NKCC1 and hypertension: Role in the regulation of vascular smooth muscle contractions and myogenic tone

Abstract

High-ceiling diuretics (HCD), known potent inhibitors of housekeeping Na⁺,K⁺,2Cl cotransporter (NKCC1) and renal-specific NKCC2, decrease [Cl⁻]_i, hyperpolarize vascular smooth muscle cells (VSMC), and suppress contractions evoked by modest depolarization, phenylephrine, angiotensin II, and UTP. These actions are absent in nkcc1 ^/ knock-out mice, indicating that HCD interact with NKCC1 rather than with other potential targets. These findings also suggest that VSMC-specific inhibitors of NKCC1 may be considered potential pharmacological therapeutic tools in treatment of hypertension. It should be underlined that side by side with attenuation of peripheral resistance and systemic blood pressure, HCD blocked myogenic tone (MT) in renal afferent arterioles. Keeping this in mind, attenuation of MT might be a mechanism underlying the prevalence of end-stage renal disease documented in hypertensive African-Americans with decreased NKCC1 activity and in hypertensive patients subjected to chronic HCD treatment. The role of NKCC1-mediated MT in protection of the brain, heart, and other encapsulated organs deserves further investigation.

Key words::

• Hypertension
• myogenic tone
• Na⁺
• K⁺
• 2Cl⁻ cotransport
• vascular smooth muscle contractions

Abbreviations
AII	=	angiotensin II
BP	=	blood pressure
Em	=
	=	electrical membrane potential
HCD	=	high-ceiling diuretics
KCC	=	K+,Cl cotransporter
MHS and MNS	=	Milan hypertensive and normotensive strains
MT	=	myogenic tone
NCC	=	Na + ,Cl cotransporter
NKCC	=	Na + ,K + ,2Cl cotransporter
PE	=	phenylephrine
SHR	=	spontaneously hypertensive rats
SPAK	=	20/SPS1-related proline/alanine-rich kinase
VSMC	=	vascular smooth muscle cells
WNK	=	‘with no lysine’ (K) kinases

Key messages

• Na⁺,K⁺,2Cl cotransport isoform 1 (NKCC1) is involved in regulation of vascular smooth muscle cell (VSMC) contractions via [Cl⁻]_i increment, membrane depolarization, and activation of voltage-gated Ca²⁺ channels that, in turn, leads to elevation of peripheral resistance.

• NKCC1 also plays a key role in regulation of myogenic tone (MT) in renal afferent arteriole.

• Thus, systems providing cell type-specific regulation of NKCC1 in resistant vessels with low MT may be considered potential pharmacological targets in hypertension, whereas attenuation of NKCC1-dependent MT might be a mechanism underlying the prevalence of end-stage renal disease documented in hypertensive with decreased NKCC1 activity and/or subjected to chronic treatment with high-ceiling diuretics.

The hypothesis of involvement of Na⁺ ,K ⁺ ,2Cl cotransporter (NKCC) in the pathogenesis of hypertension is based on two initial observations. First, in the mid-1980s, Bianchi and co-workers obtained F₁ hybrids of the Milan hypertensive and normotensive strains (MHS and MNS, respectively) and subjected them to X-ray irradiation and bone-marrow transplantation. They discerned that NKCC activity in erythrocytes from F₁ MHS bone-marrow recipients is elevated to a similar extent as in donor strains compared to MNS bone-marrow recipients (1). Second, in erythrocytes from F₂ hybrids of MHSxMNS (1) as well as spontaneously hypertensive rats developed by K. Okamoto (SHR) and normotensive WKY rats (2), NKCC activity correlates positively with blood pressure (BP) (for more details, see (3,4)).

Two NKCC isoforms providing electroneutral symport of monovalent ions with stoichiometry of 1Na ⁺ :1K ⁺ :2Cl have been cloned from vertebrate cDNA libraries. NKCC2, expressed exclusively in apical membranes of renal epithelial cells, plays a major role in renal handling of salt and osmotically obliged water. In contrast, NKCC1 is expressed in all types of cells studied so far, including vascular smooth muscle cells (VSMC) and erythrocytes (5,6). Both isoforms are inhibited by high-ceiling diuretics (HCD), such as furosemide and bumetanide (7). The renal mechanisms implicating NKCC1 and NKCC2 in BP regulation have been considered in recent publications (8,9). This mini-review focuses on the role of NKCC1 in the maintenance of systemic and local BP via regulation of VSMC contractions and myogenic responses.

Molecular mechanisms involved during NKCC1 regulation of vasoconstriction and hypertension

Under baseline conditions ([Na⁺]_o >>[Na ⁺]_i and [Cl⁻]_o² >>[Cl⁻]_i²), NKCC1 provides inwardly directed net ion flux and maintenance of [Cl⁻]_i above values predicted by the Nernst equilibrium potential (7). Importantly, unlike the dominant contribution of K ⁺ permeability (P_K) to resting membrane potential (E_m) in skeletal and cardiac muscles, P_K and P_Cl values in VSMC are somewhat similar (10), suggesting the involvement of NKCC1-mediated [Cl⁻]_i modulation in the regulation of E_m and E_m-dependent VSMC contractions. Indeed, HCD decrease [Cl⁻]_i (11,12), hyperpolarize VSMC (11), and attenuate the activation of voltage-gated L-type Ca²⁺ channels (12). These studies disclose the mechanism underlying the inhibitory action of HCD on the baseline tone of smooth muscles (13–15) as well as their contraction triggered by modest K⁺ -induced depolarization (12), electrical stimulation (16), histamine (17), angiotensin II (AII) (18), thromboxane A₂ (19,20), oxytocin (21,22), the α-adrenergic agonist phenylephrine (PE) (12,23–25), and agonists of P2Y purinergic receptors (26). We have reported that the inhibitory action of bumetanide on VSMC contractions triggered by modest depolarization, α-adrenergic stimulation, and UTP is completely abolished in nkcc1-null mice (27). These results show that HCD inhibit VSMC contractions via their interactions with NKCC1 rather than with other potential targets.

Meyer and co-workers observed that attenuation of systolic BP was accompanied by decreased left ventricular pressure without any changes in myocardial contraction parameters (28). Garg and co-workers examined the role of attenuated tone in resistance arteries under BP declines evoked by bumetanide. They found that contractile responses in PE-treated, isolated, third-order mesenteric arteries were associated with transient NKCC activation and were suppressed by bumetanide (24). Recently, the augmented inhibitory action of bumetanide on PE-induced contractions of mesenteric arteries and aortae from SHR was demonstrated by Lee and co-workers (29). The bumetanide-sensitive component of PE-induced contractions was also sharply decreased in mice lacking SPAK (20/SPS1-related proline/alanine-rich kinase) (30), playing a key role in phosphorylation and activation of NKCC1 (31). Viewed collectively, these data indicate that augmented NKCC1 activity may contribute to the pathogenesis of hypertension via its involvement in the regulation of VSMC contractions. Indeed, systolic BP measured by tail cuff of femoral artery catheter was decreased by 15–20 mmHg in nkcc1 ^/ mice compared to wild-type animals (28,32,33). More recently, Kim and co-workers applied radiotelemetry for mean arterial BP measurement. This investigation did not detect any differences between nkcc1 ^/ and wild-type mice (34). Additional studies should be performed to clarify the nature of this discrepancy.

The mechanisms underlying altered NKCC1 activity in hypertension are still a controversial matter that probably reflects its diverse regulatory pathways and the mosaic origin of hypertension. Thus, in VSMC, [Ca²⁺]_i and aldosterone elevations increase NKCC1 activity, whereas cAMP inhibits this carrier without alteration of its expression (23,35,36). Numerous early investigations have demonstrated abnormal [Ca^{2 +}]_i handling and cAMP-mediated signaling in primary hypertension (for review, see (37,38)). Side by side with [Ca^{2 +}]_i- and cAMP-mediated signaling, the role of ‘with no lysine’ (K) kinases (WNK) in the regulation of NKCC1 and other Cl -coupled ion carriers has been well documented in numerous in-vitro studies (39–41). Recently, Bergaya and co-workers reported that Wnk^+/ mice exhibit a pronounced decrease in NKCC1 phosphorylation and BP responses following α-adrenergic activation (42). However, data on mutations of genes encoding these enzymes are limited to rare monogenic (Mendelian) forms of hypertension, such as Gordon's disease and familial hypertension of pseudohypoaldosteronism type II (43). It should also be noted that mutations of WNK1 and WNK4, seen in monogenic hypertension, contribute to BP elevation via their involvement in the abnormal activity of renal Na ⁺ ,Cl and K ⁺ ,Cl cotransporters (NCC and KCC, respectively) rather than via NKCC1 (44).

Sterile 20/SPS1-related proline/alanine-rich kinase (SPAK) is a downstream substrate of WNK1 and WNK4 and an upstream regulator of NCC and NKCC (31). NKCC1 expression is increased, whereas phosphorylated NKCC1 is decreased in the aortic tissues of SPAK-deficient mice (30). Rafigi and co-workers generated knock-in mice in which SPAK cannot be activated by WNKs (45). These animals display reduced BP that was accompanied by attenuated expression in the kidney of NCC and NKCC2 protein. To the best of our knowledge, there are no reports of altered SPAK expression and/or activity in hypertension.

Recently, Lee and co-workers reported that increased nkcc1 mRNA and NKCC1 protein content of the aorta and heart of SHR is accompanied by hypomethylation of the nkcc1 gene promoter (29). Importantly, methylation of nkcc1 promoter in normotensive WKY rats was increased with age, whereas in SHR it remained hypomethylated after development of hypertension (46). Both increased nkcc1 expression and inhibitory action of bumetanide on mesenteric artery contractions were increased with age in SHR but not in WKY rats. They also found that at 18 weeks of age the activity of DNA methyltransferase 3B (DNTB3B) in the aorta of WKY was 3-fold higher than that of age-matched SHR (46). Viewed collectively, these data strongly suggest that the maintenance of hypomethylation in nkcc1 promoter due to the decreased DNTB3B activity underlies age-dependent development of hypertension in SHR via augmented expression of this carrier in VSMC that, in turn, leads to depolarization and contraction of resistance arteries.

NKCC1 as a regulator of myogenic tone (MT)

Myogenic tone (MT) is the intrinsic ability of small arteries to constrict in response to increases in intraluminal pressure in the absence of blood flow. In all types of blood vessels studied so far, МT is accompanied by depolarization of the sarcolemma that, in turn, culminates in activation of voltage-gated Ca^{2 +} channels and elevation of [Ca^{2 +}]_i (47,48); in several resistance vessels, MT is caused by heightened myosin light chain kinase sensitivity to Ca²⁺ or by С²⁺ -independent modification of actin microfilaments (49). Considering this, it is noteworthy that both the magnitude and kinetics of MT are different in different blood vessels. Thus, in renal afferent arterioles, intraluminal pressure elevation from 90 to 160 mmHg results in almost 2-fold attenuation of inner diameter within the first 5–10 s (50). In brain arteries, MT developing in 30 s of intraluminal pressure elevation from 20 to 80 mmHg leads to complete normalization of vessel diameter (51), whereas in mesenteric arteries the partial normalization of the diameter occurs in 3–5 min of pressure elevation (27). This diversity is probably elicited by a VSMC-specific set of signaling systems involved in myogenic responses. In contrast to the ubiquitous role of depolarization and voltage-gated Ca²⁺ channels, these upstream signaling systems provide specific mechanisms of adaptation to an altered supply of oxygen and blood-derived fuels to tissues, triggered by BP and blood flow modulation.

Numerous experiments have demonstrated that furosemide and other HCD suppress the autoregulation of renal blood flow via their interaction with NKCC2 that, in turn, leads to inhibition of tubuloglomerular feedback (8). Wang and co-workers were the first to report that these compounds reversibly inhibit MT in renal afferent arterioles (20). In contrast to renal afferent arterioles, bumetanide diminished MT in mesenteric arteries by 50%–60% (27). This discrepancy might be explained by the actions of HCD on targets distinct from NKCC1. Indeed, in VSMC, furosemide partially inhibited KCC (52) and thromboxane A₂ receptor-mediated signaling (19) and activated cyclic nucleotide phosphodiesterase (22). Keeping this in mind, we investigated MT regulation in mesenteric arteries isolated from nkcc1 ^/ mice. We discovered that the inhibitory action of bumetanide on MT is completely abolished in nkcc1-null mice (27). These results disclosed, for the first time, that HCD suppress MT via their interactions with NKCC1.

Myogenic tone as a drug target

Because blood flow resistance varies inversely to the fourth power of vessel diameter (53), the involvement of abnormal MT responses in protecting against hypertension-induced organ damage was the subject of numerous investigations (54). Importantly, sustained suppression of MT evokes hypertrophic vascular remodeling that, in turn, decreases the sensitivity of MT to BP variations. As a result of this vicious circle, systemic BP elevation is transferred to the microcirculation and high BP-induced damage of encapsulated organs, such as the brain, heart, retina, and kidneys (55,56). Because of this, the impact of antihypertensive drugs and other therapeutics on MT needs careful examination.

Calcium channel blockers, such as nifedipine, amlodipine, and diltiazem, which target L-type voltage-gated channels, exhibit potent and long-lasting antihypertensive actions, ranking second in the pharmaceutical market for hypertension treatment. It is generally accepted that the blood–brain barrier protects the brain through abundant voltage-gated Ca^{2 +} channels from the side-effects of these drugs. However, such is not the case with other encapsulated target organs, including the kidneys. A growing body of evidence points to a problem associated with the development of kidney insufficiency in patients treated with calcium channel blockers compared to other antihypertensive drugs, such as thiazides and renin-angiotensin-aldosterone system inhibitors (56–59). An analysis of publications has demonstrated an increased incidence of heart failure, side by side with renal complications, in hypertensive patients treated with calcium channel blockers (60). Both in-vivo and in-vitro studies strongly indicate that problems associated with the chronic administration of calcium channel blockers to hypertensive patients are caused by MT inhibition in afferent arterioles that leads to pressure elevation in the local microcirculation despite decreased systemic BP. Indeed, given in vivo, these compounds cause a greater increase in the glomerular filtration rate than in renal plasma flow. Such observations were confirmed in isolated, perfused, intact, and hydronephrotic kidneys and by direct visualization of the nephron circulation (for review, see (61–63)). The preferential afferent arteriole actions of calcium antagonists indicate a greater abundance of L-type channels compared to efferent arterioles. Indeed, Hansen and co-workers demonstrated that Cav1.2 L-type channel subunits in rabbit kidneys are expressed in afferent but not in efferent arterioles (64).

Loutzenhiser and co-workers developed a model of isolated perfused hydronephrotic kidneys that allows direct observation of the renal microvasculature in the absence of its regulation via tubuloglomerular feedback. With this model, they demonstrated that low doses of bumetanide and furosemide dose-dependently inhibited contractions of rat afferent arterioles (20). Their results as well as the absence of HCD actions on MT in mesenteric arteries isolated from nkcc1 ^/ mice (27) prompted us to propose that augmented NKCC1 activity protects the kidneys from high BP-induced damage, whereas chronic HCD usage accelerates the development of proteinuria and end-stage kidney disease (9,65). Indeed, highly active NKCC1 keeps unchanged renal blood flow via its manifestation in VSMC of afferent arteriole possessing sharp MT in spite of elevation of systolic BP caused by augmented NKCC1 in mesenteric arteries and other blood vessels with low NKCC1-sensitive MT (Figure 1). Considering this, HCD and other inhibitors of NKCC1 abolish this protective mechanism via inhibition of MT in the renal afferent arteriole. This hypothesis is consistent with decreased by more than 2-fold NKCC1 activity in erythrocytes from normotensive and hypertensive blacks compared to their Caucasian counterparts (for more details, see (6,65)) and with ∼4-fold greater prevalence of end-stage renal disease documented in hypertensive African-Americans compared to hypertensive whites (66,67). Further investigations should clarify the relative impact of NKCC1 in the regulation of [Cl⁻]_i, E_m, and contractions of VSMC from afferent and efferent arterioles as well as the spectrum of MT-mediated side-effects of chronic HCD administration that is widespread in the treatment of heart failure and other volume-expanded disorders.

Figure 1. Mechanisms of NKCC1 involvement in vascular smooth muscle cell (VSMC)-mediated blood pressure (BP) regulation and target organ damage: a working hypothesis.

Conclusions and unresolved issues

We have drawn three major conclusions from the data considered in the mini-review:

First, NKCC1 is involved in BP elevation via [Cl⁻]_i increment, membrane depolarization, and voltage-gated Ca²⁺ channel activation that, in turn, leads to augmented MT and contractions evoked by electrical stimuli and vasoconstrictors (Figure 1).

Second, cell-type specific inhibition of NKCC1 in mesenteric arteries and other resistant blood with low-active MT may be considered a potential pharmacological approach for treatment of hypertension.

Third, attenuation of NKCC1-dependent MT might be a mechanism underlying the prevalence of end-stage renal disease documented in hypertensive African-Americans with decreased NKCC1 activity and in hypertensive patients subjected to chronic HCD treatment (Figure 1). The role of NKCC1-mediated MT in protection of the brain, heart, and other encapsulated organs deserves further investigation.

Keeping in mind the dual mechanism of NKCC1 in the pathogenesis of hypertension and its complications, several questions should be answered and alternative pharmacological tools should be tested to achieve the therapeutic targeting of this carrier.

NKCC in cultured VSMC and isolated vessels is well documented to be activated by vasoconstrictors, such as PE and AII, and inhibited by cAMP-elevating compounds and other vasodilators (23,35). Does reciprocal regulation of NKCC1 by vasoconstrictors and vasodilators contribute to vascular tone control?

Side by side with vasoactive compounds, NKCC1 is reciprocally affected by cell shrinkage and swelling (68). Our studies have revealed that VSMC contractions might be evoked by anisosmotic media (12,69). Does cell volume modulation have an impact on NKCC-mediated VSMC contractions?

Endothelium-denuded vessels in an overwhelming number of investigations have demonstrated the vasodilatory actions of HCD. More recently, we reported that treatment with the nitric oxide synthase blocker L-NAME did not affect the inhibitory action of bumetanide on MT and contractions of mesenteric arteries evoked by modest depolarization and agonists of α-adrenergic and P2Y-receptors (27). In contrast, Wang and co-workers noted that L-NAME slowed the dilating action of bumetanide on the MT of renal afferent arterioles (20). This observation raises questions about the molecular origin of HCD-sensitive endothelial elements involved in the regulation of VSMC contractions.

Should we consider [Cl⁻]_i-dependent modulation of E_m to be the only physiologically relevant consequence of NKCC1 regulation? In other words, does [Cl⁻]_i affect cellular functions via its interaction with intracellular sensor(s), as shown for other monovalent ions, such as Na⁺ and HCO₃² (70)?

Autoregulation of blood flow in kidney and other target organs display remarkable differences between human and experimental rodents. Thus, the novel approaches for in-vivo measurement of the actions of HCD and other drugs on MT in humans should be developed in forthcoming investigations.

Currently available inhibitors, such as bumetanide and furosemide, exhibit similar affinity for NKCC1 and NKCC2. Importantly, the apparent affinity of NKCC for HCD is sharply increased by diverse activators of these carriers (71). Thus, NKCC1 should have reduced HCD sensitivity under baseline conditions compared to permanently active NKCC2. Because of this, vasodilatation through NKCC1 inhibition will be subjugated by massive diuresis caused by NKCC2 inhibition. Moreover, the lack of function of NKCC1 in the inner ear epithelium causes profound deafness in mice and humans (72,73). These data suggest that cell type-specific inhibitors of SPAK and other upstream NKCC1 activators are probably the preferable challenge in antihypertensive therapy.

Acknowledgements

The editorial help of Ovid Da Silva is appreciated.

Declaration of interest: This study was supported by grants from the Canadian Institutes of Health Research (MOP-81392), the Heart and Stroke Foundation of Canada, the Kidney Foundation of Canada, the Russian Foundation for Fundamental Research (09-0073/04), and the Federal Aim Program 2009–2013 ‘Research and Teaching Resources of the Innovating Russia’.