Vasopressin influences salt and water transport in renal epithelia. This is coordinated by the combined action of V2 receptor-mediated effects along distinct nephron segments. Modulation of NaCl reabsorption by vasopressin has been established in the loop of Henle, but its role in the distal convoluted tubule (DCT), an effective site for fine regulation of urinary electrolyte composition and the target for thiazide diuretics, is largely unknown. The Na+-Cl− cotransporter (NCC) of DCT is activated by luminal trafficking and phosphorylation at conserved NH2-terminal residues. Here, we demonstrate the effects of short-term vasopressin administration (30 min) on NCC activation in Brattleboro rats with central diabetes insipidus (DI) using the V2 receptor agonist desmopressin (dDAVP). The fraction of NCC abundance in the luminal plasma membrane was significantly increased upon dDAVP as shown by confocal microscopy, immunogold cytochemistry, and Western blot, suggesting increased apical trafficking of the transporter. Changes were paralleled by augmented phosphorylation of NCC as detected by antibodies against phospho-threonine and phospho-serine residues (2.5-fold increase at Thr53 and 1.4-fold increase at Ser71). dDAVP-induced phosphorylation of NCC, studied in tubular suspensions in the absence of systemic effects, was enhanced as well (1.7-fold increase at Ser71), which points to the direct mode of action of vasopressin in DCT. Changes were more pronounced in early (DCT1) than in late DCT as distinguished by the distribution of 11β-hydroxysteroid dehydrogenase 2 in DCT2. These results suggest that the vasopressin-V2 receptor-NCC signaling cascade is a novel effector system to adjust transepithelial NaCl reabsorption in DCT.
- antidiuretic hormone
- distal convoluted tubule
- sodium-chloride cotransporter trafficking
antidiuretic hormone [arginine vasopressin (AVP)] serves to control extracellular fluid homeostasis. The principal effect of AVP is found in the collecting duct where AVP increases water reabsorption. The prerequisite to ensure this function is the creation of a hypertonic interstitium via action of the thick ascending limb (TAL), which is also promoted by AVP. These epithelial effects of AVP are mediated by type 2 AVP receptors (V2R) (2, 29), whereas type 1a receptor signaling rather serves to limit the antidiuretic effects of AVP (31). Mapping receptor-specific probes to the tubular segments has revealed subtype-selective distribution along TAL, macula densa, distal convolutions, and collecting ducts (3, 28). In TAL, strong V2R expression has been observed, which agreed with increased abundance and phosphorylation of the furosemide-sensitive Na+-K+-2Cl− cotransporter (NKCC2) and enhanced NaCl reabsorption in TAL in response to AVP (8, 13, 28). V2R signaling also activates epithelial Na+ channel (ENaC)-dependent Na+ reabsorption in collecting ducts (16); AVP-induced antinatriuresis in this segment has therefore been considered as a causal element in the context of salt-sensitive hypertension (23).
Significant V2R expression has also been mapped to the distal convoluted tubule (DCT) (28), the site of the thiazide-sensitive Na+-Cl− cotransporter (NCC). The functional relevance of V2R expression in this nephron segment is not clear. Like in TAL, AVP signaling likely occurs via Gαs-coupled V2R and cAMP release (25), but DCT cell culture experiments on the role of cAMP were inconclusive (14). Chronic AVP substitution in vasopressin-deficient Brattleboro rats with central diabetes insipidus (DI) using V2R agonist markedly augmented the abundance of NCC (8), suggesting AVP sensitivity within DCT. However, the action of aldosterone may be involved as well under this condition (8), leading to an activation of NCC and ENaC along the terminal, preferentially mineralocorticoid-sensitive portion of DCT (DCT2) (1, 24, 26). Aldosterone-induced activation of NCC via specific phosphokinases has been demonstrated (5).
NKCC2 and NCC are the principal renal members of the electroneutral cation-Cl− cotransporter family (CCC) (11). The two proteins share significant structural homology, and extensive work has established that they can be activated by common mechanisms that include their surface expression and phosphorylation of conserved NH2-terminal threonines (13, 30, 33, 34, 36, 39). For NCC, an additional activating phosphorylation site (Ser71) has been identified (5). Phosphorylation of NKCC2 and NCC also utilizes the same pathways involving the serine/threonine kinases, with no lysine K (WNK) and STE20/SPS1-related Pro/Ala-rich kinase (SPAK) (5, 6, 24, 34).
The substantial contribution of AVP to activate NKCC2 led us to hypothesize the existence of a similar, V2R-mediated pathway regulating the activity of the related NCC. To clarify this issue, short-term effects of the V2R agonist desmopressin (dDAVP) on NCC trafficking and phosphorylation steps were studied in rats. Our results demonstrate significant increases of NCC surface expression and phosphorylation in response to V2R stimulation. Vasopressin-dependent activation of salt reabsorption in the DCT may therefore be regarded as a novel, potentially powerful component in renal volume conservation.
MATERIALS AND METHODS
Animals, tissues, treatments.
Adult (10–12 wk) male Brattleboro rats with central DI (Harlane; n = 25) and Wistar rats (n = 4) were obtained from the local animal facility (Charité Berlin) and kept on standard diet and tap water. DI rats received dDAVP by intraperitoneal injection [1 ng/g body wt or 0.3 pg/g body wt or vehicle (saline)]. After 30 min, rats were separated for morphological or biochemical analysis. For morphology, rats were anesthetized by an intraperitoneal injection of pentobarbital sodium (0.06 mg/g body wt). The kidneys were perfused retrogradely through the abdominal aorta using PBS/sucrose adjusted to 330 mosmol/kgH2O, pH 7.4, for 20 s, followed by 3% paraformaldehyde in PBS for 5 min. Kidneys were removed, dissected, and shock-frozen in liquid nitrogen-cooled isopentane. For biochemical analysis, rats were killed, the kidneys were removed, and cortex and outer medulla were dissected. To prepare suspensions of renal tubules, pieces of cortex from Wistar rats were digested with a mix of type II collagenase (306 U/ml; Pan Biotech) and type XIV protease (9.4 U/ml; Sigma) for 15 min at 37°C. The resulting tubular suspensions were washed and then incubated with dDAVP (10−6 M) in renal epithelial growth medium (REGM; Lonza) or REGM alone. Suspensions were kept at 37°C for 30 min under agitation and then fixed in 3% paraformaldehyde and placed on microscopic slides coated with poly-l-lysine for immunohistochemical analysis. All experiments were approved by the Berlin Senate (GO 062/05).
Rabbit antisera directed against NH2-terminal phosphorylation sites of rat NCC (anti-phospho-Thr53 [pT53-NCC] and anti-phospho-Ser71 [pS71-NCC]; both 1:1,500 dilution) (5, 40), guinea pig anti-NKCC2 antibody directed against NH2-terminal 85 amino acids (1:1,000 dilution; gift from D. H. Ellison, Portland, Oregon) (28), rabbit anti-NCC antibody (1:1,000 dilution; gift from D. H. Ellison) (28), and sheep anti-11β-hydroxysteroid dehydrogenase 2 (11HSD2) antibody (1:500 dilution; Chemicon) were the primary antibodies used. Sets of cryostat sections from each experiment were processed exactly in the same fashion with samples from the compared groups placed on the same slide. Sections or tubular suspensions were incubated with blocking medium (30 min), followed by primary antibody diluted in blocking medium (1 h). For multiple staining, antibodies were sequentially applied, separated by a washing step. Fluorescent Cy2- or Cy3-conjugated antibodies (DIANOVA) were applied for detection. Sections were evaluated in a Leica DMRB microscope or a Zeiss confocal microscope (LSM 710) and ZEN-software (Zeiss). Evaluation of luminal pS71-NCC immunoreactive signal intensity was performed in a standardized approach within a nonsaturated range of signal intensity, using constant optical field dimensions for all measurements.
For immunogold evaluation of NCC, perfusion-fixed kidneys were embedded in LR White resin. Ultrathin sections were incubated with anti-NCC antibody. Signal was detected with 5 nm nanogold-coupled secondary antibody (Amersham). With the use of transmission electron microscopy, immunogold signal in DCT (DCT1 and DCT2) was quantified by counting gold particles over the luminal plasma membrane or cytoplasmic vesicular compartment. Evaluation was performed in a blinded fashion. Quantification of NCC signals was performed on micrographs recorded at ×16,700 and printed at ×50,000 magnification according to an established protocol (38, 39). At least 10 DCT profiles containing an average of 5 cells/profile were evaluated per individual. Gold particles were attributed to the apical cell membrane when located near (within 20 nm of distance) or within the bilayer; particles found below 20 nm of distance to the membrane up to a depth of 2 μm or until the nuclear envelope were assigned to cytoplasmic localization.
Procedures were as described previously (37). Briefly, kidney cortices were excised and homogenized in buffer containing 250 mM sucrose, 10 mM triethanolamine, protease inhibitors (Complete; Roche Diagnostics), and phosphatase inhibitors (Phosphatase Inhibitor Cocktail 1; Sigma) (pH 7.5). The homogenates were subjected to sequential centrifugation steps to obtain postnuclear fractions by removing nuclei (1,000 g, 15 min) and vesicle-enriched fractions (removal of the large plasma membrane fragments at 17,000 g for 1 h and subsequent spinning at 200,000 g for 1 h). Pellets containing the vesicle-enriched fractions were separated in 8% polyacrylamide minigels, electrophoretically transferred to polyvinylidene fluoride membranes, and analyzed using primary antibodies against NCC (diluted 1:500), pS71-NCC (1:300), pT53-NCC (1:300), flotillin-1 (1:10,000; BD Biosciences), or β-actin (1:20,000; Sigma), horseradish peroxidase-conjugated secondary antibodies (1:3,000; Dako Cytomation), induction of chemiluminescence, exposure of X-ray films, and densitometric evaluation. Signals from the vesicle-enriched fractions were normalized for flotillin-1 as a membrane resident protein; signals from the postnuclear fractions were normalized for β-actin. The linear range of the detection was controlled by reducing the load of the postnuclear homogenates to 50%, which produced corresponding decreases of target protein (NCC: −58.5 ± 26%; P < 0.05) and loading control (β-actin: −57 ± 9%; P < 0.05) [Supplemental Fig. 1 (Supplemental data for this article can be found on the American Journal of Physiology: Renal Physiology website.]).
Bands obtained with pT53-NCC and pS71-NCC antibodies at 160 kDa corresponded to NCC monomers, those above 250 kDa to multimers (data not shown) (40). Both revealed similar changes in treated and control samples. However, because reducing conditions were used, only the monomer bands were used for safer densitometric evaluation.
Analysis of data.
Results were evaluated using routine parametric descriptive statistics. Groups were compared with two-way ANOVA and t-tests, Bonferroni corrected as appropriate. A probability level of P < 0.05 was accepted as significant. Values are given as means ± SD.
Effects of dDAVP on NCC trafficking.
In TAL, short-term administration of vasopressin induces significant trafficking of NKCC2 to the luminal membrane in mouse (13) and rat kidney (28, 44), which agrees with the significant expression of V2R at this site. Assuming that similar changes occur in DCT, the intracellular distribution of NCC was analyzed in DI rats receiving dDAVP (1 ng/g body wt for 30 min) or vehicle. The supraphysiological dose of dDAVP was selected for its established effects in previous work (13, 28). For comparison, a physiological range was tested as well (see below).
NCC immunoreactivity was diffuse in the apical cytoplasm of controls but clearly concentrated in the apical plasma membrane upon dDAVP administration (Fig. 1, A and B). Corresponding quantification of immunogold staining showed that the intracellular distribution of gold particles had shifted to the luminal plasma membrane of the dDAVP- vs. vehicle-treated groups (79 ± 12 vs. 56 ± 17% of total NCC immunoreactivity, respectively; P < 0.01) without concomitant change of total NCC immunoreactivity (Fig. 1, C, D, and G). NCC distribution was further evaluated by Western blot analysis of vesicle-enriched fractions from kidney cortex. Densitometric quantification of the respective signal intensities confirmed the results of the immunocytochemical analysis; administration of dDAVP produced a significant decrease in the vesicle-enriched fractions (−33 ± 9%, P < 0.05) (Fig. 1, E, F, and H). NCC total abundances in cortical homogenates were not different (see Fig. 3, I, J, and L). These findings demonstrate effective membrane translocation of the transporter upon short-term dDAVP administration in DI rats.
Effects of dDAVP on NCC phosphorylation.
Visual evaluation of immunoreactive pT53-NCC signals revealed a substantially higher proportion of positive DCT profiles in the dDAVP-treated rats than in controls (Fig. 2, A and D). Cross-reactivity of pT53-NCC antibody with NKCC2 was excluded by double-staining with anti-NKCC2 antibody (Fig. 2, A–F). Western blot evaluation of pT53-NCC signals obtained from postnuclear cortical fractions revealed significant increases in dDAVP-treated rats (+147%; Fig. 2, G–I). For a more comprehensive analysis of the changes in NCC phosphorylation, additional antibody directed against the alternative, specific phosphorylation site of NCC at Ser71 was studied; cross-reactivity of pS71-NCC antibody with NKCC2 was again excluded by double-staining using anti-NKCC2 antibody. pS71-NCC signal was markedly stronger after dDAVP (Fig. 3, A–F), and changes were confirmed by Western blot, revealing increases in pS71-NCC signal in the postnuclear fractions (+42 ± 31%, P < 0.05); by contrast, whole NCC levels were unchanged (Fig. 3, G–L).
To verify that the observed dDAVP-induced changes in NCC phosphorylation directly resulted from activation of V2R in DCT, we performed dDAVP stimulation of renal tubules in suspension. Confocal evaluation of fluorescence intensity revealed significant increases of pS71-NCC signal in dDAVP- compared with vehicle-treated DCTs (+72 ± 22%, P < 0.05) (Fig. 4).
In the treated DI rats, increases in luminal pS71-NCC signal were confined to DCT1 (+184 ± 20%; P < 0.05; Fig. 5, A–F and M), whereas 11HSD2-positive DCT2 showed no major changes in pS71-NCC signal compared with the vehicle group (Fig. 5, G–L and N).
Together, these results demonstrate that stimulation of V2R by short-term dDAVP application in DI rats induces rapid and strong phosphorylation of NCC.
Comparative application of physiological dose of dDAVP.
Because the dose of dDAVP applied for the above experiments (1 ng/g body wt), albeit validated in previous studies, rather corresponds to a supraphysiological range of AVP (2), we comparatively administered the more physiological dose of 0.3 pg/g body wt in DI rats. As a result, the lower dose produced similar changes with respect to NCC phosphorylation, thus validating the above experiments (Supplemental Fig. 2, A and B).
Vasopressin exerts both antidiuretic and antinatriuretic actions, which are primarily the result of its binding to V2R in kidney. The capacity of AVP to increase Na+ reabsorption in acute and chronic settings using tissues (8), isolated perfused tubules (15), and cell lines (44) has been appreciated for a long time. Although TAL and collecting ducts are established effector sites in this respect (2, 16), a role for DCT has not been clearly determined so far. The DCT mediates NaCl reabsorption of no more than 5–10% of the glomerular filtrate exclusively via the NCC, yet, it achieves an important part of the fine control of Na+ excretion, which should be as high as that accomplished by the ENaC-expressing late distal segments (10). Modulation of NCC functions may therefore be quite effective. This is illustrated by mutations of the cotransporter causing salt wasting in Gitelman's syndrome with decreased NCC performance (18, 22), by its hyperactivity in pseudohypoaldosteronism type II (PHA II) caused by mutations of WNK isoforms (45, 46), and by the substantial clinical efficacy of thiazide diuretics to reduce blood pressure. Modulation of DCT's reabsorptive capacity by AVP has been inferred from earlier data reporting upregulation of NCC in association with the vasopressin-escape phenomenon (9) and by chronic, systemic application of dDAVP to DI rats (8), but specific information on local AVP signaling in DCT compared with TAL or collecting duct is still scarce.
In TAL, a number of key features of AVP signaling have been identified (6, 11). Short-term regulation of NKCC2 by AVP involves phosphorylation of NH2-terminal threonine residues (12, 13, 28). SPAK and oxidative stress response kinase were found to interact with the phosphoacceptor site of NKCC isoforms (7, 27, 32, 34). Apical trafficking and surface expression of NKCC2 have further been established upon application of AVP (13, 25, 44), and regulatory domains of NKCC2 have been identified (4, 47).
Our present data provide clear evidence for a significant relation between V2R signaling and activation of NCC in terms of luminal trafficking and rapid, pronounced phosphorylation in response to dDAVP. Application of dDAVP in supraphysiological and near-physiological doses (2, 13, 28) exerted similar effects on NCC phosphorylation, supporting the physiological relevance of the observed changes.
Thirty minutes of dDAVP application in DI rats produced a clear shift of a substantial proportion of immunoreactive NCC from the subapical vesicle compartment to the plasma membrane without concomitant changes in total NCC abundance. The magnitude of the adluminal shift of NCC, as obtained by immunogold staining, fairly agreed with the densitometric differences resulting from Western blot evaluations, taking into account procedural differences between these approaches. The magnitude of these changes also agreed with previous data on endocrine stimulation of NCC (39).
So far, several lines of evidence have highlighted that trafficking of NCC per se is a major regulatory step in Na+ transport activation. Mutations of NCC found in Gitelman's syndrome caused reduced surface expression compatible with salt wasting in this disorder (18); genetic variants of WNK1 and WNK4 found in PHA II led to higher surface expression of NCC related with the increased Na+ reabsorption in this defect (19, 46). In a mouse model for PHAII, in which mutant WNK4 D561A was knocked in, NCC was more concentrated in the apical membrane as well (5). Likewise, chronic salt loading and acute ANG II infusion or the induction of hypertension caused significant redistribution of NCC (20, 38, 39). Therefore, the luminal shift of NCC in response to dDAVP, as observed in this study, likely contributes to the activation of the transporter.
Another major finding of our study is the increased phosphorylation of NCC at NH2-terminal phosphoacceptor sites after dDAVP treatment. In analogy to NKCC2, a cluster of threonine/serine residues in NCC was shown to be related with activation of the cotransporter (pT53, pT58, and pS71 in rat NCC) (30, 40). We have demonstrated increases in NCC phosphorylation with specific antibodies recognizing phospho-Thr53 or phospho-Ser71 of NCC (5, 40). The effects were predominantly confined to DCT1. As shown earlier for NKCC2 (12, 13, 28), vasopressin-induced NH2-terminal phosphorylation of NCC in this study obviously also coincided with increased luminal trafficking, supporting a role for NCC's phosphoacceptor domain as an end point in AVP-induced signaling. Physiologically, hypotonic, low-Cl− stress is an established, major stimulus to activate phosphorylation at this site in both NKCC2 and NCC, which accounts for increased Na+ transport activity (30, 33, 35). In analogy, pronounced phosphorylation of NCC in DCT1 in response to dDAVP, as observed in our study, is likely associated with increased activity of the transporter, especially when the accompanying enhanced luminal abundance of NCC is considered.
It may be argued that AVP administered in DI rats may have triggered other endocrine stimuli such as components of the renin-angiotensin-aldosterone system that in turn may activate NCC as well (17, 39). However, stimulation of enzymatically isolated suspensions of renal cortical tubules with dDAVP, as performed in our study, also resulted in significant phosphorylation of NCC. This demonstrates that V2R-mediated signaling of AVP can stimulate NCC in the absence of systemic endocrine components. It must further be considered that the effects of AVP were localized in 11HSD2-negative DCT1, which does not belong to the “aldosterone-sensitive distal nephron” as defined elsewhere (1, 21, 26).
Because AVP-induced, V2R-mediated activation of NKCC2 in TAL and NCC in DCT at short term appears to be phenomenologically similar with respect to phosphorylation and apical trafficking of the transporters, one may speculate on the similarity of the signal transduction pathways involved in the two segments. In TAL, we and others have clearly shown that the V2R-mediated activation of NKCC2 is triggered by cAMP and protein kinase A (PKA) (44). One may assume that this applies for the DCT as well, but compelling evidence for the role of cAMP/PKA is still scarce (14), probably because of the difficulty to obtain a suitable cell model for the DCT. With regard to the downstream signaling cascade, NKCC2 and NCC also appear to utilize similar pathways involving WNK and SPAK (5, 6, 24, 34). The interaction of SPAK with CCC and its WNK activators has been illustrated by the identification of an RFX[I/V] docking motif that is required for efficient phosphorylation of NCC (34, 41, 42, 43). Notably for NCC, the important role within its specific interaction with SPAK kinase and resulting increased phosphorylation was observed in WNK4 knockin mice (5). However, this occurred in an aldosterone-dependent manner and was probably confined to DCT2, which selectively expresses 11HSD2 conferring mineralocorticoid specificity to this subsegment (1, 21). Contrastingly, the present changes were occurring preferentially in DCT1. A heterogeneity in phosphokinase-dependent signaling may therefore exist among DCT subsegments, mediating either the effect of AVP in DCT1 or of aldosterone in DCT2. The particular role for WNK isoforms and SPAK herein remains to be studied.
In summary, our study extends the current knowledge on V2R-mediated regulative effects of AVP in DCT. Our data are compatible with an AVP-induced activation of NCC that may effectively contribute to distal tubular NaCl conservation. Activation occurs chiefly in early DCT. Under pathological conditions, excessive action of AVP, which has been shown to affect blood pressure homeostasis (23), may thus aggravate salt-sensitive hypertensive conditions by its action on DCT1. We consider these findings relevant for antihypertensive treatment strategies.
This work was supported by the Deutsche Forschungsgemeinschaft (DFG FOR 667/2).
No conflicts of interest are declared by the authors.
- Copyright © 2010 the American Physiological Society