engineered mice, our ability to examine the mechanism of transepithelial Cl− absorption in the CCD has been greatly expanded. However, because perfusing mouse tubules in vitro requires greater technical skill than is needed with rat or rabbit, few studies have utilized this technique to examine ion transport in mouse CCD.
Within the CCD, transepithelial transport of Cl− occurs primarily across intercalated rather than principal cells (27). Intercalated cells are classified based on whether they express the basolateral Cl−/HCO3− exchanger AE1 and the localization of the H+-ATPase within the cell (29, 41). Within mouse CCD, the predominant intercalated cell subtypes are type A and type B cells (34). Type A cells secrete H+ equivalents through the H+-ATPase, which localizes to the apical plasma membrane. Type B cells secrete OH− equivalents and absorb Cl− through an apical Cl−/HCO3− exchanger, which operates in tandem with net H+ and Cl− efflux across the basolateral membrane mediated by the H+-ATPase and a Cl− channel (29, 41). Pendrin localizes to the apical regions of type B and non-A, non-B intercalated cells and likely mediates the apical Na+-independent, Cl−/HCO3− exchange observed in these cell types (25, 26, 31, 38, 41).
Within the type B cell, pendrin and apical anion exchange are upregulated with aldosterone analogs, which enhance secretion of OH− equivalents and absorption of Cl− (26, 43). Thus pendrin-mediated transport attenuates the alkalosis observed following the administration of aldosterone analogs. Moreover, the apparent vascular volume expansion and the hypertension which follow treatment with desoxycorticosterone pivalate and a high-NaCl diet are critically dependent on pendrin expression, likely through its effect on renal Cl− absorption (38). Thus, while pendrin has an important role in acid-base balance, it also plays a novel role in the maintenance of blood pressure and fluid and electrolyte balance (26, 38).
Pendrin is also upregulated in treatment models not associated with increased circulating aldosterone (37). For example, during selective dietary Cl− restriction, pendrin expression increases, although circulating aldosterone does not change (39). However, because plasma renin concentration is elevated in this treatment model, pendrin expression might increase from enhanced angiotensin II production.
Angiotensin II plays a critical role in the regulation of systemic blood pressure as well as fluid and electrolyte balance (8). Besides acting as a powerful vasoconstrictor, angiotensin II regulates Na+, Cl−, and H2O excretion through direct effects on renal epithelia (21, 46). In the proximal tubule, angiotensin II upregulates apical Na+/H+ exchange and basolateral Na+-HCO3− cotransport, which act in series to increase absorption of Na+ and secretion of H+ equivalents (7, 17). In principal cells of the CCD, angiotensin II also increases Na+ absorption by stimulating the epithelial Na+ channel (ENaC) (23). The effect of angiotensin II on Cl− transport is poorly understood. Weiner et al. (44) showed that angiotensin II application to the bath increases luminal alkalization and increases apical Cl−/HCO3− exchange in rabbit outer CCD through an AT1 receptor-dependent process (44). Because apical anion exchange of the type B intercalated cell is most likely the gene product of Slc26a4, the gene which encodes pendrin, we were prompted to study the effect of angiotensin II in vitro on pendrin-dependent transepithelial net Cl− transport (JCl).
In the present study, we asked following questions: 1) does angiotensin II increase JCl in mouse CCD?; and 2) does angiotensin II increase JCl through pendrin-dependent transcellular transport across intercalated cells? To answer these questions, JCl and transepithelial voltage (VT) were measured in CCDs perfused in vitro from wild-type and Slc26a4 null mice.
All experiments were performed using male and female Slc26a4 (−/−) mice developed by Everett et al. (4) and in wild-type mice from the same strain (129S6/SvEv Tac; Taconic Farms), which were bred in parallel. Unless otherwise indicated, mice were fed either a standard, salt-replete rodent chow (LabDiet 5001; PMI Nutrition International, Richmond, IN) or a balanced diet (53881300; Zeigler Brothers) prepared as a gel (0.6% agar, 74.6% water, and 24.8% mouse chow) supplemented with NaCl (∼1.13 meq NaCl/day) or NaCl plus furosemide (100 mg·kg−1·day−1) for 5 days.
Measurement of blood pressure, arterial pH, and serum electrolytes.
Systolic blood pressure was measured in conscious mice by tail cuff using a BP-2000 (Visitech Systems). Arterial blood was collected through the abdominal aorta under isofluorane anesthesia. Arterial pH and Pco2 were measured using ABL5 analyzer (Radiometer America, Westlake, OH). Serum electrolytes were measured as described previously (38).
In vitro perfusion of isolated CCD.
CCDs were dissected from medullary rays and perfused at flow rates of 2–3 nl·min−1·mm−1 in the presence of symmetric, physiological solutions containing (in mM) 125 NaCl, 2.5 K2HPO4, 24 NaHCO3/5% CO2, 2 CaCl2, 1.2 MgSO4, and 5.5 glucose bubbled with 95% air-5% CO2 (26, 43). Tubules were equilibrated at 37°C for 30 min before the start of the collections. As specified below, angiotensin II (10−8 M) and bafilomycin A1 (10−8 M) were applied to the bath, while benzamil hydrochloride (3 × 10−6 M) was added to the perfusate. Stock solutions of angiotensin II and benzamil hydrochloride (10−5 M, 3 × 10−3 M, respectively) were prepared in deionized water. A stock solution of bafilomycin (10−5 M) was prepared in absolute ethanol. All chemicals were purchased from Sigma-Aldrich (St. Louis, MO).
Measurement of JCl.
Cl− concentration was measured in perfusate and collected samples using a continuous-flow fluorimeter and the Cl−-sensitive fluorophore, 6 methoxy-N-(3-sulfopropyl) quinolinium (SPQ; Molecular Probes, Eugene, OR), as described previously (5, 42, 43). JCl was calculated according to the equation where Co and CL are perfusate and collected fluid Cl− concentrations, respectively, Q is flow rate (nl/min), and L is tubule length. Net fluid transport was taken to be zero since net fluid flux has not been observed in CCDs when perfused in vitro in the presence of symmetric solutions and in the absence of vasopressin (15, 16). JCl was expressed in picomoles per millimeter per minute.
VT was measured in the perfusion pipette connected to a high-impedance electrometer through an agar bridge saturated with 0.16 M NaCl and a calomel cell, as described previously (40). The reference was an agar bridge from the bath to a calomel cell.
Calculation of chloride permeability, ionic conductance, and transepithelial electrical resistance.
Chloride permeability (PCl) was calculated from the Nernst equation where c̄ is the mean Cl− concentration present in the lumen and bath and taken to be 0.129 M, Z is the valence of the ion, F is Faraday's constant, R is the gas constant, and T is temperature in degrees Kelvin. We assumed ZF to be 96.5 C/meq and RT to be 2.57 J/meq, respectively, at 37°C.
Ionic conductance (GCl) was calculated based on the relationship (33) In the calculation of GCl, it was assumed that benzamil affects Cl− flux entirely through its effect on VT.
Kidney tissue from wild-type mice given 7 days of the NaCl-replete gel diet or diet plus furosemide was processed for preembedding peroxidase staining using the anti-pendrin antibody and embedded in the Epon as described previously (13). Tissue was cut into 1.5-μm sections, stained with hematoxylin, and examined by light microscopy.
All data are presented as means ± SE. Data from two collections from each tubule were averaged to obtain a single value. Each n used in the statistical analysis represents data from separate mice. To test for statistical significance between two groups, an unpaired Student's t-test was used. When results from more than two groups were compared, an ANOVA was used followed by Tukey's protected t-test. The criterion for statistical significance was P < 0.05.
Effect of angiotensin II on transepithelial Cl− transport, JCl, and VT.
In CCDs from wild-type mice ingesting the NaCl-replete diet, JCl and VT were not different from zero in either the presence or absence of 10−8 M angiotensin II in the bath solution (Fig. 1). Thus, under basal conditions, no appreciable net transepithelial Cl− transport was observed.
Effect of furosemide on pendrin expression.
We sought a treatment model in which pendrin protein expression is increased. Figure 2 shows that pendrin immunostaining appears more prominent in both connecting tubule and CCD in mice given a NaCl-replete diet and furosemide relative to mice ingesting the NaCl-replete diet alone, as described previously (25). Moreover, the subapical cytoplasmic area was larger in pendrin-positive cells from furosemide-treated mice, which gave the appearance that these cells “bulge” into the lumen. Thus further studies examined Cl− transport in furosemide-treated mice.
Because pendrin is critical in the renal absorption of Cl− (43), the effect of furosemide on apparent vascular volume in wild-type and Slc26a4 null mice was explored (Table 1). As shown, after loop diuretic treatment, Slc26a4 null mice developed greater weight loss, a higher blood urea nitrogen and lower blood pressure than wild-type mice over the treatment period. Thus with furosemide administration these mutant mice had greater apparent vascular volume contraction than wild-type mice. We conclude that pendrin is upregulated with loop diuretics, which makes pendrin critical for the conservation of apparent vascular volume in this treatment model.
Angiotensin II stimulates Cl− absorption in CCDs from furosemide-treated mice.
The effect of angiotensin II on JCl and VT were studied in mice given the high-NaCl diet and furosemide (Fig. 3). Cl− absorption and a lumen-negative VT were observed in CCDs from wild-type mice (Fig. 3B). With the application of angiotensin II to the bath, JCl doubled (Fig. 3A), although VT did not become more lumen negative (Fig. 3B). We conclude that angiotensin II-induced Cl− absorption is not mediated by an increase in electromotive force.
In contrast, in CCDs from furosemide-treated Slc26a4 null mice, Cl− secretion was observed, which was unaffected by angiotensin II (Fig. 3A). VT in CCDs from furosemide-treated Slc26a4 null mice was not different from zero either in the presence or absence of angiotensin II (Fig. 3B). We conclude that angiotensin II-induced Cl− absorption observed in wild- type mice is dependent on pendrin expression.
Angiotensin II does not increase Cl− absorption through paracellular transport.
Because the lumen-negative VT does not increase with angiotensin II, the driving force for paracellular Cl− absorption does not increase with application of the hormone. However, paracellular Cl− transport might increase with angiotensin II by augmenting paracellular permeability, such as through changes in tight junction properties. To test this hypothesis, the effect of angiotensin II on JCl was tested when the H+-ATPase was inhibited with application of bafilomycin (10−8 M) to the bath. Basolateral plasma membrane H+-ATPase inhibition should reduce net H+ efflux in type B cells, thus reducing intracellular HCO3− concentration and hence reducing the driving force for apical anion exchange (3, 19, 36). The effect of H+-ATPase inhibition on JCl is shown in Fig. 4. In the presence of bafilomycin, angiotensin II did not increase JCl (Fig. 4A). Because the lumen-negative VT persisted following bafilomycin application (Fig. 4B), failure of angiotensin II to stimulate JCl did not occur from a reduced electromotive force. Thus the angiotensin-induced increase in JCl does not involve increased paracellular permeability.
Cl− absorption occurs in the absence of an electromotive driving force.
We asked whether Cl− absorption occurs in the absence of an electromotive force, such as following application of the ENaC inhibitor benzamil (Fig. 5). With benzamil (3 × 10−6 M) present in the perfusate and with angiotensin II (10−8 M) present in bath, the lumen-negative VT was abolished, although Cl− absorption was reduced by only 50%. Thus significant Cl− absorption is observed in the absence of an electromotive driving force. Further experiments examined the mechanism of this “active,” benzamil-insensitive component of Cl− absorption. Following application of bafilomycin to the bath, where reduced apical Cl−/HCO3− exchange is expected, benzamil-insensitive Cl− absorption was abolished. We conclude that active Cl− absorption is H+-ATPase dependent and does not occur through paracellular transport.
Based on these data (Table 2, Fig. 5), we calculated PCl to be 0.81 × 10−5 cm/s in mouse CCD, similar to the PCl estimated in rat CCD by Weinstein (0.60 × 10−5 cm/s) (45). Based on this PCl, a chloride conductance, GCl, of 3.8 mS/cm2 was calculated. If the epithelial Na+ and Cl− conductances are comparable, then the overall CCD electrical resistance is 132 Ω·cm2 in mouse CCD, compared with a resistance of 266 Ω·cm2 observed in rabbit CCD under basal conditions (33) and 51 Ω·cm2 observed in CCDs from desoxycorticosterone-treated rats (28).
The present study demonstrates 1) mouse CCDs secrete or absorb Cl− on demand; 2) Cl− absorption in mouse CCD is increased by angiotensin II; and 3) angiotensin II-induced Cl− absorption occurs through transcellular transport across intercalated cells, through a pendrin- and H+-ATPase-dependent mechanism(s).
Previous studies in rabbit and rat demonstrated that net Cl− and HCO3− transport in the CCD are highly regulated by in vivo conditioning (9, 18). While no significant net Cl− movement is observed in rabbit CCD under basal conditions, substantial net Cl− absorption is detected in CCDs from rabbits treated with the aldosterone analog DOCA (9). Following aldosterone analogs, Cl− absorption and HCO3− secretion increase in tandem, mediated by electroneutral, Na+-independent, apical Cl−/HCO3− exchange (6, 32). This process likely occurs through increased pendrin expression in the apical plasma membrane of type B and non-A, non-B cells (26, 38, 43). Similarly, while we observed little net Cl− flux in CCDs from mice fed a NaCl-replete diet alone, with increased pendrin expression achieved with furosemide treatment in vivo (25), substantial Cl− absorption was observed. However, Cl− absorption was not observed in CCDs from furosemide-treated Slc26a4 null mice, as expected, due to the absence of pendrin-mediated Cl−/HCO3− exchange and the absence of a lumen-negative VT, which drives conductive absorption of Cl−. The absence of a lumen-negative VT in these mutant mice likely results from the reduction in ENaC expression and the low ENaC-mediated current reported previously in preliminary form within CCDs of Slc26a4 null mice (14, 22). We conclude that total and angiotensin II-stimulated Cl− absorption are pendrin dependent.
In rat and dog renal cortex, interstitial angiotensin II concentration is 3–50 nM (20). Thus 10 nM (10−8 M) falls within the physiological range of angiotensin II concentrations observed in vivo within the cortical interstitium. This angiotensin II concentration (10−8 M) was selected since Weiner et al. (44) observed that 10−7 M, but not 10−10 M angiotensin II, increased B cell apical anion exchange when applied to the bath. Since the effect of angiotensin II on JCl in the CCD has not been tested at concentrations <10−8 M, we cannot exclude the possibility that angiotensin II has a biphasic effect on JCl in the CCD as was observed in the proximal tubule (10).
The present study extends the previous study of Weiner and colleagues (44) to show that angiotensin II directly regulates JCl in the CCD through a pendrin- and H+-ATPase-dependent mechanism(s). We also show that angiotensin II stimulates intercalated cell-dependent Cl− absorption through a direct effect on the CCD, rather than through other systemic effects of angiotensin II observed in vivo, such as changes in blood pressure or changes in the distal delivery of Cl− (37).
Angiotensin II likely modulates apical anion exchange through AT1a receptors, since they represent the dominant AT1 receptor expressed in the collecting duct (1). However, the distribution of the AT1a receptor along the collecting duct is controversial. One study localized AT1 receptors along the collecting duct primarily to the apical regions of intercalated cells (12), whereas others noted diffuse localization of the receptor within both principal and intercalated cells (11). Thus angiotensin II might increase apical anion exchange directly through stimulation of intercalated cell AT1a receptors or could act on principal cell receptors, thereby changing intercalated cell transporter function through cell-cell interaction. In this regard, we cannot exclude the possibility that angiotensin II, when applied to the bath, modulates JCl in mouse CCD by diffusing across tight junctions and acting on apical plasma membrane AT1 receptors.
Since pendrin transport inhibitors are not available, we studied the role of this protein in angiotensin II-induced Cl− absorption using mice with genetic disruption of Slc26a4 (4). H+-ATPase inhibitors were also employed, which should reduce the driving force for apical anion exchange.1 While angiotensin II doubled Cl− absorption in CCDs from wild-type mice, it had no effect in tubules from Slc26a4 null mice. Similarly, inhibiting H+-ATPase-mediated transport eliminated the angiotensin II-induced increase in JCl in wild-type mice, although it did not make VT less lumen negative. Thus angiotensin II increases JCl through changes in transcellular transport across intercalated cells, rather than through changes in paracellular permeability or through changes in electromotive force.
The present study confirms and extends previous work by Terada and Knepper (35), which showed that JCl is reduced 50% with the application of amiloride to the luminal fluid and eliminated with the application of both amiloride and hydrochlorothiazide (HCTZ). While HCTZ inhibits the NaCl cotransporter in the DCT, it also inhibits carbonic anhydrase (24), which markedly reduces type B cell apical anion exchange (19). Thus apical anion exchange should fall with reduced intracellular HCO3− concentration, as expected following carbonic anhydrase inhibition, such as with HCTZ, or following H+-ATPase inhibition with bafilomycin.
In the present study, net fluid flux was assumed to be zero in the CCD (15, 16). However, furosemide-treated Slc26a4 null mice have marked apparent vascular volume contraction, which might increase vasopressin secretion. If CCDs from furosemide-treated mice absorb fluid, such as through increased circulating vasopressin, JCl reported above would underestimate true Cl− absorption.2
Pendrin-mediated Cl− absorption and HCO3− secretion might generate net ion movement into the renal interstitium, such as through changes in VT or by buffering H+ secreted into the luminal fluid. However, because pendrin is a Cl−/HCO3− exchanger, it might not generate any net ion movement. Resolving this question will require measurement of net Na+, K+, and HCO3− flux under identical experimental conditions.
In conclusion, angiotensin II increases net Cl− absorption in mouse CCD through a transcellular pathway, which is pendrin and H+-ATPase dependent, rather than through conductive paracellular pathways. The H+-ATPase provides an “active step” for transcellular Cl− transport across type B cells.
This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-52935 (to S. M. Wall). V. Pech is the recipient of an American Heart Association Postdoctoral Fellowship Award (0525384B).
↵1 Application of bafilomycin (10 nM) to the bath reduces net H+ efflux in type B intercalated cells in mouse CCD, thereby reducing intracellular pH (3). However, bafilomycin is a membrane-permeable compound (2); thus we cannot exclude the possibility that application of bafilomycin to the bath also inhibits the apical H+-ATPase, which should increase luminal HCO3− concentration (36). Either of these expects should reduce the driving force for apical anion exchange of the type B intercalated cell.
↵2 It is unlikely that our failure to detect Cl− absorption in CCDs from Slc26a4 null mice occurs exclusively from vasopressin-stimulated fluid absorption, since Cl− absorption was not observed in CCDs from Slc26a4 null mice following administration of DOCP and NaHCO3 ingestion (43). Circulating vasopressin levels are not likely elevated in Slc26a4 null mice following this treatment protocol since increased serum osmolality and apparent vascular volume contraction are not observed (38).
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