Mechanisms of angiotensin II stimulation of NCC are time-dependent in mDCT15 cells

Abstract

Angiotensin II (ANG II) increases thiazide-sensitive sodium-chloride cotransporter (NCC) activity both acutely and chronically. ANG II has been implicated as a switch that turns WNK4 from an inhibitor of NCC into an activator of NCC, and ANG II's effect on NCC appears to require WNK4. Chronically, ANG II stimulation of NCC results in an increase in total and phosphorylated NCC, but the role of NCC phosphorylation in acute ANG II actions is unclear. Here, using a mammalian cell model with robust native NCC activity, we corroborate the role that ANG II plays in WNK4 regulation and clarify the role of Ste20-related proline alanine-rich kinase (SPAK)-induced NCC phosphorylation in ANG II action. ANG II was noted to have a biphasic effect on NCC, with a peak increase in NCC activity in the physiologic range of 10⁻¹¹ M ANG II. This effect was apparent as early as 15 min and remained sustained through 120 min. These changes correlated with significant increases in NCC surface protein expression. Knockdown of WNK4 expression sharply attenuated the effect of ANG II. SPAK knockdown did not affect ANG II action at early time points (15 and 30 min), but it did attenuate the response at 60 min. Correspondingly, NCC phosphorylation did not increase at 15 or 30 min, but increased significantly at 60 min. We therefore conclude that within minutes of an increase in ANG II, NCC is rapidly trafficked to the cell surface in a phosphorylation-independent but WNK4-dependent manner. Then, after 60 min, ANG II induces SPAK-dependent phosphorylation of NCC.

the thiazide-sensitive sodium-chloride cotransporter (NCC) is the salt-reabsorptive pathway localized to the apical membrane of the mammalian distal convoluted tubule (DCT) that is responsible for reabsorbing 5–10% of the filtered load of sodium (9). Now, almost 60 years after the introduction of the first thiazide diuretic (8), pharmacological inhibition of NCC by thiazide diuretics is recommended as first line treatment for essential hypertension (6). NCC has also been shown to play a role in genetic disorders of hypotension and hypertension (7, 18, 27, 33, 34).

Over the last decade it has become clear that the renin-angiotensin-aldosterone system (RAAS) is the primary physiological regulator of NCC. First reported were the actions of aldosterone on NCC. Chronic aldosterone exposure (3–8 days) increases total and phosphorylated NCC and results in an increase in thiazide-sensitive sodium reabsorption (14, 20, 32). Acute aldosterone (12–36 h) stimulation, by contrast, increases NCC activity and phosphorylation without a change in total NCC (16). More recently, the effects of ANG II on the cotransporter have been described. ANG II has been implicated as a switch that turns WNK4 [With-No-Lysine (K) 4] from an inhibitor of NCC into an activator of NCC (4, 5, 25). Additionally, ANG II's effect on NCC appears to require WNK4. This dependence of ANG II's effects on WNK4 has been investigated chronically with transgenic animals (4) and acutely utilizing the Xenopus laevis oocyte expression system (25). Others have demonstrated that overexpression of WNK4 decreases NCC surface expression (3, 10, 29, 33, 35). We recently reported that knockdown of WNK4 in mammalian cells increases NCC surface expression and function (15). Here, we demonstrate in mammalian cells that ANG II acutely stimulates NCC surface expression in a WNK4-dependent manner.

While corroboration of the critical and intriguing role that ANG II plays in WNK4 regulation of NCC is important, clarification of the underlying mechanisms at play is equally important. Multiple investigators have confirmed that chronic ANG II stimulation of NCC results in an increase in total and phosphorylated NCC (30, 31). However, studies have reached conflicting conclusions concerning the role of phosphorylation in acute regulation of NCC by ANG II. While acute pressor doses of ANG II result in an increase in phosphorylated NCC (30), acute (20–30 min) nonpressor doses of ANG II in vivo result in increased NCC surface expression without an increase in phosphorylation (19, 26). In vitro results have conflicted with these data, indicating that acute ANG II does increase NCC phosphorylation (25, 30).

To address this conflict and further examine the effects of acute ANG II administration, this study utilizes a well-validated model of the mammalian DCT (15). By measuring NCC activity, phosphorylation, surface abundance, and total expression, we demonstrate that physiologic doses of ANG II increase NCC function and surface expression in a WNK4-dependent manner. Interestingly, our data suggest a more complex role for Ste20-related proline alanine-rich kinase (SPAK) and phosphorylation in the regulation of NCC by ANG II. Very early (15–30 min) ANG II effects are independent of SPAK and NCC phosphorylation while later effects are dependent on SPAK and associated with increases in NCC phosphorylation.

MATERIALS AND METHODS

Materials.

Materials were purchased from Sigma (St. Louis, MO), unless stated otherwise.

Cell culture and treatments.

mDCT15, SPAK-KD, or WNK4-KD cells previously generated and characterized (15, 16) were plated on cell culture dishes and grown in growth medium containing a 50:50 mix of DMEM/F12, 5% heat-inactivated fetal bovine serum (FBS), and 1% penicillin/streptomycin/neomycin, at 37°C. Experiments were conducted when the cells reached 90–95% confluence. For SPAK-KD and WNK4-KD cells, RNAi suppression of SPAK and WNK4 was reconfirmed via immunoblotting.

Assessment of NCC function in cells.

mDCT15, SPAK-KD, or WNK4-KD cells were seeded in 12-well plates and prepared as described above. The cells were then incubated in a serum-free growth media (Opti-Mem) for 24 h before being assayed. Cells were then treated with aldosterone or vehicle (DMSO) for the indicated times and concentrations. Thirty minutes before uptake, 0.1 mM metolazone (an inhibitor of NCC) or vehicle (DMSO) was added to the media to ensure NCC inhibition during the uptake period. The medium was then changed to a ²²Na⁺-containing medium [140 mM NaCl, 1 mM CaCl, 1 mM MgCl, 5 mM HEPES/Tris pH 7.4, 1 mM amiloride (an inhibitor of ENaC), 0.1 mM bumetanide (an inhibitor of NKCC), 0.1 mM benzamil (an inhibitor of ENaC and Na Ca exchanger), 1 mM ouabain (an inhibitor of the Na-K-ATPase), and 1 μCi/ml of ²²Na⁺] with or without thiazide (0.1 mM metolazone) and incubated for 20 min. Tracer uptake was then stopped via washes with ice-cold wash buffer. Cells were subsequently lysed with 0.1% sodium dodecyl sulfate (SDS). Radioactivity was measured via liquid scintillation and protein concentrations of the lysates were determined [bicinchoninic acid (BCA) Protein Assay, Pierce]. Uptakes were normalized to nanomoles per milligram. Thiazide-sensitive uptake was given by the difference of the uptakes with and without thiazide.

Cell surface biotinylation.

mDCT15, SPAK-KD, or WNK4-KD cells were incubated as above. The cells were washed with PBS and cell surface proteins were labeled with sulfo-NHS-SS-biotin (Pierce, Rockford, IL) in PBS for 30 min at 4°C. The reaction was quenched by adding 500 μl of the quenching solution (Pierce). The cells were harvested, lysed using lysis buffer containing protease inhibitor, and homogenized by sonication on ice. The cell lysates were centrifuged briefly and supernatant was collected. Eighty microliters of the supernatant from each group were stored separately at −80°C. Biotinylated proteins in the cell lysates were isolated by incubating with NeutrAvidin gel (Pierce) for 60 min at room temp. The labeled proteins were washed and eluted in SDS-PAGE sample buffer containing 50 mM dithiothreitol as per protocol outlined in the Pierce Surface Protein Isolation Kit. Protein concentrations were determined by using BCA protein assay kit (Pierce). The eluted proteins and the cell lysates were immunoblotted as detailed below.

Immunoblotting.

Cells were incubated as above. The cells were harvested, lysed using lysis buffer containing protease inhibitor, and homogenized by sonication on ice. The cell lysates were centrifuged briefly and supernatant was collected. Proteins were resolved by SDS-PAGE and then transferred electrophoretically to PVDF membranes. After being blocked with 3% BSA, the membranes were probed with corresponding primary antibodies [1:1,000 SPAK N17 (Santa Cruz Biotechnology), 1:1,000–1:5,000 NCC, T53 phospho-NCC (gift from J. Loffing, 1:500), WNK4 (University of Dundee, 5 μg/ml), actin (Santa Cruz Biotechnology, 1:1,000)] overnight at 4°C. The blots were washed in TBST. Signal detection was done either via the Odyssey Infrared Imaging (Li-Cor Biosciences) or via chemiluminescence. For the Odyssey system, secondary antibodies used were IRDye680 goat anti-mouse (Rockland Immunochemicals, dilution 1:10,000) and then subsequent scanning of the membrane by the Odyssey Infrared Imager. Intensity of the protein bands was analyzed by using Odyssey Infrared Imaging Software (Li-Cor Biosciences). For chemiluminescence, secondary antibodies were donkey anti-rabbit horseradish peroxidase (HRP)-linked antibody (Amersham, 1:5,000) and donkey anti-sheep HRP-linked antibody (Abcam, 1:3,000). Supersignal West Pico was used for chemiluminescence (Thermo Scientific). Chemiluminescence was detected with G:Box (gelbox) and analysis by Genetools software (Syngene).

Statistical analysis.

Statistical analysis was performed using the SigmaPlot software package (Systat, San Jose, CA). Data were analyzed for statistical significance using a paired t-test, ANOVA (Holm-Sidak), or Mann-Whitney Rank Sum where appropriate. A P value of <0.05 was taken as statistically significant.

RESULTS

Acute ANG II administration enhances NCC activity.

To examine the underlying mechanisms (phosphorylation and/or surface expression) mediating ANG II's effect on NCC, we first needed to establish the physiologic dose that maximally stimulates NCC in the mDCT15 cell line. This would allow us to avoid examining the surface expression and phosphorylation of NCC at dosages where the net functional effect was negligible. Therefore, we utilized mDCT15 cells to examine the dose-dependent effect of acute ANG II on NCC activity. These cells display native NCC activity and express the known regulators of NCC activity such as WNK4 and SPAK, making them an excellent model to study the effects of ANG II on NCC (15). NCC activity was measured after 30 min of incubation with various concentrations of ANG II. The resulting dose-response curve (Fig. 1A) demonstrates a bimodal response to the hormone, with peak NCC activity increases of 35 ± 4% at 10⁻¹¹ M ANG II and 25 ± 4% at 10⁻⁶ M ANG II (n = 6, P < 0.05 compared with control, one-way ANOVA by Holm-Sidak). Other dosages ANG II had variable effects, with some demonstrating no significant change in NCC function. A biphasic or bimodal effect has been shown for other ANG II effects in the kidney (2, 11, 13, 22).

Fig. 1.

Effect of ANG II on sodium-chloride cotransporter (NCC) activity. A: mDCT15 cells were incubated in ANG II at the indicated concentrations for 30 min before being incubated in uptake medium with ²²Na⁺ and either vehicle or 1 mM metolazone. Radioactive uptake was determined as in materials and methods. Thiazide-sensitive uptake was calculated as the difference in radiotracer uptake between the metolazone-containing and the metolazone-free groups. A black dot indicates mean, gray dots denote individual data points, and bars indicate standard error; n = 6, *P < 0.05 (1-way ANOVA, Holm-Sidak) compared with control. B: mDCT15 cells were treated with 10⁻¹¹ M ANG II for the indicated times before determination of radiotracer uptake as outlined above. A black dot indicates mean, gray dots denote individual data points, and bars indicate standard error; n = 6, *P < 0.01 (1-way ANOVA, Holk-Sidak) compared with 0 min, #P < 0.01 (1-way ANOVA, Holk-Sidak) compared with 0 min.

Having established that ANG II administration increases NCC activity at some dosages but not others, we further examined the effects at the maximally effective dose (10⁻¹¹ M), which was consistent with in vivo physiologic, nonpressor levels of ANG II. To establish the appropriate time frame for further studies, mDCT15 cells were treated with 10⁻¹¹ M ANG II for various lengths of time before measurement of NCC activity. As shown in Fig. 1B, ANG II significantly increases NCC activity after 15 min of exposure (n = 6, P < 0.01 compared with control). NCC activity increases further at 30 min (n = 6, P < 0.05 compared with 15 min) and is sustained at that level to 120 min. These data demonstrate that ANG II is a potent stimulator of NCC activity with a significant acute effect at physiologic concentrations of ANG II.

ANG II increases NCC surface expression.

Since both surface expression and phosphorylation have been reported to increase with acute ANG II regulation of NCC, we examined whether our increase in NCC activity was due to an increase in the amount of surface-expressed NCC cotransporter or an increase in individual NCC cotransporter activity. NCC surface expression was measured after treatment with ANG II for 15, 30, and 60 min. NCC surface expression sharply increased after 15 min of ANG II and remained elevated through 60 min (Fig. 2; n = 8, *P < 0.01 compared with 0 min, one-way ANOVA by Kruskal-Wallis). This parallels the increase in NCC activity.

Fig. 2.

Effect of ANG II on NCC surface expression and total abundance. mDCT15 cells were grown to confluence, then treated with 10⁻¹¹ M ANG II for the indicated times before labeling with biotin at 4°C. Biotinylated (cell surface) proteins were recovered using streptavidin agarose and immunoblotted for NCC. Representative blot shown with accompanying densitometry; n = 8, *P < 0.01 (1-way ANOVA, Kruskal-Wallis) compared with 0 min.

ANG II effects on NCC are dependent on the angiotensin receptor.

ANG II serves as a ligand for multiple receptors. However, the renal effects are primarily felt to be secondary to the ANG II receptor type 1 (AT₁R). To confirm that these acute effects of ANG II upon NCC were mediated by the AT₁R, NCC activity in response to ANG II was measured in the presence of 100 μM losartan, an AT₁R blocker. As shown in Fig. 3, losartan abolished any effect of ANG II on NCC activity at the previously determined maximally effective dosage. This finding confirms that at this dosage ANG II acts via the AT₁R.

Fig. 3.

Effect of ANG II receptor antagonism on ANG II-induced NCC activity. mDCT15 cells were treated with 10⁻¹¹ M ANG II (AII) for 30 min in the presence or absence of the ATII receptor antagonist losartan (100 μM, ARB). A black dot indicates mean, gray dots denote individual data points, and bars indicate standard error; n = 6, *P < 0.05 (1-way ANOVA, Holm-Sidak) compared with control.

Acute ANG II acts via WNK4 but is SPAK-independent.

WNK4 and SPAK have been described as key mediators of chronic ANG II actions on NCC (25). To assess these kinases' role in acute ANG II effects in mDCT15 cells, mDCT15 cell lines were treated with shRNA to stably reduce expression of WNK4 (shWNK4) and SPAK (shSPAK).

Western blotting for WNK4 of shWNK4 cells demonstrated a 72 ± 6% reduction in WNK4 expression, consistent with previous studies (15) (data not shown, n = 4, P < 0.05). In stark contrast to the effect of ANG II on control cells, treatment of shWNK4 cells with ANG II did not change NCC activity (Fig. 4A). Similarly, NCC surface expression was largely unchanged in response to ANG II, with no significant effect seen at 15 or 60 min (Fig. 4B). A statistically significant increase in NCC surface expression was seen at 30 min but this increase was small in magnitude, far smaller than the 2.5-fold increase seen in control cells (Fig. 4B). This small effect was likely due to the residual WNK4 expression in these cells. In addition, this was not sustained at 60 min and did not correspond to an increase in NCC activity. Since knock-down of WNK4 resulted in a loss of the ANG II effect on NCC, we conclude that ANG II is dependent on WNK4 for its acute actions on NCC.

Fig. 4.

Effect of WNK4 on ANG II-induced NCC activity. A: mDCT15 (control) and WNK4KD (WKD) cells 10⁻¹¹ M ANG II (AII) for 30 min before radioactive uptake determination as in materials and methods. Thiazide-sensitive uptake was calculated as the difference in radiotracer uptake between the metolazone-containing and the metolazone-free groups. A black dot indicates mean, gray dots denote individual data points, and bars indicate standard error; n = 4, *P < 0.05 (1-way ANOVA, Holm-Sidak) compared with control and #P < 0.05 (Holm-Sidak) compared with control + AII. Representative blot of WNK4 expression shown in inset. B: mDCT15 and WNK4KD cells were grown to confluence, then treated with 10⁻¹¹ M ANG II for the indicated times before labeling with biotin at 4°C. Biotinylated (cell surface) proteins were recovered using streptavidin agarose and immunoblotted for NCC. Representative blot shown with accompanying densitometry [n = 4, *P < 0.05 (1-way ANOVA, Holm-Sidak) compared with 0 min].

shSPAK cells show a 68 ± 4% reduction in SPAK expression, consistent with previous results (16) (data not shown, n = 4, P < 0.05). This level of knockdown was sufficient to prevent the stimulation of NCC by aldosterone (16). However, SPAK knockdown had no effect on ANG II stimulation of NCC, with ANG II increasing NCC activity by 35 ± 6% after 30 min of treatment compared with a 33 ± 4% increase in control cells (P < 0.05, n = 6; Fig. 5A). Surface expression similarly increased at 15 and 30 min of ANG II administration. Interestingly, NCC surface expression after 60 min of ANG II administration was not significantly different from control. Taken together, these data indicate that ANG II acutely increases NCC activity and surface expression. This effect is WNK dependent for all time points observed. SPAK dependence, by contrast, varies, with SPAK dependence seen only at the 1-h time point.

Fig. 5.

Effect of SPAK on ANG II-induced NCC activity. A: mDCT15 (C) and SPAKKD (SPKD) cells 10⁻¹¹ M ANG II (AII) for 30 min before radioactive uptake determination as in materials and methods. Thiazide-sensitive uptake was calculated as the difference in radiotracer uptake between the metolazone-containing and the metolazone-free groups. A black dot indicates mean, gray dots denote individual data points, and bars indicate standard error; n = 4, *P < 0.05 (1-way ANOVA, Holm-Sidak) compared with SPAK KD. Representative blot of SPAK expression shown in inset. B: mDCT15 and SPAKKD cells were grown to confluence, then treated with 10⁻¹¹ M ANG II for the indicated times before labeling with biotin at 4°C. Biotinylated (cell surface) proteins were recovered using streptavidin agarose and immunoblotted for NCC. Representative blot shown with accompanying densitometry [n = 4, *P < 0.05 (1-way ANOVA, Holm-Sidak) compared with 0 min].

Acute ANG II does not affect NCC phosphorylation.

WNK4 has been associated with both changes in surface expression and/or phosphorylation, depending on the system and model. SPAK directly phosphorylates NCC and its effects on NCC are universally associated with phosphorylation. Given the differences in dependence on these regulators of NCC, we studied the time-dependent effects of ANG II on NCC phosphorylation. For this, mDCT15 cells were again treated with 10⁻¹¹ M ANG II for 0, 15, 30, and 60 min. NCC phosphorylation in the resulting cell lysates was examined using a phospho-antibody specific for T53 NCC (28). As shown in Fig. 6, NCC phosphorylation did not significantly differ from control at both 15 and 30 min. However, at 60 min, NCC phosphorylation significantly increased (120 ± 15% increase, P < 0.01, n = 4; Fig. 6) compared with control. This directly parallels the SPAK dependence of the ANG II effect, indicating that immediate (<1 h) effects of ANG II are not dependent on SPAK-mediated phosphorylation of NCC, but that subsequent effects are.

Fig. 6.

Effect of ANG II on NCC phosphorylation. mDCT15 cells were grown to confluence and then treated with 10⁻¹¹ M ANG II for 30 min before lysis and immunoblotting for NCC Phospho T53 (P-NCC) and total NCC (NCC). Densitometry with representative immunoblot as inset; n = 4, *P < 0.01 (1-way ANOVA, Holm-Sidak) compared with 0 min.

SPAK knockdown decreases baseline NCC activity.

In addition to providing information concerning ANG II effects on NCC, the data also provide information concerning the regulation of NCC by surface expression and phosphorylation. The data above indicate that surface expression can be regulated independently of phosphorylation. Therefore, we sought to further explore this by examining the baseline effect of SPAK knockdown on NCC function and surface expression. While reducing SPAK expression via shRNA significantly decreased NCC function compared with control (Fig. 7A), NCC surface expression did not change (Fig. 7B). This demonstrates that higher activity with SPAK present is not a consequence of higher surface expression, instead enhancing individual cotransporter activity.

Fig. 7.

Effect of SPAK knockdown on NCC activity. A: NCC activity in mDCT15 (mDCT15) and SPAKKD (SPKD) cells was measured by radioactive uptake as in materials and methods. Thiazide-sensitive uptake was calculated as the difference in radiotracer uptake between the metolazone-containing and the metolazone-free groups. A black dot indicates mean, gray dots denote individual data points, and bars indicate standard error; n = 6, *P < 0.01 (1-way ANOVA, Holm-Sidak) compared with mDCT15. B: mDCT15 and SPAKKD cells were grown to confluence before labeling with biotin at 4°C. Biotinylated (cell surface) proteins were recovered using streptavidin agarose and immunoblotted for NCC. Representative blot shown with accompanying densitometry (n = 6).

DISCUSSION

Here, we used a validated cell culture model of the mammalian DCT that allows measurement of NCC activity independent of surface expression and phosphorylation to examine the effect of acute ANG II on NCC and the mechanisms underlying this effect. Utilizing this unique model, we demonstrated that ANG II increases thiazide-sensitive sodium reabsorption by NCC via an acute increase in surface expression as soon as 15 min. This increase in function and surface expression is sustained through 60 min and is dependent on WNK4 at all time points. The chronic effects (days) of ANG II on NCC also seem to require WNK4 (4, 25). We recently reported that knockdown of WNK4 increases baseline NCC function and surface expression significantly (15). This correlates well with studies demonstrating that overexpression of WNK4 inhibits NCC surface expression (3, 10, 29, 33, 35). This baseline inhibitory effect of WNK4 on NCC and yet dependence of the acute ANG II stimulatory effect on WNK4 has also been described in oocytes (25). Our findings suggest that in mammalian DCT cells, WNK4 at baseline inhibits cotransporter surface expression and activity, but in the presence of acute ANG II WNK4 facilitates an increase in surface expression and activity. This effect is presumably sustained chronically. While it is possible that WNK4 knockdown maximally stimulates NCC function so that ANG II can longer stimulate, we feel that our data are most consistent with the accumulating data that ANG II acts as switch, changing WNK4 from an inhibitor to a stimulator of NCC (4, 5, 25).

Corroborating this important hypothesis was critical, but the controversy in the field concerned the role of SPAK-mediated phosphorylation in the stimulation of NCC by ANG II. Our findings indicate that this ANG II effect is not associated with a significant increase in NCC phosphorylation at 15 and 30 min. Then at 60 min there is a significant increase in phosphorylation of NCC at the threonine 53 site. Correspondingly, knockdown of the kinase (SPAK) that phosphorylates NCC at T53 did not affect angiotensin's stimulation of NCC activity or surface expression (Fig. 5) at 15 and 30 min, but it suppressed this effect at 60 min.

These results are consistent with acute in vivo ANG II studies utilizing nonpressor doses of ANG II, which demonstrated a change in surface expression but no change in phosphorylation of NCC at 20–30 min (19, 26). Other investigators demonstrated that an acute pressor dose of ANG II increases NCC phosphorylation in vivo at 30 and 120 min (30). The 60-min time point has not been examined in vivo but multiple groups have shown chronic ANG II (days) increases NCC phosphorylation (4, 31).

While our in vitro results match Sandberg et al.'s (26) in vivo work, our results contrast with two in vitro studies examining acute effects of ANG II on NCC. This difference between in vivo and in vitro studies created a controversy in the field. Previous in vitro studies used mpkDCT cells and reported an increase in NCC phosphorylation with 15–30 min of angiotensin (25, 30). Function and surface expression were not assessed in these studies. There are a number of potential reasons for this contrasting data. Perhaps the most important of these is the model, as it affected the dose given and the experimentation that we were able to undertake. While some have taken phosphorylation as a proxy of function, studies have shown that phosphorylation can increase in situ activity of cotransporters without changing surface expression (16, 21). Conversely, other studies have shown that surface expression of NCC can be changed independent of phosphorylation (19). Thus, only in a system where you can measure function (thiazide-sensitive sodium absorption), surface expression, and phosphorylation independent of each other can you discern what is driving the change in function. Therefore, we began our studies by ascertaining what ANG II concentration in the physiologic range increased function of NCC in our model. Given that high pressor doses of ANG II can provoke different effects than physiologic nonpressor doses, we felt that determination of the concentration that is associated with change in activity was crucial. This eliminated the possibility that any perceived changes in phosphorylation were being counteracted by changes in surface expression. Additionally, we were aware that biphasic effects of ANG II have been reported. Indeed, our dose-response curve identified a biphasic response with peaks at 10⁻¹¹ M and 10⁻⁶ M, and a “valley” at 10⁻⁹ M. While tubular levels of ANG II may reach as high as 10⁻⁹ M, physiologic nonpressor serum ANG II levels are in the picomolar to nanomolar range. Therefore, we chose the physiologic dose that correlated with increased function (10⁻¹¹ M) for all future experiments. In the previous in vitro studies, the lowest phosphorylation-inducing dosage was 100-fold higher at 10⁻⁹ M. Since there are apical and basolateral AT₁R receptors in DCT, and all of the in vitro experiments assessing ANG II effects on NCC have not separated the apical membrane from basolateral, we are not able to delineate which receptors are involved. Even the in vivo experiments that have infused ANG II have not been able to address this issue as tubular and serum ANG II levels were not measured. This would likely require microperfusion of the DCT, which is an extremely technically difficult procedure that hasn't been reported in decades.

Having established the dose that changes function, we were then able to test whether this correlated with a change in phosphorylation, surface expression, or both. Similar to in vivo studies, this showed a change in surface expression at the very early time points without a change in phosphorylation. We feel that this reflects our use of the in vitro equivalent of a nonpressor dose of ANG II. Since mDCT15 cells natively express SPAK and WNK4, we were able to utilize shRNA to show that this ANG II effect is dependent on WNK4 and not SPAK at early time points and dependent on both at later time points. These data, when combined with data from these previous studies, suggest that short physiologic doses of ANG II do not involve SPAK-mediated phosphorylation of NCC but prolonged or pressor doses of ANG II do. We feel these data clarify the controversy, providing a reasonable explanation for the previously reported conflicting findings.

Another important observation reported here is that knockdown of SPAK decreases baseline NCC function significantly, reaffirming the direct stimulatory role of SPAK on NCC. This change in thiazide-sensitive sodium uptake was not accompanied by a change in surface expression. This adds to the evidence that SPAK-mediated phosphorylation acts primarily to increase activity of individual cotransporters without affecting the amount of NCC on the surface (16, 19, 21). However, there is also evidence in the literature that phosphorylation can affect surface expression (12, 24). A recent study demonstrated that phospho-mimicking NCC mutants (replacing T53, T8, and S71 with aspartic acids) have reduced endocytosis resulting in greater surface expression compared with wild-type NCC. Maximally stimulating SPAK-mediated NCC phosphorylation with chloride-free preincubation also decreased endocytosis resulting in increased NCC surface expression. The reasons for differences in these studies are likely secondary to the use of different model systems and different stimuli. Of the studies demonstrating lack of correlation of phosphorylation with surface expression only one used low-chloride incubation a stimulus for phosphorylation (21). That study was done in X. laevis oocytes so perhaps the difference in model systems explains the differing conclusions. In the SPAK knockdown experiments, we suppressed SPAK expression ∼70% and showed a change in function, but not in surface expression. The other studies all used hormonal stimuli to alter surface expression or phosphorylation (16, 19). It may be that, similar to the effect of ANG II, phosphorylation is associated with changes in surface expression only with prolonged or severe stimulation. Indeed, in our data at 60 min of ANG II, surface expression and phosphorylation are both increased. While it is clear that phosphorylation can increase NCC function without changing NCC surface expression under certain conditions, it also appears clear that phosphorylation can increase surface expression under other conditions (19). Considering all the data, we would propose that while surface expression and phosphorylation can be independent regulatory mechanisms under moderate or short stimulus, when the stimulation is severe or prolonged all stimulatory pathways are recruited.

The chronic effects of ANG II and aldosterone on NCC function have been well-described. Having recently examined the acute effects of aldosterone on NCC (16) and now the acute effects of ANG II on NCC, we now propose a model of RAAS effects on NCC. Within minutes of an increase in ANG II, NCC is rapidly trafficked to the cell surface in a phosphorylation-independent but WNK4-dependent manner. After 60 min, angiotensin induces SPAK-dependent phosphorylation of NCC. Then, after a number of hours, increases in aldosterone result in amplification of the SPAK-dependent increase in NCC phosphorylation and activity. If the low-salt or volume-depleted state persists, then at 2–8 days of RAAS activation both ANG II and aldosterone act to increase both total and phosphorylated NCC. Aldosterone presumably increases total NCC through inhibition of NEDD4-2-mediated ubiquitination of NCC (1, 23). The mechanism of the ANG II increase in total NCC is unclear, but appears to involve WNK4 (4). Then, at extremely chronic time points greater than 2 wk, there may be an increase in NCC message (17). This sequential physiologic stimulation of NCC by the RAAS provides a physiologic response that gradually increases in both magnitude and difficulty of reversibility. This allows ease of changing a physiologic response for acute events while providing the more sustained and robust changes necessary for chronic events.

GRANTS

This work was supported by National Institutes of Health Grants NIHK08 DK081728 (to B. Ko) and NIHR01 DK085097 (to R. S. Hoover). This work was also supported by the Research Service, Atlanta Veterans Affairs Medical Center (to R. S. Hoover).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: B.K. and R.S.H. conception and design of research; B.K., A.C.M., L.N.H., R.M., and R.S.H. performed experiments; B.K., A.C.M., L.N.H., R.M., and R.S.H. analyzed data; B.K. and A.C.M. interpreted results of experiments; B.K. prepared figures; B.K. and R.S.H. drafted manuscript; B.K. and R.S.H. edited and revised manuscript; B.K. and R.S.H. approved final version of manuscript.