Current progress of ferroptosis in cardiovascular diseases

Ferroptosis, a newly recognized form of nonapoptotic regulated cell death, is characterized by iron-dependent lipid peroxidation. Biological processes, such as iron metabolism, lipid peroxidation, and amino acid metabolism, are involved in the process of ferroptosis. However, the related molecular mechanism of ferroptosis has not yet been completely clarified, and specific and sensitive biomarkers for ferroptosis need to be explored. Recently, studies have revealed that ferroptosis probably causes or exacerbates the progress of cardiovascular diseases, and could be the potential therapeutic target for cardiovascular diseases. In this review, we summarize the molecular mechanisms regulating ferroptosis, inducers or inhibitors of ferroptosis, and the current progresses of ferroptosis in cardiovascular diseases. Furthermore, we discuss the emerging challenges and future perspectives, which may provide novel insights into the treatment of cardiovascular diseases.


Introduction
Cardiovascular diseases (CVDs), a group of diseases, are currently the leading causes of death globally, imposing a significant economic burden on families and communities (1,2).Therefore, understanding the pathophysiological progression of CVDs and related molecular mechanisms has become the main focus of medirelated cal studies.
Multiple types of cell death, including apoptosis, necrosis, and pyroptosis, are associated with the pathogenesis of CVDs (3).Accumulating evidence has demonstrated that ferroptosis, a newly recognized type of nonapoptotic regulated cell death (RCD), plays a central role in CVDs, which is triggered by iron-dependent lipid peroxidation (3,4).Ferroptosis was first reported in 2012, with features of loss of mitochondria crista, cell swelling, and outer membrane rupture, without the change of nucleus (5).
In this review, we focus on effects of ferroptosis and related mechanisms in the occurrence and development of CVDs, and highlight the effective strategy for targeting ferroptosis in CVDs.

Cellular mechanisms of ferroptosis
Ferroptosis is a complicated process that is moderated by a series of signaling and molecules.The occurrence and execution of ferroptosis contain three main events, iron dysregulation, GSH depletion, and lipid membrane oxidation (6).The regulatory mechanisms of these events and their roles in cardiovascular diseases are summarized in Figure 1.

Iron dysregulation
Iron homeostasis disorders are involved in the pathological processes of cardiovascular diseases.Iron as a necessary cofactor for several enzymes is essential for normal physiological activities, including cell oxygen transport, oxidation-reduction reactions, and energy metabolism (7).Circulating iron is present in two ionic forms, ferrous (Fe2 + ) and ferric (Fe3 + ) iron (8).In general, iron levels in different tissues are tightly regulated, which are accompanied by precise regulation of iron storage protein ferritin and other proteins.Extracellular iron binds to ferritin transferrin (TF) in the form of ferric iron, and is transported into cells by a cell-surface receptor, transferrin receptor 1 (TFR1) (8,9).Six-transmembrane epithelial antigens of the prostate 3 (STEAP3) catalyzes the reduction of Fe3 + to Fe2 + , and the transporter divalent metal transporter (DMT1) transports Fe2 + into the iron pool (LIP) through SLC11A2 (10,11).The surplus Fe2 + is temporarily kept in the ferritin protein, composed of the ferritin light chain (FTL) and ferritin heavy chain 1 (FTH1), or enters into the blood by ferroportin-1 (FPN) and lipocalin-2 (LCN-2) (12,13).
FPN, the sole exporter in charge of the efflux of surplus iron, is degraded after being bound to hepcidin (14).Accordingly, hepcidin-FPN axis is essential for maintaining iron homeostasis.Altered FPN expression could cause iron deficiency or overload.Knockdown of FPN promotes ferroptosis with an accumulation of intracellular iron and the production of reactive oxygen species (ROS) by the Fenton reaction (13).High level of hepcidin is induced via iron overload and chronic inflammation.Hepcidin binds to FPN and leads to degradation of FPN, thereby causing intracellular iron restriction (15).In contrast, low level of hepcidin in response to iron deficiency or hypoxia stabilizes FPN, leading to more liberation of the iron (15).
Under pathological conditions, intracellular iron homeostasis is disrupted, including increased iron intake or decreased iron stores, which leads to increased Fe 2+ concentration in extracellular space, and catalyzes the Fenton reaction to participate in the production of ROS, inducing lipid peroxidation and eventually ferroptosis (16).Recent studies have suggested that iron-bound ferritin could bind to nuclear receptor coactivator 4 (NCOA4), promoting the NCOA4-mediated autophagy pathway (namely ferrotinophagy) and resulting in lysosomal degradation of ferritin and iron release (17,18).Upregulation of NCOA4 increases ferritin degradation and promotes ferroptosis (19).Therefore, both iron overload and ferritin deficiency could trigger ferroptosis.

Lipid peroxidation
Lipid peroxidation is a way that oxidants, including free radicals or non-radical species, attack lipids containing two or more double hydrocarbon chains on the carbon chain.Especially, polyunsaturated fatty acids (PUFAs) are more prone to lipid peroxidation under oxidative stress (20,21).Cell membranes contain a large number of PUFAs that are primary targets of the attack by ROS (21).Lipid peroxidation can be non-enzymatically and enzymatically triggered.The former process is driven by free radicals, and the latter by enzymes, such as lipoxygenases (LOXs) (20).
PUFAs can be acylated by acyl-CoA synthetase long-chain family member 4 (ACSL4) to form PUFA acyl-coenzyme A (PUFA-CoA), and then reacts with lysophosphatidylcholine acyltransferase 3 (LPCAT3) to form PUFA-PE. PUFA-PE is subsequently oxidized by lipoxygenases (LOXs) to generates PUFAs-OOH (22,23).Peroxidation of PUFA-PEs mainly depends on ACSL4 and LOXs.Thus, suppressed ACSL4 or LOXs could effectively inhibit ferroptosis.A study with human haploid cells has shown that downregulation of ACSL4 and LPCAT3 inhibits ferroptosis by limiting oxidationsensitive fatty acids in cell membranes (24).Another study has also demonstrated that ACSL4 is a biomarker associated with ferroptosis sensitivity by regulating lipid metabolism (25).

Amino acid metabolism
Ferroptosis has been demonstrated to be controlled by amino acid metabolism.According to study reports, ferroptosis can be induced by erastin, a ferroptosis inducer.Erastin inhibits cystine uptake by the cystine/glutamate antiporter, therefore suppressing antioxidant defenses through GSH reduction (5,28).The pathway that erastin inhibits cystine uptake is mainly dependent on cystine/ glutamate antiporter (system Xc − ), a membrane amino acid transporter, which consists of the catalytic subunit solute carrier family 7 member 11 (SLC7A11) and the solute carrier family 3 member 2 (SLC3A2) (29).System Xc − transports glutamate and cystine between the intracellular and extracellular at a 1:1 ratio, and reduces the cystine to cysteine, therefore providing the initial materials of GSH in the cell (30).Inhibition of System Xc − could directly decrease cellular cysteine, thus repressing the generation of GSH.Thus, System Xc − plays an important role in the process of ferroptosis-related amino-acid metabolism (31).
In addition, it is widely accepted that GPX4 is another key point in the amino acid metabolic pathway.GPX4 belongs to the selenoprotein family that uses GSH as a reducing agent and turns GSH into oxidized glutathione (GSSG) (32).In the process, PUFAs-OOH is reduced to PUFAs-OH by GPX4, suggesting that GPX4 plays a vital role in inhibiting lipid peroxidation (33).Restricted GPX4 decreases GSSG and increases ROS, thereby inducing ferroptosis (34).

Regulation of ferroptosis
Lipid peroxidation and iron homeostasis, as well as specific gene interventions, have been shown to regulate ferroptosis.Many molecules in cells, including system Xc − and GPX4, have been implicated in governing ferroptosis (28).
Transcription factors have been shown to impact on the determinants of ferroptosis through regulating the antioxidant system.Kelch-like ECH-associated protein 1 (Keap1)-Nuclear factor E2-related factor 2/antioxidant response element (Keap1-Nrf2-ARE) pathway is the main survival pathway in cells, which protect cells from oxidative and electrophilic insults (35).Nrf2 is considered as a master transcription factor for the antioxidant response.Under normal conditions, Nrf2 protein levels are kept basally low by binding to Keap1, and being then degraded through the ubiquitinproteasome pathway (36).Under stress conditions, Nrf2 is released from Keap1 and translocated to the nucleus to initiate the transcription of the downstream gene (36).Notably, many proteins and enzymes responsible for the initiation of ferroptosis are the downstream of Nrf2, which are grouped into three broad categories according to their function: iron/metal metabolism, intermediate metabolism, and glutathione synthesis/metabolism (37).Therefore, the Keap1-Nrf2-ARE pathway is central to the ferroptosis.

Ferroptosis inducers and inhibitors
Ferroptosis plays a central role in CVDs, and regulating the levels of some key molecules related to ferroptosis may be helpful in treating CVDs (Table 1).Several compounds have been identified to induce or repress ferroptosis (Figure 1).

Ferroptosis inducers
Ferroptosis inducers are compounds that act by inhibiting system Xc − or altering intracellular iron levels.Many ferroptosis inducers has been found, which can be divided into four classes, class I-IV.Class I ferroptosis inducers inhibit cystine uptake through the system Xc − , such as erastin and sorafenib.Erastin binds to SLC7A5, which is one of six light chains that bind SLC3A2 to import cystine, decreasing cellular glutathione, and inducing ferroptosis (5,26,33).Sorafenib and its analogs, such as SRS13-45, SRS13-60, SRS13-67, and SRS14-98, also inhibit cystine import via system Xc − , deplete glutathione and cause lipid ROS accumulation (41).
Class II inducers directly inhibit GPX4 activity, such as RASselective lethal 3 (RSL3).Class II inducers do not deplete GSH, but bind to and inactivate GPX4, resulting in the formation of ROS (33).Class III inducers (such as FIN56), newly identified inducers of ferroptosis via chemoproteomic analysis, induce ferroptosis by two different pathways.FIN56 induces GPX4 degradation through post-translational regulation and depletion of coenzyme Q10 and modulation of mevalonate pathway (42).

Ferroptosis inhibitors
A group of small molecules have been identified as ferroptosis inhibitors, which act by iron chelation, inhibiting lipid peroxidation, or eliminating lipid peroxides.Iron chelators, such as deferoxamine (DFO) and dexrazoxane (DXZ), could deplete the cellular iron pool, and promote the excretion of excess Fe 2+ , thereby preventing ferroptosis (20,45).Ferrostatin-1 (Fer-1) and liproxstatin-1 (Lip-1), two effective inhibitors of ferroptosis, prevent lipid hydroperoxides accumulation due to the reactivity as RTAs instead of the ability as inhibitors of LOXs (45).In addition, studies have shown that liproxstatin-1 and Vitamin E could regulate ferroptosis by binding to 15-LOX with high affinity, inhibiting the 15-LOX enzymatic activity and then suppressing the generation of oxygenated PE (22).Besides vitamin E, reduced vitamin K confer strong anti-ferroptotic function via FSP1mediated reduction and act as an RTA (46).Reduced CoQ and its analogs, idebenone, also act as an RTA, suppressing lipid peroxidation and ferroptosis via potent free radical scavenging (47).

Ferroptosis and CVDs
Multiple cell death types are involved in the complicated mechanisms of CVDs.Recently, ferroptosis has been proven to play a central role in CVDs (3).The association between ferroptosis and various CVDs are summarized in the Table 2 and Table 3.

Myocardial ischemia/reperfusion injury
Acute myocardial infarction (AMI) is one of the most common high-risk CVDs (48).Timely restoration of the blood supply remains the effective treatment for ischemic myocardium, called reperfusion (49).However, during this process, the myocardium generates a large number of free radicals, further aggravating myocardium injury, which is known as ischemia/reperfusion injury (MIRI) (49).
Ferroptosis, characterized by excess iron accumulation and excessive lipid peroxides, participates in the pathogenesis of MIRI (50).Studies have suggested that ferroptosis happens mainly in myocardial reperfusion instead of ischemia, suggesting that ferroptosis intervention may have numerous beneficial effects on the reperfusion phase but not ischemic phase (51).
As a critical endogenous repressor of ferroptosis, GPX4 determines the severity of MIRI.The level of GPX4 is  significantly down-regulated in MIRI-injured myocardium, and overexpression of GPX4 attenuates ferroptosis damage in MIRI (52).GPX4 has been suggested to be a critical regulator of ferroptosis in MIRI.Recent studies have reported that iron metabolism balance also plays a crucial role in MIRI.Inhibition of TfR1 reduces iron contents and ROS levels in H9C2 cells, thereby suppressing H/R-induced ferroptosis (53).Furthermore, downregulation of TfR1 and upregulation FTH1 and GPX4 could also attenuate ferroptosis and prevent MIRI in MIRI rat model (54).In addition, knockdown of ubiquitin-specific protease 7 (USP7) inhibits ferroptosis in p53/TfR1 pathway-dependent in H/ R-treated H9c2 cells (55).A growing number of studies suggests that heme metabolism also exerts a significant role in ferroptosis.Heme oxygenase-1 (HO-1) level is increased both in the I/Rinjured myocardium and cultured cardiomyocytes.HO-1 degrades heme and therefore causes excess Fe2 + in endoplasmic reticulum (ER), triggering ferroptosis after H/R in cardiomyocytes (56).

Atherosclerosis
Atherosclerosis (AS) characterized with arterial wall lipid deposition is the pathological basis for many vascular disorders.AS has been shown to be associated with inflammation, deposition of lipid peroxides and iron (60).Ferroptosis has been shown to regulate AS and plays a vital role in the development of early and late stage (61).Inhibition of macrophage ferroptosis may significantly improve the progression of AS.In ApoE −/− mice, hepcidin-knockdown or SLC7A11-overexpression inhibits plaque macrophage ferroptosis, therefore delaying the progression of atherosclerosis (62).Inhibiting ox-LDL-induced macrophage ferroptosis through reduced Fe 2+ overload, lipid peroxidation, or elevated GPX4, FTH1, and SLC7A11 expression may alleviate atherosclerosis (63).Endothelial dysfunction is also critical in the onset and progression of AS (64).Iron overload in endothelial cells induces endothelial dysfunction by alleviating the oxidative stress (65).In human coronary artery endothelial cells (HCAECs), an enhanced PDSS2/Nrf2 axis inhibits ferroptosis by decreasing the release of ROS, providing a promising therapeutic target for AS (66).Moreover, in mouse aortic endothelial cells (MAECs), ferroptosis inhibitors could increase SLC7A11 and GPX4 levels, and reduce lipid peroxidation and iron content, therefore alleviating AS (67).Hepcidin, as a key regulator of iron homeostasis, could bind to and causes the degradation of ferroportin, thereby decreasing serum iron contents.Hepcidin deficiency reduces intracellular macrophage iron and proinflammatory phenotype of macrophages, and decreases atherosclerosis in the murine model (68).
Ethoxyquin is a lipophilic antioxidant widely applied in food preservation, and has been shown to inhibit GPx4-deficient ferroptosis and prevent DIC both in vitro and in vivo (81).In a cellular and animal DIC model, melatonin alleviates doxorubicin caused mitochondrial damage and ferroptosis by upregulating YAP expression (82).Acyl-CoA thioesterase 1 in the biosynthesis of polyunsaturated fatty acid prevents cardiomyocytes from doxorubicin-induced ferroptosis by altering the lipid composition (83).In doxorubicin-treated mice, Empagliflozin (EMPA) reduces ferroptosis, fibrosis, apoptosis and inflammation through NLRP3 and MyD88-related pathways, causing significant improvements in cardiac functions (84).Therefore, ferroptosis in DIC could be manipulated with various approaches, which might be a therapeutic target.

Others chemical-induced cardiotoxicity
Ferroptosis has been reported to play a vital role in 5-Fluorouracil and other antitumor agents induced cardiotoxicity

Diabetic cardiomyopathy
Diabetic cardiomyopathy (DCM), a major complication of type 2 diabetes, is a leading cause of heart failure and death in advanced diabetes.Ferroptosis also engages in the pathogenesis of DCM.A recent study has reported that ferroptosis is involved diabetes-induced endothelial dysfunction by activation of the p53-xCT-GSH axis (95).Astragaloside IV (AS-IV) attenuates cardiomyocyte injury through inhibiting CD36 expression, decreasing cellular lipid deposition, therefore suppressing ferroptosis in the models of DCM rats (96).Retinol dehydrogenase 10 (RDH10) metabolism disorder exists in the hearts of mice and the patients with type 2 diabetes mellitus (T2DM) (96).Reduction in RDH10 inhibits GPX4 and FSP1, promoting ferroptosis (97).In the model of T2D mice with DCM, sulforaphane (a Nrf2 activator) ameliorates cardiac ferroptosis in AMPK/Nrf2-dependent way (98).Canagliflozin (Cana), a hypoglycemic drug of SGLT2i, reduces ferroptosis both in DCM mice and cardiomyocytes by regulating iron metabolism and system Xc − /GSH/GPX4 axis (99).

Heart failure
Heart failure (HF) is a complex clinical syndrome with high mortality and morbidity (100, 101).Lipid peroxidation and cardiomyocyte ferroptosis have been reported in a rat model of heart failure (102).Since iron homeostasis is essential for the maintenance of normal function of cardiomyocytes, disrupted iron homeostasis contributes to HF. Knockdown of FTH1, which alters cardiac iron homeostasis and increases lipid peroxidation, leads to severe left ventricular (LV) hypertrophy via ferroptosis (103).Both SLC7A11 overexpression in cardiomyocytes and ferroptosis regulator Fer-1 may inhibit high-iron diet-induced severe cardiac damage (103).In a rat model of high-salt diet-induced HF with preserved ejection fraction (HFpEF), iron overloading and lipid peroxidation have been found in the myocardial tissue, as well as the expression of ACSL4, 4-HNE and NOX4 are increased, while GSH content is decreased (102).These changes eventually lead to ferroptosis, which are reversed by sodium-glucose cotransporter 2 inhibitors (SGLT2i) (102).In a transverse aortic constriction (TAC) mouse model, the circular RNA circSnx12 directly regulates iron metabolism via miR-224-5p/FTH1 regulatory network in myocardial cells, suggesting that circular RNA could be a potential therapeutic target for HF (104).Furthermore, expression of Tolllike receptor 4 (TLR4) and NADPH oxidase 4 (NOX4) is significantly increased in HF, and either TLR4 or NOX4 knockdown inhibits ferroptosis by regulating ferroptosis-related proteins, suggesting that TLR4-NOX4 pathway might be a potential therapeutic target for HF (105).

Atrial fibrillation
Atrial fibrillation (AF) is the most prevalent type of cardiac arrhythmia with higher morbidity and mortality, due to stroke, heart failure and other complications (106).Ferroptosis is related to frequent excessive alcohol consumption induced AF (107,108).Mechanistically, inhibition of atrial ferroptosis via SIRT1-Nrf2-HO-1 signaling to reducing atrial mitochondrial damage, ROS accumulation and iron overload, eventually decrease the AF susceptibility (107).FPN, an iron exporter, is involved in LPS-induced endotoxemia rat model.Knockdown FPN increases intracellular iron concentration, worsens the calcium handling proteins dysregulation and exaggerates the AF vulnerability (109).

Limitations and perspectives
Ferroptosis is a newly-identified iron-dependent form of cell death.Recent studies have suggested that ferroptosis could be a feasible therapeutic target for several cardiovascular diseases.However, the molecular mechanism of ferroptosis in cardiovascular diseases has not been fully understood.Reliable and sensitive biomarkers for ferroptosis need to be explored.Most investigations are confined to in vitro cell cultures, ex vivo models, and animal studies.The clinical studies relevant to ferroptosis in cardiovascular diseases are urgently needed.Currently, many chemical compounds have been found to affect the process of ferroptosis, including dysregulation of iron metabolism, lipid peroxidation, and abnormal amino acid metabolism.However, selectivity and side-effects of these compounds remain to be examined.It still faces great challenges in translating these compounds as future pharmacological drugs.In addition to ferroptosis, other forms of cell death, such as apoptosis, regulated necrosis and pyroptosis, have also been involved in the pathogenesis of CVDs.For examples, apoptosis and necroptosis are detected in diabetic mouse hearts at different ages (114).Palmitic acid induces the cell death of H9c2 cardiomyoblasts in a dose-and time-dependent manner (115).However, the occurrence timing of those cellular death forms in different CVDs is difficult to determine.The cross-talks may exist between those types of cell death.Therefore, it is unable to provide reliable and accurate information to guide the time window for intervention.
In conclusion, further studies are needed to identify specific biomarkers and central regulators of ferroptosis, providing solid evidence and basis for further exploration in clinical trials.Furthermore, it is important to clarify the different temporal stages of ferroptosis and other cell death forms in CVDs.

TABLE 3
Animal models of ferroptosis related genes in CVDs.