|Home | About | Journals | Submit | Contact Us | Français|
Normal contraction of cardiac muscle requires a coordinated cellular mechanical response to the electrical pacemaker signal that begins in the sinoatrial node and spreads through the conduction system and ultimately across the ventricular myocardium via gap junctions. In ventricular myocytes, depolarization by the action potential causes clusters of L-type calcium channels (LCCs) to open, allowing a small amount of Ca to enter the diadic cleft space between the sarcolemma and the sarcoplasmic reticulum (SR). When the Ca concentration in this space becomes sufficiently high to gate ryanodine receptors (RyRs), they open and release SR Ca into the cytoplasm so that it can bind to and activate myofilaments, resulting in force development. Each cluster of LCCs on the sarcolemma together with its opposing cluster of RyRs across the 10–12 nm diameter diadic cleft space together with associated SR regulatory proteins is a couplon 1. According to the local control theory of cardiac excitation-contraction (EC) coupling, couplons are activated independently. Independent activation of couplons depends upon the extent of LCC activation, providing a mechanism for regulation of contractile force by stochastic recruitment of couplons 2. In systolic heart failure, numerous cellular defects in EC coupling and Ca regulation have been identified that contribute to contractile dysfunction (and ventricular arrhythmias), though there is a host of other major abnormalities leading to loss of force development, including fibrosis 3, defective energy metabolism 4, pH changes 5, heterogeneous conductance across connexin hemichannels 6, and defects in sarcomeric proteins 7. Thus it is important to keep in mind that no single defect can explain the contractile dysfunction of systolic failure.
Recent studies of defective EC coupling in heart failure have addressed not only reductions in the extent and kinetics of Ca released in response to the action potential (i.e. reduced recruitment of couplons and reduced SR content), but also changes in the synchronization of intracellular Ca release. In remodeled ventricular myocytes isolated from the peri-infarct zone of rabbit hearts and studied with confocal microscopy, Litwin et al. 8 showed that during electrical stimulation, the local Ca release events (i.e. Ca2+ sparks9) produced by couplons no longer activate in a synchronous fashion. This results in reduced Ca flux with the expectation of reduced contractility at the cellular level. Loss of subcellular synchrony has a number of causes. One of these, reduction in couplon size due to attrition of LCCs during metabolic stress, is clearly an important factor contributing to loss of synchrony as we have shown previously 10. The effect of losing LCCs on Ca release can be explained as follows. If, for the sake of simplicity, we assume that all RyRs in a couplon can be activated by a single LCC 11, then the Ca spark’s latency will reflect the first latency of opening of this LCC. The fewer the number of LCCs in the couplon, the lower the chances of finding an LCC that opens with a short first latency. Thus the mean first latency will increase as LCCs are lost 10. LCC first latencies exhibit an exponential distribution. Such distributions exhibit a mean which is equal to the standard deviation of the distribution. As the mean of the first latency increases, the standard deviation and hence the variance of first latency also increases. Since the first latency of LCCs underlies the first latency of sparks, these too will display a greater variance, i.e. asynchrony, as we see in single cells from heart failure models.
Alternatively RyR clusters that form part of the couplon may become isolated during the remodeling process that removes t-tubules 12, 13. If these RyR clusters are in fact activated during depolarization, it must be by a common pool mechanism 2, which due to its inherent positive feedback will prevent regulation of the Ca transient by stochastic recruitment of individual couplons. Furthermore, these couplons will no longer reflect the activity and first latency of LCCs since RyRs will be physically separated. The net effect is a reduction in the synchronization of Ca release, which we will discuss in greater detail below.
In this issue of Circulation: Heart Failure, Wasserstrom et al 14 draw attention to another phenomenon contributing to contractile dysfunction that single cell studies, by their inherent nature, fail to reveal: loss of intercellular synchrony. The authors of this paper use confocal microscopy to study Ca release in the epicardial layer of ventricular cells in intact hearts from failing spontaneously hypertensive (SHR) rats. Although the spatial and temporal resolution of this technique does not allow detailed study of couplon activation, the authors nevertheless demonstrate that in failing hearts Ca release is not uniform across all ventricular cells. Indeed, they identify several different Ca release types: 1) triggered release with reduced amplitude and kinetics compared to normal rats; 2) spontaneous release either before or after the electrical trigger that is not present in normal rats; and finally 3) triggered release of slow intracellular waves that is not present in normal rats. Each of these phenomena can be identified in isolated cell studies, but until now have not been demonstrated occurring together in intact tissue. The reduced amplitude and kinetics of the Ca transient in some cells are likely the result of loss of LCCs 10 or reduced SR Ca load caused by reduced SERCA activity 15, whereas spontaneous release in other cells is probably caused by SR Ca overload and possibly leaky RyRs 16, 17. The very slow wavelike diffusion of Ca in some cells in response to electrical stimulation is suggestive of clusters of RyRs not activated directly by closely opposed LCCs in a couplon. There are at least two possible explanations for these waves. One, suggested by the authors, is that “orphaned” RyRs in their failure model are activated by diffusion of Ca through the cytoplasm from neighboring Ca release sites instead of the customary triggering of intact couplons by the electrical signal 13. Presumably if this occurs, the activation of non-junctional RyR clusters that are separated from couplons requires Ca release from intact couplons to spread to these non-junctional RyRs. Thus, we may expect that the activation of these RyR clusters is delayed. Since RyRs are largely confined to the Z-disk 18, a transverse scan is likely to display an inhomogeneous rise in Ca since couplons will be activated before non-junctional RyR clusters. Transients of this nature are apparent in Figure 8A of the article 14. Thus not only can the production of sparks be temporally asynchronous, but also they can be spatially nonuniform. This seems to be consistent with the findings of the authors.
It is also possible that the spread and shape of the action potential is abnormal in many of these cells due to faulty intercellular conduction due to abnormal Na channels, connexin hemichannels and gap junctions 6, 19, fibrosis 3, or refractoriness associated with abnormal K currents 20. All of these changes have been implicated in arrhythmogenesis, but could also lead to heterogeneous electrical activation of ventricular myocytes, which would be expected to trigger Ca release in a heterogeneous fashion. This is certainly an area that deserves further study, and which could be addressed by applying voltage sensitive dyes to the author’s multicellular preparation and then correlating changes in voltage with abnormal Ca release.
We would expect inhomogeneities at the cellular level to impair contractility. For example, if sarcomeres that are in series are activated at different times due to temporally inhomogeneous Ca release, the sarcomere that is activated first will stretch the compliance of the sarcomere with which it is in series before this neighboring sarcomere is activated. Thus the neighboring sarcomere will produce less force or shortening, and the contraction of these two sarcomeres together will be inhomogeneous. This will have the effect of reducing contractility within the cell. Sufficient serial elements in the cell activated in this fashion could greatly reduce the efficiency with which contraction occurs above and beyond that expected from reduction of the amplitude and kinetics of Ca transients alone. The observation by Wasserstrom et al. 14, that Ca transients are both temporally and spatially inhomogenous between cells, suggests an extension of this idea. If one myocyte tugs on its neighbor before the neighbor has been activated, this will have the effect of significantly reducing contractility for the reasons already mentioned. Thus we can expect contractility to be grossly impaired at both the cellular and the multicellular levels.
It is now understood that t-tubules enter the cell at the z-disk 18, 21. If a significant number of RyRs are orphaned in the z-disk by virtue of t-tubule loss 13, this means the spread of Ca across the z-disk is likely to be significantly inhomogeneous, with parallel sarcomeres quite possibly activated at different extents and at different times. The mechanical consequences of this are undisclosed, however it is unlikely to improve contractility.
This work exposes the limitations of single cell studies of EC coupling in heart failure and likely other cardiac pathologies. It also underscores the importance of simultaneously observing the behavior of collections of cells, which in the case of heart failure display a wide variety of failure modes. This leads to insights and conclusions that could not possibly be drawn from the study of single cells, for example the realization that multiple failure modes within a single diseased heart defy a single solution. This has implications for designing therapies based on cellular mechanisms contributing to systolic dysfunction. The results also suggest that the asynchronous behavior of Ca release within single cells and between adjacent cells may have a common origin. However, the precise consequences of this remain to be disclosed by rigorous experimentation.
This work was supported by the National Institutes of Health (R01HL070828), and the American Heart Association, Greater Los Angeles Affiliate 0755091Y