Introduction

For patients presenting with an acute myocardial infarction (AMI), it is well-established that early, effective myocardial reperfusion using either thrombolysis or primary percutaneous coronary intervention (PCI), remains the most powerful intervention for limiting myocardial infarct size. However, the morbidity and mortality from an AMI remains significant, necessitating the development of new strategies for cardioprotection, which can further reduce myocardial infarct size and improve clinical outcomes in this patient group. Ideally, any such novel cardioprotective strategy, would be available to be applied in conjunction with the current myocardial reperfusion therapy, and be demonstrated to confer a cardioprotective effect, in terms of myocardial infarct size reduction and improved clinical outcomes, over and above that elicited by coronary artery reperfusion and ancillary therapies (such as antiplatelet and antithrombotic treatments).

In this regard, the Reperfusion Injury Salvage Kinase (RISK) pathway, a term given to describe a group of survival protein kinases which include Akt and Erk1/2 that confer powerful cardioprotection, when specifically activated at the time of myocardial reperfusion, represents a novel target for cardioprotection in AMI patients [1, 2]. However, despite the abundance of preclinical data demonstrating effective cardioprotection with a variety of different agents given at the time of myocardial reperfusion to activate the RISK pathway, clinical studies are limited, a situation which should change with the revelation that both ischemic preconditioning and postconditioning also recruit the RISK pathway [3, 4], thereby regenerating interest in the myocardial reperfusion phase as a viable target for cardioprotection in AMI patients.

Ischemic preconditioning and postconditioning: ‘United’ by the RISK pathway

The requirement for intervening at myocardial reperfusion in AMI patients, renders ischemic preconditioning (IPC) ineffective as a cardioprotective intervention, given that its protective effect results from the application of one or more short-lived episodes of ischemia and reperfusion, applied before the index ischemic event [5], which is naturally unpredictable in AMI patients. This restricts its utility to scenarios in which the index myocardial ischemic episode can be reliably anticipated such as in those undergoing coronary artery bypass graft (CABG) surgery [6] or in unstable angina patients presenting with a threatening myocardial infarct [7]. However, emerging studies suggest that the signal transduction pathways underlying the cardioprotection elicited by IPC converge at the myocardial reperfusion phase, with the RISK pathway identified as a key component [8–10], thereby enabling the use of pharmacological agents that target components of the RISK pathway to harness the protective benefits of IPC for AMI patients.

As an interventional strategy which can be applied at the time of myocardial reperfusion, the recently introduced phenomenon of ischemic postconditioning (IPost), that describes the cardioprotective effect elicited by interrupting myocardial reperfusion with short-lived episodes of myocardial ischemia interspersed with reperfusion following the index ischemic event [11], offers an amenable strategy for cardioprotection in AMI patients, and its introduction has succeeded in renewing interest in the myocardial reperfusion phase as a target for cardioprotection [12, 13]. The clinical efficacy of IPost as a cardioprotective strategy has already been demonstrated in several small clinical studies of patients undergoing primary PCI, using an invasive IPost protocol comprising serial low-pressure coronary angioplasty inflations and deflations immediately following the deployment of the stent in the infarct-related coronary artery [14–17]. However, the widespread use of IPost in the clinical arena is likely to be limited by both its invasive nature and the fact that it is restricted to AMI patients undergoing PCI. A more amenable approach will be to mimic IPost, using pharmacological agents that target the RISK-pathway that has also been identified as underlying IPost-induced cardioprotection, thereby obviating the need for such an invasive IPost protocol.

This article will review the origins and evolution of the RISK pathway as a potential common cardioprotective pathway which can be activated by administering agents given either at the time of reperfusion or prior to the index ischemic event. The non-pharmacological activation of the RISK pathway as part of a common cardioprotective pathway which ‘unites’ IPC and IPost at the time of myocardial reperfusion, and the clinical implications of the RISK pathway, as a novel cardioprotective target will also be covered. This article will focus only on those protein kinases specifically modulated at the time of myocardial reperfusion in the setting of cardioprotection, and the reader is directed to several other reviews for a more comprehensive account detailing the contribution of protein kinases to cardioprotection [18–22].

The origins and evolution of the RISK pathway

The Reperfusion Injury Salvage Kinase (RISK) pathway emerged as a concept in the late 1990s with the recognition that apoptotic cell death contributed to lethal reperfusion injury [23–25], and the knowledge that there existed certain pro-survival anti-apoptotic protein kinases, the original members of which were Akt and Erk1/2, which when specifically activated at the time of myocardial reperfusion conferred powerful cardioprotection [1, 2]. Studies had previously demonstrated activation of these protein kinases Akt [26, 27], Erk1/2 [28] and JNK [28, 29] at the time of myocardial reperfusion in control hearts, but clearly the activation of the RISK pathway in these settings was not sufficient to confer cardioprotection, and an additional pharmacological stimulus was required to enhance the activation of the RISK pathway. Other studies had confirmed the cardioprotective potential of both Akt [30, 31] and Erk1/2 [32] using transgenic activation of these kinases.

Alongside an ever expanding list of diverse pharmacological agents, demonstrated to confer cardioprotection when administered at the time of myocardial reperfusion through the activation of the RISK pathway (see Table 1), the concept of the RISK pathway has evolved to encompass several novel features: (a) the RISK pathway can be activated by interventions instituted prior to the index ischemic event, which includes pharmacological preconditioning agents, such as isoflurane [88] and opioids [89], as well as by IPC itself [4]; (b) the RISK pathway now includes other cardioprotective reperfusion salvage kinases such as PKC (primarily the PKC-ε isoform), PKG, p70s6K, and GSK-3β; (c) there are protein kinases such as PKC-δ and rho-kinase which when activated at the time of myocardial reperfusion are pro-injurious and counteract the cardioprotection elicited by the RISK pathway. The roles played by p38 and JNK MAPK are controversial, a term frequently used when discussing these protein kinases in the context of cardioprotection [19, 20, 94], with studies reporting both cardioprotective and pro-injurious roles of these kinases at the time of myocardial reperfusion [80, 88, 90]; and (d) the RISK pathway appears to be a core component of a common cardioprotective pathway which converges on the mitochondrial permeability transition pore, that appears to ‘unite’ both IPC and IPost at the time of myocardial reperfusion [95].

Table 1 List of agents or factors which elicit cardioprotection specifically at the time of myocardial reperfusion through the activation of the RISK pathway
Full size table

The critical time ‘Window’ for RISK pathway activation

Preclinical studies clearly demonstrate that cardioprotection at the time of myocardial reperfusion through the activation of the RISK pathway whether that be by administering insulin [35] or by applying an IPost protocol [96], must be instituted at the immediate onset of reperfusion to be effective, suggesting the existence of a critical time ‘window’ for cardioprotection. During the first few minutes of myocardial reperfusion, in response to the generation of ROS, an increase in mitochondrial Ca2+, the restoration of normal pH, the mitochondrial permeability transition pore (mPTP) opens [97–99], mediating cell death by uncoupling oxidative phosphorylation and inducing mitochondrial swelling [100, 101]. Pharmacologically inhibiting mPTP opening after the first few minutes of myocardial reperfusion have elapsed is ineffective, confirming the existence of this critical time ‘window’ of cardioprotection [102]. The implications of these findings for the patient presenting with an AMI are that, to be effective as a cardioprotective intervention, any pharmacological agent used to activate the RISK pathway, needs to be administered either prior to or at the immediate onset of myocardial reperfusion, and such a requirement should be a critical feature in the design of a clinical trial.

Activation of the RISK pathway by intervening at the time of reperfusion

The clinical requirement to demonstrate that a potential cardioprotective agent can attenuate myocardial injury, when given specifically at the time of myocardial reperfusion, has resulted in an increasing number of pharmacological agents being linked to the activation of the RISK pathway (see Table 1). Growth factors were the first group of agents demonstrated to exert a cardioprotective effect at the time of myocardial reperfusion through the activation of the RISK pathway, but this group has now grown to include various other receptor ligands as well as non-pharmacological activators of the RISK pathway such as IPost.

Growth factors as activators of the RISK pathway

Among the first growth factors demonstrated to confer cardioprotection when administered specifically at the time of myocardial reperfusion through the activation of one or more components of the RISK pathway are outlined below (and see Table 1). Ligand binding at the growth factor receptor, results in the activation of its receptor tyrosine kinase, which then activates the PI3K-Akt and Ras-MEK1-2-Erk1/2 signalling cascades.

Transforming Growth Factor-β1 (TGF-β1)

A cytokine that regulates cell growth and differentiation and modulates apoptosis in many cell types, has previously been demonstrated to confer cardioprotection [103]. Baxter et al. [33] demonstrated that TGF-β1 administered at the time of myocardial reperfusion or reoxygenation reduced myocardial infarct size and attenuated apoptotic cardiomyocyte death, respectively, effects which were abolished by the MEK1/2 inhibitor PD98059;

Insulin

Insulin has been reported to exert cardioprotection when administered at the time of myocardial reperfusion through the activation of the PI3K-Akt [34–36] component of the RISK pathway and the recruitment of downstream targets including the phosphorylation of p70S6K [35], BAD [35], and eNOS [36].

Insulin-like growth factor-1 (IGF-1)

A polypeptide which regulates cell proliferation and differentiation in different cell types in response to a diverse array of stimuli, has been extensively demonstrated to cardioprotect through the activation of Akt [27, 104–108] and Erk1/2 [105] and their downstream signalling elements including the inhibition of BAX [107, 109, 110], enhanced Bcl-2 expression [109, 110], and inhibition of mPTP opening [110]. Studies have demonstrated improved recovery of LV systolic function and reduced myocardial injury with IGF-1 given specifically at the time of myocardial reperfusion in the perfused rat heart, an effect which was abrogated by the PI3K inhibitor [37]. Yamashita et al. [27] demonstrated that mice heterozygously over-expressing IGF-1 displayed enhanced cardioprotective Akt activation in response to myocardial reperfusion when compared to wild type controls.

Corticotrophin-1 (CT-1)

A member of the IL-6 group of cytokines had been originally isolated as a cardiomyocyte hypertrophic growth factor in 1995 [111], with subsequent studies implicating a cardioprotective role for this cytokine [112], through anti-apoptotic signalling pathways mediated by the activation of Erk1/2 [113]. Importantly, Brar et al. [38] found that both the pharmacological and genetic inhibition of PI3K, Akt and Erk1/2 abrogated the cardioprotective effect elicited by CT-1 when administered at the time of reoxygenation to neonatal rat cardiomyocytes, a finding confirmed using adult-rat cardiomyocytes and the perfused rat heart [39, 40].

Subsequent studies have linked growth factors such as fibroblast growth factor-2 (FGF-2) [41, 114], erythropoietin (EPO) [42, 43], and most recently the adipocytokines, leptin [45, 115], apelin (Simpkin et al. unpublished) and visfatin (Hausenloy et al. unpublished) with cardioprotection elicited at the time of myocardial reperfusion through the activation of one or more components of the RISK pathway (see Table 1). Most recently, Granulocyte Colony-Stimulating Factor (G-CSF), a cytokine that mediates the proliferation and differentiation of neutrophil progenitors, which has been found to prevent post-myocardial infarction cardiac remodelling through the JAK-STAT pathway [116], has been found to mediate cardioprotection when administered at the time of myocardial reperfusion through the activation of JAK2, STAT3, ERK1/2, Akt, and eNOS, with the suggestion that JAK2 which was reported to be upstream of the RISK pathway [44] may act as the intermediary between the cytokine receptor and the RISK pathway.

G-protein coupled receptor ligands as activators of the RISK pathway

Several agents have now been demonstrated to confer cardioprotection at the time of myocardial reperfusion by binding to their specific G-protein coupled receptor (GPCR) and activating the RISK pathway (see Table 1). Ligand binding at the GPCR, results in the transactivation of receptor tyrosine kinases which in turn activate the PI3K-Akt and Ras-MEK1-2-Erk1/2 signalling cascades.

  1. (a)

    Urocortin, a peptide related to corticotrophin-releasing factor, has been reported to protect neonatal cardiomyocytes against hypoxia-reoxygenation through the activation of Erk1/2, when given specifically at the time of reoxygenation [48], and reduce myocardial infarct size when administered at the time of myocardial reperfusion in both the perfused and in situ rat hearts [49]. Furthermore, it has been demonstrated that the cardioprotection elicited by both urocortin and its analogue, stresscopin, when given at the time of reoxygenation to neonatal rat cardiomyocytes was antagonized by the pharmacological and genetic inhibition of PI3K, Akt and Erk1/2 [46, 47].

  2. (b)

    Adenosine (through the use of various adenosine receptor agonists [51–53]), bradykinin [51, 54] and opioids [55] have all been demonstrated to reduce myocardial infarct size when administered at the time of myocardial reperfusion through the activation of the RISK pathway.

  3. (c)

    Adrenomedullin, a vasodilating peptide, which was first isolated from human phaeochromocytoma tissue in 1993 [117], binds to the calcitonin gene-related peptide like receptor. Animal studies have demonstrated that adrenomedullin given via adenoviral transfer [118, 119], acutely to the in situ rat heart [120], or specifically at the time of myocardial reperfusion [56], reduces myocardial infarct size and attenuates apoptotic cardiomyocyte death through the activation of the RISK pathway and downstream targets including BAX suppression, the phosphorylation of BAD, Bcl2 [118] and GSK-3β [119] and NO release [56].

  4. (d)

    Glucagon-Like Peptide-1 (GLP-1) is a gut incretin hormone which stimulates insulin secretion has emerged as a potential novel anti-diabetic agent [121]. Interestingly, our laboratory has demonstrated that GLP-1 administered at the time of myocardial reperfusion reduces myocardial infarct size through the activation of the RISK pathway [57].

Other receptor mediated activation of the RISK pathway

Natriuretic peptides

Both atrial natriuretic peptide (ANP) and brain natriuretic peptide (BNP), which bind to the natriuretic peptide receptor A, a membrane-bound guanylyl cyclase receptor, have been demonstrated to confer cardioprotection when given at the time of myocardial reperfusion [122, 123], although only ANP has been demonstrated to confer its protective effect through the activation of the RISK pathway. The mechanism through which ANP activates the RISK pathway is currently unclear.

Estrogens

Although the cardioprotective role of hormone replacement therapy is not clear, preclinical studies have demonstrated cardioprotection with estrogen replacement using a chronic myocardial infarction model in ovariectomized female mice through the activation of the PI3K-Akt pathway [124], and other studies have reported the cardioprotective effect of estrogens administered at the time of myocardial reperfusion [125]. More recently, the RISK pathway has been reported to mediate the infarct-limiting effects of 17-β estradiol and the phytoestrogen, genistein [60], a drug usually used to inhibit tyrosine kinase. Recent studies suggest that there exists a membrane bound estrogen receptor which has been demonstrated to be linked to activation of PI3K and MAPK’s [126, 127].

CGX-1051

A synthetic version of a peptide from Conus snail venom, has been reported by Zhang et al. [61] to reduce myocardial infarct size when administered at the time of myocardial reperfusion through MEK1/2 using the in situ rabbit heart.

Non-receptor mediated activation of the RISK pathway

Emerging studies suggest that components of the RISK pathway can be activated by pharmacological agents which exert their intracellular effect through non-receptor mediated mechanisms, the most extensively investigated being the volatile anesthetics.

Following the introduction of ischemic postconditioning as a cardioprotective strategy [11], and on the background of studies demonstrating the ability of volatile anesthetics to mimic the effects of ischemic preconditioning [128, 129] and exert cardioprotection when administered at the time of myocardial reperfusion [130, 131], a growing body of studies suggest that volatile anesthetics, when administered at the time of myocardial reperfusion are able to reduce myocardial injury through the recruitment of the RISK pathway [62], and its downstream targets, p70s6K, GSK-3β, eNOS, Bcl-2 and the mPTP [63, 68] both in the normal [63, 64, 67] and infarct-remodelled myocardium [65], thereby introducing the concept of ‘pharmacological’ or ‘anesthetic’ postconditioning.

Studies have reported that morphine was able to potentiate the cardioprotective effect of isoflurane administered at the time of reperfusion [66]. Furthermore, isoflurane [68, 132] and pharmacological inhibition of the pro-apoptotic protein p53 [133], have been found to potentiate the cardioprotective effects of IPost, suggesting the existence of a threshold level of stimulation, required to mediate cardioprotection at the time of myocardial reperfusion.

Other non-receptor mediated agents demonstrated to confer cardioprotection at the time of myocardial reperfusion through the activation of the RISK pathway include: (a) the 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitor (‘statins’), atorvastatin [69, 70]. Pravastatin, Pitavastatin, and Cerivastatin given pre-ischemically have been demonstrated to activate Akt at myocardial reperfusion [91]. Simvastatin has been shown in separate studies to confer cardioprotection when given at the time of myocardial reperfusion [134] and activate Akt [135]; (b) the antidiabetic biguanide, metformin, administered at the time of myocardial reperfusion, have been reported to mediate a reduction in myocardial infarct size through the activation of Akt in perfused non-diabetic and diabetic rat hearts, and have been demonstrated to delay mPTP opening in rat cardiomyocytes subjected to oxidative stress through PI3K (Bhamra et al. Unpublished) However, the mechanism through which these agents mediate the activation of the RISK pathway at the time of myocardial reperfusion is currently unclear.

Mechanical activation of the RISK pathway: Ischemic postconditioning

Our laboratory was the first to demonstrate the link between the cardioprotection elicited by IPost and the recruitment of the RISK pathway, in a study in which we reported that the activation of Akt and the downstream p70s6K, contributed to the infarct-size reduction in postconditioned perfused rat hearts [3]. Several subsequent studies have confirmed the contribution of Akt activation to IPost-induced cardioprotection (see Table 1) [62, 71–73, 76, 77], including in diseased MI and LVH-remodelled rat hearts [74, 75], although a single study using perfused rabbit hearts failed to demonstrate Akt activation in postconditioned hearts [78].

The other component of the RISK pathway, Erk1/2, has also been linked to IPost-induced protection with activation of Erk1/2 in postconditioned hearts [77, 78, 78, 79, 136]. Interestingly, Schwartz et al. [136] demonstrated both Akt and Erk1/2 activation using in situ porcine hearts subjected to a non-cardioprotective IPost protocol, suggesting perhaps that before protection is observed, a threshold level of activation of the RISK pathway may be required.

Subsequent studies have demonstrated the activation of PKC as a critical mediator of IPost-induced cardioprotection implicating the PKC-ε isoform as the survival kinase in this setting (see Table 1) [81, 82, 84]. Where PKC activation is situated in relation to the activation of Akt and Erk1/2 is unclear, although pharmacological activation of PKC at the time of myocardial reperfusion was suggested to be upstream of both the adenosine receptor and Akt, although this study did not examine events occurring in IPost directly [81]. Preliminary studies have implicated PKG activation in the cardioprotection elicited by IPost [73, 87]. Finally, using neonatal rat cardiomyocytes, Sun et al. [80] found that hypoxic postconditioning mediated cardioprotection through the reduction in both p38 and JNK activity, finding which do not agree with those demonstrating no effect [93] or their activation [88, 90] at the time of myocardial reperfusion, but the role of these kinases in the setting of cardioprotection has frequently courted controversy.

Intriguingly, Bopassa and colleagues [71] have reported that the infarct-limiting effect observed with low-pressure reperfusion of isolated perfused rat hearts could be abolished by reperfusing with pharmacological inhibitors of PI3K, suggesting the potential involvement of the RISK pathway in low-pressure reperfusion, although Akt activity was not measured in this study. This form of controlled reperfusion may, in common with IPost, be simply a form of modified reperfusion which cardioprotects through the RISK pathway [137].

The mechanism through which the RISK pathway is activated in the setting of IPost, is unclear although it may be due to the ligand binding of either adenosine [138] or opioids [139] with their cell surface receptor, although this has not been directly demonstrated. A more recent proposal is that a transient period of acidosis at the time of myocardial reperfusion, perhaps mediating a delayed restoration of neutral pH from reduced lactate wash-out, may contribute to both the activation of the RISK pathway and cardioprotection observed in postconditioned hearts [77], a finding that may be expected to attenuate mPTP opening given that the restoration of neutral pH at the time of myocardial reperfusion is a critical determinant of mPTP opening [99].

Activation of the RISK pathway by intervening prior to Ischemia

Interestingly, several studies suggest that components of the RISK pathway (including Akt, Erk1/2 p38 MAPK, JNK MAPK) can be activated by an intervention applied prior to the index ischemic period, whether that be by pharmacological agents such as ‘statins’ [91] (using pravastatin, pitavastatin and cerivastatin), a δ1-opioid agonist [89], isoflurane [88], pioglitazone [92] or Ag II [93], administered as preconditioning mimetics, or the cardioprotective phenomenon of IPC [4, 9, 10]. IPC or pre-treatment with ‘statins’ resulted in the activation of Akt and/or Erk1/2 at the time of myocardial reperfusion and crucially, the administration of a pharmacological inhibitor of the RISK pathway at the time of reperfusion, abolished cardioprotection [4, 91], suggesting that the activation of the RISK pathway was essential for protection.

The mechanism through which a protective stimulus applied pre-ischemically appears to recruit the RISK pathway at the time of myocardial reperfusion is unclear, although potential explanations include: (a) the activation of the RISK pathway observed at the time of myocardial reperfusion is in fact a continuation of that initiated pre-ischemically by the preconditioning stimulus. This would suggest that the kinase activation initiated by the preconditioning stimulus needs to be sustained into the myocardial reperfusion phase to confer cardioprotection. A recent study suggests that Akt activation is required for up to 50–60 min into reperfusion, whereas Erk1/2 activation is needed only the first 5–10 min of reperfusion to mediate the cardioprotection elicited by IPC [9]. Evidence in support of this explanation is provided by studies reporting that inhibiting either MEK1/2, PKC or the mKATP channel during the preconditioning phase attenuated the Erk1/2 and p38 activation observed in hearts preconditioned with either IPC, a δ1-opioid agonist or isoflurane [88, 89]; (b) the preconditioning stimulus primes the protein kinases by inducing their intracellular translocation to their sites of action, such that at the time of myocardial reperfusion, kinase activation is enhanced; (c) the preconditioning stimulus may induce the release of certain growth factors within the myocardium which then augment RISK pathway activation at the time of myocardial reperfusion in conjunction with ROS, as observed in the study demonstrating enhanced Akt activation at the time of myocardial reperfusion in mice heterozygously over-expressing IGF-1, which exhibited raised endogenous levels of IGF-1 and Akt activation [27].The enhanced Akt activation observed at the time of reperfusion was abolished in the presence of the antioxidant N-acetylcysteine [27], suggesting a potential signalling role for ROS in this setting; (d) adenosine A1/A2B receptor ligand binding at the time of myocardial reperfusion in preconditioned hearts may mediate the activation of the RISK pathway [9]. Solenkova et al. [9] reported that the administration of a non-specific adenosine receptor blocker at the time of myocardial reperfusion abrogated both the infarct-limiting effect of IPC as well as the Akt activation observed at the time of reperfusion in preconditioned hearts. Whether endogenous adenosine is generated in greater quantities in preconditioned hearts, or whether specific adenosine receptors are more sensitive to adenosine, an effect perhaps mediated by PKC, a known mediator of IPC, is currently unknown.

Interestingly, when comparing the recruitment of the RISK pathway, there appear in some cases to be differences between pharmacological preconditioning mimetics and IPC. For example, IPC was found to activate both p38 and Erk1/2 at the time of myocardial reperfusion, whereas only Erk1/2 was activated in hearts preconditioned with isoflurane [88]. Furthermore, Lecour et al. [10] reported that although the preconditioning-mimetic TNF-α, in common with IPC, exerted its cardioprotective effect through the activation of STAT-3, it did not elicit its protective effect through the conventional components of the RISK pathway, Akt and Erk1/2.

An important study by Bell et al. [93] has extended the contribution of RISK pathway activation to the phenomenon of delayed preconditioning (in which a preconditioning stimulus has been demonstrated to elicit cardioprotection 12–24 h later [140]), and demonstrated dissociation between RISK pathway activation and cardioprotection. Using AgII as their preconditioning mimetic they demonstrated the activation of both Akt and Erk1/2 at the time of myocardial reperfusion in hearts at 1, 6 and 24 h following the preconditioning stimulus, yet cardioprotection, corresponding to the time-points of classical and delayed preconditioning, was only observed at 1 and 24 h, respectively [93]. Why RISK pathway activation and cardioprotection were dissociated at 6 h is unclear, but it may suggest the requirement for another factor such as PKC to be present to mediate cardioprotection, or that the intracellular localization of components of the RISK pathway and its downstream effectors might not be optimized at the 6-h time-point.

Effector mechanisms of RISK pathway activation

When the concept of the RISK pathway was originally conceived, the antiapoptotic signalling pro-survival pathways recruited by Akt and Erk1/2, were proposed as the mechanism through which the RISK pathway conferred cardioprotection at the time of myocardial reperfusion [2, 1]. This notion has been subsequently supported by the many studies reporting, the infarct-limiting effect of pharmacologically activating the RISK pathway, to be associated with the recruitment of antiapoptotic signalling systems such as the phosphorylation and inhibition of the proapoptotic proteins BAX and BAD, the inhibition of caspase 3 activation, and the phosphorylation and activation of p70s6K (which acts to inhibit BAD [141]) and the phosphorylation and activation of the antiapoptotic protein Bcl-2 [2](see Table 1 and Fig. 1).

Fig. 1
figure 1

Scheme demonstrating the diverse variety of agents which activate the Reperfusion Injury Salvage Kinase (RISK) pathway in both a receptor and non-receptor mediated manner. Interestingly, in addition to being activated by intervening at the time of myocardial reperfusion, the RISK pathway can also be activated by interventions applied pre-ischemically such as ischemic preconditioning and volatile anesthetics, opioids, and ‘Statins’. The activation of the RISK pathway mediates cell survival through various pathways including various anti-apoptotic mechanisms, by inhibiting the opening of the mitochondrial permeability transition pore (mPTP) and by possibly inhibiting autophagy. The activation of the RISK pathway and the subsequent inhibition of mPTP opening provides a common cardioprotective pathway recruited at the time of myocardial reperfusion, ‘uniting’ the cardioprotective phenomena of ischemic preconditioning and postconditioning, which can be targeted by pharmacological agents at the time of myocardial reperfusion as a novel cardioprotective strategy in patients presenting with an AMI

Full size image

Clearly, given the size of the myocardial infarct reduction elicited by the RISK pathway activation in the majority of studies, there must be other anti-necrotic protective mechanisms contributing to the cardioprotective effects of the RISK pathway. In this regard, the inhibition of the mitochondrial permeability transition pore (mPTP), a mitochondrial channel which mediates cell death at the time of myocardial reperfusion by uncoupling oxidative phosphorylation and inducing mitochondrial swelling [101, 100], have been identified as a down stream target of the RISK pathway [45, 71, 142, 143]. However, the mechanism through which the RISK pathway inhibits the opening of the mPTP is unclear, although there are several hypotheses (see Fig. 1): (a) GSK-3β, a downstream target of the RISK pathway has been linked to the inhibition of mPTP opening in the context of cardioprotection [142]; (b) eNOS, another downstream target of the RISK pathway has the potential for inhibiting mPTP opening either through the PKG-PKC-ε-mKATP channel signalling pathway [144–147] or it may suppress mPTP opening through the generation of nitric oxide [148]; (c) the inhibition of BAX translocation to mitochondria [149] and/or the activation of mitochondrial hexokinase II [150, 151] may act in concert to inhibit mPTP opening; (d) finally, Abdallah et al. [152] have demonstrated that activating PI3K using insulin can reduce the uptake of calcium by the sarcoplasmic reticulum, which may in turn act to inhibit mPTP opening at the time of myocardial reperfusion.

Sanada et al. [91] have previously demonstrated that the administration of ‘statins’ pre-ischemically results in the phosphorylation of PI3K and the subsequent activation of ecto-5′-nucleotidase, an effect which would act to promote adenosine release at the time of myocardial reperfusion, which in itself could confer a cardioprotective effect.

Finally, an interesting recent study by Valentim et al. [50] suggests that the activation of the Akt but not the Erk1/2 component of the RISK pathway using urocortin may attenuate autophagy (a lysosomal degradative pathway that has emerged as a form of programmed cell death distinct from apoptosis), an effect which is in part due to the inhibition of Beclin-1, a critical mediator of autophagy. Furthermore, in support of a detrimental role for autophagy in the setting of myocardial ischemia-reperfusion injury, autophagocytosis has been reported to be inhibited in both preconditioned and postconditioned cardiomyocytes [153].

Pro-injurious protein kinases which counteract the RISK pathway

Somewhat intriguingly, it appears that the pro-survival RISK pathway may have its pro-injurious antithesis in the form of a collection of protein kinases that are detrimental, when activated at the time of myocardial reperfusion. This role is exemplified by the protein kinase, PKC-δ, the activation of which at the time of myocardial reperfusion increases myocardial infarct size [21]. Studies have reported that the selective inhibition of PKC-δ specifically at the time of myocardial reperfusion decreases myocardial infarct size [84, 154, 155]. Furthermore, Zatta et al. [84] have reported the inhibition of PKC-δ translocation to the mitochondria using in situ post-conditioned rat hearts.

Abnormal activation of Rho-kinase (ROCK), the downstream target of the small GTPase, Rho-A, mediates cardiovascular damage including myocardial ischemia-reperfusion injury and hypertension, and its inhibition underlies some of the pleiotropic effects of ‘statins’ [156, 157]. Studies have revealed that myocardial ischemia and reperfusion activate ROCK in ischemic myocardium and pharmacologically inhibiting its activation is cardioprotective [158] and this protective effect appears to be mediated through the activation of the PI3K-Akt-eNOS pathway [159], suggesting that ROCK exerts its pro-injurious effect by counteracting the RISK pathway. Importantly, data from Hamid et al. [160] have demonstrated that administering ROCK inhibitors specifically at the time of myocardial reperfusion reduces myocardial infarct size through the PI3K-Akt-NO pathway.

Sun et al. [80] demonstrated that the treatment of neonatal cardiomyocytes with hypoxic postconditioning reduced the activation of both JNK and p38 MAPK’s, and the pharmacological activation of these MAPK’s at the time of reoxygenation abolished the cardioprotective effect elicited by postconditioning, suggesting a detrimental role for these kinases at the time of reoxygenation. In contrast however, IPC has been reported to activate both p38 [88] and JNK [90] MAPK’s at the time of myocardial reperfusion, suggesting that the modulation of these particular kinase members of the RISK pathway in the setting of cardioprotection is more complex.

The RISK pathway as a target for cardioprotection: clinical application

As an interventional strategy which can be applied at the time of myocardial reperfusion for patients presenting with an AMI, the use of pharmacological agents to target the RISK pathway is certainly a viable proposition. The major clinical application of this cardioprotective strategy would be as adjuvant therapy to myocardial reperfusion for AMI patients undergoing either thrombolysis or primary PCI, although it could also be used to extend the ‘time window’ for intervention in those AMI patients in which a delay in myocardial reperfusion is anticipated. Patients receiving PCI for unstable angina/NSTEMI, and patients undergoing CABG surgery or cardiac transplantation as well as those patients surviving a cardiac arrest, also might experience acute myocardial ischemia-reperfusion injury and therefore might accrue benefit from such a cardioprotective strategy.

Many of the pharmacological agents linked to RISK pathway activation in basic science studies are already in clinical use today (see Tables 1 and 2), facilitating their investigation as potential cardioprotective agents. Importantly, the design of any clinical study investigating this form of cardioprotective strategy in AMI patients should ensure that the pharmacological agent is administered either prior to or at the immediate onset of myocardial reperfusion to ensure effective delivery of the agent to the reperfused myocardium and RISK pathway activation within the time-frame of the ‘window’ for cardioprotection. Several of the clinical studies reviewed below administered their cardioprotective agent to AMI patients after the onset of myocardial reperfusion, which may explain in part the lack of benefit of their treatment strategy (see Table 2).

Table 2 Activators of the RISK pathway with clinical potential if given as adjunctive therapy to myocardial reperfusion
Full size table

Several pharmacological agents now known from pre-clinical studies to confer cardioprotection when given at the time of myocardial reperfusion through the activation of the RISK pathway have been previously investigated in clinical trials and are reviewed below (see Table 2). Clearly, whether the activation of the RISK pathway actually underlies the cardioprotective effect of these agents when used in the clinical arena is not clear, although data from our laboratory suggest that the RISK pathway does operate in human myocardial tissue to confer cardioprotection, in a study in which we have demonstrated that both EPO and IPost cardioprotect human atrial trabeculae harvested from patients undergoing CABG surgery through the activation of the RISK pathway (Mudgliari et al. Unpublished).

Activators of the RISK pathway which have demonstrated clinical cardioprotection

The large multi-centred Japanese clinical study entitled J-WIND-ANP [161], which recently reported its main findings at the AHA Scientific Sessions 2006, found that a 72 h infusion of Carperitide, a recombinant form of human ANP, conferred cardioprotection in over 600 patients presenting with an AMI undergoing PCI as evidenced by a 14.7% reduction in myocardial infarct size (measured by CK and troponin T) and a 5.1% increase in ejection fraction. In addition, the secondary endpoint of reperfusion injury (assessed by the presence of malignant ventricular arrhythmias during reperfusion periods, re-elevation of ST segments, and worsening of chest pain) was decreased by 25.9% and Carperitide also reduced the incidence of cardiac death and re-hospitalization for heart failure by 73.3% compared with placebo. Although these data are promising, further large-scale clinical studies using primary clinical outcome endpoints are required to determine whether the cardioprotective benefits of this treatment strategy extend to an improvement in clinical outcomes.

The recently described phenomenon of ischemic postconditioning (IPost) had already been demonstrated to reduce myocardial injury in several small clinical studies of AMI patients undergoing PCI [14–17]. In these studies, a series of low-pressure inflations and deflations of the coronary angioplasty balloon administered upstream of the deployed stent in the infarct-related coronary artery were demonstrated to attenuate myocardial reperfusion injury and reduce myocardial infarct size as measured by cardiac enzymes and nuclear scanning [14–17]. Further large clinical studies are required to determine whether the cardioprotective benefits translate to improved clinical outcomes in this patient group.

The administration of high-dose ‘statins’ to patients with acute coronary syndromes more than 24 h following PCI has been demonstrated to improve clinical outcomes [162, 163], but whether high dose ‘statins’ given at the time of myocardial reperfusion, confers any further cardioprotective effect is unclear. A small preliminary study, has demonstrated that high-dose atorvastatin (80 mg) administered before PCI in patients with non-ST elevation MI or unstable angina conferred clinical benefit [173].

The administration of high-dose adenosine as an adjunct to myocardial reperfusion in AMI has demonstrated myocardial protection in several small clinical studies [174, 175]. Larger randomized controlled studies, in which intravenous adenosine was commenced after the onset of reperfusion therapy, reported an 11% reduction in myocardial infarct size but no improvement in clinical outcomes with this treatment strategy [164, 165]. Adequately powered larger clinical studies administering adenosine prior to the onset of myocardial reperfusion are required to provide evidence of improved clinical outcomes with this treatment strategy.

In a small clinical study comprising 10 AMI patients with impaired LV systolic function undergoing primary PCI, a 72 h infusion of glucagon-like peptide-1 (GLP-1), administered more than 3 h following reperfusion, was demonstrated to improve ejection fraction from 29 to 39% [166]. However, the effect of GLP-1 on subsequent myocardial injury was not investigated in this study. The use of GLP-1 as both an antidiabetic and potential cardioprotective agent is limited by the fact that it is rapidly broken down by endogenous dipeptidyl peptidase-IV (DPPIV). The longer acting GLP-1 analogues, which are resistant to DDPIV breakdown such as Exenatide, or the novel DPPIV inhibitors Sitagliptin and Vildagliptin, which act to augment endogenous GLP-1, may offer more promise, but their cardioprotective effect needs to be first determined in preclinical studies.

Activators of the RISK pathway which have failed to demonstrate clinical cardioprotection

The cardioprotective potential of the glucose-insulin-potassium (GIK) cocktail was comprehensively examined in the large multi-centred randomized clinical trial comprising 20,201 patients undergoing PCI or thrombolysis for an ST-elevation MI, and was found to confer no beneficial effect in terms of mortality, cardiac arrest, cardiogenic shock and re-infarction at 30 days [176]. Potential explanations for the lack of cardioprotection include: (a) the preclinical studies demonstrating a reduction in myocardial infarct size used insulin alone on the most part [35], with only one study demonstrating cardioprotection with GIK therapy at the time of myocardial reperfusion [177]; (b) a delay in the administration of GIK therapy in relation to the onset of myocardial reperfusion therapy and the prolonged myocardial ischemic time, with pre-clinical studies suggesting benefit with insulin therapy at the immediate onset of myocardial reperfusion [35, 178], and following a shorter myocardial ischemic time [178].

Several small clinical studies have demonstrated reduced myocardial injury [179] and preserved left ventricular function [180] with the use of volatile anesthetics as preconditioning agents in patients undergoing CABG surgery. However, a recent meta-analysis has reported no beneficial effect of volatile anesthetics on rates of myocardial infarction and mortality in patients undergoing CABG surgery [181]. Larger clinical trials are required to determine whether volatile anesthetics are beneficial in cardiac surgery.

A recent small clinical trial examining G-CSF as adjunctive therapy to PCI for an acute anterior MI, in which subcutaneous G-CSF was injected daily for 5 days commencing after myocardial reperfusion, but within 24 h, failed to show any clinical benefit [182], with other clinical studies reporting high restenosis rates [183] and serious side effects [184] with this treatment strategy.

Activators of the RISK pathway that have the potential to demonstrate clinical cardioprotection

Erythropoietin (EPO) which is already in clinical use for raising haematocrit in anemic patients with chronic renal and cardiac failure shows great promise as both a neuroprotective and cardioprotective agent. Ehreneich et al. [167] demonstrated that an intravenous infusion of 33,000 iu of recombinant EPO administered daily for 3 days, was safe and improved functional recovery and showed a trend for reducing cerebral infarct size in 20 patients presenting with an acute ischemic stroke. A preliminary study has demonstrated that the long-acting EPO analogue, darbopoeitin alfa is safe when administered as a single 300 μg intravenous bolus prior to primary PCI in patients presenting with an AMI [168]. Clinical studies examining the cardioprotective potential of EPO in AMI patients and in patients undergoing CABG surgery are now underway.

Whether hormone replacement therapy is cardioprotective in post-menopausal women is unclear. However, a small clinical study comprising both men and women has demonstrated that estrogen administration has the ability to reduce ST-segment shift and reduce chest pain in patients undergoing single vessel elective PCI [169], but whether this translates to any cardioprotective benefit in patients presenting with an AMI remains to be examined.

Clinical trials are underway examining the cardioprotective potential of pharmacological PKC-ε activators and pharmacological PKC-ε inhibitors in different clinical settings of acute myocardial ischemia-reperfusion injury.

Several small clinical studies have demonstrated that the pharmacological inhibition of rho-kinase (ROCK) using fasudil, to be beneficial as an anti-anginal agent [170], to be neuroprotective in acute ischemic stroke [171] and to improve endothelial function in heart failure patients [172]. Whether ROCK inhibitors administered at the time of myocardial reperfusion to AMI patients, are beneficial, remains to be determined.

Conclusions

The Reperfusion Injury Salvage Kinase (RISK) pathway, when originally described, referred to the protein kinases, Akt and Erk1/2, which when specifically activated at the time of myocardial reperfusion conferred powerful cardioprotection against lethal reperfusion injury. The list of pharmacological agents identified as mediating their cardioprotective effect at the time of myocardial reperfusion through the activation of the RISK pathway is ever expanding and now includes growth factors, G-protein coupled receptor ligands, and other non-receptor acting agents. The RISK pathway has evolved to encompass other pro-survival kinases such as PKC-ε, p70S6K, PKG and GSK-3β, and constitutes the core component of a common cardioprotective pathway recruited at the time of myocardial reperfusion by both ischemic preconditioning and postconditioning which converges on the inhibition of the mitochondrial permeability transition pore, as one of the potential cardioprotective mechanisms of the RISK pathway.

The concept of the RISK pathway as a target for intervening at the time of myocardial reperfusion in patients presenting with an AMI, along with the effective clinical application of ischemic postconditioning in AMI patients, has succeeded in re-igniting interest in the myocardial reperfusion phase as a target for cardioprotection. The RISK pathway also provides a pharmacological target for intervention in other patient groups experiencing acute myocardial ischemia-reperfusion injury such as patients undergoing CABG surgery, patients having cardiac transplant surgery, and those patients surviving a cardiac arrest. Several small clinical studies have demonstrated myocardial protection with pharmacological agents known to activate the RISK pathway. Large placebo-controlled randomized clinical studies are now required to determine whether the administration of pharmacological agents, known to activate the RISK pathway, as adjunctive therapy to myocardial reperfusion confers any cardioprotective benefit, in terms of meaningful clinical out comes, for patients presenting with an AMI. Clearly, the valuable information obtained from the wealth of pre-clinical data that exist, should form the centrepiece in the design of such clinical trials.