FUNDC1-mediated mitophagy is regulated by reversible phosphorylation. Under physiological condition, Ser¹³ in FUNDC1 is phosphorylated by CK2 kinase and Tyr¹⁸ is phosphorylated by SRC kinase, and their high phosphorylation status inhibits their interaction with LC3 to prevent mitophagy. Under the stimulation of oxygen deficiency or carbonyl cyanide 4-(trifluorome-thoxy) phenylhy-drazone (FCCP), dephosphorylation of Ser¹³ and Tyr¹⁸ enhanced the interaction between FUNDC1 and LC3, triggering mitophagy (Liu et al., 2014). Therefore, phosphorylation of Tyr¹⁸ is deemed a molecular switch for FUNDC1-mediated mitophagy (Kuang et al., 2016). In addition, concerted studies using ULK1 defective mutants with FUNDC1 knockout model suggest that FUNDC1 is the mitochondrial localization substrate of ULK1 and may have a ULK1 adapter property (Wu et al., 2014a). Further studies showed that SRC kinase inhibited the binding of ULK1 to mitochondria and decreased FUNDC1 phosphorylation of ULK1 on Ser¹⁷. Thus, ULK1 and FUNDC1 are both necessary for mitophagy, and regulate mitophagy synergistically, in contrary to inhibition of mitophagy through phosphorylation of Ser¹³ and Tyr¹⁸. Therefore, phosphorylation of Ser¹³, Ser¹⁷, and Tyr¹⁸ on FUNDC1 residues influences their interaction with LC3 together, thereby promoting or inhibiting mitophagy.

It is well known that collapse of mitochondrial membrane potential causes the accumulation of PINK1, which recruits Parkin onto damaged mitochondria to turn on mitophagy (Narendra et al., 2010). Studies proposed that the receptor-mediated machinery is involved in mitophagy induction by a decline in mitochondrial membrane potential (Chen et al., 2014). PGAM5, the phosphatase of Ser¹³, dephosphorylates Ser¹³ under hypoxia, and plays a synergistic role with PINK1-Parkin pathway in mitophagy. It was observed that PINK1 interacts with PGAM5, and deficiency of PGAM5 inhibits PINK1-mediated mitophagy (Imai et al., 2010). Knockout of FUNDC1 reduces Parkin translocation to mitochondria (Chen et al., 2014). Notably, PGAM5 activation is closely related to the anti-apoptotic protein BCL2L1. Wu and others found that BCL2L1 interacts with PGAM5 through BH3 domain to inhibit PGAM5 activation and prevents Ser¹³ dephosphorylation under normal condition. When hypoxia occurs, BCL2L1 degrades and releases PGAM5 to prompt Ser¹³ dephosphorylation, resulting in the initiation of FUNDC1-mediated mitophagy (Wu et al., 2014b). Levels of BCL2L1 govern FUNDC1 dephosphorylation and mitophagy mediated by PGAM5 through gain- and loss-of- function studies. Knockdown of PGAM5 inhibited mitophagy regardless of BCL2L1 level. Therefore, the BCL2L1-PGAM5-FUNDC1 signaling axis is vital for mitophagy under hypoxic condition.

MARCH5, a E3 ubiquitin ligase in OMM, participates in the regulation of mitochondria through ubiquitination of mitochondrial fusion protein 1 (Mfn1), mitochondrial fusion protein 1 (Mfn2), and MiD49 (Sugiura et al., 2013; Park et al., 2014; Xu et al., 2016). Recently, MARCH5 was also shown to participate in FUNDC1-mediated mitophagy. Under hypoxia, aggregated MARCH5 interacted with FUNDC1, which degrades FUNDC1 through Parkin-mediated ubiquitination and results in reduced binding of FUNDC1 with the LC3 motif, and desensitization of mitochondria (mitophagy) to hypoxia (Chen et al., 2017). These results suggest that MARCH5 may negatively regulate FUNDC1-mediated mitophagy, to maintain mitophagy at a certain level to avoid excessive autophagy damage of intracellular homeostasis. In addition, microRNA-137 (mir-137) may also inhibit mitophagy by down-regulating FUNDC1 under hypoxia both in vivo and vitro (Li et al., 2014). Interestingly, overexpression of mir-137 inhibited mitophagy without influencing autophagy of other organelles and proteins. These results indicate that mir-137 specifically regulates FUNDC1-mediated mitophagy, revealing promises for drug development targeting FUNDC1-mediated mitophagy.

Mitochondria are highly dynamic organelles, with their division, fusion and selective autophagy governing morphology and quality of mitochondria. Mitochondrial dynamics refers to a process in which mitochondria maintain dynamic balance through fusion and fission, and imbalance between the two is usually the prerequisite for mitophagy (Chen et al., 2016). After mitochondrial division, polarized and depolarized mitochondria are generated. The polarized mitochondria could be refused, while the depolarized daughter mitochondria become the target of mitophagy and could be cleared away (Mattenberger et al., 2003; Twig et al., 2008). Mitochondrial fusion is mainly mediated by Mfn1 and Mfn2 in OMM and optic atrophy 1 (OPA1) in the inner mitochondrial membrane (IMM). On the other hand, mitochondrial fission is mainly mediated by dynamic related protein 1 (DRP1) (Ding and Yin, 2012). OPA1 interacts with FUNDC1 through Lys70 residue. When Lys70 is mutated, the interaction of OPA1 and FUNDC1 is disengaged to promote mitophagy. Under mitochondrial stress, dephosphorylation of FUNDC1 promotes the dissociation of FUNDC1-OPA1 complex to bind with DRP1 to recruit DRP1 in mitochondria. In addition, Wu and coworkers noted division of mitochondria prior to the engulfment by autophagic pro-body (Wu et al., 2016). FUNDC1 aggregates onto the mitochondrial membrane to interact with endoplasmic reticulum (ER) calcium-binding protein (CANX) under hypoxia. The N-terminus of CANX binds to the hydrophilic region of FUNDC1, an indirect process due to the localization of N-terminus of CANX within the ER lumen. Therefore, contribution from intermediate proteins is speculated between FUNDC1 and CANX. As mitophagy proceeds, FUNDC1 would dissociate from CANX and recruit DRP1 to drive mitochondrial fission in response to hypoxic stress. Compared to CANX, DNM1L also binds to the FUNDC1 hydrophilic zone. In addition, deletion of FUNDC1, DRP1 or CANX in hypoxic cells may provoke mitochondrial elongation, elevated number of elongated mitochondria, decreased co-location of autophagosomes and mitochondria, thus preventing mitophagy. FUNDC1 is a novel receptor for DRP1 and may be an important molecule in the regulation of mitochondrial fission and autophagy under hypoxia (Wu et al., 2016). These results suggest that FUNDC1 regulates mitochondrial fission/fusion and mitophagy by interacting with DRP1 and OPA1, and functions as a springboard between mitochondrial dynamics and mitophagy.

Cardiomyocytes are rich in mitochondria for energy production, in order to maintain normal systolic and diastolic function of the heart (Barth et al., 1992). If mitochondria are injured, cardiomyocyte function will be compromised to trigger senescence of cardiomyocytes (Marzetti et al., 2013; Ikeda et al., 2014). In addition, severe damage of mitochondria may also release pro-apoptotic factors, leading to programmed cell death and myocardial injury (Zhou et al., 2018a). Injured cardiomyocytes could stimulate mitophagy to remove long-lived or damaged mitochondria, for the maintenance of cellular homeostasis and myocardial function (Kubli et al., 2013; Feng et al., 2016). When H9c2 cells are subjected to hypoxic challenge, WD repeat domain 26 (WDR26), a protein evoked by ischemic preconditioning, translocates Parkin to mitochondria by increasing mitochondrial membrane potential, thus promoting mitochondrial protein ubiquitination, mitophagy and myocardial protection (Feng et al., 2016). However, oxidative stress leads to excess mitophagy, which results in the degradation of healthy mitochondria despite the initial goal of selective clearance of damaged mitochondria. Unchecked mitophagy often causes irreversible damage to cardiomyocytes. Jin and team depicted that dual specificity protein phosphatase1 (DUSP1) was capable of inhibiting BNIP3-mediated mitophagy through a JNK-mediated mechanism, enhancing cardiomyocyte function following MI/R (Jin et al., 2018). These studies further consolidated an important role for mitophagy (FUNDC1-dependent or -independent) in cardiovascular system, suggesting the utility of appropriately enhanced mitophagy or inhibition of excessive mitophagy in the prevention of myocardial injury under pathological condition.

The main function of platelets is to participate in hemostasis and coagulation, in order to protect against vascular endothelial damage and atherosclerosis from external stimuli or endogenous reactions. However, platelets are over-activated by coagulation factors, ADP and inflammatory cells during MI/R. Activated platelets are spontaneously aggregated to form microthrombus, to block microvasculature and trigger myocardial cell injury or MI (Randriamboavonjy et al., 2015; Zhang et al., 2017). Platelet activation evokes dramatic energy-demanding morphological changes. Thus, maintenance of energy balance and mitochondrial integrity is essential to platelet activation and aggregation (Chen et al., 2016; Zhang et al., 2017). On one hand, mitophagy may avert further activation of platelets and formation of thrombosis by removing damaged mitochondria in platelets, offering a protective role in diabetic patients (Lee et al., 2016). On the other hand, insufficient mitophagy causes mitochondrial dysfunction of platelet and poor ATP production, leading to dissimilation of platelet activation and aggregation (Jang et al., 2015). It was reported that mitophagy may be facilitated through post-transcriptional modification of FUNDC1 in platelets, as evidenced by the absence of mitophagy activity in mice with knockout of platelets (Zhang et al., 2016). In addition, Zhang and others proposed that platelet mitophagy plays a dual role in the regulation of platelet activity and MI/R injury (Zhang et al., 2016). In the early stage of hypoxia, FUNDC1 binds with LC3 to induce mitophagy, which removes long-lived or damaged mitochondria to sustain mitochondrial function, in conjunction with accelerated mitochondrial metabolism and ATP production during initial myocardial repair. In addition, with prolonged ischemia, mitophagy is overactivated to prevent further deterioration of myocardial injury at the expense of removing parts of normal mitochondria (Zhang et al., 2016). These results suggest that the heart may execute a self-protective machinery through mitophagy induction in response to adverse external stimuli. These studies suggest a vital role for FUNDC1-mediated mitophagy in platelet activation and ischemia-reperfusion injury. During the ischemia period, FUNDC1-mediated mitophagy was upregulated in platelets and cardiomyocytes in response to ischemia and hypoxia challenge (Zhang et al., 2016).

FUNDC1 acts as a mitophagy receptor under hypoxia, through phosphorylation of Ser¹³, Ser¹⁷ and Tyr¹⁸. By interaction with IP3R2 (inositol 1, 4, 5-phosphoinositol receptor), FUNDC1 knockout in cardiomyocytes compromises mitochondrial function and disrupts mitochondrial-associated membranes (MAMs) and Ca²⁺ influx into mitochondria and cytosols (Wu et al., 2017). PGAM5 dephosphorylates FUNDC1 at Ser¹³ and activates mitophagy through binding to LC3 motif on phagocytes (Chen et al., 2014). PGAM5 also provides cytoprotection against necroptosis by promoting PINK1-mediated mitophagy, and PGAM5 injury aggravates necroptosis caused by MI/R injury in both hearts and brains (Lu et al., 2016). In addition, phosphorylation of FUNDC1 at Tyr¹⁸ remains virtually unchanged during post-MI/R phase, whereas phosphorylation of FUNDC1 at Ser¹³ is progressively upregulated, suggesting that inactivation of FUNDC1 through Ser¹³ following reperfusion exposure. It is important to realize that levels of phosphorylated FUNDC1 at Tyr¹⁸ and Ser¹³ are downregulated during ischemia prior to a progressive upregulation during refusion. Thus, inhibition of FUNDC1-mediated mitotic phagocytosis is a pathologic cause of MI/R injury, and phosphorylation of FUNDC1 at Ser¹³ and Tyr¹⁸ is closely associated with myocardial injury, indicating their diagnostic and prognostic potential as biomarkers for acute MI.

Dysfunctional mitochondria are closely related to the occurrence and progression of various CVDs, amongst these findings play a critical role. Upregulation of mammalian STE20-like kinase 1 (MST1) inhibits FUNDC1-mediated mitophagy through the cAMP response element binding protein (MAPK/ERK-CREB) cascade in MI/R injury, leading to ROS production, cell apoptosis, and aggravation of myocardial injury. Ablation of MST1 gene preserves FUNDC1 levels and mitophagy, thus reducing myocardial infarction size and preserving cardiac function (Yu et al., 2019). It was reported that casein kinase 2α (CK2α) levels were elevated in MI/R injury, which inhibited FUNDC1-mediated mitophagy. Cardiac-specific deletion of CK2α was reported to rescue against mitochondrial damage and protect cardiac function by reversing mitophagy in a FUNDC1-dependent manner (Zhou et al., 2018a). In addition, NR4A1 activates CK2α during MI/R, leading to suppressed FUNDC1-mediated mitophagy and aggravated MI/R injury (Zhou et al., 2018b). Similarly, reperfusion injury disrupted FUNDC1-mediated mitophagy and led to caspase 9-related apoptosis through upregulation of receptor-interacting serine/threonine kinase 3 (RIPK3). RIPK3 deficiency protected against MI/R injury by activation of mitophagy and inhibition of apoptotic pathways (Zhou et al., 2017b). Zhou and others reported that loss of peroxisome proliferator-activated receptor γ (PPAR γ) activates FUNDC1-mediated mitophagy and enhances mitochondrial function and ATP generation, leading to platelet aggregation and cardiac dysfunction in MI/R injury (Zhou et al., 2017a).

Aldehyde dehydrogenase 2 (ALDH2) is the most active isoenzyme in the ALDH superfamily, which mainly resides in mitochondria. ALDH2 possesses three main enzymatic functions, including dehydrogenase, esterase and reductase, among which dehydrogenase function (in removing aldehyde metabolites) is the most well-studied. Studies demonstrated that important cardioprotective effects in various risk factors of heart failure, such as coronary artery disease (CAD), hypertension, diabetes, alcoholism and other susceptibilities (Sun et al., 2014; Shen et al., 2015; Wang et al., 2016; Shen et al., 2017). Alpha-lipoic acid (α-LA) is a coenzyme in mitochondria. It is a cofactor of pyruvate dehydrogenase complex, ketoglutarate and amino acid hydrogenase complex which belongs to B vitamins. Studies have shown that α-LA restores ALDH2 activity through reduction of disulfide at its active site in diabetic cardiomyopathy and acute MI/R injury, thus restoring ALDH2 activity and improving nitrate tolerance (Wang et al., 2011; He et al., 2012). In the transverse aortic contraction (TAC) model, α-LA was shown to reduce cardiac hypertrophy (Zhang et al., 2014). As a highly efficient metabolic enzyme in mitochondria, increasing evidence supports an important role of ALDH2 in heart failure (Marino and Levi, 2018). Increased levels or activity of ALDH2 provides cardiac protection of ischemia and atherosclerosis through aldehyde detoxification (Chen et al., 2008; Yang et al., 2018). Li and others found that reduced α-LA in TAC-induced cardiac hypertrophy and cardiac fibrosis in an ALDH2-dependent manner. Mechanistically, α-LA activates FUNDC1 through NRF1 signaling (Li et al., 2020). However, it is not clear whether ALDH2 regulates FUNDC1 or other forms of mitophagy under pathological pressure overload settings. FUNDC1 mitophagy was found downregulated in pressure-overload induced heart failure, the effect of which was restored by α-LA treatment, in an ALDH2 dependent manner (Li et al., 2020). From the microarray data, certain mitochondria and autophagy related genes, such as Atpif1, BECN2 and BNIP3 were robustly altered, suggesting a role for these genes in α-LA- and ALDH2-offered action (Li et al., 2020).

Recent studies confirmed that FUNDC1 promotes mitochondria-associated ER membranes (MAMs) stability and regulates ER release of Ca²⁺ ions into mitochondria and cytoplasm through interaction with IP3R2 (Wu et al., 2017). MAMs refer to membranous contact of mitochondrial ER connection, with an important role in maintaining Ca²⁺ homeostasis (Rowland and Voeltz, 2012; Giorgi et al., 2015). FUNDC1 regulates the formation and maintenance of MAMs, and the destruction of FUNDC1-IP3R2 axis on MAMs promotes development of heart failure (Wu et al., 2017). Knockout of FUNDC1 in murine cardiomyocytes leads to more abundant slender mitochondria, and accumulation of defective mitochondria with suppressed cardiac function. Echocardiography showed cardiac systolic and diastolic dysfunction, myocardial fibrosis and increased expression with stress genes, including cardiac natriuretic peptide and brain natriuretic peptide, in FUNDC1 knockout mice. When subjected to acute myocardial infarction (AMI), FUNDC1 knocked-out mice exhibited worsened heart failure, shortened survival and greater mortality. Further studies suggested that FUNDC1 knockout exacerbated AMI-induced heart failure by inhibiting MAMs formation. This supports the idea that FUNDC1 and MAMs are involved in the development of heart failure (Giorgi et al., 2015). Therefore, restoration of MAMs integrity may be a new therapeutic target for heart failure.

In order to maintain a functional mitochondrial network, mitochondria develop specific repair pathways, including mitochondrial bacteriophage and UPR^mt (Zhu et al., 2021a; Gottlieb et al., 2021). UPR^mt prevents abnormal protein accumulation in mitochondria by normalizing folding and degradation of mitochondrial protein (Shpilka et al., 2021). UPR^mt controls the dynamic import and export of mitochondrial proteins and fine-tunes mitochondrial behavior (Lim et al., 2021). Recently, a role of UPR^mt in myocardial stress was reported (Smyrnias et al., 2019; Wang et al., 2019). In a mouse model of chronic stress-mediated cardiac hypertrophy (Smyrnias et al., 2019), UPR^mt markers such as ATF5, CHOP, mtDNAj, ClpP and LonP1 were significantly elevated at the mRNA level, suggesting a role for UPR^mt as an adaptive responsive molecule to myocardial stress. Interestingly, further activation of UPR^mt by nicotinamide nucleoside supplementation was associated with normal mitochondrial respiration and reduced cardiomyocytes in vitro (Smyrnias et al., 2019). In addition, cardiomyocyte apoptosis was lessened and cardiac function was improved in response to nicotinamide nucleoside, suggesting a cardioprotective effect activated by UPR^mt (Smyrnias et al., 2019). Loss of FUNDC1 significantly upregulated the transcription of UPR^mt markers, indicating that it acts as a compensatory mechanism in response to mitophagy inactivation. Wang and others reported that activation of FUNDC1 mitophagy alleviated mitochondrial injury and cardiomyocyte death in a cell model of septic cardiomyopathy (Wang et al., 2021). Increased mitochondrial quality control activates UPR^mt to optimize the import and export of mitochondrial proteins by upregulating gene transcription or promoting protein degradation (Wang et al., 2020b; Wang and Zhou, 2020). If UPR^mt cannot repair mitochondrial damage, mitochondrial division is induced to segregate the damaged regions from the healthy mitochondrial network (Zhou et al., 2018c; Wang et al., 2020c).

Mitophagy is used as a scavenger to remove structurally damaged mitochondria (Zhou et al., 2017b; Zhou et al., 2018a). Due to mitophagy-mediated mitochondrial removal is usually accompanied by reduction of mitochondria, mitochondrial integrity is enhanced by mitophagy to jack up ATP synthesis (Zhou and Toan, 2020; Chang et al., 2021). Therefore, UPR^mt and mitotic phagocytosis can be considered as distinct mitochondrial repair pathways. The former controls mitochondrial proteomics, while the latter alters mitochondrial quantity. Both regulatory and pathophysiological roles of mitotic phagocytosis were identified in vascular diseases (Zhu et al., 2021b; Zhou et al., 2021). Moderate activation of mitotic phagocytosis alleviates myocardial stress, while induction of abnormal mitotic phagocytosis is unexpectedly associated with pronounced cardiomyocyte death due to abrupt drops in residual mitochondria and intracellular ATP. Activation of FUNDC1-associated mitotic phagocytosis was shown to protect the heart against lipopolysaccharides (LPS)-induced sepsis through preserving mitochondrial function and structure (Wang et al., 2021). Moreover, loss of FUNDC1 inhibits mitotic phagocytosis, resulting in mitochondrial dysfunction and myocardial cell death. This finding further sheds some lights on the role of FUNDC1-dependent mitotic phagocytosis in myocardial stress (Zhou and Toan, 2020). Based on this finding, targeting FUNDC1-dependent mitotic phagocytosis should be a potential strategy for reconciliation of cardiac dysfunction in sepsis.

In MetS, glucose and lipid metabolism can be improved by reducing oxidative damage and promoting mitochondrial networking (Kelley et al., 2002). Under hypoxia, reducing mitochondrial number prevents production of reactive oxygen species (ROS) evoked by the mismatch between oxygen consumption and mitochondrial abundance (Lowell and Shulman, 2005; Giorgi et al., 2009). In addition, metabolic stress-mediated vascular endothelial injury can be prevented by promoting mitophagy in obese and diabetic situation (Vance et al., 1997). Tong and associates demonstrated that preservation or activation of mitophagy may prevent HFD-induced diabetic cardiomyopathy (Tong et al., 2019). Likewise, FUNDC1 deficiency was shown to promote obesity, insulin resistance, MetS, cardiac remodeling, and even cell death due to defective control of mitochondrial quality (Wu et al., 2019a; Ren et al., 2020). These findings suggest a central role for FUNDC1 in the regulation of cardiac function in obese MetS and the role of FUNDC1-mediated mitophagy as a main regulator of mitochondrial function, particularly in response to hypoxia-mediated mitochondrial injury commonly found in metabolic diseases.

Studies has revealed that high glucose-induced AMPK inhibition contributes to the onset of diabetic cardiomyopathy by upregulation of FUNDC1, FUNDC1-governed MAMs, and rises in mitochondrial Ca²⁺ in the heart. Mitochondrial Ca²⁺ controls cell metabolism and cell death through necroptosis, apoptosis, and autophagy. In particular, transient fluctuations of Ca²⁺ activate three matrix dehydrogenases (pyruvate dehydrogenase, α ketoglutarate dehydrogenase, and isocitrate dehydrogenase) governing oxidative phosphorylation and dehydrogenase which stimulate ATP generation (Denton, 2009). Recent findings also suggested an essential role for FUNDC1 in maintaining MAMs structures and ensuring adequate Ca²⁺ transfer from ER to mitochondria (Wu et al., 2017). Wu and others reported that FUNDC1 interacted with IP3R2 and thus inhibited IP3R2 ubiquitination and proteasome degradation, leading to activation of high glucose-driven adenylate activated protein kinase (AMPK). Interestingly, downregulation of FUNDC1 loosened FUNDC1-governed MAMs to reverse diabetic cardiomyopathy through inhibition of mitochondrial Ca²⁺ overload and mitochondrial injury (Wu et al., 2019b). In line with this finding, results from the British Prospective Diabetes Study suggested that metformin, one common antidiabetic drugs for type 2 diabetes and AMPK activator (Zhou et al., 2001; Xie et al., 2008; Xie et al., 2011; Meng et al., 2015), improved cardiac function and reduced incidence of myocardial infarction in diabetic patients (Effect of intensive blood, 1998). In contrary to its role in diabetic hearts, deficiency of FUNDC1 was found to accentuate obesity cardiomyopathy through regulation of mitochondrial Ca²⁺ integrity in an IP3R3-depednent manner. Data from our group revealed that FUNDC1 directly interacts with subunit of human SCF ubiquitin ligase complex FBXL2 to ease degradation of IP3R3 and reduce mitochondrial Ca²⁺ overload (Ren et al., 2020). These findings suggest an apparent paradoxical role of FUNDC1 in distinct settings of metabolic heart anomalies.