Mitochondrial iron metabolism and neurodegenerative diseases

Abstract

Iron is a key element for mitochondrial function and homeostasis, which is also crucial for maintaining the neuronal system, but too much iron promotes oxidative stress. A large body of evidence has indicated that abnormal iron accumulation in the brain is associated with various neurogenerative diseases such as Huntington’s disease, Alzheimer’s disease, Parkinson’s disease, and Friedreich’s ataxia. However, it is still unclear how irregular iron status contributes to the development of neuronal disorders. Hence, the current review provides an update on the causal effects of iron overload in the development and progression of neurodegenerative diseases and discusses important roles of mitochondrial iron homeostasis in these disease conditions. Furthermore, this review discusses potential therapeutic targets for the treatments of iron overload-linked neurodegenerative diseases.

1. INTRODUCTION

Iron, an essential metal nutrient, is a fundamental element for many metabolic processes, including electron transfer, oxygen transport and storage, mitochondrial function, and cellular growth and differentiation. Such biochemical reactions in the body are processed by iron-containing proteins such as hemoglobin, iron-sulfur enzymes, and heme-containing enzymes. Thus, iron homeostasis is controlled via multiple iron regulatory processes: iron uptake, storage, transfer, and export to other cells or organs.

In the brain, iron is involved in many physiological activities such as mitochondrial respiration and energy production, myelin synthesis, neurotransmitter synthesis and metabolism, and DNA replication (1). Given that neuronal mitochondrial respiration is in charge of about 20% of the total body oxygen consumption (2), the maintenance of physiological iron levels is critical for proper brain function. Indeed, dysregulated iron levels have been reported to be associated with several neurodegenerative diseases, including Friedreich’s ataxia (FA), Huntington’s disease (HD), Alzheimer’s disease (AD), Parkinson’s disease (PD), and neurodegeneration with brain iron accumulation (NBIA) (3–7). Studies have also pointed out that many neurodegenerative disorders are characterized by abnormal iron status in the mitochondria, the major iron recipient of the cell (8–10). In this review, we aim to summarize important roles of mitochondrial iron in neuronal function and to provide current understanding about the pathophysiology related to mitochondrial dysfunction in neurodegenerative diseases.

2. MITOCHONDRIA AND IRON

2.1. Biochemistry of Iron

Iron is a d-block transition metal whose primary oxidation states in biological conditions are ferrous (II), ferric (III), and ferryl (IV) states. The unoccupied d orbitals also allow iron to bind to ligands, including oxygen, nitrogen, and sulfur atoms (11, 12). Thus, iron has a tendency to easily donate its electrons, which contributes to various biochemical processes (13). In normal cellular conditions, ferric iron (Fe³⁺) constitutes the common stable form of bioavailable iron that is associated with various biological complexes. Meanwhile, ferrous iron (Fe²⁺) can be found in ferritin, heme synthesis, and transmembrane transport. Despite its importance in biochemical processes, free ferrous iron is highly reactive and can cause cellular damage (13, 14), and is therefore stored and chaperoned throughout the cell with the help of several iron binding and transport proteins (e.g., ferritin or hemosiderin) (13). For example, ferritin is a highly conserved protein among both vertebrate and invertebrate species, and two subunits of ferritin, L-ferritin (FTL) and H-ferritin (FTH), are essential for iron storage in vertebrate cells.

2.2. Cellular Iron Homeostasis

Iron, after its intestinal absorption or export out of iron-storage organs (e.g., liver), binds to transferrin (Tf) as a ferric form in the plasma, and Tf-bound iron (TBI) is taken up by the cell through endocytosis mediated by transferrin receptor 1 (TfR1), which is the major iron transport mechanism under normal physiological conditions (15–17). Intracellular Fe³⁺ is then reduced by ferrireductase Steap3 to Fe²⁺ in the endosome and is transported to the cytoplasm by the divalent metal transporter 1 (DMT1). This reduced iron (Fe²⁺) is either stored in ferritin protein, labile iron pool or delivered to mitochondria (18).

In the cytosol, iron concentration is regulated by iron regulatory proteins (IRP1 and IRP2) that bind to the iron-responsive elements (IREs) within the 5’ or 3’ untranslated regions (UTR) of target mRNAs and regulate the expression of various iron processing proteins, including TfR1 and iron storage protein (i.e., ferritin). Thus, the levels of these iron-associated proteins are precisely regulated by IRPs, and any decrease in extracellular iron levels can promote the IRP-IRE interaction, resulting in increased stability of TfR1 mRNA and decreased translation of ferritin (19). Conversely, when the cell is iron sufficient, a 4Fe-4S cluster is bound to IRP1, while IRP2 undergoes iron-mediated proteasomal degradation. The depletion of IRP2 not only causes the degradation of TfR1 mRNA, but also promotes ferritin translation. Those outcomes lead to a decrease in iron transport into the cell, as well as an increase in iron storage by ferritin, thereby preventing iron-induced oxidative damage. The IRP2-mediated iron response is thought to be dependent on oxygen level as well as FBXL5 protein binding, which further activates ubiquitination-dependent IRP2 degradation (20). However, a recent study suggests that IRP2 may not require FBXL5 protein binding to modulate its iron response and it can regulate cellular iron levels independent of IRP1 (21). The underlying mechanism has been proposed to be dependent on iron-sulfur (Fe-S) cluster binding to the IRP2 protein itself, since an inhibition of Fe-S biogenesis in IRP1-knockout cells had similar iron responses as in IRP2-knockout cells (21). However, more studies are warranted to fully understand the Fe-S dependent iron response linked with IRP2, as well as its potential applications in iron-associated disorders (Fig. 1A).

Fig. 1: Cellular and mitochondrial iron homeostasis.

A: Iron dependent cellular iron homeostasis. From circulation, iron is imported into the cell 1) mostly as transferrin-bound iron (TBI) through TfR1-mediated endocytosis, which releases iron from the transferrin due to the difference in pH and is transferred to the cytosol as Fe²⁺ by DMT 1 or 2) to lesser degree as non-transferrin-bound iron by DMT1 directly into the cytosol. Iron is then transported to various organelles or labile iron pool, and excess iron is stored in ferritin protein. Cellular iron regulatory system is dependent on the availability of iron itself in the cell. During increased iron levels, there is an increased availability of Fe-S clusters (dependent on iron level) bound to the IRPs, therefore decreasing IRP-IRE interactions. These interactions result in decreased iron uptake by transferrin receptor (TfR), increased iron storage by ferritin (FTN), and increased iron export by ferroportin (FPN).

B: Mitochondrial iron regulation. Iron is imported into the intermembrane space via several proteins (e.g., porins, DMT1). Mitoferrin (MFRN) then imports the iron into the mitochondrial matrix, where it is utilized for Fe-S cluster biogenesis by the iron sulfur cluster assembly complex, as well as for energy production by the ETC. Excess iron is not only stored in mitochondria specific ferritin, but also exported by iron exporters such as ABCB7/8 in either elemental form or Fe-S clusters. Upon iron overload in the mitochondria, free iron content level is elevated, by which the Fenton reaction can produce highly reactive oxygen species, leading to increased oxidative damage.

2.3. Mitochondrial Iron Homeostasis

Mitochondria are vital for the survival of nearly all eukaryotic cells and organisms since they are essential for energy metabolism along with the generation of various intermediate metabolites (e.g., acetyl CoA, citrate, succinate, and fumarate) (22, 23) and the regulation of trace elements (e.g., copper, manganese, zinc, and iron) (24–26). Therefore, mitochondrial regulations of these elements and metabolites are critical for proper mitochondrial functions.

2.3.1. Iron and mitochondrial iron-sulfur cluster.

Iron plays an important role in the mitochondria where it is utilized in three major pathways: heme synthesis, Fe-S cluster biogenesis, and iron storage in the mitochondria-specific ferritin (FtMt) (15). Heme synthesis and Fe-S cluster biogenesis are highly interdependent because enzymes (e.g., ferrochelatase) involved in the heme synthesis pathway contain an Fe-S cluster (27, 28). Fe-S clusters are versatile prosthetic groups for the proteins involved in several essential cellular processes such as electron transfer, energy production, gene expression, and iron metabolism. These clusters, composed of ferrous/ferric iron and inorganic sulfide, are allowed for delocalization of electron cloud over both Fe and S, which is essential for the versatility of Fe-S clusters as co-factors (29). The synthesis of Fe-S clusters is facilitated by over 20 different protein components in the cytosol, mitochondria, and nucleus, where sulfur is sourced from cystine and is combined with iron to form a Fe-S cluster on a scaffold protein (30, 31). This Fe-S cluster is then incorporated into an apoprotein, which is utilized for required biochemical activities. It has been suggested that the machinery of Fe-S cluster synthesis was inherited from bacterial ancestors of mitochondria, and this system has been conserved in eukaryotes from yeast to human (31). Because of the versatility of the Fe-S cluster biosynthesis and its importance for essential cellular functions, any disruption in the Fe-S cluster formation can impair the overall iron homeostasis of the mitochondria and can be linked with severe diseases such as FA and Fe-S assembly enzyme myopathy (32, 33).

2.3.2. Mitochondrial Iron Regulation.

Mitochondria are a bilayer organelle. The outer mitochondrial membrane (OMM) is highly permeable to ions and small uncharged molecules via pore-forming membrane proteins (porins), while an inner mitochondrial membrane (IMM) is impermeable to most ions and small molecules. Porins in the OMM, such as voltage-dependent anion channel (VDAC), allow the transport of most of the metabolites involved in energy production, including pyruvate, ATP, cations, and others (34). Although the impermeability of IMM creates a high electrochemical gradient between the intermembrane space and the matrix, ions and molecules can be transported only through IMM-specific proteins. As a result, mitochondrial iron uptake is regulated by IMM protein mitoferrins (MFRNs), members of the mitochondrial carrier family (MCF) (15). MFRN1, also referred to as SLC25A37, is involved in erythroid cell development. Meanwhile, MFRN2, or SLC25A28, is ubiquitously expressed in different tissues (35–38), although the exact roles of MFRN2 have not been completely understood (35, 39).

In addition to the iron import and utilization mechanisms, it is also important for mitochondria to regularly export iron in the form of heme, Fe-S clusters, as well as elemental iron. Failure of proper iron export from the mitochondria can cause severe mitochondrial iron overload as seen in FA (27, 40) and various cardiomyopathies (41, 42). Although the mechanism of mitochondrial iron export is not clearly determined, it has been suggested that the members of the mitochondrial ABC family (e.g., ABCB7 (43) and ABCB8 (42)) are involved in mitochondrial transport of iron. In particular, ABCB7 is found to export mitochondrial iron via the formation of the Fe-S cluster (44, 45). Notably, it has been reported that ABCB7 could contribute to the regulation of cytosolic iron status since the deletion of ABCB7 gene hampers the formation of cytosolic proteins containing the Fe-S cluster (43, 46). The exact roles of ABCB8 have yet to be defined, although it appears to support mitochondrial export of iron (42) (Fig. 1B).

3. ROLE OF IRON IN NEURONAL FUNCTION

3.1. Iron and neurochemical processes

Iron is involved in the regulation of various neuronal functions, in which several neurotransmitters such as dopamine, norepinephrine (NE), epinephrine, and 5-hydroxytryptamine (5-HT; serotonin) are involved, and these neurotransmitters are synthesized by several iron-containing enzymes, including tyrosine hydroxylase, phenylalanine hydroxylase, and tryptophan hydroxylase (12). For example, studies have shown that iron-deficient children are more vulnerable to anxiety and/or depression (47). These conditions appear to have a significant relationship with altered brain homeostasis, including myelination, neurotransmission (e.g., monoamine metabolism), and energy metabolism (48).

The dopaminergic system is tightly regulated by changes in iron status. It has been suggested that decreased whole-brain iron levels may result in altered dopaminergic system in different brain regions. The dopamine receptor 1 (D1R) was reported to be lowered in the caudate putamen and prefrontal cortex in postweaning iron-deficient rats (11, 49). Nevertheless, contradictory findings have been also reported, as increased or decreased extracellular dopamine levels are observed in the striatum or prefrontal cortex, respectively, of rats exposed to postweaning iron deficiency (50, 51).

5-HT and NE are also reported to be altered by cellular iron status, as radioligand binding studies have demonstrated reduced 5-HT transporter densities in the striatum of iron-deficient mice (52). Moreover, extracellular NE levels are also found to be increased in iron-deficient rats, while NE transporter levels in the brain are decreased in the iron-deficiency condition (53, 54).

In addition, iron is required for the production and maintenance of myelin, which is the fatty white matter that insulates axons and helps preserve signaling (12). Brain iron deficiency leads to abnormal expression of myelin-related proteins, lipids, and cholesterol, since 80% of the brain cholesterol content is expressed as myelin (55–58). Lack of myelination causes delayed neuronal conduction, which was shown to be associated with retardation of reflexes in iron-deficient infants (59, 60).

3.2. Brain iron trafficking and regulation

Dietary iron (Fe³⁺) is reduced in the small intestine by the duodenal cytochrome B to Fe²⁺, which is then taken up by DMT1 into the duodenal epithelium and exits to the blood by a basolateral membrane transporter ferroportin (FPN). In the blood, Fe²⁺ is oxidized back to Fe³⁺ by ceruloplasmin or hephaestin and then binds to circulating transferrin (Tf). As described above, the iron-bound Tf interacts with TfR1 on the microvasculature of the blood-brain barrier (BBB) of the brain, which allows iron to enter the brain by endocytosis and transcytosis within various cell types in the central nervous system (CNS) (61–63). Once in the brain, iron is utilized for cellular metabolism or stored in ferritin when in excess. Alternatively, other cells including astrocytes and oligodendrocytes can take up excess iron in the low molecular weight form, such as iron-citrate, ATP, or ascorbate, which collectively represent the non-transferrin-bound iron (NTBI)-mediated transport pathway (64, 65).

In the brain, iron is more abundantly distributed in the white matter compared with the gray matter since significant iron staining is observed in oligodendrocytes of the matured brain (62, 66). Iron is highly required in oligodendrocytes which exhibit high metabolic rate and are responsible for axon myelination, and the Tim-2 receptor-mediated endocytosis of interstitial FTH has been identified as the major source of iron for oligodendrocytes due to their lack of TfRs (65, 67, 68). However, little is known about how and where iron is released from FTH and transported across the membrane into the cytosol. Although oligodendrocytes can synthesize Tf, its secretion from the cells is very limited (1). Instead, Tf is mainly synthesized and secreted by the choroid plexus, so that iron (Fe³⁺) is further transported to glia and neurons (11, 64, 65). Neurons can take up iron from Tf primarily via TfR1 as well as DMT1, and can export excess iron via FPN, the only known iron exporter, out of neuronal cells (8). Dysregulation of proteins involved in iron export may lead to iron accumulation in the brain and is further associated with neurodegeneration disorders such as aceruloplasminemia (ACP) (69) and PD (70) (Fig. 2).

Fig. 2: Brain Iron Homeostasis.

After entering the brain capillary endothelial cells (BCEC), transferrin-bound iron (TBI) is either 1) directly released into the brain via transcytosis or 2) converted to Fe²⁺ in the endosome after internalization of transferrin receptor 1 (TfR1) that is released to cytosol via divalent metal transporter 1 (DMT1) and then exported to the brain through ferroportin (FPN), forming the labile iron pool (LIP). Hepcidin could downregulate the expression of FPN and consequently trap iron in the cells. The labile iron in the brain is either stored in ferritin (FTN) or oxidized to Fe³⁺ by ceruloplasmin (CP), and Fe³⁺ is further bound to extracellular Tf secreted from the choroid plexus. Fe³⁺ that is released from astrocytes can bind to low molecular weight compounds such as ATP, ascorbate, and citrate (i.e., NTBI, non-transferrin-bound iron). Astrocytes also take up labile irons via DMT1 or other pathways such as TfR1, wherein those irons are then either stored in FTN or exported via FPN. CP can promote Fe²⁺ release by FPN. The extracellular TBI is taken up by neurons by TfR1-mediated endocytosis and undergoes internalization and acidification in endosomes where iron is released. The released Fe³⁺ is further reduced to Fe²⁺ and exported into the cytosol by DMT1, contributing to the LIP of neurons. Iron is exported from neurons by FPN, which is stabilized by amyloid precursor protein (APP). Microglia takes up iron and export ferritin through unknown pathways (possibly DMT1 and FPN), and the exported ferritin is further imported into oligodendrocytes by T-cell immunoglobulin mucin domain 2 protein (Tim-2).

3.3. Dysregulation of brain iron

Although iron status is tightly regulated in the brain, abnormally high or low iron levels are directly or indirectly involved in the pathophysiology of several neurodegenerative diseases, including AD, HD, PD, FA, and amyotrophic lateral sclerosis (ALS). Iron deficiency may impair the synthesis of essential iron-containing proteins and consequently hamper normal physiological functions. Untreated systemic iron deficiency can lead to iron-deficiency anemia, which is the most common type of anemia since the body lacks iron required to produce hemoglobin (71). Nevertheless, it has been shown that anemia is not only caused by iron deficiency, but also linked with chronic inflammation (72).

Recently, hepcidin, the master regulator of systemic iron homeostasis, is often elevated during inflammation (73) and has also been identified as a key player in brain iron homeostasis by controlling FPN expression post-translationally (74, 75). One of the possible sources of hepcidin in the brain is systemic hepcidin that can cross the BBB via transcytosis, although the exact mechanism remains unknown (76–78). In the brain, it has been also suggested that local cellular source of hepcidin is available from astrocytes or microglia (74, 76). Like systemic hepcidin, brain hepcidin appears to regulate iron homeostasis by downregulating the expression of FPN and iron importers such as DMT1. Both cell (79) and animal (80) studies have shown that increased hepcidin expression in the brain leads to both decreased cellular iron uptake (by decreasing TfR1 and DMT1) and decreased release (by decreasing FPN), although mixed results have been reported by different groups. A recent study has shown that overexpression of brain hepcidin in astrocytes leads to decreased brain iron level in AD mice (81). On the other hand, in studies using 6-hydroxydopamine (6-OHDA)-induced PD cell culture models, elevated levels of both hepcidin and DMT1 along with decreased expression of FPN were noted, resulting in iron accumulation (82). This suggests an independent regulation of DMT1 by 6-OHDA-induced inflammation apart from the hepcidin pathway, as reported by Jiang et al. (83). It was also reported that hepcidin knockdown in neuronal cells attenuated cellular iron burden by upregulation of FPN1 expression (82, 84), which is inconsistent with findings in the AD model. More studies are needed to identify and verify the molecular mechanisms of hepcidin and FPN in brain iron homeostasis.

Iron overload is also associated with several neuronal diseases. It has been well-recognized that age-related brain iron accumulation is linked with neurodegenerative diseases (e.g., PD and AD), and other iron overload diseases are caused by genetic mutations, which include hemochromatosis (i.e., Hfe), aceruloplasminemia (i.e., CP), and FA (i.e., FXN) (64, 85, 86). It is generally believed that excess free iron in the brain interacts with oxygen molecules to form reactive oxygen species (ROS) through the Fenton and Haber-Weiss reactions, which leads to oxidative stress in the neuronal cells (61, 87). Therefore, increased non-heme iron promotes the generation of ROS and consequently cellular damage and tissue fibrosis in the systemic organs (88).

Abnormal iron status can cause neurodegenerative diseases via mitochondrial dysfunction since iron is required for the formation of Fe-S clusters that are critical factors for oxidative phosphorylation (89). It has been reported that disruption of iron homeostasis results in altered mitochondrial morphology and function, as well as impaired mitochondrial DNA (90, 91). As a result, mitochondrial dysfunction has been recognized as an important hallmark for neurodegenerative diseases with iron dyshomeostasis. For instance, Huang et al. reported that iron overload results in reduced mitochondrial energy production along with induction of apoptosis, which is accompanied by mitophagic signaling (e.g., PINK1) (89).

3.4. Brain Iron and Mitochondrial Dynamics

Mitochondria are highly dynamic organelles so that they alter their functional and structural characteristics in response to various environmental changes. The process of fission or fusion allows mitochondria to divide into multiple smaller mitochondria or to combine into larger mitochondria, respectively, which not only contributes to attaining functional structures of mitochondria within a cell, but also facilitates the maintenance of mitochondrial homeostasis (92–94). Although the detailed mechanisms of mitochondrial fission and fusion are out of the scope of this review, it is important to understand how these mitochondrial dynamics are regulated by cellular iron status.

Mitochondrial fission and fusion are dependent on the common primary regulator, dynamin-related family proteins. The DRP1 (Dynamin-Related Protein 1) is a primary fission protein that is present in the cytosol and interacts with fission protein 1 (FIS 1) (95). During fission, DRP1 forms a ring structure around the mitochondria to constrict the OMM at the site of fission, creating smaller spherical mitochondria. On the other hand, the process of mitochondrial fusion is mediated by two primary proteins, MFN (Mitofusin) 1 and 2, and the dynamin-like GTPase protein Optic Atrophy 1 (OPA1) (94). The MFN proteins mediate the fusion of OMM, whereas the IMM fusion is mediated by the OPA1 protein (96). In normal cells, mitochondrial fission-fusion is well balanced, however this balance is disturbed in many disease conditions including neurodegenerative diseases. Following the impaired mitochondrial fission-fusion balance, mitochondrial morphology and activity are abnormally regulated due to diminished respiratory and bioenergetic capacities (97).

Along with the important roles of iron in mitochondrial oxidative function (98), dysregulated iron homeostasis has been also shown to be linked with an imbalance in mitochondrial fission and fusion in the central area. For example, several neurodevelopmental and neurodegenerative diseases, such as autism spectrum disorder (ASD), attention-deficit/hyperactivity disorder (ADHD), and PD, have been found to be associated with prenatal and postnatal iron deficiency that results not only in depleted mitochondrial ATP production, but also in dysregulated mitochondrial dynamics and density in the neuronal cells (99, 100). In particular, iron-deficient neurons were shown to accelerate the fission process, as evidenced by decreased mitochondrial size, as well as reduced expression of fusion (OPA1) and increased expression of fission (DRP1) proteins (97).

In addition to iron deficiency, iron overload is also commonly observed in many neurodegenerative diseases (e.g., AD, PD, and HD) (101–103), and this excess iron has been shown not only to be linked with oxidative stress, but also to alter mitochondrial dynamics to prefer fission, rendering more fractioned mitochondria (104, 105). Although exact mechanisms are poorly understood, the iron overload-induced mitochondrial fission-fusion imbalance is likely to be secondary to the primary cause of neurodegenerative diseases, except for the Charcot-Marie-Tooth disease in which MFN2 defect plays a primary role (106). Although elevated DRP1 activity (i.e., fission) has been reported in iron overload-related neurodegenerative diseases (104, 107–109), there is no direct evidence that increased iron levels alter the fission-fusion balance towards fission. Nevertheless, it has been suggested that two mechanisms may be involved in iron overload-induced mitochondrial fission. First, iron overload-associated ROS production can stimulate mitochondrial fission through activating DRP1, as the amelioration of excess ROS was shown to prevent mitochondrial fragmentation as well as neurodegeneration process (109, 110). Furthermore, in neuronal cells with a deficiency of peroxiredoxin 5 that is important for eliminating H₂O₂, mitochondrial fragmentation and neurodegenerative characteristics were observed (108). Second, in neuron, iron overload-induced mitochondrial fission is probably linked with calcium signaling, as DRP1 activity has been shown to be regulated by cellular calcium levels. For example, elevated iron levels appear to accelerate calcineurin activity along with increased intracellular calcium levels, which could upregulate mitochondrial fission by DRP1 dephosphorylation at serine 637 and affect neurodegeneration (107, 111).

Collectively, altered mitochondrial dynamics are hallmarks of neurodegenerative diseases that are affected by either iron deficiency or overload, although more studies are warranted to fully understand a causal relationship between iron status and mitochondrial fission-fusion balance in neuronal cells.

4. NEURODEGENERATIVE DISEASES AND IRON OVERLOAD

Multiple factors contribute to the development of neurodegeneration, including aging, oxidative stress, neuroinflammation, and neuronal or protein dysregulation (e.g., dopaminergic neurons in PD and protein aggregation in AD). In addition, it has been revealed that both iron dysregulation and mitochondrial dysfunction are critical causal factors for neurodegenerative diseases. Ferroptosis, as an iron-dependent programmed cell death, has also been implicated in neurodegenerative disorders, including AD and PD (112). Ferroptosis is characterized by depletion of glutathione (GSH) and lipid peroxidation, with distinct morphological changes such as shrunken mitochondria (113) and ruptured mitochondrial outer membrane (114). Glutathione peroxidase 4 (GPX4) plays an important role in the regulation of ferroptosis, by catalyzing the reduction of lipid peroxides along with GSH (112). Furthermore, GSH binds to Fe²⁺ in the labile iron pool (LIP) and prevents it from being oxidized, acting as a ligand for ferrous iron and providing a substrate in the synthesis of the Fe-S cluster proteins (115).

While brain iron levels increase with age, upregulated iron-related proteins are observed in multiple brain regions of the elderly (64). For example, higher levels of ferritin and heme oxygenase-1 (HO-1) are detected in the substantia nigra (116) and hippocampus (117), respectively, which is linked with increased oxidative stress in the brain of older people. Nevertheless, the causality between iron accumulation and aging-related neurodegeneration has not been completely understood. Iron accumulation in the brain is likely to alter iron-related gene expression, which contributes to neurodegenerative diseases. Many of these disease conditions are also significantly correlated with mitochondrial dysfunction in the CNS, which is affected by iron transport and metabolism. In the following sections, several notable neurodegenerative diseases associated with iron overload are discussed with a focus on mitochondrial iron regulation.

4.1. Friedreich’s Ataxia

Friedreich’s ataxia (FA) is an autosomal recessive disorder that is caused by the mutation of FXN gene located on the ninth chromosome that codes for mitochondrial frataxin protein (118, 119). The mutation for FXN is most commonly identified by the intronic expansion of GAA trinucleotide repeat of the FXN gene (119). The trinucleotide repeats in FA patients can reach over 1000 GAA copies, and it has been shown that the number of these repeats is inversely proportional to the age of onset for the disease (118). However, trinucleotide expansion is not the only cause for abnormal FXN expression since several epigenetic factors (120), FXN point mutation (121, 122), and compound heterozygosity for GAA expansion (123) have also been observed in a smaller fraction of FA patients. This disease is commonly characterized by the loss of positional awareness, muscle dystrophy, hypertrophic cardiomyopathy, and diabetes.

Mature human FXN is a 210-amino acid protein (~23 kDa), which is essential for iron homeostasis and metabolism (124), as well as many enzymatic actions as co-factor. Frataxin is a nuclear encoded protein that is later transported to mitochondria (32) wherein it contributes to the production of Fe-S clusters since FXN-ablated cells show reduced abundances of Fe-S cluster proteins (125). FXN has also been shown to interact with ISD11, a component of the Fe-S cluster assembly complex, supporting the idea that FXN is required for the formation of Fe-S clusters (126, 127). These clusters are also involved in several key proteins including mitochondrial respiratory chain components (e.g., complex I and II), aerobic metabolic enzymes (e.g., aconitase, citrate, fumarase), and other proteins in the nucleus (for DNA synthesis and repair mechanism) and cytoplasm (for ribosome biogenesis) (128, 129).

Under normal iron status, healthy cells promote the production of Fe-S clusters, while inhibiting the import of extra iron by the action of iron regulatory proteins (IRP1 and IRP2) (130, 131). However, in FA, even in normal iron status, the Fe-S cluster formation is diminished, and other iron regulatory system (IRP1, TfR, and ferritin) is also dysregulated. Interestingly, a recent study showed that a long-term running exercise has an effect to restore IRP1 expression without changing FXN levels using KIKO mice, a Knock in-Knock out FA mouse model (81), suggesting that a molecular mechanism of mitochondrial iron homeostasis may be independent of FXN in FA.

Reactive oxygen species (ROS) are mainly produced from the mitochondrial electron transport chain (ETC) as byproducts of energy production. In FA, it has been reported that ROS are produced in an uncontrolled manner, which can react with proteins, nucleic acids, and lipids, causing their damages (132). In FXN-deficient cells, Fe-S cluster formation is ablated in the ETC, which not only increases electron leakage but also results in ROS production in the mitochondria (133). For instance, the ratio of reduced to oxidized glutathione (GSH/GSSH) was significantly decreased in FA cells (134), suggesting FA-induced increase in oxidative stress (e.g., H₂O₂ production). The increased H₂O₂ and free iron in the mitochondria may facilitate the production of hydroxyl radical (•HO) via the Fenton reaction (132). However, hydroxyl radicals are not the only ROS that are produced in FA cells, and the peroxides also react with membrane lipids to produce carbon centered radicals (R•) that will further react with oxygen to produce reactive hydroperoxyl radicals (ROO•) (133). They are highly reactive species that can transfuse through membranes, which causes oxidative stress-induced damages. Furthermore, it has been shown that a loss of FXN in FA dorsal root ganglia blunts the expression of Nrf2 (Nuclear Factor Erythroid 2-related Factor 2), which in turn inhibits the expression antioxidants such as glutathione S-transferase, peroxiredoxin, and glutaredoxin (135). Together these changes decrease the antioxidative capacity and sensitize the cells to oxidative stress.

Neurodegeneration in FA is known to lead speech impairment, progressive ataxia, and an overall decrease in cognitive capabilities (135). It has been shown that FA-linked neurodegeneration is associated with defects in Fe-S cluster assembly unit where FXN plays a key role for the cluster formation with other proteins (136). Nevertheless, myopathy patients with Fe-S cluster assembly mutations are found to have no neurodegenerative symptoms, although they have FA-like characteristics such as iron accumulation as well as decreased aconitase activity (137). Thus, it may be plausible that other factors are responsible for FA-linked neurodegeneration other than FXN-based Fe-S cluster formation unit. More studies are required to elucidate the mechanisms of disease.

Furthermore, ferroptosis has been identified as a possible mechanism for neurodegeneration in FA patients (138, 139). In neuronal cells, Nrf2 (a key regulator of ferroptosis) has been shown to regulate expression of several oxidative stress resistance genes such as HO-1, thioredoxin (Trx), and glutamate cystein ligase (GCL) (135, 140, 141). Although Nrf2 is usually located in the cytoplasm, it is translocated to the nucleus under oxidative stress where it binds to the antioxidant-responsive element (ARE) in the DNA and induces transcription of antioxidant/cytoprotective genes (140). Both in vivo and in vitro studies have shown that Nrf2 expression is dependent on FXN levels in neuronal cells and tissues (138, 142). In FA, lower FXN levels appear to downregulate Nrf2 expression in neurons, which renders these cells more susceptible to oxidative damage and results in ferroptosis (138, 143). These studies suggest Nrf2-linked ferroptosis as a chief pathway for FA neurodegeneration (138, 142). Other than its role in ferroptosis, Nrf2 has also been shown to be important for neurogenesis (143). A recent study shows that reduced Nrf2 expression in FA neural stem cells inhibits impairment in neurogenesis (138, 143). However, upon recovery of Nrf2 function, the proliferation and regeneration pattern of these stem cells were significantly improved (138, 143). Recovering of Nrf2 also triggers axonal re-growth, which can possibly help in the formation of better neural networks (143). The increasing evidence for the role of Nrf2 in FA neurodegeneration suggests its possibility as a therapeutic target. On the other hand, oxidative stress has been shown to alter the mitochondrial fission/fusion balance in FA (144), so it is plausible that restoration of Nrf2 expression could correct the impaired mitochondrial dynamics possibly by alleviating oxidative stress, although more studies are needed to verify this idea.

Currently, pharmacological treatments for FA include agents that are designed to reduce oxidative stress as well as to enhance mitochondrial function (e.g., idebenone, coenzyme Q₁₀, A001, and omaveloxolone), anti-inflammatory agents (e.g., methylprednisolone), iron chelator (e.g., deferiprone), FXN modifiers or mimetics (e.g., carbamylated erythropoietin, ubiquitin competitors, and CTI-1601), and agents that induce FXN gene expression (e.g., RG2833, nicotinamide, interferon gamma, and resveratrol) (145). Recently, mitochondrial fragmentation (i.e., DRP1) has been identified as a therapeutic target for FA patients, as reversal of mitochondrial fragmentation appeared to restore impaired bioenergetics and ATP production in FA patient-derived cells (146). Nevertheless, further studies are required to determine the efficacy and safety of these therapeutic options in FA patients.

4.2. Huntington’s Disease

Huntington’s Disease (HD) is a progressive neurodegenerative disease caused by the expansion of a CAG trinucleotide repeat in exon 1 of the huntingtin (HTT) gene (147). Western European ancestry are more vulnerable to HD, with the prevalence of ~12 individuals per 100,000, due to their genetic differences in the HTT gene (148). Typical clinical symptoms of HD include chorea (involuntary movements), neuropsychiatric disturbances, dementia, and weight loss (149), which may be related to impaired mitochondrial ATP synthesis (150). Mutant HTT not only induces neuronal loss in the basal ganglia, but also leads to movement and cognitive deficits (151). In early stage of HD, destruction of neurons and astrogliosis in the striatum is reported to accompany iron overload (152, 153). As the disease develops to the advanced stage, the neuronal damage starts to affect the cortex and hippocampus, as well as respiratory chain (154, 155).

Studies have reported that healthy HTT is responsible for mediating endocytosed Fe²⁺ that is required for oxidative energy production (149, 156). Therefore, its dysregulation is responsible for the functional impairment of mitochondria. For example, multiple studies have reported that HTT mutation induced-neuronal death is linked with disrupted mitochondria energy metabolism, including decreased mitochondrial respiratory complexes II-IV and aconitase activity (157, 158), abnormal mitochondrial trafficking (159), and impaired mitochondrial ATP production (160). Since HTT is involved in nuclear gene expression, its mutation is believed to be correlated with disrupted transcription, as well as import of mitochondrial proteins (161–164). Mutant HTT is likely to increase dynamin-related protein-1 (Drp1), a mitochondrial fission GTPase protein, in the postmortem brain of HD patients (165), which results in favor of fission in the fission-fusion dynamic balance of mitochondria (166).

Iron overload, in the case of HD, appears to be an early event instead of a founding event (151) (Fig. 3B). Magnetic resonance imaging (MRI) studies showed changes in brain iron metabolism in early stage of HD (167, 168). Iron accumulation is observed in the basal ganglia (153, 169), microglia (170), and myelinating oligodendrocytes of HD patients, which further leads to increased oxidative stress and damage to adjacent neurons (171). Consistent with HD patients, HD animal models also show increased iron and ferritin levels in neurons and glia (170, 172). One hypothesis of the relationship between increased brain iron and HD is the dysregulated iron homeostatic pathways that are affected by mutant HTT. Niu et al. demonstrated that mutation of HTT induces upregulation of IRP1, Tf, and TfR in both in vivo and in vitro studies, along with increased iron levels in the striatum and cortex in 12-week-old HD mice, although no HD symptoms were observed at this age (173). Multiple studies using different chelators have been conducted to investigate whether the progression of HD can be ameliorated by removal of excess brain iron. Notably, chelation of excess brain iron by deferoxamine (DFO) improves behavioral and pathological symptoms in HD mice (172) and HD cells (174), respectively. Another metal-binding compound clioquinol was shown to ameliorate neurodegenerative symptoms in HD mice, as well as to downregulate mutant HTT expression in PC12 cells (175). However, most studies are limited to early-stage HD, and more studies are needed to determine the role of iron in late-stage HD.

Fig. 3: Neurodegenerative diseases with brain iron accumulation.

A: Friedreich’s ataxia. In Friedreich’s ataxia (FA), mutations of the mitochondrial protein frataxin (FXN) lead to impaired Fe-S cluster biogenesis which further leads to disruption of iron metabolism along with increased mitochondrial iron contents. Excessive accumulation of iron leads to the production of reactive oxygen species (ROS), causing lipid peroxidation and cell death. Mutation of frataxin has also been shown to impair the translocation of nuclear factor erythroid 2-related factor 2 (Nrf2), a key player for antioxidant gene expressions, into the nucleus, which not only impairs antioxidant response, but also contributes to the ROS production.

B: Iron dysregulation in Huntington’s disease. In Huntington’s disease (HD), the presence of mutant Huntington (mHTT) leads to increased sensitivity of NMDA receptors, which not only increases influx of extracellular calcium (Ca²⁺), but also results in the activation of neurotoxic Rhes, rendering more toxic soluble mHTT. The mHTT interacts with mitochondria and it results in the impaired mitochondrial membrane permeability as well as the efflux of Ca²⁺ from mitochondria. Increased cytosolic Ca²⁺ leads to the activation of neuronal nitric oxide synthase (nNOS), which binds to NMDA receptors via Dexras1, releasing NO that eventually forms peroxynitrite radicals in the presence of superoxides. The nNOS also binds to divalent metal transporter 1 (DMT1), which results in enhanced Fe²⁺ influx, as well as increased oxidative stress via the Fenton reaction, while interacting with H₂O₂ released from superoxide dismutase (SOD). These reactions ultimately lead to increased lipid peroxidation and neuronal cell death.

C: Iron dysregulation in Alzheimer’s disease. Alzheimer’s disease (AD) is characterized by the aggregation of Aβ plaque and phosphorylated Tau protein (neurofibrillary tangle, NFT). In normal condition, APP stabilizes ferroportin (FPN) to facilitate iron export, and Tau protein supports the trafficking of APP transport to the cell surface. In AD, Tau is phosphorylated and aggregated to NFT, leading to less Tau available to facilitate APP transport, and APP is cleaved to generate Aβ, so the iron efflux via FPN is impaired, leading to iron accumulation in cells. Iron can bind to Aβ and promotes its aggregation, leading to disrupted mitochondrial membrane permeability, producing oxidative stress within the mitochondria. Iron dysregulation also alters Aβ metabolism at post-transcriptional level and causes DNA damage and mutation. Taken together, these disruptions lead to increased release of Ca²⁺, apoptosis-inducing factor (AIF) and cytochrome c, resulting in apoptosis and neuronal cell death.

D: Iron dysregulation in Parkinson’s disease. In Parkinson’s disease (PD), Fe²⁺ levels are elevated by both increased DMT1-mediated import and neuroinflammation, which leads to oxidative damage by the Fenton reaction. The reduction from Fe³⁺ to Fe²⁺ is accelerated in the presence of α-synuclein, while the oxidation from Fe²⁺ to Fe³⁺ is inhibited due to a lack of ceruloplasmin (CP). The Fe²⁺ overload induced-oxidative stress exacerbates the aggregation of misfolded α-synuclein, by which Lewy bodies are accumulated and it could be the cause of dementia. Disrupted Tf/TfR2 transport system can lead to an iron accumulation in mitochondria, which may induce mitochondrial dysfunction along with mitochondrial oxidative stress. Another characteristic of PD is decreased release of dopamine, which also promotes the misfold of α-synuclein, either directly or indirectly leading to neuronal cell death.

Apart from iron accumulation, impaired iron regulatory system is likely to be linked with mitochondrial dysfunction in the brain of HD patients where it is featured by increased expression on mitoferrin 2 (MFRN2) (176), a mitochondrial iron importer from the cytosol (36). To test whether mitochondria are a key site of iron dysregulation in HD, Agrawal et al. found increased labile iron in the mitochondria of both human and mouse HD brain at an advanced stage, which was shown to be rescued by a chelating agent, deferiprone (DFP), suggesting that mitochondrial labile iron overload might be a mediator of HD mitochondrial dysfunction, as well as disease progression (149). In addition, the expression patterns of heme-based mitochondrial respiratory chain complexes and Fe-S cluster enzymes are altered in the striatum of HD patients (177, 178). Also, FXN was found to be downregulated in the brains of mouse and human HD subjects, which is a common feature of FA (described above) (149). It appears plausible that both deficient FXN and oxidative stress could play an important role in impaired mitochondrial Fe-S cluster biosynthesis in HD (159).

4.3. Alzheimer’s Disease

Alzheimer’s Disease (AD) is a chronic multifactorial neurodegenerative condition associated with progressive dysfunction and degeneration of neurons, which leads to progressive cognitive decline and dementia (179). In the brain of AD patients, the accumulation of extracellular β-amyloid (Aβ) plaques and the deposit of neurofibrillary tangles (NFTs) formed by hyperphosphorylation of tau (pTau) proteins are considered to be the main pathophysiological features. The etiology of AD is diverse with a combination of genetic and environmental factors so that the specific mechanisms of its onset and corresponding treatment are still unclear. Growing evidence has shown that iron plays an important role in the early progression of AD and is involved in the development of Aβ plaques and NFTs (180).

Abnormal iron levels result in diminished iron regulatory enzyme activity, which not only increases Aβ production through its binding with amyloid precursor proteins (APP) and tau proteins (181), but also promotes oxidative stress and cytotoxicity along with the plaques (182, 183). The intracellular iron has been reported to accelerate Aβ deposition and production at the post-transcriptional level. For instance, the increased iron level upregulates the expression of APP proteins via the regulation of the atypical IRE in 5’UTR of APP transcripts, leading to increased Aβ level as well as aggregation (184–186). However, it has been argued that the Aβ conjugated with iron could be a protective response of the neuron from oxidative stress through sequestering iron (187, 188). Nevertheless, the cellular mechanisms of iron in the progression of AD are still unclear, which warrants further studies.

Similar to Aβ, the aggregation of pTau leads to the deposition of NFTs via tauopathy, which is induced by Fe³⁺ (189). In physiological conditions, tau promotes the stabilization and regulation of microtubule and axonal transport, which is critical for neurotransmission as well as iron homeostasis (190). However, upon CDK5/P25 complex and GSK-3b, the Fe³⁺-induced pTau hinders the microtubule stabilization in NFTs (191). Meanwhile, the aggregation of tau in NFTs induces the expression of HO-1 and also increases the release of Fe²⁺ (192), which elevates oxidative stress and further promotes the aggregation of pTau (193). In addition to cellular damage, the iron-associated tau protein deposition hampers iron export, which in turn results in iron trapping in the cells and leads to a vicious cycle (194). Therefore, iron chelation therapies have been applied and shown to have promising efficacies on ameliorating AD symptoms. For example, Guo et al. revealed that DFO has a significant effect to inhibit the iron-induced pTau aggregation in an AD mouse model (191). Other chelators such as clioquinol and 5,7-dichloro-2-((dimethylamino)methyl)quinolin-8-ol (PBT2) were also shown to decrease Aβ1–42 levels and improved cognitive function in AD patients (195, 196).

In addition, oxidative stress has been suggested to contribute to the production of Aβ (197, 198), which appears to be linked with abnormal mitochondrial function and morphology. These dysregulated mitochondria have been identified in the early stage of AD, which is featured by reduced ATP production in the brain (8). In the mitochondria, for example, an active γ-secretase complex cleaves APP into Aβ, which inhibits mitochondria complex IV through Aβ-bound alcohol dehydrogenase (ABAD) that produces ROS (192). Thus, mitochondria are recognized as major organelle for ROS and oxidative stress (151), but further studies are required to better understand how iron-related oxidative stress is linked with mitochondrial dysfunction in AD (Fig. 3C).

4.4. Parkinson’s Disease

Parkinson’s Disease (PD) is the second common neurodegenerative disease of the elderly, in which dopaminergic neuronal loss and iron accumulation in the substantia nigra (SN) pars compacta have been recognized as pathological features (199). Iron accumulation within the SN has been well reported using different analytical methods, including inductively-coupled plasma mass spectrometry (ICP-MS) (190), MRI (200), and atomic absorption spectroscopy (201, 202). The exact molecular mechanism of PD has not yet been fully understood due to the complex pathology of the disease, although other factors such as misfolded proteins, impaired ubiquitin-proteasome, dysfunction of mitochondria, and oxidative stress are considered for the pathogenesis of PD (203).

In PD, iron binds to α-synuclein, an abundant protein in dopamine neurons, and accelerates its aggregation, causing the production of toxic hydroxyl radical in the SN (12). In the affected dopaminergic neurons, iron is also thought to contribute to the formation of Lewy bodies, which are intraneuronal proteinaceous inclusions that result in an aggregation of α-synuclein and are considered to be a pathological hallmark of PD (199). This clustered α-synuclein disturbs cytosolic and mitochondrial environments and dopamine transporters, which further results in activation of cell death (183) (Fig. 3D). Moreover, PD patients also show increased levels of ferritin-loaded microglia in the SN, where degenerating and neuromelanin-loaded dopaminergic cells are shown to have more reactive microglia (204). The upregulation of both glial cells by the debris and toxins released by dying neurons appear to contribute to increased neuromelanin, which, upon binding to iron, leads to the release of neurotoxic factors including tumor necrosis factor-α, inflammation marker interleukin-6 (IL-6), and nitric oxide (183). The immunoreactivity of HO-1 is also reported to be increased in the neuropil of the substantia nigra (205).

As a redox-active iron pool, mitochondrial dysfunction in PD has been studied as early as in the 1980s, when four college students developed Parkinson-like symptoms after the injection of drugs that contain MPTP, a compound whose metabolite MPP+ inhibits the mitochondrial complex I capacity (206, 207). More studies have also confirmed dysregulated mitochondrial complex I activity and content levels in the brain of PD patients (208, 209). This lessened mitochondrial complex I capacity seems to elevate mitochondrial iron uptake and ROS production (210). It has been identified that iron chelation- and gene-based therapies are effective to restore ferritin and to further rescue the parkinsonism symptoms in animal models of PD (12). In a pilot clinical trial with early-stage PD patients, the DFP treatment reduced iron overload in the substantia nigra and significantly improved the motor ability (211), indicating that chelation therapies have the potential to slow down the disease progression and provide treatment options for further long-term therapy.

PD has been shown to compromise iron regulatory system along with mutations in iron-related proteins such as Tf (212), IRP2 (213), ferritin (214), and DMT1 (215). For instance, ferritin levels are decreased in the brain of PD patients, indicating reduced iron storage capacity (216). DMT1 appears to be increased in nigral dopamine neurons and microglia, which may not only contribute to the increased capacity of Fe²⁺ transport from the extracellular space into the cytosol, but also increase the production of ROS (217). Therefore, iron accumulation is likely to exacerbate mitochondrial dysfunction in PD patients (199).

4.5. Neurodegeneration with Brain Iron Accumulation (NBIA)

Neurodegeneration with brain iron accumulation (NBIA) is a group of rare inherited monogenic diseases characterized by iron accumulation in different parts of the brain, especially within the basal ganglia, with the symptoms of painful dystonia, parkinsonism, neuropsychiatric abnormalities, and early death. There is no cure for NBIA disorders. NBIA displays a marked genetic heterogeneity, wherein the mutations of each specific gene lead to the failure of each protein for essential cellular functions. Ten different subtypes of NBIA are summarized in Table 1 (218, 219). The majority of these genes are found in the mitochondria, including PANK2, CoPAN, PLA2G6, C19orf12, and ATP13A2, and it is thus important to understand iron regulatory mechanisms and its connections between mitochondria function and neurodegeneration in the brain (220).

Table 1.

Subtypes and associated genes of NBIA (218, 219)

NBIA disorder	Gene	Symptoms
PKAN (Pantothenate Kinase-Associated Neurodegeneration)	PANK2	Classic PKAN: gait problems, progressive dystonia, retinal degeneration, dysarthria, rigidity, spasticity, hyperreflexia, and extensor toe signs Later onset PKAN: speech difficulty, psychiatric symptoms
PLAN (PLA2G6-Associated Neurodegeneration)	PLA2G6	INAD (infantile neuroaxonal dystrophy): hypotonia, spasticity, optic atrophy aNAD (atypical neuroaxonal dystrophy): speech delay, dystonia, behavior changes PLA2G6-related dystonia-parkinsonism: dystonia, neuropsychiatric changes, slowness, poor balance and rigidity
MPAN (Mitochondrial membrane Protein-Associated Neurodegeneration)	C19orf12	Dystonia, spasticity, weakness, optic atrophy, and neuropsychiatric changes
BPAN (Beta-propeller Protein-Associated Neurodegeneration)	WDR45	Childhood development delay and seizures, slow motor, and cognitive gains (mutant on the X chromosome, lethal in most males)
FAHN (Fatty Acid Hydroxylase-associated Neurodegeneration)	FA2H	Leg dystonia, weakness and falling, optic atrophy, dysphagia (difficulty swallowing), progressive intellectual impairment
CoPAN (COASY Protein-Associated Neurodegeneration)	COASY	Early in the disease: spastic-dystonic paraparesis in lower limbs Later in the disease: dystonia of the mouth and jaw, speech problems caused by dysarthria
Woodhouse-Sakati Syndrome	DCAF17	In Saudi Arabian population: dystonia, hair loss, diabetes, hearing loss, gonadal dysfunction, and mental retardation
Neuroferritinopathy	FTL	Dystonia, chorea (jerky movements), subtle cognitive decline
Aceruloplasminemia	CP	Iron accumulation in brain as well as other organs, retinal degeneration, diabetes, ataxia, involuntary movement, anemia
Kufor-Rakeb syndrome	ATP13A2	Juvenile parkinsonism, dementia, abnormal eye movements and involuntary jerking of facial and finger muscles

Although being called NBIA, only two subtypes are directly related to iron metabolism: aceruloplasminemia (ACP), which is associated with recessive mutations in ceruloplasmin (CP), and neuroferritinopathy, which is associated with mutations in ferritin light chain 1 (FTL1). These two proteins and their subtypes are involved in different mechanisms, such as lipid metabolism, mitochondrial function, coenzyme A (CoA) metabolism, and autophagy (6).

Ceruloplasmin, encoded by CP gene, is the only existing ferroxidase that is expressed as a glycosylphosphatidylinositol (GPI)-linked form in astrocytes and catalyzes the peroxidation of Fe²⁺-Tf to Fe³⁺-Tf (220, 221). Aceruloplasminemia (ACP) that is caused by the mutation of CP gene induces the absence of the corresponding protein in the plasma and is characterized by extrapyramidal symptoms, ataxia, progressive CNS, and retinal neurodegeneration (183, 222). The mutations of the CP gene cause neuronal structural modifications, so that the type I copper binding site fails to incorporate copper into the apo-CP, an unstable apo form of CP that will be rapidly degraded in the plasma, leading to retention of CP proteins in the endoplasmic reticulum (ER), as well as impaired ferroxidase activity (223, 224). Consequently, iron enters the CNS as Fe²⁺ where it is not oxidized, leading to excess NTBI. In addition, mutations of CP also contribute to the destabilization of FPN on the cell surface, so that iron export is impaired (223). Collectively, in ACP patients, cellular iron is accumulated in astrocytes and is not delivered to the neurons, which causes neuronal cell death due to iron deficiency (225).

Neuroferritinopathy is an autosomal inherited, late-onset basal ganglia disease linked to mutations in FTL1 that is one of the two ferritin subunits. Individuals with neuroferritinopathy show nuclear and cytoplasmic ferritin accumulation in both neurons and glia of the striatum, as well as cerebellar cortex, which contributes to increased cellular labile iron pool along with elevated ROS production and oxidative stress (183, 226, 227). Consistently, it has been suggested that the iron accumulation-linked oxidative stress causes mitochondrial dysfunction and cell damage, as in the neurodegenerative rodent model, markers of oxidative stress (e.g., lipid peroxidation) are increased along with the dysregulated iron regulatory system (e.g., TfR1 and IRP1) (228).

5. CONCLUSIONS

There has been increased attention on the roles of iron in the CNS since mitochondrial function and structure are largely dependent on the optimal range of iron level in the central tissues. In particular, neurodegenerative diseases have shown to be notably linked with irregular iron status, which also causes impaired mitochondrial function along with oxidative stress. Here, we discussed how dysregulated iron homeostasis alters the mitochondrial system in multiple neurodegenerative diseases: Alzheimer’s disease, Huntington’s disease, Parkinson’s disease, neurodegeneration with brain iron accumulation, and Friedreich’s ataxia. Remarkably, we underscore that iron overload may be an important causal and/or susceptibility factor for these neurodegenerative diseases through impinging on mitochondrial functions and structures. The present review also summarized current understanding on therapeutic targets of these neurodegenerative diseases, while focusing on the mitochondrial iron system. In addition to approaches with various iron chelators, we have also discussed mitochondria-specific targets that include mitochondrial dynamics (i.e., fission and fusion), as well as mitochondrial iron regulatory system. More mechanistic and systematic studies are required to better understand the underlying mechanisms of neurodegenerative diseases associated with iron overload and associated mitochondrial dysfunction, and these efforts will help to identify novel therapeutic strategies that can prevent the development of neurodegenerative diseases and ameliorate disease symptoms or slow down the disease progression.

Abbreviations:

ACP

aceruloplasminemia

AD

Alzheimer’s disease

APP

amyloid precursor protein

BBB

blood-brain barrier

BCEC

brain capillary endothelial cells

CNS

central nervous system

CP

ceruloplasmin

DMT1

divalent metal transporter 1

ER

endoplasmic reticulum

ETC

electron transport chain

FPN

ferroportin

FA

Friedreich’s Aataxia

FTH

ferritin heavy chain

FTL

ferritin light chain

FtMt

mitochondria specific ferritin

FXN

frataxin

GPX

glutathione peroxidase

GSH

glutathione

HD

Huntington’s disease

HO-1

heme oxygenase 1

HTT

huntingtin

IMM

inner mitochondrial membrane

IRE

iron responsive element

IRP

iron regulatory protein

MFRN

mitoferrin

NBIA

neurodegeneration with brain iron accumulation

Nrf2

nuclear factor erythroid 2-related factor 2

NTBI

non-transferrin-bound iron

OMM

outer mitochondrial membrane

PD

Parkinson’s disease

ROS

reactive oxygen species

SN

substantia nigra

TBI

transferrin-bound iron

Tf

transferrin

TfR1

transferrin receptor 1

Tim-2

T-cell immunoglobulin mucin domain 2 protein

UTR

untranslated region