Mitochondrial Pathways,Permeability Transition Pore,and Redox Signaling in Cardioprotection: Therapeutic Implications

A. Acute myocardial infarction and reperfusion injury

Acute myocardial infarction (AMI) continues to be a major cause of morbidity and mortality, and infarct size is the major determinant of patients' prognosis. Millions of people worldwide each year die from AMI. With the in-hospital mortality rate of 5%–6% among ST segment elevation myocardial infarction (STEMI) patients and about 4% among non-ST segment elevation myocardial infarction (NSTEMI) patients, AMI remains the first cause of death in Western countries and is the leading cause of chronic heart failure and cardiovascular diseases (302). All this happens despite the incredible improvements in the care of patients with AMI in recent years. In fact, prompt myocardial ischemic reperfusion is a very good strategy to reduce the infarct size and reduce all manifestations of postischemic injury resulting in improved outcomes. The culprit coronary artery can be opened by percutaneous coronary intervention (PCI) or fibrinolytic agents, with or without stenting. Although early reperfusion is the only way to save an ischemic organ, reversible and irreversible organ damage occurs during the early moments of reperfusion (that is reperfusion injury).

Myocardial ischemia is characterized by severe hypoxia, acidosis, energy depletion, and ion homeostasis alterations leading to cardiac dysfunction and, ultimately, to cell death. Mitochondria are abundant in cardiomyocytes, and they use oxygen as the principal substrate, so that it is not surprising that they play a role of protagonist in the I/R scenario during hypoxia/ischemia/reoxygenation. Mitochondria also play a pivotal role in ion homeostasis. In fact, the key events that occur in cardiomyocytes during I/R are imbalanced and altered exchange of ions. These are precipitated by the metabolic and chemical changes of ischemia, which then triggers mitochondrial dysfunction during reperfusion. Therefore, mitochondria play a critical role in the regulation of cardiac function in both health and disease, and are also increasingly recognized as end effectors for various cardioprotective signaling pathways. Therefore, understanding the role of mitochondria in this scenario is a requisite to find an appropriate therapeutic approach.

Myocardial injury that follows ischemia and reperfusion is, of course, related to the duration of ischemia. Successful reperfusion has been shown to dramatically reduce the infarct size; the sooner it is performed, the greater the amount of saved myocardium. However, it has also been seen that a large component of the damage takes place during the first minutes of reperfusion when an amplification of ischemic injury or additional damage occurs (12, 115, 116, 256, 396). Intracellular Ca²⁺ overload, inadequate resynthesis of adenosine triphosphate (ATP), oxidative stress by ROS, and loss of membrane phospholipids have been suggested as contributing to reperfusion injury (21, 285, 383). Therefore, both ischemia and reperfusion contribute to organ damage. Reperfusion injury includes arrhythmias, transient mechanical dysfunction of the heart or myocardial stunning, microvascular damage, and no reflow, as well as inflammatory responses. In reperfusion, cell death can occur by apoptosis, necrosis, and autophagy [the reader is redirected to extensive review on this topic (e.g., 115, 260, 269, 270, 290, 347, and 369)]. However, in contrast to necrosis and apoptosis, which inevitably lead to cell death, autophagy is not simply a destructive phenomenon, but under certain conditions, autophagy can be considered a protective mechanism against I/R injury (290, 347). The vulnerability to I/R injury is likely to be heavily influenced by the autophagic control of protein and organelle quality, including mitochondria. For these reasons, autophagy is briefly discussed in a dedicated section of this review (see section IX).

A relevant role in exacerbating myocardial injury by favoring mitochondrial permeability transition pore (mPTP) opening is played by ROS, which are generated at various sites in the cell and within mitochondria. The control of mPTP opening and ROS generation appears to modulate the I/R injury. To control mitochondrial function, a plethora of reactions occurs outside and within the mitochondria themselves. In Figure 1 are reported some of the elements in play, which will be responsible for cell death via mPTP opening: Figure 1A displays the balance of factors controlling mPTP opening in ischemia; Figure 1B illustrates the conditions favoring the mPTP opening and the main consequences upon reperfusion.

OPEN IN VIEWER

FIG. 1.

Schematic representation of factors controlling mitochondrial permeability transition pore (mPTP) open probability, during ischemia (A) and during reperfusion (B). Ischemia-induced intracellular acidosis may shift the equilibrium toward the closed state. Upon reperfusion, the recovery of pH toward neutral values facilitates the concomitant elevation of matrix [Ca²⁺] and reactive oxygen and nitrogen species (ROS/RNS) formation in promoting mPTP opening. For other acronyms, see the list of Abbreviations Used.

B. Strategies to reduce injury

Since recent data indicate that different forms of cell death are probably correlated (115, 116), the best strategy for developing cardioprotective agents is not to define the mode of cell death and its proportion occurring during I/R, but to identify mediators active in all forms of cell death. In this context, mitochondria have emerged as the relevant targets. Interestingly, the change in the mitochondrial membrane permeability occurring in early reperfusion appears to be among the most important regulators of all forms of cell death.

The timing of the reperfusion injury occurrence explains the efficacy of the PostC interventions in limiting the infarct size and in ameliorating the mechanical recovery of the heart (174, 329). The involved factors and the consequences of PostC are discussed in the following sections. Timing of cardioprotective interventions is of paramount importance for limiting mPTP opening. In fact, both ischemic PreC and ischemic PostC have been demonstrated to significantly attenuate I/R injury. Although PreC can be obtained with one or more brief coronary occlusions (a few min each) before the infarcting ischemia, PostC may be performed with one or more brief occlusions (a few seconds each), starting immediately after the end of the ischemia in animals and humans (e.g., 61, 133 –136, 174, 260, 329, and 333). Moreover, PreC initiates an immediate protective response (early PreC) and 12–24 h later, a more modest protection against the infarct size (SWOP), which will be considered briefly in the section VIII (Fig. 2).

OPEN IN VIEWER

FIG. 2.

Schematic diagrams illustrating the protocols of preconditioning (PreC) and postconditioning (PostC). PreC is triggered by short periods of ischemia (a few minutes) applied before the onset of a sustained period of ischemia. PreC consists of two chronologically and pathophysiologically distinct phases: an early phase (Early PreC), which develops very quickly, within a few minutes from the exposure to the stimulus, but lasts only 1–2 h [this is the phenomenon originally described by Murry et al. (239)], and a late phase (Late PreC), that is the so-called second window of protection (SWOP), which develops more slowly (requiring 6–12 h), but lasts much longer (3–4 days) (35, 37, 342, 348). PostC is triggered by short periods of ischemia (a few seconds) applied immediately after the end of the sustained ischemia [this is the phenomenon originally described by Zhao et al. (393)]. PostC maneuvers are not protective if they are carried out after the first minutes of reperfusion (260, 392). For other acronyms, see the list of Abbreviations Used.

The necessity of the early intervention for PostC to be protective has emphasized the importance of reperfusion in inducing I/R injury. Since the PostC maneuvers are not protective if they are carried out after the first minutes of reperfusion (260, 392), this period seems to be confirmed as the interval when most damage takes place. The main advantage of PostC with regard to PreC consists in the more frequent possibility of clinical application. In fact, due to the unpredictability of an ischemic event, PostC can be utilized with reperfusion procedures, which may be under the control of physicians (212, 333).

Cardioprotection may be due to passive mechanisms, which are responsible for a reduction in the reperfused heart of (i) endothelial cell activation and dysfunction; (ii) neutrophil activation and accumulation; (iii) tissue ROS generation; and (iv) microvascular injury and tissue edema (269, 368, 369). However, cardioprotection by PreC and PostC triggers also active mechanisms, which consist of the activation of intracellular signaling pathways that lead to limitation of cell death (268, 269, 369, 393). These signaling pathways initiate before sustained ischemia for ischemic PreC, but importantly, they initiate also at the very start of reperfusion for both PreC and PostC. In fact, both ischemic PreC and PostC reduce myocardial damages due to ischemia and reperfusion.

While in ischemic PreC, the brief cycles of protective I/R are employed before the sustained infarcting ischemia; in ischemic PostC, they are employed at the onset of reperfusion after sustained ischemia; however, both ischemic PreC and PostC are protective phenomena both targeting reperfusion (Fig. 3); as such, they share certain signaling elements in experimental analyses (137, 144, 145, 269, 382). At reperfusion, there are three populations of cells: (i) those that were killed by sustained ischemia, (ii) those that had sublethal injury and will survive, and (iii) those that are alive, but will die mainly from mPTP opening. Both PreC and PostC target this last group of cells. In fact, it is now thought that the actual PreC protection occurs in the reperfusion, rather than the ischemic phase, and repopulation of membrane receptors and activation of multiple kinases are critical events (61, 64). In PreC, we recognize a trigger phase in which adenosine, bradykinin, opioids, and possibly, other surface receptors couple through multiple pathways to activate mK_ATP and PKC. In the ischemic phase, PKC may act as a memory, and in reperfusion phase, signal transduction pathways act to prevent mPTP opening (Fig. 3A). Also, PostC protects because it maintains acidosis during early reoxygenation (167), which inhibits mPTP formation and allows ROS signaling (Fig. 3B) that gives the heart enough time to activate signaling pathways, and thus to precondition itself against reperfusion injury.

OPEN IN VIEWER

FIG. 3.

Current theory for PreC (A) and for PostC (B). In the PreC, we recognize a Trigger phase: adenosine (Ade), bradykinin (BK), and other surface receptors couple through multiple pathways to activate PKC and possibly other kinases; ischemic phase (sustained ischemia): kinases act as a memory; reperfusion phase (early reperfusion): signal transduction pathways act to prevent mPTP opening. In the PostC, intermittent PostC ischemias do not allow a rapid recovery of pH in early reperfusion, thus keeping mPTP closed and allowing the heart to precondition itself. It has been suggested that low pH in early reperfusion plays a role also in PreC (139). For other acronyms, see the list of Abbreviations Used.

The signaling elements recruited by these cardioprotective phenomena, especially those of the PreC trigger phase, have been extensively studied. It has become clear that different pathways finally target the mitochondria, and that preservation of mitochondrial structure and function is central for cardioprotection [for reviews see (27, 81, 260)]. In other words, the mitochondrial structure and function are preserved by the cardioprotective maneuvers, because they are able to affect the cross-talk between cytosolic and mitochondrial elements. Besides the survival kinases of the RISK (Akt/ERK1/2/GSK-3β) and protein kinase G (PKG)/PKC pathways, also SAFE (TNF-α/JAK/STAT-3) pathways, as well as the AMP-activated protein kinase (AMPK), Pim-1, and B-cell lymphoma-2 (Bcl-2) family may contribute to preserve the mitochondrial function and to avoid cell death.

Actually, the majority of the mitochondrial proteins is synthesized in the cytosol and has to be imported (52, 100). This import is influenced by protective maneuvers. Also, mitochondrial factors/products (e.g., ATP, cytochrome c [Cyt c], and ROS, mainly hydrogen peroxide [H₂O₂]) may be released in to the cytosol, affecting cell life and death. Both the import toward and the release from mitochondria are influenced by stressors like I/R and by cardioprotective maneuvers like PreC and PostC (e.g., 27, 137, and 139). Several of the aforementioned signaling and mitochondrial components take part to this cross-talk between the cytosol and mitochondria, constituting a mitochondrial network. Here, we will consider some of these elements, which are modified by I/R and cardioprotective maneuvers.

A. Important kinases of the mitochondrial network involved in I/R injury and cardioprotection

The heart should constantly adjust energy production to energy supply and utilization, and is a high-energy consumer. For this reason, the heart greatly depends on oxidative metabolism for adequate energy production and on efficient energy-transfer systems. Therefore, the function of mitochondria is finely regulated by intrinsic (e.g., mPTPs and mK_ATP channels) and extrinsic factors (e.g., GSK-3β and PKC). Before taking into account the role of mitochondria in the scenario of I/R, it is useful to consider how some of these factors are involved in the regulation of mitochondrial functions in physiological conditions and during stress.

2. PKC family in the I/R scenario and cardioprotection

The serine/threonine PKC family was first identified as intracellular receptors for the tumor-promoting agents phorbol esters. There are multiple isoforms of PKC that function in a wide variety of biological systems. The conventional PKC isoforms (PKC-α, β1, β2, and γ) are activated by phosphatidylserine (PS), calcium, and diacylglycerol (DAG), or phorbol esters such as phorbol 12-myristate 13-acetate (PMA), whereas novel PKCs (PKC-δ, ɛ, θ, and η) are activated by PS, DAG, or PMA, but not by calcium. The atypical PKCs (PKC-ζ and ι/λ) are not activated by calcium, DAG, or PMA; PKCλ is for mouse; the human homolog is termed PKCι.

Cardioprotective mechanisms involve the activation of several membrane receptors of agonists such as adenosine, bradykinin, and opioids. All these agonists eventually target PKC: adenosine through the activation of phospholipase C, and the other agonists through a more complex pathway that includes serial activation of PI3K/Akt, nitric oxide synthase (NOS), guanylyl cyclase, PKG, and opening of mK_ATP channels, and finally production and release of ROS, which target PKC within and outside mitochondria (66, 67, 113, 296). Mitochondrial ROS are represented by superoxide anion (O₂ ^−•) from which can derive its products H₂O₂ and hydroxyl radical (OH^•). Actually, H₂O₂ might be the only oxidant released from mitochondria to the cytosol in physiological conditions; the negatively charged O₂ ^−• does not permeate the lipid bilayer of biological membranes. However, it is able to pass through the pore of anion channels (Fig. 5), and in some cases, it may play a role as such or after a reaction with nitric oxide (NO^•) to form peroxynitrite (ONOO⁻) (8, 9, 79, 81, 205, 399).

OPEN IN VIEWER

FIG. 5.

ROS/RNS are products of multiple enzymes and reactions within and outside the mitochondria. The immediate product of multiple enzymes, including the enzymes of mitochondrial oxidative phosphorylation, may be the superoxide anion (O₂ ^−•), mainly generated at complexes I and III of the electron transport chain (ETC); however, due to spontaneous and enzymatic formation by p66Shc and dismutation by SOD, hydrogen peroxide (H₂O₂) is also rapidly generated. While H₂O₂ easily cross the biological membranes, the negatively charged O₂ ^−• does not permeate the lipid bilayer of membranes. However, O₂ ^−• is able to pass through the pore of anion channels (ACs), such as voltage-dependent mitochondrial AC. H₂O₂ can be transformed to the more dangerous OH^•. Superoxide anion can react with NO^• to generate ONOO⁻ and other RNS. Either redox stress or redox signaling can occur on both sides of membrane. However, it must be borne in mind that the switch from redox signaling to redox stress (within and outside mitochondria) does not depend only by the type of ROS/RNS but also on the amount of ROS/RNS and the vicinity of targets and antioxidants. Several isoforms of SOD degrade O₂ ^−• to H₂O₂, including manganese (MnSOD) in the matrix and copper/zinc SOD (CuZnSOD) in the intermembrane space and cytosol. In the matrix, H₂O₂ is further detoxified to water primarily by glutathione peroxidase (GPX) and by catalase (CAT). The thickness of dotted lines represents the ability to diffuse through the membrane. For other acronyms, see the list of Abbreviations Used.

The PKC isoforms mainly studied in I/R and cardioprotection are PKCα, PKCδ, and PKCɛ. After I/R, PKCδ enhances and PKCɛ reduces irreversible myocardial injury in rodents. PKCα is the isoform involved in I/R injury and protection from it in pigs (328).

Activation of PKCδ induces cell death (apoptosis and oncosis) through the regulation of mitochondrial function. It has been suggested that increased levels and catalytic activity of PKCδ in the mitochondrial fraction are associated with hyperphosphorylation of PDH (58). This and other mechanisms may be involved in PKCδ-induced cell death, whereas its degradation has been involved in protective mechanisms (55, 57).

Activation of PKCɛ protects mitochondrial function and diminishes cell death. Upon a PreC stimulus, PKCɛ translocates from the cytosol to the particulate fraction, but also to the mitochondria (251). Also, the cardioprotection by ischemic PostC was lost in isolated rat hearts treated with aspecific and specific PKCɛ inhibitors (276, 388).

Importantly, as said, PKCɛ may be activated by ROS signaling (260, 288, 341). Within mitochondria, PKCɛ is located in the intermembrane space, and it is associated with the IMM, but the presence of a second PKCɛ pool within the matrix has been suggested (66). PKCɛ is imported into mitochondria by heat shock protein 90 (Hsp90) and outer membrane translocase 20 (Tom20), playing a role in cardioprotection from I/R injury (42). Within mitochondria, PKCɛ has several targets, including mK_ATP channels and aldehyde dehydrogenase (ALDH). The latter contributes to cardioprotection by inhibition of toxic aldehyde formation (56). It has been reported that PKCɛ phosphorylates VDAC when HKs (I and II) are overexpressed. In fact, HKs play a pivotal role in promoting cell survival by binding specifically to the VDAC, whereas HK removal triggers cell death (50). PKCɛ also targets mitogen-activated protein kinases (MAPKs) and ERK, forming a module that induces the phosphorylation and inactivation of the pro-apoptotic protein Bad (9).

In summary, PKCs are important for the I/R and for the cardioprotection by ischemic PreC and PostC. In particular, mitochondrial PKCɛ contributes to cardioprotection by regulating the activity of several mitochondrial proteins (e.g., VDAC, connexin-43 [Cx43], mK_ATP channels, and ALDH). On the contrary, PKCδ may play a pivotal role in I/R injury.

6. AMP-activated protein kinase, a metabolic regulator in health and disease

AMPK phosphorylates multiple targets, including several biosynthetic enzymes and cardioprotective elements, such as acetyl-CoA carboxylase, hydroxymethylglutaryl-CoA reductase, glycogen synthase, and endothelial nitric oxide synthase (eNOS) (53). In Figures 6 and 7 are reported some of the multiple targets of activated AMPK.

OPEN IN VIEWER

FIG. 6.

AMP-activated protein kinase (AMPK) activation. AMPK is formed by α-, β-, and γ-subunits. An increased AMP-to-ATP ratio leads to a conformational change in the γ-subunit, leading to increased phosphorylation and decreased dephosphorylation of AMPK. This activation facilitates the activity of PKC of which AMPK is a substrate. Additional regulation is affected by Ca²⁺–calmodulin-dependent kinase kinase β (CaMKKβ), which phosphorylates and activates AMPK in response to increased Ca²⁺. Activated/phosphorylated AMPK enhances free fatty acid (FFA) oxidation, improves NO^• availability, and may limit ROS production via several different pathways (see also Fig. 7). Once activated, AMPK can download the NFκB-signaling system. For other acronyms, see the list of Abbreviations Used.

OPEN IN VIEWER

FIG. 7.

AMPK signaling in controlling the metabolic state. AMPK regulates protein synthesis, glycolysis, fatty acid oxidation, and lipolysis. AMPK activity is influenced by exercise and hypoxia, likely with the intervention of hormones and metabolites. Here are reported some of the metabolic targets and signaling pathways of AMPK. Activated AMPK has a plethora of targets; for instance, AMPK is a central regulator for cell growth and metabolism via multiple targets, including mTOR (the target of rapamycin), which resides in the two functionally distinct complexes TORC1 and TORC2 (defined by their adaptors Raptor and Rictor, respectively). The two TORCs are affected with unknown modalities. AMPK is also implicated in suppression of initiation and progression of cancers in various tissues via the regulation of the inhibitor of growth (ING) family, an evolutionarily conserved set of proteins. ROS may be activators of AMPK, and activated AMPK may limit ROS production (see Fig. 6 and text). In bold are reported some targets directly involved in cardioprotection (see text). Therefore, AMPK may play a pivotal role in protection and deserves more attention by researchers. PFK2, phosphofructokinase 2; P70 SGK, glucocorticoid-inducible kinase. For other acronyms, see the list of Abbreviations Used.

Usually AMPK is activated when the AMP concentration increases as a result of insufficient ATP production or unmatched energy demand. Therefore, it plays a role of paramount importance in the regulation of mitochondrial function during and after ischemia when the ATP/ADP/AMP undergoes abrupt changes. Although, its main role is to monitor the mitochondrial function and cellular energy status responding to changes in the AMP/ATP ratio, AMPK can also be activated independently of adenine nucleotides, by changes in calcium concentrations as well as by increased production of ROS (16).

AMPK is a heterotrimeric complex that contains a catalytic (alpha) and two regulatory (beta and gamma) subunits. Each subunit has multiple isoforms (α1 and 2, β1 and 2, and γ1, 2, and 3). Therefore, 12 possible combinations of AMPK holoenzyme are described in murine and human hearts. The catalytic alpha subunit includes both the protein kinase domain and a threonine residue (Thr172) whose phosphorylation by kinases is responsible for AMPK activation. Figure 6 shows the main factors controlling the shift from inactive to active AMPK. It seems that in human hearts, both AMPK α1 and 2 catalytic subunits contribute to the total AMPK activity. On the contrary, in mouse hearts, α2 subunit accounts for about 80% of total AMPK activity (16). The β subunit acts as a scaffold for the other two subunits. It also contains a glycogen-binding domain whose physiological role might be to control glycogen metabolism. AMPK is present in the nucleus and the cytoplasm; however, the mechanisms that regulate the intracellular localization of AMPK are unknown.

It seems that environmental stresses regulate the intracellular localization of AMPK, and upon recovery from heat shock or oxidant exposure, AMPK accumulates in the nuclei. Treatment with an AMPK activator increases the expression of peroxisome proliferator-activated receptor γ coactivator (PGC)-1α and manganese superoxide dismutase (MnSOD) mRNAs, which may inhibit ROS production in mitochondria (204). Since AMPK serves as an energy sensor, it is at the center of control for a large number of metabolic reactions, thereby playing a crucial role in metabolic syndrome (MS), Type-2 diabetes, and other human diseases.

The activation of AMPK induces acceleration of mitochondrial biogenesis in physiological conditions. Conditions leading to changes in AMP concentrations and AMPK activation are directly related to changes in ATP concentrations. Accordingly, ischemia, similarly to mitochondrial respiration inhibitors, activates AMPK within a few minutes (131). Increased ATP demand also leads to AMPK activation, especially when combined with decreased ATP supply, as is the case in contracting muscle during hypoxia. Also, intense exercise, norepinephrine, phenylephrine, isoproterenol, or vasopressin activates heart AMPK (183). In Figure 7 are reported some of the pathways able to activate AMPK and some of the AMPK metabolic targets.

When activated, AMPK aims to restore the cellular energy charge by switching off anabolic ATP-consuming pathways. Although the mechanism remains to be elucidated, AMPK could inhibit (i) cJUN kinase activation; (ii) endoplasmic reticulum stress; and (iii) oxidative stress in several cellular models, including cardiomyocytes (86). Activated AMPK may also inhibit glucose-induced oxidative stress and NADPH oxidase activation in endothelial cells (51). On the other hand, AMPK may switch on catabolic ATP-producing pathways. For these reasons, it may play a role in cell protection. Recently, it has been demonstrated that the AMPK-dependent mitochondrial protection of resveratrol against oxidative stress may be associated with the downstream inhibitory phosphorylation of GSK-3β (322). Although the mechanism of resveratrol's cytoprotection involves AMPK activation, resveratrol does not directly activate AMPK in vitro (15). It is likely that redox conditions play a role in this mechanism (see section II.C).

B. Important components of mitochondria in the network involved in I/R and cardioprotection

Kinases may be imported into mitochondria and/or may affect directly and indirectly the function of mitochondrial components (e.g., mPTPs and mK_ATP channels). Here, we consider more details of these components. However, a challenge is that many mitochondrial channels and transporters have yet to be identified at the molecular level.

1. Mitochondrial permeability transition pore

This pore is a high-conductance megachannel, which, as said, plays a role of paramount importance in the I/R scenario. The modulation of mPTP is also important physiologically in regulating the calcium homeostasis. The pore modulation and opening will be considered with more detail below (sections III.A, III.B, and III.C); here, we consider the putative components of mPTPs and their interaction. The pore was first described in 1976 by Hunter et al. (167) and is located between the inner and OMMs. When an mPTP is formed, it opens and allows communication between the cytoplasm and the mitochondrial matrix. The molecular identity of the proteins that form this pore is still unknown. It has been suggested that the mPTP is formed by the VDAC in the OMM, the ANT in the IMM, and Cyp-D in the matrix of mitochondria. When the mPTP is in the closed state, the matrix protein Cyp-D is detached from IMM, whereas HK II is attached to OMM components of the pore (Fig. 8A). Opening of mPTP (Fig. 8B) appears to be facilitated by the binding of Cyp-D to the IMM in a process regulated by both Ca²⁺ and inorganic phosphate (Pi) (78). However, experiments with transgenic mice in each of the components of the putative mPTP achieved controversial results. In fact, neither the deletion of the gene nor knockdown of VDAC or ANT prevents mPTP opening in response to Ca²⁺ overload of mitochondria (7, 8, 197, 243). In fact, experiments with VDAC and ANT KO mice demonstrated substantial permeability transition during Ca²⁺ stimulation, indicating that VDAC and ANT are not essential for the permeability transition (7, 8, 197). However, these experiments are not conclusive, and the dispute is not resolved. While in fibroblasts and liver mitochondria of KO mice Cyp-D Ca²⁺-induced mPTP opening is overdue (7, 12), the sensitivity of mPTP opening in response to adenine nucleotides or oxidative stress is similar in Cyp-D-deficient and wild-type mitochondria, indicating that Cyp-D is important for mPTP opening, but that the permeability transition can occur even in the absence of Cyp-D (12). The possibility that Cyp-D has a role in the regulation of apoptotic proteins in a manner that is independent of the mPTP has been suggested (98).

OPEN IN VIEWER

FIG. 8.

mPTP composition and regulation. In Figure 1 are reported the factors controlling the mPTP-opening probability. Here are evidenced the putative elements that form the pore. The mPTP is believed to be composed of the ANT in the IMM, the VDAC in the OMM, and Cyp-D in the matrix. (A) In physiological conditions, Cyp-D is detached from IMM, and HK is attached to the OMM. Notably during ischemia, mPTP opening hardly occurs because mPTP is strongly inhibited by acidosis. (B) mPTP typically opens in reperfusion when Ca²⁺ overload, generation of ROS, and pH normalization occur. Ca²⁺ overload may stimulate the interaction of Cyp-D with other mPTP components, which triggers permeability transition. A part Ca²⁺ overload, also inorganic phosphate (Pi), contributes to mPTP opening through the binding of Cyp-D to the inner membrane. Also, HK detachment from mitochondria triggers apoptosis through mPTP opening. In fact, pore opening leads to cell death through the release of proapoptotic factors and via RIRR (ROS-induced ROS release). Other important factors that regulate pore formation are Bax/Bad/Bcl-2 and GSK-3β. Cytochrome c and other mediators are simultaneously released that may play a role in committing the cell to death. These mediators include Smac/Diablo and AIF. For further explanations, see the text and Figure 9. For other acronyms, see the list of Abbreviations Used.

2. Putative mitochondrial ATP-sensitive potassium channels

Abundant in myocardial sarcolemma, where they were originally discovered (305), ATP-sensitive potassium (K_ATP) channels have also been reported to reside within intracellular membranes, including endoplasmic/sarcoplasmic reticulum, nuclei, secretory granules, and mitochondria (78). K_ATP channels are biosensors that enable high-fidelity readout of metabolic distress signals (115, 347, 369). It is likely that these channels behave like checkpoints that perform a rheostat-like operation adjusting membrane potential-dependent functions to match energetic demands of the working heart (124, 128, 236). K_ATP channel complexes are formed by KCNJ11-encoded Kir6.2 pore subunits coassembled with the regulatory ATP-binding cassette sulfonylurea receptor; channel deficit impairs tolerance to endurance challenge, hemodynamic load, or sympathetic discharge (70, 126, 200, 207, 285, 286). Both sarcolemmal and putative mK_ATP channels comprise the protective benefit of ischemic PreC and PostC, (7, 27, 82, 146, 197), whereas disruption of the K_ATP channel blunts this protective response (197, 243). Because of the central role of mitochondria in cardioprotection, much attention has been devoted to mK_ATP channels.

Mitochondrial ATP-sensitive K⁺ channels are located in the IMM, are considered targets of protective signaling pathways, and play a pivotal role in ROS production, mainly O₂ ^−• derived from complex I of the ETC (27, 117, 274, 310, 383). Opening of the mK_ATP channels and subsequent generation of ROS are considered to be a pivotal step in the mechanisms of PreC and PostC (27, 270, 382). We evidenced that ROS signaling is downstream of mK_ATP channel opening in isolated rat hearts subjected to I/R with an intermittent infusion of mK_ATP channel opener, diazoxide, or diazoxide plus the ROS scavenger mercaptopropionylglycine (MPG) at the onset of reperfusion. In fact, MPG attenuated diazoxide-induced protection (271).

In the context of cardioprotection, it has been reported that Akt-dependent phosphorylations and subsequent PKG activation lead to mK_ATP channel activation. Actually, Akt and PKG are a part of the so-called signalosomes (113), which are vesicular, multimolecular signaling complexes, involving the assembly and regulation of multiproteins, which compartmentalize and convey the intracellular signal (296). Whether or not these proteins form a signalosome, they also include eNOS, guanylyl cyclase, and cyclic guanosine monophosphate (cGMP), which activates PKG. Indeed, it has been proposed that activated PKG phosphorylates an unknown protein of the OMM (113). This causes the signal to be transmitted to PKCɛ already bound to the IMM, which in turn causes the mK_ATP channel opening. (66 –68, 113).

For the mK_ATP channel opening, the protective signal is transmitted from the OMM to the IMM. This may be accomplished by a series of intermembrane-signaling steps that includes PKCɛ activation, so that stimulation of mitochondrial K⁺ influx via the mK_ATP channels is cardioprotective, and activation of the mK_ATP channel is mediated by mitochondrial PKCɛ (Fig. 9). The resulting ROS then activate a second PKC pool, which through another signal transduction pathway causes inhibition of mPTPs (one of the end effectors) and reduction in cell death. In summary, the inhibition of mPTP opening occurs through intermediary steps involving PKCɛ, mK_ATP channels, K⁺ entry into mitochondria, matrix alkalization, and generation of ROS with a protective signaling role (Figs. 3 and 9) (66 –68, 72, 113, 253, 270, 382). Pharmacological opening of mK_ATP channels by diazoxide contributes to the formation of small amounts of ROS inducing cardioprotection (74, 109, 129, 142, 143). Also NO^• donors can activate mK_ATP channels in rabbit ventricular myocytes and can potentiate the protective effect of the mK_ATP channel opener diazoxide (310). Besides cGMP/PKG-dependent phosphorylation, mK_ATP channels can be opened by direct reaction of NO^• and derivatives and by the action of H₂S, leading to S-nitrosylation (SNO) and S-sulfhydration of proteins, respectively [for review see (260)].

OPEN IN VIEWER

FIG. 9.

Signaling pathways in cardioprotection: reperfusion injury salvage kinases (RISKs) and survivor-activating factor enhancement (SAFE). Activation of cell surface receptors (e.g., Ade, BK, erythropoietin [EPO], tumor necrosis factor [TNF], and interleukin-6 [IL6]) in response to a PreC or PostC stimulus may recruit both the RISK and the SAFE pathways. Ligands of membrane receptors initiate cardioprotection by activating a cascade of kinases, leading to mPTP inhibition, and thus protecting the cell against apoptotic and necrotic death. These two pathways are activated at the time of reperfusion and will cross-talk. See also the text. For other acronyms, see the list of Abbreviations Used.

It has been suggested that mK_ATP channels are activated by ROS: O₂ ^−• and H₂O₂, but not other peroxides. Yet, certain RNS can activate mK_ATP via a mechanism involving mitochondrial complex II. Moreover, mK_ATP channels are inhibited by NADPH. Thus, to sum up, mK_ATP channels are activated by S-nitrosothiols, nitroxyl, and nitrolinoleate (102, 110, 292, 389). The latter two species also inhibit mitochondrial complex II. Nitroxyl protects cardiomyocytes against IR injury in an mK_ATP-dependent manner. Overall, these results suggest that the mK_ATP channels are activated by ROS and RNS and inhibited by NADPH. The redox modulation of mK_ATP channels may be an underlying mechanism for its regulation in the context of PreC and PostC (295).

However, controversy exists on the nature, existence, and opening of mK_ATP channels, which may also be a deleterious process (72, 74). In fact, for instance, mK_ATP channels are suggested to be involved in angiotensin II-induced redox stress via the depolarization of mitochondrial membrane potential and consequent respiratory dysfunction (182, 216). Moreover, the mK_ATP channel opener diazoxide caused a nitrate tolerance-like phenomenon, whereas the K_ATP channel inhibitor glibenclamide improved tolerance in nitroglycerin-treated animals (71). Finally, diazoxide and/or 5-HD, channel agonist and antagonist, respectively, may also have direct effects on mitochondrial ETC and mitochondrial bioenergetics (129, 130). Thus, these drugs lack specificity for the channel, and cautions are needed in this respect.

Hence, it seems that on the one hand, mK_ATP channel opening induced by PKC activation stimulates ROS production, whereas on the other hand are ROS generated in the cytosol that stimulate mK_ATP channel opening. A simple model in which the first mechanism may be protective while the second would be detrimental could be suggested. However, PKC activation leading to the opening of mK_ATP channels has been challenged by the Halestrap group: they demonstrated that PreC inhibits opening of the mPTPs in situ by an indirect mechanism probably involving decreased ROS production and Ca²⁺ overload at reperfusion (74, 128). Therefore, based on these observations, one can speculate that mK_ATP channel activation by PKC is not required to induce protection. Nevertheless, several studies have shown that redox-dependent opening of the channel is a protective mechanism [for reviews see (66, 67, 295)].

Clearly, further work in this area, including the molecular identification of the mK_ATP channel itself and the redox-sensitive residues within it, will facilitate a better understanding of the role that regulation of the channel plays in events such as cardioprotection.

4. Mitochondrial uncoupling proteins

Other mitochondrial components associated with ROS production are the uncoupling proteins (UCPs). Under physiological conditions, mitochondrial oxygen consumption (VO₂) is coupled to ATP synthesis. Reducing equivalents, resulting from energy substrate oxidation, deliver electrons to the mitochondrial ETC. The energy resulting from electron transfer to oxygen atom is used to generate an electrochemical gradient by pumping protons from the mitochondrial matrix into the intermembrane space. Under physiological conditions, the protons re-enter the matrix via F₀F₁-ATPase, which uses the energy to regenerate ATP from ADP (Fig. 10). In addition, a small proportion of H⁺ can bypass F₀F₁-ATPase, so that VO₂ is not strictly coupled to ATP synthesis. In the 1970s, H⁺ traslocases were identified in the IMM of brown adipose tissue and named UCPs, which are present in the mitochondria of different tissues and have different homologs. There are four UCP variants believed to induce inward H⁺ leak in energized mitochondria. Their main role is to direct the mitochondria to produce heat rather than ATP, that is, there is an H⁺ influx into the mitochondrial matrix without phosphorylation of ADP. As such, UCPs are important metabolic regulators in permitting fat oxidation and in attenuating free-radical production. Since ROS production increases with increasing membrane potential, UCP-mediated uncoupling has been proposed to play a role in regulating mitochondrial ROS production and may represent a mechanism by which mitochondria protect themselves from oxidative damage (40, 317). Among ROS-derived lipid peroxidation products, 4-hydroxynonenal (HNE) is considered an important mediator of free-radical damage (101). It has been suggested that cytotoxic HNE is also able to act as a signaling agent, inducing mitochondrial uncoupling via the UCP1, UCP2, UCP3, and ANT (94). The mechanism of induction of proton leak by HNE is unclear. It has been suggested that HNE may induce covalent modification and conformational changes in the UCPs that allow the passage of protons back into the mitochondrial matrix (95). Both UCP2 and 3 are expressed in the heart, but their role is unclear. Although controversy exists on the cardioprotective role of uncoupling, mild uncoupling secondary to activation of UCPs has been described to confer cardioprotection under several conditions, including myocardial reperfusion, likely by decreasing ROS production (157). For instance, both ONOO⁻ and electrophilic lipids can activate cardioprotective mitochondrial mild uncoupling (41, 229) Moreover, conditions occurring during PreC, such as elevated NO^•, transient ROS generation, acidic Ph, and the activation of both mitochondrial phospholipase A2 and lipoxygenases (63, 237, 376), could favor nitroalkene generation from the polyunsaturated fatty acids in mitochondrial membranes. Nitroalkenes formed in mitochondria during ischemic PreC nitroalkylated ANT and UCP2, leading to mild uncoupling and cardioprotection against I/R injury (240). Yet mitochondrial uncoupling contributes to reduced ROS formation (306). Intriguingly, it has been suggested that transient complex I inhibition during reperfusion is cardioprotective via attenuated ROS production (324). Moreover, conditions leading to elevated NO^•, transient ROS generation, acidic pH, and thus to mitochondrial uncoupling may occur also in PostC (see below). Nevertheless, experimental evidence supporting the involvement of uncoupling in PostC protection has yet to be provided.

OPEN IN VIEWER

FIG. 10.

F₀F₁-ATPase and ETC function in ischemia and reperfusion with and without protection. F ₀ F₁-ATPase uses the energy of the proton–electrochemical gradient (ΔH⁺) across the IMM to regenerate ATP from ADP. During ischemia, it can work in a reverse mode and consumes ATP. During reperfusion, F₀F₁-ATPase and ETC restart to produce ATP and ROS in a variable quantity, whether the heart is protected or not. In nonprotected reperfusion when mPTP is opened, F ₀ F₁-ATPase can work in a reverse mode. The thickness of arrows indicates the prevalent fuel utilized (in bold) and the molecules produced (in bold-italic) by mitochondria. TCA, tricarboxylic acid cycle. For other acronyms, see the list of Abbreviations Used.

C. ROS/RNS: from mitochondria to activation of kinases of the network

ROS and RNS such as O₂ ^−•, H₂O₂, and ONOO⁻ are small and highly reactive molecules having both physiological (ROS/RNS signaling) and pathological effects (ROS/RNS stress) (Fig. 5). In several systems, various pathways, particularly those involving MAPKs (e.g., JNK, ERK, and p38 MAPK) and PKCs, are modulated by ROS and RNS. It is now clear that normal levels of cellular ROS and RNS play important roles in cell-signaling pathways and are vital for physiological functions. For instance, H₂O₂ of mitochondrial origin are of paramount importance in metabolic coronary vasodilatation (307).

ROS are generated in different cellular compartments and by several enzymes, including NADPH oxidases at the plasma membrane (211) and cytosolic xanthine oxidases (111). Although ROS are produced by several extracellular and intracellular processes, in cardiomyocytes, mitochondria represent the most relevant site for ROS formation (91, 238, 363 –365). It has been suggested that ROS-mediated signals (mainly by H₂O₂) arising within mitochondria can also generate a retrograde response that is conveyed to the nucleus, causing the upregulation of nuclear genes encoding mitochondrial proteins and leading to the induction of mitochondrial biogenesis (48). Also, NO^• is known to induce the production of O₂ ^−• and H₂O₂ by mitochondria, and may trigger redox signaling (46, 48, 240, 241, 294, 300) (see also below).

Within mitochondria, the largest amount of oxygen is reduced to water at respiratory complex IV. Indeed, even under physiological conditions, a minor fraction of oxygen (<0.1%) is transformed into O₂ ^−• at the level of complexes I and III; SOD then rapidly converts O₂ ^−• to H₂O₂ that is freely diffusible through membranes (110). The production of H₂O₂ can increase during myocardial challenging, such as during increased cardiac work load (307). The further reduction of H₂O₂ to OH^• in the presence of transition metals, such as Fe²⁺ or Cu²⁺, represents a dangerous step, because an increase in toxicity can occur. In fact, no enzyme is available for the removal of OH^• that can only be scavenged by antioxidants with the formation of less-dangerous radical species.

The mitochondrial formation of ROS might be modulated by NO^• (46, 294, 300), which reversibly inhibits cytochrome oxidase (20, 363). This inhibition can be transformed into irreversible alterations of respiratory chain when NO^• reacting with O₂ ^−• generates a great amount of ONOO⁻, an RNS that can produce irreversible nitration of proteins, including proteins from the oxidative phosphorylation system (18, 220, 362, 391). Under certain conditions, such as scarcity of substrate and/or cofactors, NOS can become uncoupled, resulting in the generation of O₂ ^−• and/or OH^•, instead of NO^• (18, 379). Recent evidence suggested neuronal NOS (nNOS) as a protein completely incorporated to the IMM in a physiological context. In particular, evidence on the presence of an nNOSα variant in heart mitochondria was provided by Kanai et al. (188), with the electrochemical determination of the Ca²⁺-induced NO^• release from a single mouse heart mitochondrion, a process that was absent in nNOS^−/− KO mice. Mitochondrial NOS uncoupling has also been postulated (106), and it could play a role in I/R injury, though this remains to be demonstrated.

Apart from the ETC, ROS can also be produced in the intermembrane space by the apoptosis promoter p66Shc, by monoamine oxidases (MAOs), and by AIF located in the OMM (Fig. 5). The p66Shc oxidizes reduced Cyt c, thus inducing the partial reduction of oxygen to peroxide. While under physiological conditions p66Shc is present in the cytosol, under stress conditions, it can be phosphorylated by PKCβ and translocated into the mitochondrial intermembrane space. The monoamine catabolism generates aldehydes, ammonia, and H₂O₂. In fact, MAOs transferring electrons from amine compounds to oxygen generate H₂O₂. AIF comprises NADH oxidase activity and can generate superoxide anion (82, 91, 114, 256).

Besides being a relevant site for ROS formation, mitochondrial function and structure are profoundly altered by oxidative stress (82), especially when mPTPs were undergoing prolonged opening. In fact, mPTPs play a central role in the so-called ROS-induced ROS release (RIRR) (396 –398); excessive ROS facilitate mPTP opening, which in turn favors ROS formation by inhibiting the respiratory chain because of the mPTP-induced loss of Cyt c and pyridine nucleotides (82, 127). This vicious cycle is likely to be established at the onset of a rapid reperfusion when a large increase in ROS formation occurs along with pH recovery and Ca²⁺ overload, thus inducing injury amplification, as mentioned above and discussed below (Fig. 1).

ROS are important regulators of PKC by reacting with thiol groups associated with the zinc-finger region of the molecule (199). RNS-dependent activation of PKC, possibly via a redox-sensible SNO process, has been also suggested, a process also occurring within mitochondria (260, 288, 341). A recent hypothesis considers that H₂O₂ and aminoimidazole carboxamide ribonucleotide (AICAR) treatments phosphorylate and activate AMPK (99). The H₂O₂-mediated activation of AMPK is likely mediated via the ROS-induced decrease in ATP levels. Endogenously produced ROS within skeletal muscle cells are important for the maintenance of PGC-1α mRNA expression. The PGC-1 family of regulated coactivators, consisting PGC-1α and PGC-1β as well as PGC-1α-related coactivator, may control mitochondrial respiratory function and biogenesis. A potential link between ROS and the induction of mitochondrial biogenesis has been established. In fact, the PGC-1α gene expression is regulated by ROS, and PGC-1α plays a central role in regulating the mitochondrial content within cells (312, 335). Also, kinases are involved in mitochondrial biogenesis, and the most important of these include calcium-/calmodulin-dependent protein kinase IV (378), AMPK (177), and p38 MAP kinase. At least two separate mechanisms are involved in increasing PGC-1α mRNA expression in response to AICAR. One involves AMPK activation, leading to increases in PGC-1α promoter activity and mRNA expression. The other is mediated by a reduction in cellular ROS levels, which leads to a reduction in the level of PGC-1α mRNA expression. Therefore, it is likely that the effect of AICAR on PGC-1α gene expression is a balance between the positive effects of AMPK activation on promoter activity and its negative effect of ROS, which may enhance mRNA decay (176).

Also, AMPK-mediated protection could be redox-sensitive: both H₂O₂ and increased production of radicals activate AMPK (99), possibly by oxidation of two cysteine residues in the alpha subunit of AMPK (395). For instance, it has been demonstrated that metformin, a drug used to treat type II diabetes, inhibits complex I of the respiratory chain to generate mitochondrial O₂ ^−•, and then ONOO⁻, which leads to AMPK activation via a c-Src- and PI3K-dependent pathway (400). Subtoxic concentrations of ONOO⁻ generated by metformin activate AMPK to precondition the cells, which subsequently reduces the excessive oxidant stress triggered by hyperglycemia, so that a cross-talk exists between redox conditions, activity of kinases, and factors promoting mitochondrial biogenesis, which, in turn, may regulate ROS and RNS production in acute and chronic conditions.

Therefore, the key message here is that the signal transduction pathways converge on mitochondria, and are enhanced by ROS/RNS. That is, they act as secondary messengers to control a variety of physiological responses (redox signaling), such as the regulation of vascular smooth muscle tone, the regulation of mitochondrial respiration, and the progression/enhancement of signaling pathways. The latter involves kinases, such as PKC, MAPK, and AMPK, as well as some transcription factors. Yet, sources of superoxide such as the mitochondrial ETC, in the RIRR mechanisms, and xanthine oxidase are not tightly regulated and may become increasingly relevant in the pathology (redox stress) of I/R injury.

A. mPTP opening in acute I/R

Of note, mPTP opening can be temporary, intermediate, or prolonged, depending on the complex balance between inducers and antagonists (282, 283). The transient opening of mPTP is able to generate reversible cellular changes, so that this temporary opening has been suggested to be involved in physiological processes and cardioprotection (185). In reality, the transient pore opening is involved in physiological events such as the reversible intracellular trafficking of NAD⁺ (80), the rapid discharge of excessive intramitochondrial Ca²⁺ (21, 171, 172), and the transient formation of ROS (374) (see also section III.C).

Prolonged opening of the mPTP represents a key event leading to cell death in the pathophysiology of I/R injury (337, 377). In fact, the opening of mPTP is central to the oxidative damage associated with reperfusion. Many studies have highlighted an important contribution of mPTP opening and cell death that correlated with the release of Cyt c after Bax and enhanced levels of ROS. When an mPTP is formed, mitochondria are uncoupled, and ATP production is blocked, and the matrix swells and can cause the OMM rupture (Figs. 1 and 8) [for reviews see (78, 81, 82)].

It is important to note that the probability of mPTP opening of mitochondria de-energized is drastically reduced below pH 7.4, a condition that occurs during prolonged ischemia (64, 62, 112, 274). Low pH also favors the activation of both mitochondrial NHE and NCE, and consequently reduces the uptake and extrusion of Ca²⁺ from the mitochondrial matrix (175). Still, the low pH can also activate UCPs. These would uncouple ATP synthesis from oxygen consumption (96). Whether UCPs play a deleterious or protective role in tolerance against ischemia is controversial (24, 157). Low pH can also inhibit glycolysis and the production of pyruvate, resulting in a lower production of acetyl-CoA for the tricarboxylic acid cycle. Finally, this will result in a slower feeding of the respiratory chain complexes. Therefore, low pH prevents mPTP opening, mainly in ischemia.

Upon reperfusion, very different conditions are established depending on whether or not Δψm rapidly recovers. In the case of energized respiring mitochondria, a low pH stimulates the uptake of Pi, raising its intramitochondrial content that acts as a potent inducer of mPTP opening. For example, in brain-energized mitochondria, low pH has been reported for mPTP opening (202). In contrast, in the presence of mitochondrial membrane depolarization, prolonged opening of mPTP takes place when a rapid normalization of tissue pH occurs. This last condition (i.e., membrane depolarization and rapid normalization of pH) is the most common scenario during abrupt reperfusion. The mPTP opening is facilitated by the presence of elevated levels of Ca²⁺, Pi, and ROS and by the presence of low levels of NO^•. In fact, when reperfusion occurs, metabolic and biochemical changes take place in the postischemic cardiomyocytes, including a rapid restoration of normal pH by membrane exchangers (especially NHE), generation of ROS (including RIRR), an intracellular Ca²⁺ overload, resulting in mitochondrial dysfunction and possibly cell death (78, 79, 286, 398). In addition to its direct action on mitochondria, the opening effect of Ca²⁺ overload is also due to indirect aspects such as swelling, band necrosis, and enzyme activation. For example, activation of phospholipase A2 (319), which facilitates the release of arachidonic acid from phospholipids, and activation of calpain proteinases are both Ca²⁺ dependent (325). These enzymes facilitate further mPTP opening (Fig. 8) (175).

B. Prevention of prolonged mPTP opening

All these opening factors are countered by physiological mPTP antagonists, such as adenine nucleotides (particularly ADP), high concentrations of protons (i.e., pH below 7.4), increased mitochondrial membrane potential, and magnesium ions, as well as by physiological levels of NO^• (78, 372). The pore quickly closed if Ca²⁺ is chelated (21, 78). The events promoting mPTPs are prevented by some drugs (see section X), including CsA, which is an mPTP desensitizer at nM concentrations (12). Paradoxically, it was suggested that under some conditions, Pi may be an inhibitor of mPTP. In the absence of Pi, desensitizing effects of CsA or Cyp-D ablation are no longer present (12).

In summary, the prolonged opening of mPTP may be one of the key events in transition from reversible to irreversible reperfusion injury, or cell death, during the first minutes of reperfusion. In fact, the opening of these channels results in a further production of ROS and mitochondrial uncoupling of oxidative phosphorylation and mitochondrial swelling, resulting in cellular apoptosis and necrosis of the myocardium. Despite a modest increase in the levels of Pi and mitochondrial de-energization, Ca²⁺ overload, and the formation of ROS during ischemia, the long-term opening of mPTP is antagonized by acidosis. On the contrary, and despite an initial recovery of Δψm, on reperfusion, mPTP opening is strongly facilitated by pH recovery, increased matrix Ca²⁺ concentration, and accumulation of ROS (Figs. 1 and 8). This underlines the importance of the mPTP inhibition by acidosis in cardioprotection.

Actually, cardioprotective interventions induce several cytosolic signaling events resulting in suppression of the mPTP opening, but it is still uncertain how these events are conveyed to mitochondria. It has been proposed that kinases (e.g., PKC, PKG, and GSK-3β) may play a direct pivotal role in the interaction with the mPTP after that these kinases are phosphorylated by the upstream protective signals, such as membrane receptor activation and Akt phosphorylation in the RISK pathway (e.g., 117, 137, 141, and 185) (see section IV).

C. Transient opening of mPTP can be protective

As mentioned above, transient opening of mPTP occurs in physiological conditions, and it can be protective (135, 308). Transient mPTP opening has been proposed to serve as a Ca²⁺-release mechanism by which mitochondria avoid matrix Ca²⁺ overload (170, 171, 180). It has been also proposed that transient mPTP openings occur asynchronously in the mitochondrial population in a stochastic fashion (198). This may allow this population of asynchronously cycling mitochondria to tolerate markedly extramitochondrial Ca²⁺ overload without collectively depolarizing, whereas transient increased ROS production due to transient mPTP opening may engage cardioprotective signaling.

A recent in vitro study (169) suggested that both transient formation and inhibition of the mPTP can be considered for therapeutic purposes, and that there is a defined therapeutic window, with the first few minutes of reoxygenation being a crucial period to achieve protection. This is the first study describing direct transient opening of the pore as a possible clinical target for cardioprotection in a postischemic phase. It is necessary to confirm in future work the observations reported in this study in in vivo animal models and in humans.

Thus, the emerging picture is that similar to ROS, (i) mPTP opening is a two-edged sword, with both protective (transient opening) and deleterious (prolonged opening) actions in both the pre- and postischemic phases. However, this picture needs to be supported by future in vivo studies; (ii) a more substantial picture is the one that sees the transient opening of mPTP to be protective in the preischemic and the prolonged opening to be deleterious in the postischemic phase. Therefore, the pharmacological closure of the pores at this latter stage is considered to be highly protective. These two scenarios are not mutually exclusive.

D. Chronic ischemia and mitochondria

Although the review focuses on acute I/R and protection, here we will briefly consider the effects on the mitochondrial structure and function and the possible therapeutic target in chronic ischemia. For a more detailed insight into characteristics of mitochondria in chronic ischemia, we kindly refer the reader to related reviews elaborating on this point (e.g., 312).

Myocardium can respond to a prolonged decrease in perfusion by achieving a balanced state of downregulation of metabolic demand, termed “hibernating” myocardium, which is characterized by several alterations of the mitochondrial structure and function. A number of small mitochondria within the area of reversible lesions have been described in the myocardium chronically underperfused (209). Two key events have been observed after acute ischemia reperfusion and chronic ischemia. These events are the decrease of the stimulatory effect of creatine on metabolic activity and the increase of OMM permeability to Cyt c and ADP. These effects indicate the alteration of the intermembrane space architecture, leading to the loss of intracellular adenine nucleotide compartmentation and possibly of functional coupling of mitochondrial creatine kinase and ANT. These alterations result in the impairment of intracellular energy transfer from mitochondria to ATP-utilizing sites located in the cytosol. This may play a significant role in ischemic injury and alterations in the heart function (87). Preservation of high-energy phosphate levels and reduction in MVO2 have been observed after ischemia in chronic hibernating myocardium (160). These functional alterations are usually supported by structural alterations characterized by mitochondrial clustering and swelling associated with membrane rupture. Yet, favorable mitochondrial adaptations in a swine model of chronic myocardial ischemia have been observed. In this model, isolated mitochondria from the ischemic tissue demonstrate preserved state-3 respiration after anoxia/reoxygenation (228). This is consistent with a stress-resistant state, which is characterized by a mild degree of uncoupling under basal conditions and decreased O₂ ^−• generation. In fact, UCP 2 expression is enhanced in these mitochondria, providing a potential mechanism for the favorable mitochondrial adaptations (228). It may be argued that highly localized, specific mitochondrial enzyme changes (beneficial/detrimental) may result from chronic myocardial ischemia, which may make the differences between surviving/hibernating or dying tissue.

Protective strategy: the transcriptional coactivator PGC-1α coactivates downstream transcription factors such as estrogen-related receptors, nuclear respiratory factors, and peroxisome proliferator-activated receptors (PPARs). These factors are known to regulate many aspects of energy metabolism, including mitochondrial biogenesis, fatty acid oxidation, and antioxidant detoxification (218, 312). Since PGC-1α is upstream of the regulation of energy metabolism, targeting it may represent an approach allowing an improvement of energy metabolism in multiple sites (233). Improving mitochondrial biogenesis by upregulating PGC-1α is a strategy largely at work in metabolic diseases, and it has been proposed for chronic ischemic disease. PGC-1α is regulated in a cardiac-specific manner (233), and more work is needed to understand its role and regulation in the normal and failing heart. Because ischemic disease is also a metabolic disease, such a strategy could be beneficial in this pathology.

A. PreC and PostC

As stated in the introduction, in 1986, Murry et al. (239) reported that four 5-min circumflex occlusions, each separated by 5 min of reperfusion, followed by a sustained 40-min occlusion (index ischemia=infarcting ischemia) in the dog heart dramatically attenuated I/R injury. This phenomenon was named PreC. In 2003, Zhao et al. (393) reported that three episodes of 30 s of reperfusion/30 s of ischemia performed immediately after index ischemia (60-min coronary occlusion) in the dog heart drastically attenuated reperfusion injury. This phenomenon was named PostC. It was soon clear that the later the application of the first PostC ischemia, the lower the protection (Fig. 2). The discovery of ischemic PostC provided strong evidence supporting the existence of reperfusion injury (154). It is commonly believed that cardioprotective signals mediate protection by acting on the mitochondria to inhibit mitochondrial-mediated cell death. Therefore, an understanding of the mechanisms of cardioprotection is strictly linked to an understanding of the mechanisms by which mitochondria regulate cell death.

The protective effects observed with PostC are comparable to those observed with the powerful PreC (269, 393). PostC has been shown to have salubrious effects on different tissue types within the heart (cardiomyocytes and endothelium) and protect against various pathological processes, including necrosis, apoptosis, contractile dysfunction, arrhythmias, and microvascular injury or no reflow. The mechanisms by which PostC alters the pathophysiology of reperfusion injury is exceedingly complex, and involves physiological mechanisms (e.g., delaying realkalinization of tissue pH, triggering release of autacoids, redox signaling, and opening and closing of various channels) and molecular mechanisms (especially activation of kinases of the RISK and SAFE pathways; see also below) that impact on cellular and subcellular targets or effectors (260, 367, 393).

The term “PostC” has highlighted the importance of intervening at the beginning of myocardial reperfusion to protect the postischemic heart. This is a clinically relevant timepoint for intervention in patients presented with AMI. As such, its clinical application has been rapid for both STEMI patients undergoing primary PCI (e.g., 333 and 356) and for patients undergoing on-pump cardiac surgery (223) (see also section XI).

Intensive investigation of the signaling pathways underlying PreC and PostC has identified a number of signal transduction pathways conveying the cardioprotective signal from the sarcolemma to the mitochondria, some of which are common in PreC and PostC. In fact, both PreC and PostC induce activation of signaling elements during the early reperfusion after the index ischemia (90) (Figs. 3 and 11). In this phase, a great attention has been focused on the cGMP/PKG pathway (e.g., 153, 268, and 271), on the RISK pathway (e.g., 132, 139 and 326), which involves the kinases Akt and ERK1/2, and more recently, on the SAFE pathway that contributes to PostC protection through the activation of TNF-α, its receptor type-2, JAK, and STAT-3 (e.g., 137 and 212). All these pathways in PreC and PostC converge on mitochondria via the modulation of several kinases, including GSK-3β, Bcl-2/Bax/Bad, and PKCɛ (e.g., 137 and 236). In Figure 9 are reported only the principal modalities of mitochondrial control by cytosolic kinases. Nevertheless, these modalities are still controversial (66, 67, 113, 296).

OPEN IN VIEWER

FIG. 11.

Key mechanisms proposed for PreC- and PostC-induced decrease in susceptibility to mPTP opening during reperfusion. Key events are transient physiological mPTP opening in the preischemic phase and acidosis plus redox signaling in reperfusion. These events avoid prolonged mPTP opening, allow the activation of cardioprotection pathways, and lead together to mPTP inhibition to the prevention of cell death and organ protection. For other acronyms, see the list of Abbreviations Used.

As said, mPTP has been shown to open at the start of reperfusion, and its opening is generally attributed to an overload of matrix Ca²⁺ and/or ROS increase. Therefore, it is likely that cardioprotective signaling reduces matrix Ca²⁺ and/or ROS, rather than by direct modulation of mPTP components. However, redox signaling and acidosis in early reperfusion are two cardioprotective processes operating in the very early reperfusion in both PreC and PostC (Figs. 3 and 1). Therefore, these two processes may act, first, directly on mPTP components, limiting their opening and then may activate signaling pathways that have been suggested to converge again on mitochondria, definitely decreasing susceptibility to mPTP opening and mediating protection. More details on these two processes are given in the next sections.

B. Redox signaling and acidosis in early reperfusion

We must consider the detrimental effects of ROS within the heart, before discussing the beneficial role of redox signaling.

A. ROS/RNS signaling may be modulated by antioxidants

Reactive species function as trigger molecules of protection by activating protein kinases such as MAPK and AMPK, as well as PKC, within and outside the mitochondria, including PKCɛ (68, 72, 199, 260, 262, 270, 288, 341) and the MAPK p38 and/or JAK/STAT (387). As said, several mitochondrial components are targeted by ROS/RNS via oxidative/nitrosative processes. Accordingly, many large-spectrum scavengers of ROS/RNS, such as ascorbic acid, MPG, or NAC, attenuate infarct size reduction by ischemic or pharmacological PreC or PostC, in several animal species (e.g., 109, 253, 260, 262, 269, 270, and 276). Since a target of ROS/RNS in redox signaling is the PKC (199, 260, 288, 339, 341), the hearts can be preconditioned by simply infusing free-radical generators into the coronary arteries, and the resulting protection can be blocked by either a PKC antagonist or a ROS scavenger (61, 358, 382). Evidence exists that ROS-activated PKC will also protect the reperfused heart (262, 276). This sequence would explain the observation that the PKC activator could rescue hearts experiencing acidic and hypoxic reperfusion (62). Moreover, chelerythrine, a wide-spectrum PKC antagonist, blocks PostC protection (276, 388).

Indeed, though it has been reported that ROS/RNS can activate kinases, and ROS/RNS scavenger may attenuate cardioprotective pathways, recently, we have observed in an ex vivo study that PostC and acidosis induce downregulation of SOD, whereas catalase activity does not change in the early reperfusion phase (275). Moreover, PostC reduces 3-nitrotyrosine (3-NT) and increases S-nitrosylated protein levels, thus contributing to cardioprotection triggering (275). Indeed, in addition to activating cGMP/PKG-dependent signaling pathways, NO^•/RNS can modify sulfhydryl residues of proteins through their SNO, which has emerged as an important post-translational protein modification. Thus, SNO of proteins is the result of a transnitrosylation reaction, or alternatively, donors of NO⁺ (e.g., N₂O₃) can nitrosylate a cysteine moiety. SNO of mitochondrial proteins has attracted the attention of researchers in cardioprotection (62, 64, 112, 175, 252, 274, 275). Therefore, SNO has been proposed to be very important in both ischemic PreC and PostC cardioprotection, limiting oxidative stress. As such, SNO of proteins can be seen as a result of redox signaling, leading to a potentiation of the antioxidant system.

B. Role of SNO in regulating mitochondrial function in I/R and cardioprotection

Within the context of redox signaling, the increase in NO^• occurring during PreC (18, 401) or PostC (173, 205, 262, 274 –276) can lead to an increase in SNO of several mitochondrial proteins (47, 340). These are a subproteome of SNO proteins (see below), which include complex I (47) and other proteins involved in mitochondrial energetics, calcium transport, and metabolism (340). Intriguingly, induction of the mitochondrial permeability transition by N-ethylmaleimide has been seen to depend on secondary oxidation of critical thiol groups. In fact, it has been reported that mPTP putative components (see above) may have several important cysteines residues, and that mPTP opening is blocked by low levels of N-ethylmaleimide (21, 79). Depending on the NO^• level, these cysteines may be nitrosylated and the activity of mPTP regulated. Furthermore, SNO of proteins is rapidly reversible so that changes would be rapidly reversed on reperfusion. It has been reported that cardioprotection by PreC or by treatment with the naturally occurring transnitrosilating agent S-nitrosoglutathione (GSNO) resulted in an increase in SNO of several mitochondrial proteins involved in regulation of mitochondrial energetic and calcium transport, including acyl-CoA dehydrogenase, alpha-ketoglutarate dehydrogenase, complex I, creatine kinase, F₀F₁-ATPase α1 subunit, and malate dehydrogenase (340). It has been also shown that PreC results in SNO of complex I, which may result in less ROS generation in the setting of ischemia and reperfusion (46, 47). We have shown that also PostC results in SNO of mitochondrial proteins (275). For further discussion on the importance of SNO in cardioprotection, we kindly refer the reader to related articles elaborating on this point (e.g., 260, 274, 340, and 341).

A. Mitochondria and MS

Since the MS is an archetypal condition with an altered mitochondrial function, and since it has an increasing prevalence in the patients with ischemic cardiovascular disease (97), this condition will be considered with more details. Other risk factors of heart diseases are briefly discussed; for further details, the reader may see (104, 259, 368).

MS is now defined as a complex syndrome, which includes several interrelated disturbances of glucose and lipid homeostasis, which may lead to diabetes (108). Since the MS induces a significant variation of O₂ consumption, alterations in the mitochondrial function and mitochondrial morphology play a major role in this syndrome (43). Several studies associated mitochondrial dysfunction and the MS; for example, Wisløff et al. (377) proposed an association between low expression of the protein involved in mitochondrial function, in the skeletal muscle, with the occurrence of multiple cardiovascular risk factors (insulin resistance, hyperlipidemia, and hypertension) in rats with low oxidative capacity. In humans, for instance, Peterson et al. (281) found that the increase in Body Mass Index was associated with an increase in myocardial VO₂, reduced cardiac efficiency (ratio of cardiac work to VO₂), and impaired glucose tolerance correlating with increased fatty acid utilization. A reduced cardiac phosphocreatine/ATP ratio, an impaired high-energy phosphate metabolism, and a cardiac energy deficit have been reported (314).

In diabetic and obese individuals, the increased myocardial fatty acid oxidative capacities are mediated, at least in part, by increased activity of PPARs (43), which are regulators of fatty acid oxidation in the heart. In particular, PPARα increases the expression of genes involved in several steps of cardiac fatty acid utilization (11). It has been reported that either high-fat diet or streptozotocin-induced diabetes may increase UCP expression in the striated muscles (54, 187). This increased expression may play a pivotal role in supporting increased oxidative stress. Moreover, type-2 diabetes patients have been shown to possess reduced ETC capacities and reduced citrate synthase activity (189, 316). The preferential metabolism of fatty acids, reducing glucose utilization (280), causes gradual dysfunction of pancreatic beta and other cells and upregulation of mitochondrial UCP2 and UCP3, thus resulting in an increased uncoupling of mitochondrial respiration, a reduced ETC activity, and ATP production, with subsequent skeletal muscle fatigue (126, 317). Whether this occurs in the myocardium is not clear. Preferential metabolism of fatty acids and reduced glucose utilization correlate with the reduced muscle ETC activity (189) and with the decreased whole-body anaerobic capacity (315). Studies demonstrated disparate effects of oxidative stress on mitochondria and implicated oxidative stress as a cause of mitochondrial dysfunction in diabetic hearts (Fig. 12).

OPEN IN VIEWER

FIG. 12.

Main alterations leading to mitochondrial and cardiac dysfunction in metabolic syndrome (MS). MS is considered as one of the archetypal conditions that may alter mitochondrial function, thus compromising the possibility to induce cardioprotection. For other acronyms, see the list of Abbreviations Used.

The low levels of O₂ ^−• and H₂O₂ normally produced by mitochondria are overproduced in MS and associated diseases (219, 316, 317). In fact, in MS and diabetes, ROS may be predominantly derived from mitochondria as opposed to cytosolic origin. As described above (Fig. 5), excess O₂ ^−• can (i) directly damage iron–sulfur center-containing enzymes, (ii) be converted to H₂O₂ (and ultimately to OH^•), and (iii) react with NO^• to produce ONOO⁻, a potentially dangerous RNS (19). As said above, fatty acids are particularly sensitive to ROS/RNS oxidation, resulting in the formation of lipid peroxides, which are cytotoxic and lead to free-radical damage of other lipids, proteins, and DNA, especially mitochondrial DNA (119, 316). Even before the diagnosis of MS or type-2 diabetes, the accumulation of oxidized fatty acids in mitochondria can result in a progressive oxidative damage. In fact, continually excessive ROS production by mitochondria has a role in the development and progression of diabetes and its complications, including cardiac pathologies.

Mitochondrial proteome changes with diabetes types 1 and 2 (38, 320). Moreover, MS and diabetes are also associated with alterations in kinase phosphorylation, in particular in ERK1/2 and Akt phosphorylation, possibly due to alteration in phosphatase activity (164, 166, 293, 359, 368).

Taken together, in diabetes and MS, in general, altered mitochondrial substrate flux, increased activation of UCPs, reduced activation of kinases, and excessive ROS/RNS production may play important roles in the development and progression of the disease and in contributing to cardiac dysfunction (Fig. 12). Thus, novel treatments that are targeted to these alterations might lead to new therapeutic approaches for the prevention of cardiac dysfunction.

B. Cardioprotection in Aging

Clearly, the cardioprotective effect of PreC and PostC and pharmacological conditioning decrease with aging (1 –3, 25, 30, 292).

Oxidative stress is considered causal for the process of aging (330, 375). Aging cardiomyocytes are subjected to enhanced ROS production, which damages mitochondria, reduces mitochondrial fission, and contributes to the formation of larger mitochondria (351, 375). These giant mitochondria are not effectively removed by mitophagy, and therefore accumulate within cells. They often contain mutated DNA and altered proteins of the respiratory chain, and therefore contribute to excessive ROS formation and further oxidative protein damage (330, 351, 375).

Age and associated comorbidities are characterized by changes in the expression and activity of typical cardioprotective signaling proteins (e.g., ERK1/2, PI3K-Akt, STAT-3, p38 MAPKs, and PKCs), both under baseline conditions and in response to stressors such as I/R (123, 168, 200, 201, 242). It is not yet clear if any of this variability in response originates at the mitochondrial level. However, it has been suggested that Cx43 and STAT-3 are reduced not only in ventricular total protein extracts but also in mitochondria isolated from aged mice (28, 30). Moreover, the amounts of nuclear encoded mitochondrial proteins are also affected by age and associated diseases. These include proteins involved in ATP synthesis and mitochondrial respiration, as well as STAT-3 and Cx43, which are considered important for cardioprotection (24, 31, 32). In fact, with aging the transcription of genes encoding proteins with different functions, including those of ETC (23), is reduced, thus contributing to the development of mitochondrial dysfunction (27).

A decreased level of STAT-3 in aged hearts may affect the expression level of STAT-3-target genes. Indeed, the expression of MnSOD is decreased in aged rat hearts (105). However, in mice, mRNA and protein levels of the STAT-3-target iNOS are elevated with age (14, 381). It is likely that increased levels of TNF-α in aged mouse hearts may contribute to the enhanced transcription of iNOS via enhanced O₂ ^−• formation and subsequent NFκB activation (14). Since the cardioprotection by SWOP and pharmacological PreC is dependent on iNOS (35, 265), the enhanced iNOS levels may contribute to the preservation of SWOP in aged hearts.

On the whole, in aging, there is an increased susceptibility to I/R, which is likely due to enhanced oxidative stress (31, 330, 351). Moreover, ischemic PreC and PostC lose their effectiveness. The mechanisms responsible for the loss of protection in the aged heart include alterations in gene/protein expression, signal transduction cascades, and altered mitochondrial function (e.g., ROS formation and respiration).

However, some contrasting results are reported; for example, a limitation of I/R injury with PreC has been reported in rabbits aged 2–4 years (227, 291) and in sheep aged 5–8 year (45); rats aged 16–18 months did not lose the possibility to protect the heart with ischemic or pharmacological PostC (323, 385). We cannot rule out that an inappropriate definition of aged cohorts may explain some of these discrepant findings. In fact, we should keep in mind that aging is a continuum; thus, a temporal phase of adaptability in cardioprotective signaling, including alternative mediators or signaling pathways, may explain some positive results in models of initial aging (291). In this regard, prolonged, preischemic administration of opioids, as well as caloric restriction and exercise, has been seen to contribute to adaptations that prevent the age-induced loss of cardioprotection (1, 2, 256).

C. Cardioprotection in hypertension and hypertrophy

Left ventricular hypertrophy as a consequence of sustained elevation of cardiac work load is an important form of cardiac remodeling, which occurs in many pathological conditions, among which arterial hypertension is the most common. There is experimental evidence that hypertrophied myocardium is at a greater risk of sustaining injury after I/R [for review see (104)]. However, there are few studies in hypertensive animals with infarct size as the principal end-point, and these studies have not clearly demonstrated an increased infarct size in the hypertrophied heart (93, 104, 278). Nevertheless, several biochemical and metabolic alterations have been observed in the hypertrophic myocardium. These include altered mitochondrial energetics and ATP production and alterations in glycolytic metabolism with lactate accumulation during ischemia (4, 250, 257). Moreover, increased ROS generation and reduced antioxidant potential have been described (13). Although these data have not been consistent, they may explain the increased sensitivity of hypertrophied myocardium to I/R injury (104).

The cardioprotective effect of pharmacological PreC is blunted in models of hypertension and hypertrophy. Yet from studies on ischemic PreC, it seems that the protection is conserved in the early stages of hypertrophy and lost as hypertrophy increases (92, 93, 104). Interestingly, initial reports also suggest that the beneficial effects of PostC are largely maintained in models of hypertension and/or hypertrophy (103, 215, 278, 394). Among the studies in which the infarct size was assessed, one has demonstrated robust, PostC-induced cardioprotection in a mouse model of transverse aortic constriction via the ERK1/2 pathway (215); another study has shown that PostC protects remodeled hypertrophic myocardium via the PI3K-PKB/Akt pathway (394). We observed a nonsignificant trend toward reduction of infarct size with PostC in spontaneously hypertensive rats (278). Moreover, in this study, we have shown that protection by PostC cannot be added to the beneficial effects of antihypertensive therapy with captopril. In other studies (267, 279), we have shown that during nandrolone treatment, which induces a progressive myocardial hypertrophy, the efficacy of ischemic PostC was preserved or even enhanced in a not-yet hypertrophic heart; this is reminiscent of the supernormal period (with increased resistance to I/R) observed during experimental diabetes development. However, the protective efficacy of PostC was lost in hearts clearly hypertrophic. In this study, we have shown that the levels of survival kinases change during hypertrophy development (279).

Whether infarct size reduction with ischemic PreC and PostC is, indeed, preserved in hypertension and/or hypertrophic myocardium remains to be substantiated in future studies. Moreover, the possible alterations of the cardioprotective pathways, especially those of PostC, in the hypertrophic/hypertensive heart need to be further studied.

A. The possible reasons of variable outcomes with ischemic PostC

In our opinion, it is not surprising that a variability (ranging from highly protective to negative results) in the magnitude of myocardial salvage can be observed among some clinical and experimental studies. Both the I/R injury and the PostC protection may be influenced by a plethora of conditions that is not easy to keep under the control of the operators (physicians, cardiologists, and researchers), especially in the clinical arena. Particularly relevant are the variables correlated to artery reopening. For example, in determining whether I/R injury may be relevant whether or not the artery is fully reopened. This is also relevant in determining PostC cardioprotection: an artery not completely reopened in the I/R setting (control group) may exert a sort of gentle reperfusion that may reduce the infarct size, thus restricting the difference with hearts protected by ischemic PostC (stuttering reperfusion in the treatment group). On the other hand, an artery not completely reopened during PostC maneuvers may reduce cardioprotective effects, thus further restricting the difference with hearts treated with full reperfusion only. Many of the negative studies with PostC did not check whether or not the artery was fully reopened during the stuttering reperfusion (PostC maneuvers). Moreover, as said above, factors that are known to influence/limit the cardioprotective benefits of both PreC and PostC include age, gender, and the use of drugs (e.g., 46, 161, 235, 277, and 350). In fact, a number of medications can limit I/R injury, including ACE inhibitors, adenosine, AT1-blockers, statins, β-blockers, dronedarone, ivabradine, and nitroglycerine (60, 149, 150, 272, 278, 327). Moreover, drugs such as COX inhibitors can interfere with PostC molecular mechanisms (273). Therefore, it is not only the attenuation of protection in the PostC cohort but also the potential recruitment of protection in the control group, which may minimize any difference between PostC and control. Moreover, another possible confounder is the presence of comorbidities (MS, diabetes and hyperlipidemia, and hypertension), including increased age, which are all risk factors for cardiovascular diseases (see above). In fact, patients exhibiting one or more of these risk factors are the actual populations in which the incidence of AMI is greatest, and thus cardioprotection is required (104, 221).

Therefore, due to the complexity of alterations in the presence of comorbidities and due to complexity and variability of ischemic PostC responses, this procedure will be translated to a clinical routine only when the signaling pathway is fully understood. Moreover, we need to understand the reasons for the incongruity between experimental models and clinical outcome, especially in the settings with comorbidities. In fact, it seems that the majority of comorbidities limit cardioprotection in the experimental setting, whereas comorbidities do not appear as a major problem in a clinical setting (179, 284, 333, 353). How can this discrepancy be reconciled? One difference may be due to the fact that humans with comorbidities are receiving adequate treatments. There are encouraging clinical data suggesting that patients with AMI may benefit from preserved mitochondrial function, thereby emphasizing the paramount importance of mitochondria as therapeutic targets to limit I/R injury. The beneficial effects are obtained despite an important incidence of comorbidities in these patients.

B. Pharmacological PostC

Since mPTP is a major factor in determining cell death and is considered one final step in cardioprotective signaling cascade, mPTP inhibition affords significant cardioprotection (10, 81, 127, 142); in fact, this concept has been successfully translated into clinical practice (e.g., 179 and 284). Yet, ischemic PostC may be limited to angioplasty and surgery, and may be associated with increased complications of the operative procedure, and thus the most constructive strategy would be pharmacological PostC.

Pharmacological PostC would avoid the unfavorable consequences linked with intermittent artery cross-clamping and provide a simple method of myocardial protection subsequent to all cardiac procedures, including coronary reopening with thrombolytics (384). However, caution must be used; in fact, previous attempts to intervene at the onset of reperfusion using pharmacological agents as a strategy for protecting against myocardial reperfusion injury have been extensively investigated, and in terms of translation, the treatment strategy into clinical therapy has been largely disappointing (33, 384). For instance, adenosine has been considered for long time as the ideal candidate for protection against I/R damage, based on several experimental evidence of its protective effects, which however did not solve the conundrum of effective receptors: adenosine has at least four receptors that can change their affinity for the agonist at different times of ischemia and reperfusion, and at present, it is unclear which ones of the different receptors are more important at different stages (60, 61, 368). Despite intensive investigation with adenosine and specific receptor agonists, it is difficult to conclude as to a protective role of adenosine in STEMI patients (179, 303). Glucagon-like peptide-1 is another drug that has been demonstrated to improve the left ventricular ejection fraction by about 35% in patients with severely reduced left ventricular function after AMI (245), but the study was not designed to reduce lethal reperfusion injury, as the drug was administered several hours after reperfusion, that is, far from the therapeutic window for reperfusion injury. However, initial progress has been made with more novel approaches for preventing myocardial reperfusion injury administering the drug in the first minutes of reperfusion (135). As said, preliminary clinical data indicate that drugs targeting mPTP or RISKs may confer benefits to patients with AMI, with and without comorbidities, increasing that provided by myocardial reperfusion alone. Very good results are obtained with drugs such as CsA as an mPTP desensitizer (230, 284, 333), as well as with other drugs targeting RISKs, such as erythropoietin and its analogs (e.g., 22 and 44). Moreover, the protective effects of the K_ATP channel opener, nicorandil, on the one hand, and natriuretic peptide, on the other hand, were studied in J-WIND trials (195). Treatment with a natriuretic peptide was associated with a small but significant reduction in the infarct size and improvement in the ejection fraction, corresponding to a better clinical outcome. In this study, nicorandil did not affect the size of infarct size, though oral administration of nicorandil during follow-up increased the left ventricular ejection. In another small study, a dose of nicorandil that was three times higher than that used in J-WIND decreased the infarct size and reduced re-admission to hospital for chronic heart failure or the rate of cardiovascular death in patients with AMI (178).

The treatment with CsA in the setting of AMI in conjunction with reperfusion therapy was safe and significantly reduced the size of the infarction (179). In a follow-up study, Mewton et al. (230) recently examined whether CsA might have modified LV remodeling in that subgroup of patients with appreciable effects on the infarct size. The authors described that a reduction persisted after 6 months, and that CsA did not exert any deleterious effect on LV remodeling.

Importantly, similar cardioprotective effects were obtained with other drugs acting on mPTP, confirming the relevance of this approach. For instance, derivatives of CsA, such as [N-methyl-ala⁶]CsA, [N-methyl-Val⁴]CsA, Debio 025, NIM811, or sanglifehrin A (126 –128, 133, 134, 136 –139, 141, 142, 230), also prevent myocardial I/R injury in an experimental setting. TRO40303 is a new drug that inhibits mPTP triggered by oxidative stress. Its efficacy in an animal model of AMI makes TRO40303 a promising new drug for the reduction of myocardial I/R injury, which acts on mitochondrial translocator protein 18 kDa (TSPO) at the cholesterol site. TSPO is a five-transmembrane-domain protein that is localized primarily in the OMM, and it is involved in the translocation of cholesterol from the OMM to the IMM. Since TRO40303 binds to the TSPO, its mode of action differs from that of other mPTP desensitizer, such as CsA, thus providing a new promising pharmacological approach to study mPTP regulation (313). In our opinion, a fortunate aspect of these drugs is that they are not inhibitors, but desensitizers, which do not abolish ROS production in any cellular conditions, but only inhibits it, thus, might not interfere with the protective redox signaling.

One molecule that has received much attention is nitrite (NO₂ ⁻) (324). A strong debate surrounds the mechanisms of nitrite-mediated cardioprotection, with a central focus on the identification of the nitrite reductase responsible for the biotransformation of this molecule in hypoxia. Candidates for this activity include deoxyhemoglobin, deoxymyoglobin, and various components of the respiratory chain itself. In addition, the mechanism by which the hypoxic metabolites of nitrite bring about protected phenotype is also matter of controversy, but interestingly, it is thought that mPTP, ROS-modulation, and SNO of proteins may play a role (306, 309). Notably, regardless of these mechanisms, nitrite is now entering clinical trials for AMI. Much attention in the cardioprotection field has also focused on the effects of volatile anesthetics, which are largely shown to be cardioprotective in many species, including humans (130, 165, 214, 298, 336). It has been shown that some of these anesthetics can inhibit mitochondrial respiration, whereas others are thought to activate K⁺ channels in the mitochondrial membrane (214, 336). Yet, irrespective of the exact mechanism, anesthetics currently represent one of the most widely used, safest, and most clinically applicable cardioprotective agents.

C. Features of a successful cardioprotective approach in early reperfusion

Due to the abundance of mitochondria in cardiomyocytes and to the compelling dependence of myocardial function on oxidative phosphorylation, it is hardly surprising that treatments targeting these organelles are successful in myocardial cardioprotective interventions. Besides the more appropriate target (mitochondrial elements) reached by the drugs, it is likely that novel approaches can achieve more success, because the essential factors are understood and considered recently. First of all, the time window for ischemic PostC is a critical factor (193, 269, 270). In fact, during angioplasty, the treatment (either ischemic or pharmacological PostC) should be applied before (drugs) or within the first minute after direct stenting of the culprit coronary artery. Several other major confounding factors that have been shown to deeply influence the results of all infarct size reduction studies must be taken into consideration, including the timing of the protective intervention with respect to reflow and the major determinants of infarct size (duration of ischemia, size of the area at risk, collateral circulation, and microembolic complications) (147). Besides the phosphorylation, also the redox signaling, the SNO and the other forms of signaling in the early reperfusion should be borne in mind for a successful treatment in reperfusion. The disappointing results with antioxidants may be due, at least in part, to limitation of protective ROS in the very early phase of reperfusion (260). The possibility also exists that intermittent treatment might be helpful (271 –273). In fact, we have shown that brief repetitive (intermittent) administration of bradykinin during early reperfusion can trigger PostC through a redox signaling (271). Finally, Cohen et al. (64) have highlighted the fact that staccato reperfusion reintroduces intermittent reoxygenation to perpetuate myocardial acidosis and to initiate protection.

Therefore, a pharmacological approach in the very early phase of reperfusion, which targets mitochondria, allows a ROS signaling, and preserves acidosis, may represent an important goal for strategies protecting ischemic–reperfused myocardium. Of course, the presence of comorbidities and drugs already used by the patient must be considered and targets adjusted accordingly.