Targeting Late Na⁺ Current: Too Late or Better Late than Never?

Ventricular arrhythmias are associated with significant morbidity and mortality in both ischemic and non-ischemic heart failure (HF) patients. Even with catheter ablation, many patients require anti-arrhythmic medication to prevent defibrillator shocks yet currently available drug options are not ideal¹. Despite tremendous progress made in discovering genetic associations with disease and unraveling complex molecular mechanisms and signaling pathways, there have been no new anti-arrhythmic drugs approved in the US since 2009. In this issue of the Journal, Hegyi et al.² investigate an old arrhythmic target, the late sodium current (I_Na,L), and its regulation by a novel class of channel interacting proteins called fibroblast growth factor homologous factors (FHF). They also report the action of a protective FHF isoform-derived peptide called FixR (FHF inhibiting x Region) as a potential anti-arrhythmic therapeutic strategy.

What is FHF?

FHF is a sub-family of Fibroblast Growth Factor (FGF)³ that has several intriguing properties. Unlike FGF, FHF is largely not secreted and, thus, relies on intracellular signaling, which may benefit selectivity. Importantly, FHF interacts with the C-terminal domain of the pore-forming Na⁺ channel α-subunit, a region that is rife with sites for post-translational modifications and mutations that are associated with arrhythmia. This interaction can affect multiple Na⁺ channel gating properties including inactivation³. There are four FHF genes (FHF1–4) that vary in expression between species where, for example, FHF1 is predominant in humans and FHF2 in mice. Most interesting, FHF isoforms undergo alternative splicing at the N-terminus, and expression of these splice-variant isoforms differ in species and have distinct action on I_Na,L.

What is already known about FHF and its action on I_Na,L?

As a reminder, I_Na is responsible for the rapid action potential upstroke in cardiac and neuronal cells. Disruption of this channel, in particular, impaired inactivation can lead to a depolarizing late Na⁺ current (I_Na,L) during the action potential plateau that can prolong repolarization, which by itself is a potent substrate for arrhythmia. Making matters worse, the increase in late Na⁺ entry increases Ca²⁺ entry due to reverse mode Na⁺/Ca²⁺ exchanger, which can promote Ca²⁺ overload (also an arrhythmia substrate), and other harmful sequela such as mitochondrial dysfunction, oxidative stress, and disruptive post-translational modifications. Increased I_Na,L is associated with long QT type 3 (LQT3) mutations and also acquired disease. In humans, mutations of FHF1 can cause ventricular arrythmias⁴ and neurological disorders⁵. Prior to the study by Hegyi et al.² in this issue of the Journal, the action of different isoform/splice variants were shown to “tune” I_Na,L⁶. For example, in humans, the long FHF isoforms can reduce I_Na,L (“protective” isoforms), whereas the short FHF isoforms do not exhibit potent I_Na,L inhibition⁶. In that same prior study, it was also shown that FixR, a peptide that mimics a protective FHF isoform, can mitigate I_Na,L due to mutations in LQT3 and calmodulin interaction sites in mice and human cardiac myocytes. Whether FHF isoforms are germane in acquired heart disease associated with increased I_Na,L such as HF, and if this can be effectively targeted to prevent arrhythmia were investigated by Hegyi et al.² in this issue of the Journal.

Are results similar for acquired disease that have increased I_Na,L?

Hegyi et al.² rigorously tested three different acquired models of HF including HFrEF in rabbits with aortic insufficiency/constriction, HFrEF in mice with transverse aortic constriction (TAC), and HFpEF in Lepr^db/db mice with infusion of d-aldosterone, along with testing FHF isoform expression in human HF tissue samples. While previous studies suggest FHF expression is altered in HF^7,8, Hegyi et al.² showed specifically that the protective, I_Na,L inhibitor, isoforms are reduced in rabbits, mice and, importantly, humans, with HF. All the acquired models they tested have increased I_Na,L, and in all cases FixR was able to mitigate I_Na,L in vitro. Finally, in a separate genetic model of HFrEF mice due to overexpression of CaMKIIδC, they also show FixR markedly inhibited I_Na,L, despite less change in FHF compared to TAC mice. These results show that a protective FHF isoform can mitigate increased I_Na,L that is associated with acquired HF and CaMKII activity, which may broaden its therapeutic application beyond mutations alone.

Can FixR mitigate arrhythmias in acquired disease?

The models of acquired disease tested by Hegyi et al.² in this issue of the Journal demonstrate multiple cellular arrhythmia substrates in vitro including APD prolongation, delayed after depolarizations (DAD), short-term variability (STV) of APD, and APD alternans. These are well established substrates for arrhythmia due to both abnormal impulse formation (e.g., DADs) and conduction (e.g., APD alternans). FixR was able to mitigate most of these, and GS967, a selective I_Na,L inhibitor, had similar action on cellular arrhythmia substrates. These results are, however, from isolated cells, where it is impossible to demonstrate arrhythmia substrates associated with reentrant excitation that is common in HF, such as slow conduction, spatial heterogeneities, and discordant alternans⁹. To address this, the mice overexpressing CaMKIIδC that exhibit ventricular arrhythmia were also tested in vivo. Without any treatment, the ECG demonstrated a wide QRS, QTc prolongation, and atrial and ventricular arrhythmic events that were further aggravated by isoproterenol and caffeine. Adeno-associated virus 9 (AAV9) encoding FixR was administer to CaMKIIδC mice and WT controls. Importantly, FixR in CaMKIIδC mice attenuated QTc prolongation, and resulted in less arrhythmias as well as reducing mortality following isoproterenol and caffeine treatments. There were no changes in ECG parameters in WT controls with FixR, further demonstrating its selectivity. It is also interesting that FixR reduced atrial arrhythmias. Details are sparse, but atrial conduction has been shown to be impaired in FHF deficient mice⁸ and Ranolazine can suppress AF¹⁰. Thus, FixR can mitigate arrhythmia substrates associated with acquired disease in vitro, and it demonstrates some atrial and ventricular anti-arrhythmic behavior in vivo.

How does FixR compare with clinical drugs that target I_Na,L?

Ranolazine, originally developed as an anti-anginal and anti-ischemic metabolic effector, also inhibits I_Na,L, and in animal models it was effective in suppressing arrhythmia¹¹. Hegyi et al.² and others⁶ showed FixR is much more potent than Ranolazine. Furthermore, in both control and HFrEF rabbit ventricular myocytes they show that FixR has no effect on transient Na⁺ current (I_Na,T), the L-type Ca²⁺ current (I_Ca,L), transient outward K⁺ current (I_to), rapid (I_Kr) and slow (I_Ks) delayed rectifier K⁺ current, and the inward rectifier K⁺ current (I_K1). In contrast, Ranolazine is known to inhibit peak I_Na, I_Kr, and the L-Type calcium current (I_Ca,L)¹². The superior selectivity of FixR might be due to, in part, its action on intracellular cytoplasmic domains⁶. This is key, because non-specific effects can raise safety concerns, which is a major reason why some anti-arrhythmic drugs never progress¹³.

Limitations

The ventricular arrhythmias shown Hegyi et al.² (largely PVCs) are often triggers of fatal VT/VF in vulnerable substrates such as HF, however the mechanism by which FixR tended to reduce mortality when mice were challenged with caffeine and isoproterenol is not demonstrated. Additionally, it is important to remember that there are many well-established causes of arrhythmia in patients with HF that are probably unrelated to I_Na,L including, for example, downregulation of Ca²⁺ handling proteins, reduced repolarization reserve, and altered cell-to-cell coupling. Consequently, even with perfect I_Na,L blockade, complete arrhythmia suppression in HF may not be achieved. For a more realistic assessment of FixR, an essential next step would be to perform preclinical studies in animal models that better represent fatal arrhythmia observed in humans that are adequately powered to assess clinical outcomes.

Final thoughts

Inhibition of I_Na,L by FixR has several promising characteristics that may make it very appealing for further testing including broad targeting (mutations and acquired disease), high potency (efficacy), and selectivity (safety). Compared to available drugs that target I_Na,L such as Ranolazine, FixR appears to be superior in all of these aspects. In addition, compared to Ranolazine, FixR mechanism of action is better understood^3,6, which is critical for moving forward. While intracellular delivery of FixR can be challenging, several options are promising including cell-penetrating peptides such as trans-activating transcriptor (TAT)⁶. Finally, in addition to the therapeutic implications, Hegyi et al.² in this issue of the Journal, further develop mechanistic insight on FHF splice isoform-dependent regulation of I_Na.

Is this enough to move the needle on anti-arrhythmic drug progress? It’s still too early to tell as additional studies in more clinically realistic models are needed. Yet, even if efficacy and safety are clearly demonstrated in such models, additional barriers including pharmaceutical buy in and regulatory approval may be challenging¹³. Given that the clinical experience with Ranolazine as a stand-alone anti-arrhythmic has been somewhat disappointing¹⁴, is it too “late” for any new I_Na,L inhibitors to be considered? Maybe not, because the novel targeting of I_Na,L demonstrated by Hegyi et al.² is very encouraging, and the new mechanistic insight is very intriguing, so probably better “late” than never.