Potassium Supplementation Prevents Sodium Chloride Cotransporter Stimulation during Angiotensin II Hypertension

Abstract

Ang II hypertension increases distal tubule Na-Cl cotransporter (NCC) abundance and phosphorylation (NCCp), as well as epithelial Na⁺ channel (ENaC) abundance and activating cleavage. Acutely raising plasma [K⁺] by infusion or ingestion provokes a rapid decrease in NCCp that drives a compensatory kaliuresis. The first aim tested whether acutely raising plasma [K⁺] with a single 3 hr 2% potassium meal would lower NCCp in Sprague Dawley rats after 14 days of AngII (400 ng/kg/min). The potassium-rich meal neither decreased NCCp nor increased K⁺ excretion. AngII infused rats exhibited lower plasma [K⁺] versus controls (3.6 ± 0.2 vs. 4.5 ± 0.1 mmol/L, p < 0.05) suggesting that Ang II mediated ENaC activation provokes K⁺ depletion. The second aim tested whether doubling dietary potassium intake from 1% (A1K) to 2% (A2K) would prevent K⁺ depletion during AngII infusion and, thus, prevent NCC accumulation. A2K fed rats exhibited normal plasma [K⁺] and 2-fold higher K⁺ excretion and plasma [aldosterone] versus A1K. In A1K rats, NCC, NCCpS71, and NCCpT53 abundance increased 1.5- to 3-fold versus controls (p< 0.05). The rise in NCC and NCCp abundance was prevented in the A2K rats, yet blood pressure did not significantly decrease. ENaC subunit abundance and cleavage increased 1.5- to 3-fold in both A1K and A2K; ROMK abundance was unaffected by Ang II or dietary K⁺. In summary, the accumulation and phosphorylation of NCC seen during chronic AngII infusion hypertension is likely secondary to potassium deficiency driven by ENaC stimulation.

Introduction

The angiotensin II (AngII) infusion model of experimental hypertension has been implemented in thousands of studies and has provided insights into mechanisms that cause the resultant hypertension and cardiovascular injury. ^1–3 We and others have defined key intrarenal mechanisms controlling the blood pressure by profiling the regulation of sodium transporters expressed along the nephron. In brief, we found that AngII increases abundance and stimulates transporters beyond the macula densa including the apically expressed cortical Na,K-2Cl cotransporter (NKCC2),⁴ NCC and both cortical and medullary ENaC, while the resultant hypertension depresses the abundance and activation of cortical and medullary Na-H exchanger (NHE3), and medullary NKCC2.⁵ In rodent models that exhibit blunted hypertensive responses to AngII, transporter profiling reveals blunting of either distal transporter activation and/or proximal transporter inhibition.^6–8 Even so, a direct association between distal co-transporter abundance and hypertension is not evident in mice overexpressing NCC, indicating the importance of integrating multiple mechanisms along the entire nephron.⁹ Chronic AngII regulation of NCC and its phosphorylation have been attributed to the stimulation of a WNK-SPAK kinase cascade.^{10, 11} Indeed, our own studies show that renal cortical SPAK (Ste20/SPS-1 related proline-alanine rich kinase) is stimulated during chronic AngII infusion hypertension and that if intrarenal production of AngII is prevented, both the NCC and SPAK stimulation are blocked and the rise in blood pressure is blunted.^{4, 8}

The beneficial effect of raising dietary potassium to lower blood pressure is indicated in both epidemiology and interventional studies in humans and laboratory animals.^12–14 It is now evident that the blood pressure lowering property may, at least in part, be linked to renal responses activated to excrete potassium: raising plasma potassium acutely (via a potassium-rich meal, oral gavage or potassium infusion) drives a rapid decrease in NCC phosphorylation and activity, which lowers Na⁺ reabsorption by the distal convoluted tubule (DCT) NCC and drives Na⁺ and volume downstream for reabsorption by cortical collecting duct (CCD) ENaC, which provokes K⁺ secretion by ROMK and BK channels in the same region, the impact of which is to match K⁺ excretion to K⁺ intake.^{15, 16} Thus, an increase in potassium input acts like a thiazide diuretic to suppress NCC activity which not only raises K⁺ excretion but also decreases Na⁺ reabsorption by NCC which can lower blood pressure set point.^{14, 17, 18} In fact, evidence suggests that homeostatic control of potassium regulation has a higher hierarchical importance than control of sodium and volume.^19–21

During AngII dependent hypertension, DCT NCCp increases 2- to 4-fold as does NCC mediated Na⁺ reabsorption.^{4, 8} Motivated by these findings, we tested the hypothesis that the NCC phosphorylation, elevated by AngII infusion hypertension, could be reversed by raising potassium intake. We discovered that AngII infusion per se provokes a kaliuresis and potassium depletion, most likely secondary to ENaC activation: Mamenko et al have determined that AngII infusion hypertension increases ENaC membrane abundance and activation far above the physiologic stimulation observed when animals were fed near-zero salt diet. ²² In support of the connection between inappropriate stimulation of ENaC by AngII driving potassium loss and NCC activation, we discovered that doubling potassium intake normalizes potassium homeostasis and prevents the rise in NCC and NCCp observed in response to AngII infusion.

Concise Methods (expanded in Supplement)

Animal protocols

All animal procedures were approved by the Institutional Animal Care and Use Committee of the Keck School of Medicine of the University of Southern California. Male Sprague Dawley rats (225–250 g body weight, Harlan Laboratories, San Diego, CA) were anesthetized intramuscularly with 200 μl of 1:1 volume ratio of ketamine (Phoenix Pharmaceuticals, St. Joseph, MO) and xylazine (Lloyd laboratories, Shenandoah, IA) and randomized to two groups, implanted with osmotic minipumps (Alzet, model 2002, Cupertino, CA) subcutaneously to deliver AngII (400 ng/kg/min; Sigma) for 14 days; or subjected to the same sham surgery. Rats were acclimated to handling, metabolic cages (Tecniplast, Italy) and to blood pressure measurement protocols before data collections.

Protocol 1. Acute K⁺-rich meal

On day 13, at 3:00 PM, rats were placed in metabolic cages with water and without food. On day 14, at 7:00 AM, overnight urine was collected and rats were given a gelled meal prepared from powdered chow (TD 88239, Harlan-Teklad, Madison, WI) containing 0.74% NaCl and supplemented to either 2% potassium with KCl (A+2K) or with no potassium added (A+0K, C+0K) as previously described.¹⁶ After 3 hr of free access to this food and water, rats were removed and anesthetized for blood and tissue collection. Bladder urine was pooled with urine collected in metabolic cage before analysis.

Protocol 2. Chronic K⁺-rich diet

During the 14 days of AngII infusion, rats were fed either the 2% potassium diet described above (A2K) or the same diet with 1% potassium (C1K, A1K, as in control chow). On day 13, at 5:00 PM, rats were placed in metabolic cages with free access to their assigned diets and water for collection of overnight urine. On the morning of day 14, rats were returned to their standard cages with water but without food for 5 hours before they were anesthetized for collection of blood and tissues.

Physiological measurements

Plasma and urine [Na⁺] and [K⁺] were assessed by flame photometry (Cole-Parmer). Blood pressure was measured by tail cuff plethysmography in acclimated rats (Visitech BP2000, Apex, NC).

Transporter profiling

Kidney homogenates were prepared and assayed as described in detail ²³ and detailed in Table S1 in the online-only Data Supplement. Signals were analyzed with Odyssey Infrared Imaging System (Li-COR) and software. Arbitrary density units were normalized to mean intensity of control group, defined as 1.0.

Statistical Analyses

Results are presented as individual measurements along with mean ± SEM. One-way ANOVA followed by a Tukey’s multiple comparisons post test was used to analyze differences between three groups. Statistical tests were calculated using GraphPad Prism (San Diego, CA). A p value of <0.05 was considered significant.

Results

AngII infusion leads to potassium depletion

In a previous study in male rats, we established that a 3 hr 2%K meal raised plasma [K⁺] from 4.0 mmol/L to 5.2 mmol/L and increased urine K⁺ excretion (UKV) more than 10-fold versus rats fed a 3 hr 0%K meal; additionally, NCCp abundance decreased 50%, consistent with a response to shift Na⁺ downstream to CCD ENaC where Na⁺ reabsorption drives K⁺ excretion.¹⁶ Our first aim was to determine whether this same 2%K meal protocol would raise plasma [K⁺] and depress NCCp abundance elevated during AngII infusion. Rats infused with AngII (400 ng/kg/min,14 days) were fasted overnight and then fed a 0%K meal (A+0K) or 2%K meal (A+2K), and non-AngII infused rats were fed the 0%K meal (C+0K). A+0K, A+2K, and C+0K groups consumed: 14 ± 2, 10 ± 1 and 14 ± 1 gms/3 hr, respectively. Unexpectedly, plasma [K⁺] was significantly lower in A+0K vs. C+0K fed rats (3.6 ± 0.2 vs. 4.5 ± 0.1 mmol/L), Figure 1. The protocol was carried out twice (and data pooled) with the same lower plasma K⁺ detected in A+0K vs. C+0K. This decrease was not detected in our previous studies of AngII infused rats,⁴ likely because those rats were fed ad lib the night before plasma was collected, preserving K⁺ intake, while in this series rats were fasted overnight during which they lost more K⁺ than they took in (K⁺output > K⁺intake). Thus, the overnight potassium deprivation unmasked K⁺ depletion in the AngII infused rats, which is not evident in the control rats.

Figure 1. Effects of AngII infusion and a single K⁺-rich meal on plasma and urinary Na⁺ and K⁺.

Rats infused with AngII for 14 days were fasted overnight and then fed either a meal containing 0% K⁺ (A+0K) or 2% K⁺ (A+2K) for 3 hr; non-AngII infused rats were fed 0% K⁺ meal (C+0K). During feeding, urine was collected in metabolic cages. UKV = urinary K⁺ excretion, UNaV = urinary Na⁺ excretion. At 3 hr, plasma was collected. Values represent individual measurements and means ± SEM; *P < 0.05 vs. C+0K; # P < 0.05 vs. A+0K after correction for multiple comparisons. NS indicates not significant. Ang II lowers plasma [K⁺] in A+0K group which was restored to C+0K level after the K⁺ rich meal (A+2K). In A+0K group, urine K⁺ is elevated vs. C+0K, evidence for kaliuresis, and does not increase after 2%K⁺ meal, evidence for K⁺ conservation.

In AngII infused rats fed a K⁺ rich meal (A+2K), plasma [K⁺] rose to 4.7 ± 0.1 mmol/L. In a previous study using an identical protocol in control rats, plasma [K⁺] rose to 5.2 ± 0.2 mmol/L.¹⁶ Urinary K⁺ excretion (UKV), collected in metabolic cages during the meal and pooled with bladder urine, was 4-fold higher in A+0K than in C+0K rats, consistent with an AngII driven K⁺ losing phenotype. Surprisingly, despite the 2%K-rich meal and rise in plasma [K⁺], there was scant evidence for increased UKV in A+2K vs. A+0K groups (p=0.21). For comparison, in the previous study in control rats fed a 3 hr 2% K meal, UKV increased seven-fold (Fig. 2B Rengarajan et al).¹⁶ Plasma [Na⁺] and urinary Na⁺ excretion (UNaV) were not different between C+0K, A+0K, A+2K groups. The results demonstrate that chronic AngII infusion increases acute postprandial potassium excretion whether the meal contains K⁺ or not, evidence that AngII infusion promotes a potassium deficient state which is detected only after an overnight fast. Acutely restoring plasma [K⁺] with a K-rich meal raises plasma [K⁺] but does not prompt the kaliuresis observed in control rats fed the same amount of potassium,¹⁶ suggesting that a significant fraction of the ingested K⁺ is used to replenish intracellular K⁺ stores.

Figure 2. Effects of AngII infusion and a single K⁺ -rich meal on NCC, SPAK and γENaC in renal cortex.

A) Immunoblots of renal cortex homogenates. Both 1 and ½ amounts of protein were loaded and the greater amount is displayed (see antibody table S1 for details). FL=full length, CL=cleaved. SPAKpS373 and OSR1pS325 were detected with the same antiserum. OSR1 was not detected at levels sufficient for quantitation. Density values, normalized to C+0K group = 1.0, displayed as means ± SEM. B) Values calculated from the normalized densities of both 1 and ½ amounts loaded, analyzed by immunoblot. Means ± SEM *=P<0.05 vs. C+0K group after correction for multiple comparisons. No differences were detected in A+2K vs. A+0K groups. Transporters not changed by AngII are summarized in Figure S3.

Acute potassium ingestion does not reduce NCC abundance or phosphorylation in AngII infused rats

To determine whether the 1 mmol/L rise in plasma [K⁺] in A+2K meal-fed rats lowered NCC phosphorylation compared to the A+0K meal-fed rats, renal transporter responses were assessed by immunoblot in cortical homogenates prepared immediately after the 3 hr meal (Figure 2) as described.¹⁶ Chronic AngII infusion (in A+0K vs. C+0K) more than doubled abundance of NCC and NCC phosphorylated at T53, S71, and S89, as well as SPAK (the kinase that phosphorylates NCC)²⁴ and its activated form SPAKp (ratios of phosphorylated to total NCC, but not SPAK, were increased by AngII, Table S2). AngII also increased full length and cleaved (activated) ENaC gamma subunit 1.7- and 3.3-fold, respectively. Thus, there is evidence for significant activation of both NCC and ENaC during AngII infusion, as previously reported.^{4, 22} The 3 hr 2%K meal (A+2K) did not reduce NCC phosphorylation despite the 1 mmol/L rise in plasma [K⁺], consistent with the lack of a large increase in UKV after the meal, supporting the notion that a significant fraction of the ingested K⁺ is used to replenish intracellular stores. The >3-fold increase in γENaC cleavage during AngII infusion can account for the increased driving force for K⁺ secretion and kaliuresis as ENaC cleavage increases channel activity and amiloride sensitive natriuresis;^25–27 the resultant K⁺ loss would stimulate the increase in NCCp. Figure S3 summarizes abundance of NHE3, NKCC2 and ROMK. NKCC2 was stimulated by AngII, as reported before,⁴ NHE3, NKCC2, NKCC2-P and ROMK were unaltered by the K-rich meal. Urinary albumin, a marker of renal injury, detected by Coomassie staining gels, was elevated in all the AngII infused rats.

Doubling dietary potassium during AngII infusion prevents NCC stimulation

The second aim tested the hypothesis that doubling dietary K⁺ intake would normalize potassium homeostasis during AngII infusion, obviating the drive to increase NCC and NCCp abundance, allowing Na⁺ delivery to ENaC to drive K⁺ excretion as in non-AngII infused rats. A preliminary study in control Sprague Dawley rats (not infused with AngII) established that chronically doubling chow K⁺ content from 1% to 2% doubled UKV but did not significantly increase UNaV, plasma [K⁺] or aldosterone (Table S4), nor did 2%K chow significantly lower NCC or NCCp abundance (Figure S4). Figure 3 and Table S5 compare the physiologic responses in AngII infused rats with doubled K⁺ intake (A2K) to responses in control (C1K) and AngII infused (A1K) rats fed 1%K chow. Plasma [K⁺] was similar in C1K and A1K groups, indicating that K⁺ depletion is not unmasked in A1K rats fed ad lib the night before plasma collection, and tended to increase in A2K versus A1K, although insignificantly. In A2K vs. A1K, UKV (not UNaV) was doubled, confirming similar dietary intake. Compared to the C1K rats, plasma aldosterone increased 10-fold in A1K, reflecting AngII stimulation, and increased 20-fold in A2K, reflecting AngII plus potassium stimulation of aldosterone synthesis. As reported and discussed previously,⁴ urinary Na⁺, K⁺ and volume are higher and plasma [Na⁺] lower in the AngII infused rats, reflecting increased food intake and thirst.

Figure 3. Physiologic responses to doubling dietary potassium throughout AngII hypertension.

Groups: control rats fed 1% K⁺ diet (C1K, n=8), AngII infused rats fed 1% K⁺ (A1K, n=7) or 2% K⁺ (A2K, n=8) diet during 14 days of AngII infusion. For systolic BP: C1K: n=5; A1K: n=3; A2K: n=4. UNaV = urine Na⁺ excretion, UKV = urine K⁺ excretion, UV = urine volume. Values represent individual measurements and means ± SEM. *P < 0.05 vs. C1K; # = P < 0.05 vs. A1K after correction for multiple comparisons. NS indicates not significant. Tabular format presented in Table S5. AngII infusion raised BP and heart weight. Doubling dietary K⁺ did not significantly lower systolic BP, shown for both days 11, 12, 13 separately as well as mean values summarized for individual rats in adjacent panels; nor did K⁺ alter heart weight.

Blood pressure (BP) and heart weight were elevated in A1K vs. C1K, as expected. Doubling dietary K⁺ (in A2K versus A1K) did not significantly lower systolic BP or reduce cardiac hypertrophy, likely due to the persistent overstimulation of ENaC by AngII.²² Immunoblots of renal cortex homogenates from C1K, A1K and A2K are summarized in Figure 4 and Figure S5, S6. As predicted by our hypothesis, doubling K⁺ intake prevented the increase in NCC, NCCpT53 and NCCpS71 abundance levels during AngII infusion: levels in A2K were significantly less than levels in A1K and not significantly more than baseline levels in un-infused C1K group; both NCC total and NCC phosphorylation were suppressed about 40% in A2K versus A1K, yet ratio of NCCpT53/total NCC remained elevated in A2K suggesting that the pool of total NCC was suppressed greater than the pool of NCCpT53 (Table S3). In A1K, levels of the NCC phosphorylating kinase SPAK and SPAKp were significantly elevated 1.7- and 1.5-fold, respectively, consistent with activation during K⁺ depletion. In A2K, SPAKp (1.2 ± 0.1) did not increase above the levels in C1K, but abundance was not significantly lower than levels in A1K (1.5 ± 0.1). The results are consistent with the interpretation that SPAKp increases during AngII infusion due to the accompanying K⁺ depletion,^{23, 28, 29} and is not activated during AngII infusion when K⁺ depletion is prevented by increasing K⁺ intake. Nonetheless, the results still provide evidence for significant AngII stimulation of total SPAK in the K⁺-restored A2K group (1.5 ± 0.1) vs. C1K. The AngII stimulated increases in α-, β-, and γ-ENaC abundance and activation (cleavage), evident in A1K group, were not further stimulated by doubling K⁺ intake despite the significantly increased plasma aldosterone and urinary K⁺ excretion in the A2K. This finding reflects the unregulated over-activation of ENaC due to the AngII infusion alone, sufficient to drive the 2-fold increase in K⁺ secretion. Figure S6 summarizes the abundance of NHE3, NKCC2, NKCC2-P and ROMK, all unchanged by doubling K⁺ intake in A2K compared to A1K. Additionally, the figure shows that renin is similarly suppressed, and urinary albumin similarly increased in all the AngII infused rats.

Figure 4. Doubling potassium intake throughout AngII infusion prevents NCC activation during AngII hypertension.

Control and AngII infused rats were fed 1% or 2% K⁺ diets during the 14 days of AngII infusion. A) Immunoblots of renal cortex homogenates. Both 1 and ½ amounts of protein were loaded and the greater amount is displayed (see antibody table S1 for details). FL=full length, CL=cleaved. Density values, normalized to C1K group =1.0, displayed as means ± SE. B) Values calculated from the normalized densities of both 1 and ½ amounts loaded, analyzed by immunoblot. Means ± SEM. * =P<0.05 vs. C1K, # = P<0.05 A2K vs. A1K after correction for multiple comparisons. Scatter plots in Figure S5. NCC and NCCp abundance was reduced to levels not different from C1K in A2K. SPAKpS373 also returned to baseline in A2K, but wasn’t significantly less than A1K. ENaC levels were not further increased in A2K, showing that the elevation is driven by AngII and not high K⁺.

Discussion

By testing the hypothesis that an acute potassium-rich meal would reduce NCC phosphorylation elevated during chronic AngII infusion, we unexpectedly discovered that rats become potassium depleted by AngII infusion, manifest as kaliuresis and lower plasma [K⁺] after an overnight fast and a 0%K⁺ meal. Chronically doubling potassium intake during AngII infusion normalized potassium homeostasis and prevented the rise in NCC and NCCp. The finding illustrates that some phenotypes associated with AngII infusion, including stimulation of NCC, may be secondary to excessive urinary potassium output and prevented by doubling dietary potassium. AngII stimulation of ENaC abundance and cleavage is the likely culprit that drives the unregulated K⁺ secretion and excretion, which secondarily increases NCC abundance and phosphorylation to reduce Na⁺ delivery to ENaC to blunt K⁺ loss.

That ENaC activity is elevated by AngII is supported by many studies. In isolated tubules, AngII directly and acutely increases ENaC: channel translocation to the plasma membrane, channel activation and Na⁺ transport mediated by AT1R and superoxide generation,^30–32 all independent of aldosterone action.³⁰ Chronic AngII infusion (at a dose and time similar to that used in this study) raises ENaC activity and abundance in the plasma membrane far above that observed with physiologic stimuli, also independent of mineralocorticoid action,²² and amiloride treatment reduces the progression of AngII hypertension suggesting a non-redundant role for AngII activation of ENaC in the hypertension in this model.³³ While direct ENaC channel activity was not assessed in this study, we did assess cleavage and abundance of the alpha and gamma subunits. Cleavage is well established to activate channels in vitro and in vivo,^{25, 26} evident as greater amiloride sensitive natriuresis.²⁷

There are two main routes to reach potassium deficiency: reduce K⁺ intake rate below K⁺ output rate or increase K⁺ excretion above K⁺ intake by increasing the driving force for K⁺ excretion, e.g. by activating ENaC activity and/or increasing tubular flow rate to increase K⁺ secretion via ROMK or BK channels.¹⁴ AngII is reported to inhibit ROMK channel activity during dietary K⁺ deficiency but not in normal K⁺ diet.³⁴ Thus, in the setting of AngII infusion, we expect that both the driver (ENaC) and the effector of K⁺ secretion (ROMK) are stimulated; the “escape” is activation of NCC phosphorylation to reduce Na⁺ delivery to ENaC, all occurring in the setting of “normal” dietary K⁺ intake. Why wasn’t K⁺ depletion apparent in previous studies of AngII hypertension? Because in our previous studies controls and AngII infused rats were fed normal chow ad lib the night before blood was collected and plasma [K⁺] was 4.5 mmol/L in both groups.⁴ In protocol 1 of the current study both groups were fasted overnight and then fed a K⁺ free meal before sampling blood, revealing significantly lower plasma [K⁺] in the AngII infused versus non-infused rats (3.6 versus 4.5 mmol/L K⁺). During the 3 hr 0%K⁺ meal, urinary K⁺ excretion was not different from zero in the control group and 3-fold higher in the AngII infused group, evidence for K⁺ output > K⁺ input despite zero K⁺ intake, which likely led to the fall in plasma [K⁺]. Why didn’t the acute K⁺-rich meal, with accompanying rise in plasma [K⁺], decrease NCC phosphorylation as reported previously?^{15, 16} We postulate that in AngII infused rats, negative K⁺ balance signals (unidentified) are transmitted to the DCT where they override the signals emanating from the meal and the rise in plasma [K⁺], thus maintaining NCC phosphorylation and favoring sodium reabsorption at the DCT rather than at the CCD. This notion is supported by the small increase in urinary K⁺ excretion in the A+2K group; in comparison, UKV increased seven-fold in control rats subjected to the same protocol.¹⁶ The AngII infused rats fed 2% K⁺ containing chow chronically had only a tendency to higher plasma [K⁺] (p=0.07, Figure 3, Table S5). Likewise, doubling K⁺ intake in control rats (not infused with AngII) did not increase plasma [K⁺] nor NCC (Table S4, Figure S4). Terker and Ellison report that NCC and NCCp abundance are a linear function of plasma [K⁺] when dietary K⁺ is chronically varied between 0 and 5%.³⁵ This relationship is also apparent in control versus AngII infused rats in this study after an overnight fast, but not apparent in the K⁺-rich meal fed AngII group (Protocol 1, Figure S5A), which is consistent with the lack of significant kaliuresis in this group. We predict that if enough K was consumed by the rat to actually raise plasma [K⁺] significantly, NCCp would decrease, mediated by the mechanisms recently summarized.²⁸

In the rats chronically fed 2% versus 1%K diet during AngII infusion, there is only a weak trend of lower NCC and NCCp with higher plasma [K⁺], not surprising since there is 40% lower NCC and NCCp and only a borderline change in plasma [K⁺] in the A2K group (Protocol 2, Figure S7B). There are numerous potential reasons for the lack of a clear association between plasma [K⁺] and NCC in AngII infused rats: while Terker and Ellison changed K⁺ intake to drive the change in plasma K⁺ and NCC activation, AngII stimulation of ENaC (and potentially ROMK³⁴) in the face of normal dietary K⁺ intake provokes the K⁺ deficiency. When provided with acute or chronic K⁺ supplementation, the AngII infused rats appear to retain most of it, likely to replenish cellular K⁺ stores. Nonetheless, there is a relationship between NCC activation and urinary K⁺ excretion in the AngII infused rats: no change in either in response to an acute K⁺ meal (despite rise in plasma [K⁺]), and decreased NCC activation and increased K⁺ excretion in response to chronic K⁺ supplementation. As a future direction, analysis of muscle sodium pumps and potassium pools during AngII infusion (with and without K⁺ supplementation) would reveal the extent of the potassium depletion as well as the mechanisms activated to replenish the cellular K⁺ stores.

A new study from Terker, Ellison and colleagues³⁶ shows that knocking out mineralocorticoid receptors along the nephron in mice (KS MR^−/−) reduces the abundance and activation of both ENaC and NCC accompanied by salt wasting and hyperkalemia. Analogous to the findings in our current study, the deficient NCC expression in KS MR^−/− was corrected by normalizing plasma [K⁺], in this setting by restricting dietary K⁺. The findings indicate that the low NCC abundance in KS MR^−/− is secondary to hyperkalemia, not directly due to the loss of MR stimulation.

A recent study by van der Lubbe et al²¹ tested the inverse of the issue addressed in this study, specifically, whether the decrease in NCC abundance provoked by chronic high K⁺ diet was preserved in rats subsequently infused with AngII. The results showed that AngII infusion returned NCC levels to control and doubled NCCp abundance, which the authors attributed to AngII stimulation, and also further increased γ-ENaC cleavage. While plasma potassium and urinary potassium were not measured in this study, an interpretation consistent with the findings of this study is that the AngII driven ENaC activation (stimulated above that seen during high K⁺ diet alone), provoked excessive K⁺ secretion driving K⁺ loss and compensatory increase in NCCp.

Low K⁺ diets have been reported to increase SPAK abundance,²³ attributed to lower DCT cell [Cl⁻]^{10, 28} and, thus, to contribute to the increase in NCCp. Wade and colleagues fed SPAK knockout mice a K⁺ deficient diet and found that the NCC regulatory response was blunted but not eliminated, indicating that other kinases may play a role.³⁷ We found that, like NCC phosphorylation, the regulation of SPAK increased 50% during AngII infusion and was blunted by doubling dietary K⁺ during AngII infusion but the response was more subtle: SPAKp was reduced to levels not different from baseline, nor different from the A1K group, and total SPAK remained at 50% above baseline. In our previous studies of AngII infused rats and mice, SPAK increased after only 3–4 days of AngII infusion, before there was any significant increase in ENaC subunit cleavage or blood pressure,^{8, 38} which suggests that SPAK may be directly stimulated by AngII,³⁹ that is, before K⁺ excretion is increased.

Does AngII regulate NCC independent of changes in plasma [K⁺]? A consensus is developing that acute stimulation of NCC by AngII initially occurs independent of increasing phosphorylation. Infusion of anesthetized rats with AngII for 20–30 min (with co-infusion of an angiotensin converting enzyme inhibitor to block local AngII production and prevent hypertension) drives acute trafficking of NCC in multimeric complexes within the apical membrane without any change in NCCp/NCC ratio.⁴⁰ In cultured kidney epithelial cells, NCC is rapidly trafficked to the cell surface in a phosphorylation-independent manner within minutes of an increase in AngII, then, after an hour, AngII induces SPAK-dependent NCC phosphorylation.⁴¹ The time course of chronic treatment with AngII provides support for direct activation of NCC phosphorylation by AngII: in 3 day AngII- infused rats, NCC and NCCp increase 0.5-fold in the absence of increased ENaC activating cleavage, urinary K⁺ loss or K⁺ depletion;⁴² as mentioned above, in 4 day AngII infused mice, SPAK and SPAKp increased significantly prior to NCC or NCCp increases in the same samples, suggesting that SPAK regulation precedes NCC regulation by AngII;⁸ a key caveat is that potassium balance was not measured.

Why doesn’t K⁺ supplementation lower AngII induced hypertension? Recent population studies show a clear association between higher dietary K⁺ and lower blood pressure.^{13, 14} In two recent studies, mice fed K⁺ deficient, sodium replete diets exhibited elevated blood pressures.^{2828, 29} In contrast, in this study, blood pressure had only a tendency to decrease despite pronounced blunting of NCC activation by dietary K⁺ supplementation. The simple explanation for this finding is extreme ENaC activation driving inappropriate Na⁺ reabsorption during AngII infusion – with or without K⁺ supplementation – necessitating hypertension to drive pressure natriuresis.⁵

Perspectives

The angiotensin II hypertensive model has been widely used to understand the influence of the central nervous system, blood vessels, immune cells and kidneys in blood pressure regulation. Here, we provide evidence that treating male Sprague Dawley rats with AngII for 14 days provokes a K⁺ deficient state, likely due to ENaC activation which drives K⁺ secretion along the cortical collecting duct. This unexpected effect was unmasked after an overnight fast. We previously showed that raising plasma K⁺ by intravenous infusion or a single K⁺-rich meal triggers renal natriuretic responses likely driven by decreased NCCp, a surrogate of its activity. This could explain, at least in part, the blood pressure lowering effects of high K⁺ diets. However, during Ang II infusion, raising dietary K⁺ did not decrease the blood pressure, presumably because of the unregulated ENaC activation. Besides the very well-known role of the renin angiotensin aldosterone (RAAS) system in controlling the activity of sodium transporters, we now illustrate a parallel indirect effect of RAAS system activation on renal sodium handling driven by altered potassium status. Further, our data provide additional mechanistic support for the cardiovascular benefits of a K⁺ rich diet to suppress NCC activity.

Novelty and Significance.

What is new?

Ang II infusion provokes kaliuresis, leading to a K⁺ deficient state. Acute K⁺ intake does not decrease NCCp nor significantly increase K⁺ excretion during AngII infusion, as observed in normokalemic rodents. Rather, the ingested K⁺ is likely used to restore plasma [K⁺] and K⁺ intracellular pools.

Evidence is presented that the AngII stimulation of NCC is secondary to K⁺ deficiency driven by ENaC stimulation since doubling dietary K⁺ during AngII infusion prevents increases in NCC and NCCp abundance. The results also indicate that elevation in blood pressure is independent of NCC activation during AngII hypertension, likely due to elevated ENaC activity.

What is relevant?

When studying AngII hypertension, animals become K⁺ deficient and mildly hypokalemic, therefore, some of the phenotypes observed may be K⁺ rather than AngII or blood pressure dependent. If interested in detailing AngII or hypertension specific effects, investigators are advised to consider assessing K⁺ homeostasis and/or feeding rodents a K⁺ enriched diet during AngII infusion to maintain K⁺ balance prevent K⁺ loss.

Summary

Our results suggest that the accumulation and phosphorylation of NCC measured during chronic AngII infusion hypertension is secondary to potassium deficiency driven by ENaC stimulation.