Late I_Na in the Heart: Physiology, Pathology, and Pathways

In this issue of Circulation, Hund and colleagues¹ make an important contribution to understanding the physiological and pathophysiological roles of late sodium current (I_Na) in the heart, with a focus on a key pathway regulating late I_Na amplitude. They conducted well designed and detailed studies with two new genetically engineered mouse lines: one a S571A mouse that ablates phosphorylation by Ca²⁺/calmodulin-dependent kinase II (CaMKII)² selectively at serine 571 in the cardiac Na channel pore-forming protein Scn5a, and the other a S571E mouse that mimics phosphorylation at serine 571. Serine 571 was shown previously to be a target for phosphorylation by CaMKII and this phosphorylation enhanced late I_Na². The present studies in “knock-in” mice expressing either S571A or S571E have distinct advantages over other earlier studies in heterologous expression systems including cultured myocyte models because they allow the study of whole animal and organ phenotypes, as well as the study of cellular and molecular biophysical properties in a more native environment. What these new in vivo studies reveal is that despite the extensive network of CaMKII targets, phosphorylation of S571 selectively regulates late I_Na and in particular enhanced late I_Na in failing heart.

Peak I_Na is the large inward current flowing mainly through the cardiac Na⁺ channel pore formed by Scn5a, which is part of a larger sodium channel macromolecular complex. Members of this macromolecular complex act to localize the complex and regulate I_Na³. With the onset of the action potential (AP) in the myocardium the peak I_Na rapidly rises and decays to nearly zero over several ms. This I_Na spike underlies excitability and conduction in working myocardium and the Purkinje conduction system. In contrast to peak I_Na, late I_Na is a small inward current, usually less than 0.5% of peak I_Na, that flows throughout the action potential (AP) plateau. Although the amplitude of late I_Na is small, it plays a role to maintain the AP plateau because competing repolarizing potassium currents are also small. Increased late I_Na can directly affect cardiac electrophysiology by prolonging refractoriness and predisposing to triggered activity as early after-depolarizations or EADs, observed clinically as long QT arrhythmia. Because late I_Na flows for much longer time than peak I_Na (~300 to 400 ms for late I_Na) it is predicted to play a greater role in Na⁺ loading than peak I_Na⁴. Increased Na⁺ loading increases intracellular Ca²⁺ levels through effects on Na⁺-Ca²⁺ exchange, and thereby affects contractility and relaxation⁵. Increased intracellular Ca²⁺ levels affect the electrophysiology of the cell via a number of mechanisms including delayed after-depolarizations, or DADs. Late I_Na is increased under many conditions including inherited disorders such as in inherited long QT syndromes (LQT 3, 9, 10, 12), and also in acquired conditions, as in hypertrophy, heart failure, ischemia and diabetes, where it plays roles in the pathogenesis of arrhythmia, heart failure, and angina⁶ and it has attracted much attention as a therapeutic drug target^7,8,9. Therefore understanding the properties and pathways regulating late I_Na has the potential to help us understand the pathogenesis and provide avenues for treatment of many disease processes in clinical cardiology.

In this commentary we consider key unanswered and partially answered questions about late I_Na, and discuss how the genetically engineered mice developed and characterized by Hund and colleagues¹ have addressed or could be used to address them.

What are the signaling pathways that regulate late I_Na? How do they interact?

Two pathways that enhance late I_Na act by post-translational modification of Scn5a involve CaMKII dependent phosphorylation¹⁰ and nNOS dependent nitrosylation.¹¹ A third pathway that may involve direct phosphorylation of Scn5a or other regulatory protein, involves phosphoinositide 3-kinase (PI3K), which acts to suppress late I_Na.¹² The CaMKII pathway is presently the most studied and best defined with the key phosphorylation site affecting late I_Na known to be S571. The nNOS pathway appears to involve direct nitrosylation of the Scn5a channel, but the Cys site(s) have not yet been determined. Whether and how these different pathways interact are unknown. Are they independent and additive? Do they share common features? It is not known whether or not the PI3K pathway acts directly by phosphorylation of Scn5a¹², or whether it may somehow involve the CaMKII or nNOS or other pathways. Although these questions were not directly addressed in the present study, it is interesting to note that the S571A mouse retains a signficant proportion of WT late I_Na (Fig. 2 in Glynn et al. ¹) suggesting a componenet of late I_Na that is not regulated by the S571 site. The S571 mouse models should be useful to address other questions about pathway interactions. For example, would inhibition of the PI3K pathway result in an increase in late I_Na in the S571A model? If not, this would support the idea that PI3K activation ultimately acts by supressing phosphorylation of S571. In addition to the above named pathways PKC-dependent phosphorylation at S1503 altered I_Na kinetics in a way that enhances a type of late I_Na called “window current”¹³ and PKC inhibition blocked increased late I_Na that was caused by calcium loading the cell ¹⁴, suggesting roles for PKC that may be interact with the CaMKII pathway. The S571A and S571E models will be useful tools to further define the relationships and relative importance among these pathways.

What signaling pathways are involved in inherited and acquired diseases with increased late I_Na?

Enhanced late I_Na occurs in numerous inherited cardiac disorders (mutations in the Scn5a complex for LQT3, LQT9. LQT19, LQT12), and in acquired conditions (hypertrophy, heart failure, ischemia, diabetes), and can also arise due to changes in metabolites and other molecules (acidosis, carbon monoxide (CO), reative oxygen species (ROS) and drugs (PI3K inhibitors)) ⁹. The detailed mechanisms for the causes of late I_Na in disease and the pathways involved have been investigated in only a few of these diseases. Activation of the nNOS pathway to increase late I_Na has been implicated in the pathogenesis of LQT9 involving caveolin3 mutations¹⁵ and LQT12 involving α1-syntrophin mutations¹¹. An important finding in the present study¹ is that stress induced heart failure in the S571A mouse failed to develop the increased late I_Na seen in wild-type mice¹, strongly supporting the idea that phosphorylation of S571 via the CaMKII pathway is required for the late I_Na in this model of heart failure. Further experiments in this model could extend these important insights about the role of the CamKII pathway in causes of late I_Na. For example, would late I_Na be enhanced by ischemia, CO, ROS, PI3K inhibitors in the S571A mouse? If not, this would provide evidence that they all work through the CaMKII pathway.

What are the biophysical and structure function mechanisms for late I_Na?

Overall the detailed biophysical mechanism(s) for late I_Na at the level of Scn5a are not clear. The prevailing idea is that it involves the inactivation gate on the D3–4 linker, or in the inactivation receptor involving the S4–5 linker and residues on S5. But LQT3 mutations occur over much of the Scn5a topography. A lack of “hot spots” suggests complexity in the structures affecting complete inactivation of I_Na. In particular, how does phosphorylation of S571 increase late I_Na? Is it linked in some way to inactivation structures? While the present study does not address these structure-function and biophysical issues directly, the finding that late I_Na can be regulated by phosphorylation at S571 independently of other gating changes is of interest, and must be accounted for by any proposed structure-function model. Two LQT3 mutations close to S571 were postulated to cause late I_Na by mimicking the charge near this site ¹⁶ and provide additional clues to the structure-function of late I_Na. Other nearby CaMKII-dependent phosphorylation sites (S516 and T594) do affect kinetics of gating but do not appear to contribute specifically to late I_Na.¹⁷ These key and interesting findings might be investigated by additional electrophysiological studies including single channel analysis of the S571 knock-in mouse lines.

Another key observation was the rate-dependent decrease in late I_Na in these mice. This property was seen with the canonical LQT3 mutation delta KPQ¹⁸, and has been postulated to be protective. The late I_Na found in some Scn5a mutations, such as those found in sudden Infant death syndrome (SIDS)¹⁹, do not have this property and may account for greater lethality.

Is physiological late I_Na (in contrast to the enhanced late I_Na found in inherited and acquired disorders) important for regulating excitability and contractility in the absence of disease?

Most studies of late I_Na have been conducted in models of pathologically enhanced late I_Na, but late I_Na is also present in normal hearts (usually 0.2 to 0.5% of peak I_Na). Is this physiological late I_Na important for normal electrophysiology and regulation of contractility? Might drugs that block too much of the late I_Na lead to undesired effects? The recognition that late I_Na plays a role in normal cardiac physiology goes back ~50 years when it was shown that the selective Na⁺ channel blocker tetrodotoxin shortened the AP plateau in normal myocytes²⁰. But in normal rabbit and rat hearts specific late I_Na blockers had no important effects on contractility and conduction.²¹ Small but significant decreases in ejection fraction were observed in the S571A mouse¹, supporting the idea that CaMKII-dependent late I_Na has importance in regulating contractility in normal heart. But “S571A APD was not significantly different than WT at baseline” ¹ suggesting modest if any effects on electrophysiology. It is possible that the complex pathways for regulation of late I_Na evolved to deal exclusively with stress or pathological conditions, but it is more likely that these physiological levels of late I_Na are under tight regulatory control for important reasons related to normal cardiac physiology. More studies are needed on this less well studied issue.

Can pathologically enhanced late I_Na be better selectively targeted based on pathway mechanism?

The authors of the present study¹ emphasized how CaMKII-dependent S571 phosphorylation specifically regulates late I_Na and may represent an attractive target for blocking pathological late I_Na. Currently used drugs such as flecainide, amiodarone, and ranolazine are all pore blockers and would presumably block late I_Na regardless of mechanism of generating late I_Na⁶. They appear to get their selectivity to block late I_Na over peak I_Na because of state-dependent block. Even more selective blockers of late I_Na are in development²². Regulating the phosphorylation S571 could in theory be a novel and important specific regulator of late I_Na but it is not yet clear if small molecules can be found for this target.

Caveats and importance

As the authors of the present study correctly point out, a mouse model may not be completely translatable to human physiology. Despite this limitation this study¹ has generated insights into unanswered questions about the regulatory pathways and characteristics of both pathological and physiological late I_Na, and the models developed in these studies have the potential to generate more insights.