Gain and Cardiac E-C Coupling
Revisited and Revised
See related article, pages 590–597
A phenomenon that has fostered much experimental investigation and theoretical speculation is “gain” in cardiac E-C coupling. So-called “macroscopic” or “whole-cell” gain may be defined as the ratio of the total flux through the SR Ca2+ release channels (RyR) to that through the L-type Ca2+ channels (LCC). Experimentally, gain was found early on to be relatively high, and this observation, together with the seemingly incompatible fact that Ca2+-induced Ca2+ release (CICR) is normally tightly controlled in cardiac muscle, led to the development of the modern understanding of cardiac E-C coupling, the “local control” theory. Gain reflects not only the operation of the fundamental processes that underlie normal E-C coupling, but also those involved in important pathological conditions of the heart, particularly those produced by uncontrolled SR Ca2+ release, such as triggered arrhythmias.1
In the heart, it can be said that “not all Ca2+ currents (ICa) are created equal”; ICa at negative potentials is much more efficacious in triggering SR Ca2+ release than is equivalent (peak) ICa at positive potentials. Therefore, gain decreases as activating voltage is made more positive. The conventional and widely accepted explanation of this phenomenon is based on the voltage-dependence of the single-channel (unitary) Ca2+ current (iCa), and was stated succinctly by Stern and his colleagues,2 “gain decreases with voltage because the efficacy of the L-type current to trigger release from the RyR depends on the unitary current of the L-type channel, which decreases as the calcium reversal potential is approached”. Indeed this is a cornerstone of the local control theory of cardiac E-C coupling. However, early on it had been recognized3 (also by Stern) that “the discrepancy between these two curves (Ca2+ current and SR Ca2+ release) is actually a clue to the fact that the number of sarcolemmal channels activated, on the one hand, and the magnitude of their unitary current, on the other, play fundamentally different roles in controlling calcium release… ”. In this issue of Circulation Research, 15 years later, Altamirano and Bers4 report real progress in defining experimentally for the first time the different roles of these 2 factors, the number of L-type Ca2+channels activated (N*Po), and the magnitude of the Ca2+ current through each (iCa). In the process, they have overturned the conventionally held notion about the origin of macroscopic “gain”, and have shown that E-C coupling, under physiological conditions, is dominated by different rules than just those that are known to govern the isolated interaction of a single L-type Ca2+ channel and nearby SR Ca2+ release channels. When it comes to macroscopic gain of cardiac E-C coupling, it turns out that having neighbors, at least of the L-type Ca2+ channel variety, makes a big difference.
The conceptual framework of cardiac E-C coupling has been well established by numerous studies; Ca2+ release through ryanodine receptors (RyR) at individual cardiac dyads (viz a Ca2+ “spark”) is triggered locally, via Ca2+-induced Ca2+ release, specifically by the Ca2+ that have entered the cell through a single, coassociated L-type Ca2+ channel (LCC).5,6 An unknown number, but several at least, RyR become activated, perhaps in “concerted” fashion. The control of SR Ca2+ release is thus exerted in the subspace between LCC and RyR, and in the crystal-lattice array of RyR. The whole cell Ca2+ transient that activates contraction arises from the spatial and temporal summation of individual Ca2+ sparks that have diffused outward to fill the cytoplasm. The whole cell Ca2+ current is given by the well known equation: iCa=N*Po*iCa. Until now, experimental examination of the details of Ca2+ release triggering by L-type Ca2+ channels has almost invariably involved using conditions in which the number of active LCCs (N*Po-) was markedly reduced, through the use of Ca2+ channel antagonists (reduce N), or by working at the foot of the channel voltage-activation curve (where Po is small). This is necessary as a practical expedient, as it vastly reduces the number of Ca2+ sparks that are elicited by depolarization, and makes accurate counting of Ca2+ sparks possible. Most importantly, it also makes tractable the analysis of the relationship between unitary Ca2+ current and Ca2+ sparks because it creates a situation in which the total Ca2+ current at a given dyad is only iCa. These conditions are ideal, and necessary for examining the rules of the intermolecular interaction between a single LCC and the RyR. These conditions preclude observation of any effect that the number of active channels might have on E-C coupling or gain. Under these conditions, several phenomena have been established. First, the probability of triggering a Ca2+ spark (designated PS) has been found to vary proportionally to the square of the amplitude of the unitary Ca2+ current (iCa).7 This is a highly satisfying result that is consonant with the known dynamics of [Ca2+] in the subspace near an open channel and the Ca2+-dependence of RyR activation8 (its likely that 2 Ca2+ are required to activate a RyR and initiate Ca2+ release). Also under these same conditions, all available studies suggest that spark production is linearly related to the probability of L-type Ca2+ channel opening; sparks are generally triggered by the opening of only 1 L-type Ca2+ channel. Once a spark is triggered at a dyad, that dyad becomes refractory, and the probability of another spark being triggered recovers relatively slowly.9 These are the rules of the LCC-RyR interaction when the probability of LCC activation is low.
The experimental situation described above is not that which obtains during normal E-C coupling; each dyad may contain perhaps a dozen L-type Ca2+ channels, and strong depolarization activates most. In their new work,4 Altamirano and Bers took a different experimental approach to assessing the voltage-dependence (and the issue of gain) of cardiac E-C coupling in cardiac myocytes. With their approach it was not necessary to reduce vastly the number of active Ca2+ channels. In fact, their goal was to separate, for the first time, the influence of active channel number (N*Po) and the magnitude of iCa on the voltage dependence of cardiac E-C coupling. To avoid the necessity of counting sparks at low probability, they measured Ca2+ “spikes”, which are a rather direct indication of SR Ca2+ release.10 Ca2+ spikes are obtained by “trapping” released Ca2+ with high concentrations of Ca2+ chelator (EGTA) and observing local (dyadic) [Ca2+] with a fast, low-affinity Ca2+ indicator (Oregon green-5N). They were thus free of Ca2+ channel blockers and able to use a broad range of membrane voltage. The amplitude of iCa was controlled by the rather simple expedients of rapid changes in extracellular [Ca2+], or use of different repolarization test potentials. With these protocols, N*Po is always the same at the times of measurements. In complementary experiments, N*Po was then changed without changes in iCa, by the standard technique of inducing steady-state channel inactivation through the use of different holding potentials, before a standard activating pulse.
The first experiments showed that increasing N*Po decreased gain, defined as the active dyads per unit of LCC current (active junctions/ICa). Unitary LCC current was constant in these experiments, so this new result seems unequivocal. They suggest that increasing N*Po decreases gain because increased channel openings become “redundant”, in the sense that such openings cannot trigger additional sparks at the dyad (which becomes refractory, after the first one).
The relationship between magnitude of iCa and SR Ca2+ release, with constant N*Po was investigated next. Under these experimental conditions, the first surprising result was that SR Ca2+ release, as given by the fraction of active dyads, was sensitive to the magnitude of iCa only when iCa was less than its “normal” value at 0 mV (ie, when external [Ca2+] was ≤1.0 mmol/L). Increasing iCa, by elevation of [Ca2+] to 10.0 mmol/L, did not increase SR Ca2+ release, contrary to conventional expectation. Whether iCa was changed through the use of driving force or changed external [Ca2+], the macroscopic gain (active junctions/ICa), decreased as iCa increased.
The question should be asked: How is it possible that the probability of an active junction (ie, one that produces a spark), is in fact proportional to (iCa)2, as shown earlier,7 and, at the same time, that increasing iCa actually decreases macroscopic gain?
Consideration of the simple statistics of spark production in a dyad provides some insight to this question. Assume the “rules” of the isolated LCC-RyR interaction as given above, Ca2+ influx through one LCC can trigger a spark after 2 Ca2+ bind to an RyR, and only 1 spark can be triggered at a dyad. Subspace [Ca2+] is proportional to iCa. When one LCC is open (n=1.0, Po=1.0), the probability of an active junction is proportional to (iCa)2 (thin line, Figure, A), in agreement with experimental data.7 Note that for RyR activation, and hence spark production rate to be proportional to (iCa),2 the subspace [Ca2+] (proportional to iCa) must be well below the KD for the binding of 2 Ca2+. Hence the maximum probability considered is 0.5. Now assume that a dyadic junction contains 12 LCC, and that depolarization to 0 mV opens all of them (n=12, Po=1.0). This describes a hypothetical case in which all LCC of the dyadic junction are open at the peak ICa. The probability of the junction being active (Pj), or producing a spark, is calculated as: Pj=1.0−(1.0−PS)N*Po, (where PS=iCa)2, as above. The quantity, (1.0 − PS), is the probability that a spark will not be triggered by the opening of a particular LCC. When raised to the power, N*Po, it is the probability that no spark will be produced when N*Po channels are open. This is then subtracted from 1.0 to obtain the probability that the junction will be active. In the case of more than 1 open LCC at peak ICa of a junction, the probability of the junction producing a spark is much higher than when only 1 LCC is open. “Macroscopic” gain (Figure, B) is calculated as PJ/ICa, where ICa=N*Po*ica. Gain is linearly related to iCa for the single open LCC, as expected (thin line, Figure, B). In qualitative agreement with the new data of Altamirano and Bers, gain at a given iCa is reduced when N is increased, and most significantly, gain can decrease when iCa increases (over the appropriate range of iCa), even though PS is proportional to (iCa)2 for all iCa. (The decrease in gain is most evident with the larger N, and occurs at a lower iCa).
In summary, Altamirano and Bers have certainly disproved any idea that the characteristics of macroscopic gain derive entirely from the voltage-dependence of the the unitary Ca2+ channel current. Macroscopic gain is clearly influenced by the number of active channels in the cluster, in ways not appreciated fully before. Nevertheless, the experimental results are not inconsistent with the previous findings that the probability of triggering a Ca2+ spark is a function of the unitary Ca2+ current. Thus, the “local control” theory of cardiac E-C coupling3 remains intact, but we do have a new, quite different, explanation for the phenomenon of macroscopic gain.
Sources of Funding
The author is supported by NIH grant HL 73094.
The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association.
Ter Keurs HEDJ, Boyden PA. Calcium and arrhythmogenesis. Physiol. Rev. 2007; 87: 457–506.
Stern MD, Song LS, Cheng H, Sham JS, Yang HT, Boheler KR, Rios E. Local control models of cardiac excitation-contraction coupling. A possible role for allosteric interactions between ryanodine receptors. J Gen Physiol. 1999; 113: 469–489.
Altamirano J, Bers DM. Voltage-dependence of cardiac excitation-contraction coupling: unitary Ca2+ current amplitude and open channel probability. Circ Res. 2007; 101: 590–597.
Lopez-Lopez JR, Shacklock PS, Balke CW, Wier WG. Local calcium transients triggered by single L-type calcium channel currents in cardiac cells. Science. 1995; 268: 1042–1045.
Cannell MB, Cheng H, Lederer WJ. The control of calcium release in heart muscle. Science. 1995; 268: 1045–1049.
Santana LF, Cheng H, Gomez AM, Cannell MB, Lederer WJ. Relation between the sarcolemmal Ca2+ current and Ca2+ sparks and local control theories for cardiac excitation-contraction coupling. Circ Res. 1996; 78: 166–171.
Zahradnikova A, Zahradnik I, Gyorke I, Gyorke S. Rapid activation of the cardiac ryanodine receptor by submillisecond calcium stimuli. J Gen Physiol. 1999; 114: 787–798.