An Il‐lumen‐ating look at ryanodine receptor modulation by disruption in triadin and calsequestrin interactions in cardiac myocytes

In cardiac muscle, ryanodine receptors (RyR) govern the release of calcium from the sarcoplasmic reticulum (SR). Among the various proteins that are involved in the functional modulation of single channels, triadin (Trd), junctin (Jn) and calsequestrin (CASQ2) are three that reside in the SR lumen. CASQ2, a high capacity calcium buffer, also has a secondary role as a luminal calcium sensor that inhibits RyR opening at low luminal calcium levels. Interactions of CASQ2 with RyR appear to be mediated through Trd and Jn as these proteins have binding sites for both CASQ2 and RyR. The domain structures of Trd and Jn are very similar: both proteins contain a KEKE protein–protein binding motif that is thought to be involved in interactions of Trd and Jn with each other and with RyR2 and CASQ2. The region identified as the CASQ2 binding region in Trd is a 25 amino acid site (200–224) that contains a single KEKE motif. While our current knowledge of the three abovementioned proteins and of luminal calcium sensitization is the result of the research carried out by many groups, Sandor Gyorke's laboratory is one of the first to identify these auxiliary proteins as exerting a complex functional control over RyR sensitivity to luminal calcium. The complex structural formation of Trd and Jn to anchor CASQ2 close to RyR channels was discerned by Larry R. Jones's laboratory while implications of this formation on RyR sensitivity with respect to calcium was studied in collaboration with the group of S. Gyorke (Zhang et al. 1997; Gyorke et al. 2004). The identification of the luminal calcium sensor was done by adding these proteins sequentially to purified single RyR channels on the trans (luminal) side and looking at the combined and the individual effects of these proteins on the open probability of a single RyR channel. From these bilayer experiments, it was clear that when Trd and Jn are added together or separately to a single purified RyR channel the open probability of the channel increased fivefold but this combination of proteins was unable to react to luminal calcium changes. On the other hand, CASQ2 alone did not have a significant effect on RyR open probability. However, when acting in conjunction with Trd and Jn, the complex was able to modulate RyR open probability as well as sense changes in luminal calcium levels. These single channel modulations have considerable physiological implications since clusters of RyR channels produce the elementary units of SR calcium release, calcium sparks, and local depletion of luminal calcium is thought to be involved in terminating these events (Sobie et al. 2002). The discussion that follows focuses on the group of S. Gyorke's most recent publication (Terentyev et al. 2007), which examined disruption of Trd–CASQ2 interactions. This was achieved experimentally by inducing cultured adult rat myocytes to overexpress a SR-tagged competitive inhibitor peptide (TrdPt) that contained the binding region of the rat cardiac Trd to CASQ2. To establish that TrdPt interacts only with CASQ2 and does not affect the RyR channels directly, the effects of this peptide on single RyR channels in planar lipid bilayers were investigated. The peptide induced an insignificant change in the open probability of RyR when it was added to the trans side of the channel (their Fig. 6). Further evidence that the peptide affected the channel's open probability only indirectly by interacting with CASQ2 was obtained when CASQ2 failed to inhibit RyR channels after the addition of the peptide. This confirmed the inability of TrdPt to directly interact with RyR and emphasized that the paper's other results reflected the lack of Trd–CASQ2 interaction. Several experiments were done to understand the functional significance of this disruption. The voltage dependence of calcium release triggered by calcium current (ICa), spontaneous calcium sparks, and existence of calcium waves are all means of quantifying the mishandling of luminal calcium and any change in the function of RyR channels. The response of these cells to ICa at different voltages in the form of calcium transients lacked the characteristic bell shaped curve (their Fig. 2). More calcium release at positive potentials suggests that addition of TrdPt enhances the functional activity of RyR channels so that they can respond to smaller calcium triggers. The frequency of spontaneous sparks and calcium waves increased in myocytes expressing TrdPt when compared to control, while the peak amplitudes of these events decreased. Both these results suggest stimulated activity of RyR channels (their Figs 4 and 5). However, whether the increased activity is due to increased SR calcium content (which is usually the case in such observations) or sensitization of the RyR channels is unclear. An assessment of the SR content revealed that in fact the SR calcium was lowered in the presence of TrdPt, which can be attributed to calcium leak from the SR due to increased spontaneous spark probability. This further confirmed the sensitization of RyR channels to luminal calcium due to the lack of Trd–CASQ2 interaction. In accordance with previous observations of cellular arrhythmia in the case of CASQ2 mutations, and the overexpression of Trd, similar arrhythmogenic disturbances in calcium transient and membrane action potential are seen in the myocytes expressing TrdPt when they are paced while exposed to isoproterenol. In this case also the amplitude of the calcium transient was decreased relative to the control (their Fig. 3), again suggesting stimulated RyR channel activity. (Terentyev et al. 2007) provides a thorough analysis of the effects of TrdPt as a competitive inhibitor that prevents Trd–CASQ2 interaction. The situation becomes more complicated when one tries to place these results in the context provided by previous studies, namely the paper that examined overexpression of full length and a mutated triadin (Trddel200–224) (Terentyev et al. 2005). Overexpression of Trd also resulted in cellular arrhythmias, and most of the experiments resulted in similar observations as in the current paper. The question raised then is how in two almost opposing scenarios, one where Trd–CASQ2 interaction is disrupted (in Terentyev et al. 2007) and the other where this interaction is not disrupted but more triadin is added, the end result is the same. In Fig. 1 of this paper, we have tried to summarize the four sets of results: (1) Control conditions; (2) Trd overexpression; (3) Trd del200–224 and; (4) TrdPt addition. This summary is based on the lipid bilayer data available on single RyR activity and observations at the cellular level seen in these papers. One major assumption that we make here is that if a certain combination of these proteins results in increased open probability (compared to purified RyRs) in lipid bilayers at 20 μm trans calcium, then this combination will also cause increased spontaneous calcium release at cellular SR calcium levels (∼700–1000 μm). The box at the top of the figure contains the four possible configurations in which the RyR complex can be found where A represents Trd binding to RyR and CASQ2; B, Trd interacting only with RyR; C, absence of Trd but with CASQ2 present; and D, RyR in the absence of both Trd and CASQ2. The thickness of the arrow in each case reflects the average open probability of RyRs in that configuration. We hypothesize that the majority of RyRs in the control case are in form A while C and D are present in small numbers. Continuing with this hypothesis, it is possible that B is missing or present in an insignificant quantity under control conditions, as any Trd would be bound to CASQ2 due to the abundance of the latter protein. Cumulative behaviour of RyR channels under different protein interaction modifications The box on the top contains the four possible configurations in which the RyR complex can be found where A represents Trd binding to RyR and CASQ2; B, Trd interacting only with RyR; C, absence of Trd but with CASQ2 present; and D, RyR in the absence of both Trd and CASQ2. The thickness of the arrow in each case reflects the average open probability of RyR in that configuration. Four set of results are summarized by representation of different combinations of RyR complex configuration. In the case of full-length Trd overexpression, the amount of A remains the same while B increases due to Trd binding to previously unoccupied RyRs (state D). This would make some RyRs more sensitive and can be the cause of the observed calcium changes. Overexpression of Trddel200–224 has no significant effect on RyR channel activity, suggesting that this mutant protein cannot interact normally with the channel. Hence A is in a majority and no change is seen in other conformations. In the case of TrdPt, the Trd–CASQ2 interaction is disrupted, so A is changed to B, and RyRs that had been inhibited by CASQ2 are now stimulated by Trd. Thus, this model predicts that full-length Trd and TrdPt can produce similar functional effects, but via different mechanisms. One intriguing observation from (Terentyev 2005 & 2007) is the fact that Trddel200–224 loses its ability to increase the open probability of RyR channel in lipid bilayers, while the peptide that consists mostly of this region also fails to have any effect on RyR channel open probability. The simplest explanation can be that due to the deletions in Trddel200–224, the mutated protein was unable to fold correctly which resulted in its inability to bind to RyR channels. However, if this is not the case, then this region is crucial not only for binding to CASQ2 but also facilitates interactions between Trd and RyR channels. Overall, the paper by Terentyev et al. (2007) not only provides a thorough analysis of Trd–CASQ2 interaction, it also helps put together a more complete picture of the luminal side of the RyR story. These findings contribute greatly to our understanding of normal and pathological cardiac cellular physiology. We are grateful to Dr Eric A. Sobie for his encouragement, insightful discussions and review of the manuscript.