Translational issues for mitoprotective agents as adjunct to reperfusion therapy in patients with ST‐segment elevation myocardial infarction

Abstract Pre‐clinical studies have indicated that mitoprotective drugs may add cardioprotection beyond rapid revascularization, antiplatelet therapy and risk modification. We review the clinical efficacy of mitoprotective drugs that have progressed to clinical testing comprising cyclosporine A, KAI‐9803, MTP131 and TRO 40303. Whereas cyclosporine may reduce infarct size in patients undergoing primary angioplasty as evaluated by release of myocardial ischaemic biomarkers and infarct size imaging, the other drugs were not capable of demonstrating this effect in the clinical setting. The absent effect leaves the role of the mitochondrial permeability transition pore for reperfusion injury in humans unanswered and indicates that targeting one single mechanism to provide mitoprotection may not be efficient. Moreover, the lack of effect may relate to favourable outcome with current optimal therapy, but conditions such as age, sex, diabetes, dyslipidaemia and concurrent medications may also alter mitochondrial function. However, as long as the molecular structure of the pore remains unknown and specific inhibitors of its opening are lacking, the mitochondrial permeability transition pore remains a target for alleviation of reperfusion injury. Nevertheless, taking conditions such as ageing, sex, comorbidities and co‐medication into account may be of paramount importance during the design of pre‐clinical and clinical studies testing mitoprotective drugs.


| INTRODUC TI ON
Modern reperfusion therapy has improved outcome for patients with ST-elevation myocardial infarction (STEMI) tremendously. 1 Over the past 5 years, however, mortality reduction has levelled out 1 and the decline in the incidence of post-MI heart failure is modest. 2 So, there still may be a need to reduce infarct size to further improve outcome.
Because infarct size depends on ischaemia time, the most important way to diminish it and improve outcome remains a reduction in the ischaemic time by reducing any delay and insuring rapid revascularization in STEMI patients. Beyond this focus, a major target may be an attempt to reduce infarct size by addressing the reperfusion injury that occurs, when injuring mechanisms are activated upon opening of the coronary artery. 3,4 Mitochondria in the heart are crucial for the generation of adenosine triphosphate (ATP) necessary to sustain contractile function and for the dynamic adjustment of the cardiomyocytes' metabolic demand and ionic homeostasis. Hence, the organelle is considered an important target for cardioprotection of the myocardium exposed to an acute ischaemia-reperfusion injury. In particular, acute opening of the mitochondrial permeability transition pore (MPTP) has been involved in ischaemia-reperfusion injury 5,6 because of its disruptive role in mitochondrial respiratory coupling and ATP production. 7,8 Experimental studies indicate that pharmacological approaches aimed at preventing MPTP have cardioprotective effects in the context of myocardial ischaemia reperfusion. 9,10 However, translation of this concept into the clinic has been disappointing, [11][12][13][14][15][16][17] suggesting that targeting a single intracellular molecule, such as the MPTP or dynamin-related protein 1 (Drp1), 18 may not be sufficient to create cardioprotection. 19,20 It emphasizes that more mechanistic insight about the mode of action of cardioprotective modalities is needed.
Lack of efficacy might also reflect that clinical outcome in STEMI patients undergoing primary percutaneous coronary intervention (PCI) is excellent by modern reperfusion therapy, such that ischaemia reperfusion as a target for protection has diminished. Median infarct size with current reperfusion therapy is small-in the order of magnitude of 7% and in anterior infarcts 16% of the left ventricle. 21 Infarct sizes up to 17% rarely translate into clinical symptoms manifesting as cardiac death and hospitalization for heart failure, 22 which are the most appropriate clinical end-points for evaluating the efficacy of cardioprotective pharmacological agents. 23 The aims of the present review were to provide an overview of the pharmacological agents that have advanced to clinical testing and to identify obstacles for a clinical benefit in order to clarify whether pharmacological mitoprotection is a useful way to pursue for improving outcome in STEMI patients undergoing reperfusion therapy.

| PATHOPHYS IOLOG IC AL BACKG ROUND FOR INTERVENTI ON AG AIN S T MITOCHONDRIAL DYS FUN C TION IN ISCHAEMIA-REPERFUS ION INJ URY
In a clinical context, mitochondrial dysfunction has been reported in cardiac diseases including ischaemia-reperfusion injury as well as in comorbidities associated with ischaemic cardiomyopathies such as diabetes or obesity. Under aerobic conditions, mitochondria are indispensable for cell function and viability primarily through ATP production, regulation of cellular redox potential and control of apoptosis. Because of their close functional and anatomical association with the sarcoplasmic reticulum, mitochondria play an active role in calcium uptake, which in turn is a critical regulator of the Krebs cycle and therefore of mitochondrial respiration and NAPDH-dependent antioxidant regeneration. [24][25][26] Mitochondrial calcium uptake is driven by the mitochondrial calcium uniporter, whose low affinity is counteracted by the high calcium concentration present at the sarcoplasmic reticulum-mitochondria interface. 27 Mitochondrial dysfunction can be of various origins and may alter cell homeostasis or viability through several mechanisms including reduction in ATP production, enhanced oxidative stress and release of pro-apoptotic molecules. Following prolonged ischaemia reperfusion, the accumulation of calcium within the mitochondrial matrix and increased reactive oxygen species (ROS) production favour the opening of the MPTP. The opening of this non-selective mega-channel dissipates the mitochondrial electrochemical gradient necessary for ATP production and precipitates energy exhaustion and mitochondrial matrix swelling. Rupture of the outer mitochondrial membrane favours the release of pro-apoptotic factors. During ischaemia, MPTP remains inhibited because of the acidic pH, but upon restoration of blood flow, rapid normalization of pH in the presence of calcium overload and excessive ROS triggers MPTP and exacerbates reperfusion-induced cell injury. 5

| Cyclosporine A
Apart from the immunosuppressive action related to its binding to the cytosolic calcineurin and subsequent inhibition of the transcription factor nuclear factor of activated T cells (NFAT), cyclosporine A (CsA) inhibits MPTP opening following binding to the mitochondrial matrix chaperone CypD ( Figure 1). CypD, which is known to bind to the oligomycin sensitivity conferral protein (OSCP), acts on the MPTP by facilitating the removal of the F 1 domain from the c subunit in a CsA-sensitive manner during pore opening. 31 The concept of a putative cardioprotection by CsA was also based on the observation that CypD-deficient mice develop significantly smaller infarcts after a prolonged ischaemic insult. 32,33 Hausenloy et al first reported that administration of CsA at the time of reperfusion could reduce infarct size in the isolated rat heart model. 34 In vivo administration of NIM811, a non-immunosuppressive CsA derivative, was able to inhibit MPTP opening in mitochondria isolated from reperfused rabbit myocardium and limit infarct size when administered at the time of reperfusion. 35 CsA also reduced infarct size in anaesthetized pigs with 90-minutes regional ischaemia, when given intravenously just before reperfusion. 36 However, CsA failed to induce robust cardioprotection in other studies with efficacy dependency on the experimental conditions, such as duration of ischaemic period. 37,38 Also, CsA administered during reperfusion fails to restore cardioprotection in pre-diabetic Zucker obese rats in vivo. 39 Overall, CsA variably and inconsistently seems to reduce infarct size across species in experimental models of reperfused myocardial infarction. 40

| KAI-9803
Although contentious, 41 the protein kinase C (PKC) family of isoenzymes has been involved in pre-conditioning protection against ischaemia-reperfusion injury. 42 The KAI-9803 peptide (delcasertib) inhibits δ-PKC activity and prevents translocation of δ-PKC to the mitochondria during prolonged ischaemia reperfusion. Whereas administration of KAI-9803 may preserve mitochondrial function, it has no direct action on MPTP opening, but would rather prevent apoptosis by limiting the accumulation and dephosphorylation of the pro-apoptotic Bcl-2-associated death promoter. 43 In the in vivo pig model, intracoronary administration of KAI-9803 immediately prior to reperfusion reduced infarct size and improved contractile function recovery. 44

| TRO 40303
TRO 40303 was initially presented as an inhibitor of MPTP opening. 10 In vitro experiments suggested that this compound might reduce oxidative stress and subsequently prevent opening of the MPTP. Experimental evidence suggests that TRO40303 acts through its binding to the translocator protein (TSPO) located in the outer mitochondrial membrane (Figure 1). Importantly, Sileikyte et al demonstrated using calcium retention capacity measurements in isolated mitochondria that TRO40303 has no direct effect on the MPTP. 45 However, it is important to emphasize that no targets and mechanisms for TRO40303 have been defined. Intravenous administration of 2.5 mg/kg of TRO40303 immediately prior to reperfusion reduced infarct size by 38% in the in vivo rat model of ischaemia reperfusion. 10 However, in saline-perfused rat hearts and in anaesthetized pigs, TRO40303 did not reduce infarct size when given at reperfusion. 46

| MTP131
MTP-131 (Bendavia) may reduce infarct size, when administered at the time of reflow in various animal models. In isolated cardiac mitochondria, Zhao et al suggested that MTP-131 was able to limit ROS production. 47 Alternatively, an enhanced ROS scavenger capacity of MTP-131 during ischaemia reperfusion has been discussed, but no specific target or mechanisms have been defined.
In a rat model of acute kidney injury, MTP-131 binds to cardiolipin, prevents its peroxidation by cytochrome c, thereby protecting mitochondrial cristae membranes during renal ischaemia reperfusion. 48 Conflicting results regarding infarct size reduction have been reported suggesting a not so clear putative cardioprotection potential. 49

| CLINIC AL S TUD IE S OF DRUG S TARG E TING MITOCHONDRIAL FUN C TION A S AN ADJ UN C T TO REPERFUS I ON IN S T-S EG MENT ELE VATI ON MYO C ARD IAL INFARC TION
An overview of clinical studies of mitoprotective drugs is given in the Table 1.

| Cyclosporine A
The first seminal clinical proof of concept by Piot et al demonstrated that CsA 2.5 mg/kg iv <10 minutes before primary PCI with direct stenting yielded a 40% reduction in creatine kinase (CK) release over 72 hours in 58 patients with reperfused STEMI. 50 The reduction on infarct size persisted at 6 months and was associated with less detrimental remodelling. 51 In a subgroup of 27 patients, the absolute mass of the area of hyperenhancement on cardiac magnetic resonance imaging (MRI) was significantly reduced in the CsA group corresponding to a reduction in infarct size from 18% to 15% of the left ventricle. 50 In the subsequent CIRCUS phase III trial, no benefit in clinical outcome (composite of all-cause mortality, worsening of heart failure during the initial hospitalization, hospitalization for heart failure and adverse left ventricular [LV] remodelling) was demonstrated within 1 year after reperfused acute anterior STEMI by 2.5 mg/kg CsA intravenously before primary PCI in 971 patients. 12 Because the study demon- trial. 13 A single intravenous CsA bolus (2.5 mg/kg) before primary PCI had no effect on ST-segment resolution or high-sensitive cardiac troponin T, and it did not improve clinical outcome or LV remodelling up to 6 months. 13 In patients undergoing reperfusion therapy with thrombolysis, CsA did not have any effect on troponin T and CK myocardial band (CK-MB) release or percentage ST resolution at 90 minutes after treatment. 11 A more recent meta-analysis did not demonstrate any significant differences between the CsA and placebo in terms of all-cause death (OR: 1.21, 95% CI: 0.78-1.87) and cardiovascular death (OR: 1.05, 95% CI: 0.66-2.49). 54 F I G U R E 1 Cyclosporine A and TRO40303 inhibit opening of mitochondrial permeability transition pores (MPTP). Proteins implicated in MPTP formation include the matrix cyclophilin D (CyD), the inner membrane (IMM) and the outer mitochondrial membrane (OMM). Additional proteins such as the translocator protein 18 kDa (TSPO), located in the OMM, interact with proteins implicated in MPTP formation. Under pathophysiological conditions, such as high Ca 2+ concentration and increased oxidative stress, the complex forms an open pore between the inner and outer membranes that ultimately result in mitochondrial swelling, mitochondrial Ca 2+ efflux and the release of apoptogenic proteins. Cyclosporine A targets matrix CyD, where Ca 2+ overload triggers MPTP opening. TRO40303 binds to TSPO in the outer membrane Whereas the lack of translation may reflect that improvement in treatment and outcome has diminished ischaemia reperfusion as a target for protection, it was surprising in both studies that it was not possible to confirm the immediate reduction in biochemical myocardial necrosis marker release observed in the original proof-of-concept study by Piot et al. 50 However, this finding is not unique in trials evaluating cardioprotection in humans. The CONDI-1 study 21   our focus on infarct size reduction to improve clinical outcome may have been too narrow. 58 We must pay more attention to also reduce coronary microvascular obstruction. [59][60][61] In fact, in the NHLBI-sponsored trial on ischaemic post-conditioning, coronary microvascular obstruction on MRI was reduced along with better clinical outcome. 57 Pharmacological agents that specifically target microvascular obstruction would be of interest, administered alone or in association with infarct size reducing agents. The so far mostly disappointing data do not exclude that ischaemia-reperfusion injury can be a target for modification. We must possibly address more targets to translate protection into a clinical benefit.
As for mitochondria, we must also direct our attention not only to cardiomyocyte, but also to endothelial fibroblast, and smooth muscle cell mitochondria. 62 Modern reperfusion therapy also includes treatment with platelet inhibiting P 2Y12 antagonists, which possess inherent cardioprotective capacity 63,64 such that the efficacy of additional cardioprotection may be increasingly difficult to demonstrate. 65 In the CIRCUS trial, more than 90% of the patients were treated with P 2Y12 antagonists. 12 During Noteworthy, there has been no evidence from the above-mentioned trials that in these clinical conditions of STEMI, CsA had any detectable biological effect, thereby questioning our ability to deliver it in due time at the appropriate dose to the right molecular target.
Although the magnitude of reperfusion injury may be too small for a modification that translates into a clinically important benefit, the emerging evidence that CsA may not have effect on infarct size leaves the role of the MPTP in reperfusion injury in humans unanswered because no MPTP inhibitors are available. As long as its molecular structure remains unknown and specific inhibitors of its opening are lacking, MPTP remains a candidate to alleviate reperfusion injury.

| KAI-9803
The selective inhibitor of delta-protein kinase C (delta-PKC), KAI-9803, was first studied in a phase I dose-escalation safety study of 154 patients with acute STEMI (DELTA-MI trial). 15

| TRO 40303
The single clinical study of TRO 40303 was the multi-centre to area at risk did not decrease with the highest TRO40303 dose but increased significantly in the vehicle group-not because of a change in infarct size but because of a change in area at risk.
As a consequence, the relative reduction in infarct size compared with vehicle did not seem convincing. Concentrations in rats were lower than those achieved in humans. In saline-perfused rat hearts and in anaesthetized pigs, TRO40303 did not reduce infarct size when given at reperfusion. 46 Circulating levels of the drug were not measured in the MITOCARE study, and adjudicated side effects were frequent in the treatment group perhaps indicating that the dose might not have been optimal. However, the lack of effect and demonstration of side effects have prevented further testing of TRO40303.

| CHALLENG E S IN TR AN S L ATING MITOPROTEC TI ON INTO THE CLINIC AL S E T TING
The lack of clinical efficacy by agents that target mitochondria to confer cardioprotection might be understood in terms of redundancy of mitochondrial pathways, insufficient pathophysiological knowledge on the role of mitochondria in cell death/survival pathways, limited information about their cellular origin (cardiomyocyte vs coronary vascular), their structures and functions, and the complex interplay of the conditions closely related to cardiovascular disease that affects mitochondria, such as ageing, comorbidities and co-medication. 69-71

| Ageing
As lifespan has extended, age-related conditions accumulate in elder people. Some of these conditions disrupt mitochondrial functions.
Ageing by itself induces a progressive impairment of mitochondrial respiratory efficiency as well as changes in mitochondrial morphology and mitochondrial pool 72 that collectively account for the exercise intolerance seen in elderly patients. 73 The aerobic capacity of the human myocardium is depressed in elderly people (>75 years) because of an excessive mitochondrial calcium accumulation that eventually leads to less number of respiring mitochondria. 74 This mechanism may facilitate the transition from a healthy to a failing cardiomyocyte and could underlie the reduced tolerance to ischaemic damage observed in elderly patients 74 and aged animals. 75 The interfibrillar mitochondria are particularly sensitive to the effects of ageing, whereas subsarcolemmal mitochondria remain rather preserved. 76,77 During ageing, mitochondria also experience structural changes, including reduced inner membrane surface and disarrangement of cristae. 78 Furthermore, impairment of the electron transfer chain complexes III and IV may be the cause of the disorganized 'respirosomes' observed in the aged heart. 79 Other functional alterations of mitochondria of aged cardiac cells are excessive ROS production that favours mitochondrial DNA (mtDNA) damage, 80 diminished mitophagy 81 and an increased susceptibility to undergo MPTP opening upon reperfusion. 75 Telomerase is not only present in the nucleus, but also in mitochondria, and telomerase abundance typically decreases with ageing and pre-disposes to myocardial ischaemiareperfusion injury. 82 Telomerase-deficient rat hearts have increased infarct size. 83 As a result of the wide range of altered mitochondrial mechanisms because of the natural process of ageing, pharmacological agents aimed at promoting mitoprotection might fail.

| Sex differences
Experimental observations have confirmed the results of epidemiological studies investigating sex-specific differences in cardiac tolerance to ischaemia. 84 Female gender appears to favourably influence cardiac remodelling after ischaemia-reperfusion injury. Detailed mechanisms of sex-related differences remain unknown and may involve genomic and non-genomic effects of sex steroid hormones, particularly the oestrogens, which have been the most extensively studied, but also by influences during the early-phases of ontogenetic development.
Experimental studies of the mitochondrial proteome have identified a number of mitochondrial proteins that have male/female differences in post-translational modification. 85 Specifically, males have increased phosphorylation of the pyruvate dehydrogenase (PDH)-E1α subunit, whereas females have increased phosphorylation of mitochondrial aldehyde dehydrogenase-2 (ALDH2), an enzyme involved in post-ischaemic reperfusion and remote conditioning pathways, 86,87 and the E2 subunit of α-ketoglutarate dehydrogenase. 85 Therefore, females exhibited reduced ROS generation on reoxygenation. Similar to CsA, oestrogen may contribute to maintaining mitochondrial function during ischaemia reperfusion by stabilizing the mitochondrial membrane potential and inhibiting MPTP opening. 88 Also, the protection during conditions of increased contractility seems to involve an increase in nitric oxide signalling that leads to S-nitrosylation of the L-type calcium channel. The nitrosylation reduces calcium loading during ischaemia and early reperfusion, and hence moderates ischaemia-reperfusion injury. 89 Hence, sex-related differences in cardiac sensitivity to ischaemia-reperfusion injury may influence mitoprotective strategies in patients with acute coronary syndrome.

| Comorbidity and risk factors
Diabetes seems to attenuate the cardioprotective effects of pharmacological and ischaemic conditioning manoeuvers. 90 This may partly be because of dysregulation of mitochondrial homeostasis, by impairing autophagy, overproduction of ROS, lipotoxicity and activation of calpain that has been linked to F-ATPase proteolysis. 91,92 After an AMI, cardiomyocytes of type 2 diabetes mellitus rats present a significant down-regulation of genes involved in mitochondrial fusion and autophagy when compared with non-diabetic animals, manifesting as accumulation of incompletely degraded mitochondria and enhanced apoptosis of cardiomyocytes. These findings correlate with a severe form of heart failure and increased mortality among diabetic rats. 93 Accordingly, cardiac cells of type 2 diabetic mice exhibit degenerated swollen mitochondria with disintegrated cristae and sparse autophagosomes, linked with the reduction in the phosphorylated state of the 5' adenosine monophosphate-activated protein kinase (AMPK). 94 In non-diabetic animals, two-week administration of resveratrol reverses heart remodelling after myocardial infarction via up-regulated autophagy mediated by AMPK activation. 95 Additionally, resveratrol improved LV diastolic and endothelial function in patients with a history of myocardial infarction, among whom nearly one-third were diabetic. 96 To what extent these last findings are generally applicable to diabetic patients remains to be proven. Loss of cardiomyocyte-restricted insulin signalling decreases the mitochondrial capacity of fatty acid catabolism. 97 Dyslipidaemia affects the myocardium directly, obscuring and preventing cardioprotection provided by ischaemic conditioning and pharmacological agents. 98 This might be consequence of MPTP opening secondary to decreased levels of antioxidant enzyme expression 99 or to reduced glycogen synthase kinase (GSK)-3β phosphorylation. 100 The MPTP opening inhibitor CsA cannot by its own offer cardioprotection in hypercholesterolaemic animals. 101 Another target of mitoprotection is the mitochondrial potassium-ATP (K ATP ) channel located at the inner membrane. In hyperlipidaemic rats, neither cromakalim nor diazoxide, two K ATP channel activators, maintain their cardioprotective effects in the context of ischaemia reperfusion. 102

| Co-medication
At the same time as comorbidities may influence the efficacy of mitoprotective interventions, drugs used as treatments may alter mitochondrial processes by interference with the mitochondrial respiratory chain (eg uncoupling) or inhibition of important mitochondrial processes including oxidative phosphorylation, mitochondrial DNA replication and ADP/ATP translocation.

| Antidiabetic drugs
The antidiabetic compounds glibenclamide and other sulphonylureas promote MPTP opening, causing calcium efflux to the cytosol and resulting in swollen mitochondria. 103

| Lipid-lowering drugs
Statins inhibit mitochondrial respiration, stimulate ROS production and may facilitate MPTP opening. 110,111 A proprotein convertase subtilisin/kexin 9 (PCSK9) inhibitor as well as high doses of atorvastatin lowers the ratio between phosphorylated and total Drp-1, a key factor in mitochondrial fission, whereas only the PCSK9 inhibitor restores mitochondrial ROS levels in insulin-resistant, dyslipidaemic rats. 112 Expression of PCSK9 is increased in the viable reperfused myocardium and its pharmacological inhibition can reduce infarct size, possible through autophagy, in the in vivo mouse model of myocardial infarction. 113 Paradoxically, although some statins such as pravastatin seem to facilitate MPTP opening, atorvastatin reduces infarct size in the isolated mouse-perfused hearts by activating the Akt-eNOS pathway, 114 suggesting that MTPT modulation may vary between statins. Mitochondrial damage is not restricted to cardiomyocytes, but it is also encountered in other cell types. In pigs with metabolic syndrome, mitochondrial density is diminished in LV endothelial cells, and accordingly, endothelium-dependent vasorelaxation of coronary arteries is compromised. Of note, regular exercise seems to exert mitoprotective effects even in dyslipidaemic conditions. 99

| Chemotherapeutics
Widely prescribed chemotherapeutics display a plethora of adverse side effects mediated by mitochondria damage with cardiotoxicity as the most feared. 115,116 This is the case for the inhibitor of topoisomerase II, doxorubicin and other anthracyclines, which stand as the main cause of chemotherapy-related cardiotoxicity. 117 Doxorubicin diverts electrons from respiratory chain complex I and other dehydrogenases that induce ROS overproduction. 118 Despite its intended effect is preventing DNA replication at the nucleus of cancer cells, doxorubicin also reduces mtDNA levels and thus mitochondrial biogenesis in all tissues, including the myocardium. 119 Indirectly, doxorubicin triggers the irreversible accumulation of mtDNA adducts through ROS production in a cardioselective manner. 120 The modes of action of mitoxantrone and doxorubicin present a high degree of similarity. 121 Mitoxantrone hampers oxidative phosphorylation, whereas its associated ROS production is relatively limited and secondary to ATP depletion. 122 Nevertheless, it mitoxantrone inhibits the mitochondrial calcium uniporter by a direct interaction. 123 The tyrosine kinase inhibitors, trastuzumab, is an antibody directed against the human epidermal growth factor receptor 2 that has as an adverse side effect of the heightened risk of congestive heart failure. 124 Trastuzumab cardiomyopathy could be because of the disruption of critical signalling that sustains cardiomyocyte survival, 125 but mitochondrial dysfunction and ROS imbalance cannot be dismissed. 126 Other anticancer drugs, including tamoxifen, flutamide, alkylating agents and taxanes, produce considerable alterations in essential mitochondrial functions. On the other hand, anti-proliferative drugs such as rapamycin/sirolimus that are used in drug-eluting stents after revascularization could owe their success partially to induced mitophagy and reduced apoptosis after ischaemia reperfusion, as observed in other tissues. 127 Despite their modulatory effect on mitochondrial function, the interaction between chemotherapeutics and specific mitoprotective strategies in the clinical setting remains unknown.

| Non-steroidal anti-inflammatory drugs
Non-steroidal anti-inflammatory drugs (NSAIDs) such as naproxen, diclofenac and celecoxib have been associated with an increased risk of cardiovascular disease, predominantly coronary thrombosis, because of their variable affinity to the cyclooxygenase 1 and cyclooxygenase 2 (COX-1 and COX-2) enzymes that may alter thrombogenicity. 128 Recent experimental studies on isolated rat heart mitochondria have demonstrated that NSAIDs may also increase ROS formation, mitochondria membrane collapse, mitochondria swelling, lipid peroxidation, and glutathione and ATP depletion, 129 which may play important roles in developing cardiotoxicity. MPTP sealing agents and antioxidants may prevent mitochondrial toxicity. 129 However, the clinical implications of these observations remain unknown.

| PER S PEC TIVE AND CON CLUS I ON
Despite pre-clinical evidence for a cardioprotective effect, most firmly established for CsA, neither of the clinically tested mitoprotective drugs has demonstrated protective capability on clinical outcome beyond that provided by rapid revascularization alone.
Whereas it may relate to favourable outcome with current optimal therapy, risk factors, comorbidity and concurrent medications may also alter mitochondrial function, sometimes in an irreversible way and often by multiple mechanisms. In consequence, targeting one single mechanism to provide mitoprotection may be ineffectual.
However, as long as the molecular structure of the MPTP remains unknown and specific inhibitors of its opening are lacking, it remains a candidate to alleviate reperfusion injury. Nevertheless, taking conditions such as ageing, sex, comorbidities and co-medication into account may be of paramount importance during the design of pre-clinical and clinical studies testing mitoprotective drugs.

CO N FLI C T O F I NTE R E S T
None.

AUTH O R CO NTR I B UTI O N S
All authors contributed equally to the manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data sharing is not applicable to this article as no new data were created or analysed in this study.