Gaining respectability: membrane-delimited, caveolar-restricted activation of ion channels.

Voltage-gated sodium channels play a key role in the excitability of myocardial cells and in impulse propagation.1,2 An increase in Na current is, for example, responsible for the generation of the rapid upstroke of the action potential (phase 0). Mutations in the gene encoding this channel (SCN5A) have been linked to the pathogenesis of several cardiac channelopathies, including the long-QT and the Brugada syndromes,1,2 the discoveries of which have helped in the functional characterization of these channels. In addition, the signaling pathways that regulate Na channel activity, such as the sympathetic nervous system, have been the subject of numerous studies.3–6 Indeed, stimulation of -adrenergic receptors can both enhance conduction in normal ventricular myocardium but also induce arrhythmic events in a number of cardiac disease states.1,2 Several mechanisms that contribute to the regulation of Na channels by -adrenergic stimulation have been documented (see Figure). All of them involve G proteins, although diffusible second messengers are not necessarily involved. For example, besides the coupling of G s to adenylate cyclase and downstream protein kinase A (PKA)-mediated phosphorylation events,3,4 the so-called “membrane-delimited” pathway does not involve diffusible cytosolic factors. The G s subunit can, indeed, directly modify Na channel activity.5,6 This type of regulation is not unique to G s, as the direct binding of the G protein subunit to several Ca , K , and Cl channels has been extensively documented (for review, see Dascal7). Shibata and colleagues have shown previously that the application of peptides derived from the G s sequence in inside-out macropatches enhanced isoproterenol-evoked sodium current (INa), attributable to an apparent increase in the number of functional channels.5,6 In this issue of Circulation Research, Shibata and colleagues (Yarbrough et al8) report that the source of these new channels could be caveolae. Using antibodies directed against caveolin-3, the musclespecific isoform of caveolin, these investigators were able to block the direct effect of isoproterenol on the Na channel (ie, the PKA-independent, G s membrane-delimited pathway). Importantly, this effect seems specific, given that antibodies directed against other caveolin isoforms did not affect isoproterenol activation of Na channel activity. Moreover, Yarbrough et al8 also documented, in cardiac myocytes, the apparent preferential enrichment into caveolar microdomains of Na channels and G s. Although these data led the authors to conclude that an increase in sympathetic nervous system activity could lead to the recruitment of Na channels out of caveolar microdomains to the sarcolemma, thereby increasing INa, several caveats regarding this hypothesis should be noted. Definitive evidence for the translocation of Na channels out of caveolae upon isoproterenol stimulation would require further subcellular fractionation and/or electron microscopic studies. In the absence of such experiments, there is no reason to think that the G s-activated Na channels necessarily would need to exit caveolar microdomains to be active. The use of the anti–caveolin-3 antibody, although specific for this caveolin isoform, may have had effects on other signaling proteins in myocyte caveolae that affected Na function. Moreover, it is unlikely that the antibody resulted in the disruption of mature caveolae. This specific antibody, which is directed against an epitope in the nonscaffolding N-terminal end of caveolin-3, probably acts via steric hindrance to prevent G s translocation to the Na channels into the caveolar microenvironment. Alternatively (or additionally), because -adrenergic receptors have been found to be enriched in cardiac myocyte caveolae,9,10 accumulation of the antibody could have altered receptor coupling to G s (and its consecutive targeting to Na channel) upon agonist stimulation. In either case, the argument that blockade of Na channel translocation out of caveolae is required to explain the decrease in INa does not necessarily hold. Indeed, sarcolemmal caveolar microdomains could be an important site of Na channel activation. Several lines of evidence support this concept. First, a caveat: the term “caveolae” should be used with caution in the context of cardiac muscle. In contrast to skeletal muscle, where caveolin-3 is only observed in association with sarcolemmal caveolae, caveolin-3 in cardiac myocytes has been observed to be associated with both the plasma membrane and the T-tubular system11,12 (Prof R.G. Parton, written communication, January 2002). Interestingly, several groups have reported the location of cardiac Na channels in T-tubules to be in close proximity with the Na -Ca exchanger and L-type Ca channels (which are in close contact with the sarcoplasmic reticulum [SR]) (for review, see Scriven et al13). The data of Yarbrough et al8 could therefore be interpreted as complementary evidence that Na channels are colocalized with other key regulators of excitation-contraction coupling within caveolin-enriched structures in T-tubules. The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association. From the Unit of Pharmacology and Therapeutics (O.F.), Department of Medicine, University of Louvain Medical School, Brussels, Belgium; and Genzyme Corporation (R.A.K.), Cambridge, Mass. Correspondence to Olivier Feron, University of Louvain Medical School, Unit of Pharmacology and Therapeutics, UCL-FATH 5349, 53, Avenue E. Mounier, B-1200 Brussels, Belgium. E-mail [email protected] (Circ Res. 2002;90:369-370.) © 2002 American Heart Association, Inc.