Calcium flux balance across cell membranes in the heart: important unanswered questions with implications for the role of ryanodine receptors

This Opinion article makes the case that current widely held views concerning interpretations of experimental observations on calcium handling in the heart are oversimplified and may cause us to miss opportunities for developing new therapeutic agents for heart failure and arrhythmias. Unanswered questions considered here concern aspects of calcium flux balance across cardiac membranes, particularly the sarcolemma and sarcoplasmic reticulum (SR). As discussed in detail below, this calcium flux balance may determine the importance of any change in the responsiveness of the ryanodine receptor (RyR2) Ca2+ release channels in the SR membrane. The aim of this Opinion article is to encourage additional experiments to be done to clarify our understanding of these mechanisms. In tackling these questions, the first point to address concerns the particular actions of caffeine on cardiac myocytes. Rapid application of caffeine leads to an increase in the amplitude of Ca2+ transients (CaTs) triggered by action potentials (APs), followed by a decline which is accompanied by a fall in SR Ca2+ content (Fig. 1A, adapted from Greensmith et al., 2014; reviewed in Eisner et al., 2017). What is cause and effect? Might some or all of the reduction in SR Ca2+ content caused by caffeine have occurred whether or not the myocyte was stimulated to activate CaTs? Are the actions of caffeine specific to this molecule or can they be generalized? The careful and important experiments of Eisner and colleagues were made at or close to body temperature on dog and sheep cardiac ventricular myocytes which are expected to be relevant to humans. As shown in Fig. 1B it was argued that '(i) The increase of RyR opening increases the amplitude of the Ca transient (ii) leading to increased Ca efflux (iii) and a decrease of SR Ca content (iv) which returns the amplitude of the Ca transient to control levels (v).' The italics have been added to emphasize the core of the argument that caffeine actions during CaTs are essential for the decline in SR content. This argument may be taken to mean that the effect of an agent to increase CaT amplitude by an increase in RyR opening cannot be maintained because there is a compensation mechanism in which any increase in CaT amplitude necessarily results in an increased extrusion of Ca2+ and a consequent fall in SR Ca2+ content until the amplitude of CaT returns to pre-caffeine levels. The increase in Ca2+ extrusion is expected to be driven by sodium–calcium exchange (NCX) in its 'normal' mode with Na+ influx and Ca2+ efflux (3 Na+ for each Ca2+) across the sarcolemma. An important aspect of this argument is that in the steady state the balance between Ca2+ entry and Ca2+ extrusion across the sarcolemma must be maintained. The interpretation of the effects of caffeine was extended to include all agents acting on RyR2s to cause enhancement of CaTs and contractions triggered by APs in cardiac myocytes (Eisner et al., 2017). While the observations shown in Fig. 1 can be explained by increased Ca2+ extrusion via NCX during the CaT when CaT amplitude is increased, is this the only possibility? Might there be another mechanism that at least contributes, or perhaps even provides an alternative explanation? High concentrations of caffeine (commonly 10 mm) are frequently used to empty the SR of Ca2+ even in the absence of stimulated APs, as discussed in Eisner et al. (2017), and used as an experimental tool in Greensmith et al. (2014). Although the concentration of caffeine used in the experiment shown in Fig. 1 was much smaller (0.5 mm), can we be sure that this concentration of caffeine is without the effect commonly seen in unstimulated cells to cause a fall in SR Ca2+ by a mechanism that does not depend on CaTs? Caffeine crosses the sarcolemma to exert its effects on SR content very rapidly. This very membrane-permeant molecule also crosses the SR membrane and has been suggested to influence RyR2 from the luminal side (Kong et al., 2008) as discussed in more detail later. In Fig. 1, there appears to be an approximately exponential decline in both CaT amplitude and SR Ca2+ content over the period of caffeine exposure. Might the approximately exponential fall in SR Ca2+ have occurred even without CaTs triggered by APs? More weight is given to this possibility by the simultaneous observations of cytosolic Ca2+ shown in the top trace of Fig. 1, since there seems to be a small blip of increased Ca2+ occurring just after the start of caffeine exposure between the first and second CaTs, and this appears to be followed by a barely perceptible exponential fall. A line has been added to the original observations marking the time of caffeine exposure and positioned just below the resting cytosolic Ca2+ level to guide the eye in detecting this apparent small fall in cytosolic Ca2+ after the initial rise at the start of caffeine exposure. There is an approximate parallel between the suggested effects of caffeine on the cytosolic Ca2+ and the clear decline in SR Ca2+ content. An effect of caffeine that occurs whether or not APs are stimulated, and therefore continues to occur between CaTs in stimulated cells, may be described for convenience as a 'diastolic' effect, although it is recognized that contractions are not measured and will show a slower time course. A diastolic effect of caffeine to cause a fall in SR Ca2+ would itself reduce CaT amplitude. The change in SR Ca2+ is shown enlarged in Fig. 1C to aid discussion. Can a 'diastolic' effect of caffeine that does not depend on CaTs triggered by APs provide a complete explanation of the fall in SR Ca2+ shown in Fig. 1? This seems unlikely at the caffeine concentration used for these experiments which produced such a large (approximately fourfold) increase in cytosolic CaT amplitude. Indeed, observations shown in Fig. 1D made by Greensmith et al. (2014) in the same cell as Fig. 1A provide evidence for additional Ca2+ extrusion via NCX during the largest CaT (marked with an arrow). The AP accompanying the first enlarged CaT shows an elevation of the plateau at times slightly greater than 20 ms, consistent with an additional depolarization driven by inward NCX current associated with Ca2+ extrusion (Fig. 1D, also marked with an arrow). These observations support the proposal that under these conditions increased extrusion of Ca2+ via NCX is a contributor to the observed progression of the fall in SR Ca2+. However, the observations do not preclude the possibility that there is an additional fall in SR Ca2+ between CaTs. The relative contribution of these two mechanisms remains to be explored, and the case is made here for an urgent need to carry out experiments to test for these possibilities. It would be interesting to see AP shape for the fourth or fifth CaT in Fig. 1 where the increase in CaT amplitude is more modest (a doubling or slightly less). Figure 1E shows another example from Greensmith et al. (2014) where the peak effect of caffeine on CaT amplitude was about a doubling and in this case there was no elevation of the plateau over at least the first 100 ms when re-uptake of Ca2+ into the SR is expected to be at its greatest. There was, however, the appearance of a 'late' plateau phase at times ∼200 ms or greater after the AP upstroke, consistent with additional depolarizing current associated with Ca2+ extrusion via NCX at this stage (arrow). It seems that the balance of caffeine actions occurring during or between CaTs may vary with the extent of increase in CaT amplitude, and therefore the caffeine concentration. Calcium handling in the heart is complex (Eisner et al., 2017; Terrar, 2023), but to aid discussion a simple model is presented in Fig. 2. This includes the key elements of transport pathways which influence calcium flux balance across SR and sarcolemmal membranes. These processes determine the fate of Ca2+ ions during the return to baseline of each CaT. Re-uptake of Ca2+ into the SR by SERCA (sarcoplasmic/endoplasmic reticulum Ca2+-ATPase) will be opposed by mechanisms to extrude Ca2+ from the cytosol across the sarcolemma. For most of the AP and at the resting potential the balance between Ca2+ extrusion and Ca2+ re-uptake into the SR is mainly determined by competition between NCX and SERCA (Fig. 2A). There will be a role for a surface membrane Ca2+-ATPase, but although this is important for calcium balance over long periods, its contribution during CaTs is relatively small (see references in Eisner et al., 2017) and constant. It is therefore not included in the following discussion. Unlike SERCA, the effectiveness of NCX in extruding Ca2+ will depend on the membrane potential as well as the concentrations of Na+ and Ca2+ on either side of the sarcolemma. As far as the membrane potential component is concerned, at positive potentials early in the AP plateau the driving force for Na+ entry (and therefore Ca2+ extrusion) is especially small and this driving force increases progressively during repolarization reaching a maximal level at the resting potential. The balance between Ca2+ extrusion via NCX and Ca2+ uptake via SERCA is therefore expected to vary during the time course of the AP, and from these considerations alone it seems that SERCA may be dominant early in the plateau. However, an important additional factor to be taken into account is evidence for 'reverse mode' NCX reported by several authors including Lipp and Niggli (1994). The mechanism depends on Na+ entry via voltage-gated Na+ channels during the upstroke of the AP leading to Na+ accumulation beneath the sarcolemma to the extent that the direction of NCX is for Na+ efflux and Ca2+ entry into the cytosol. This reverse mode NCX has been thought to be especially important as a component of excitation–contraction coupling. It is argued here that the existence of reverse mode NCX is also crucial in determining the efficiency of Ca2+ re-uptake into the SR, particularly close to the peak of the CaT. NCX cannot oppose SERCA while it is in reverse mode (Fig. 2B). Lipp and Niggli argue that the effects of reverse mode NCX last for ∼20 ms. Presumably there will be period during which NCX is neutral or negligibly small when SERCA is still unopposed before a gradual return to normal mode when NCX can start to make a significant impact on the balance between Ca2+ extrusion and uptake. The timing of the period when SERCA dominates remains to be determined. Difficulties in measuring CaT using fluorescent probes (which must bind Ca2+ to show a detectable signal) mean that the timing of the peak of the CaT accompanying APs remains uncertain, but it is suggested here that the time to peak is ∼10 ms. From the arguments presented above the following hypothesis is proposed: (1) reverse mode NCX during at least the first ∼20 ms after the upstroke of each AP allows uptake of Ca2+ into the SR by SERCA without opposition by 'normal' mode NCX at this time, thus ensuring maximal efficiency of reload of SR Ca2+ at the peak of the CaT triggered by the AP (peak of CaT at about 10 ms); and (2) 'diastolic' effects of caffeine (between CaTs) make an important contribution to the decline in SR Ca2+ content observed with this agent. It is hoped that this Opinion article will provoke experiments to test the above hypothesis, and indeed the author intends to seek support for such an undertaking. A suggested experimental test is to apply caffeine under the same conditions with and without stimulation of APs. In the experiments without APs during caffeine the cells would nevertheless be stimulated up to the point of caffeine application to achieve the same level of SR Ca2+ load as experiments where stimulation of APs is continued during caffeine. The following points should be taken into account in designing future experiments. It is essential that the experiments are carried out at or very close to body temperature. This is important because at room temperature the timing of the CaT peak relative to the AP, and therefore the balance between Ca2+ uptake and extrusion mechanisms, will be changed. Any delay in CaT relative to the AP would favour Ca2+ extrusion because more negative membrane potentials at any stage of the CaT would enhance Ca2+ extrusion by 'normal' mode NCX. It would also be beneficial to use Ca2+ probes with the lowest affinity consistent with a good signal to noise ratio to avoid unnecessary slowing of the CaT (for similar reasons as for the avoidance of room temperature). It is important to measure both SR and cytosolic Ca2+. Although any 'diastolic' effect of caffeine to increase Ca2+ release from the SR is expected to cause a rise in cytosolic Ca2+, this may be hard to detect because at the resting potential 'normal' mode NCX is maximally effective. Therefore, as Ca2+ is released from the SR by caffeine a large proportion of the additional cytosolic Ca2+ will be rapidly extruded across the sarcolemma by NCX (unlike the situation at positive membrane potentials early in the AP plateau, as discussed above). However, a luminal SR Ca2+ probe would nevertheless detect any fall in SR Ca2+ content. Whatever the outcome of the above experiment it may be questioned whether actions of caffeine provide a universal model for the actions of all agents which increase CaT accompanying APs by an effect to increase Ca2+ release via RyR2. As mentioned above, caffeine is an unusually membrane-permeant molecule and it has been suggested that a major site of action is within the lumen of the SR promoting 'leakage' of Ca2+ (Kong et al., 2008) in addition to enhancement of Ca2+ release during APs. Caffeine markedly reduced the threshold for activation of RyR2 by luminal Ca2+ (Kong et al., 2008). Caffeine has also been shown to increase RyR2 opening by cytosolic Ca2+, and very low cytosolic Ca2+ concentrations (80 pm) suppressed caffeine effects (Sitsapesan & Williams, 1990). These observations and those of Porta et al. (2011) show that RyR2 opening may increase because of combined effects of caffeine to increase both the cytosolic and luminal sensitivity to Ca2+. The mechanisms of action of caffeine may therefore differ from those of many other agents activating RyR2s. An example of another agent which has been argued to increase CaT amplitude at least in part by a potentiating action on RyR2 is cADP-ribose (cardiac actions reviewed in Terrar, 2023 and a more general review in Galione & Chuang, 2020). This Ca2+-mobilizing agent is a polar molecule which is thought to act in the heart at a cytosolic site on RyR2 by a mechanism which probably involves FK binding proteins (as discussed in Terrar, 2023). Over the first 3 min of exposure to cADP-ribose there was a maintained effect of cADP-ribose on CaT amplitude and contractions in guinea pig cardiac myocytes although there were no changes in L-type calcium channel (LTCC) or SR Ca2+ load tested by exposure to a high concentration of caffeine (Macgregor et al., 2007). In rat permeabilized myocytes cADP-ribose increased Ca2+ spark frequency over the first 3 min but again there appeared to be no increase in Ca2+ load both from the lack of change in Ca2+ spark amplitude and from the lack of change in response to a high concentration of caffeine. A longer exposure (10 min) to cADP-ribose did increase spark amplitude and caused a more rapid spark decay consistent with enhancement of SERCA activity. It remains to be established whether this additional effect at long exposure times is secondary to a prolonged increase in Ca2+ release. It seems that the effects of any agent with an action on Ca2+ handling are unlikely to be simple. cADP-ribose contributes to the physiological effects of beta-adrenoceptor stimulation, and excessive levels of cADP-ribose have been found to be arrhythmogenic (Terrar, 2023). Under physiological conditions cADP-ribose acts in concert with another Ca2+-mobilizing agent, NAADP, with complementary actions (see Terrar, 2023). In the heart both can be synthesized by CD38. The above discussion may be relevant to other agents which increase the sensitivity of RyR2 to cytosolic Ca2+. Ca2+ flux balance across SR and sarcolemmal membranes also has implications for analysis of the effects of beta-adrenoceptor stimulation. The relative contribution of changes in RyR2 sensitivity as compared to increases in LTCC and the regulation of SERCA by phospholamban remains to be fully established (see e.g. Houser, 2010; O'Brien et al., 2015), and is strongly influenced by Ca2+ flux balance. This merits discussion in a longer article, and cannot be included here. Tetracaine has been compared to caffeine since it reduces CaT amplitude and this has been suggested to increase SR Ca2+ as a consequence of reduced Ca2+ extrusion across the sarcolemma during CaTs (Eisner et al., 2017). However, actions of tetracaine may include a 'diastolic' effect on SR Ca2+ which is not dependent on CaTs (see fig. 6B in Greensmith et al., 2014). Tetracaine increased SR Ca2+ content even in permeabilized cells, where the balance between Ca2+ extrusion across the sarcolemma via NCX and SR Ca2+ uptake by SERCA cannot be a factor (Zima et al., 2008). From the above arguments it seems that the experiments proposed here are essential to distinguish between different interpretations of caffeine actions, and to determine whether caffeine provides a universal model for actions of agents which increase the sensitivity of RyR2 to cytosolic Ca2+. Revision of general models of calcium handling in the heart would provide new insights concerning the control of contraction and arrhythmias. This in turn would aid development of appropriate clinical treatments. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article. No competing interests declared. D. A. T.: conception or design of the work; drafting the work or revising it critically for important intellectual content; final approval of the version to be published; agreement to be accountable for all aspects of the work. None.