Regulation of mitochondrial iron homeostasis in tumor cells

Abstract

Iron is an indispensable trace element for all living organisms, existing in the form of free iron (Fe²⁺/Fe³⁺) and bound iron, and it plays a crucial role in a variety of cellular processes, including biomolecule synthesis, epigenetic regulation, immune modulation, cellular senescence, and mitochondrial respiration. Iron homeostasis is meticulously regulated within biological systems to avert the detrimental effects of both iron overload and deficiency, with imbalances leading to a multitude of diseases. Numerous studies have underscored the importance of rebalancing iron homeostasis in cancers, emphasizing its crucial role in tumorigenesis. Within the cell, mitochondria serve as the central hub for iron metabolism, primarily accountable for the synthesis of heme and iron-sulfur (Fe-S) clusters, the storage of excess iron via mitochondrial ferritin, and the regulation of iron-dependent cell death pathways such as ferroptosis and cuproptosis. Despite decades of in-depth research into the biological functions of iron and its homeostatic regulation, numerous scientific questions remain unresolved. This review offers a comprehensive and integrated analysis of the detailed regulatory mechanisms of mitochondrial iron metabolism and its profound influence on cancer metabolism. Based on the current research progress, we have summarized the challenges and limitations in this field and proposed new conceptual frameworks and research directions to enhance our understanding of mitochondrial iron biology in cancer.

Introduction

Iron, as the most abundant trace metal element in the human body, exists in various forms, including free iron (Fe²⁺), iron-containing proteins and enzymes, Fe-S cluster-containing proteins, and heme-containing proteins, etc. Iron deficiency is closely related to a variety of diseases, including iron deficiency anemia (IDA), pica, neurocognitive development problems in children, immunodeficiency, and restless legs syndrome (RLS) (Borgna-Pignatti and Zanella 2016; Cheng et al. 2022). However, excessive intake of iron or abnormal accumulation of iron in the body can also lead to a myriad of health issues, such as hemochromatosis, cardiovascular and endocrine system impairments, liver and joint damage, and a heightened susceptibility to diabetes and cancer (Sawicki et al. 2023; Mahmoud et al. 2021; Sohal and Kowdley 2024; EASL Clinical Practice Guidelines on haemochromatosis 2022). Therefore, systemic and cellular iron homeostasis is essential for life and health.

Iron in the human body is primarily derived from dietary sources, which include two absorbable forms: heme iron—derived from hemoglobin and myoglobin in red meat—and non-heme iron, predominantly found in plant-based foods such as vegetables and cereals, which is released during digestion (Li et al. 2017). Approximately 80% of absorbed iron is used for hemoglobin synthesis during erythropoiesis. In non-erythroid cells, the cellular uptake, storage, and utilization of iron are precisely regulated to meet the metabolic requirements of the cell while preventing the toxic effects of iron overload. Most of the iron within the cell is directed into the mitochondria for the synthesis of heme and iron-sulfur (Fe-S) clusters, which in turn participate in many important biological functions. Consequently, mitochondrial dysfunction can disrupt cellular iron homeostasis through a cascade of secondary effects.

Within various types of tumor cells, there is an increased requirement for iron to support heightened metabolic activity and the rapid cell proliferation. Extensive evidence indicates that iron metabolism is intimately linked to tumor initiation and progression (Guo et al. 2021). These mechanisms underscore iron’s pivotal role in cancer progression and highlight its potential as a therapeutic target.

This review summarizes the recent advances in iron homeostasis research, with a particular emphasis on elucidating the regulatory mechanisms of mitochondrial iron metabolism and its intricate role in the progression of cancer. Additionally, we discuss current challenges in mitochondrial iron metabolism research and propose new conceptual frameworks and research directions to advance progress in this field.

Cellular iron homeostasis regulation

Cellular iron is essential for oxygen transport, energy metabolism and DNA synthesis. However free iron can also be toxic by generating reactive oxygen species (ROS). The labile iron pool (LIP)—chelatable, redox-active iron (Fe²⁺ and Fe³⁺) not bound to storage or transport proteins—is widely used to assess iron supply–demand balance in cells and also represents the metabolically active fraction of cellular iron (Fig. 1). Cellular LIP levels are in the low micromolar range but rise markedly during systemic iron overload, making them a key target for chelator screening and protection against oxidative damage. In iron sensing and feedback regulation, the LIP acts as a central hub that links iron uptake, storage, utilization, and export. Pathological LIP elevation in specific organs or cells usually results from impaired iron utilizations such as defects in Fe-S cluster assembly or faulty iron storage—which drive labile iron accumulation and promote oxidative damage via Fenton chemistry. In susceptible contexts, this can culminate in ferroptosis, an iron dependent form of regulated cell death driven by lipid peroxidation (Dixon et al. 2012; Lei et al. 2024; Zhou et al. 2024; Zhang et al. 2022; Jiang et al. 2021a). Therefore, understanding the mechanisms that govern cellular iron balance is fundamental not only for basic physiology but also for pathological conditions involving iron dysregulation.

Fig. 1.

Intracellular iron metabolism process. LCN2 is a secreted protein that has a high affinity for iron bound to iron carriers. SLC22A17 is an LCN2 transporter expressed in various cell types. LCN2 is internalized by binding to the SLC22A17 receptor, leading to the release of iron and subsequently increasing intracellular iron concentration. Iron ions form a complex with transferrin to become the complex Tf-Fe³⁺, which enters the cell via TfR1-mediated endocytosis. Under acidic conditions, iron ions are released from Tf, reduced to Fe²⁺ by STEAP3 reductase, and transported into the cytoplasm by DMT1. With the mediation of Dcytb, vitamin C provides electrons to extracellular ferric iron, reducing it to ferrous iron, which enters the cell via DMT1 and is oxidized to DHA itself. During autophagy, NCOA4 guides the degradation of Ferritin, releasing iron ions into the iron pool. Free iron within the cell is transported into the mitochondria by MFRN proteins, performing various functions, including heme synthesis, iron-sulfur cluster assembly, and mitochondrial ferritin assembly. The functions of intracellular free iron include the Fenton reaction, gene regulation, epigenetic modification, and mitochondrial respiration. The export of iron from the cell is accomplished by FPN, which is considered the only pathway for iron to enter the plasma from the cell. Heme Absorption: Red blood cells are cleared through erythrophagocytosis, releasing heme. HRG-1 protein is involved in the transport of heme from the lysosome to the cytosol and may contribute to the regulation of the intracellular heme pool. Hemopexin binds free heme to form a heme-hemopexin complex, which undergoes endocytosis via receptors such as CD91 and is subsequently degraded in lysosomes, releasing heme for cellular use or recycling. Heme can be transported to lysosomes for digestion or transported via endocytosis by FLVCR2 protein. Heme is degraded through the HO-1/2 pathway, with products including carbon monoxide, biliverdin, and Fe.²⁺.Ferritin: Ferritin plays a key role in regulating iron storage and release, preventing the toxic effects of intracellular free iron. Fenton Reaction: The Fenton reaction plays a complex role in iron metabolism, both promoting the recycling and storage of iron and potentially triggering oxidative stress, affecting iron transport and utilization. FPN: Ferroportin, FPN is considered the only pathway for iron to enter plasma from the cell; LCN2: Lipocalin 2; TfR1: Divalent Metal Transporter 1; STEAP3: Six Transmembrane Epithelial Antigen of the Prostate 3; MT1: Divalent Metal Transporter 1; DHA: Dehydroascorbic Acid; NCOA4: Nuclear Receptor Coactivator 4; MFRN: Mitoferrin; FPN: Ferroportin; HRG-1: Highly Regulated in Cancer 1; FLVCR2: Folate Receptor 2; HO-1/2: Heme Oxygenase 1/2

Absorption of iron

Tf-TfR1-STEAP3-DMT1 system

In plasma, iron binds reversibly to the carrier protein transferrin (Tf), forming diferric transferrin (Tf-Fe³⁺), which is specifically recognized by transferrin receptor 1 (TfR1). The Tf-Fe³⁺-TfR1 complex is then internalized via clathrin-mediated endocytosis (Wang and Pantopoulos 2011). Acidification of the endosome releases Fe³⁺ from transferrin, and both apo-transferrin (apo-Tf) and TfR1 are recycled to the plasma membrane (Collawn et al. 1990; Cheng et al. 2004). This pathway requires multiple trafficking molecules such as membrane transporters ZIP14 and mucolipin-1, vacuolar protein sorting-associated protein 35 (VPS35), sorting nexin-3 (SNX3), as well as the vacuolar-type H⁺-ATPase (V-ATPase) and its assembly factor coiled-coil domain containing 115 (CCDC115) (Chen et al. 2013; Sobh et al. 2020); however, the precise mechanism of acidification in endocytic vesicles is still unclear at present. Fe³⁺ is reduced to Fe²⁺ by members of the six-transmembrane epithelial antigen of the prostate (STEAP) family or via ascorbate-dependent reduction within the endosome, and the resulting Fe²⁺ is transported into the cytosolic LIP by divalent metal transporter 1 (DMT1) (Sendamarai et al. 2008).

Notably, Tf circulates with 20–50% of its iron-binding sites occupied, and the di-ferric isoform exhibits the highest affinity (TfR1). Clinically, iron status is gauged via unsaturated iron-binding capacity and transferrin saturation. TfR1 governs cellular iron uptake; its activity is suppressed by stearic acid (C18:0)-catalyzed stearoylation, a modification that concurrently accentuates mitochondrial fragmentation (Senyilmaz et al. 2015). TfR1 expression is markedly up-regulated in multiple malignancies (pancreas, breast, bladder, lung) to satisfy the heightened iron demand of proliferating cancer cells, prompting therapeutic exploitation of anti-TfR1 antibodies (Marques et al. 2016). Emerging data implies TfR1 participation in biological processes beyond iron metabolism (Jeong et al. 2016; Chen et al. 2015).

Dcytb mediated Fe²⁺ absorption

Duodenal cytochrome b (Dcytb, also known as CYBRD1) is a member of the human cytochrome b561 family, and serves a vital role in the ascorbate-dependent reduction of inorganic iron in duodenal enterocytes. It uses cytosolic ascorbate (also known as Vitamin C) as the electron donor to catalyze the reduction of extracellular Fe³⁺ to Fe²⁺ during the process of iron absorption (Zhou et al. 2025). Fe²⁺ is then transported into cytosol by DMT1, SLC39A14 (ZIP14) (Paterek et al. 2019; Liuzzi et al. 2006) or SLC39A8 (ZIP8) (Wang et al. 2012).

Dcytb is widely expressed in human tissues and highly expressed in the duodenal brush border membrane. Its ferric reductase activity is believed to play a physiological role in dietary iron absorption. However, it’s also reported that Dcytb is not an essential component of the intestinal iron absorption apparatus in mice even in the case of iron deficiency (Gunshin et al. 2005). Therefore, the detailed in-vivo function of Dcytb requires further investigation. Meanwhile, it has been demonstrated that hypoxia can strongly upregulate Dcytb expression through a hypoxia-inducible factor-2α (HIF-2α)-dependent mechanism, which contributes to cancer development through increased iron absorption. Notably, other iron-regulated proteins, such as DMT1, ferroportin (FPN), and nuclear receptor coactivator 4 (NCOA4), have also been reported as transcriptional targets of HIF-2α (Sendamarai et al. 2008; Berezovsky et al. 2022; He et al. 2022). Evidence shows that Dcytb expression is closely correlated with patient prognosis (Chen et al. 2021; Lemler et al. 2017; Brookes et al. 2006).

Heme iron uptake and degradation

Heme is a type of porphyrin compound containing iron ions. As a key protein cofactor, it plays an important role in cellular energy metabolism and signal transduction. Heme in bloodstream mainly comes from the aged or damaged red blood cells digested by macrophages. Although the molecules mediating heme uptake has not been identified, considerable evidence indicates that heme enters cells through endocytosis mediated by specific receptors, such as CD163 and low-density lipoprotein receptor-related protein 1 (LRP1 or CD91) (Schaer et al. 2006; Wang et al. 2017a), or through heme importer, such as feline leukaemia virus subgroup C receptor2 (FLVCR2), heme responsive gene-1 (HRG1) and heme carrier protein 1 (HCP1) (Rajagopal et al. 2008; Duffy et al. 2010). The cellular heme can be catabolized into biliverdin, carbon monoxide and free iron by heme oxygenase 1 or 2 (HO-1/2) (Poss and Tonegawa 1997).

Iron uptake by Lipocalin2 (LCN2) and SLC22A17

LCN2, a secreted protein featuring a beta barrel structure, has a high affinity for ferric iron complexed with siderophores (Yang et al. 2002). SLC22A17 is an LCN2 transporter expressed in diverse cell types. LCN2 is internalized upon binding to the SLC22A17 receptor, resulting in the release of iron and a subsequent increase in intracellular iron concentration.

Iron storage and distribution

The LIP, which consists of unused or retained iron not utilized by the cell, is stored and detoxified within ferritin in the cytoplasm. This protein is composed of 24 subunits of heavy ferritin chain (FTH1) and light ferritin chain (FTL), forming a hollow shell-like structure that can accommodate up to 4,500 trivalent iron atoms. In normal cells, excess iron will be stored in ferritin to prevent excessive production of ROS. When cellular iron utilization is required, ferritinophagy is induced to degrade ferritin stores, thereby liberating bioavailable iron for metabolic needs.

NCOA4-mediated ferritinophagy in iron mobilization

The iron sequestered within ferritin can be mobilized by the process of ferritinophagy, an autophagy turnover process mediated by NCOA4. This is a feedback regulation mechanism for intracellular iron utilization, and its activation will result in an increase in the content of available iron in cells. Mechanistically, NCOA4 functions as a selective autophagy receptor and binds to FTH1 of ferritin to facilitate the transportation of intracellular ferritin to autophagy lysosomes, ultimately releasing free iron. When iron is abundant, HERC2 (E3 ubiquitin ligase) mediates the degradation of NCOA4 in a ubiquitin-dependent manner to decrease the ferritinophagy flux for balancing iron levels (He et al. 2022).

Poly(rC)-Binding Proteins (PCBPs) mediated cytosolic iron distribution

PCBPs are members of the heterogeneous nuclear ribonucleoprotein (hnRNP) family. To date, five mammalian hnRNP family members have been identified: PCBP1, PCBP2, PCBP3, PCBP4, and hnRNP K. Notably, four PCBP isoforms have recently been characterized as cytosolic iron chaperones (Leidgens et al. 2013; Sen et al. 2022). PCBP1 and PCBP2 are cytosolic iron chaperones that deliver Fe²⁺ to ferritin, non-heme iron enzymes such as HIF prolyl hydroxylases and deoxyhypusine hydroxylase, and the iron exporter FPN thereby converting labile iron into catalytically active or protein-binding forms. PCBP2 acquires its iron directly from the DMT1 transporter via a cytoplasmic N-terminal interaction whereas the iron source for PCBP1 remains undefined. Loss of PCBP1 or PCBP2 reduces ferritin iron loading that resulting in elevating cytosolic free Fe²⁺ levels (Bogdan et al. 2016). While PCBP3 and PCBP4 similarly demonstrate iron chaperone activity, their precise functions in cellular iron metabolism have yet to be fully characterized (Leidgens et al. 2013).

MitoNEET-mediated mitochondrial iron-redox coupling

MitoNEET (CDGSH iron sulfur domain 1, CISD1) is an outer mitochondrial membrane protein that harbors a redox-active [2Fe-2S] cluster. It functions as a redox enzyme, catalyzing electron transfer from reduced flavin mononucleotide (FMNH₂) to either oxygen or ubiquinone, thereby promoting the oxidation of cytosolic NADH and modulating mitochondrial redox balance. The [2Fe-2S] cluster of MitoNEET is remarkably stable and resistant to oxidation by H₂O₂, allowing it to undergo reversible oxidation–reduction cycles in response to the cellular redox environment. Critically, MitoNEET is essential for maintaining mitochondrial iron and ROS homeostasis; its suppression leads to uncontrolled accumulation of both iron and ROS within mitochondria, resulting in decreased mitochondrial membrane potential, activation of autophagy, and significantly impaired proliferation of tumor cells. Thus, MitoNEET acts as a key regulator that couples mitochondrial redox state to iron handling, preserving organelle integrity and cellular viability through its [2Fe-2S] cluster–dependent activity (Wang et al. 2017b; Sohn et al. 2013; Tam and Sweeney 2023).

Iron export

Ferroportin (FPN)

FPN is the sole known cellular iron exporter and serves as the central regulator of systemic and cellular iron efflux (Donovan et al. 2005). Its expression and stability are tightly controlled through multiple regulatory mechanisms, most critically by hepcidin-induced internalization and degradation in response to elevated iron (Drakesmith et al. 2015; Jiang et al. 2021b). By mediating iron export, FPN maintains systemic iron homeostasis, and its dysregulation contributes to iron metabolism disorders and cancer progression (Berezovsky et al. 2022).

Other iron expulsion mechanisms

In addition to FPN, tumor cells utilize alternative strategies—including exosomal ferritin secretion, copper-iron co-export via transporters such as ATP7A, and lactylation-mediated upregulation of iron-handling genes—to expel excess iron. Despite their mechanistic diversity, these pathways collectively reduce the intracellular LIP and confer resistance to ferroptosis, representing a broadly conserved adaptive response that supports therapy resistance across cancer types Li et al. 2021; Sun et al. 2023; Zhang et al. 2024a; Hassannia et al. 2019; Drakesmith and Prentice 2012).

Iron utilization

Intracellular iron utilization constitutes a rigorously orchestrated axis: uptake, trafficking, storage, and utilization are interlocked to safeguard non-redundant processes such as oxygen transport (Kerins and Ooi 2018); electron transfer, a fundamental aspect of cellular energy production (Lill et al. 2012); and enzymatic reactions, which play a vital role in various biochemical processes within the cells (Wise et al. 2017); DNA demethylation, a process mediated by iron dependent Ten-Eleven Translocation enzymes (TETs) family (Matuleviciute et al. 2021). Each cell tailors iron flux to its metabolic demand, thereby preserving systemic and intracellular homeostasis.

Intracellular iron is primarily utilized in mitochondria for the synthesis of Fe-S clusters and heme (Fig. 1). The detailed processes of iron utilization will be elaborated individually in the subsequent sections.

Iron sensors

Cells employ compartment-specific iron-sensing systems to maintain iron homeostasis: in the cytosol, iron-regulatory proteins IRP1 and IRP2 bind iron-responsive elements (IREs) to coordinately regulate transferrin receptor 1 (TFRC) and ferritin expression according to iron availability (Muckenthaler et al. 2017). Unlike IRP2, IRP1, also termed ACO1, is a [4Fe-4S] cluster-binding protein that exhibits aconitase activity under iron-replete conditions. Conversely, under iron-depletion conditions, cluster disassembly exposes its RNA-binding domain, enabling it to bind IRE-containing mRNAs to promote iron uptake and release, by enhancing TFRC expression and inhibiting ferritin expression, respectively. (Fig. 2) Thus, IRP1 is bifunctional and can perform distinct functions depending on the cellular [4Fe-4S] cluster “content” (Rouault and Maio 2017).

Fig. 2.

FBXL5-IRP axis–mediated iron sensing and homeostasis. Under iron-replete conditions, FBXL5 is stabilized via iron binding to its N-terminal hemerythrin-like (Hr) domain, which harbors a di-iron center. Concurrently, in the presence of oxygen, its C-terminal substrate-binding domain assembles an oxidized [2Fe–2S]²⁺ cluster, enabling high-affinity recognition of iron regulatory protein 2 (IRP2). FBXL5, as the substrate-recognition subunit of an SCF-type E3 ubiquitin ligase, targets IRP2 for polyubiquitination and proteasomal degradation, thereby permitting ferritin (FTH1/FTL) translation and promoting TFRC mRNA decay. Under iron deficiency, the Hr domain fails to bind iron, triggering FBXL5 auto-ubiquitination and degradation; under hypoxia or reducing conditions, the C-terminal [2Fe-2S] cluster becomes reduced, disrupting IRP2 binding. In both scenarios, IRP2 accumulates, binds iron-responsive elements (IREs) in the 5′UTR of ferritin mRNAs to repress translation, and stabilizes transferrin receptor 1 (TFRC) mRNA via IREs in its 3′UTR, enhancing cellular iron uptake. Parallel to IRP2, IRP1 (ACO1) functions as a cytosolic aconitase when its cubane [4Fe-4S] cluster is intact under iron-sufficient conditions; upon iron depletion or oxidative stress, cluster disassembly unmasks its RNA-binding domain, allowing it to bind IREs and cooperate with IRP2 in suppressing ferritin synthesis and upregulating TFRC expression. Together, FBXL5, IRP2, and IRP1 form an integrated iron- and oxygen-sensitive regulatory network that dynamically balances cellular iron acquisition, storage, and utilization. FBXL5: F-box and leucine-rich repeat protein 5; IRP1/2: Iron regulatory protein 1/2; IRE: Iron-responsive element; TFRC: Transferrin receptor 1; FTH1/FTL: Ferritin heavy/light chain; SCF: Skp1–Cul1–F-box; [2Fe–2S]/[4Fe-4S]: Iron–sulfur clusters; Hr: Hemerythrin-like domain; UTR: Untranslated region

In the nucleus, FBXL5 (F-box and leucine-rich repeat protein 5) senses iron through N-terminal hemerythrin (Hr)-like domain, thereby controlling IRP2 stability to modulate iron homeostasis (Ruiz and Bruick 2014). As the substrate-recognition component of the SCF (SKP1–CUL1–F-box protein) ubiquitin ligase complex, FBXL5 targets IRP2 for proteasomal degradation. This FBXL5-IRP2 axis operates through an iron-dependent mechanism: under iron-replete conditions, iron binding stabilizes the Hr domain, maintaining the overall stability of FBXL5; conversely, during iron deficiency, the inability of the Hr domain to bind iron induces conformational changes that trigger ubiquitination and subsequent proteasomal degradation of FBXL5, thereby allowing IRP2 to accumulate and bind to iron-responsive elements (IREs) in target mRNAs (Fig. 2). Beyond iron sensing, FBXL5 harbors a unique redox-sensitive [2Fe-2S] cluster within its C-terminal substrate-binding domain. High-affinity interaction between FBXL5 and IRP2 occurs exclusively when the [2Fe-2S] cluster is in the oxidized [2Fe-2S]²⁺ state—a condition regulated by ambient oxygen levels. When oxygen is limited or the cluster becomes reduced, the C-terminal conformation of FBXL5 is altered, abrogating its ability to bind IRP2. Together, these two mechanisms enable FBXL5 to sense iron availability through its N-terminal Hr domain to control its own stability, and to sense oxygen through its C-terminal [2Fe-2S] cluster to bind IRP2. This two-tiered system ensures precise maintenance of cellular iron homeostasis in response to varying microenvironmental conditions (Muckenthaler et al. 2017; Ruiz and Bruick 2014; Salahudeen et al. 2009).

At the plasma membrane, TFRC monitors extracellular transferrin-bound iron, while HIF-2α—stabilized under low iron or hypoxia—transcriptionally upregulates DMT1 and DCYTB to enhance iron uptake (Ortmann 2024). Within mitochondria, the efficiency of Fe-S cluster synthesis serves as an indicator of bioavailable iron, feeding back to adjust cellular iron acquisition programs (Zhang et al. 2025). Through these sensors, cells exert fine-grained control over iron, providing multiple nodes for targeted intervention.

The iron demand of cancer cells

Increasing iron absorption in cancers

The principal characteristic of cancer, which is hyper-proliferation, essentially makes them far more dependent on iron compared to normal cells. The abnormal and excessive growth of cancer cells creates a heightened need for iron to support their rapid and unregulated division and proliferation.

Cancer cells can increase iron uptake through various ways, including upregulating the classical Tf-TfR1-STEAP3-DMT1 system, and activating the non-traditional Fe²⁺ absorption mediated by Dcytb, ZIP8, ZIP14 or LCN2, as well as the uptake of heme iron (Torti and Torti 2013; Yang et al. 2009). A considerable number of proteins that are closely related to the uptake of iron have been found to play a crucial and significant role in the entire process of the occurrence and subsequent development of tumors (Table 1).

Table 1.

Research on the mechanism by which iron homeostasis affects tumor progression

Protein	Inducer/modification	Roles in cancers	Reference
TF	IL-18/N-glycosylation	As a drug carrier for cancer treatment	Trbojević-Akmačić et al. 2023; Park et al. 2009; Wigner et al. 2021)
TfR1	FGFR1- IRP2/stearoylation	Regulating of mitochondrial morphology; as a valuable tool for cancer-targeted drug delivery	Senyilmaz et al. 2015; Mojarad-Jabali et al. 2022; Lin et al. 2024a)
STEAP3	HIF-1α/DNA methylation	Activating Wnt/β-catenin signaling; regulating histone acetylation; facilitating nuclear trafficking of EGFR	Zhou et al. 2022; Lv et al. 2024; Wang et al. 2021; Zhao et al. 2024)
DMT1	Hippo-YAP, HO-1, HIF-1α/ubiquitination	As a biomarker of tumor prognosis; Activate the HIF1α/HRE pathway	Giorgi et al. 2023; He et al. 2023a; Xiong et al. 2022; Liu et al. 2025)
Dcytb	HIF-2α/–	Up-regulate the expression of other ferritransport proteins and activate HIF-2α signaling pathway	Zhu et al. 2022; Li et al. 2024)
ZIP14	TNF-α, TGF-β, HO-1/–	As a potential target for the treatment of metastatic cancer cachexia	Giorgi et al. 2023; Wang et al. 2018)
ZIP18	–	A metal transporter	Vungutur et al. 2024)
CD163	CSF1, CCL2/–	Secrete growth factors to promote tumor proliferation/immunosuppression	Stadler et al. 2021)
CD91	HSP/Phosphorylation	Initiating an anti-tumor response by releasing cytokines and responding to other immune mediators	Harkness et al. 2024)
FLVCR2	–	–	–
HRG1	ErbB, YAP, GPR30, Fn14, AhR/–	Activating the NF-κB pathway	Yang et al. 2008)
HCP1	–	Inhibiting HSP90 activity, inducing apoptosis, and blocking autophagy	Wei et al. 2018)
HO-1/HO-2	PERK, NRF2/–	Activating PERK-Nrf2-HO-1 triggers ferroptosis	Wei et al. 2021)
LCN2	IFN, JAK–STAT, ISG/–	Enhancing cancer's susceptibility to oncolytic viruses	Barer et al. 2023)
SLC22A17	LCN2–SLC22A17–MMP9/DNA methylation	LCN2 binds SLC22A17 to boost IL-10 and decrease MHCII in macrophages	Candido et al. 2022; Lavoro et al. 2024; Czech et al. 2024)

Altered iron metabolism in cancer cells

Both high and low expression levels of ferritin have been reported in cancer cells (Schonberg et al. 2015; Shpyleva et al. 2011). Owing to rapid proliferation and active lipid metabolism, cancer cells chronically endure high-level oxidative stress; to buffer this pressure they remodel ferritin expression or promote ferritinophagy, thereby re-calibrating the LIP.

NCOA4-mediated ferritinophagy is a pivotal process in cancer progression, particularly in maintaining iron homeostasis and promoting tumor growth. For instance, in pancreatic ductal adenocarcinoma (PDAC), upregulated ferritinophagy has been shown to sustain iron availability, thereby promoting tumor progression (Santana-Codina et al. 2022). Elevated ferritinophagy signatures predict poor prognosis in PDAC patients. In glioblastoma (GBM), the coatomer protein complex subunit zeta 1 (COPZ1) plays a significant role in cancer cell proliferation. Inhibition of COPZ1 leads to decreased proliferation of GBM cells by upregulating NCOA4 and promoting ferritin degradation (Chen et al. 2024). Additionally, in renal clear cell carcinoma, reduced NCOA4 expression can control adverse cancer outcomes and impaired immune cell infiltration (Mou et al. 2021).

FPN is down-regulated in breast cancer, lung cancer, ovarian cancer and prostate cancer (Basuli et al. 2017; Xue et al. 2015; Babu and Muckenthaler 2016; Zhang et al. 2014). The decreased expression of FPN reduces iron efflux and increases LIP in cancer cells, and the additional intracellular iron promotes their growth and proliferation. Hepcidin produced by tumors or liver leads to the degradation of FPN and also contributes to the growth and progression of cancer. Multiple studies on solid tumors have shown abnormalities in hepcidin levels, with elevated expression in tumors such as breast cancer and renal cancer, and decreased expression in hepatocellular carcinoma (Pinnix et al. 2010; Greene et al. 2017).

Beyond the canonical axes, HIF-1α/HIF-2α, NRF2/KEAP1, and AMPK/mTORC1 modules further sculpt the altered iron metabolism of cancer cells by transcriptionally reprogramming iron-transporter expression, ferritin biosynthesis, and ferroptosis-regulatory circuitry (Wahida and Conrad 2025).

Mitochondrial iron metabolism in cancer

Mitochondria serve as the central hub for iron utilization in cancer cells, directly linking iron metabolism to tumor bioenergetics, redox balance, and therapeutic resistance.

Free iron

The majority of cytosolic Fe²⁺ is imported into mitochondria primarily via mitoferrins, establishing a critical supply line for iron-dependent processes (Seguin et al. 2020). Inside mitochondria, iron is predominantly channeled into two essential biosynthetic pathways: heme synthesis and Fe-S cluster assembly (Napier et al. 2005; Dietz et al. 2021). Heme not only functions as a prosthetic group in respiratory complexes but also yields antioxidant byproducts—such as bilirubin and carbon monoxide—upon degradation, thereby mitigating oxidative stress and promoting tumor cell survival. In contrast, Fe-S clusters are obligate cofactors that must be incorporated into apoproteins via the Fe-S clusters assembly machinery to support electron transport, TCA cycle activity, and DNA repair.

Heme

Heme, an iron-containing prosthetic group, continuously incorporates, shuttles, and recycles iron throughout its synthesis, turnover, and degradation, constituting an indispensable hub of cellular iron metabolism (Fig. 3).

Fig. 3.

Intracellular heme synthesis and metabolic regulatory network. The sources of heme include both extracellular uptake and autogenous synthesis. Heme then participates in many important functions by constituting essential proteins, regulating metabolic pathways, or being broken down into other products within and outside the mitochondria, and is eventually exported out of the cell via heme transport proteins

Heme biogenesis

The synthesis of heme is a highly conserved process involving a series of enzymatic reactions. It begins with the condensation of glycine and succinyl-CoA to form 5-aminolevulinic acid (ALA), a reaction catalyzed by ALA synthase 1 (ALAS1), which acts as the rate-limiting step (Stojanovski et al. 2019). Subsequently, porphobilinogen deaminase (PBGD) catalyzes the formation of porphobilinogen (PBG), which is then converted to uroporphyrinogen III through a series of intermediate steps (Zhang et al. 2015). Uroporphyrinogen III is decarboxylated to coproporphyrinogen III by uroporphyrinogen decarboxylase (UROD) and further oxidized to protoporphyrinogen IX by coproporphyrinogen oxidase (CPOX). The final step involves the insertion of iron into protoporphyrin IX by ferrochelatase (FECH) to form heme (Medlock et al. 2021; Sachar et al. 2016; Swenson et al. 2020). The cytosolic heme level is sustained by the rate of hemoprotein production, the activity of heme importers and exporters, and the rate of heme degradation facilitated by heme oxygenase (Furfaro et al. 2016).

Heme onco-regulation

Heme is not merely a passive cofactor but an active regulator of tumor cell physiology. It influences drug metabolism via cytochrome P450 enzymes, regulates transcription factors such as BTB and CNC homology 1 (Bach1) and nuclear receptor Rev-erbα (NR1D1; nuclear receptor subfamily 1, group D, member 1) that control cell proliferation and circadian metabolism, and fine-tunes redox signaling through its degradation by heme oxygenase-1 (HO-1), which generates the antioxidant molecules biliverdin and carbon monoxide (Ponka 1999; Shibayama et al. 2020). Critically, the intracellular heme pool is tightly controlled by synthesis (via ALAS1/2), degradation (by HO-1/HO-2), incorporation into hemoproteins, and intercompartmental trafficking. In cancer, this balance is frequently disrupted—elevated heme synthesis or impaired degradation can amplify oxidative stress, whereas excessive HO-1 activity may confer a survival advantage by suppressing ferroptosis.

Elevated heme levels are a hallmark of multiple cancers, reflecting an adaptive rewiring of metabolism rather than a passive accumulation. Tumor cells upregulate heme synthesis to simultaneously modulate bioenergetics, redox balance, and signaling: by consuming succinyl-CoA—a TCA cycle intermediate—and sequestering mitochondrial iron, heme biosynthesis attenuates Fe-S cluster assembly and slows TCA flux, thereby redirecting metabolic intermediates toward anabolic pathways (Fukuda et al. 2017; Wang et al. 2017c; Tolosano 2019). Beyond metabolic reprogramming, heme directly influences tumor suppressor networks; it stabilizes p53 and modulates the activity of heme-binding apoptotic regulators (Consoli et al. 2024; Vávra et al. 2023). Moreover, heme shapes the tumor microenvironment (TME) by polarizing tumor-associated macrophages (TAMs) toward pro-tumorigenic phenotypes, further supporting immune evasion and disease progression (Costa Silva et al. 2017; Vinchi et al. 2016).

Fe-S cluster

Fe-S clusters, serving as iron-containing catalytic centers, continuously sequester and cycle mitochondrial free iron, functioning as an indispensable hub for iron unloading and intracellular recycling. When Fe-S cluster biogenesis is impaired, the resulting cofactor deficiency is independently sensed by IRP1 and the FBXL5-IRP2 axis, triggering an iron-starvation response that enhances iron import and mobilization, thereby expanding the LIP. In cancers, this iron metabolic rewiring also incidentally sensitizes tumor cells to ferroptosis (Erdem et al. 2021).

Fe-S cluster assembly system

The assembly of Fe-S clusters and its regulatory mechanisms have been predominantly investigated in structurally simple model organisms, such as Escherichia coli and yeast. In eukaryotic cells, the process of assembly and trafficking of Fe-S clusters, as well as their insertion in apoproteins, require the coordinated participation of more than 30 proteins (Braymer and Lill 2017). Among them, the Fe-S cluster assembly scaffold protein ISCU2 relies on six other core ISC (Iron-Sulfur Cluster assembly) proteins: the cysteine desulfurase complex NFS1–ISD11–ACP1, frataxin (FXN), ferredoxin (FDX2), and its reductase (FDXR), to construct the [2Fe-2S] center. The research on the Fe-S cluster assembly mechanism mainly includes the following hypotheses (Table 2).

Table 2.

Fe-S cluster assembly mechanism

Name	Mechanism Summary	Key Molecules Involved	Reference
ISC-mediated Assembly Mechanism	The ISC system assembles Fe-S clusters stepwise on a scaffold protein via enzymatic reactions, with chaperones facilitating transfer to target proteins	Mitochondrial ISC system components:the cysteine desulfurase complex NFS1–ISD11–ACP1, frataxin (FXN), ferredoxin (FDX2), FDXR	Sheftel et al. 2012; Urbina et al. 2001)
Chemical Self-Assembly Hypothesis	Fe-S clusters can form spontaneously without enzyme catalysis in a suitable chemical environment	Fe2⁺/Fe3⁺, S2⁻, Cysteine, Reducing Agents	Jordan et al. 2021; Ren et al. 2021)
Mitochondrial Central Hypothesis	In eukaryotes, an unknown sulfur precursor generated by the mitochondrial ISC system is exported to the cytosol via the transporter Atm1	Mitochondrial ISC components, ABC transporter Atm1	Gyula Kispal1, P.C., Corinna Prohl and Roland Lill2 1999; Kispal, R.L.a.G. 2000; Schaedler et al. 2014)
[2Fe-2S] “Precursor Fusion” Mechanism	[2Fe-2S] cluster is formed by the fusion of two [1Fe-1S] precursor units	Mitochondrial ISC system components	(Sylvian Gervason et al. 2025)

Chemical self-assembly hypothesis

In suitable chemical environments, Fe-S clusters can be formed spontaneously without enzyme catalysis. Under very low concentration (micromolar level) of cysteine, anaerobic, neutral pH and normal temperature conditions, Fe-S clusters can also be spontaneously and quickly assembled from simple inorganic matter (Fe²⁺ and S²⁻), and these clusters (mainly [2Fe-2S] and [4Fe-4S]) have natural Fe-S Protein-like redox activity (Jordan et al. 2021; Ren, et al. 2021). This hypothesis supports the “iron-sulfur world” theory, which believes that Fe-S clusters are the core of early life molecules. However, the intracellular environment (aerobic, low ion concentration, oxidative stress) is not conducive to self-assembly and cannot solve the problems of specificity and toxicity. Therefore, it explains the “possibility”, but it cannot explain the “actual operation” of modern organisms.

Mitochondrial central hypothesis

In eukaryotic cells, mitochondria is the initiator and center of cytoplasmic and intranuclear Fe-S cluster biosynthesis. The mitochondrial ISC system produces an unknown “precursor substance” (X-S) output by the ABC transporter protein Atm1, which is necessary to activate the cytoplasmic CIA system. This explains why mitochondrial dysfunction directly affects the stability of the nuclear genome and cytoplasmic metabolism (Kispal, Prohl and Lill 1999; Kispal 2000; Schaedler et al. 2014).

The “precursor fusion” mechanism of the 2Fe-2S cluster

Recently, a more subversive model is proposed: the ISC machine does not directly construct the [2Fe-2S] cluster, but first synthesizes [1Fe-1S] as the core precursor, and then generates the final [2Fe-2S] product through the fusion of the two precursors (Sylvian Gervason et al. 2025). The [2Fe-2S] cluster may be formed by the fusion of two [1Fe-1S] precursor units, which challenges the traditional model of “adding atoms one by one on a scaffold protein”.

The ISC system

During the assembly, NFS1 catalyzes the transfer of sulfur atoms from cysteine to Fe-S cluster assembly scaffold proteins (ISCU2 or ISCA2), forming a [2Fe-2S] cluster, which is then reduced and fused to form a [4Fe-4S] cluster. The assembled Fe-S cluster is transferred to the target protein with the assistance of molecular chaperones Hsc20 and Hsp70 (Sheftel et al. 2012; Urbina et al. 2001).

Fe-S cluster functions

Fe-S clusters are primarily synthesized in the mitochondria and serve as cofactors for many enzymes, participating in electron transfer, DNA synthesis, cellular respiration, and other biological processes (Boncella et al. 2022; Bak and Weerapana 2023). They play a particularly crucial role in maintaining the function of mitochondria (Fig. 4) (Bak and Weerapana 2023; Boland et al. 2025; Bandara et al. 2021; Hird et al. 2024; Yang et al. 2023a; Boniecki et al. 2017; Garcia et al. 2022).

Fig. 4.

Regulatory mechanisms of ferroptosis within cells. Fe²⁺ is imported into the mitochondrial matrix by MFRN1/2 and fed into the ISC assembly complex (NFS1, ISD11, U-type scaffold ISCU, NFU1, BOLA3, GLRX5, HSCB, FDX2) to synthesize Fe–S clusters. These clusters are inserted into respiratory-chain complexes I (NDUFS1/2/3/7/8 with [2Fe-2S] and [4Fe-4S]), II (SDHA-SDHB [2Fe-2S], [4Fe-4S], [3Fe-4S]), and III (UQCRFS1 Rieske [2Fe-2S]) where they transfer electrons to cytochrome c (CYCS) and onward to complex IV, undergo redox cycling, sense mitochondrial ROS, and bind the inhibitor rotenone. Fe–S deficiency disrupts the electron-transport chain, increases lipid peroxidation and sensitizes cells to ferroptosis. Electrons from fatty-acid β-oxidation are delivered to coenzyme Q via ETFA-ETFB-ETFDH. GPX4, using GSH generated by GCLC/GCLM and supplied by SLC25A39, reduces lipid hydroperoxides; FSP1 (coQ-NAD(P)H oxidoreductase) regenerates CoQH2; both inhibit ferroptosis. Copper ions (Cu²⁺) reduced to Cu⁺ by COX17 and SLC25A33 promote cuproptosis, which overlaps with ferroptosis through lipid peroxidation. The figure thereby compiles all annotated proteins, metabolic pathways, redox reactions and mitochondrial functions that collectively regulate ferroptosis. ASN: Fatty acid synthetase, catalyzes the conversion of acetyl-CoA to fatty acids. CIC: Mitochondrial carrier for fatty acids, transports fatty acids into the mitochondria. Pyr: Pyruvate, derived from glucose (Glc) during glycolysis. It enters the mitochondria through MPC (pyruvate carrier) and is converted to acetyl-CoA, participating in the tricarboxylic acid cycle (TCA cycle). Cit: Citrate; α-KG: alpha-ketoglutarate, plays an important role in DNA demethylation, working with the TET protein family to convert 5-methylcytosine (5mC) to 5-hydroxymethylcytosine (5hmC), 5-formylcytosine (5fC), and 5-carboxylcytosine (5caC); Suc-CoA: Succinyl-CoA; Fum: Fumarate; Mal: Malate; OA: Oxaloacetate acid

Fe-S micro-iron hub

Fe-S clusters act as quantitatively small yet metabolically critical iron pools that couple cellular iron homeostasis to mitochondrial energy metabolism. SDHB (Succinate dehydrogenase), ACO2 (Aconitate hydratase 2) and LIAS (Lipoyl synthase) each bind canonical Fe-S cofactors whose assembly consumes labile iron and whose stability is rapidly lost when intracellular iron falls or when the ISC-export/cytosolic CIA pathways are saturated. In pancreatic ductal adenocarcinoma cells, oncogenic KRAS amplifies c-Myc-driven transcription of the scaffold protein ISCU2; the resulting rise in Fe-S output stabilizes SDHB, ACO2 and α-KGDH activities, accelerates α-ketoglutarate oxidation and thereby expands the redox TCA flux that fuels tumor growth (Ricci et al. 2023; He et al. 2023b; Ren et al. 2023; Li et al. 2022; Zhu et al. 2023).

LIAS itself coordinates two [4Fe-4S] centers to generate lipoate, the obligate cofactor for PDH and α-KGDH (Hendricks et al. 2021); its Fe-S occupancy is rate-limiting for overall lipoylation and is gated by FDX1, a ferredoxin that donates electrons—and, indirectly, Fe-S-derived iron—to the lipoate synthase reaction. Because FDX1 also delivers Cu(I) required for cuproptotic lipoylated-protein clustering, fluctuations in Fe-S availability (reflected by ISCU2, LIAS or FDX1 levels) simultaneously report mitochondrial iron sufficiency and dictate whether excess copper triggers death or is safely buffered by the Fe-S-centered respiratory chain (Wang et al. 2022; Lin et al. 2024b; Dreishpoon et al. 2023).

Fe-S clusters at the nexus of copper and ferroptosis

Fe-S clusters constitute the molecular hinge linking iron metabolism to both ferroptosis and cuproptosis. Copper-dependent generation of Cu⁺ and ROS within mitochondria rapidly blocks Fe-S assembly, disabling TCA cycle enzymes and precipitating cuproptosis while simultaneously derepressing IRP2 to raise labile iron, thereby amplifying lipid peroxidation and ferroptotic death (Lin et al. 2024b; Lin et al. 2023; Zheng et al. 2022; Tang et al. 2022; Yang et al. 2024; Yang et al. 2023b; Zhang et al. 2024b; Qi and Zhu 2023). Disruption of Fe-S biogenesis therefore converts copper stress into iron dyshomeostasis, merging the two cell death pathways through a shared metabolic lesion (Zheng et al. 2022; Tang et al. 2022; Qi and Zhu 2023; Gatto et al. 2023; Tsvetkov et al. 2022).

Iron metabolism in cancer: key nodes and therapeutic targets

Recent studies have increasingly demonstrated that the heightened dependence of cancer cells on iron renders iron metabolism pathways promising targets for therapeutic intervention. Dysregulation has been observed across multiple steps of iron handling in tumors, including uptake, storage, export, and utilization. This review focuses on core aspects of iron metabolism and systematically examines key molecules involved in maintaining iron homeostasis and regulating ferroptosis. These molecules act at critical nodes, including transferrin receptor–mediated iron uptake, ferritin-dependent iron storage, ferroportin-mediated iron export, and mitochondrial iron utilization. We systematically summarize the targeted pathways and associated molecules discussed herein, with the aim of providing a reference for precision therapies based on iron metabolism modulation (Table 3).

Table 3.

Key proteins and targeted pathways in cellular iron metabolism

Metabolic Step	Key Proteins	Primary Function	Regulatory Mechanism
Cellular iron uptake	TfR1, DMT1	Mediate transferrin-dependent and non-transferrin iron import	Post-transcriptionally stabilized by IRP binding to 3′ IREs under iron deficiency
Cytosolic iron chaperoning	PCBP1, PCBP2	Deliver Fe2⁺ to ferritin, non-heme iron enzymes, and ferroportin	PCBP2 acquires iron directly from DMT1; loss disrupts mitochondrial iron supply
Iron storage	Ferritin	Sequester excess iron as a mineralized Fe3⁺ core	Translation repressed by IRP binding to 5′ IRE under low iron; degraded via NCOA4-mediated ferritinophagy
Mitochondrial iron utilization	MFRN1/2, MitoNEET (CISD1)	MFRN1/2 transport Fe2⁺ into the matrix for heme/Fe–S biogenesis; MitoNEET maintains redox-iron homeostasis via its [2Fe–2S] cluster	MitoNEET dysfunction causes mitochondrial iron/ROS accumulation, loss of membrane potential, and impaired cell proliferation
Cellular iron export	Ferroportin (FPN)	Sole cellular iron exporter	Degraded upon hepcidin binding; translation inhibited by IRP via 5′ IRE
Systemic iron sensing	FBXL5, IRP2	FBXL5 targets IRP2 for degradation in iron- and oxygen-replete conditions	FBXL5 stability controlled by N-terminal hemerythrin-like domain and C-terminal [2Fe–2S] cluster

Scientific problems encountered in the research of mitochondrial iron metabolism

How is the [4Fe-4S] cluster “reassembled” inside the mitochondrion?

Although the ISC pathway is mapped, we still lack high-resolution structural–functional evidence for when and where a transient [2Fe-2S] template switches Fe²⁺/Fe³⁺ valence and synchronously incorporates sulfide to yield [4Fe-4S], limiting precision interventions for assembly-failure disorders such as Friedreich ataxia.

How does cysteine, the sulfur source, cross the inner mitochondrial membrane?

No specific cysteine transporter has been found in mammals; neither the cystine–cysteine redox shuttle nor direct transport by an unidentified SLC25 member has been validated, leaving the “sulfur-supply” step a potential bottleneck for Fe-S assembly flux.

Where do the Fe-S clusters that catalyze Fe-S synthesis come from when the enzymes themselves require Fe-S clusters?

NDOR1 (NADPH-dependent diflavin oxidoreductase 1), FDX1/2 (Ferredoxin 1/2), CIAO3 (Cytosolic iron-sulfur assembly component 3) and others must pre-load [2Fe-2S] or [4Fe-4S] to deliver electrons or sulfur, yet their own synthesis depends on the still-absent Fe-S machinery, creating a chicken-and-egg loop that remains a fundamental riddle for defining the origin of Fe-S biogenesis.

Discussion

Iron deficiency and iron overload are opposite extremes, yet they converge at the mitochondrial level. In both states, Fe-S clusters destabilize, oxidative phosphorylation collapses, and the electron transport chain leaks electrons that generate ROS bursts. The resulting lipid peroxidation propagates a ferroptotic signal that is amplified by an IRP-hepcidin-ferroportin axis that is rewired in either direction. A self-sustaining “energy failure-inflammation loop” then precipitates secondary injury in heart, liver, and other organs. Tumor cells exploit the same vulnerability: they accumulate mitochondrial iron to sensitize themselves to ferroptosis while simultaneously up-regulating protective pathways. Thus, preserving mitochondrial Fe-S homeostasis is not only the key to preventing the shared pathology of systemic iron disorders but also the Achilles heel that next-generation ferroptosis inducers can target cancer therapy.

Authors’ contributions

C.X. ，X.S. and Z.Z. completed the writing of the main text; C.X. and X.S. prepared Figs. 1, 2 and 3 and Tables 1 and 2; G.j., W.j. and W.j. collected and organized the preliminary material; J.L., L.J. and R.X. provided professional guidance and critical review. All authors have read and approved the final manuscript.

Funding

This study was supported by grants from Zhejiang Provincial Natural Science Foundation (Grant Number: LMS25H160005) and the National Natural Science Foundation of China (Grant Number: 82302631).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Competing interests

The authors declare no competing interests.