CaMKII-dependent late Na⁺ current increases electrical dispersion and arrhythmia in ischemia-reperfusion

Abstract

The mechanisms underlying Ca²⁺/calmodulin-dependent protein kinase II (CaMKII)-induced arrhythmias in ischemia-reperfusion (I/R) are not fully understood. We tested the hypothesis that CaMKII increases late Na⁺ current (I_Na,L) via phosphorylation of Na_v1.5 at Ser⁵⁷¹ during I/R, thereby increasing arrhythmia susceptibility. To test our hypothesis, we studied isolated, Langendorff-perfused hearts from wild-type (WT) mice and mice expressing Na_v channel variants Na_v1.5-Ser571E (S571E) and Na_v1.5-Ser571A (S571A). WT hearts showed a significant increase in the levels of phosphorylated CaMKII and Na_v1.5 at Ser⁵⁷¹ [p-Na_v1.5(S571)] after 15 min of global ischemia (just before the onset of reperfusion). Optical mapping experiments revealed an increase in action potential duration (APD) and APD dispersion without changes in conduction velocity during I/R in WT and S571E compared with S571A hearts. At the same time, WT and S571E hearts showed an increase in spontaneous arrhythmia events (e.g., premature ventricular contractions) and an increase in the inducibility of reentrant arrhythmias during reperfusion. Pretreatment of WT hearts with the Na⁺ channel blocker mexiletine (10 μM) normalized APD dispersion and reduced arrhythmia susceptibility during I/R. We conclude that CaMKII-dependent phosphorylation of Na_v1.5 is a crucial driver for increased I_Na,L, arrhythmia triggers, and substrate during I/R. Selective targeting of this CaMKII-dependent pathway may have therapeutic potential for reducing arrhythmias in the setting of I/R.

NEW & NOTEWORTHY Ca²⁺/calmodulin-dependent protein kinase II (CaMKII) phosphorylation of Na_v1.5 at Ser⁵⁷¹ leads to a prolongation of action potential duration (APD), increased APD dispersion, and increased arrhythmia susceptibility after ischemia-reperfusion in isolated mouse hearts. Genetic ablation of the CaMKII-dependent phosphorylation site Ser⁵⁷¹ on Na_v1.5 or low-dose mexiletine (to inhibit late Na⁺ current) reduced APD dispersion, arrhythmia triggers, and ventricular tachycardia inducibility.

INTRODUCTION

Each year, ∼1,000,000 people suffer from myocardial infarction (MI) and an additional 700,000 patients experience cardioplegic arrest during completion of a cardiac procedure (38). While rapid return of normal coronary circulation is paramount to survival of these patients, the return of blood flow itself induces damage and arrhythmia (7, 14, 41). Therefore, it is of great importance to better understand the mechanisms underlying ischemia-reperfusion (I/R) injury. Intracellular Ca²⁺ overload secondary to reverse-mode Na⁺/Ca²⁺ exchanger activity plays a central role in tissue damage and arrhythmias after I/R (10). A critical step that leads to reverse-mode Na⁺/Ca²⁺ exchanger activity in I/R is an initial Na⁺ overload arising from a variety of sources, most notably the Na⁺/H⁺ exchanger (2, 17), although the precise sequence of events remains unknown. Normal cardiac function depends on voltage-gated Na⁺ (Na_v) channels, essential for the rapid action potential (AP) upstroke and an important source of Na⁺ entry into the cell (34). Under normal conditions, Na_v current (I_Na) rapidly activates and then inactivates to allow for AP repolarization. However, even under normal (nondiseased) conditions, a minor population of Na_v channels fails to inactivate, giving rise to a small Na⁺ current that persists throughout the AP [late Na⁺ current (I_Na,L)]. Importantly, amplification of I_Na,L in the setting of disease has been shown to increase arrhythmia susceptibility (33), including in I/R, where increased I_Na,L has been linked to elevated intracellular Na⁺ (45). At the same time, Ca²⁺/calmodulin-dependent protein kinase II (CaMKII) activity has been found to be upregulated during I/R, with possible proarrhythmic effects on several downstream targets (47), including the dominant cardiac Na⁺ channel isoform Na_v1.5 (54). Previous work from our group has shown that CaMKII phosphorylates Na_v1.5 at Ser⁵⁷¹ to increase I_Na,L and arrhythmia susceptibility (19, 22, 29).

Despite considerable work in this area, a potential role for direct CaMKII-dependent phosphorylation of Na_v1.5 has not been studied in I/R. To explore the importance of this pathway for I/R-induced arrhythmia, we use two knockin mouse models where the Ser⁵⁷¹ site on Na_v1.5 was either ablated (S571A) or constitutively activated [phosphomimetic (S571E)] (19). An important aspect of these models is that they allow for specific targeting of I_Na,L without altering other critical channel properties. We anticipate that results from this study will advance our understanding of the role of I_Na,L in I/R injury and identify new therapeutic strategies to reduce arrhythmias after I/R without altering normal physiology.

MATERIALS AND METHODS

Mouse models.

The following mouse models were used: Scn5a knockin mice (C57BL/6 background) with a S571E or S571A point mutation in Na_v1.5 or wild-type (WT) controls (19). Adult male mice were anesthetized by isoflurane; anesthesia was confirmed by the absence of response to probing of the lower extremities. Mice were intraperitoneally injected with heparin to avoid blood clots during heart extraction and cannulation. The whole heart was extracted and perfused via the aorta using a rolling pump at 2 ml/min and oxygenated Tyrode solution [containing (in mM) 140 NaCl, 1.0 MgCl₂·6H₂O, 1.2 NaH₂PO₄, 4.0 KCl, 1.8 CaCl₂·2H₂O, 5.6 glucose, and 10 HEPES sodium salt] maintained at 36.5°C and titrated to pH 7.4 using HCl. Experiments were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals following protocols that were reviewed and approved by the Institutional Animal Care and Use Committee of The Ohio State University.

Optical mapping of ex vivo heart preparations.

Optical mapping techniques were used to measure the membrane electrical potential on the anterior portion of the isolated heart. The heart was perfused for 15 min with blebbistatin (from a stock solution of 5 mg/ml in DMSO) to decrease motion for clearer signal production without significantly altering the electrophysiology (51). Perfusion continued with the voltage dye di-4-ANEPPS (4 μM) for 15 min (4, 56). Images were taken to ensure that the dye reached the extremity tissues of the organ. The fast voltage-sensitive dye was chosen based on previous work showing a good signal-to-noise ratio, negligible light interference, and reduced autofluorescence signals (which contribute to inaccurate readings) (5). Global ischemia was induced by stopping perfusion for 15 min; the heart was then reperfused with Tyrode solution for the remainder of the experiment.

To excite the dye, the LEX2-LZ4-G light source (SciMedia) was used in conjunction with a 531/40 filter. Signals were collected at 1,000 Hz (1 ms/frame) using MiCAM05 complementary metal oxide semiconductor cameras and filtered with a 600-nm long-pass filter. The area of the resulting images was 1.06 ×1.06 cm², with 100 × 100 pixels. A custom-designed MATLAB code (provided by Dr. Steven Poelzing) was used to obtain values from the signals measured during experimentation. AP duration at 80% repolarization (APD) was measured from optical signals and reported as an average value throughout the mapping area. APD spatial heterogeneity was estimated as the standard deviation of the mean value over the mapping area. Conduction velocity was also measured in the transverse and longitudinal directions, and activation maps were created using RHYTHM (18).

Electrocardiogram activity was measured using three leads placed around the temperature-controlled dish, adjacent to the heart, and was analyzed for spontaneous arrhythmia events [e.g., premature ventricular contractions (PVCs)] throughout the I/R protocol. Arrhythmia susceptibility was also assessed after 5 min of reperfusion using a S1S2 pacing protocol. Briefly, after 6 s of S1 pacing (150-ms cycle length), a S2 beat was delivered at an initial S1S2 interval of 100 ms and then decreased by 10 ms until 1:1 capture was not possible or ventricular tachycardia (VT) was observed. Pacing was performed using a unipolar lead at the center of the anterior heart wall and a stimulus amplitude equal to twice threshold at baseline (with amplitude held constant for the entire I/R period). VT in response to S1S2 pacing was defined as three or more nonpaced beats in rapid succession.

Immunoblot analysis.

Ventricular lysates were analyzed by SDS-PAGE and immunoblotting, as described elsewhere (19, 22, 23, 29). Equal protein loading was achieved using standard BCA protocols and verified by Ponceau staining of immunoblots. The following antibodies were used: total CaMKII (Badrilla), phosphorylated CaMKIIδ antibody (Santa Cruz Biotechnology), and total and phosphorylated Na_v1.5 (custom antibodies) (22).

Data and statistical analysis.

SigmaPlot 13.0 was used for statistical analysis. One-way ANOVA was used for multiple comparisons with Tukey’s test for post hoc testing. Values are means ± SE. Contingency data were analyzed using a χ²-test. The null hypothesis was rejected for P < 0.05.

RESULTS

Phosphorylation of Na_v1.5 at Ser⁵⁷¹ contributes to the generation of arrhythmias during I/R.

To test our hypothesis that increased CaMKII-dependent phosphorylation of Na_v1.5 promotes increased I_Na,L, Na⁺/Ca²⁺ overload, and arrhythmias during I/R, we subjected isolated Langendorff-perfused hearts to global ischemia followed by reperfusion (Fig. 1). We first measured levels of total/phosphorylated CaMKII and Na_v1.5 at baseline and during the I/R protocol in WT hearts but not in S571A and S571E hearts because of loss of the phosphorylated Na_v epitope in these models (19). Consistent with previous findings (47), autophosphorylated (activated) CaMKII was significantly increased after 15 min of ischemia. Na_v1.5 also showed a significant increase in phosphorylation at Ser⁵⁷¹ (CaMKII site) after ischemia (Fig. 2). Surprisingly, CaMKII and Na_v1.5 phosphorylation levels rapidly returned to baseline with reperfusion; normal values were restored by 5 min (Fig. 2). These results demonstrate that global I/R induces a transient increase in CaMKII-dependent phosphorylation of Na_v1.5.

Fig. 1.

Top: proposed hypothesis linking Ca²⁺/calmodulin-dependent protein kinase II (CaMKII)-dependent phosphorylation of dominant cardiac Na⁺ channel isoform Na_v1.5 and arrhythmia susceptibility. p-CaMKII, phosphorylated CaMKII; p-Na_v1.5, phosphorylated Na_v1.5; I_Na, Na⁺ current; APD, action potential duration at 80% repolarization. Bottom: experimental protocol to study arrhythmia susceptibility in isolated, Langendorff-perfused mouse hearts.

Fig. 2.

A and B: representative immunoblots and densitometric measurements for phosphorylated Ca²⁺/calmodulin-dependent protein kinase II (p-CaMKII) and phosphorylated Na⁺ channel isoform Na_v1.5 (p-Na_v1.5) expressed as ratio to total CaMKII and Na_v1.5, respectively, in wild-type (WT) mice at baseline, after 15 min of ischemia (Isc), and after reperfusion (I/R) for 5 and 15 min. Values are means ± SE; n = 3 in each group. *P < 0.05 vs. baseline; #P < 0.05 vs. reperfusion for 5 min; ^P < 0.05 vs. reperfusion for 15 min.

Phosphorylation of Na_v1.5 at Ser⁵⁷¹ prolongs APD and increases APD heterogeneity during I/R.

To assess the role of CaMKII-dependent phosphorylation of Na_v1.5 in creating a substrate for I/R-induced arrhythmias, we used knockin mouse models with the CaMKII phosphorylation site at Ser⁵⁷¹ replaced with glutamine (S571E) to mimic constitutive phosphorylation or alanine (S571A) to prevent phosphorylation (19). Optical recordings revealed significant APD prolongation in S571E hearts at baseline, after 15 min of ischemia, and after 5 min of reperfusion compared with WT or S571A hearts (Fig. 3). WT and S571A hearts showed similar APD except immediately after ischemia, when APD was significantly shorter in S571A than WT hearts (Fig. 3, A and B) Consistent with the known action of ATP-sensitive K⁺ channels, APD was significantly shorter in all groups after ischemia than at respective baselines (P < 0.05 for all groups) (16). Interestingly, although WT and S571A hearts had a similar mean APD after 5 min of reperfusion, WT hearts showed a significant increase in APD dispersion compared with S571A hearts (measured as standard deviation of APD across the mapping area) (Fig. 3, C and D). APD dispersion was also high in S571E compared with S571A hearts during reperfusion (similar to the level in WT hearts). In contrast, conduction velocity was not significantly different between the three groups at baseline or after 5 min of reperfusion (Table 1). These data suggest that CaMKII-dependent phosphorylation of Na_v1.5 affects APD and APD dispersion to potentially alter the triggers and substrate for arrhythmia during I/R.

Fig. 3.

A and B: optically recorded action potentials (A) and summary data (B) for action potential duration at 80% repolarization (APD) in wild-type (WT), S571E, and S571A hearts at baseline and during the ischemia (Isch)-reperfusion (I/R) protocol. Data are also shown for a subset of WT hearts pretreated with mexiletine (WT + mex). Scale bars = 50 ms. C and D: APD maps and summary data showing APD standard deviation as an index of spatial dispersion during the I/R protocol. Maps span a change in APD of 30 ms for all groups, with a shift in the absolute range to facilitate comparison of dispersion independent of differences in average APD. Approximate locations of right ventricle (RV), left ventricle (LV), base, and apex are labeled on APD maps. A pacing cycle length of 150 ms was used for all measurements. Values are means ± SE; n = 5 independent preparations for all groups. *P < 0.05 vs. WT; # P < 0.05 vs. S571A; ^P < 0.05 vs. WT + mex.

Table 1.

Ca²⁺/calmodulin-dependent protein kinase II-dependent phosphorylation of Na_v1.5 does not alter conduction velocity during ischemia-reperfusion

	WT		S571E		S571A		WT + Mexiletine
	Longitudinal	Transverse	Longitudinal	Transverse	Longitudinal	Transverse	Longitudinal	Transverse
Baseline	57.6 ± 1.7	35.6 ± 1.5	53.8 ± 3.0	27.4 ± 2.8	57.6 ± 1.7	38.2 ± 1.6	56.2 ± 1.8	36.2 ± 0.14
Ischemia-reperfusion (5 min)	56.8 ± 2.0	36.4 ± 2.1	53.2 ± 3.6	26.2 ± 3.0	57.6 ± 1.4	38.0 ± 0.7	54.0 ± 1.1	35.0 ± 1.3

Values (in cm/s) are means ± SE; n = 5 independent preparations for all groups. WT, wild-type. P = not significant.

To determine whether CaMKII-dependent phosphorylation of Na_v1.5 affected arrhythmia susceptibility during reperfusion, electrocardiogram measurements were analyzed during spontaneous (unpaced) activity. Although spontaneous VT was not observed in any strain during reperfusion, the incidence of arrhythmia triggers (PVCs) was significantly higher in S571E and WT than S571A hearts (Fig. 4). Arrhythmia inducibility was also assessed in WT, S571E, and S571A hearts after 5 min of reperfusion using a S1S2 pacing protocol. Interestingly, reentry giving rise to VT was readily induced in WT and S571E, but not S571A, hearts (Fig. 5). Notably, arrhythmias were not inducible in S571A hearts for a wide range of S1S2 intervals (decreased until the tissue could no longer be captured). Together, these findings indicate that CaMKII-dependent phosphorylation of Na_v1.5 alters excitability to promote both triggers and substrate for arrhythmia during I/R.

Fig. 4.

A and B: representative electrocardiogram recordings after 5 min of reperfusion (scale bars = 500 ms) and summary data for spontaneous arrhythmia events over the entire reperfusion period for wild-type (WT), S571E, and S571A hearts. Data are also shown from a subset of WT and S571E hearts pretreated with mexiletine (WT + mex). Representative arrhythmia events (premature ventricular contractions) are identified with red arrowheads. Values are means ± SE; n = 5 independent preparations for all groups. *P < 0.05 vs. WT; #P < 0.05 vs. S571E.

Fig. 5.

A and B: summary data and representative action potentials during the S1S2 protocol to induce arrhythmia following 5 min of reperfusion for wild-type (WT), S571 mutant (SE), and S571 mutant (SA) hearts. Data are also shown for a subset of WT and S571E hearts pretreated with mexiletine (WT + mex and S571E + mex). Cycle length of 150 ms was used for S1 pacing. The red arrow denotes the initiation of S2 protocol. VT, ventricular tachycardia. Values are means ± SE; n = 5 independent preparations for all groups. *P < 0.05 vs. WT; #P < 0.05 vs. SA. C: representative activation maps showing the spatial spread of activation after the last S1 beat, subsequent S2 beat, and first unpaced beat. WT and S571E maps demonstrate reentrant activation driving ventricular tachycardia after S2 (reentry path indicated by white arrows). Map for S571A shows conduction block near site of S2 (no unpaced activity after S2). The WT + mex map displays a single unpaced beat (no unpaced activity after the first unpaced beat). The location of pacing for S1 and S2 is indicated by a white asterisk.

Treatment with mexiletine reduces APD dispersion and arrhythmia susceptibility after I/R.

To test the hypothesis that CaMKII-dependent phosphorylation promotes arrhythmia after I/R via increased I_Na,L, we examined the effects of the selective Na⁺ channel blocker mexiletine on WT hearts subjected to our I/R protocol (Figs. 3–5). We used 10 μM mexiletine on the basis of previous reports showing preferential block of I_Na,L over peak at this dose (49). Interestingly, WT hearts treated with mexiletine showed APD changes with I/R similar to those observed in S571A hearts: mexiletine resulted in a shorter APD after ischemia without significant differences at baseline or after reperfusion compared with untreated hearts (Fig. 3, A and B). Mexiletine also reduced APD dispersion after reperfusion similar to that measured in S571A hearts (Fig. 3, C and D). At the same time, mexiletine reduced spontaneous arrhythmia events during reperfusion in both WT and S571E hearts (38 ± 9% and 49 ± 7% reduction, respectively, although differences were not significant; Fig. 4) while completely eliminating inducibility of VT (Fig. 5). These results support our hypothesis that CaMKII-dependent phosphorylation of Na_v1.5 enhances I_Na,L to alter arrhythmia triggers and substrate during I/R.

DISCUSSION

In the United States, ∼1,000,000 patients suffer MI each year, yet the mechanisms contributing to arrhythmogenesis in the setting of I/R remain incompletely understood (38, 52). While efforts have been made to limit Na⁺/Ca²⁺ overload after I/R, effective targeting of related pathways without altering normal physiology remains a challenge for the field. In this study we define a novel mechanism to limit arrhythmias after I/R by specifically targeting CaMKII-induced I_Na,L. Consistent with previous studies (28, 47), we found that I/R promotes CaMKII activation (autophosphorylation). For the first time, we report a parallel increase in phosphorylation of Na_v1.5 at Ser⁵⁷¹. Through the use of Scn5a knockin mouse models (S571E and S571A), we show that Ser⁵⁷¹ on Na_v1.5 is critical for CaMKII-dependent regulation of I_Na,L, tissue excitability, and arrhythmias during I/R. Our findings demonstrate that ablation of CaMKII-dependent phosphorylation of Na_v1.5 or pharmacological inhibition of I_Na,L with mexiletine reduces APD heterogeneity, arrhythmia triggers, and inducibility in response to I/R. Our data support the hypothesis that CaMKII-dependent hyperphosphorylation of Na_v1.5 promotes both triggers and the substrate for arrhythmia during I/R. We propose that Na_v1.5 hyperphosphorylation promotes spontaneous arrhythmia events via increased I_Na,L and consequent Na⁺/Ca²⁺ overload.

Earlier work identified CaMKII as an important contributor to cardiac arrhythmias in heart failure and chronic disease states (3, 12, 13, 52a, 57). These previous studies showed that, through various mechanisms (e.g., β-adrenergic stimulation and oxidative stress), CaMKII activity is increased, leading to abnormal cardiac excitability and function. While CaMKII has received considerable attention for its role in chronic disease, it also has the ability to promote arrhythmogenesis under acute stress relevant to acute MI (6, 28, 37, 46, 47). Studies of acute CaMKII activation in I/R showed an increase in phosphorylated CaMKII after as little as 15 min of I/R, with measurable effects on the downstream targets phospholamban and ryanodine receptor 2 (9, 47, 48). Other work pointed to Na⁺/Ca²⁺ overload as a source of arrhythmias in I/R (6, 10). Consistent with these studies, we have shown an increase in CaMKII autophosphorylation after 15 min of ischemia, with a concomitant increase in arrhythmia susceptibility after reperfusion. Beyond this previous work focusing heavily on CaMKII and its regulation of Ca²⁺ cycling after I/R, our study sought to elucidate Na_v1.5 as a potential upstream source of Na⁺ overload (and subsequent Ca²⁺ overload during I/R). In line with this objective, we found that phosphorylated Na_v1.5 is upregulated, together with CaMKII autophosphorylation, after 15 min of ischemia. Interestingly, we found that CaMKII and Na_v1.5 phosphorylation levels are rapidly normalized during reperfusion (by 5 min), although the proarrhythmogenic effects on AP repolarization (and, presumably, ion homeostasis) remain. Related to these findings, previous work in rats showed that intracellular Na⁺ concentration doubles within 5 min of the start of global ischemia and continues to steadily increase for ≥30 min (25). Furthermore, intracellular Na⁺ concentrations return to normal within 10 min of reperfusion. In part, these rapid changes in intracellular Na⁺ are due to alterations in Na⁺/H⁺ exchanger activity, but I_Na,L also likely plays a role (27). Investigation of the upregulation of phosphorylated Na_v1.5 at shorter and multiple time points for I/R would help elucidate the role of I_Na,L in mechanisms of Na⁺ overload and how it relates to changes in phosphorylated CaMKII and other Ca²⁺-handling proteins. It is important to note that, in addition to changes in phosphorylation status, CaMKII and Na_v1.5 are likely subject to posttranslational modification downstream of a myriad of stress signals during I/R. While our experiments focused on CaMKII phosphorylation of Ser⁵⁷¹ on Na_v1.5, a host of phosphorylation sites have been identified for multiple signaling pathways (including additional sites for CaMKII) with potential to regulate Na_v1.5 function in I/R (35). Furthermore, reactive oxygen species overproduction during reperfusion likely contributes to CaMKII activation and downstream effects but may also directly affect Na_v1.5 (and other channel) behavior (30, 55). In this vein, it would be interesting to examine whether reactive oxygen species scavenging could reduce CaMKII activity, restore normal Na_v1.5 activity, and mitigate arrhythmias produced by their interaction during I/R.

Beyond direct effects on ion homeostasis, it is interesting to consider the multiple ways that I_Na,L alters excitability and arrhythmogenesis during I/R. We provide interesting data that alteration of Na_v1.5 phosphorylation impacts both the triggers (spontaneous PVCs) and substrate (inducibility of tachycardia) for reperfusion arrhythmias. The effect of Na_v1.5 phosphorylation on arrhythmia triggers is likely mediated by changes in ion homeostasis, although this remains to be tested directly. It is less clear how changes in Na_v1.5 phosphorylation and I_Na,L impact the substrate for arrhythmia. Our data showing decreased epicardial APD dispersion (base to apex) during reperfusion in S571A compared with WT or S571E hearts hint at a potential mechanism (Fig. 3, C and D). Na_v1.5 has variable regional expression in the mouse heart, although base-to-apex changes are less well characterized than transmural or across chamber (e.g., right to left ventricle) changes (44, 53). At the same time, others observed that interventricular electrical heterogeneities are amplified by ischemia, exacerbating the arrhythmia substrate (36, 50). It is possible that the increase in APD dispersion and arrhythmia susceptibility observed with I/R-induced augmentation of phosphorylated Na_v1.5 reflects the inherent heterogeneous expression of Na_v1.5. It will be interesting in the future to determine the expression pattern of total and phosphorylated Na_v1.5 throughout the heart at baseline and during I/R.

Previous efforts to prevent arrhythmogenesis associated with Na⁺/Ca²⁺ overload have focused primarily on inhibiting the Na⁺/Ca²⁺ exchanger (10, 24, 32). Much less attention has been given to I_Na,L. as a source of Na⁺ overload upstream of the Na⁺/Ca²⁺ exchanger. Recently, it was shown that I_Na,L activity increases to promote intracellular Na⁺ accumulation during I/R, providing a potential link between dysregulation of I_Na,L and arrhythmogenesis (45). At the same time, it was shown that CaMKII exerts specific regulation over Na_v1.5 through phosphorylation, and when CaMKII activity is increased, this leads to a higher amount of I_Na,L (54). Our study builds on this previous work by identifying CaMKII-dependent phosphorylation of Na_v1.5 at Ser⁵⁷¹ upstream of increased I_Na,L, Na⁺ overload, and arrhythmias in I/R. Our investigation depended on the use of novel S571E and S571A knockin mice, which allowed for targeted perturbation of CaMKII-dependent regulation of Na_v1.5 and I_Na,L. Interestingly, we were able to alter repolarization and arrhythmogenesis during I/R by manipulating the Ser⁵⁷¹ site. These data support the notion that CaMKII-dependent phosphorylation of Na_v1.5 is likely an important upstream event in I/R-induced arrhythmias.

Clinical relevance.

Despite efforts to reduce arrhythmogenesis after MI, there remains a great need for new and improved therapies (15). High-profile examples of past failures in this area include the use of Na_v channel blockers, such as flecainide and encainide, which were successful in reducing PVCs but increased the risk of fatal reentrant arrhythmias in high-risk patients (8a). While improved outcomes have been observed for a Na⁺ channel inhibitor, such as ranolazine (more specific for late than for peak Na⁺ current), concerns remain about efficacy and off-target effects that can affect normal physiology (1, 8, 11, 20). Although implantable devices have shown greater efficacy in post-MI patients, they have several limitations, including cost, durability, and invasiveness (39). There has also been an effort after MI to treat patients with nonspecific anti-inflammatory steroids to assist recovery, but these drugs are linked with worsened ventricular and clinical outcomes (21, 42). As such, there is certainly a need for more specific targeting of pathogenic pathways in post-MI patients to combat these life-threatening arrhythmias (14). Our study attempts to explore this need by investigating a novel arrhythmogenic pathway involving CaMKII and Na_v1.5 in the setting of I/R. Our findings suggest that targeting of I_Na,L may be an effective way to reduce arrhythmia triggers without increasing risk for reentrant arrhythmias.

Limitations.

While our findings provide novel insights into the mechanism linking CaMKII-dependent phosphorylation of Na_v1.5 with increased I_Na,L and arrhythmias in I/R, we acknowledge multiple study limitations. First, our study used a global ischemia model in which perfusion to the entire heart is stopped for a period of time. In reality, ischemia is a much more heterogeneous condition, with regions of poor perfusion surrounded by nonischemic tissue. We anticipate that the role of the CaMKII-dependent pathway identified here will be just as important in the more heterogeneous case as in the global ischemia model. In vivo studies using temporary ligation of a coronary artery would be important for extending these findings to more a physiological situation. The use of blebbistatin is an additional limitation, particularly in the ischemia model, because of its mechanical uncoupling action, which would be expected to conserve ATP and aid in recovery during the reperfusion period (49). Another limitation is the difference in electrophysiology (especially repolarization) between mice and humans (40). However, the CaMKII-dependent pathway under investigation in this study also is highly conserved across species (31, 33).

GRANTS

This work was supported by National Heart, Lung, and Blood Institute Grants HL-135096, HL-134824, and HL-114893 (to T. J. Hund) and HL-135754, HL-134824, and HL-114383 (to P. J. Mohler) and by an American Heart Association Postdoctoral Fellowship (to A. Greer-Short).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

T.H., A.G.-S., T.S., N.P., D.N., P.J.M., and T.J.H. conceived and designed research; T.H., A.G.-S., T.S., N.P., and D.N. performed experiments; T.H., A.G.-S., T.S., N.P., D.N., P.J.M., and T.J.H. analyzed data; T.H., A.G.-S., T.S., N.P., D.N., P.J.M., and T.J.H. interpreted results of experiments; T.H., A.G.-S., T.S., and T.J.H. prepared figures; T.H., A.G.-S., T.S., and T.J.H. drafted manuscript; T.H., A.G.-S., T.S., N.P., D.N., P.J.M., and T.J.H. edited and revised manuscript; T.H., A.G.-S., T.S., N.P., D.N., P.J.M., and T.J.H. approved final version of manuscript.