Potassium Homeostasis and WNK Kinases in the Regulation of the Sodium-Chloride Cotransporter: Hyperaldosteronism and Its Metabolic Consequences

Shortly after the discovery of aldosterone (1), Jerome Conn and colleagues characterized a form of hypertension, indelibly associated with Conn’s name, that exhibited excessive aldosterone production for the degree of sodium (Na⁺) intake, severe hypertension, hypokalemia, and metabolic alkalosis (2). Primary hyperaldosteronism continues to be a major cause of hypertension, with a greater degree of risk for cardiovascular complications than can be attributed to the severity of hypertension (3). An important element of this increased risk is persistent, and inappropriate, mineralocorticoid receptor (MR) activation and its attendant consequences; these consequences include effects on many organs, including the heart, vasculature, kidney, liver, skeletal muscle, and adipocytes. Common burdens of MR activation include increased oxidative stress, inflammation, and insulin resistance that weigh heavily on all organs, but cardiac fibrosis and hypertrophy, vascular endothelial dysfunction, renal podocyte dysfunction, and proteinuria are particularly pernicious to the cardiorenal axis (4).

Whether this greater risk reflects these other metabolic effects of hyperaldosteronism, the frequent occurrence of hypokalemia or metabolic alkalosis, or a combination of these factors, is not clear. Although hypokalemia is frequently present in hyperaldosteronism, the use of the aldosterone/renin ratio as a screening test has expanded the recognition that primary hyperaldosteronism without hypokalemia is not so rare as early studies proposed and represents a substantial percentage of patients with refractory or severe hypertension (5,6).

However, the presence of hypokalemia typically exacerbates the severity of hypertension in hyperaldosteronism and is associated with a more severe disease progression (7). The adverse cardiovascular and renal effects of hypokalemia, and low potassium (K⁺) diets in general, are well known (8–11). Conversely, diets rich in K⁺ have been shown to reduce systemic BP and improve hypertension, slow the decline of renal function, and lower the incidence of cardiovascular complications (11–19).

Part of the benefit of K⁺-rich diets is attributed to their natriuretic effect (20,21). Reduced dietary K⁺ content promotes Na⁺ retention, whereas K⁺ supplementation has the opposite effect. Because low-K⁺ diets reduce aldosterone production, their effect normally supersedes aldosterone’s action to cause Na⁺ retention (9,22). In contrast, K⁺ supplementation reduces mineralocorticoid-induced Na⁺ retention and hypertension (4,23–32).

K⁺ depletion, in addition to promoting sodium chloride (NaCl) retention, consistently causes an increase in renal vascular resistance (33,34), which predisposes to salt-sensitive hypertension (24). Structurally, the earliest renal changes occur in the outer medullary collecting duct (35–37), with net K⁺ absorption (38–40), so that a portion of distal cortical net K⁺ secretion is reabsorbed (recycled) within the renal medulla (41,42). However, the degree of medullary K⁺ recycling increases during K⁺ loading (43), whereas two maneuvers that reduce net cortical K⁺ secretion (dietary K⁺ restriction and administration of amiloride) reduce K⁺ recycling (41–44). Absorptive K⁺ efflux in the cortex also participates in the regulation of tubular net K⁺ transport (45–48). Stokes (49) proposed that K⁺ recycling serves a physiologically important role to regulate medullary interstitial concentration, but data also indicate a role to modulate renal vascular resistance via tubuloglomerular feedback (50,51).

One notable effect of hypokalemia is to phosphorylate and activate the thiazide-sensitive NaCl cotransporter, NCC (SLC12A3). This conclusion comes from the work of numerous investigators, and from the synthesis of genetic, physiologic, and biochemical studies that provide our current explanation for the role of aldosterone in NCC activation.

Previous work identified that the NCC is one of the target effectors of aldosterone (52), and current evidence supports both direct and indirect effects of aldosterone to stimulate the NCC. In cultured distal convoluted tubule (DCT) cells, aldosterone acutely stimulated NCC activity and phosphorylation. The effect is mediated by MR via a serum and glucocorticoid kinase 1 and serine-threonine kinase 39 (also known as SPAK) pathway (53). These and other data support a direct effect of aldosterone on NCC (54). However, other studies suggest that aldosterone has indirect effects that may also be an important mechanism to stimulate NCC (55–57). Genetic studies identified that mutations in genes encoding two kinases with atypical lysine locations, labeled with-no-lysine (WNK) kinases were associated with pseudohypoaldosteronism type II (PHAII) (58). This led to the identification of the WNK kinase signaling pathway that acts to modulate NCC activity (59–63).

The WNK kinase pathway regulates NCC activity (64) via SPAK and oxidative stress response kinase 1 (65,66) that phosphorylate specific serine and threonine residues in the amino terminus of the NCC. However, overexpression of NCC does not recapitulate the PHAII phenotype in mice, so additional mechanisms appear to be important in the development of PHAII (67).

NCC, like the closely related cation transporters NKCC1 and NKCC2, is activated by phosphorylation at critical amino acids in the amino-terminal domain (68,69). Cell shrinkage or low intracellular chloride (Cl⁻) concentration were known to phosphorylate and activate NKCC1 at specific amino-terminal sites. Pacheco-Alverez et al. (70) examined maneuvers that reduced intracellular Cl⁻ activity and found they resulted in NCC phosphorylation and activation, as measured by ²²Na uptake.

Subsequent work identified that the WNK1 kinase bound Cl⁻, which stabilizes the inactive conformation of WNK1, preventing kinase autophosphorylation (71). This observation, the identification of genetic mutations in inward rectifier K⁺ channels that exhibit a hypokalemic phenotype (72–74), and the localization of these K⁺ channels to the basolateral membrane in proximal and distal tubules (75–77), suggested a potential role for basolateral membrane voltage as a mechanism by which extracellular K⁺ concentration ([K⁺]) affected intracellular [Cl⁻], and hence WNK activity. Studies subsequently reported that in vitro changes in extracellular [K⁺] altered plasma membrane voltage, cellular Cl⁻ efflux, WNK kinase activity, and NCC phosphorylation and activity (55,78).

These considerations suggest that a low plasma K⁺ activates, whereas high plasma K⁺ inhibits, NCC activity. Specifically, a low extracellular K⁺ results in membrane hyperpolarization (via Kir4.1/5.1) and Cl⁻ exits via a Cl⁻ channel. If basolateral Cl⁻ exit sufficiently exceeds Cl⁻ entry (via NCC or other mechanisms), the resultant reduction in intracellular Cl⁻ will relieve the inhibition of WNK autophosphorylation and allow the WNK-SPAK pathway to increase the rate of phosphorylation of NCC. This leads to further NCC activation and enhanced NaCl reabsorption. Of the four WNK isoforms, WNK4 is a likely candidate for the isoform responsible for this pathway on the basis of its level of expression in the DCT and because a low-K⁺ diet does not result in NCC phosphorylation in mice with disruption of WNK4 gene (WNK4 knockout) (79).

Conversely, a K⁺ load by increasing plasma K⁺ depolarizes the basolateral membrane of the DCT, which could increase intracellular Cl⁻ activity. A sufficient increase in intracellular [Cl⁻] could suppress WNK-SPAK and NCC activity, which would increase luminal Na⁺ concentration beyond the early DCT to increase K⁺ secretion in the late DCT, connecting segment, and cortical collecting duct. This presumes that K⁺ secretion is limited by Na⁺ delivery, so that the shift of Na⁺ absorption from the early DCT to these downstream segments enhances electrogenic Na⁺ absorption and K⁺ secretion. However, the studies of Good and Wright (46) would suggest that flow rate, and not luminal Na⁺ concentration, is the critical factor that affects K⁺ secretion, and studies by Hunter et al. (80) also suggest that inhibiting NCC with hydrochlorothiazide does not necessarily increase electrogenic Na⁺ reabsorption or produce a kaliuresis. Thus, alternative explanations merit consideration.

Further work proposed that differences previously observed between in vitro studies (which suggested that WNK4 inhibits NCC) and in vivo studies in a WNK4-knockout mouse (which suggested that WNK4 stimulated NCC), could be explained by the unique properties of WNK4 inhibition by intracellular [Cl⁻] (56). Inhibition of WNK4 phosphorylation of SPAK occurred at smaller [Cl⁻] in vitro than similar experiments for WNK1 or WNK3. Stated differently, in vitro [Cl⁻] had a higher affinity to inhibit WNK4 than WNK1 or WNK3. Because all three kinases are expressed in the DCT, the authors proposed that WNK4 was the critical WNK kinase that modulated NCC activity in the DCT and explained the discordant data. In mice adapted to different levels of K⁺ content for 10 days, plasma [K⁺] inversely correlated with phosphorylated NCC, supporting this hypothesis (56). Testing this hypothesis under conditions of active transport, Yang et al. (81) observed that a low K⁺ intake increased NCC phosphorylation and functional activity in wild-type but not WNK4-knockout mice. However, they also found that increased luminal NaCl delivery increased NCC activity in both wild-type and WNK4-knockout mice, and concluded that alterations in WNK4 mediated by intracellular [Cl⁻] was not the only mechanism that regulated NCC. More recent studies have provided additional refinements to this proposed mechanism and evidence that WNK4 signaling involves mechanisms besides intracellular [Cl⁻] (82,83).

Nevertheless, reports by Preston et al. (21) suggest that the role of increased NCC activity as an explanation of Na⁺-sensitive hypertension requires further study.

It is notable that several studies have showed in humans and rodents that an oral K⁺ load induces a kaliuresis with a protocol that did not significantly change plasma K⁺ or aldosterone, or was associated with changes in K⁺ homeostasis attributable to loss of pituitary function (84–87). Thus, it is likely that the body possesses multiple mechanisms to preserve K⁺ homeostasis. Whether changes in luminal flow to vasopressin-responsive segments, as might be expected with alterations in NCC-mediated transport, may be responsible for the kaliuresis is plausible, experimentally testable, and supported by a significant number of studies (88–93).

This brings us to the current study (94) that extends previous work by these investigators (95) showing effects of intravenous NaCl to reduce NCC abundance and phosphorylation in urinary vesicles, and builds on the growing translation of experimental animal studies into studies of human disease (96). The data and the conclusions support an effect of K⁺ to modulate NCC, which is consistent with previous studies, some of which are cited herein.

Disclosures

All authors have nothing to disclose.

Funding

C.S. Wingo is supported in part by NIH Award R56DK128271. J.G. Johnston is supported by American Heart Association Award 872573.

Author Contributions

J.G. Johnston reviewed and edited the manuscript, and C.S. Wingo wrote the original draft.