Neurohormonal Regulation of IKs in Heart Failure: Implications for Ventricular Arrhythmogenesis and Sudden Cardiac Death.

Heart failure (HF) results in sustained alterations in neurohormonal signaling, including enhanced signaling through the sympathetic nervous system and renin-angiotensin-aldosterone system pathways. While enhanced sympathetic nervous system and renin-angiotensin-aldosterone system activity initially help compensate for the failing myocardium, sustained signaling through these pathways ultimately contributes to HF pathophysiology. HF remains a leading cause of mortality, with arrhythmogenic sudden cardiac death comprising a common mechanism of HF-related death. The propensity for arrhythmia development in HF occurs secondary to cardiac electrical remodeling that involves pathological regulation of ventricular ion channels, including the slow component of the delayed rectifier potassium current, that contribute to action potential duration prolongation. To elucidate a mechanistic explanation for how HF-mediated electrical remodeling predisposes to arrhythmia development, a multitude of investigations have investigated the specific regulatory effects of HF-associated stimuli, including enhanced sympathetic nervous system and renin-angiotensin-aldosterone system signaling, on the slow component of the delayed rectifier potassium current. The objective of this review is to summarize the current knowledge related to the regulation of the slow component of the delayed rectifier potassium current in response to HF-associated stimuli, including the intracellular pathways involved and the specific regulatory mechanisms.

action potential duration

angiotensin II receptor blockers

angiotensin II type 1 receptor

calcium‐bound calmodulin

calcium/calmodulin‐dependent protein kinase II

exchange protein activated by cyclic adenosine monophosphate

rapid component of the delayed rectifier potassium current

slow component of the delayed rectifier potassium current

voltage‐gated sodium current

mineralocorticoid receptor

phosphatidylinositol 4,5‐bisphosphate

protein kinase A

protein kinase C

epsilon isoform of protein kinase C

renin‐angiotensin‐aldosterone system

sudden cardiac death

sympathetic nervous system

small ubiquitin‐like modifier

beta‐adrenergic receptor

Heart failure (HF) refers to a clinically heterogeneous group of pathologic conditions that result in reduced cardiac output and a subsequent reduction in the systemic circulation of oxygenated blood. HF affects nearly 6 million people in the United States and constitutes a leading cause of morbidity, mortality, hospitalization, and healthcare expenditure. HF imparts a 6‐ to 9‐fold increase in the risk for sudden cardiac death (SCD) that occurs secondary to the development of ventricular arrhythmias, such as ventricular tachycardia (including torsades de pointes) and ventricular fibrillation. , Animal models of HF as well as cardiomyocytes isolated from failing hearts consistently demonstrate prolongation of cardiac action potential duration (APD), which is a risk factor for the development of ventricular arrhythmias. , Though HF mortality occurs through a variety of mechanisms, arrhythmogenic SCD accounts for up to 40% of mortality in patients with severe HF.

HF is associated with pathological structural (ie, changes in the expression of contractile and extracellular matrix proteins) and electrical (ie, changes in the expression and function of calcium handling proteins, gap junctions, and ion channels) cardiac remodeling that predispose the failing heart to arrhythmogenesis. These remodeling processes largely stem from HF‐associated changes in neurohormonal signaling and dysregulated calcium homeostasis. In particular, sustained elevations in the sympathetic nervous system (SNS) and renin‐angiotensin‐aldosterone system (RAAS) pathways, as occur in HF, are intimately involved in HF pathophysiology. The roles of the SNS and RAAS in mediating HF disease progression are evidenced by the fact that medications targeting increased catecholaminergic (β‐AR [beta‐adrenergic receptor] blockers) and angiotensin II (ACE [angiotensin‐converting enzyme] inhibitors or angiotensin II receptor blockers [ARBs]) signaling confer mortality benefit and remain standards of care in HF therapy. , , While the mechanistic basis for the pathophysiologic role of alterations in neurohormonal signaling in HF is not completely understood, various investigations have demonstrated that sustained SNS and RAAS activation inhibit a variety of species of ion channels that are important regulators of the cardiac action potential. In particular, the potential for HF‐associated changes in neurohormonal signaling to inhibit L‐type calcium channels and delayed rectifier potassium channels, including those responsible for the slow component of the delayed rectifier potassium current (IKs), have been well established.

Cardiac ion channels mediate propagation of autonomous electrical signals in the heart, thereby providing coordinated electrical stimuli that trigger synchronized contraction of cardiac muscle. In this way, ion channels are critical physiologic regulators of cardiac rhythm and contractility. The human ventricle contains many species of ion channels that contribute to cardiomyocyte membrane potential during specific phases of the ventricular action potential via conductance of sodium, calcium, and potassium ions, as illustrated in Figure 1A in the healthy heart. Currents conducted by delayed rectifier potassium channels, including IKs (Figure 1B), are the primary contributors to the rapid repolarization phase of the ventricular action potential (phase 3) that is responsible for returning cardiomyocytes to resting membrane potential in preparation for subsequent depolarizing electrical stimuli (ie, heartbeats). Accordingly, alterations in ventricular ionic currents in heart failure (summarized in Figure 1C), including reductions in delayed rectifier potassium currents, prolong ventricular APD and thereby prolong the QT interval on the surface ECG.

Figure 1

Depiction of the ventricular action potential with the major ionic currents that contribute to each phase in the healthy heart and in heart failure.

A, Diagram of the ventricular action potential in the healthy heart, demonstrating the contributions of ion channels species to each phase of the ventricular action potential, including the upstroke phase (phase 0), the early repolarization phase (phase 1), the plateau phase (phase 2), the late repolarization phase (phase 3), and the resting phase (phase 4). Inward currents are indicated by arrows pointing downward (representing ions moving from the extracellular to intracellular environment), and outward currents are indicated by arrows pointing upward (representing ions moving from the intracellular to extracellular environment). B, A simulated trajectory of outward IKs current density during the ventricular action potential in the healthy heart. C, Diagram of the ventricular action potential in the failing heart, with HF‐mediated alterations in ionic currents indicated by enlarged (enhanced current) or decreased (reduced current) arrows depicted current direction. ICa indicates voltage‐gated calcium currents; Ih, hyperpolarization‐activated cyclic nucleotide–gated current; IK1, inward rectifier potassium current; IKCa, calcium‐activated repolarizing potassium current; IKr, rapid component of the delayed rectifier potassium current; IKs, slow component of the delayed rectifier potassium current; INa, voltage‐gated sodium current; Ito, transient outward potassium current; and NCX, sodium‐calcium exchanger.

Given the essential role of ion channels in regulating cardiac rhythm and contractility, dysregulation of ion channel function by HF‐associated stimuli underlies increased arrhythmogenesis in HF, including increased development of ventricular arrhythmias. , Though conflicting data exist, research investigations have largely found reductions in both species of ventricular delayed rectifier potassium currents in limited human studies and in animal and cellular HF models. Specifically, both increased angiotensin II signaling and sustained β‐AR stimulation have been demonstrated to inhibit both the rapid component of the delayed rectifier potassium current (IKr) and IKs. , , , , These findings are supported by studies in various models, including isolated ventricular cardiomyocytes from patients with HF, that demonstrate reductions in IKs and prolongation of APD in HF. HF‐associated reductions in IKs may be of particular clinical importance since IKr is also reduced in HF, making the contributions of IKs to phase 3 repolarization in the ventricle more critical. ,

Significance of IKs in Health and Disease

IKs is conducted through a voltage‐gated potassium channel that is formed via protein‐protein interaction of the pore‐forming subunit, KCNQ1 (gene product of KCNQ1; located on chromosome 11p), with the beta subunit, KCNE1 (product of KCNE1; 21q). KCNQ1 and KCNE1 are expressed in both human atria and ventricles, where IKs plays an essential role in phase 3 cardiac repolarization. In the ventricular action potential, IKs is a major contributor to the ventricular “repolarization reserve” by providing redundant current to maintain rapid repolarization during increases in cardiac rate or during conditions in which other ventricular potassium currents are reduced (eg, pharmacological inhibition of IKr). In accordance with its contributions to the repolarization reserve, IKs contributes only modestly to repolarization during normal resting heart rates (during which time IKr is the predominant repolarizing current). However, IKs has the potential to be acutely activated by the SNS to increase the rate of ventricular repolarization to compensate for increases in cardiac rate. , , This role is evidenced by the fact that reductions in IKs may result in APD prolongation and the development of potentially fatal ventricular arrhythmias particularly during sustained adrenergic tone. For example, loss‐of‐function mutations in KCNQ1, referred to as long QT syndrome phenotype 1, constitute the most common causes of congenital long QT syndrome and predispose patients to the development of ventricular arrhythmias and SCD that typically occur during periods of sustained adrenergic stimulation (eg, swimming).

Channel activation to increase IKs in response to acute release of norepinephrine and other catecholamines comprises a key physiologic mechanism to enhance repolarization during rapid heart rates. Increases in IKs during rapid heart rates involves multiple mechanisms, including enhancement of IKs current amplitude during individual ventricular depolarizations and accumulation of IKs during rapid pacing attributable to incomplete current deactivation. IKs activation during acute or intermittent SNS activation is mediated by catecholaminergic activation of β‐ARs (primarily cardiac β1‐ARs) and involves a PKA (protein kinase A)‐dependent signaling complex that has been well characterized.

In contrast to acute and intermittent β‐AR signaling, sustained β‐AR stimulation, as occurs as part of the pathophysiology of HF, has consistently been demonstrated to reduce IKs. , A number of studies have also shown that sustained increases in angiotensin II signaling, a major signaling molecule of the RAAS pathway that is also upregulated in HF, also result in reduced IKs. , , Regardless of mechanism, reductions in IKs have been demonstrated to prolong APD and increase arrhythmia development in animal HF models. In addition, cardiomyocytes isolated from explanted hearts from patients with HF have also demonstrated reduced IKs. , Disturbances in IKs may contribute to known mechanisms of arrhythmia development, including transmural dispersion of repolarization, , since KCNQ1 and KCNE1 have both demonstrated transmural differences in expression in human ventricular myocardium. , Experiments in explanted human hearts have demonstrated that the contribution of IKs to ventricular repolarization is enhanced both during β‐AR stimulation and in conditions where IKr is reduced. Given its increased contribution to repolarization in HF, reductions in IKs have been shown to potentiate transmural dispersion of repolarization, though conflicting reports exist. , , Therefore, reductions in IKs in response to various neurohormonal triggers involved in HF pathogenesis may account for increases in the risk of ventricular arrhythmogenesis and SCD seen in HF.

Cellular Signaling Pathways Involved in IKs Regulation in HF

HF results in a multitude of alterations in neurohormonal signaling that, while initially are adaptive to maintain cardiac output in the failing heart, ultimately lead to cardiac remodeling and disease progression. In particular, chronic activation of the SNS and RAAS signaling pathways are key mediators of HF pathophysiology, and pharmacological targeting of these pathways remains the standard of care for preventing disease progression and HF‐related morbidity and mortality. Sustained activation of the SNS and RAAS pathways have also been demonstrated to regulate IKs, potentially contributing to enhanced arrhythmogenesis in patients with HF. Table 1 summarizes changes in IKs in response to HF‐associated neurohormonal signaling.

Table 1

HF‐associated Neurohormonal Signaling Molecules Demonstrated to Regulate IKs

Signaling Molecule	Receptors Involved	Duration	Effect	Reference
Angiotensin II	Angiotensin II type 1 receptor	Acute	↑ IKs	Zankov et al 40
				Matavel et al 41
			↓ IKs	Guo et al 17
				Si et al 42
				Daleau et al 43
		Chronic	↓ IKs	Matavel et al 41
Aldosterone	Angiotensin II type 1 receptor	Chronic	↓ IKs	Lv et al 44
	Mineralcorticoid receptor	Chronic	↓ IKs	Lv et al 45
Catecholamines	α1‐Adrenergic receptor	Acute	↓ IKs	Kurakami et al 46
				Kienitz et al 47
	β‐Adrenergic receptor	Acute	↑ IKs	Marx et al 30
	β‐Adrenergic receptor	Chronic	↓ IKs	Aflaki et al 15
				Shugg et al 19
Endothelin	Endothelin receptor	Acute	↑ IKs	Habuchi et al 48
			↓ IKs	Washizuka et al 49

Multiple investigations have identified regulation of IKs in response to SNS signaling, mediated through cardiac α‐ARs and β‐ARs. Alterations in IKs by enhanced catecholaminergic signaling through α‐ARs, mediated primarily through cardiac α1‐ARs, are less well established relative to β‐AR signaling. Acute activation of α1‐ARs has been demonstrated to reduce IKs by a mechanism that is mediated through PKC (protein kinase C)– and AMP kinase–dependent signaling and involves (1) downregulation of KCNQ1 and possibly KCNE1 , and (2) impaired IKs via depletion of phosphatidylinositol‐4,5‐bisphosphate (PIP2). , The implications of sustained α1‐AR stimulation on IKs remain largely unknown, however; while sustained α1‐AR stimulation might be expected to reduce IKs via both subunit downregulation of KCNQ1 and KCNE1 and depletion of PIP2, evidence has shown (1) the potential for desensitization of PIP2 hydrolysis downstream of chronic α1‐AR signaling and (2) the ability of CaM (calcium‐bound calmodulin) to interact with the PIP2 binding site on the KCNQ1 carboxyl terminus to stabilize IKs. Acute β‐AR stimulation has been consistently shown to enhance IKs through the aforementioned PKA pathway. In contrast, sustained β‐AR stimulation has been demonstrated to reduce IKs through 2 distinct pathways: the first involving activation of the Epac (exchange protein activated by cAMP) and the second involving activation of CaMKII (calcium/calmodulin‐dependent protein kinase II) . ,

Sustained elevations in RAAS signaling, mediated largely through activation of the AT1 (angiotensin II type 1 receptor) by angiotensin II, have also been demonstrated to affect IKs. Conflicting data exist regarding the role of acute angiotensin II treatment on regulating IKs, with investigations demonstrating both enhancements, mediated by phosphorylation of KCNE1 S102 by classical PKC isoforms, , , and reductions, mediated by PKC‐ε‐dependent phosphorylation of KCNQ1 S95, T96, or both and KCNE1 S102. , , Sustained treatment with angiotensin II, though less widely studied, has been demonstrated to reduce IKs through a mechanism involving cellular depletion of PIP2 in response to Gq‐coupled AT1 signaling. Chronic aldosterone exposure has also been found to reduce IKs via decreased expression of both KCNQ1 and KCNE1. Interestingly, aldosterone’s inhibitory effects on IKs have been shown to be mediated by signaling through both the MR (mineralocorticoid receptor) and AT1. , This finding supports previously described crosstalk between MR activation and subsequent activation of angiotensin II/AT1 signaling. , ,

HF also results in chronic elevation of other signaling hormones, including atrial and brain natriuretic peptides, arginine vasopressin, and endothelin. Results from 2 studies assessing the effects of acute endothelin exposure on IKs within isolated guinea pig cardiomyocytes are conflicting, with one study demonstrating IKs enhancement and the other showing IKs reduction. , The effect of chronic endothelin signaling on IKs has not been investigated. Similarly, the potential for acute or sustained exposure of atrial natriuretic peptide, brain natriuretic peptide, and vasopressin to regulate IKs have not been studied.

Potential for Crosstalk Among IKs Regulatory Pathways

Given that HF‐associated changes in neurohormonal signaling have been demonstrated to regulate IKs through a variety of pathways, there exists the potential for crosstalk among signaling pathways that regulate IKs. Most notably, the potential for crosstalk between cardiac α1‐ARs and AT1 receptors, as detailed in Figure 2A, has been consistently demonstrated in response to HF stimuli. , Specific to IKs, acute activation of both α1‐ARs and AT1 receptors have been demonstrated to produce reductions in IKs through pathways that involve PKC activation and phosphorylation of KCNE1 S102. , , Accordingly, simultaneous signaling through acute activation of both pathways may bypass homeostatic mechanisms (eg, downregulation of receptors) involved in either pathway and thus further reduce IKs.

Figure 2

Potential for cardiac‐specific crosstalk between signaling pathways demonstrated to regulate IKs.

Schematic diagrams show putative pathways involved in (left) activation of PKC downstream of acute activation of both AT1 and α1‐AR receptors, (center) activation of δCaMKII downstream of sustained activation of both AT1 and β‐AR activation, and (right) activation of AT1 receptors downstream of sustained MR receptor activation. AC indicates adenylyl cyclase; ACE, angiotensin‐converting enzyme; Aldo, aldosterone; Ang I, angiotensin I; Ang II, angiotensin II; AT1, angiotensin II type 1 receptor; ATP, adenosine triphosphate; Ca2+, calcium; cAMP, cyclic adenosine monophosphate; DAG, diacylglycerol; Epac, exchange protein directly activated by cAMP; Gq, Gq protein alpha subunit; Gs, Gs protein alpha subunit; IP3, inositol triphosphate; JAK‐STAT, Janus kinase‐signal transducer and activator of transcription proteins pathway; MR, mineralcorticoid receptor; NE, norepinephrine; NOX2, NADPH oxidase 2; P, phosphorylation; PIP2, phosphatidylinositol 4,5‐bisphosphate; PKC, protein kinase C; PKC‐ε, epsilon isoform of protein kinase C; PLC, phospholipase C; PLC‐ε, epsilon isoform of phospholipase C; S, serine; SR, sarcoplasmic reticulum; α1‐AR, alpha‐1 adrenergic receptor; and β‐AR, beta‐adrenergic receptor; and δCaMKII, delta isoform of calcium/calmodulin‐dependent protein kinase II.

Evidence also supports crosstalk between AT1 receptor and cardiac β‐AR signaling, including the potential for heterodimerization of the G‐protein alpha subunits Gq and Gs downstream of AT1 signaling. , , As previously mentioned, sustained activation of β‐AR and AT1 receptors have been demonstrated to reduce IKs through distinct mechanisms. , , However, these signaling pathways have been demonstrated to share a common downstream signaling molecule, CaMKII, that is known to be upregulated in HF and has been implicated in HF pathophysiology. , The delta isoform of CaMKII (δCAMKII) is known to be activated downstream of AT1 receptor signaling in cardiomyocytes via a mechanism that involves oxidation of methionine 281 or methionine 282 on the enzyme. , , Conversely, sustained β‐AR signaling activates the delta isoform of CaMKII through a pathway that involves autophosphorylation at T287 downstream of Epac and PKC‐ε signaling. , , , Therefore, as described in Figure 2B, IKs reductions may also result from δCAMKII activation downstream of both AT1 and β‐AR signaling.

Finally, results from Lv et al suggest the potential for AT1 activation downstream of sustained aldosterone activation of the MR. Mechanistically, these results are supported by previous findings that prolonged activation of the MR by aldosterone resulted in increased AT1 receptor density and increased gene expression of ACE. , Both of these effects are predicted to increase signaling through AT1, as shown in Figure 2C, which has been consistently shown to contribute to pathologic cardiac remodeling processes in HF. , In addition, sustained MR activation has been independently linked to cardiac fibrosis and adverse HF outcomes. ,

Mechanisms of IKs Electrical Remodeling in HF

A multitude of investigations have explored potential mechanisms for how HF‐associated stimuli, including alterations in neurohormonal signaling, may pathologically regulate IKs. These mechanistic analyses have largely focused on (1) changes in the expression of KCNQ1, KCNE1, and other KCNE isoforms; (2) posttranslational regulation of KCNQ1 and KCNE1 subunits; and (3) effects on the interactions among IKs‐related proteins, including interaction of KCNQ1 and KCNE1 with each other and with various accessory proteins. Regardless of mechanism, the arrhythmogenic potential of reductions in IKs has been corroborated within investigations that have shown the development of ventricular arrhythmias during pharmacological inhibition of IKs and in canine and rabbit HF models wherein IKs is reduced. Below, we discuss the existing literature for each regulatory mechanism.

Changes in KCNQ1 and KCNE Expression in HF

Although IKs reductions and APD prolongation have been consistently observed clinically and in cellular and animal HF models, the mechanism supporting IKs reduction remains elusive. In particular, changes in the expression of KCNQ1 and KCNE1 within animal HF models and samples from patients with HF have shown conflicting results, as summarized in Table 2. The results from animal studies, in which HF has been simulated by atrioventricular blockade, tachycardiac pacing, coronary artery occlusion, coronary artery microembolization, or treatment with the compounds isoproterenol (β‐AR agonist) or aldosterone (MR agonist), have generally been mixed. Multiple studies have found reductions in mRNA and protein expression of KCNQ1 and KCNE1, while others have demonstrated no change in expression. , , , , , , , , In contrast, human studies that have analyzed ventricular tissue explanted from patients with HF have demonstrated either no change or increases in the mRNA expression of KCNQ1 and KCNE1 relative to healthy controls. ,

Table 2

Studies Assessing Changes in the Expression of KCNQ1 and KCNE1 in Animal or Human Models of HF

HF Simulation Method	Experimental Model	Percent Change in mRNA Expression		Percent Change in Protein Expression		Authors
		KCNQ1	KCNE1	KCNQ1	KCNE1
Chronic aldosterone treatment	Guinea pig left ventricle	↓60%*	↓60%*	↓40%*	↓40%*	Lv et al 45
Chronic atrioventricular block	Canine ventricle	↓80%	↓70%	↓50%	↓60%	Ramakers et al 34
Chronic atrioventricular block + bradycardiac pacing	Rabbit left ventricle	↓53%	↓43%	↓53%	↓33%	Tsuji et al 21
Chronic atrioventricular block + bradycardiac pacing	Rabbit right ventricle	↓76%	↓58%	↓72%	↓37%	Tsuji et al 21
Chronic atrioventricular block + tachycardiac pacing	Rabbit left ventricle	↓51%	↓43%	↓53%	↓37%	Tsuji et al 21
Chronic atrioventricular block + tachycardiac pacing	Rabbit right ventricle	↓64%	↓74%	↓78%	↓45%	Tsuji et al 21
Chronic ischemia (coronary artery occlusion)	Rabbit myocardium	↔	↔	↓80%	NS†	Guo et al 77
Chronic ischemia (coronary microembolization)	Canine myocardium	NS	NS	↔ ‡	↓37%	Liu et al 76
Chronic isoproterenol treatment	Guinea pig ventricle	↔	↓45%	↔	↓51%	Aflaki et al 15
Chronic isoproterenol treatment	Guinea pig ventricle	↔	NS	↔	NS	Soltysinska et al 74
Tachycardiac pacing	Canine left ventricle	↔	↔	↔	↔	Akar et al 78
Tachycardiac pacing	Rabbit ventricle	↔	↔	↔	↔	Rose et al 75
None (clinical heart failure)	Human left ventricle	↑160%	NS	NS	NS	Borlak et al 79
None (clinical heart failure)	Human left ventricle	↑400%*	↑400%*	NS	NS	Lundquist et al 81
None (clinical heart failure)	Human right ventricle	↑400%	NS	NS	NS	Borlak et al 79
None (clinical heart failure)	Human right ventricle	↔	↑25%	NS	NS	Watanabe et al 32

↑ indicates increased expression; ↓, decreased expression; ↔, no significant change in expression; HF, heart failure; and NS, not studied.

Numerical results for the change in expression of KCNQ1 and KCNE1 were not presented in the Lv et al and Lundquist et al articles. The presented percentages were estimated based on the presented graphical results.

Because of limitations in detecting KCNE1 within protein immunoblot experiments, Guo et al did not report results for changes in KCNE1 protein expression.

Liu et al found no change in expression of the predominant KCNQ1 isoform (KCNQ1.1), but did observe an increase in expression of a truncated KCNQ1 isoform (KCNQ1.2), which they predicted to inhibit IKs in a dominant‐negative fashion.

Additionally, multiple KCNE isoforms, including KCNE2, KCNE3, KCNE4, and KCNE5, are expressed in human ventricular cardiomyocytes and may form heteromeric channels with KCNQ1 and KCNE1 to alter IKs. , , Coexpression of KCNE2, KCNE4, and KCNE5 along with KCNQ1 and KCNE1 reduce IKs amplitude, whereas coexpression of KCNE3 with KCNQ1/KCNE1 has been shown to enhance IKs amplitude. , , While data describing HF‐associated changes in these KCNE isoforms are sparse, one investigation found that mRNA expression of KCNE2 and KCNE5 were reduced approximately 3‐ and 4‐fold, respectively, in explanted left ventricular tissue from patients with dilated or ischemic cardiomyopathy relative to healthy tissue. In contrast, mRNA expression of KCNE3 and KCNE4 were enhanced approximately 5‐ and 2‐fold, respectively, in the cardiomyopathic left ventricle.

In summary, reductions in mRNA and protein expression of KCNQ1, KCNE1, and other KCNE isoforms in patients with HF and in animal HF models have not been consistently reproduced. These inconsistent findings are likely attributable to differences related to the HF model and the method of HF simulation. Regardless, these inconsistencies, along with the fact that the limited clinical data do not support reduced expression of KCNQ1 and KCNE1, suggest that alternative or additional molecular mechanism may mediate reduced IKs in HF.

Posttranslational Regulation of IKs in HF

A number of investigations have assessed the potential for aberrant posttranslational regulation of KCNQ1 and KCNE1 subunits to mediate reductions in IKs in response to HF‐associated changes in neurohormonal signaling. These studies have identified regulatory sites on the intracellular amino and carboxyl termini of KCNQ1 and the intracellular carboxyl terminus of KCNE1 with functional implications for IKs. Figure 3 diagrams the KCNQ1 and KCNE1 subunits, showing demonstrated sites of posttranslational regulation in response to HF‐related stimuli.

Figure 3

Demonstrated sites of posttranslational regulation on KCNQ1 and KCNE1.

Subunit diagram of KCNQ1 and KCNE1 showing sites of post‐translational regulation, the type of posttranslational regulation, and the signaling molecules involved in response to HF‐related stimuli. E3 ligase indicates E3 ubiquitin ligase (the specific ligase responsible for SUMO linkage to KCNQ1 K424 has not been elucidated); L, leucine; NEDD4‐2, neural precursor cell expressed developmentally down‐regulated 4‐like; P, phosphorylation; PKA, protein kinase A; PKC, protein kinase C; PKC‐ε, epsilon isoform of protein kinase C; S, serine; SUMO, small ubiquitin‐like modifier; T, threonine; Ub, ubiquitination; Y, tyrosine; and δCaMKII, delta isoform of calcium/calmodulin‐dependent protein kinase II.

Biology of KCNQ1

The KCNQ1 subunit is 676 amino acids in length and contains a 120‐residue intracellular amino terminus and a 325‐residue intracellular carboxyl terminus. These termini are separated by a transmembrane region with 6 membrane‐spanning domains, with the region connecting the fifth and sixth segments serving as the pore and multiple positively charged residues in the fourth segment serving as the voltage sensor. The canonical IKs channel is formed when 4 KCNQ1 subunits coassemble via protein‐protein interactions with 2 KCNE1 auxiliary subunits. The intracellular amino and carboxyl termini of KCNQ1 contain residues and domains that are critical for channel function, including domains that affect (1) the biophysical properties of IKs current, (2) coassembly of KCNQ1 tetramers, (3) trafficking of KCNQ1 channels, and (4) interaction of KCNQ1 subunits with KCNE1 and other interacting proteins and molecules (eg, PIP2 and CaM). , , , , Given the established functional importance of the intracellular domains of KCNQ1, multiple investigations have assessed the potential for experimental models of HF to regulate the intracellular regions of KCNQ1 and thereby affect IKs.

Posttranslational Regulation of KCNQ1 in HF

Posttranslational regulation of the intracellular amino and carboxyl termini of KCNQ1 has been demonstrated to regulate IKs in cellular and animal models in response to HF‐related signaling cascades. Specifically, the KCNQ1 amino terminus contains residues, including S27 and possibly S92, that mediate activation of IKs during acute β‐AR stimulation via a well‐characterized signaling complex that includes PKA, alpha‐kinase anchoring protein 9 (also called yotiao), protein phosphatase 1, phosphodiesterase 4D3, and adenylate cyclase 9. , , , , However, it is has been consistently demonstrated that responsiveness to β‐AR signaling becomes reduced or “uncoupled” during sustained β‐AR stimulation by a mechanism that involves downregulation of Gs; , , these findings suggest that PKA‐mediated enhancement of IKs during acute β‐AR stimulation, which plays a critical physiologic role in shortening ventricular repolarization during rapid heart rates, is diminished during HF and may contribute to pathological electrical remodeling. Site‐specific phosphorylation of the KCNQ1 amino terminus has also been shown to regulate IKs in response to enhanced angiotensin II signaling. Acute increases in angiotensin II signaling alter IKs via a signaling cascade that involves PKC, with a recent investigation by Gou et al implicating the PKC‐ε (epsilon isoform of protein kinase C) as the regulator via a pathway that involves phosphorylation of S95, T96, or both on the amino terminus of KCNQ1 and a carboxyl residue (S102) on KCNE1. It should be restated that the implications of acute angiotensin II exposure on IKs have demonstrated conflicting results, with 2 studies demonstrating IKs enhancement , and 3 others finding IKs reductions. , ,

Posttranslational regulation of the KCNQ1 carboxyl terminus in response to HF‐related signaling has also been described. Sustained β‐AR stimulation was recently shown to reduce IKs via a mechanism that involves δCAMKII‐mediated phosphorylation of S484 on the KCNQ1 carboxyl terminus. Ubiquitination of a PY motif by the E3 ubiquitin‐protein ligase NEDD4‐2 (neural precursor cell expressed developmentally downregulated 4‐like) in the distal KCNQ1 carboxyl terminus (L659‐Y662) has also been demonstrated to reduce IKs via downregulation of KCNQ1 plasma membrane expression. , More recently, findings by Kurakami et al elucidated the signaling pathway involved in KCNQ1 ubiquitination, which involves AMP kinase and PKC and occurs secondary to acute α1‐AR activation. The effects of sustained α1‐AR activation on KCNQ1 ubiquitination and plasma membrane expression have not been studied. Finally, an investigation by Xiong et al has identified the SUMO (small ubiquitin‐related modifier) protein as a key regulator of activation gating. “SUMOylation,” the process by which an interacting protein is covalently bound by a SUMO protein, has been demonstrated at K424 on each KCNQ1 subunit (incorporating up to four SUMO proteins in a functional IKs channel) in a process that is KCNE1‐dependent. The regulatory role of SUMOylation on IKs may be of significance in HF since past investigations have found that SUMO proteins, and thereby the process of SUMOylation, are downregulated in HF; decreased SUMOylation in HF has been shown to result in pathological changes in processes involved in cardiac contractility and stress adaptation, including specific pathological regulation of the sarco/endoplasmic reticulum Ca2+‐ATPase. ,

Biology of KCNE1

The KCNE1 subunit is 129 amino acids in length and contains a 44 residue extracellular amino terminus and a 58 residue intracellular carboxyl terminus. These termini are separated by a single membrane‐spanning transmembrane domain. Functional interaction with KCNQ1 involves multiple locations within the transmembrane domain and carboxyl terminus of KCNE1 and produces the characteristic properties of physiologic IKs, including slowing of channel activation, increasing conductance, and reducing or eliminating channel inactivation relative to currents produced by KCNQ1 alone. , , Based on the necessity of KCNQ1‐KCNE1 interaction for proper channel function, investigations have assessed the functional implications of posttranslational modifications of KCNE1 in response to HF‐related signaling pathways.

Posttranslational Regulation of KCNE1 in HF

Multiple investigations have elucidated a role for α1‐AR signaling in mediating reductions in IKs via a mechanism that involves PKC‐dependent phosphorylation of KCNE1 S102. , While the exact isoform of PKC that phosphorylates S102 has not been fully elucidated, it has been demonstrated to be a calcium‐dependent PKC isoform, such as alpha (PKC‐α), which is the predominant isoform expressed in the heart, or beta (PKC‐β), which is upregulated in HF. , , Regardless of the specific PKC isoform involved, IKs reduction in response to α1‐AR stimulation mechanistically involves enhanced clathrin‐ and dynamin‐dependent endocytosis of plasma membrane‐bound KCNQ1‐KCNE1 channel complexes. ,

HF Effects on Interaction Among IKs‐Related Proteins

In addition to changes in KCNQ1 and KCNE1 expression or posttranslational subunit regulation, HF‐associated changes in neurohormonal signaling have also been demonstrated to modulate the interaction of KCNQ1 and KCNE1 with each other and other interacting proteins. These alterations may have important implications for IKs in HF.

HF‐associated Modulation of the KCNQ1‐KCNE1 Subunit Interaction

HF‐associated signaling changes may modulate the protein‐protein interaction between KCNQ1 and KCNE1 subunits and thereby reduce IKs. Work by Dvir et al, which investigated the mechanisms for how long QT syndrome mutations in the KCNQ1 and KCNE1 carboxyl termini affect IKs, demonstrated an important role for PIP2 in facilitating interaction of KCNQ1 and KCNE1. The importance of PIP2 in regulating IKs was also demonstrated by Kienitz et al. Their work found IKs reductions following sustained α1‐AR stimulation via enzymatic depletion of cellular PIP2 through a pathway involving chronic PKC activation, which is a known pathologic regulator in HF. , Accordingly, HF‐associated increases in SNS signaling are predicted to disrupt the KCNQ1‐KCNE1 subunit interaction by depleting cellular stores of PIP2, thereby reducing IKs.

HF‐associated Modulation of KCNQ1‐KCNE1 Interaction With Other Interacting Proteins

HF‐associated changes in neurohormonal signaling have also been demonstrated to impact the functional interaction between KCNQ1‐KCNE1 complexes and other interacting proteins necessary for channel function. For instance, an investigation by Aye et al demonstrated that, relative to tissue from control hearts, explanted heart tissue in patients with HF showed impaired interaction of PKA with alpha‐kinase anchoring protein 9, thus potentially blunting IKs via PKA‐mediated (and alpha‐kinase anchoring protein 9–dependent) phosphorylation of KCNQ1 S27 during acute or intermittent rises in adrenergic signaling. Reductions in alpha‐kinase anchoring protein 9–dependent IKs responsiveness to acute β‐AR stimulation are predicted to exacerbate an already decreased responsiveness to β‐AR stimulation in HF (ie, because of PKA uncoupling secondary to reduced Gs expression). Together, these HF‐mediated signaling adaptations are predicted to act in concert to impart an increased risk of QT prolongation and ventricular arrhythmogenesis.

HF‐associated changes may also influence interaction of KCNQ1 with CaM, a calcium‐binding second messenger that is a regulator of IKs inactivation gating and channel assembly. Specifically, CaM has been shown to interact with KCNQ1 via 2 CaM‐binding domains within helices A and B of the KCNQ1 carboxyl terminus in a process that is required for channel tetramerization. , Given the essential role of CaM in regulating IKs, reduced expression of CaM, as has been consistently demonstrated in HF, may be involved in the pathologic regulation of IKs in HF. ,

Conclusion

HF imparts an enormous health burden, and cardiac electrical remodeling underlies a major mechanism of HF‐related pathophysiology. , HF is associated with alterations in neurohormonal signaling, including sustained elevations in SNS and RAAS signaling, that prolong APD and predispose to arrhythmogenic SCD via pathologic regulation of ion channels, including IKs. , Herein, we present the current experimental evidence related to the regulation of IKs during normal physiology and in response to HF‐associated stimuli. Specifically, we focus on (1) the cellular receptors and intracellular pathways involved in HF pathophysiology and (2) the specific mechanisms through which HF‐associated neurohormonal signaling impact IKs. By aggregating findings from the basic science and clinical literature, we provide a comprehensive review of both the pathologic regulation of IKs in HF and the resulting clinical implications on electrical instability and arrhythmogenesis.

Although this review summarizes the literature related to pathological remodeling to alter IKs, HF is also known to pathologically regulate a wide variety of ion channels and calcium‐handling proteins, which contribute to HF‐associated ventricular arrhythmogenesis. Past investigations have convincingly demonstrated pathological regulation of ventricular sodium (eg, late INa), calcium (eg, ICa,L), and potassium currents (eg, the transient outward potassium current, IKr, the inward rectifier potassium current) along with calcium‐handling proteins, including the ryanodine receptor, the sodium‐calcium exchanger, and sarco/endoplasmic reticulum Ca2+‐ATPase, in HF. , , , , , , HF‐associated changes in affected ionic currents are summarized in Figure 1C. Given that HF results in deficits in other repolarizing ionic currents, including IKr and the inward rectifier potassium current, the proarrhythmogenic potential of reductions in IKs is enhanced in HF. , , Indeed, it has been demonstrated that HF‐associated reductions in IKs contribute to the proarrhythmogenic effects of other calcium‐related ion carriers that are activated in response to β‐AR stimulation (ie, ICa,L, the sodium‐calcium exchanger). , Therefore, the enhanced ventricular electrical instability and arrhythmogenesis characteristic of HF stem from complex interaction of multiple ionic currents that undergo pathological remodeling in HF.

While the therapeutic success of targeting SNS and RAAS signaling in HF is well documented, , progress toward the understanding of HF pathophysiology continues to elucidate novel potential therapeutic targets. In particular, advances in basic science research have revealed putative therapeutic targets downstream of membrane‐bound receptors that have the potential to overcome challenges associated with signaling pathway crosstalk, such as Epac and CaMKII. , As detailed in this review, the potential benefits of these promising therapeutic targets may be, at least in part, attributable to their ability to prevent or ameliorate pathological IKs remodeling. Further research is needed to explore the therapeutic potential of these novel targets in preventing and improving clinical outcomes in HF, including arrhythmogenic SCD.

Sources of Funding

This work was supported in part by a grant from the National Institutes of Health, National Heart, Lung, and Blood Institute; R03 HL141619.

Disclosures

None.