Targeting mitochondrial fusion and fission proteins for cardioprotection

Abstract New treatments are needed to protect the myocardium against the detrimental effects of acute ischaemia/reperfusion (IR) injury following an acute myocardial infarction (AMI), in order to limit myocardial infarct (MI) size, preserve cardiac function and prevent the onset of heart failure (HF). Given the critical role of mitochondria in energy production for cardiac contractile function, prevention of mitochondrial dysfunction during acute myocardial IRI may provide novel cardioprotective strategies. In this regard, the mitochondrial fusion and fissions proteins, which regulate changes in mitochondrial morphology, are known to impact on mitochondrial quality control by modulating mitochondrial biogenesis, mitophagy and the mitochondrial unfolded protein response. In this article, we review how targeting these inter‐related processes may provide novel treatment targets and new therapeutic strategies for reducing MI size, preventing the onset of HF following AMI.


| INTRODUC TI ON
Acute myocardial infarction (AMI) and the heart failure (HF) that can follow are among the leading causes of death and disability worldwide. Although mortality following AMI is on the decline, the prevalence and severity of HF is rising. Therefore, new treatments are required to protect the myocardium against the detrimental effects of acute ischaemia/reperfusion (IR) injury in order to reduce myocardial infarct (MI) size, preserve left ventricular (LV) function and prevent the onset of HF. 1 Maintenance of healthy mitochondria is of critical importance, given the high energy demands required for normal cardiac contractile function. 2 Under conditions of energy stress such as experienced during acute myocardial ischaemia/reperfusion injury (IRI) following AMI, damaged mitochondria generate less ATP and produce reactive oxygen species (ROS) that are detrimental to cell survival. As such, new therapies that are able to preserve mitochondrial function during acute myocardial IRI may provide novel strategies for cardioprotection. 2 The mitochondrial fusion and fission proteins have been shown to play critical roles in several processes related to mitochondrial quality control including mitochondrial biogenesis, mitophagy and the unfolded protein response (UPR). Mitochondrial fission is required to selectively remove damaged mitochondria by mitophagy (which are then replaced by mitochondrial biogenesis), and mitochondrial fusion allows fragmented mitochondria which are still viable to re-enter the mitochondrial network (reviewed in . The mitochondrial UPR is a cytoprotective signalling pathway triggered by the mitochondrial accumulation of toxic unfolded proteins under conditions of cellular stress. 5 In this article, we review how targeting these inter-related processes related to mitochondrial quality control may provide novel treatment targets and new therapeutic strategies for reducing MI size and preventing HF following AMI.

| MITOCHONDRIAL MORPHOLOGY
Mitochondria are dynamic organelles, constantly changing their morphology or shape between a fragmented disconnected phenotype by undergoing fission, and an elongated interconnected morphology by undergoing fusion, processes which are coordinated by the mitochondrial fission and fusion proteins, respectively (Figure 1) (reviewed in . Mitochondrial fission is essential for cell division and is required to remove damaged mitochondria by mitophagy. In contrast, mitochondrial fusion enables the replenishment of damaged mitochondrial DNA and facilitates intracellular energy distribution. A brief overview of these mitochondrial shaping proteins is given here, with the main focus being on their roles in acute myocardial IRI, and as potential therapeutic targets for acute cardioprotection. Mitochondrial fission involves the division of a single mitochondrion into 2 individual fragmented and disconnected mitochondria by the fission proteins, dynamin-related protein 1 (Drp1), 6 human fission protein 1 (hFis1), 7 mitochondrial fission factor (Mff) 8 19 sepsis-related cardiomyopathy, 20 post-AMI cardiomyopathy 21 and diabetic cardiomyopathy. 22 In this article, we focus on the role of the mitochondrial fission proteins as potential therapeutic targets for cardioprotection against acute myocardial IRI.
Mitochondrial fusion is characterized by the tethering of two adjacent mitochondria via the OMM fusion GTPase proteins, mitofusin 1 (Mfn1) 23 and mitofusin 2 (Mfn2), 24 which then mediate fusion of the OMMs in a GTP-dependent manner. Subsequently, the inner mitochondrial membrane (IMM) fusion GTPase protein, optic atrophy 1 (OPA1), 25,26 mediates fusion of the IMMs resulting in sharing of mitochondrial matrix material, and the formation of a single elongated mitochondrion. Importantly, Mfn2 has been reported to have a number of non-fusion-related functions including tethering ER to mitochondria, 27 mediating mitophagy 28 and cellular autophagy, 29 and contributing to the stress-related UPR 30 (Figures 2 and   3). Similarly, OPA1 has been demonstrated to play several pleiotropic non-fusion roles related to its effects on cristae remodelling the consequences of which are to regulate mitochondrial cytochrome release 31 and to improve mitochondrial respiratory efficiency by facilitating the assembly of electron transport supercomplexes. 32 In addition, OPA1 has been shown to stabilize mitochondrial cristae shape and favour ATP synthase oligomerization and needs ATP synthase to protect mitochondria from stress-induced respiratory chain inhibition. 33 Differentiating the pro-fusion roles of Mfn2 and OPA1 from their non-fusion roles can be quite challenging and needs to be taken into consideration when investigating their roles in the context of acute myocardial IRI and cardioprotection, where both their pro-fusion and non-fusion roles may be important.  37 In this article, we focus on the role of the mitochondrial fusion proteins as potential therapeutic targets for cardioprotection against acute myocardial IRI.

F I G U R E 1
Mitochondrial fission and fusion proteins as targets for cardioprotection. Mitochondrial fission induced by acute myocardial ischaemia/reperfusion injury can be targeted by mdivi-1 and Drpitor, pharmacological inhibitors of Drp1 and P110, a peptide inhibitor of the interaction between Drp1 and hFis1 to confer cardioprotection. A number of other factors (such as SB203580, PKA activators, succinate, erythropoietin and melatonin) have also been shown to confer cardioprotection by targeting the fusion and fission proteins. Inset upper box: This scheme shows the mitochondrial fission machinery, comprising Drp1 and its outer mitochondrial membrane receptors, MiD49/MiD51, Mff and hFis1. Pre-constriction by the endoplasmic reticulum (ER) via INF2, actin and Spire1C initiates the Drp1-driven mitochondrial fission process. Inset lower box: This scheme shows the pleiotropic non-fusion effect of Mfn2 in tethering mitochondria to the ER

| Relevance of mitochondrial fission and fusion proteins to the adult heart
In adult cardiomyocytes, mitochondria are distributed in 3 main locations: intermyofibrillar spaces, where they are densely packed and mainly provide energy for contractile function (interfibrillar mitochondria), just beneath the plasma membrane, where they provide energy for the sarcolemmal ion channels (subsarcolemmal mitochondria), and around the nucleus, where they provide energy for nuclear transcription (perinuclear mitochondria). 38  opening. [45][46][47] Genetic ablation of Mff in mice has been shown to induce a lethal dilated cardiomyopathy at 13 weeks associated with increased heterogeneity in mitochondrial shape and abundance, perturbed mitochondrial respiratory function, increased myocardial apoptosis and interstitial fibrosis. 48 Interestingly, this effect was reversed if either one of the mitochondrial fusion proteins, Mfn1 or Mfn2, were also deleted suggesting that the pro-fusion effect of Mff ablation requires the presence of one of the mitofusins. This finding also supports the notion that restoring the balance in mitochondrial fission and fusion in the heart can normalize cardiac function. 48 With regard to the mitochondrial fission proteins, MiD49 and MiD51, genetic ablation of these proteins is embryonically lethal, 10 but the effect of their deletion in the adult heart is not known, and needs to be investigated.
With respect to the mitochondrial fusion proteins, cardiomyocyte-specific deletion of Mfn2 resulted in a mild cardiac phenotype with modest LVH, mild LV systolic dysfunction, no change in mitochondrial respiratory function and unexpected changes in mitochondrial morphology and MPTP opening susceptibility, with pleomorphic and enlarged subsarcolemmal (but not interfibrillar) mitochondria, and resistance to MPTP opening. 49 Interestingly, the effect of Mfn2 ablation on mitochondrial morphology in neonatal rat cardiomyocytes was as expected, with mitochondrial fragmentation and increased susceptibility to MPTP opening, suggesting F I G U R E 3 The mitochondrial unfolded protein response and cardioprotection. Schematic model of the mitochondrial unfolded protein response (UPRmt). Nicotinamide ribose, ischaemic preconditioning (IPC), oligomycin and doxycycline are therapeutic strategies that have been shown to reduce acute myocardial ischaemia/reperfusion injury (IRI) by enhancing the UPRmt, and Mfn2 has been shown to play a key role in the UPR cell-specific effects on mitochondrial morphology and MPTP opening. 49 In addition, to having enlarged mitochondria, Mfn2-deficient cardiomyocytes have also being shown by 3D electron microscopy to have fewer mitochondria-junctional sarcoplasmic reticulum (SR) contacts, and an increase in the distance between mitochondria and junctional SR, when compared to wild-type cardiomyocytes. 50 Overexpression of cardiomyocyte-specific Mfn2 in the adult heart had a very mild phenotype with minor mitochondrial enlargement with no detrimental effects on either mitochondrial respiratory function or LV size or function. 47 Interestingly, genetic ablation of cardiomyocyte-specific Mfn1 did result in smaller more fragmented mitochondria, but had no effect on cardiac function or mitochondrial respiratory function, although it did prolong time to induce ROSmediated MPTP opening. 51 The differential effects of Mfn1 versus Mfn2 ablation on mitochondrial size may relate to the pleiotropic non-fusion effects of Mfn2 such as acting as a tether to the SR. 27 The mild phenotype observed in hearts deficient in either cardiomyocyte Mfn1 or Mfn2 suggests that these proteins have a redundant function. Genetic ablation of both Mfn1 and Mfn2 mid-gestation resulted in a severe cardiomyopathy at day 7 of post-natal period suggesting that the mitofusins are needed for the mitochondrial remodelling which occurs in the first week following birth. 52 Cardiomyocytespecific deletion of both Mfn1 and Mfn2 in the adult heart resulted in mitochondrial fragmentation, impaired mitochondrial respiration, a mitochondrial UFR and a dilated cardiomyopathy. 39,46,50,52 Although genetic deletion of OPA1 is embryonically lethal, mice with heterozygous ablation of OPA1 survive and have enlarged mitochondria with disorganized cristae, increased time to induce MPTP opening, but had no effect on cardiac function or mitochondrial respiration, although the hearts were more susceptible to LVH induced by total aortic constriction. 53,54 Heterozygous OPA1 mice did however develop a late-onset cardiomyopathy associated with fragmented mitochondria with disorganized cristae, impaired mitochondrial respiratory function and increased mitochondrial ROS. 55 Genetic mouse models with deletions of OMA1 and/or YME1L (the proteases responsible for cleaving L-OPA1 to S-OPA1) have been used to investigate the effect of modulating myocardial OPA1 levels on mitochondrial morphology, mitochondrial respiratory function and cardiac function. 56 Genetic deletion of Yme1l in adult cardiomyocytes was shown to induce mitochondrial fragmentation (due to OMA1 cleaving L-OPA1 to S-OPA1), but did not affect mitochondrial respiration, and resulted in dilated cardiomyopathy at 20 weeks. 56 Interestingly, additional deletion of OMA1 reversed the detrimental effects of YME1L deletion by restoring OPA1 levels. 56 These studies support an important role for OPA1 in normal cardiac function.
It is interesting and surprising to note that genetic ablation of both the mitochondrial fission protein (Drp1), 16   respectively. 66 Mice deficient in NR4A1 were demonstrated to be protected against acute coronary microvascular injury and mitochondrial dysfunction, when compared with wild-type mice. 66 In summary, a number of different factors have been shown to contribute to the observed mitochondrial fission which occurs in response to acute myocardial IRI, thereby providing multiple therapeutic targets for inhibiting mitochondrial fission as a cardioprotective strategy.

| Therapeutic targeting of the mitochondrial fission proteins for cardioprotection
It has been shown in HL-1 cardiac cells that transfection with Drp1 dominant-negative mutant-induced mitochondrial elongation prevented MPTP opening and reduced cell death following simulated IRI, 16 demonstrating Drp-1 mediated mitochondrial fission to be a therapeutic target for cardioprotection. In this regard, it has been shown that acute inhibition of mitochondrial fission using the putative Drp1 GTPase inhibitor, mitochondrial division inhibitor 1 (mdivi-1), reduced cell death in isolated adult murine cardiomyocytes subjected to simulated IRI and reduced MI size in murine hearts subjected to in vivo acute myocardial IRI. 16 Mdivi-1 has also been shown to inhibit IR-induced mitochondrial fragmentation in a diabetic mouse model of acute myocardial IRI 67 and inhibited mitochondrial fission and restored the cardioprotective effects of sevoflurane under conditions of high glucose in neonatal rat cardiomyocytes. 68 It has also been demonstrated that mdivi-1 reduced cell death in human W8B2 + cardiac stem cells subjected to simulated IRI, although in that study no beneficial effects were observed on mitochondrial morphology so the mechanisms underlying the cardioprotective effect in these cells are unclear. 69 Other pharmacological inhibitors of mitochondrial fission have also been shown to be cardioprotective including Dynasore (a non-specific inhibitor of dynamins), 70 and P110, a peptide inhibitor that inhibits the interaction between Drp1 and hFis. 71 Importantly, pharmacological inhibition of mitochondrial fission at the onset of reperfusion, a clinically relevant time-point, using either P110 71 or mdivi-1 72 was shown to reduce MI size in small animal models of acute myocardial IRI. In Table 1, we present a summary of the main experimental studies investigating cardioprotection with pharmacological agents targeting mitochondrial fission proteins.
Recent studies suggest that mdivi-1 has off-target effects that are independent of its inhibitory effects on Drp1 GTPase activity, TA B L E 1 Summary of major studies investigating cardioprotection with pharmacological agents targeting mitochondrial fission proteins Melatonin, a well-established cardioprotective factor, has been shown to reduce coronary microvascular injury, as evidenced by  In summary, these studies implicate IR-induced mitochondrial fission to be a critical mediator of cardiomyocyte death following AMI that can be targeted indirectly via a wide variety of cardioprotective agents.

| Targeting mitochondrial fusion proteins for cardioprotection
Given the data supporting mitochondrial fission to be a key deter-  In contrast, it has been shown in the adult mouse that genetic overexpression of OPA1 stabilized mitochondrial cristae thereby improving mitochondrial respiratory efficiency, preventing mitochondrial dysfunction, attenuating cytochrome c release and ROS production, and through these actions, it reduced susceptibility to acute myocardial IRI. 106 However, the effect of OPA1 overexpression on preventing IR-induced mitochondrial fragmentation was not assessed, as the cardioprotective effects of OPA1 were attributed to its non-fusion effects on mitochondrial cristae remodelling and improving mitochondrial respiratory efficiency.
Importantly, in this mouse model of OPA1 overexpression, there were beneficial effects in other organs with protection from muscular atrophy, brain ischaemia and liver apoptosis. 106 Interestingly, the role of OPA1 as a cardioprotective target has been apoptosis. 110 Similarly, the brain-derived neurotrophic factor mimetic, 7,8-dihydroxyflavone, has been reported to confer cardioprotection via inhibition of IR-induced mitochondrial fission and decreased apoptosis. 111 Finally, modulation of OPA1 levels by genetic ablation of OMA1 has also been shown to protect against acute renal IR acute injury. 112 In terms of therapeutic targeting of OPA1 as a cardioprotective strategy, a recent study has identified epigallocatechin gallate to be a novel pharmacological inhibitor of OMA1. Treatment of mouse embryonic fibroblasts was able to prevent cleavage of L-OPA1 to S-OPA1, inhibit mitochondrial fission, prevent apoptosis and reduce cell death following simulated IRI, providing a potential therapeutic strategy for reducing MI size following acute myocardial IRI.
In summary, these studies implicate the mitochondrial fusion proteins, Mfn2 and OPA1, to be potential targets for cardioprotection, but the mechanisms underlying their beneficial effects appear to be related to their non-fusion pleiotropic functions rather than their pro-fusion effects.

| TARG E TING MITOPHAGY FOR C ARDIOPROTEC TION
Under basal conditions, mitophagy is required for mitochondrial quality control and plays a key role in removal of damaged mitochondria. 113  In summary, enhancing mitophagy during acute myocardial IRI is an important strategy for cardioprotection. Given the close interplay between mitochondrial morphology and mitophagy, therapeutic targeting of the mitochondrial fission and fusion proteins may provide a strategy for boosting mitophagy following AMI.

| TARG E TING THE MITOCHONDRIAL UNFOLDED PROTEIN RE S P ON S E FOR C ARDIOPROTEC TION
The mitochondrial UPR (UPR mt ) is a cytoprotective signalling pathway triggered by the mitochondrial accumulation of toxic unfolded proteins under conditions of cellular stress that induces mitochondrial dysfunction (Figure 3). The UPR mt , in turn, up-regulates cyclic AMP-dependent transcription factor ATF-5 (ATF5) to restore mitochondrial proteostasis and respiratory function. 131  It has been shown in mice hearts subjected to pressure-overload hypertrophy and in hypertrophied human heart tissue due to aortic stenosis, that the UPR mt is increased as evidenced by up-regulation of ATF5 and the C/EBP transcription factor CCATenhancer-binding protein homologous protein (CHOP), which encodes adaptive proteins of the UPR mt . 134 Importantly, from a therapeutic perspective, it was reported that the UPR mt could be enhanced using nicotinamide riboside (which augments NAD + pools) in cardiomyocytes and prevented cardiomyocyte death and preserved mitochondrial respiratory function and cardiac contractile function. 134 Pharmacological induction of the UPR mt using either oligomycin or doxycycline reduced MI size in wildtype mice but not ATF5-deficient mice, suggesting that ATF5mediated UPR mt confers cardioprotection against acute IRI. 5 The mitochondrial protease, LonP1, which ensures mitochondrial proteostasis and regulates adaptive responses to cell stress, has been shown to contribute to the cardioprotection elicited by ischaemic preconditioning (IPC), an endogenous cardioprotective intervention in which brief cycles of non-lethal ischaemia and reperfusion protect the heart against an episode of acute lethal IRI. 135 IPC was shown to up-regulate LONP1 in wild-type mice, and heterozygous LONP1± mice were shown to be more susceptible to acute myocardial IRI as evidenced by mitochondrial dysfunction and increased MI size, and were found to be resistant to IPC-induced cardioprotection. 136 In contrast, genetic overexpression of cardiomyocyte-specific LONP1 prevented IR-induced mitochondrial dysfunction and reduced MI size, implicating mitochondrial LONP1 as an endogenous mediator of cardioprotection. 136 In summary, enhancing the UPR mt during acute myocardial IRI has emerged as a novel therapeutic cardioprotective strategy.
Although pharmacological strategies such as nicotinamide riboside, oligomycin and doxycycline have been shown to confer cardioprotection by enhancing the UPR mt , novel more specific activators of the UPR mt are needed.

| CON CLUS IONS
Preventing mitochondrial dysfunction during acute myocardial IRI following AMI is a major therapeutic strategy for cardioprotection in terms of reducing MI size, preserving cardiac function and preventing the onset of HF. In this regard, the mitochondrial fission and fusion proteins which play key roles in process involved in mitochondrial quality control (such as mitochondrial morphology, mitophagy and UPR mt ), act to preserve normal mitochondrial respiratory function in the setting of acute myocardial IRI, positioning them as key targets of cardioprotection. Substantial experimental data suggest that acute inhibition of mitochondrial fission can reduce MI size and preserve cardiac function in small animal models of AMI. Further studies are needed to test this therapeutic approach in clinically relevant large animal AMI models and to discover novel more specific inhibitors of the mitochondrial fission machinery. However, it is important to note that chronic inhibition of mitochondrial fission is detrimental to both susceptibility to acute IRI and cardiac function, as it suppresses mitophagy and results in the accumulation of damaged mitochondria. Although the mitochondrial fusion protein, OPA1, has been shown to be a target for acute cardioprotection (with its overexpression protecting against acute myocardial IRI), its beneficial effects have been attributed to non-fusion pleiotropic roles rather than its effects on mitochondrial morphology. The role of the mitofusins as targets for acute cardioprotection has been unexpected, as it appears that ablation of Mfn1 and/or Mfn2 confers a protective response against acute myocardial IRI, and again this may be due to their non-fusion pleiotropic roles (such as tethering SR), rather than their effects on mitochondrial morphology. Overall, targeting the mitochondrial fusion and fission proteins have emerged as important targets for cardioprotection and have the therapeutic potential to reduce MI size, preserve cardiac function and prevent HF in patients presenting with AMI.

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.

AUTH O R CO NTR I B UTI O N
SRH, HG, FP, DJH, GD, all contributed to the design, writing and critical reading of the paper.