Calcium-mediated cell death during myocardial reperfusion

Abstract

Reperfusion may induce additional cell death in patients with acute myocardial infarction receiving primary angioplasty or thrombolysis. Altered intracellular Ca²⁺ handling was initially considered an essential mechanism of reperfusion-induced cardiomyocyte death. However, more recent studies have demonstrated the importance of Ca²⁺-independent mechanisms that converge on mitochondrial permeability transition (MPT) and are shared by cardiomyocytes and other cell types. This article analyses the importance of Ca²⁺-dependent cell death in light of these new observations. Altered Ca²⁺ handling includes increased cytosolic Ca²⁺ levels, leading to activation of calpain-mediated proteolysis and sarcoplasmic reticulum-driven oscillations; this can induce hypercontracture, but also MPT due to the privileged Ca²⁺ transfer between sarcoplasmic reticulum and mitochondria through cytosolic Ca²⁺ microdomains. In the opposite direction, permeability transition can worsen altered Ca²⁺ handling and favour hypercontracture. Ca²⁺ appears to play an important role in cell death during the initial minutes of reperfusion, particularly after brief periods of ischaemia. Developing effective and safe treatments to prevent Ca²⁺-mediated cardiomyocyte death in patients with transient ischaemia, by targeting Ca²⁺ influx, intracellular Ca²⁺ handling, or Ca²⁺-induced cell death effectors, is an unmet challenge with important therapeutic implications and large potential clinical impact.

Myocardial infarction, Reperfusion injury, Calpain, Mitochondria

1. Background: from Ca²⁺ paradox to oxygen paradox

Myocardial reperfusion injury, defined as a kind of cell death occurring secondary to transient ischaemia that is preventable by interventions applied at the time of reperfusion,^1–6 is now recognized as an important element of the pathophysiology of myocardial ischaemia–reperfusion. However, its cellular and molecular mechanisms are far from being completely elucidated.³ Alterations in cellular Ca²⁺ homeostasis have been considered part of these mechanisms during the past few decades, and several hypotheses have been proposed to explain their causative role in cell death.

Electron microscopy studies performed at different times after coronary reperfusion demonstrated rapid and massive Ca²⁺ deposition in the mitochondrial matrix of reperfused myocardium.^7,8 The initial hypotheses on the mechanism of reperfusion injury conceded a capital importance to cytosolic Ca²⁺ overload. In fact, the so-called Ca²⁺ paradox (hypercontracture and death of cardiomyocytes occurring upon restoration of Ca²⁺ after a Ca²⁺-free perfusion period) was meant to be a first cellular model of reperfusion injury.⁹ Hypercontracture was recognized as the cause of the characteristic histological appearance of reperfused infarcts known as contraction band necrosis.¹⁰ Hypercontracture could also be observed in isolated perfused hearts submitted to reoxygenation after transient hypoxia, a phenomenon that was named ‘oxygen paradox’.¹¹ Hypercontracture was proposed to be a triggering cause of cell death in isolated cardiomyocytes¹² and intact tissue,¹³ as contractile blockers administered at the time of reperfusion were able to prevent it.

Subsequent studies identified mitochondrial permeability transition (MPT) as an important cause of reperfusion injury.^14–16 Although MPT was originally described as a consequence of increased mitochondrial Ca²⁺ load,¹⁷ it was soon realized that other triggers were also involved. The role of Ca²⁺ as a cause of MPT lost ground in favour of reactive oxygen species (ROS), and Ca²⁺ is nowadays more often seen to play a secondary role in the pathogenesis of reperfusion injury.^18–20 A prominent recent review on reperfusion injury (covering not only myocardium) contains no reference to Ca²⁺ at all.⁶ In the present article, we revisit the role of Ca²⁺ in reperfusion-induced cardiomyocyte death in the light of recently available information.

2. Alterations of Ca²⁺ handling during ischaemia

2.1 Na⁺ and Ca²⁺ overload during ischaemia

Anaerobic glycolysis and H⁺ released by ATP breakdown produce a progressive decline in intracellular pH (pHi) during the initial minutes of ischaemia.^{21 31}P-NMR spectroscopy studies demonstrated pHi values of about 6.4 after 20 min of ischaemia in isolated rat hearts.²² These initial changes are associated with a sustained rise in intracellular Na⁺ as determined by ²³Na-NMR spectroscopy.^23,24 This increase in intracellular Na⁺ has been attributed to a decrease in the energy-dependent Na⁺ extrusion due to the inhibition of Na⁺/K⁺-ATPase, and to an increased Na⁺ influx associated with the activation of H⁺ extrusion mechanisms by Na⁺/H⁺-exchanger (NHE) and Na⁺-HCO₃ cotransporters,^25–27 but also to persistent (non-inactivating) Na⁺ channels.^28,29 The relative contribution of NHE and persistent Na⁺ channels has not been satisfactorily resolved.²⁸ The fact that NHE activity is reduced by low extracellular pH as occurs during ischaemia,³⁰ and that the use of the first non-selective NHE inhibitors, such as amiloride, also inhibited persistent Na⁺ channels,²⁸ has resulted in the role of NHE in the rise in intracellular Na⁺ levels being questioned. In this sense, initial studies by Xiao and Allen argued that NHE1 is substantially inhibited during ischaemia and that it becomes activated during early reperfusion.^31,32 According to these authors, activation during this phase would be critical for reperfusion injury. However, more recent studies using a new generation of NHE inhibitors^33,34 and mice with null mutation in the NHE1 gene³⁵ have shown convincingly a contribution of NHE to cytosolic Na⁺ overload during ischaemia.

Studies in isolated cardiomyocytes labelled with fluorescence markers of Ca^2+36,37 or isolated perfused hearts using surface fluorescence³⁸ or 19F-NMR spectroscopy³⁹ have demonstrated a sustained rise in Ca²⁺ during ischaemia, which starts after the Na⁺ rise in close temporal association with rigor onset. When the ATP concentration reaches a critically low threshold (below 100 μmol/L), a force-generating crossbridge cycling is initiated, and rigor contracture develops.⁴⁰ Rigor contracture at this low ATP concentration is essentially Ca²⁺ independent,^40,41 but its onset marks the beginning of the cytosolic Ca²⁺ rise and represents a key event in the progression of ischaemic injury.^42,43 These low ATP concentrations would impair the activity of sarcolemmal Na⁺/K⁺-ATPase and sarcoplasmic reticulum Ca²⁺-ATPase (SERCA), thus acting as a trigger for both cytosolic Na⁺ and Ca²⁺ overload.

The contribution of sarcolemmal NCX to cytosolic Ca²⁺ overload during ischaemia has been consistently reported in studies involving pharmacological inhibition of NCX, its genetic ablation, or reduced expression via adenovirally delivered shRNA.^39,44–47 The direction of NCX operation depends on the difference between transmembrane potential and the reversal potential of NCX, and is therefore determined by the intra- and extracellular concentrations of Na⁺ and Ca²⁺.⁴⁸ Under physiological conditions, the exchanger operates mainly in its forward mode to extrude Ca²⁺ from the cytosol during diastole,⁴⁹ but membrane depolarization and a reduced Na⁺ gradient, as a consequence of the intracellular Na⁺ rise, determine a net Ca²⁺ influx through reverse NCX transport during ischaemia. However, the fact that prevention of Na⁺ overload only attenuates the rise in cytosolic Ca²⁺ during ischaemia²⁵ suggests a contribution of additional, Na⁺-independent ways of Ca²⁺ entry, e.g. via the L-type Ca²⁺ channel.⁵⁰

Under normal conditions, the concentration of Ca²⁺ in the mitochondrial matrix is very low and similar to that in cytosol.⁵¹ Ca²⁺ enters the matrix mainly through the Ca²⁺ uniporter, dependent on the transmembrane electrochemical gradient, and is mainly extruded via the mitochondrial NCX, which is influenced by the matrix Na⁺ concentration⁵² and which, in turn, depends on the H⁺ concentration and the activity of mitochondrial NHE.⁵³ The H⁺ gradient across the mitochondrial membrane, built up by respiration and dissipated by ATP synthesis, is modulated by different ion transporters and exchangers, prominently the KHE.⁵⁴ During myocardial ischaemia, the rise in cytosolic Ca²⁺ levels tends to increase mitochondrial Ca²⁺ influx, but dissipation of the transmembrane potential has the opposite effect. However, dissipation of the mitochondrial membrane potential is incomplete during a prolonged ischaemic period. Thus, it is generally admitted that mitochondrial Ca²⁺ influx is important during initial ischaemia and that it remains increased thereafter.^55,56

2.2 Consequences of Ca²⁺ overload during ischaemia

2.2.1 Cell-to-cell uncoupling

In the ischaemic cardiomyocyte, the onset of the rise of cytosolic Ca²⁺ is closely followed by cell-to-cell electrical uncoupling.^43,57 It is still unclear whether the effects of Ca²⁺ are direct or mediated through intracellular messengers. Ca²⁺ probably induces connexin43 (Cx43) gap junctional closure by activation of calmodulin, which may act directly as a gating particle.⁵⁷ pHi is also an important regulator of Cx43 gap junction channel permeability, and acidosis probably contributes to uncoupling.⁵⁷ Closure of gap junctional channels by intracellular Ca²⁺ during ischaemia may protect cells from membrane depolarization and leakage of metabolites through gap junctions by disconnecting them from damaged cells.⁵⁸ However, reduced cell coupling may cause arrhythmias.⁵⁹

2.2.2 Opening of connexin hemichannels

Both extracellular and intracellular Ca²⁺ concentrations modulate unopposed connexin hemichannel gating. A decrease in extracellular Ca²⁺ has been shown to induce hemichannel opening^60,61 and increase Cx43 hemichannel pore diameter.⁶² Intracellular Ca²⁺ also regulates hemichannel function in a biphasic manner: moderate increases in Ca²⁺ can induce Cx43 hemichannel opening, whereas larger increases in Ca²⁺ inhibits hemichannel activity.⁶³ The opening of Cx43 hemichannels triggered by moderate intracellular Ca²⁺ levels has been suggested to involve multiple intermediate signalling steps⁶⁴ and intramolecular Cx43 loop/tail interactions.⁶³ Ischaemia induces opening of connexin hemichannels, which may contribute to cell injury and arrhythmias⁶⁵ through deleterious Ca²⁺ influx.⁶⁶ Hemichannel opening induces release of intracellular metabolites, such as ATP, inositol 1,4,5-triphosphate (IP₃), cAMP, NAD⁺, or glutamate, to the extracellular space^67,68 that may be involved in paracrine cardioprotective signalling.^68,69

2.2.3 Calpain translocation to the cell membrane

Calpains represent a wide family of non-lysosomal, Ca²⁺-dependent thiol proteases implicated in basic cellular processes including differentiation, proliferation, and cell migration and are tightly regulated by the cellular control of Ca²⁺ and its endogenous inhibitor calpastatin. Loss of Ca²⁺ homeostasis results in an unregulated overactivation of calpain.⁷⁰ The contribution of calpains to ischaemia/reperfusion injury has been consistently reported by several groups,^71–74 and the kinetics of calpain activation have been recently determined.⁷⁵ It has been proposed that under physiological conditions, translocation of calpains to the sarcolemma in response to transient cytosolic Ca²⁺ elevation is an obligatory step in the process of their activation.⁷⁶ However, translocation seems not to be essential under situations leading to a dysregulation of intracellular Ca²⁺ levels.⁷⁵In vitro studies have shown that calpain activity is highly influenced by pHi.⁷⁷ Recent results from our group demonstrated that intracellular acidosis prevents calpain activation during ischaemia, despite high intracellular Ca²⁺ concentrations.⁷⁵

2.2.4 Mitochondrial Ca²⁺ overload

It is generally assumed that mitochondrial Ca²⁺ accumulation has detrimental cellular effects, facilitating mitochondrial membrane permeabilization and energetic collapse. Prevention of mitochondrial Ca²⁺ uptake has a protective effect in ischaemia–reperfusion injury.^78,79 However, normoxic healthy mitochondria exhibit a striking ability to accumulate enormous amounts of Ca²⁺ and efficiently participate in cellular Ca²⁺ buffering along with the sarcoplasmic reticulum (SR).⁸⁰ Mitochondrial Ca²⁺ uptake and storage capacity may be particularly relevant when cytosolic Ca²⁺ is elevated, as during ischaemia. Nevertheless, the concept that mitochondria can shape the spatio-temporal pattern of cytosolic Ca²⁺ during ischaemia has been controversial. While mitochondrial membrane depolarization is expected to preclude Ca²⁺ uptake through the mitochondrial Ca²⁺ uniporter,⁸¹ significant mitochondrial Ca²⁺ accumulation has been documented during ischaemia in the presence of residual mitochondrial membrane potential.^55,56 Moreover, pharmacological inhibition of Ca²⁺ uptake during ischaemia has been shown to have a pernicious effect on cell survival upon re-energization, an effect that is associated with an impairment of cytosolic Ca²⁺ handling,⁵⁵ supporting the concept that mitochondrial Ca²⁺ uptake helps to delay a rise in cytosolic Ca²⁺ levels.

3. Intracellular Ca²⁺ handling during reperfusion

The main mechanisms and consequences of altered Ca²⁺ handling during reperfusion are summarized schematically in Figure 1.

Figure 1

Mechanisms and consequences of altered Ca²⁺ handling in cardiomyocytes during initial reperfusion. Main events are connected through black lines, whereas red lines indicate important modulating factors. GCPR, G-coupled protein receptors; IP3, inositol trisphosphate; NOS, nitric oxide synthase; ROS, reactive oxygen species.

3.1 Na⁺ and Ca²⁺ influx during initial reperfusion

The intracellular Na⁺ overload observed in reperfused cardiomyocytes is largely the consequence of Na⁺ gain during prior ischaemia,^23,24 but also of additional Na⁺ influx at the onset of reperfusion associated with pHi correction or Na⁺ influx from adjacent myocytes via gap junctions.⁸²

Reperfusion removes extracellular H⁺ and rapidly reactivates intracellular H⁺ efflux mediated by the lactate-H⁺ cotransporter⁸³ and the activity of NHE and Na⁺HCO₃ cotransporter,⁸³ resulting in a fast pHi recovery.^{22 23} Na-NMR studies demonstrate a fast correction in intracellular Na⁺ levels during initial reperfusion even in the presence of NHE inhibitors,⁸⁴ which can probably be explained by the compensatory action of bicarbonate transporters and the reactivation of Na⁺/K⁺-ATPase.^22,85,86

Reactivation of Na⁺/K⁺-ATPase appears to be a key determinant of the kinetics of Na⁺ recovery at reperfusion.⁸⁷ Whereas after short periods of ischaemia functionality of the Na⁺/K⁺-ATPase is preserved immediately upon reperfusion,^86,88 it is impaired after prolonged occlusions.^88,89 It has been proposed that oxidation of sulfhydryl groups on the protein,⁹⁰ and more recently, calpain-dependent loosening of its alpha subunit from the membrane–cytoskeleton complex,⁸⁸ contribute to reperfusion-induced Na⁺/K⁺-ATPase dysfunction. During the first minutes of reperfusion, cytosolic Na⁺ overload provides a large driving force for Ca²⁺ influx through the reverse mode of NCX. Inhibitors of NCX applied at the time of reperfusion reduce the frequency of Ca²⁺ oscillations (see below), infarct size, and myocardial stunning in intact hearts, confirming that early reperfusion results in an additional Ca²⁺ influx through the reverse-mode operation of NCX.^5,91,92

3.2 SR-driven Ca²⁺oscillations and hypercontracture

The SR is the main intracellular Ca²⁺ store in cardiac myocytes, orchestrating excitation–contraction coupling through rapid Ca²⁺ uptake (SERCA ATPase) and release transport system (ryanodine receptor, RyR) facilitated by a close juxtaposition with T-tubules and mitochondria.⁹³ Previous studies have indicated that in the presence of an abnormally high cytosolic Ca²⁺ concentration, such as that occurring in the very first minutes of reperfusion, a rapid SR Ca²⁺ cycling may be the genuine cause of reperfusion-induced hypercontracture.^94,95 Ca²⁺ oscillations are the consequence of the reactivation of mitochondrial respiration and efficient ATP transfer to the SR, subsequent Ca²⁺ uptake by SERCA ATPase, and Ca²⁺ release through RyR when SR storage capacity is exceeded.⁹⁴ Importantly, upon resumption of mitochondrial energy synthesis, SERCA ATPase may initiate this futile Ca²⁺ cycling before other ATP-dependent sarcolemmal Ca²⁺ extruders reduce cytosolic Ca²⁺ load, in part due to the privileged spatial communication between SR and mitochondria (see below). Ca²⁺ release from a SR unit may be taken up by an adjacent one, and the resulting Ca²⁺ oscillation can propagate throughout the cell as Ca²⁺ waves that induce hypercontracture. Accordingly, a decrease in the synthesis of ATP necessary to activate SERCA ATPase,⁹⁶ pharmacological inhibition of SERCA or RyR,^94,97,98 or a reduction in the cytosolic Ca²⁺ load^5,99 decreases the frequency and amplitude of SR-induced Ca²⁺ sparks and oscillations and reduces hypercontracture and cell death in different experimental models. Alternatively, the improvement of SR Ca²⁺ sequestration capacity by protein kinase G (PKG)-dependent phosphorylation of phospholamban (resulting in more efficient SERCA activity) may decrease reperfusion-induced SR Ca²⁺ oscillations, favour the recovery of cytosolic Ca²⁺ control, and prevent excessive myofibrilar activation during reperfusion.^95,100 These results are in agreement with recent observations indicating that Ca²⁺ wave development depends on the balance between SERCA-dependent Ca²⁺ reuptake and threshold SR Ca²⁺ content¹⁰¹ and underline the importance of energy-dependent SR-driven Ca²⁺ oscillations in the pathophysiology of ischaemia–reperfusion.

3.3 Role of mitochondria–SR interaction in hypercontracture and MPT

Two subpopulations of mitochondria, subsarcolemmal (SSM) and interfibrillar (IFM), have been described,¹⁰² with differences in morphology,¹⁰³ functional capacity,^102,104 and, more importantly, differential interactions with other cellular components, as only IFM is in close contact with the SR. Mitochondria and SR are structurally and functionally interconnected, and this interplay has important pathophysiological consequences that go beyond the individual role of each organelle.^105,106 It has been demonstrated that SR Ca²⁺ uptake may be altered when creatine kinase activity from IFM is specifically depressed.¹⁰⁷ As a consequence, the highly specialized energy transfer and Ca²⁺ exchange system formed by SR, IFM, and myofilaments becomes less efficient and may account for the mismatch between energy demand/utilization and Ca²⁺ uptake/release (contractile force) that is observed under certain pathological conditions, such as heart failure.¹⁰⁷ During reperfusion, microdomains of high Ca²⁺ concentration around RyR of the SR have been proposed to be involved in the permeabilization of mitochondria located in close proximity.^97,98 Accordingly, pharmacological blockade of SR Ca²⁺ load with thapsigargin/ryanodine impacted on mitochondrial Ca²⁺ uptake kinetics in intact cardiac myocytes, but had no effect in isolated mitochondria, where the contribution of SR is expected to be negligible.⁹⁸ The distance between SR and mitochondria has been estimated to be 10–50 nm,¹⁰⁸ and mitofusin 2 has been identified as the main mediator of the physical link among the two organelles.¹⁰⁹ Chemical disruption of the microtubule network increases the inter-organelle distance and reduces the noxious effect that the SR-induced Ca²⁺ oscillations exert on mitochondrial integrity during the first minutes of reperfusion.⁹⁸ The differences in Ca²⁺ crosstalk with SR may explain in part why IFM have a higher Ca²⁺ retention capacity,¹⁰² and this difference in Ca²⁺ tolerance may exacerbate cellular stress during reperfusion, when damaged mitochondria (releasing Ca²⁺ as well as other molecules) coexist along with intact mitochondria (capable of sustaining ATP synthesis) within the same cell. Moreover, the relationship between SR and mitochondria appears to be bidirectional: SR-induced Ca²⁺ cycling may favour mitochondrial permeabilization and energetic collapse on one hand,^97,98,110 whereas mitochondrial permeabilization may trigger SR Ca²⁺ oscillations, hypercontracture, and cell death during reperfusion by increasing cytosolic Ca²⁺ overload¹¹¹ on the other hand. Thus, early pharmacologic inhibition of SR Ca²⁺ uptake and release upon re-energization reduces MPT, hypercontracture, and cell death,⁹⁸ and genetic or chemical blockade of mitochondrial permeabilization decreases hypercontracture and infarct size.¹¹²

3.4 Normalization of intracellular pH

A rapid correction of acidosis during reperfusion contributes to Ca²⁺ influx and, furthermore, may precipitate the adverse effects of Ca²⁺ overload that remain inhibited at low pHi. Normalization of pHi allows the activation of critical proteins involved in the regulation of Ca²⁺ handling, including NCX,^22,113 the L-type Ca²⁺ channel,¹¹⁴ the RyR,¹¹⁵ and SERCA.¹¹⁶ Even a brief prolongation of extracellular acidosis during the early phase of reoxygenation reduces Ca²⁺ overload.¹¹⁷ Recovery from pHi also relieves the inhibition of cardiomyocyte contractility caused by a reduced Ca²⁺ binding to troponin C,¹¹⁸ the occurrence of MPT,^119,120 activation of calpains,⁷⁵ and blockade of gap junctions.¹²¹ Therefore, the relative timing of the correction of intracellular Ca²⁺ levels and pHi during the first minutes of reperfusion has been proposed to determine cell death (recovery of pHi occurs before that of Ca²⁺) or survival (recovery of Ca²⁺ control occurs before pHi normalization). This hypothesis may help explain the cardioprotective effect of brief acidic reperfusion and ischaemic postconditioning.¹²²

3.5 Extracellular signalling modulating intracellular Ca²⁺

The intracellular Ca²⁺ concentration can be modulated by a number of extracellular signals during reperfusion. Activation of membrane-bound guanylyl cyclase with natriuretic peptides or soluble guanylyl cyclase with NO donors increases the cGMP synthesis, which in turn activates PKG. PKG has been shown to activate SERCA through phosphorylation of phospholamban, increasing Ca²⁺ accumulation by the SR and preventing cytosolic Ca²⁺ oscillations.¹²³

G protein-coupled receptors (GPCRs) might also play a relevant role under ischaemia–reperfusion in modulating intracellular Ca²⁺ levels through two opposing mechanisms. Activation of G_αq/11 by most GPCRs results in the hydrolysis of membrane-bound phosphatidylinositol 4,5-bisphosphate (PIP2) by phospholipase Cβ, releasing inositol 1,2,5-trisphosphate (IP3). This molecule binds to its intracellular receptors (IP3R), inducing the release of Ca²⁺ from the endoplasmic reticulum to the cytosol.¹²⁴ On the other hand, binding of a ligand to its GPCR also leads, probably through the βγ subunits of the G proteins, to serial activation of phosphoinositide 3-kinase (PI3K), Akt, ERK, nitric oxide synthase (NOS), and guanylyl cyclase, which results in increased levels of intracellular cGMP, activation of SERCA through PKG-mediated phosphorylation of phospholamban, and uptake of Ca²⁺ by the SR.¹²³ Adenosine, ATP (P2Y receptors), opioids, bradykinin, adrenaline, acetylcholine, insulin, erythropoietin, oestrogens, transforming growth factor-β1, angiotensin II, and adrenomedullin are endogenous ligands that may modulate intracellular Ca²⁺ by binding to GPCR. Most of them have been shown to activate the PI3K/Akt/ERK/NOS signalling pathway and to be involved in cardioprotection by pre- and postconditioning.¹²⁵ Similarly, extracellular Ca²⁺-sensing receptors are GPCR that can play a role in regulation of the intracellular Ca²⁺ concentration and have been shown to be involved in ischaemia/reperfusion-induced apoptosis in rat cardiomyocytes by increasing the intracellular Ca²⁺ concentration.¹²⁶ On the other hand, Ca²⁺-sensing receptors have been suggested to play a role in preconditioning protection in isolated mouse hearts.¹²⁷ Protease-activated receptors, such as thrombin receptors, are specialized GPCR. Thrombin activates its receptor by cleaving part of its extracellular amino-terminal domain, promoting coupling with G_αβγ proteins, and modulating the intracellular Ca²⁺ concentration. Thrombin has been shown to increase cytosolic Ca²⁺ and twitch amplitude in isolated rat cardiomyocytes¹²⁸ and to cause cell death after ischaemia–reperfusion.¹²⁹

4. The end-effectors of Ca²⁺-dependent cardiomyocyte death during reperfusion

Altered Ca²⁺ handling may trigger processes that directly cause cell death such as hypercontracture, proteolysis, and MPT.

4.1 Hypercontracture

The role of mechanical stress generated by an excessive contractile activation, known as hypercontracture, in the development of cardiomyocyte death during reperfusion is supported by ample experimental evidence.^10,130,131 Hypercontracture has been observed during the first minutes of reperfusion in vitro by microscopic techniques, and in vivo by intramyocardial ultrasonometry.^130,132 There is a close correlation between the magnitude and time course of hypercontracture and enzyme release in reperfused myocardium.^112,130 Brief contractile inhibition at the onset of reperfusion prevents enzyme release for the time the inhibition is present, while prolonged contractile inhibition limits final infarct size.^88,133,134

Hypercontracture is induced by recovery of energy production in the presence of a high cytosolic Ca²⁺ concentration¹³² (Figure 2). Cytosolic Ca²⁺ oscillations lead to sustained and uncontrolled activation of the contractile apparatus that results in hypercontracture.^131,135 Isolated, unrestrained cardiomyocytes, but not those restrained by attachment to a micropipette,¹³⁶ can maintain sarcolemmal and metabolic integrity after hypercontracture.^12,137 However, in tissue, the mechanical forces resulting from hypercontraction of adjacent cells lead to mutual cellular disruption and necrosis.^138,139

Figure 2

(A) Calcium and energy-dependent hypercontracture. Simulated ischaemia (SI, exposure to hypoxia at pH 6.4) and reperfusion in myocytes from adult rat heart labelled with JC-1 (5′,6,6′-tetrachloro-1,1′,3,3′- tetraethylbenzimidazolylcarbocyanine iodide) to study mitochondrial membrane potential. Ischaemia causes loss of membrane potential (disappearance of the red signal) and rigour shortening. Reperfusion causes recovery of mitochondrial membrane potential and hypercontracture. (From ref. 112; used with permission). (B) Analysis of recovery of cell energy (% of phosphocreatine, PCr) by NMR spectroscopy during the initial minutes of reperfusion in rat hearts submitted to 30 min of ischaemia. There is an initial recovery followed by de-energization that is prevented by contractile blockade. The inset shows that de-energization occurring in the control group is coincident with the development of hypercontracture that is manifested as an increase in left ventricular end-diastolic pressure (LVEDP) with lactic dehydrogenase (LDH) release reflecting sarcolemmal rupture, both prevented by contractile blockade with BDM. Altogether, these results indicate that de-energization is a consequence of hypercontracture-mediated cell death and not a cause of it. (Modified from ref. 112; with permission).

NMR spectroscopy shows rapid and virtually complete but transient energy recovery in hearts undergoing hypercontracture and important necrosis after transient ischaemia, demonstrating that hypercontracture is not due to low ATP levels in those experiments.¹¹² Furthermore, contractile blockade prevents cell death (LDH release) and cell de-energization, indicating that the fall in energy is not a cause but a consequence of hypercontracture.

It has been proposed that rigor-type, Ca²⁺-independent activation of the contractile machinery contributes to hypercontracture, at least under certain conditions.¹⁴⁰ Rigor-type contracture may be activated during reoxygenation if re-energization of the ischaemic cardiomyocytes occurs at a very low rate as after prolonged or severe ischaemia.^40,141 It is unclear, however, under which pathophysiological conditions in vivo this state of very slow, but non-zero re-energization required for rigor-type contracture may actually prevail. It is not expected to occur when ATP levels are rapidly restored after brief ischaemia or when reactivation of mitochondrial respiration completely fails or is effectively suppressed.^142,143

4.2 Calpain-mediated proteolysis

During reperfusion calpains hydrolyze proteins from the sarcolemma and the cytoskeleton, including α-fodrin and ankyrin. α-Fodrin forms the backbone of the membrane cytoskeleton. Its degradation correlates with increased fragility of the membrane, reducing the tolerance of the sarcolemma to the mechanical stress associated with hypercontracture and acute cell swelling during reperfusion.^71,74,144 Ankyrin has a central domain that binds to α-fodrin and an N-terminal domain that interacts with several receptors and channels, including the α subunit of Na⁺/K⁺-ATPase.¹⁴⁵ Binding to ankyrin connects Na⁺/K⁺-ATPase to the fodrin-based membrane cytoskeleton and determines its specific localization in the membrane and its correct function.¹⁴⁶ During reperfusion, calpain degradation of both fodrin and ankyrin not only causes sarcolemmal fragility but also detachment of the Na⁺ pump from its anchorage to the fodrin-based membrane skeleton, inducing dysfunction of the sarcolemmal Na⁺ pump. This results in impaired normalization of the cytosolic Na⁺ concentration and in further Ca²⁺ influx via reverse-mode NCX.⁸⁸ It has been proposed that calpain can also modulate Ca²⁺ handling by cleaving RyR¹⁴⁷ and SERCA2a.¹⁴⁸ In addition, calpain-dependent activation of Bid has been described to induce the release of cytochrome c and other proapoptotic factors.¹⁴⁹ Calpain inhibition reduces infarct size in different models^72,75 and has been proposed to be an important element in the cardioprotective effect of pre- and postconditioning.^23,71

4.3 Mitochondrial permeability transition

Upon reperfusion, cytosolic Ca²⁺ overload in the presence of mitochondrial repolarization (although incomplete) provides the electrochemical force necessary to activate mitochondrial Ca²⁺ uptake through the Ca²⁺ uniporter. However, cumulative Ca²⁺ capacity of reperfused mitochondria is severely impaired, mainly as a consequence of the concurrence of oxidative stress and low cellular ATP concentration, conditions that may initiate mitochondrial failure upon Ca²⁺ uptake when intracellular pH is normalized.^14–16 Mitochondrial failure occurs as a consequence of the abrupt increase in the permeability of mitochondrial membranes that are incompatible with efficient respiration and ATP synthesis. This phenomenon, known as mitochondrial membrane permeabilization, was initially described in vitro as a response specifically induced by Ca²⁺.¹⁷ It appears to be mediated by cyclophilin D (CyD), a trans-isomerase protein located at the inner mitochondrial membrane, although other indirect mechanisms of mitochondrial permeabilization have been described.^150,151 Genetic ablation of CyD or its pharmacological inhibition by CsA or other drugs increases mitochondrial tolerance to Ca²⁺ overload and reduces cell death after transient ischaemia in different experimental models.^152–154 In fact, tolerance to external Ca²⁺ pulses of mitochondria isolated from reperfused myocardium has been used as an index of the susceptibility of the myocardium to MPT under different conditions, including cardioprotective interventions.¹⁵² However, the causal role of mitochondrial Ca²⁺ tolerance as a mechanism of attenuated MPT during in vivo reperfusion remains obscure, and other triggers, in particular ROS, have been demonstrated to be more important in intact myocardium.¹⁵⁵ Indeed, the Ca²⁺ threshold upon which isolated mitochondria experience acute failure in vitro (high micromolar range) could only be achieved in vivo as the consequence of sarcolemmal disruption, resulting in massive Ca²⁺ entry from the extracellular space. However, studies using engineering-based fluorescent techniques have demonstrated that Ca²⁺ concentrations can be dramatically increased in specific microanatomical domains, reaching levels that are well above the concentration achieved in the bulk of cytoplasm,¹⁵⁶ and more recent evidence indicates that these microdomains may explain the occurrence of reperfusion-induced mitochondrial permeabilization in some cellular regions.^97,98 Moreover, the possibility that a subpopulation of intact, energy-producing mitochondria coexist with severely damaged mitochondria undergoing membrane permeabilization has been proposed as the mechanism reconciling both ATP-dependent hypercontracture and mitochondrial failure as the cause of reperfusion-induced cell death, because membrane permeabilization of mitochondria previously overloaded with Ca²⁺ may further impair cytosolic Ca²⁺ handling.¹¹¹

4.4 Relative importance of Ca²⁺-driven hypercontracture and MPT as causes of cell death

Occurrence of MPT in a subset of mitochondria releases Ca²⁺ into the cytosol, favouring Ca²⁺ oscillations/overload that eventually lead to hypercontracture. Conversely, uncontrolled SR-triggered Ca²⁺ oscillations may induce MPT in mitochondria located in close proximity to the SR. In addition, it is also possible that both mechanisms of cell death operate separately: altered Ca²⁺ handling can cause cell death through calpain-mediated proteolysis and hypercontracture independently of MPT, whereas MPT can be the primary cause of cell death during reperfusion. In fact, reperfusion-induced cell death occurs in cells other than cardiomyocytes that lack significant contractile machinery, such as hepatocytes, neurons, or modified cardiomyocyte cell lines.^120,157 The question of the relationship and relative importance of these different cell death mechanisms during myocardial reperfusion is thus relevant. We recently proposed that the relative contribution of MPT to cell death depends on the duration of prior ischaemia. After brief periods of ischaemia, inhibition of mitochondrial permeabilization with CsA or genetic ablation of CyD failed to protect isolated cardiomyocytes or intact hearts from reperfusion injury, whereas prevention of Ca²⁺-induced, energy-dependent hypercontracture was effective. Conversely, after more prolonged ischaemia, inhibition of MPT was strongly protective.¹¹²

4.5 Role of apoptosis

Ca²⁺ overload is an established trigger of apoptosis, and it has been proposed that it can induce apoptosis of cardiomyocytes even when its magnitude is lower than that required to induce necrosis, as may occur, for example, during inotropic stimulation.¹⁵⁸ It is thus theoretically possible that Ca²⁺ overload can induce apoptotic cell death in cardiomyocytes surviving the initial minutes of reperfusion. There are, however, reasons to believe that this is not a relevant cause of cell death. Most studies agree that infarct size, determined early during reperfusion, does not increase significantly during the following hours.^159,160 Furthermore, there is solid evidence that the caspase pathway responsible for mitochondria-driven apoptosis is not operative in adult cardiomyocytes.¹⁶¹

5. Pharmacological targets against Ca²⁺-mediated reperfusion injury

5.1 Improving Ca²⁺ handling

The value of cell systems involved in exaggerated Ca²⁺ influx during reperfusion as pharmacological targets to be used against reperfusion injury has been discussed in previous sections and in other articles.¹⁶² We will focus here on the potential value of systems involved in intracellular Ca²⁺ handling as therapeutic targets.

SERCA activity is specifically regulated by phospholamban (PLB), which in its dephosphorylated form exerts an inhibitory effect, susceptible to be relieved after its phosphorylation by either cAMP- or cGMP-dependent protein kinases or the Ca²⁺-calmodulin-dependent protein kinase (CaMKII).¹⁶³ Because ischaemia induces dephosphorylation of PLB,¹⁶⁴ modulation of the phosphorylation status of PLB has attracted much attention as a potential therapeutic strategy to reduce reperfusion injury. Calcineurin inhibitors (e.g. CsA) have been demonstrated to prevent PLB dephosphorylation by inhibiting PKC-α translocation.¹⁶⁴ Hsp20 has been recently identified as a modulator of PLB phosphorylation,¹⁶⁵ but its potential value as a pharmacological target remains to be established. Natriuretic peptides increase PLB phosphorylation via PKG and are cardioprotective.¹⁶⁶ B-type natriuretic peptide also inhibits the mitochondrial Ca²⁺ uniporter, reduces ROS generation, and improves mitochondrial energy recovery.¹⁶⁷ It may seem paradoxical that an increase in CaMKII activity has been found to decrease cell viability in rat hearts and isolated myocytes subjected to transient ischaemia–reperfusion, an effect that could be reverted in the presence of the CaMKII inhibitor KN-93 or the CaMKII inhibitory peptide AIP¹⁶⁸: the adverse effect of CaMKII may, however, be due to its effect on targets other than PLB.

Regulation of the crosstalk between SR and mitochondria is emerging as a new target to limit cell death. A recent report indicates that postconditioning may protect cardiomyocytes from apoptotic death induced by transient ischaemia by preserving crosstalk between mitochondria and the SR.¹⁶⁹ In some cases, drugs known for their effects on sarcolemmal ionic exchangers have been shown to inhibit mitochondrial transporters. It has been suggested that part of the protective effect of cariporide against ischaemic damage is due to its inhibitory effect on mitochondrial NHE exchange that delays ΔΨm dissipation and Ca²⁺ disturbances associated with energy depletion.⁵³ Also, the plasma membrane NCX inhibitor KB-R7943 may inhibit the mitochondrial Ca²⁺ uniporter and prevent mitochondrial Ca²⁺ overload during ischaemia–reperfusion in addition to its effect on cytosolic Ca²⁺ levels.¹⁷⁰

5.2 Attenuating the consequences of Ca²⁺ overload

A major limitation of strategies aimed towards limiting Ca²⁺ overload during reperfusion is that they cannot prevent its accumulation occurring during preceding ischaemia. An alternative approach is to inhibit Ca²⁺-dependent effects of reperfusion injury.

Contractile inhibitors prevent enzyme release during initial reperfusion^88,133 and limit infarct size.^134,171 The drug most extensively used for this purpose has been 2,3-butanedione monoxime (BDM), a reversible blocker of actomyosin-ATPase. Recently, a new contractile inhibitor, blebbistatin, has been shown to inhibit contractility with a potency that is three orders of magnitude higher.¹⁷² The dramatic effect blebbistatin has on the reduction of infarct size when administered to isolated rat hearts (unpublished results from our group) makes this drug a potential candidate for translation.

The cardioprotective effects obtained with some strategies aimed at preventing mechanisms other than hypercontracture could be explained, at least in part, by their actions on contractility. Delay in pHi recovery induced by pharmacological treatments,^22,85 transient reperfusion with respiratory or metabolic acidosis^4,22 or postconditioning²³ inhibits myofibrillar contractility during initial reperfusion. Similarly, cardioprotection afforded by stimulation of the cGMP/PKG pathway^100,173,174 has been related to PKG-dependent effects on the sensitivity of myofibrils to Ca²⁺.^175,176 Moreover, activation of PKG may delay pHi normalization through inhibition of NHE.¹⁷⁷

The fact that Ca²⁺-dependent overactivation of calpains occurs during reperfusion after pHi normalization makes them suitable pharmacological targets for the prevention of reperfusion injury. It has been reported that the administration of calpain inhibitors during the acute phase of reperfusion is effective in reducing infarct size.^75,178,179 However, some of the data that have been generated when the effects of prolonged calpain inhibition are investigated have been more controversial. Whereas in some studies, calpastatin overexpression in diabetic and hypertensive models attenuates the development of cardiac hypertrophy,^180,181 in other studies chronic inhibition results in cardiomyocyte degeneration and heart failure.¹⁸² Further confirmation of the effectiveness of calpain inhibition in more clinically relevant experimental models and the development of more selective inhibitors are needed to confirm the therapeutic potential of calpain inhibition.

Inhibition of MPT opening with cyclosporin A may, under some conditions, reduce the release of mitochondrial Ca²⁺ and attenuate hypercontracture.¹¹¹ Prolongation of acidosis by postconditioning is supposed to inhibit MPT.¹⁸³ Also, protection induced by stimulation of the PKG signalling pathway has been interpreted as being mediated by the inhibition of MPT,¹⁸⁴ although a direct effect of this pathway on SR Ca²⁺ handling is also likely to contribute.^100,185 It is important to remark that the mechanisms by which inhibition of MPT may prevent sarcolemmal disruption in the first minutes of reperfusion, when the cytosolic Ca²⁺ concentration remains abnormally elevated, is not known.

6. Clinical studies

Even though L-type Ca²⁺ channels may play some role in the development of Ca²⁺ overload in ischaemic cardiomyocytes (see above), there is no evidence of a protective effect of Ca²⁺ channel blockers in the setting of acute myocardial infarction. In a small clinical study (n = 36), intracoronary nisoldipine given at the onset of reperfusion increased the ejection fraction.¹⁸⁶ In contrast, in the DATA trial (n = 59), no effects on ejection fraction or enzyme release were observed after i.v. administration of diltiazem.¹⁸⁷

No NCX inhibitors have been approved for human use, and thus, there are no clinical trials on this target. However, calderet (MCC-135) is an intracellular Ca²⁺ handling modulator that is expected to inhibit, among other targets, the reverse-mode NCX and SR Ca²⁺ uptake. Calderet has been shown to reduce infarct size in the in situ dog heart when given i.v. at the end of ischaemia.¹⁸⁸ In the CASTEMI and EVOLVE randomized trials (n = 387), i.v. administration of calderet (MCC-135) did not modify infarct size or ejection fraction.^189,190

The pharmacologic approach that has been more widely tested in patients to limit Ca²⁺ overload is NHE inhibition. The GUARDIAN study (n = 11 590) did not show any benefit of cariporide on death or myocardial infarction in patients undergoing thrombolysis or coronary revascularization.¹⁹¹ Similar lack of efficacy after eniporide administration at reperfusion was reported in the ESCAMI trial (n = 1389).¹⁹² This lack of efficacy when given at the onset of reperfusion is in contrast with results obtained in trials in which NHE inhibitors were given before ischaemia, as in the EXPEDITION study,¹⁹³ and are fully consistent with preclinical studies showing that NHE inhibition protects against ischaemic injury but not against reperfusion injury, i.e. these agents need to be given before ischaemia to be effective.

Improved Ca²⁺ handling could be beneficial based on the positive results of the J-WIND-ANP trial,¹⁹⁴ in which patients with acute myocardial infarction receiving reperfusion therapy were given i.v. ANP or placebo. Patients treated with ANP showed a reduction in infarct size and an increase in ejection fraction. On the other hand, clinical evidence regarding the use of NO donors in the setting of acute myocardial infarction remains inconclusive: intracoronary sodium nitroprusside has been shown to reduce major adverse cardiac events,¹⁹⁵ whereas no benefit was observed with isosorbide mononitrate.¹⁹⁶

Finally, pharmacologic inhibition of Ca²⁺-induced MPT with cyclosporin A has been found able to prevent cell death during reperfusion in patients with acute myocardial infarction submitted to coronary angioplasty.¹⁹⁷ However, these promising results await confirmation in larger clinical trials.

In summary, there is a lack of clinically promising therapeutic strategies specifically addressing Ca²⁺-mediated myocardial reperfusion injury, but in all strategies found to be protective in patients the beneficial effect can be explained at least in part by well-demonstrated effects on Ca²⁺-mediated injury. A clear example of this is ischaemic postconditioning, a sequence of brief periods of ischaemia–reperfusion induced by repeated cycles of balloon-catheter inflation and deflation immediately applied after reopening of an occluded vessel. In experimental studies, ischaemic postconditioning has been shown to prolong acidosis and activate PKG signalling,¹⁷⁷ which result, through the mechanisms discussed above, in improved Ca²⁺ handling and reduced Ca²⁺-mediated calpain activation, hypercontracture, and permeability transition.¹²²

7. Other potential modulators of Ca²⁺ handling during reperfusion

Ca²⁺-mediated reperfusion injury could be modulated by age, sex, comorbidities, and life style, and other clinically relevant conditions. The reported reduced tolerance to ischaemia of aged hearts¹⁹⁸ appears to be partially due to impaired Ca²⁺ handling. Aged hearts showed worsened Ca²⁺ overload and reduced functional recovery in isolated heart models.¹⁹⁹ The ability of SR to take up Ca²⁺ appears to decrease with advanced age, while the contribution of NCX appears to increase.²⁰⁰ However, the effect of age could be different on different pathways of reperfusion injury. A recent report describes attenuated hypercontracture and necrosis and improved functional recovery after brief (30 min) global ischaemia in adult when compared with young guinea-pig hearts.²⁰¹

There is ample evidence that sex may influence Ca²⁺ handling and tolerance to ischaemia–reperfusion injury. Testosterone increases Ca²⁺ transient amplitude and accelerates Ca²⁺ removal during relaxation,²⁰² and ovariectomy increases RyR Ca²⁺ release and NCX activity in rat hearts.²⁰³ Part of the influence of sex on Ca²⁺ handling and cardioprotection could be mediated through enhanced PIK3-Akt-mediated PKG signalling in female hearts.²⁰⁴ Of course, the protective effect of estrogens on reperfusion injury may be mediated to a large extent by primarily Ca²⁺-independent mechanisms.²⁰⁵

Lifestyle (diet, exercise, stress, contamination) influences Ca²⁺ regulation in ways that can modify myocardial injury during ischaemia–reperfusion, and it may contribute to their effect on cardiovascular mortality. Animal studies and epidemiological evidence are consistent with the notion that exercise, when intense enough (approximately 75% of maximal oxygen consumption), activates a protective phenotype that limits infarct size secondary to subsequent transient coronary occlusion²⁰⁶ and reduces spontaneous Ca²⁺ waves in post-infarction failing cardiomyocytes.²⁰⁷ Exposure to carbon monoxide (CO) reduces tolerance to ischaemia in part by reducing SERCA and thus altering SR function,²⁰⁸ in agreement with epidemiological evidence linking mortality secondary to myocardial infarction with exposure to urban CO.

8. Conclusions

Reperfusion injury has become an important part of the current scientific paradigm of tissue damage secondary to ischaemia in the heart, and this applies also to other organs. At the same time, the role of alterations in intracellular Ca²⁺ handling in its genesis has lost weight in favour of other mechanisms that can operate, mainly through MPT. Although the importance of these mechanisms is unquestionable, available data indicate that, in cardiomyocytes, altered Ca²⁺ handling may directly cause cell death through calpain-mediated proteolysis, hypercontracture, and, likely, MPT. These mechanisms appear to be critical in necrotic cell death occurring during the initial minutes of reperfusion, particularly after brief periods of ischaemia. Developing effective and safe treatments to prevent Ca²⁺-mediated cell death in patients with acute myocardial infarction remains a challenge that needs to be met. The main difficulty in this task is the small-time window after reperfusion onset during which Ca²⁺-mediated injury can be prevented. However, the widespread use of primary angioplasty in patients with acute myocardial infarction offers the opportunity to selectively apply treatments to the area at risk at the very onset of reperfusion.

Conflict of interest: none declared.

Funding

Supported by the Spanish Ministry of Science and Instituto de Salud Carlos III (RETICS-RECAVA RD06/0014/0025; CICYT SAF/2008-03067, FIS-PI080238, and PS09/02034). A.R.-S. is recipient of a contract from Generalitat de Catalunya (Programa d'estabilització d'investigadors Miguel Servet, Departament de Salut).

Author notes

This article was guest-edited by Klaus-Dieter Schlüter, Justus Liebig University of Giessen, Giessen, Germany.

This article is part of the Spotlight Issue on: Reducing the Impact of Myocardial Ischaemia/Reperfusion Injury