© 1998 by European Society of Cardiology
Copyright © 1998, European Society of Cardiology
The control of Ca release from the cardiac sarcoplasmic reticulum: regulation versus autoregulation
Department of Veterinary Preclinical Sciences, University of Liverpool, Liverpool L69 3BX, UK
* Corresponding author. Tel.: +44 (151) 794 4228; Fax: +44 (151) 794 5347; E-mail: eisner@liv.ac.uk
Received 22 January 1998; accepted 24 February 1998
 |
Abstract |
|---|
 Top Abstract 1 Introduction2 Calcium-induced calcium...3 What regulates e-c...4 Is modulation of...References |
|---|
This review discusses the mechanism and regulation of Ca release from the cardiac sarcoplasmic reticulum. Ca is released through the Ca release channel or ryanodine receptor (RyR) by the process of calcium-induced Ca release (CICR). The trigger for this release is the L-type Ca current with a small contribution from Ca entry on the Na–Ca exchange. Recent work has shown that CICR is controlled at the level of small, local domains consisting of one or a small number of L-type Ca channels and associated RyRs. Ca efflux from the s.r. in one such unit is seen as a ‘spark’ and the properties of these sparks produce controlled Ca release from the s.r. A major factor controlling the amount of Ca released from the s.r. and therefore the magnitude of the systolic Ca transient is its Ca content. The Ca content depends on both the properties of the s.r. and the cytoplasmic Ca concentration. Changes of s.r. Ca content and the Ca released affect the sarcolemmal Ca and Na–Ca exchange currents and this acts to control cell Ca loading and the s.r. Ca content. The opening probability of the RyR can be regulated by various physiological mediators as well as pharmacological compounds. However, it is shown that, due to compensatory changes of s.r. Ca, modifiers of the RyR only produce transient effects on systolic Ca. We conclude that, although the RyR can be regulated, of much greater importance to the control of Ca efflux from the s.r. are effects due to changes of s.r. Ca content.
|
1 Introduction |
|---|
|
TopAbstract1 Introduction2 Calcium-induced calcium...3 What regulates e-c...4 Is modulation of...References |
|---|
Contraction of cardiac muscle is initiated by a systolic rise of intracellular calcium concentration ([Ca2+]i). Changes in the magnitude of this increase of [Ca2+]i are the major mechanism for regulating the force of contraction of the heart and cardiac output. Most of the systolic increase of calcium occurs as a result of release from an intracellular store, the sarcoplasmic reticulum, rather than coming directly from the extracellular fluid. The primary purpose of this review is to discuss recent evidence not only on the mechanism of this Ca release but, most importantly, of how release is regulated. The overall conclusion is that although the release process can be regulated, under many physiological conditions the effects of this are secondary to a form of autoregulation in which the balance of cellular Ca fluxes determine the s.r. Ca content and therefore the release.
In order to put s.r. Ca release into context, it is impossible to ignore completely other mechanisms which control intracellular calcium. However the reader who is interested in a more complete account of cardiac cellular Ca balance is referred to the excellent monograph by Bers [1].
|
2 Calcium-induced calcium release |
|---|
|
TopAbstract1 Introduction2 Calcium-induced calcium...3 What regulates e-c...4 Is modulation of...References |
|---|
This area has been reviewed extensively and will, therefore, only be considered briefly. It has been known for many years that calcium induced release of calcium (CICR) exists in the heart. Evidence for this came, originally, from work showing that elevating calcium in skinned muscle preparations produced contractions due to s.r. Ca release [2–4]. Subsequent work showed the likely subcellular mechanism for CICR to be a Ca-dependent opening probability of the s.r. Ca release channel (now commonly referred to as the ryanodine receptor (RyR)). This is shown most directly by incorporating the RyR into artificial lipid membranes so that the opening and closing of individual channels can be observed. Under these conditions the open probability of RyR is increased by raising [Ca2+]i on the cytoplasmic face [5].
2.1 CICR in intact cells and the trigger role of the L-type Ca current
The evidence reviewed above demonstrated the existence of CICR. Such studies, however, suffer from the fact that, in work on either skinned cells or isolated s.r. channels, the solutions used are not identical to the cytoplasm of the intact cell. It was therefore important to demonstrate CICR in intact cells. The following evidence has established this. (i) Removal of external Ca immediately abolishes contraction despite the fact that the s.r. still contains Ca which can be released by manoeuvres such as the application of caffeine [6, 7](ii) Flash photolysis induced release of calcium from a ‘caged’ calcium compound produces a contraction which requires s.r. Ca release [7–10]. (iii) At least under some conditions, the voltage-dependence of the systolic Ca transient is similar to that of the L-type Ca current. Thus, as the magnitude of depolarization is initially increased, there is an increase in the magnitude of the current (as more channels open). However as depolarization is further increased then the Ca current decreases (as the driving force for Ca entry into the cell declines). This is accompanied by first an increase and then a decrease of the systolic Ca transient. On repolarization the driving force for Ca entry is instantaneously restored whereas the voltage-dependent closing of Ca channels takes a few ms to occur and there is therefore an entry of Ca ions which is associated with a Ca transient. This correspondence between Ca entry into the cell and s.r. Ca release is perhaps the strongest evidence for CICR [11–13].
2.2 Does Ca entry by routes other than the L-type Ca current trigger CICR?
The work reviewed in the previous section suggests that the L-type Ca channel is the trigger for s.r. Ca release. It has, however, been suggested that Ca entry on the Na–Ca exchange may also trigger release.
2.2.1 Ca entry on Na–Ca exchange produced by a local increase of Na
Leblanc and Hume [14]showed that, even when the L-type current was inhibited, depolarization could produce s.r. Ca release. This Ca release required Na entry into the cell through a TTX-sensitive channel and also the presence of extracellular calcium. They explained this as resulting from (i) Na entry through the TTX-sensitive channels followed by (ii) Na efflux and Ca entry on Na–Ca exchange. As pointed out subsequently, the expected Na entry via the TTX-sensitive channel would produce a very small increase of mean cytoplasmic Na concentration which would be insufficient to affect the Na–Ca exchange. If the Na–Ca exchange is to sense a significant increase of [Na+]i then the Na entry must be into a small fraction of the cell. It was therefore suggested that Na influx would occur into a sub-sarcolemmal or ‘fuzzy’ space [15]. The idea that the Na concentration near the membrane can be different from that in the bulk cytoplasmic solution has been supported by direct electron microprobe measurements [16]as well as by measuring changes in the reversal potential of the Na–Ca exchange current (and therefore of subsarcolemmal Na concentration) during changes of Na-K pump activity [17](see [18]for review). There is controversy as to the extent to which this Ca release in the absence of the L-type Ca current is an artefact of loss of voltage control [19, 20]. Subsequent studies in this area have also produced contradictory results with one reproducing the dependence of the systolic Ca transient on Na entry [21]and others finding no effect [22, 23].
2.2.2 Ca entry on Na–Ca exchange due to depolarization
The work of Leblanc and Hume [14]suggests that the Ca entry on Na–Ca exchange is initiated by a local increase of Na concentration. In contrast, other work has suggested that depolarization may directly stimulate Ca entry on the exchanger. As a consequence of the fact that the Na–Ca exchange transports 3 Na+ ions in exchange for each Ca2+, depolarization will produce Ca entry and Na efflux. It has long been known that changes of contraction or [Ca2+]i can be produced by the effects of membrane potential on Na–Ca exchange [24–26]. The issue which is still controversial is whether this Ca entry can trigger Ca release from the s.r. which is comparable to that produced by the L-type Ca current.
Some experimental work has shown that depolarization can still produce s.r. Ca release even when the L-type channel is apparently inhibited. Furthermore the voltage-dependence of contraction or systolic [Ca2+]i does not always show the bell-shaped behaviour expected (see above) if Ca entry via the L-type current is the main mechanism for triggering release [27–29]. In many experiments, the magnitude of the Ca transient increases with depolarization up to a maximum, saturating level. It has been suggested that the Na–Ca exchange dependent triggering may not be seen under the room temperature conditions used in many studies [30, 31]. In contrast to these studies, it has been reported that (admittedly at room temperature) the deviation from a bell-shaped relationship between voltage and contraction was due to voltage clamp errors [20]. However, with the exception of this last report, there is reasonable agreement that Ca entry on Na–Ca exchange can trigger s.r. Ca release to some extent. What is therefore needed is an assessment of the quantitative importance of this mechanism (compared to the L-type channel) in triggering release. A recent study using transgenic mice overexpressing the Na–Ca exchange has found that, even when the Na–Ca exchange is overexpressed up to nine fold, Ca entry on the exchange produces a much smaller systolic rise of [Ca2+]i than does the L-type current [32]. A quantitative assessment of the relative efficacies of Na–Ca exchange and Ca current has been provided in a recent paper which has measured not only the Ca release produced by the Na–Ca exchange but also the magnitude of this Na–Ca exchange current itself [33]. This showed that, for a given amount of Ca entering the cell, the Na–Ca exchange was only 25% as efficient as the L-type channel in promoting Ca release from the s.r. Given that at most potentials the Ca entry via the L-type current is much greater than that via the exchange, it was concluded that Ca entry on Na–Ca exchange would have a negligible role as a trigger of s.r. Ca release. This may be because the Na–Ca exchange is excluded from the release ‘microdomains’ which contain the L-type channel and RyR [34].
2.3 Is there a role for voltage-activated release of Ca from the s.r?
Ferrier and Howlett [35]showed that depolarization could produce Ca release even under conditions when Ca entry through both the Ca channel and Na–Ca exchange were apparently inhibited. They therefore suggested the possible existence of a voltage-dependent Ca release mechanism perhaps similar to that which is known to produce s.r. Ca release in skeletal muscle. This release process was inactivated by prior depolarization perhaps explaining why it has not been observed in much previous work which has typically used holding potentials of –40 mV in order to inhibit the Na current. Further work on this question has recently appeared [36]suggesting that the voltage-dependent release mechanism requires cAMP. Interesting though this data is, it should be noted that it is sometimes difficult to be sure that Ca entry has been completely inhibited. There is always the concern that the residual small percentage of unblocked Ca entry can trigger Ca release from the s.r. One must assess the maximum level of this residual Ca entry and to what extent this could trigger Ca release. It is also important to note the physiological state of the cell. Thus it is well known that, as the s.r. Ca content is increased, it becomes easier to trigger release of Ca from the s.r. For example brief pulses which were unable to trigger Ca release from a lightly loaded s.r. could release Ca under more heavily loaded conditions [37]. Furthermore, when the s.r. Ca load is increased sufficiently, Ca release occurs spontaneously [38, 39]. It is therefore possible that, under very heavily loaded conditions, a very small residual Ca entry is sufficient to cause release. Under conditions of elevated [cAMP], it is likely that phosphorylation of phospholamban will result in an increased s.r. Ca content. Furthermore phosphorylation of the RyR may stimulate CICR and the combined effect may be that a very small residual Ca current will produce significant CICR. In the presence of isoprenaline (which like the conditions reviewed above will increase [cAMP]), a decrease of the L-type Ca current produces very little, if any, reduction of the Ca transient [40]and it is therefore likely that even an undetectably small Ca current will trigger Ca release.
Based on all the work reviewed above, it appears most likely that, under normal conditions the L-type current is the major trigger for Ca release from the s.r. although there may be a small contribution from Ca entry on Na–Ca exchange.
2.4 What terminates calcium-induced Ca release?
It has long been appreciated that, at least as originally described, CICR is a positive feedback system. Therefore the Ca which is released from the s.r. will lead to further opening of the release channels and the system will never terminate. In contrast to this prediction, there is good evidence that CICR is graded. Thus (as mentioned above) the magnitude of the systolic Ca transient is graded with that of depolarization and the resulting Ca current. Two types of mechanisms have been proposed to account for the lack of positive feedback. Fabiato showed in skinned muscle that, following the Ca-induced activation of release, there was a subsequent inactivation [4]. This inactivation could be responsible for terminating release although subsequent work on intact cells has not been able to demonstrate such inactivation [9]. Work on RyR incorporated into artificial lipid membranes has found a related mechanism termed ‘adaptation’. [Ca2+]i was elevated quickly by flash photolysis of a caged Ca compound. This produced an initial increase of open probability which then decayed despite the maintained presence of elevated [Ca2+]i [41, 42]. In the initial experiments the development of this adaptation was slow but subsequent work using physiological Mg concentrations found more rapid adaptation [43].
2.5 Sparks and local control theories
A fundamental theoretical advance was made by Stern [44]who pointed out that any system in which there was a common pool for calcium was inherently unstable. In other words stability requires that the opening of one L-type Ca channel triggers one or a small number of ryanodine receptors. On this model the graded nature of CICR comes about by recruiting more and more of these basic units whereas each unit responds in an all or none manner. Furthermore he suggested that release could be terminated by the process of ‘stochastic attrition’ in which the channels close spontaneously (at a rate given by the intrinsic kinetics of the RyR). If there are only a small number of RyR in the pool then there is a finite possibility that, at one instant, all RyR will be closed and the local [Ca2+]i will immediately decay to levels too low to support RyR opening. Of course the Ca-dependent adaptation process described above will accelerate the rate of this attrition and therefore increase the probability that release will be terminated.
The theoretical approach of Stern was confirmed by the introduction of confocal microscopy. Using fluo-3 to measure [Ca2+]i in resting cells with a confocal microscope occasional flashes of light (termed Ca sparks) were observed [45, 46]. The sparks were subsequently shown to arise at sites adjacent to transverse tubules [47, 48]. The importance of these data for modern theories of excitation-contraction coupling is that, as pointed out by Cannell et al. [49], the spark provides both analog and digital gain. The former is due to the fact that the flux through the RyR is much larger than that through the L-type channel. The latter (digital gain) is due to the RyR staying open for longer than the L-type channel. As discussed by these authors, this provides high amplification but without the instability problems referred to above. For a review of work in this area see [50].
|
3 What regulates e-c coupling? |
|---|
|
TopAbstract1 Introduction2 Calcium-induced calcium...3 What regulates e-c...4 Is modulation of...References |
|---|
A scheme for e-c coupling which arises from the work reviewed above is shown in Fig. 1. The obvious potential control points are as follows. (i) The size of the Ca current and therefore that of the ‘trigger’ increase of [Ca2+]i. (ii) The properties of the ryanodine receptor and, in particular, its activation by [Ca2+]i. (iii) The Ca content of the sarcoplasmic reticulum. For reasons which will become obvious below, we will consider these potential control mechanisms in the reverse order.
|
3.1 s.r. Ca content as a regulator of Ca release
Many studies have shown that the size of the systolic Ca transient or contraction is increased under conditions which increase the s.r. Ca content. For a complete discussion of work in this area, the reader is referred to Bers [1].
3.1.1 Measurement of s.r. Ca content
Systematic analysis of the relationship between s.r. Ca content and systolic Ca requires being able to measure s.r. Ca content. This has been achieved in a variety of ways. Perhaps the most direct method is to use the electron microprobe [51]. Another approach is to measure changes of either total or radioactive Ca content of the tissue and look for changes produced by manoeuvres expected to change s.r. Ca content [52, 53]. However, these methods destroy the tissue under study and cannot be used for repeated measurements from the same cell. To overcome this problem, qualitative techniques have been employed. The general principle is to release all the s.r. Ca content into the cytoplasm and then measure the magnitude of the resulting increase of [Ca2+]i. The calcium can be released either by adding caffeine [6, 54, 55]or by rapid cooling [56, 57].
For many purposes a qualitative measure of s.r. Ca content may be adequate. However in some cases it is important to obtain quantitative measurements. Two ways have been used to quantify the caffeine approach. (i) The bulk of the calcium which is released from the s.r. is pumped out of the cell on the Na–Ca exchange. Since the Na–Ca exchange is electrogenic (as a result of transporting 3 Na+/Ca2+), Ca release is accompanied by an inward current [58–60]and therefore the amount of Ca pumped out of the cell can be estimated by integrating the Na–Ca exchange current (after correcting for Ca removal by non Na–Ca exchange mechanisms [60]). (ii) A different approach is to measure the increase of free Ca produced by caffeine and, with a knowledge of the relation between free and total Ca, obtained independently [61], calculate the increase of total Ca produced by caffeine.
Whichever method is used to measure s.r. Ca content, there is general agreement that an increase of s.r. Ca content results in an increased systolic Ca transient. The exact nature of the relationship varies between studies. For example some studies report a linear relationship between Ca release and s.r. Ca content [62]. Other work has, however, reported that the relationship can be steeper or supralinear [61, 63, 64]. A supralinear relationship could arise because, as the intra-s.r. Ca buffers saturate, a given increase of total s.r. Ca content results in a larger increase of free s.r. Ca and, therefore, in a larger driving force for Ca flux into the cytoplasm.
3.1.2 How does s.r. Ca content affect s.r. Ca release?
There are at least two possible explanations: (i) The increase of s.r. Ca content will increase the driving force for Ca efflux from the s.r. and the efflux from the s.r. This effect may be further amplified by the Ca released from the s.r. binding to the cytoplasmic activation site of the RyR. Evidence that this may be important under cellular conditions has recently been provided by work investigating the ability of calcium to release either strontium or calcium from the s.r. It was found that the fraction of the strontium content of the s.r. which was released was independent of load whereas the fraction of Ca increased with the degree of Ca load. This was attributed to Ca released from the s.r. stimulating further release with strontium being unable to mimic this effect [65](ii) In addition, it has been suggested that an increase of s.r. Ca content, by binding to a site at the internal (lumenal) surface of the s.r. may increase the opening probability of the ryanodine receptor [66, 67]. Other workers have, however, suggested that the effect of lumenal Ca, at least in skeletal muscle, is due to Ca accumulating at a cytoplasmic site (in other words as in (i) above) [68]but this interpretation is disputed [67, 69]. For the present purposes, no matter whether the effect of lumenal Ca is direct or mediated via a cytosolic site, an increase of s.r. Ca content can increase Ca efflux from the s.r.
3.2 What regulates the s.r. Ca content?
It is important to realize that the s.r. Ca content depends not only on the properties of the s.r. Ca transport pathways (specifically the ryanodine receptor and the Ca-ATPase) but also on the level of cytoplasmic Ca concentration. s.r. Ca content will therefore be affected indirectly by sarcolemmal Ca fluxes.
3.2.1 Modulation of s.r. pathways affect the s.r. Ca content
Phosphorylation of the regulatory protein phospholamban [70]will remove its inhibitory action on the s.r. Ca-ATPase and thereby increase the accumulation of Ca by the s.r. This accelerates the rate of decay of systolic [Ca2+]i [71, 72]and would be expected to increase s.r. Ca content as a larger fraction of the cytoplasmic Ca will be taken up by the s.r. rather than being pumped out of the cell. Recent work has indeed found that catecholamines increase the s.r. Ca content as measured by the resulting caffeine-evoked increase of [Ca2+]i [40].
3.2.2 Changes of sarcolemmal Ca fluxes affect the s.r. Ca content
Anything which increases Ca entry into the cell or decreases Ca efflux will result in more Ca being made available to the s.r. Ca-ATPase and will thereby increase s.r. Ca content. Thus slowing the Na–Ca exchange (by for example decreasing external Na or increasing intracellular Na) will increase s.r. Ca content [6, 56]. Changes in the rate or pattern of stimulation have long been known to vary the force of contraction of the heart. Thus, in most species, an increase of heart rate increases the force of contraction (positive force frequency relationship) and interruption of regular stimulation by a rest decreases contraction (rest decay). Opposite effects are generally observed in rat heart. These effects on contraction have been correlated with corresponding changes of s.r. Ca content as assessed by a variety of methods [73–75]. An increase in the rate of stimulation will have two opposing effects on cellular Ca balance. (i) There will be an increase of Ca entry into the cell per unit time due to the increased frequency of action potentials. (ii) The increased frequency of systolic Ca transients will increase the Ca efflux from the cell. Which of these effects dominates depends on the species. It has been suggested that the negative force frequency relationship in the rat arises because it has a higher resting [Na+]i than other species and it also has a short action potential [76]. The latter difference means that, in the rat, the membrane potential is essentially fully repolarized at a time when [Ca2+]i is still elevated. Since Ca efflux on Na–Ca exchange is stimulated at negative membrane potentials, this means that there is a net loss of Ca during systole. In contrast, in most other species, the bulk of the Ca transient occurs during the plateau of the action potential and the Ca efflux is therefore less. Evidence in favour of this hypothesis has been presented in voltage clamped cells where the negative systolic Ca-frequency relationship observed with short (100 ms duration) pulses is abolished with longer pulses [77].
3.2.3 Quantitative comparison of sarcolemmal Ca fluxes and s.r. Ca content
This sort of analysis of the factors governing cellular Ca loss and gain is qualitative. Recent work has been able to quantify the Ca fluxes. The Ca entry into the cell via the L-type Ca current can be measured by integrating this current. The Ca efflux on Na–Ca exchange can be estimated from the resulting inward current. It is impossible to measure this current during depolarization due to temporal overlap with the L-type current. Instead, it is best measured from the inward current tail on repolarization [77, 78]. If the pulse duration is kept short then most of the Ca transient and therefore the Ca efflux on Na–Ca exchange will occur at the negative holding potential (particularly since the Na–Ca exchange is stimulated by hyperpolarization). This will minimize the error due to unresolved Ca efflux during depolarization [77]. It should also be noted that this approach ignores any Ca influx on Na–Ca exchange. This method allows one to compare Ca entry and efflux under steady state stimulation. An initial study in rat myocytes found that the influx was 40% of the efflux [77]. Given that the cell must be in Ca balance over the whole stimulation cycle, this result suggests that the cell is gaining Ca between stimuli. This would be consistent with the observed rest potentiation. However more recent work on both ferret [64]and rat [79]finds that the fluxes are balanced such that the amount of Ca which enters the cell during the Ca current is equal to that which is pumped out of the cell during the tail of Na/Ca exchange current. A similar conclusion has been reached by comparing the Ca influx on the L-type current with the efflux measured isotopically on Na–Ca exchange [80].
3.2.4 Changes of s.r Ca release control sarcolemmal Ca fluxes
The mechanism by which s.r. Ca content is regulated can be investigated using the techniques described above. A simple approach is to empty the s.r. using caffeine and then investigate its refilling (see Fig. 2). Qualitative studies have shown that s.r. refilling is accelerated by stimulation. Indeed in many species, unless the cell is stimulated, the s.r. does not refill to maximum levels [64, 81]. The first stimulus (pulse 1) following removal of caffeine results in a small systolic Ca transient due to the decreased s.r. Ca content. This is associated with a larger L-type Ca current and a smaller Ca efflux on Na–Ca exchange than is seen in the steady-state [64]. The increased L-type Ca current is due to decreased Ca-induced inactivation of the Ca current [34, 82, 83]whereas the decreased Na–Ca exchange Ca efflux is simply due to the smaller Ca transient decreasing the degree of activation of the Na–Ca exchange. The changes in these currents mean that, since the cell was in Ca balance in the steady state, the first Ca transient following removal of caffeine results in a net increase of s.r. Ca. The calculated net increase of Ca was found to agree quantitatively with the increase of s.r. Ca content as measured directly from the integral of the caffeine-evoked current. On successive pulses the systolic Ca transient increases in magnitude (due to the increased s.r. Ca content) and this is accompanied by a decrease of the L-type Ca current and increase of Na–Ca exchange efflux to steady-state levels. In the steady state (pulse 9) the Ca gain via the L-type current exactly balances the loss on Na–Ca exchange. This sort of analysis shows that the Ca dependence of the L-type Ca current and the Na–Ca exchange are adequate to account for the regulation of s.r. Ca content. This mechanism (summarized in Fig. 3) for regulating s.r. Ca content also provides an obvious means to change it. An increase of the L-type Ca current or a depression of Na–Ca exchange will lead to an increase of cellular (and s.r.) Ca content. This will increase the size of the systolic Ca transient. This in turn will decrease the Ca current and increase the Ca efflux from the cell until the net Ca efflux equals the influx again and a new steady-state is reached at an elevated systolic Ca transient. We will show below that this flux balance analysis has important implications for manoeuvres which affect the ryanodine receptor.
|
). The middle graph shows the net Ca gain on each pulse and the bottom graph the cumulative Ca gain obtained by adding together the net gains from each pulse. Taken from
|
3.3 Modulation of the ryanodine receptor
Many studies have shown that the Ca fluxes through the ryanodine receptor can be affected by a variety of both endogenous and exogenous ligands. Direct evidence for this comes from studies on the isolated ryanodine receptor either by measuring radioactive Ca flux from vesicles or, even more directly, from electrophysiological measurements of currents through single ryanodine receptors incorporated into artificial membranes. This area has been reviewed comprehensively [84–86].
3.3.1 ATP and other metabolites
ATP increases the opening probability of the RyR [5, 87]. Thus a decrease of [ATP], as occurs during ischaemia, may be expected to depress s.r. Ca release. The acidification which occurs in ischaemia and hypoxia will also depress the open probability of the ryanodine receptor [88, 89]. The decrease of [ATP] will result in an increase of [Mg2+]i which will also depress RyR opening [89, 90]. The effects of the accompanying increase of inorganic phosphate concentration (Pi) are less well understood. In skeletal muscle s.r. vesicles (but not in cardiac s.r.), Fruen et al. [91]reported that Pi increased maximal calcium- activated binding of ryanodine. In saponin-permeabilized myocardium, Pi was found to produce Ca release from the s.r. [92]. However phosphate removal produced a transient increase in spontaneous s.r. Ca release activity. This would be consistent with removal of a phosphate-mediated inhibition of the RyR. On the other hand a recent report has shown that Pi increased the open probability of the cardiac RyR [93]. Whatever the effects of phosphate, the net effects of the metabolite and pH changes which occur in ischaemia are likely to decrease CICR [89].
3.3.2 cADP-ribose
The endogenous pyridine nucleotide metabolite cyclic ADP-ribose (cADPR) was first shown to cause Ca release from a ryanodine sensitive, IP3-insensitive store in sea urchin eggs [94–96]. Subsequent biochemical analysis demonstrated that cADPR is present in many tissues including cardiac muscle [97]raising the possibility that cADPR may act as an endogenous regulator of CICR. In isolated cardiac cells antagonists of cADPR have been shown to reduce the amplitude of contraction [98]suggesting a role for cADPR in potentiating CICR, however, another study failed to demonstrate any effect of cADPR or its antagonist 8-amino-cADPR on the systolic Ca transient [99]and concluded that cADPR does not regulate CICR in intact cardiac myocytes. It now appears that the effects of cADPR on intact ventricular myocytes are temperature dependent [100]and this may be one factor accounting for the lack of any effect observed by Guo et al. [99].
More direct evidence that cADPR acts as a modulator of the RyR receptor was provided by planar lipid bilayer experiments in which the open probability of the RyR was increased by cADPR [95], although this effect is reversed by the presence of ATP [101]. It has also been suggested that cADPR may induce Ca release from the s.r. independently of the RyR [102, 103]. However, it appears that the actions of cADPR may require and/or be regulated by other factors including lumenal and cytosolic Ca, calmodulin, Mg, ATP and even an accessory protein [101, 104–107]. Given the diversity of modulators of cADPR it is clear that interpreting the effects of cADPR requires careful consideration of the conditions within each set of experiments.
3.3.3 Phosphorylation of RyR
The ryanodine receptor can be phosphorylated by β-adrenergic stimulation in rat ventricular myocytes [108]and the purified RyR can also be phosphorylated [109]. Phosphorylation at multiple sites by both protein kinase A and Ca/calmodulin-dependent protein kinase II (CaMKII) has been reported and, in each case, relieves Mg-dependent block of the channel and may provide a mechanism for recruiting more Ca release sites [110]. Phosphorylation results in increased Ca efflux from the s.r. in skinned cardiac trabeculae [111]and an increase in the open probability and ryanodine binding of isolated RyR [110, 112]. Phosphorylation of the channel has been suggested to increase Ca release and phosphatase application to decrease the gain of CICR in patch clamped cells [113]. Similarly an inhibitor of CaMKII (KN-93) decreased the gain of CICR in cells [114].
3.3.4 Immunophillins and the FK binding protein
Some immunosuppresant drugs such as rapamycin or FK-506 may bind to a protein associated with the RyR (FKBP12). Work on skeletal RyR expressed in insect cells showed that FKBP12 increased the number of channels which showed full conducting as opposed to subconducting states but decreased the channel opening in response to caffeine [115]. These effects could be reversed by binding rapamycin or FK-506. Subsequent work on cardiac RyR found that rapamycin increased the open probability of the RyR (while decreasing the current amplitude) and therefore that the FK binding protein might be depressing Ca release [116]. Work on intact cells showed that FK506 increased the duration of Ca sparks and increased both the duration and magnitude of the systolic Ca transient. It was concluded that the FK binding protein may be involved in the normal closing of the RyR [117]. Another study on intact cells [118]found that FK-506 increased the magnitude of the systolic Ca transient (even when the SR Ca load and L-type Ca current were constant). However, in addition to this potentiation of CICR, the authors also reported an inhibition of the Na–Ca exchange. A further complication to analysis of the effects of FK-506 is provided by a report that the effects of this compound on the systolic Ca transient are due to inhibition of the K current and prolongation of the action potential [119]but this cannot, of course, account for the effects on Ca sparks.
3.3.5 Pharmacological tools
A large number of chemicals affect the RyR. We will briefly mention two groups which have been used extensively in research on excitation-contraction coupling and are relevant to the later sections of this review.
Caffeine increases the open probability of RyR and thereby increases the sensitivity to cytoplasmic Ca [120, 121]. If applied at concentrations less than 0.5–1.0 mM, the main effect of caffeine is to potentiate the systolic Ca transient. At higher concentrations of caffeine release of s.r. Ca occurs even in the absence of electrical stimulation [122]. Caffeine has been used extensively in studies of excitation-contraction coupling (see Section 3.4.2). The naturally occurring compound carnosine which has some structural similarity to caffeine [123]also increases the open probability of the RyR [124].
The local anaesthetics procaine and tetracaine decrease the open probability of the RyR [125–127]and thereby depress contraction of both cardiac and skeletal muscle [128–131].
3.4 The effects of agents which affect the RyR on systolic [Ca2+]i
The evidence reviewed above has shown that the ryanodine receptor can be modulated by a variety of manoeuvres and agents. In this section we will consider the effect that such modulation will have on the contraction of the myocyte or heart.
3.4.1 Maintained modulation of the RyR produces transient effects on systolic [Ca2+]i
At first sight it might appear that stimulation of CICR will increase the amount of calcium released and the force of contraction. Indeed, in much of the work reviewed above this conclusion has been reached. The problem is, however, that the magnitude of systolic release depends not only on the degree of activation of CICR but also on the s.r. Ca content. The significance of this was first noted for the effects of low concentrations of caffeine on systolic [Ca2+]i. At concentrations below 1 mM, caffeine was found to produce a transient increase in the magnitude of the systolic Ca transient and contraction [122]. During steady-state application of caffeine, the magnitude of the systolic Ca transient returned to control levels (see Fig. 4). On removal of caffeine the systolic Ca transient was transiently less than the control value before returning to the control magnitude. This effect of caffeine was enhanced under conditions that lowered the fraction of the s.r. Ca store released e.g. if external Ca concentration was decreased. It is to be expected that an agent such as caffeine which stimulates CICR should have more effect when the initial trigger is small.
|
3.4.2 Cellular flux balance during modulation of RyR
The explanation for the transient nature of the response to low concentrations of caffeine is related to the requirement for the cell (and indeed the s.r.) to maintain calcium balance. In the steady-state the amount of Ca which enters the cell from the extracellular fluid (primarily via the Ca current) must exactly balance that which leaves (mainly on Na–Ca exchange). If the properties of the Ca removal mechanisms are not affected then the magnitude of the Ca efflux is determined by the magnitude of the systolic rise of calcium concentration. Therefore the magnitude of the systolic Ca transient must be such that the Ca efflux exactly balances the Ca entry. If CICR is stimulated the increase of the systolic Ca transient will increase the Ca efflux from the cell such that it is greater than the entry into the cell. This has been shown directly by comparing the Ca entry (via the L-type Ca current) and efflux from the cell [132]. As a consequence of this there is a decrease of s.r. Ca content which has the same magnitude as that expected from the changes of Ca entry and efflux [132]. Therefore the amount of Ca available for release on the next stimulus will be decreased and the Ca transient, although larger than the previous steady-state one will be decreased. There will still, however, be more efflux than influx until, in the steady-state, efflux, once again, balances entry. This can only be achieved at the same level of systolic Ca as in the steady-state (if influx is unchanged). This therefore explains why the application of caffeine only produces a transient increase in the magnitude of systolic [Ca2+]i. The transient undershoot of contraction on removal of caffeine is due to the fact that the stimulation of the RyR has been removed but the s.r. Ca content is transiently below control levels. By the same argument as that above, the smaller Ca transient will result in greater Ca entry and smaller Ca efflux and the cell will gradually gain Ca until the steady state is reestablished. The transient effect on contraction of modulation of CICR is not restricted to the effects of caffeine. Tetracaine which (see above) depresses the opening of RyR produces a depression of the systolic contraction which is only transient (Fig. 5) [79]. This decrease of contraction, and the decrease of [Ca2+]i results in a decrease of Ca efflux on the Na–Ca exchange and therefore a net gain of cellular Ca (b). This, in turn increases s.r. Ca content and increases the systolic Ca transient until flux balance is reestablished (c). On removal of tetracaine the magnitude of the systolic contraction was transiently increased above control levels as a result of the combination of increased s.r. Ca content and normal RyR function. In other words the effects of tetracaine are exactly opposite to those of caffeine.
|
Strictly speaking, the above argument assumes that the duration of the Ca transient is not affected by modifying the RyR. A steady-state increase of the Ca transient could be produced by stimulation of the RyR so long as this was accompanied by a decrease in the duration of the Ca transient such that the Ca efflux from the cell was unaffected. For example, it is possible that the decrease of s.r. Ca content resulting from RyR stimulation will allow the s.r. Ca-ATPase to operate more quickly and thereby compete more effectively with the Na–Ca exchange. This will serve to maintain the s.r. Ca content and result in a maintained increase of the systolic Ca transient.
One should consider why some previous work has found that manoeuvres which are thought to stimulate CICR produce maintained effects on the systolic Ca transient. There are two major possibilities. (i) It is possible that some of these agents may also affect processes other than CICR. A maintained increase of the systolic Ca transient could result from stimulation of either the s.r. Ca-ATPase or Ca entry pathways into the cell or inhibition of Ca efflux pathways such as Na–Ca exchange. Indeed, as mentioned above, some of the maintained effect of FK-506 on systolic [Ca2+]i has been attributed to inhibition of Na–Ca exchange [118]and prolongation of the action potential [119]. (ii) The above steady-state analysis may not be relevant to the particular experimental conditions used. If low rates of stimulation are used then the s.r. Ca content may reach a steady-state with respect to the diastolic level of calcium. Changes in the amount of calcium gained or lost during systole will therefore have no effect on the steady-state s.r. Ca content. From this point of view it is therefore important that work be carried out at appropriate heart rates. For example the normal heart rate of a guinea pig is about 400 min–1, a value much greater than the stimulation rates used in most voltage clamp studies.
3.5 Modulation of the sarcolemmal Ca current
Increasing the sarcolemmal Ca current will have two effects on excitation-contraction coupling. (i) A direct effect to increase the trigger Ca which stimulates s.r. Ca release. (ii) The increase of Ca entry on each Ca transient will increase the Ca loading of the cell and the s.r. These two roles of the Ca current have been previously identified by Fabiato [3]. Using an argument identical to that above, it is easy to show that a simple increase of the trigger function of the Ca current can only produce a transient increase of the systolic Ca transient. In other words, increasing the trigger function of the Ca current is identical in action to stimulating the CICR process. We therefore conclude that, in the steady-state, the increase of contraction produced by an increase of the magnitude of the L-type Ca current is due to increased loading of the s.r. and not via any effect on the trigger function.
|
4 Is modulation of RyR or the trigger function of the L-type Ca current useful? |
|---|
|
TopAbstract1 Introduction2 Calcium-induced calcium...3 What regulates e-c...4 Is modulation of...References |
|---|
4.1 Model
Given the above conclusion that stimulation of CICR or the trigger function of the L-type Ca current will have no steady-state effect on the systolic Ca transient, it is worth considering why CICR is modulated. One possibility is that the transient increase of contraction is useful and provides an immediate increase of contraction while other slower effects develop [132]. The predictions of a simple model for this are shown in Fig. 6Fig. 7.
|
|
We assume that immediately before the nth stimulus the Ca content of the s.r. is s.r.n. A certain fraction, f, of this Ca is released from the s.r. The resulting increase of systolic Ca is proportional to the sum of this released Ca and the entry of Ca on the L-type current (iCa). Therefore
|
| (1) |
Of this Ca, a certain amount (proportional to p) will be pumped back into the s.r. whereas the remainder (proportional to q) will be pumped out of the cell. The amount of Ca pumped out of the cell is therefore given by:
|
|
Therefore the amount of Ca in the s.r. before the n+1 systole will be given by:
|
|
|
| (2) |
The increase of systolic Ca will be:
|
| (3) |
|
| (4) |
|
| (5) |
This shows that the steady-state Ca transient is independent of f, the fraction of Ca released from the s.r. This, of course agrees, with the experimental data for the effects of caffeine and tetracaine. We consider the non steady-state cases below.
4.2 Simulation of the effects of stimulating the RyR versus the s.r. Ca-ATPase
Fig. 6 shows calculations based on this model designed to simulate the effects of phosphorylation as might be produced by catecholamines. The 
show the predicted beat by beat changes of s.r. and systolic Ca produced by a stimulation of the ryanodine receptor (equivalent to increasing f above). This results in a transient increase of the systolic Ca transient and a maintained decrease of the s.r. Ca content. On restoring the original value of f, there is a transient undershoot of systolic Ca as the s.r. Ca content is restored. In contrast the 
show the effects of stimulation of the s.r. Ca-ATPase by phosphorylating phospholamban. Although the stimulation of the s.r Ca-ATPase is assumed to be instantaneous, the onset of the inotropic effect is delayed. This is because it takes several beats for the s.r. Ca content to increase. Finally the 
shows the effects of both the phosphorylation of phospholamban and stimulation of the RyR. With the parameters chosen there is an almost immediate and maintained increase of contraction. This results from the combination of an overshooting, transient response to stimulation of the RyR and a delayed response to phosphorylation of phospholamban. It should be noted that we have adjusted the parameters to give no change of s.r. Ca content. This exact balance is, of course, unlikely to occur in any physiological situation. The use of different parameters will give results intermediate between pure stimulation of the s.r. Ca-ATPase (increase of s.r. Ca content) and pure stimulation of trigger (decrease of s.r. Ca content). This can be used to model, for example, the effects of a compound such as FK-506 which inhibits Na–Ca exchange [118](equivalent to decreasing r) as well as stimulating the RyR.
4.3 Simulation of the effects of increasing the trigger versus the loading function of the L-type Ca current
Almost identical conclusions arise from comparison of the trigger and Ca loading functions of increasing the Ca current. In Fig. 7 an increase of the trigger function results in a transient increase of the systolic transient and a decrease of s.r. Ca content, (). In contrast an increase in the Ca loading function produces a delayed increase of both s.r. Ca content and the systolic Ca transient, (). When both effects are combined there is an abrupt and maintained increase of systolic Ca transient with little effect on s.r. Ca content (). It will be important in future experiments to examine whether the changes of contraction produced by, for example catecholamines can be attributed to the combination of the transient and delayed components predicted above.
The results of this calculation have implications for voltage-activated Ca release. On this scheme, if Ca release is stimulated, there will only be a transient increase of systolic Ca. Whereas CICR automatically provides a mechanism to match Ca entry via the Ca current to Ca release, this is not the case for voltage-activated release. While one can argue that a common mediator (perhaps cAMP) as well as stimulating VACR could also increase Ca entry into the cell or uptake by the s.r., the regulation is much less straightforward than is the case for CICR.
4.4 How does one tell whether CICR has been modulated in the cell?
Agents such as tetracaine and caffeine enter the cell very rapidly and produce transient effects on contraction. If a compound enters the cell more slowly, no effect on contraction would be expected. This means that the lack of any effect of a substance on systolic [Ca2+]i does not mean that it has no effect on CICR. The question then arises as to how one can tell whether such a substance has had any effect on CICR. One way to do this is to investigate the amount of Ca released from the s.r. in response to a given trigger Ca current. The problem with this is that it is necessary to bring the s.r. Ca content (and possibly the magnitude of the L-type Ca current) back to control levels. In our opinion, a much more straightforward way, as described below, is to investigate the effects on spontaneous s.r. release.
Much work has shown that, under conditions when the cell is ‘overloaded’ with calcium, spontaneous release of Ca from the s.r. ensues [38, 39, 133, 134]. Consistent with them originating from s.r. Ca release, they persist even when the surface membrane has been removed [135]. Some of the calcium which is released from the s.r. is pumped out of the cell on the Na–Ca exchange [136, 137]. This results in the arrhythmogenic transient inward current and afterdepolarizations [138–140].
Manoeuvres which increase the Ca load of the cell increase the frequency of spontaneous Ca release [141]. In contrast there is no obvious effect on the magnitude of the Ca release. This is quantified in experiments such as that shown in Fig. 8A which shows that large changes of the frequency of spontaneous release are accompanied by a constant integral of the Na–Ca exchange current. In other words it appears that the Ca load of the cell affects only the frequency and not the magnitude of the Ca release [142]. This could be explained if spontaneous release occurs once some critical level of s.r. Ca has been reached. Increasing Ca influx will only increase the rate at which this critical level is reached and therefore the frequency. This work examined only the effects of changing external Ca concentration but it is probable that other manoeuvres which affect only sarcolemmal Ca fluxes (and not the properties of the s.r. itself) will have no effect on the magnitude of the spontaneous Ca release. Completely different results (Fig. 8B) are obtained when the RyR is interfered with. Tetracaine decreases the opening probability of the RyR. This produces an initial abolition of spontaneous release before spontaneous release resumes [126, 143]. Fig. 8B shows Na–Ca exchange currents activated by spontaneous s.r. Ca release. The magnitude of this release (as assessed either by the accompanying contraction or the integral of the Na–Ca exchange current) is increased by tetracaine [143]. This effect was attributed to an increase in the critical s.r. Ca content which must be reached for spontaneous Ca release to occur. Once release occurs it is larger as a result of the greater s.r. Ca content. The increase of s.r Ca release is accompanied by a decrease of frequency such that the time-averaged Ca efflux from the cell is unaffected. It is this decrease of frequency which allows the maintained effect of tetracaine on spontaneous in contrast to stimulated release. Therefore the magnitude of the spontaneous release may be a convenient index of RyR status.
|
4.5 Cardiac hypertrophy and failure
The data reviewed above show that a number of pharmacological and physiological interventions alter the properties of the RyR isolated from normal hearts. Cardiac hypertrophy and failure is associated with a decrease of systolic Ca [144, 145]which is associated with a decrease in the expression of RyR [144–147]and it has been suggested that the decrease in RyR may contribute to the decreased systolic Ca. A recent study [148]suggested that during cardiac hypertrophy and failure there is a change in the ability of ICa to trigger release of Ca from the SR and attributed this to a change in the cellular microarchitecture of the dyad. This observation implies that the triggering function of ICa is reduced. However, the results presented in Figs. 6 and 7
predict that altering either the trigger function or the RyR alone will only have transient effects. One would therefore expect that if either the density of RyR or dyad structure is changed the only net result will be an increase of s.r. Ca content. Several studies using both animal models of cardiac failure and explanted failing human hearts have also shown a decrease in SR Ca-ATPase at both the mRNA and protein levels [144, 145]and this (see Fig. 6) would produce a maintained decrease of systolic Ca. Taken together, the decrease in SR Ca uptake and the decrease in RyR density would be expected to reproduce the abnormal systolic Ca transients observed in failing cardiac muscle. Time for primary review 28 days.
|
Acknowledgements |
|---|
The authors work was supported by grants from the British Heart Foundation and The Wellcome Trust.
|
References |
|---|
|
TopAbstract1 Introduction2 Calcium-induced calcium...3 What regulates e-c...4 Is modulation of...References |
|---|
- Bers, DM. Excitation-contraction coupling and cardiac contractile force. Dordrecht/Boston/London: Kluwer Academic Publishers, 1991.
- Fabiato A. Rapid ionic modifications during the aequorin-detected calcium transient in a skinned canine cardiac Purkinje cell. J Gen Physiol (1985) 85:189–246.
[Abstract/Free Full Text] - Fabiato A. Simulated calcium current can both cause calcium loading in and trigger calcium release from the sarcoplasmic reticulum of a skinned canine cardiac purkinje cell. J Gen Physiol (1985) 85:291–320.
[Abstract/Free Full Text] - Fabiato A. Time and calcium dependence of activation and inactivation of calcium-induced release of calcium from the sarcoplasmic reticulum of a skinned canine cardiac Purkinje cell. J Gen Physiol (1985) 85:247–289.
[Abstract/Free Full Text] - Rousseau E, Smith JS, Henderson JS, Meissner G. Single channel and
flux measurements of cardiac sarcoplasmic reticulum calcium channel. Biophys J (1986) 50:1009–1014.[Abstract/Free Full Text] - Smith GL, Valdeolmillos M, Eisner DA, Allen DG. Effects of rapid application of caffeine on intracellular calcium concentration in ferret papillary muscles. J Gen Physiol (1988) 92:351–368.
[Abstract/Free Full Text] - Näbauer M, Callewaert G, Cleemann L, Morad M. Regulation of calcium release is gated by calcium current, not gating charge, in cardiac myocytes. Science (1989) 244:800–803.
[Abstract/Free Full Text] - Valdeolmillos M, O'Neill SC, Smith GL, Eisner DA. Calcium-induced calcium release activates contraction in intact cardiac cells. Pflüg Arch (1989) 413:676–678.[CrossRef][ISI][Medline]
- Näbauer M, Morad M. Ca2+-induced Ca2+ release as examined by photolysis of caged Ca2+ in single ventricular myocytes. Am J Physiol (1990) 258:C189–C193.[ISI][Medline]
- Niggli E, Lederer WJ. Voltage-independent calcium release in heart muscle. Science (1990) 250:565–568.
[Abstract/Free Full Text] - London B, Krueger JW. Contraction in voltage-clamped, internally perfused single heart cells. J Gen Physiol (1986) 88:475–505.
[Abstract/Free Full Text] - Cannell MB, Berlin JR, Lederer WJ. Effect of membrane potential changes on the calcium transient in single rat cardiac muscle cells. Science (1987) 238:1419–1423.
[Abstract/Free Full Text] - Beuckelmann DJ, Wier WG. Mechanism of release of calcium from sarcoplasmic reticulum of guinea pig cardiac cells. J Physiol (London) (1988) 405:233–255.
[Abstract/Free Full Text] - Leblanc N, Hume JR. Sodium current-induced release of calcium from cardiac sarcoplasmic reticulum. Science (1990) 248:372–376.
[Abstract/Free Full Text] - Lederer WJ, Niggli E, Hadley RJ. Sodium-calcium exchange in excitable cells: Fuzzy Space. Science (1992) 248:283.[CrossRef]
- Wendt-Gallitelli MF, Voigt T, Isenberg G. Microheterogeneity of subsarcolemmal sodium gradients, electron probe microanalysis in guinea pig ventricular myocytes. J Physiol (London) (1993) 472:33–44.
[Abstract/Free Full Text] - Bielen FV, Glitsch HG, Verdonck F. Changes of the subsarcolemmal Na+ concentration in internally perfused cardiac cells. Biochim Biophys Acta (1991) 1065:269–271.[Medline]
- Carmeliet E. A fuzzy subsarcolemmal space for intracellular Na+ in cardiac cells? Cardiovasc Res (1992) 26:433–442.[ISI][Medline]
- Sham JSK, Cleemann L, Morad M. Gating of the cardiac Ca2+ release channel: the role of Na+ current and Na+–Ca2+ exchange. Science (1992) 255:850–853.
[Abstract/Free Full Text] - Bouchard RA, Clark RB, Giles WR. Role of sodium-calcium exchange in activation of contraction in rat ventricle. J Physiol (London) (1993) 472:391–413.
[Abstract/Free Full Text] - Lipp P, Niggli E. Sodium current-induced calcium signals in isolated guinea pig ventricular myocytes. J Physiol (London) (1994) 474:439–446.
[Abstract/Free Full Text] - Evans AM, Cannell MB. The role of L-type Ca2+ current and Na+ current-stimulated Na/Ca exchange in triggering SR calcium release in guinea pig cardiac ventricular myocyte. Cardiovasc Res (1997) 35:294–302.
[Abstract/Free Full Text] - Sipido KR, Carmeliet E, Pappano A. Na+ current and Ca2+ release from the sarcoplasmic reticulum during action potentials in guinea pig ventricular myocytes. J Physiol (London) (1995) 489:1–17.
[Abstract/Free Full Text] - Chapman RA, Tunstall J. The interaction of sodium and calcium ions at the cell membrane and the control of contractile strength in frog atrial muscle. J Physiol (London) (1980) 305:109–123.
[Abstract/Free Full Text] - Chapman RA, Tunstall J. The tension-depolarization relationship of frog atrial trabeculae as determined by potassium contractures. J Physiol (London) (1981) 310:97–115.
[Abstract/Free Full Text] - Eisner DA, Lederer WJ, Vaughan Jones RD. The control of tonic tension by membrane potential and intracellular sodium activity in the sheep cardiac Purkinje fibre. J Physiol (London) (1983) 335:723–743.
[Abstract/Free Full Text] - Kohmoto O, Levi AJ, Bridge JHB. Relation between reverse sodium-calcium exchange and sarcoplasmic reticulum calcium release in guinea pig ventricular cells. Circ Res (1994) 74:550–554.
[Abstract/Free Full Text] - Levi AJ, Brooksby P, Hancox JC. A role for depolarisation induced calcium entry on the Na–Ca exchange in triggering intracellular calcium release and contraction in rat ventricular myocytes. Cardiovasc Res (1993) 27:1677–1690.
[Abstract/Free Full Text] - Wasserstrom JA, Vites AM. The role of Na+–Ca2+ exchange in activation of excitation-contraction coupling in rat ventricular myocytes. J Physiol (London) (1997) 493:529–542.[ISI]
- Vornanen M, Shepherd N, Isenberg G. Tension-voltage relations of single myocytes reflect Ca release triggered by Na/Ca exchange at 35°C but not at 23°C. Am J Physiol (1994) 267:C623–C632.[ISI][Medline]
- Nuss HB, Houser SR. Sodium-calcium exchange-mediated contractions in feline ventricular myocytes. Am J Physiol (1992) 263:H1161–H1169.[ISI][Medline]
- Adachi-Akahane S, Lu L, Li Z, Frank JS, Philipson KD. Calcium signalling in transgenic mice overexpressing cardiac Na+–Ca2+ exchanger. J Gen Physiol (1997) 109:717–729.
[Abstract/Free Full Text] - Sipido KR, Maes M, Van de Werf F. Low efficiency of Ca2+ entry through the Na+–Ca2+ exchanger as trigger for Ca2+ release from the sarcoplasmic reticulum. Circ Res (1997) 81:1034–1044.
[Abstract/Free Full Text] - Adachi-Akahane S, Cleemann L, Morad M. Cross-signaling between L-type Ca2+ channels and ryanodine receptors in rat ventricular myocytes. J Gen Physiol (1996) 108:435–454.