T-tubules and ryanodine receptor microdomains: on the road to translation

The concept of localized signalling in the coupling between L-type Ca2+ channels, LTCC, in the sarcolemma and ryanodine receptors, RyR, in the sarcoplasmic reticulum was introduced in the late 1980s to address one of the conundrums of excitation–contraction coupling. The tightly regulated Ca2+ signals observed in isolated cardiac myocytes under voltage clamp conditions, or in skinned myocytes, were hard to explain in a global control cell model in which intracellular calcium is uniform everywhere at all time. However, they could be understood in a concept of local control within a microdomain as proposed in the theoretical studies by Stern1 and others. Concomitant insight into the ultrastructure of the dyadic cleft and gating properties of single LTCC and RyR channels provided the critical elements to develop the microdomain concept where clusters of RyR interact with clusters of LTCC in the dyad. Such a functional unit was termed a couplon, consisting of clusters of RyR and LTCC in close juxtaposition (∼12 nm) of the two membranes. Sparks, spontaneous Ca2+ release events from RyR, were the first visualization of the local events2 and were typically associated with the dyad near T-tubules (TTs). Spurred by improved imaging of Ca2+ signals and of myocyte ultrastructure, and by computational modelling and genetic manipulations, our knowledge on the characteristics and functional impact of microdomain signalling near TTs has grown tremendously. The evidence that modulation and remodelling of these microdomains is involved in global cardiac dysfunction in hypertrophy and failure has further fuelled research. This spotlight issue gives an update of current knowledge emphasizing the outlook for signalling and cardiac remodelling, and highlighting the novel and emerging research directions. Early models pictured a couplon as consisting of geometrical arrays of RyRs with estimates for the number of molecules between 15 and >100 RyR, a number that was derived from electron microscopy data combined with theoretical derivations and mixed approaches such as fitting of spark data. The number of Ca2+ channels is generally at 1:6 for RyR or less, but this number is even less certain. RyR are now thought to form highly variable clusters not restricted to the dyad.3 Novel imaging methods have fast-forwarded our knowledge in this area.4 The work by Scriven et al.3 has further focused on the different partners of RyR and LTCC in the dyad, contributing to further understanding of privileged interactions between these partners. They also highlight the existence of RyR outside the dyad, and these ‘rogue’ RyR were shown to be functionally implicated in regulation of the SR Ca2+ content by providing a leak pathway that was not detected in classic spark recordings.5,6 Much less is known about the LTCC organization, but recent discovery of BIN1, a protein involved in trafficking, has revealed a highly organized and fast turnover system. The potential link to regulation of the dyad itself and the TT formation is discussed by Shaw and Colecraft.7 There is also growing insight that branching and axial components of the tubular membrane invaginations may all participate in signalling, hence the proposed use of the term transverse and axial tubular system or TATS, rather than TT. Changes in TT structure in biopsy material of patients with end-stage heart failure had been described by pathologists more than two decades ago, but the functional appreciation of changes in TT structure came from animal models. In vitro studies using artificial detubulation or spontaneous loss of TT during culture identified the importance of TT in providing synchrony of excitation–contraction coupling but also of location of transporters and ion channels. The next wave of studies turned to the changes in TT structure in disease and their impact on myocyte function.8–10 Animal models from mice and rats, to dogs and pigs have documented changes in TT structure during hypertrophic remodelling and in heart failure. Studies on human tissues are scarcer and examine fixed or frozen material but support the presence of similar changes in human heart failure, with some distinctions between ischaemic and dilated cardiomyopathy.9,11 In the animal models, the structural changes vary from a loss of a strict organization to actual loss in the number and surface area of the TTs. In either case this is linked to loss of synchrony and other functional changes, e.g. time course of NCX-dependent Ca2+ removal.12 Further insight into the functional consequences is gained from computational approaches that can identify critical parameters in altered spark behaviour in disease.13 Mathematical models of altered configuration of the dyad and properties of the individual RyR and transporters can provide testable hypotheses and direct new experiments. Within the normal healthy heart not all regions have a similar density of TTs. From studies in small mammals, it was long thought that atrial myocytes in general had no TTs. Novel observations indicate that atrial cells can, depending on the species, also exhibit a more or less developed TATS.14 On a broader perspective, this heterogeneity of the TT system in different regions of the heart suggests that the complexity of TT/TATS organization is a continuum, which is controlled by a regulatory pathway in the cardiac myocyte. This is also evident during the post-natal development of ventricular myocytes, which develop a TT system that is near-absent at birth. The mechanisms that underlie this normal development, and which also may be involved in the loss of normal TT structure with disease, are only beginning to unfold. Junctophilin-2, JP-2, positioned near RyR and in the dyadic cleft, has been invoked as a key regulator though much of the evidence is still associative.3,7,9 Indeed, it is very difficult to establish the causal relation and untangle the myocyte dysfunction and the loss of TTs from the loss of JP-2 as they occur together in models of heart failure. However, a strong argument in favour of the critical role of JP-2 is the observation that overexpression was able to mitigate the loss of TTs and rescue the cardiac dysfunction in the JP-2 knockdown mouse.15 In line with the concept of a highly regulated process, the loss of TTs can be reversed and has a close correlation to mechanical load.10 Though JP-2 is an appealing candidate, other proteins in or near the dyadic cleft may be involved. The targeting of LTCC and BIN-1 may also influence TT formation.7 Likely, there are additional pathways. Loss of triadin modulates ultrastructure of the dyad and the consequent alteration in the TT microdomain is related to arrhythmogenesis16 The RyR itself can be modulated in a variety of ways during cardiac remodelling. Early studies mostly focused on the global assessment of RyR expression and reports varied from reduction to no change. Novel data may provide a more subtle answer in examining the RyR cluster structure. Original work in this issue addresses a pathway for RyR turnover via chaperone-mediated autophagy that may be active during oxidative stress.17 Recent studies have emphasized the impact of post-translational modifications, including phosphorylation, nitrosylation, and oxidative changes.18,19 ROS signalling has a very wide range from normal physiology to pathology, depending on location, intensity, and timing of ROS production. The relation to RyR function is similarly complex, with novel roles discovered in stretch-sensing and communication with the mitochondria.19,20 The TT structure and associated microdomains are not limited to excitation–contraction coupling. The close co-regulation of metabolism and contraction has been linked to Ca2+ signals to the mitochondria, but the involvement of the mitochondria in global Ca2+ regulation has been more controversial. Several new data point to a close link in mitochondrial function and Ca2+ and the key may lie in looking from the vantage point of the mitochondria, i.e. the change in mitochondrial Ca2+, rather than changes in cytosolic Ca2+ consequent on mitochondrial Ca2+ uptake. At least part of the mitochondrial surface is close to the Ca2+ microdomains near TTs. Whether there is a specialization in transporters in these areas is currently unknown, but these areas could form a structural microdomain like the dyads in excitation–contraction coupling. Disrupting the relationship among the trios of TT, SR, and mitochondrion may undermine not only the microdomain Ca2+ signalling, but also the co-regulation of contraction and bioenergetics. In addition, receptors and signalling cascades may have preferential distribution and create local signals such as the beta-adrenergic signalling pathway.8 When TT structure is altered with disease, these cascades are also affected. Caveolae provide another microdomain of the sarcolemma that is not linked to RyR but harbours a unique pathway for Ca2+ signalling.7 Modifying RyR function and reducing excessive open probability with small molecule therapy has become one of the fast-track paths towards translation, following SERCA gene therapy for heart failure. It is beyond the scope of this spotlight issue to discuss all the potential benefits and caveats in this area, but the issue contains two original contributions that illustrate the on-going efforts and debate. Kalyanasundaram et al.21 illustrate potential risks with an indiscriminate increase in SERCA, whereas Sikkel et al.22 examine the mode of action of flecainide a multi-target drug to treat arrhythmias related to RyR dysfunction (see also the Letter to the Editor and Reply). Stokke et al.23 revisit the potential benefit of LTCC antagonism. The road to translation directed at restoring TT structure is only at the beginning, but has potential. Fast developments in imaging are already improving our knowledge of the structure and changes with disease.24 In parallel, a more in depth characterization of the functional microdomains is needed: near TT, at non-coupled RyR, in caveolae and in other signalling microdomains. Currently, our insights in the microdomain Ca2+ still mostly rely on computational approaches, as we lack tools for measuring local Ca2+ on the nanoscopic scale. Local ROS likewise is still beyond our measuring capacity. However, new FRET tools and targeted reporters hold promise for the near future. For translation, better understanding of the control mechanisms governing TT formation and turnover, and reversibility of TT changes with disease is needed. Precisely targeted genetic manipulations, as currently studied for JP-2, will hopefully help to elucidate further the critical pathways. Small molecule development, as successfully done for RyR, then will be the last step. Conflict of interest: none declared. This work was supported by the European Community's Seventh Framework Program FP7/2007-2013 under grant agreement no HEALTHF2-2009-241526, EUTrigTreat (to K.R.S.), and the National Key Basic Research Program of China (2013CB531200 and 2011CB809100) (to H.C.).