Calcium-dependent regulation of voltage-gated sodium channels in cardiac myocytes: just the beginning?

The Ca2+-binding protein calmodulin (CaM) and the Ca2+/CaM-sensitive kinase (CaMKII) play an important role in the regulation of cardiac excitation–contraction coupling, regulating ryanodine receptor and phospholamban activity.1 CaM is also known to regulate a variety of ion channels; voltage-dependent Ca2+ channels are the classical example, and CaM has been identified as the mediator of Ca2+-dependent channel inactivation. The discovery that Ca2+-dependent inactivation is associated with a specific region in the C-terminus containing a consensus ‘IQ’ motif2 initiated the pursuance of calcium-dependent regulation of other ion channels. The recognition of the presence of an IQ-like motif within the C-terminus of voltage-gated sodium channels (VGSCs) in mammalian cells3 triggered interest in the possibility that VGSCs may also undergo regulation via this mechanism. In the heart, VGSCs are integral to conduction within the myocardium and maintenance of normal cardiac rhythm, underlying the action potential upstroke and velocity of impulse propagation. Impairment of VGSC function is frequently associated with clinical conditions, including Brugada syndrome, and conditions involving augmented late sodium VGSC current (INaL) such as long-QT syndrome (LQT3).4 At present, however, there is limited information regarding the effects of CaM/CaMKII modulation on activity of cardiac VGSCs. The results of a few individual studies that focused on the mechanisms of the effect of CaM on the electrophysiological gating characteristics of the cardiac VGSC isoform (NaV1.5) heterologously expressed in mammalian cell lines have presented conflicting evidence. Deschênes et al.5 found no effect of CaM on human NaV1.5 co-expressed with or without a β-subunit in HEK293 cells, whereas the skeletal muscle μ1 isoforms induced a hyperpolarizing shift in the voltage dependence of inactivation using the same expression system. Conversely, when CaM was co-expressed with NaV1.5 in Chinese hamster ovary cells, a hyperpolarizing shift in the voltage dependence of activation was observed.6 These differences may be due to differences in the expression systems, protocols, and/or protein isoforms used by the different groups, or, alternatively, they may indicate the presence of more than one regulatory calcium-binding site. The latter could be supported by the evidence presented by Balser's group who demonstrated that a hyperpolarizing (leftward) shift in the availability of the human H1 VGSC isoform could involve CaM binding to the IQ motif,7 whereas a depolarizing (rightward) shift could be mediated by the direct effect of Ca2+ on the EF-hand motif in the channel.8 These molecular reports not only provide invaluable insight into the mechanistic function of the VGSC molecule, but also instigate the exploration of the mechanisms of calcium-dependent regulation of endogenous cardiac VGSCs. Aiba et al.9 have comprehensively studied the effect of CaM and CaMKII on voltage-dependent characteristics of VGSC current (INa) in isolated guinea-pig ventricular myocytes using an electrophysiological approach. The authors demonstrated that cell dialysis with both Ca2+/CaM and CaMKII increased the peak INa and selectively altered inactivation, but not activation, causing a depolarizing (rightward) shift in INa availability and decreased entry into a slow inactivation. However, only CaMKII, but not Ca2+/CaM, was able to accelerate recovery from inactivation of INa and also caused a significant increase in INaL. The differential effects of Ca2+/CaM and CaMKII on the properties of the native INa are consistent with the presence of more than one Ca2+-dependent modulatory site in endogenous cardiac VGSCs. The complexity of the phenomenon of Ca2+-dependent modulation of VGSCs is further demonstrated in this study by the finding that the cardiac sodium channel α-subunit is phosphorylated in a CaMKII-dependent fashion at the I–II linker and C-terminus, which is different from previously reported Ca2+/CaM-dependent phosphorylation at the III–IV domain linker.10 The effects of CaM/CaMKII on the INa peak and inactivation described by Aiba et al. in guinea-pig ventricular myocytes in this issue are consistent with a recent report where the effect of specific CaMK inhibitors on INa in rat ventricular myocytes was investigated.11 Conversely, Wagner et al.12 demonstrated that overexpression of the cardiac-specific CaMKIIδc isoform in cultured rabbit ventricular myocytes caused effects on INa inactivation opposite to those described by Aiba et al.9 One possible explanation for such controversy may be the use of the neuronal CaMKIIα isoform in the latter study. However, the lack of any significant effect of increased cytosolic [Ca2+] on both activation and inactivation dependencies of INa recently reported in rabbit ventricular myocytes13 might suggest that this may not be the only explanation for the differences in the effects described in the two studies mentioned above, and species differences may also be important. Indeed, the paradoxical findings of the inhibition of peak INa by elevated cytosolic [Ca2+] in rabbit ventricular myocytes,13 and the enhancement of the current in guinea-pig ventricular myocytes,9 may be explained by the findings of Aiba et al. that suggest the existence of more than one Ca2+-dependent mechanism regulating the activity of VGSCs. Another important observation presented by Aiba et al. is the CaMKII-mediated enhancement of INaL. This effect was not mimicked by Ca2+/CaM and by activation of protein kinase A. The enhancement of INaL is also CaMKII isoform independent and species independent as it was observed with CaMKIIδc12 and in human and canine ventricular myocytes.14 Previous studies have shown that INaL is augmented and contributes to action potential prolongation and the generation of arrhythmogenic early after depolarization in heart failure.14 Recently, Sag et al.15 demonstrated that CaMKII is involved in the generation of arrhythmogenesis resulting from heart failure in transgenic mice overexpressing the CaMKIIδC isoform. The increased action potential duration in the presence of CaMKII reported by Aiba et al. provides further support for CaMKII-mediated enhancement of INaL as a potential contributing mechanism to arrhythmogenesis. Thus, the study by Aiba et al. only begins to lift the curtain on this new and exciting area of cardiac electrophysiology and to uncover the complexity of the molecular mechanisms behind the Ca2+-dependent modulation of VGSCs, demonstrating their functional importance in cardiac myocytes. Future focused studies will be necessary to fully understand these mechanisms and their exact role in the aetiology of cardiovascular diseases.