T‐tubule trouble

Abstract

During the cardiac action potential, Ca2+ influx via L-type Ca2+ channels activates Ca2+ release channels (ryanodine receptors; RyRs) in adjacent sarcoplasmic reticulum (SR) membrane, causing Ca2+ release from the SR and thus contraction. There is increasing evidence that in ventricular myocytes the transverse (t-) tubules – invaginations of the surface membrane – play a specialized role in this process: early electron microscopy studies showed ‘feet’ (now known to be RyRs) abutting the t-tubule membrane; subsequent immunohistochemical studies showed localization of key proteins at the Z-line and thus, by implication, in the t-tubule membrane; and more recent work has demonstrated the functional localization of proteins, for example the Ca2+ channel, in the t-tubule membrane, and that the t-tubules underlie spatial and temporal synchronization of Ca2+ release (see Brette & Orchard, 2003 for review). T-tubule structure is complex, but it is also labile: the t-tubules appear during development, disappear when cardiac myocytes are cultured and change in animal models of heart failure, for example induced by rapid pacing (He et al. 2001), and in human heart failure, although finding suitable control cells is, as always for such studies, problematic. However the importance of these changes in shaping the Ca2+ transient during failure was unknown. Changes in Ca2+ handling are well known to occur during heart failure, and underlie many of the associated contractile abnormalities. Although the literature is vast, several common themes emerge, including decreased SR Ca2+ content and a smaller and slower Ca2+ transient, as the result of changes in the expression and activity of proteins involved in excitation–contraction coupling, including the L-type Ca2+ channel, Na+–Ca2+ exchanger, SR Ca2+ pump, phospholamban and RyR (see Bers et al. 2003 for brief review). In this issue of The Journal of Physiology, Louch et al. (2006) demonstrate that changes in t-tubule structure in a murine model of congestive heart failure also contribute to the altered Ca2+ transient. In this model, t-tubule structure becomes more disorganized and areas devoid of t-tubules develop; these areas exhibit delayed Ca2+ release. Interestingly, there was also greater beat-to-beat variability in the Ca2+ release profile in areas that didn't show such delays. Thus it appears that changes in t-tubule structure cause desynchronization, while other, unidentified, mechanisms contribute to increased beat-to-beat variability. These changes are important because they desynchronize and slow the Ca2+ transient and thus decrease contraction, although the effect on time to peak appears to be relatively small compared to other mechanisms that slow the Ca2+ transient. However, such changes in t-tubule structure may have different effects on Ca2+ release in other forms of heart failure. If delayed Ca2+ release in areas lacking t-tubules is the result of Ca2+ diffusion from adjacent release sites, then it will depend on cellular Ca2+ load, which alters the extent and velocity of propagation (Brette et al. 2005). In the murine model, SR Ca2+ load is increased or unchanged; however, in most failure models SR Ca2+ content is decreased, which may accentuate the effect of t-tubule loss on desynchronization. It is also suggested that ‘orphan’ RyRs (Song et al. 2006) or changes in dyadic structure may alter the efficacy with which the Ca2+ current can trigger SR Ca2+ release, and cause local changes of Ca2+ release (Song et al. 2006). Such changes are unlikely to produce long-term alterations in the amplitude of the whole cell Ca2+ transient, which depends on the Ca2+ load of the cell, particularly the SR, which is determined by transmembrane Ca2+ fluxes (Eisner et al. 2000); however, the role of cellular microdomains in maintaining whole cell Ca2+ balance is unknown. Failing heart muscle is, in addition, exposed to increased sympathetic stimulation in vivo. This has previously been shown to increase synchronization of Ca2+ release between RyRs in normal cardiac muscle (Brette et al. 2004), but had little effect on synchronization in the murine model of heart failure; whether this is the result of increased basal stimulation or changes in signalling pathways during failure is also unknown. Thus the full implications of the current data for heart failure remain to be explored. The present data do, however, add another facet to the changes of Ca2+ handling that occur during heart failure. They also raise fundamental questions for the future: why is t-tubule structure so labile, how is it normally maintained and why does it change during heart failure? Does this remodelling reflect, as suggested for other aspects of heart failure, reversion to a fetal pattern of gene expression? Does the distribution of protein function between the t-tubule and surface membranes also change? If so, what causes this redistribution and how does it affect cell function? Clearly t-tubule trouble has only just started!