Relationship between ferroptosis and mitophagy in cardiac ischemia reperfusion injury: a mini-review

MIRI and mitophagy

Because of the heart’s high demand for energy, normal mitochondrial function is essential for heart development (Pohjoismaki & Goffart, 2017; Zhang et al., 2020a). MIRI is often accompanied by mitochondrial damage and dysfunction in cardiomyocytes. Assuring myocardial cells have enough mitochondria to fulfill their physiological needs, a variety of quality control mechanisms have evolved in mitochondria including mitophagy, biogenesis, mitochondrial dynamics, etc. (Li et al., 2021a). In 2005, Lemasters proposed “mitophagy” firstly to emphasize the non-random nature of the mitochondrial selective autophagy process (Lemasters, 2005). Mitophagy is the main mechanism for maintaining mitochondrial homeostasis in cardiomyocytes by degrading the dysfunctional mitochondria.

The occurrence of MIRI will go through two stages: ischemia and reperfusion. In the myocardial ischemia stage, hypoxia environments caused by ischemia can affect the process of oxidative phosphorylation of mitochondria, resulting in insufficient myocardial energy supply (Killackey, Philpott & Girardin, 2020). The lack of energy activates the AMPK pathway, and then mitophagy is activated (Kim et al., 2011; Laker et al., 2017). Activated mitophagy is mainly used to degrade aging mitochondria to cope with the energy crisis. Meanwhile, ROS accumulation is gradually induced by hypoxia at this stage. As ischemia and hypoxia continue, the lack of energy leads to the inability of Ca²⁺ to be excreted by the calcium pump, resulting in the accumulation of Ca²⁺ in cardiac myocytes. In order to maintain Ca²⁺ homeostasis in cardiomyocytes, mitochondria will absorb excessive Ca²⁺ from cytoplasm, resulting in Ca²⁺ overload in mitochondria. Ca²⁺ overload (Kinnally et al., 2011) and ROS accumulation (Leucker et al., 2011; Pravdic et al., 2009) lead to the mitochondrial mPTP opening, then mitochondrial membrane potential collapse (Zorov, Juhaszova & Sollott, 2014). In the reperfusion stage, the restoration of oxygen supply leads to ROS burst, which in turn prolongs the opening time of mitochondrial mPTP, further damaging the mitochondria. The ROS released by damaged mitochondria induces more ROS generation (Zorov et al., 2000), creating a vicious cycle. At this point, mitophagy, which can reduce the production of ROS by degrading the damaged mitochondria, is very important for MIRI mitigation. It should be noted that although most studies have shown that promoting mitophagy can alleviate MIRI, some studies have also shown that excessive mitophagy also damages cardiac myocytes, and inhibition of mitophagy is required at this time (Huang et al., 2022; Wu et al., 2020).

Ferroptosis and MIRI

Ferroptosis is a new type of programmed cell death that was first discovered by Dolma in 2003 and named by Dixon in 2012 (Dolma et al., 2003; Dixon et al., 2012). Programmed cell death, such as apoptosis, necrosis, and other forms, clears out damaged or infected cells, allowing surrounding healthy cells to perform their functions better (D’Arcy, 2019). Unlike reported forms of programmed cell death, ferroptosis is an iron-dependent form of cell death that is accompanied by massive iron accumulation and lipid peroxidation (Li et al., 2020a). Nowadays, Ferroptosis has been shown to exist in the pathological process of a variety of diseases, such as in cancers (Qiu et al., 2022; Xu et al., 2019), brain diseases (Weiland et al., 2019), kidney diseases (Tang & Xiao, 2020), MIRI (Zhao et al., 2021) and other diseases. For some ferroptosis-susceptible tumors, activation of ferroptosis is a potential treatment strategy (Yu et al., 2017). But in cardiac tissue, ferroptosis which can lead to cardiomyopathy needs to be suppressed (Fang et al., 2019).

The relationship between ferroptosis and MIRI was first revealed by Gao et al. (2015). Inhibition of ferroptosis via inhibiting glutaminolysis can protect heart tissue from MIRI in vitro heart model. Inhibition of myocardial ferroptosis can also alleviate MIRI in diabetic rats by inhibiting endoplasmic reticulum stress (Li et al., 2020b). According to recent studies, ferroptosis which is iron-dependent is accompanied by lipid peroxide (LPO) accumulation (Dixon et al., 2012; Stamenkovic, Pierce & Ravandi, 2019). Ferroptosis inhibitor Fer-1 has been reported can inhibit peroxidation, and prevent the accumulation of LPO thereby inhibiting ferroptosis and subsequently alleviating MIRI (Dixon et al., 2012; Li et al., 2019; Miotto et al., 2020). Fer-1 can also protect the heart from cardiomyopathy by maintaining mitochondrial function (Fang et al., 2019). Fang et al. found another ferroptosis inhibitor Lip-1 could reduce myocardial infarct sizes and maintain mitochondrial structure and function to prevent MIRI (Feng et al., 2019). In addition, on the outer mitochondria membrane, ischemia and other pathological stimuli can be protected against by the mammalian target of rapamycin (mTOR). When mTOR is overexpressed, erastin (a ferroptosis inducer) induced cell death is inhibited, while mTOR deletion will exaggerate the cell death (Baba et al., 2018). According to these data, mitochondria are important for ferroptosis-induced cardiomyocyte death. At present, the main systems for inhibiting ferroptosis include: the Cyst (e)ine/GSH/GPX4 Axis, the NAD (P)H/FSP1/CoQ10 System, and the GCH1/BH4/DHFR System (Zheng & Conrad, 2020), but the research focus is mainly on iron homeostasis, system Xc- and GPX4 (Li et al., 2021b; Cao & Dixon, 2016).

Absorption and utilization of iron

Iron homeostasis is a vital element for fundamental biological functions of human body, accumulating evidences have shown that iron dyshomeostasis is involved in the pathogenesis of cardiovascular diseases (Wei et al., 2022). When iron homeostasis is disrupted because of iron deficiency or overload, it can lead to rapid lipid peroxidation of cells due to lack of GPX4 (Ouyang et al., 2021), resulting in cardiovascular cellular damage and accelerating the occurrence of various diseases including atherosclerosis, MIRI, coronary heart disease and so on (Gao et al., 2019a; Kobayashi et al., 2018). Until now, regulating iron acquisition, recycling, and storage is the main method of controlling system iron levels for human (Wallace, 2016), which is mainly by unidirectional recycling of iron from senescent red blood cells to the erythroid bone marrow through macrophages, the cycling of iron from hepatocytes to the blood and vice versa, and iron absorption through duodenal and upper Jejunum (Piperno, Pelucchi & Mariani, 2020).

Normally, human body absorbs iron from food or other nutritious in the type of heme and non-heme (or inorganic) forms except a small amounts of iron are lost through skin exfoliation, gastrointestinal exfoliation, and urine and bile excretion (Gulec, Anderson & Collins, 2014). After entering the body from food, iron experience various metabolic processes before it can be used (Fig. 1). Heme iron (Fe²⁺) can be absorbed directly via heme/folate transporter 1 at the apical membrane of intestinal epithelial cells (West & Oates, 2008; Zhang et al., 2019a). Non-heme iron (Fe³⁺) in food is partly reduced and dissolved by gastric acid and ascorbic acid, and the rest is reduced to Fe²⁺ by cytochrome B, which is then transported to intestinal epithelial cells by divalent metal-ion transporter 1 (DMT 1, encoded by the SLC11A2 gene) for absorption. Subsequently, after traversing the basolateral membrane via ferroportin 1, Fe²⁺ is oxidized to Fe³⁺ by Hephaestin (HEPH), and then binds with transferrin (TF) to form TF-Fe³⁺ complex for utilization by organs (Gulec, Anderson & Collins, 2014). After the TF-Fe³⁺ complex binds with the transferrin receptor (TFR) and enters the endosome through endocytosis, Fe³⁺ is released from the TF-Fe³⁺ complex, then is reduced to Fe²⁺ by STEAP3 and crosses the endosomal membrane into the cytoplasm by DMT (Sendamarai et al., 2008). The imported Fe²⁺ enters a metabolically cytosolic labile iron pool, which is used for incorporation into prosthetic groups of iron-dependent enzymes and proteins, incorporation into heme and iron-sulfur cluster biogenesis, and storage in ferritin. The excess iron is exported back to the circulation by ferroportin 1, during which Fe²⁺ is oxidized to Fe³⁺ by HEPH in plasma and recombined with TF (Manz et al., 2016).

Iron overload

Iron overload is an important factor in activating ferroptosis. Iron overload usually occurs from a genetic disease or iatrogenic (Godbold & McFarland, 2021; Murphy & Oudit, 2010; Piperno, Pelucchi & Mariani, 2020). In pathological conditions caused by some diseases, iron overload can result from increased iron intake, increased gastrointestinal absorption (Godbold & McFarland, 2021), and accumulation of non-heme iron through heme degradation (Fang et al., 2019), etc. When the body is in a pathological condition of iron overload, the capacity of plasma transferrin to bind iron is saturated, leading to the accumulation of non-transferrin bound iron (Brissot et al., 2012). The accumulation of non-transferrin bound iron in plasma accelerates the deposition of iron in tissues, particularly excitable tissues containing Ca²⁺ channels which are known to conduct Fe²⁺ into cells (Zhang et al., 2019a). Therefore, iron overload may be an important cause of MIRI-induced cardiac ferroptosis, because cardiac tissue contains high levels of functional voltage-gated Ca²⁺ channels.

Iron overload is often accompanied by the imbalance of iron storage and release in our bodies. Ferritin, including ferritin heavy chain 1, ferritin light chain, as well as TFR, is currently the focus of ferroptosis research due to its ability to store iron (Li et al., 2022a; Zhang et al., 2021b). Ferroptosis inducer RSL3 increased iron uptake by upregulating the expression of TFR, while down-regulating the expression of ferritin heavy chain 1 and ferritin light chain, reducing iron storage, leading to the release of large amounts of free iron and thus inducing ferroptosis (Yang & Stockwell, 2008). Additionally, hypoxia inducible factor 1 (HIF-1) and iron regulatory protein (IRP, also known as IREB) have also been reported can increase the expression of TFR and increase iron uptake (Cheng et al., 2015; Tacchini et al., 1999; Torti & Torti, 2013). Tang et al. (2008) further found that HIF-1 α activation not only induced TFR expression increasing, but also increased transferrin uptake and iron accumulation, exacerbated oxidative damage that increased the lipid peroxidation. Therefore, inhibition of ferritin (Torti & Torti, 2013) or ferritin deficiency (Fang et al., 2020) can induce ferroptosis.

The ferroptosis during MIRI is closely related to mitochondrial ROS. When ferritin is not expressed enough, excessive intracellular free iron will cause oxidative stress and impaired mitochondrial function in the heart, manifested by decreased mitochondrial respiration, depolarization of mitochondrial membrane potential, and mitochondrial swelling (Sumneang et al., 2020). The mitochondrial dysfunction leads to a large number of ROS production, mainly including O_2⁻ andH₂O₂, which can promote the Fenton reaction. Fenton reaction consists of three reactions, Fe²⁺ + H₂O₂ → Fe³⁺ + OH⁻ + OH•, H₂O₂ + 2Fe³⁺ → 2Fe²⁺ + O₂ + 2H⁺ and O_2⁺ Fe²⁺ → Fe³⁺ + O_2⁻. We know that H₂O₂, OH•, O_2⁻ all belong to ROS. The continuous production of ROS by Fenton reaction further damages the mitochondria and leads to more ROS generation, creating a vicious cycle. In addition, the accumulating ROS will damage cellular proteins, lipids and DNA, causing cell and tissue damage and eventually lead to ferroptosis (Valko et al., 2016). In a limited number of studies, cardiac ferroptosis has also been studied. However, the mechanistic association between cardiac ferroptosis and iron overload needs further investigation (Sumneang et al., 2020).

Lipid peroxidation

A large number of ROS produced mainly through Fenton reaction continues to participate in lipid peroxidation (Fig. 2). Lipid peroxidation has been implicated in almost all human diseases associated with oxidative stress of cause in MIRI, and it has been used to assess the degree of ferroptosis (Chen et al., 2021). Attenuating lipid peroxidation can inhibit ferroptosis and thus alleviates CVD (Bai et al., 2020; Tadokoro et al., 2020).

As a result of lipid peroxidation, polyunsaturated fatty acids (PUFA) and phosphatidylethanolamine (PE) are oxidatively decomposed. PUFA is the main component of phospholipids in cell and organelle membranes, it is also an important substrate for the synthesis of PE, the main component in the inner layer of the phospholipid bilayer. PUFA has a high affinity with ROS. ROS of hydroxyl radicals (OH•) and hydrogen peroxide (H₂O₂) first acquire hydrogen atoms in PUFA to produce Lipid ROS (L-). Then, the Lipid radicals react with the oxygen molecule to form Lipid peroxyl radicals (LOO-). Lipid peroxyl radicals extract hydrogen atoms from other PUFA to form a new LOO- and Lipid hydroperoxide (LOOH). LOO- can continuously react with PUFAs, which makes the lipid peroxidation of PUFAs have the characteristics of cascade reaction (Ma et al., 2021). PUFA, arachidonic acid and adrenal acid are synthesized into poly-unsaturated fatty acid-phosphatidyl ethanolamine (PUFA-PE). With the participation of Lipoxygenase, PUFA-PE underwent lipid peroxidation reaction in the plasma membrane and endoplasmic reticulum, and finally formed LPO (Wang & Li, 2019). The lipid peroxidation reaction of ROS with PUFA and PE destroys the fluidity and stability of cell membrane, increases the permeability of cell membrane, and eventually leads to cell death.

GPX4 and system Xc- in preventing ferroptosis

To prevent ferroptosis, oxidative stress caused by LPO needs to be inhibited, and GPX4 is the key regulator in this process (Fig. 2) (Park et al., 2019). GPX4 is a unique intracellular antioxidant enzyme that can directly reduce LPO production in cell membranes to non-toxic lipid alcohols (Imai et al., 2017; Ursini & Maiorino, 2020). Under the catalytic action of GPX4, H₂O₂ and LPO were reduced, and GSH was oxidized to disulfide-oxidized form (GSSG) (Lu, 2013). When GSH is depleted, GPX4 will be inactivated leading to LPO accumulation and ultimately ferroptosis (Xie et al., 2016; Yang et al., 2014).

GSH is composed of glycine, glutamate and cysteine. And cysteine uptake depends on the activation of cystine/glutamate reverse transport system Xc-. System Xc- is a heterodimeric cell surface amino acid antiporter composed of two subunits, a light-chain subunit SLC7A11 (xCT) and a heavy-chain subunit SLC3A2 (CD98, 4F2hc), which are linked by an extracellular covalent disulfide bond and play different roles. System Xc- imports extracellular cystine in exchange for intracellular glutamate at a ratio of 1:1 (Cao & Dixon, 2016; Liu, Zhu & Pei, 2021). Intracellular cystine is first reduced to cysteine, then cysteine and glutamate form γ-glutamylcysteine under the catalysis of glutamate-cysteine ligase. Next, GSH is formed from γ-glutamylcysteine and glycine under the catalysis of GSH synthase. Also, GSH can regulate glutamate-cysteine ligase through negative feedback (Lu, 2013).

When system Xc- dysfunction occurs, the cell redox becomes unbalanced. A study of glioma cells showed that knockdown of SLC7A11 increased ROS production and decreased glutathione production, resulting in increased cell death under oxidative and genotoxic stress, and overexpression of SLC7A11 leads to increased resistance to oxidative stress (Polewski et al., 2016). The system Xc- can be inhibited irreversibly by ferroptosis inducer erastin (Sato et al., 2018). In cancer cells, silencing SLC7A11 makes them more sensitive to ferroptosis induced by erastin, while overexpressing SLC7A11 makes them more resistant to it (Dixon et al., 2012). In addition, the tumor suppressor gene P53 can inhibit the expression of SLC7A11, and then inhibit the uptake of cystine by cells, ultimately leading to cell ferroptosis (Jiang et al., 2015). Overexpressing SLC7A11 in cardiomyocytes can restore cardiac GSH and cystine levels and reduce ferroptosis (Fang et al., 2020). In summary, GPX4 and system Xc- are both key regulators of ferroptosis.

Relationship between ferroptosis and mitophagy in MIRI

We have analyzed the relationship between mitophagy and MIRI as well ferroptosis and MIRI respectively, and the mechanism of ferroptosis. Mitophagy also has complex regulatory mechanisms. Classic mitophagy pathways include PINK1/Parkin, BNIP3/Nix and FUNDC1 pathway (Qiu et al., 2021). Mitophagy and ferroptosis are closely related to ROS during MIRI’s pathological process (Fig. 3). Moderate mitophagy can degrade damaged mitochondria and reduce excessive ROS production (Yang et al., 2020), whereas ferroptosis is accompanied by a large amount of ROS production and ultimately leads to myocardial cell death (Nakamura, Naguro & Ichijo, 2019; Su et al., 2019). It is an interesting question whether regulating mitophagy and reducing mitochondrial ROS production can inhibit ferroptosis in MIRI. From the perspective of mechanism, we found that HIF-1, mTOR and NLRP3 all play important roles in regulating mitophagy and ferroptosis, acting as “messengers”.

HIF in ferroptosis and mitophagy

HIF-1, which has three subtypes in mammals including HIF-1 α, HIF-2 α, and HIF-3 α, is a heterodimer transcription factor that plays a key role in mediating adaptive responses to hypoxia. HIF is closely associated with ferroptosis. In the hypoxic environment, HIF-1 is involved in the increase of TFR gene transcription in Hep3B human hepatoma cells (Tacchini et al., 1999). In colorectal cancer, activation of HIF-2 α potentiates oxidative cell death by increasing cellular iron (Singhal et al., 2021). In mouse testis, accumulation and stabilization of HIF-1 α induced by a widely used plasticizer (di (2-ethylhexyl) phthalate, DEHP) lead to ferroptosis in Leydig and Sertoli cells (Wu et al., 2022a). In the MIRI model, HIF-1 α has also been reported to induce TFR expression and iron absorption, exacerbate cellular oxidative damage and increase lipid peroxidation (Tang et al., 2008). These studies suggest that HIF especially HIF-1 α overexpression can induce ferroptosis mainly through storage, absorption and accumulation of iron.

HIF also plays important role in maintaining normal mitochondrial function. The HIF-1 α could improve mitochondrial function, decrease cellular oxidative stress, activate cardio-protective signaling pathways (Zheng et al., 2021). When MIRI occurs, promoting the expression of HIF-1 α and BNIP3 can promote BNIP3-mediated mitophagy, thus alleviating MIRI (Liu et al., 2019b; Zhang et al., 2019b; Zhu et al., 2020). BNIP3 and Nix (BNIP3L, homologous protein of BNIP3) are proteins on the surface of the mitochondrial membrane (Dorn, 2010). When mitophagy is activated, the LC3-interacting region (LIR) on BNIP3 and Nix can bind to LC3 on the membrane of the autophagosome, promoting the combination of damaged mitochondria with the autophagosome to complete mitophagy (Marinkovic, Sprung & Novak, 2021). Specifically, in the ischemia stage of MIRI, BNIP3 acts as a mitochondrial sensor of oxidative stress (Kubli et al., 2008) and HIF-1 initiates mitophagy mainly by activating BNIP3 and Nix (Martinez Vicente, 2017). In the reperfusion stage, BNIP3 is further activated due to ROS outburst, which promotes the initiation of mitophagy (Kubli et al., 2008; Ma et al., 2012).

Although both ferroptosis and mitophagy can be regulated by HIF. Paradoxically, increased HIF expression induces ferroptosis to aggravate MIRI, and increased HIF expression promotes mitophagy to alleviate MIRI. Some researchers inhibited GPX4 activity with ferroptosis inducers, and the production of lipid peroxides in the cells began to increase, followed by mitochondrial damage, mitochondrial ROS increased, and eventually lead to cell ferroptosis. When mitochondria targeted ROS scavenger agent (MitoQ) was used, mitochondrial morphology and function were preserved and cell death was prevented, despite the GPX4 inhibition and lipid peroxidation remained (Jelinek et al., 2018). Therefore, we suggest that ferroptosis in MIRI and even in ischemic cardiomyopathy may be mainly mitochondrial ROS dependent. Appropriate mitophagy can degrade damaged mitochondria and reduce ROS production, and the reduction of mitochondrial ROS makes Fenton reaction lack the necessary substrate H₂O₂, so even if HIF expression is increased, ferroptosis is not actually promoted.

Potential compounds for the treatment of MIRI

Considering the important role of inhibiting ferroptosis and regulating mitophagy in alleviating MIRI, we have summarized the potential drugs that can inhibit ferroptosis and/or regulate mitophagy (Table 1). In pathological conditions, since mitochondria are not the only ROS source, theoretically compounds that can both regulate mitophagy and inhibit ferroptosis should have better medicinal potential than compounds that inhibit only ferroptosis or regulate mitophagy alone. But it is a pity that although there are many compounds that can alleviate MIRI by targeting mitochondrial ROS clearance (Peng et al., 2022), there are few natural compounds have been reported to alleviate MIRI by regulating mitophagy or inhibiting ferroptosis. This indicates that although ferroptosis inhibition and mitophagy regulation play important roles in MIRI relief from a mechanism perspective, the study of active compounds for the treatment of MIRI based on ferroptosis and mitophagy needs to be further studied. Berberine alleviates MIRI through HIF-1 α/BNIP3 pathway (Zhu et al., 2020), and pentauterine B alleviates MIRI by inhibiting phosphorylated mTOR (Lu et al., 2019), providing direct evidence for the important role of mTOR and HIF in regulating mitophagy in MIRI alleviation. Therefore, referring to our previous analysis, we believe that HIF-1, mTOR and NLRP3 may become important targets for screening effective drugs to treat MIRI, based on the simultaneous regulation of mitophagy and ferroptosis.