An update on the role of ferroptosis in the pathogenesis of osteoporosis

Abstract

• Ferroptosis is a novel form of programmed cell death, distinguished from apoptosis, autophagy, and programmed necrosis and has received much attention since it was defined in 2012.

• Ferroptotic cells physiologically exhibit iron metabolism dysregulation, oxidative stress, and lipid peroxidation. Morphologically, they show plasma membrane disruption, cytoplasmic swelling, and mitochondrial condensation.

• Osteoporosis is taken more and more seriously as the proportion of the aging population continues to increase globally. Interestingly, ferroptosis has been demonstrated to be involved in the development and progression of osteoporosis in many extant studies.

• The review summarizes iron metabolism, lipid peroxidation, and the different regulatory signals in ferroptosis. Changes in signaling mechanisms within osteoblasts, osteoclasts, and osteocytes after ferroptosis occur are explained here.

• Studies showed ferroptosis play an important role in different osteoporosis models (diabetes osteoporosis, postmenopausal osteoporosis, glucocorticoid-induced osteoporosis). Inhibitors and EC (Exos) targeting ferroptosis could ameliorate bone loss in osteoporotic mice by protecting cells against lipid peroxidation. Shortly, we hope that more effective and appropriate clinical therapy means will be utilized in the treatment of osteoporosis.

Introduction

There have been a great many discoveries about cell death in the last few decades. Under different forms of death, cells behave differently in terms of physiology and morphology. Cell death has been classified into two categories: non-programmed cell death and programmed cell death (PCD). Non-programmed cell death, known as necrosis, is the passive cell death occurring when cells are exposed to an environmental irritant. PCDs have something in common as they are active cell death processes that can be blocked by inhibitors of cell signaling, including apoptosis, autophagy, programmed necrosis, pyroptosis, and ferroptosis (1, 2). Ferroptosis has received much attention as a novel type of iron-dependent PCD.

Before the emergence of the concept of ferroptosis, ferroptosis inducers were identified in high-throughput small-molecule screening research as selective lethal compounds in RAS mutant tumor cells. Research showed that erastin is lethal to human foreskin fibroblasts engineered to mutate the RAS oncogene in 2003 (3, 4). Later, in 2008, RSL 3 and RSL 5 were discovered as RAS-selective lethal small molecules that killed human foreskin fibroblasts in a non-apoptotic manner. It was found that erastin-suppressed cystine intracellular uptake through the cystine/glutamate antiporter (system X_c ⁻) and contributed to cell death. This novel form of cell death was defined as ferroptosis in 2012 (2, 5). Although ferroptosis, as a novel PCD, is different from apoptosis, necrosis, and autophagy, most scholars agree that cells undergoing ferroptosis behave morphologically in the same way as cells undergoing necrosis (2, 6). Ferroptotic cells exhibit a loss of plasma membrane integrity, cytoplasmic swelling, cytoplasmic organelles swelling, and medium chromatin condensation. In addition, morphologic abnormalities of mitochondria are found on transmission electron microscopy, including condensation or swelling, increased membrane density, and reduced or absent cristae (7).

Over the past decade, ferroptosis has been shown to have great therapeutic potential for many diseases, including drug-resistant cancers, ischemic organ injuries, and other degenerative diseases like osteoporosis (OP). The results of existing literature studies have proved that some means targeting ferroptosis, such as melatonin, metformin, and endothelial cell-secreted exosomes, could alleviate bone loss in osteoporotic mice (8). Here, we summarize the research progress of ferroptosis in OP, from cell physiology to animal models, to provide a comprehensive foundation for future related studies.

Mechanism of ferroptosis

Dysregulated iron metabolism and lipid peroxidation are the two most important causes of ferroptosis. Here, we briefly conclude that cellular iron metabolism and fatty acid metabolism process in ferroptosis cause cell membrane lipid peroxidation (Fig. 1).

Figure 1.

Mechanism of ferroptosis: (1) Iron-bound TF (Fe³⁺) enters the cell via TFR1 and Fe³⁺ is reduced to Fe²⁺ in the endosome by STEAP3 and DMT1. Fe²⁺ that enters the cytoplasm from the endosome could bind to ferritin or be excreted outside the cell via exosomes and FPN. Fe²⁺ in the LIP extracts the hydrogen atoms in PUFAs via the Fenton reaction and causes oxidative stress to proteins and membrane lipids. (2) AA and AdA as major PUFAs undergo a series of catalytic reactions to form PE-AA-OOH and PE-AdA-OOH, which act on membrane phospholipids and result in lipid peroxidation. ACSL4, acyl-CoA synthetase long-chain family member 4; DMT1, divalent metal transporter 1; FPN, ferroportin; LOXs, lipoxygenases; LIP, labile iron pool; LPCAT3, lysophosphatidyl-choline acyltransferase 3; ROS, reactive oxygen species; STEAP3, six-transmembraneepithelial antigens of the prostate 3; TF, transferrin; TFR1, transferrin receptor 1.

Iron metabolism

Iron is present in the body as ferrous ions (Fe²⁺) and trivalent iron ions (Fe³⁺) and transferrin (TF) is primarily responsible for the transportation of extracellular iron (3). Iron-bound TF (Fe³⁺) enters the cell via TF receptor 1 (TFR1) and is subsequently endocytosed in the endosome (9). Under the influence of six-transmembrane epithelial antigens of the prostate 3 (STEAP3) and divalent metal transporter 1 (DMT1) in acidic endosomes, Fe³⁺ is reduced to Fe²⁺ and exported to the cytoplasm (10). Fe²⁺ in the labile iron pool (LIP) of the cytoplasm and mitochondria could play an important role in cellular metabolism or be stored as ferritin (11). Fe²⁺ is oxidized back to Fe³⁺ via oxidative enzymes and subsequently combined with ferroportin (FPN). FPN is responsible for transporting iron ions outside the cell (12). The liver-derived iron-regulating hormone hepcidin can selectively bind to FPN, inhibiting its activity and reducing iron ion transportation (13). Prominin 2 was found to activate the production of ferritin-containing multivesicular bodies including exosomes and attenuate ferroptosis (14). Iron belongs to a redox-active metal that is involved in the formation of free radicals, causing lipid peroxidation (15). The above proteins involved in iron homeostasis, including its import, export, and storage, have been shown to modulate sensitivity to ferroptosis.

Iron is an essential trace element for the maintenance of basic life activities. When the regulation of iron metabolism is disturbed, large accumulations of iron can cause damage to cells and tissues (16). Fe²⁺ extracts hydrogen atoms in polyunsaturated fatty acids (PUFAs) via the Fenton reaction. The hydroxyl radicals generated by this reaction cause oxidative damage to DNA, proteins, and membrane lipids, ultimately resulting in cell death (17). Iron bound to ferritin behaves inertly and does not participate in the Fenton reaction. Thus, a higher level of intracellular ferritin can resist ferroptosis. In addition, if ferritin is depleted and Fe²⁺ is released into the LIP, cells become more sensitive to ferroptosis (18).

Lipid peroxidation

Cellular lipid metabolism cannot occur without the involvement of fatty acids, which perform several functions including acting as signaling molecules, maintaining the morphology of cell membranes, and participating in energy supply (19). Dysregulation of fatty acid metabolism may contribute to cellular damage via pyroptosis and ferroptosis (20). PUFAs are involved in the composition of cell membranes and are easily oxidized by reactive oxygen species (ROS) due to their weak C-H bond at the diallyl position (21). If the chemical structure of PUFA contains a higher number of double bonds, the easier it is oxidized (22). Many hypotheses exist as to the cause of cellular ferroptosis due to lipid peroxidation. On the one hand, the extensive lipid peroxidation of PUFAs disrupts the chemical structure of the lipid bilayer, which is an important component of the cell membrane, resulting in altering the permeability of the cell membrane and destroying the membrane barrier (23). On the other hand, PUFAs are capable of producing toxic derivatives via the breakdown of enzymatic and non-enzymatic pathways, including 4-hydroxynonenal (4-HNE) and malondialdehyde. These toxic products could destroy DNA bases, proteins, and other nucleophilic molecules and further amplify intracellular ROS accumulation, leading to severe cytotoxicity (24, 25).

Arachidonic acid (AA) and adrenic acid (AdA) are the main PUFAs involved in the process of ferroptosis (26). Briefly, AA is catalyzed by acyl-CoA synthetase long-chain family member 4 to bind CoA to form CoA-AA. Then the AA-phosphatidylethanolamines (PE-AA) are generated by esterification with lysophosphatidylcholine acyltransferase 3 (2, 27, 28, 29). The formed PE-AA is involved in the synthesis of PE-AA-OOH in the presence of lipoxygenases (LOXs), ultimately leading to lipid peroxidation. AdA undergoes the same series of reactions and acts on cell membrane phospholipids. Among them, LOXs belong to the family of lipid peroxidases that play a role in various types of PCD. Some of them are non-heme iron-dependent dioxygenases that could directly oxidize PUFAs on biological membranes and enhance susceptibility to ferroptosis (8, 30).

Regulation of ferroptosis

In recent years, numerous researchers have invested in the field of ferroptosis, and many regulatory mechanisms have been discovered. Among them, the system X_c ⁻/GSH/GPX4 axis, FSP1 pathway, and GCH1/DHFR/BH4 axis play important roles in the regulation of ferroptosis (Fig. 2).

Figure 2.

Regulation of ferroptosis: (1) The system X_c⁻/GSH/GPX4 axis: cystine is transported into the cell via system X_c⁻ distributed in the plasma membrane and reduced to cysteine by TXNRD1. GSH is synthesized from glutamate, cysteine, and glycine under the catalytic action of GCL. GPX4 is widely present in the cytoplasm, nucleus, mitochondria, and other organelles and suppresses the process of lipid peroxidation via both oxidation and reduction phases. GSH is used to supplement the active site residues to maintain the reduced state of GPX4. P53 and OTUB1 could promote system X_c⁻ activation, and BAP1 plays an inhibitory role. GPX4 inhibitors, including RSL3, FINO2, FIN56, ML162, and DPI7, could lead to lipid peroxidation. (2) The FSP1 pathway: the reduced CoQH2 plays an essential role as a free radical-trapping antioxidant in the fight against lipid peroxidation. (3) GCH1/DHFR/BH4 axis: BH4 could act alone or with vitamin E to protect against membrane phospholipid peroxidation. In this pathway, GCH1 and DHFR are the rate-limiting and catalytic enzymes of BH4, respectively. BAP1, BRCA1-associated protein 1; DHFR, dihydrofolate reductase; GCL, glutamate–cysteine ligase; GCH1, GTP cyclohydrolase-1; GSS, glutathione synthetase; GSH, glutathione; OTUB1, ubiquitin aldehyde-binding 1.

System X_c⁻/GSH/GPX4 axis

According to previous studies, inhibition or disruption of system X_c⁻ causes intracellular accumulation of lipid peroxides and their degradation products, which damage the plasma membrane and organelles (31, 32). As an important antioxidant system in cells, system X_c⁻ is widely distributed in the phospholipid bilayer of the plasma membrane (33). System X_c⁻ is a cystine/glutamate reverse transporter consisting of SLC3A2 and SLC7A11 that facilitates the exchange of cystine and glutamate across the plasma membrane (34). After intracellular transport of cystine, glutathione (GSH), and thioredoxin reductase 1 (TXNRD1) can reduce it to cysteine (35). In addition, as a semi-essential amino acid, cysteine can be produced intracellularly from methionine via the transsulfur pathway (36). Related studies have shown that over-expression of the p53 gene inhibits SLC7A11 function, restricts cystine transport, and induces ferroptosis in tumor cells (37). It has been found that BRCA1-associated protein 1 and ubiquitin aldehyde-binding 1 are able to act on system X_c⁻ to regulate the sensitivity of tumor cells to ferroptosis and control tumor cell proliferation (38, 39). This suggests the regulation of system X_c⁻ is a viable means of inhibiting tumor cell proliferation.

GSH is synthesized from glutamate, cysteine, and glycine following a two-step catalysis by the cytosolic enzymes glutamate–cysteine ligase and glutathione synthetase and is involved in the regulation of lipid peroxidation (40). According to previous studies, intracellular GSH was significantly reduced after system X_c⁻ was selectively suppressed, causing ROS accumulation and ultimately leading to ferroptosis occur (41, 42).

GPX4 is a selenocysteine-containing and GSH-dependent enzyme capable of reducing specific lipid hydroperoxides to lipid alcohols (43). GPX4 is a multifunctional protein that reduces lipid peroxides in free or complex form and regulates cellular ferroptosis in coordination with GSH. GPX4 is widely present in the cytoplasm, nucleus, mitochondria, and other organelles and suppresses the process of lipid peroxidation via both oxidation and reduction phases. The first stage involves the reduction of lipid peroxides by the active site of selenocysteine or cysteine, with the simultaneous oxidation of GPX4. In the second stage, the reducing substrate GSH is used to supplement the active site residues to reduce the oxidized GPX4 active site, and GSH is oxidized at the same time (44). Related studies showed that GPX4 inhibitors including RSL3, FINO2, FIN56, ML162, and DPI7 were able to cause lipid peroxidation of cell membranes and induce ferroptosis (32, 33).

The FSP1 pathway

Related studies demonstrated that GPX4 inhibitor-induced ferroptosis is highly variable across cell lines. The CRISPR/Cas9 screen revealed the existence of other pathways regulating cellular ferroptosis, of which FSP1 is a recently identified important regulatory gene (45, 46). FSP1 is also named apoptosis-inducing factor (AIF)-like mitochondrion-associated inducer of death (AMID) or AIF mitochondria-associated 2 (AIFM2) because of its similar amino acid sequence to the human AIF (47, 48). With the involvement of NADPH, FSP1 catalyzes the production of the downstream product CoQ10, which could attenuate ROS aggregation (49). Coenzyme Q10 (CoQ10), a lipid-soluble quinone compound, is the only lipid-based antioxidant that can be autosynthesized in vivo. CoQ10 is widely distributed in mammalian cell membranes, and non-mitochondrial CoQ10 is responsible for electron transportation as a reversible redox carrier in the plasma membrane and Golgi membrane (50, 51). CoQ10 exists in organisms in three forms: the oxidized ubiquinone form (CoQ), the semi-oxidized semiquinone form (CoQH), and the fully reduced ubiquinone form (CoQH₂). Among them, the reduced CoQH₂ plays an essential role as a free radical-trapping antioxidant in the fight against lipid peroxidation (52, 53).

It has been demonstrated that the addition of CoQ10 to cell lines in vitro could prevent ferroptosis (52). Several studies have shown that the oral drug CoQ10 can treat diseases such as cardiomyopathy, Parkinson’s disease, and diabetes (54, 55, 56). However, deficiencies in CoQ10 biosynthetic enzymes or related enzymes may result in limited CoQ10 synthesis and make people more susceptible to the above diseases (57, 58, 59).

GCH1/DHFR/BH4 axis

In addition to the above two major ferroptosis regulatory pathways, the system X_c⁻/GSH/GPX4 axis and the FSP1 pathway, investors have recently identified that GTP cyclohydrolase-1 (GCH1) and its metabolic derivatives tetrahydrobiopterin/dihydrobiopterin (BH4/BH2) are involved in the regulation of lipid peroxidation (60). GCH1 is the rate-limiting enzyme of BH4. Chemically, BH4 possesses two polyunsaturated fatty acyl groups which act alone or with vitamin E to inhibit plasma membrane phospholipid oxidation (61). Dihydrofolate reductase (DHFR) plays a catalytic role in the production of the derivative BH4. If DHFR expression is attenuated, cells become more sensitive to ferroptosis. The activation of the GCH1/DHFR/BH4 axis acts directly on peroxides or phospholipid radicals in the lipid bilayer to prevent lipid peroxidation (8). In conclusion, a growing body of evidence has demonstrated that the GCH1/DHFR/BH4 axis is very important for cellular energy metabolism and lipid metabolism (62, 63).

Bone cell physiology and ferroptosis

The maintenance of bone homeostasis requires the participation of both osteoclasts and osteoblasts, which are in dynamic balance. Excessive osteogenic capacity of osteoblasts may result in osteolithiasis, while overactive bone resorption by osteoclasts leads to OP (64). Osteocytes are terminally developed osteoblasts embedded in the mineralized bone matrix that signal with the surrounding microenvironment (65). In this section, we intend to describe these different types of cells in the setting of ferroptosis: osteoblasts, osteoclasts, and osteocytes.

Osteoblast in the setting of ferroptosis

The osteoblast lineage originates from mesenchymal precursor cells and eventually differentiates into osteoblasts through the sequential action of transcription factors. Osteoblasts secrete extracellular proteins, including osteocalcin, alkaline phosphatase, and type I collagen, the latter of which accounts for more than 90% of bone matrix proteins (64).

Luo et al. simulated in vitro ferroptosis in osteoblasts using ferric ammonium citrate, which upregulates cellular TFR1 and DMT1 gene expression to increase intracellular iron uptake. Ferroptosis inhibitors like iron chelator protect osteoblasts and maintain their differentiation by restoring the traditional Wnt signaling pathway (66). Iron overload may suppress ERK, AKT, and Stat3 phosphorylation and induce apoptosis in MC3T3-E1 cells (67). Lin et al. demonstrated that rats fed a high-glucose and high-fat diet exhibited bone mass loss, significantly increased serum ferritin levels, and decreased expression of SLC7A11 and GPX4. High glucose and palmitic acid-treated osteoblast cell lines were less capable of osteogenic differentiation and mineralization and underwent ferroptosis. Knockout of methyltransferase-like 3 (METTL3), one of the major m6A methyltransferases, inhibited osteoblast ferroptosis by blocking the METTL3/ASK1-P38 signal pathway (68). Research showed that osteoblast ferroptosis is an important factor in age-related OP. Immunofluorescence analysis of vitamin D receptor knockout mice indicated the suppression of the Nrf2/GPX4 signal pathway (69).

Osteoclast in the setting of ferroptosis

Mature osteoclasts are multinucleated giant cells originating from the monocyte/macrophage lineage. Osteoclasts break down and absorb bone matrix by secreting H⁺ and proteolytic enzymes like cathepsin K (69).

It was found that the iron-starvation response and ferritinophagy contribute to osteoclast differentiation under normoxia. After targeted inhibition of HIF-1α, the iron-starvation response and ferritinophagy were attenuated, and bone loss was reduced in ovariectomy (OVX)-induced OP mice (70). Zhong et al. found that osteoclast precursors (pre-OCs) were more sensitive to TXNRD1 inhibitor than bone marrow-derived monocytes during RANKL-induced osteoclast differentiation in vitro. The nuclear factor of activated T-cells 1 was able to upregulate SLC7A11 expression in pre-OCs and enhance the sensitivity of pre-OCs to TXNRD1. In vivo experiments demonstrated that TXNRD1 inhibitor protects bone microstructure in OVX mice by increasing the uptake of cystine and reducing oxidative stress (71). Qu et al. demonstrated that after bisphosphonates treatment, F-box protein 9 could act on P53, leading to ubiquitination and degradation of P53 and ultimately inducing ferroptosis in osteoclasts (72).

Osteocytes in the setting of ferroptosis

Osteocytes are terminally developed osteoblasts that take responsibility for communicating with their surroundings after being embedded in the mineralized bone matrix. Osteocytes can act on osteoblasts and osteoclasts by transmitting signaling molecules (65, 73).

It was shown that in diabetes OP (DOP) mice, osteoblasts underwent ferroptosis and their femur bone mass was significantly reduced. Molecularly, the expression of heme oxygenase-1 (HO-1) was significantly enhanced in osteocytes, which was regulated by the interaction of upstream signaling NRF2 and c-JUN transcription factors (74). Resveratrol could suppress NF-κB signaling via the SLC7A11/GPX4 axis and alleviate diabetic periodontitis, which leads to osteocyte ferroptosis (75). Thus, ferroptosis has a great impact on osteocyte activity in skeletal diseases.

Ferroptosis and osteoporosis animal models

OP is a systemic bone metabolism disorder characterized by decreased bone mass and destruction of bone microstructure, resulting in increased bone fragility and fracture susceptibility (76). OP is divided into two categories: primary OP and secondary OP. Primary OP mainly includes senile OP and postmenopausal OP (PMOP). Many diseases may cause secondary OP, such as type 2 diabetes OP (T2DOP) and glucocorticoid-induced OP (GIOP). Ferroptosis is a novel form of cell death regulation that has been discovered in recent years and is involved in the development of a variety of diseases including drug-resistant cancers, ischemic organ injuries, and other degenerative diseases (8). Numerous studies demonstrated that OP progression is closely related to oxidative stress, ROS accumulation, and lipid peroxidation (77, 78). As mentioned earlier, iron metabolism dysregulation and ROS accumulation leading to membrane phospholipid peroxidation are key contributors to ferroptosis. Therefore, targeted inhibition of cellular ferroptosis is a potential therapeutic method to slow the progression of OP. Although there are no clinical trials to prove the efficacy and safety of anti-ferroptosis drugs used to treat OP patients, researchers continue to look for evidence from animal models of OP. The results demonstrated that anti-ferroptosis drugs could alleviate bone mass loss to some extent in animal models of OP (70, 79, 80) (Table 1).

Table 1.

Interventions against ferroptosis in the treatment of osteoporosis.

Interventions	Cell types	Animal models	Mechanisms	References
Melatonin	MC3T3-E1 cells	T2DOP mice	Suppressed osteoblasts ferroptosis via activating Nrf2/HO-1 pathway	(83)
FtMt	hFOB1.19 cells	T2DOP mice	Protected osteoblasts against ferroptosis by reducing ROS accumulation	(79)
Metformin	Bone marrow progenitor cells	DOP mice	Inhibited osteoblast ferroptosis possibly mediating Runx2/Cbfa1 and AMPK activation	(84)
2ME2	Bone marrow-derived macrophages	PMOP mice	Targeted HIF-1α and induced osteoclasts ferroptosis	(70)
EC-Exos	Bone marrow-derived macrophages	PMOP mice	Attenuated osteoclast formation and ameliorated bone mass loss in OVX mice	(99)
EC-Exos	MC3T3-E1 cells	GIOP mice	Inhibited glucocorticoid-induced osteoblasts ferroptosis	(80, 98)

EC-Exos, endothelial cell-secreted exosomes; FtMt, mitochondrial ferritin; 2ME2, 2-methoxyestradiol.

DOP and ferroptosis

Diabetes can be classified into two types: type 1 diabetes mellitus due to insulin deficiency caused by damaged pancreatic B-cells, and type 2 diabetes mellitus due to insulin resistance resulting in defective insulin secretion (81). As it stands, DOP is due to deficiencies in glucose/insulin metabolism, the buildup of advanced glycosylated end products, and a lack of bone microvasculature resulting in an imbalance of bone metabolism, which is a novel syndrome (82). Recent studies have demonstrated that ROS aggregation and lipid peroxidation in ferroptosis are closely associated with DOP progression (74).

According to related studies, signal molecules and pathways such as the Nrf2/HO-1 axis have been shown to alleviate T2DOP in vivo and in vitro. Ma et al. found that the expression of GPX4 in bone tissue was attenuated in the DOP rat model, but melatonin could suppress osteoblast ferroptosis and protect the bone microarchitecture in vivo (83). Yang et al. showed that HO-1 expression was significantly enhanced in osteoblasts undergoing ferroptosis by RNA sequencing screen. After the inhibition of the upstream interaction between Nrf2 and c-JUN transcription factors, intracellular ROS accumulation and lipid peroxidation were alleviated, thus avoiding osteoblast ferroptosis and protecting bone tissue in DOP mice (74). The research found that the presence of large amounts of ferrous ions in the T2DOP rat model may cause oxidative stress, ROS aggregation, and osteoblast ferroptosis. Mitochondrial ferritin was able to attenuate osteoblast ferroptosis and reduce bone loss in T2DOP rats (79). Interestingly, the glucose-lowering drug metformin enhanced alkaline phosphatase activity, type I collagen synthesis, and osteocalcin expression in streptozotocin-induced diabetic rats. In vitro experiments showed that metformin enhanced the expression of the osteoblast transcription factor Runx2/Cbfa1 by activating AMPK signaling (84).

PMOP and ferroptosis

Estrogen deficiency leads to a disruption of the balance between osteoblastic bone formation and osteoclastic bone resorption in postmenopausal women. The eventual reduction of bone mass and destruction of bone microstructure in the skeleton of the organism are defined as PMOP (85). It is undeniable that the level of iron in the body plays an important role in the development of PMOP disease. By examining serum mineral levels in a PMOP population, Emre et al. revealed that the decrease in bone mineral density in the osteoporotic group was strongly associated with low serum iron concentrations (86).

Bone volume fractions in iron-deficient rats are significantly lower than those in rats with normal iron supplementation. Patrizia et al. found a positive correlation between iron metabolism and low-density lipoprotein oxidation. Haptoglobin phenotypes (Hp), which act as cellular antioxidants, are expressed highly in patients with low serum iron levels and are an important protective factor against osteoporotic fragility fractures (87). However, no direct association between bone mass and iron metabolism has been found in humans. The study by Jaime et al. conducted on iron-overload C57/BL6 mice showed for the first time that iron overload caused thinning of the bone cortex and reduction of the bone trabeculae in the femur compared to the control group (88). Ferroptosis was found to be involved in the process of RANKL-induced osteoclast differentiation. Mechanistically, iron metabolism dysregulation and ferritinophagy could induce osteoclast differentiation. In vitro studies demonstrated that an HIF-1α inhibitor suppressed osteoclast formation and differentiation by acting on ferritinophagy. The HIF-1α specific inhibitor 2ME2 (2-methoxyestradiol) could delay OVX-induced bone mass loss in vivo. Therefore, targeting HIF-1α and ferritinophagy may be a potential treatment method for PMOP (70).

GIOP and ferroptosis

Glucocorticoids (GCs) are widely used to treat a variety of inflammatory diseases (89). However, GCs also bring with them several toxic side effects, including Cushing’s-like syndrome, stomach ulcers, and OP (90, 91). In patients treated with GCs for a long time, more than 10% have been diagnosed with clinical fractures, and 30–40% of these patients have radiographic findings of vertebral fractures (92, 93). Bone loss resulting from GIOP is positively correlated with the dose and duration of GC usage. Bone loss and bone trabecular destruction were alleviated in some patients after discontinuation of the drug (94, 95, 96). Iron metabolism was found to be disturbed in dexamethasone (Dex)-induced GIOP mice. The level of oxidative stress and lipid peroxidation was enhanced, and ferroptosis played an important role in GIOP (97).

In the Dex-induced GIOP in vitro model, GPX4 expression was significantly reduced in osteoblasts, leading to increased cellular oxidative stress and lipid peroxidation. Similarly, researchers showed that GIOP model mice exhibited thinning of the bone cortex, reduction of bone trabeculae, and loss of bone mass. Endothelial cell-secreted exosomes (EC-Exos) were able to rescue osteoblasts that underwent ferroptosis and protect the bone microstructure in GIOP model mice. Interestingly, EC-Exos could also attenuate osteoclast activity and ameliorate bone-mass loss in OVX-induced OP mice (80, 98, 99).

Discussion

Ferroptosis is an iron-dependent PCD that causes intracellular oxidative stress, ROS accumulation, and lipid peroxidation. Morphologically, ferroptotic cells manifest plasma membrane disruption, cytoplasmic swelling, and mitochondrial condensation (2, 8). In recent years, numerous studies have demonstrated that ferroptosis plays an important role in the progression of OP. This review shows the role of iron metabolism, lipid metabolism, and the regulatory mechanisms in ferroptosis. According to existing findings, possible changes in signaling mechanisms within osteoblasts, osteoclasts, and osteocytes after the initiation of iron metabolism dysregulation are explained here. Studies have identified oxidative stress and lipid peroxidation in related OP models. Inhibitors and EC-Exos targeting ferroptosis were able to ameliorate bone mass loss in osteoporotic mice by protecting cells against ferroptosis (Fig 3).

Figure 3.

Ferroptosis plays an essential role in the progression of osteoporosis: from cell physiology to animal models. (1) Possible changes in signaling mechanisms within osteoblasts, osteoclasts, and osteocyte in the setting of ferroptosis. (2) Different types of osteoporotic mice (DOP, PMOP, GIOP) exhibit osteoclasts’ overactivity and bone microstructure destruction. Inhibitors and EC-Exos targeting ferroptosis were able to ameliorate bone mass loss in vivo. DOP, diabetes osteoporosis; GIOP, glucocorticoid-induced osteoporosis; PMOP, postmenopausal osteoporosis.

However, recognizing ferroptotic cells at an early stage is relatively difficult owing to the lack of molecular mechanism information (98). Mechanistic studies on the molecular biology of ferroptosis are still superficial, and the crosstalk between related signaling pathways remains difficult to explain (100). Signaling molecular detection and therapeutic tools are not currently available for clinical application. The underlying mechanisms of ferroptosis in the development and progression of OP still need to be further explored. We hope that more effective and appropriate clinical therapeutic means will appear in the near future.

ICMJE Conflict of Interest Statement

The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the study reported.

Funding Statement

This work was supported by National Key R&D Program of China (grant no. 2022YFC2504300), National Natural Science Foundation of China (grant no. 82372395) and The Special Project for Promoting High-Quality Development of Industries of Shanghai Municipal Commission of Economy and Informatization (grant no. 2023-GZL-RGZN-01037).

Author contribution statement

JL and LZ conceived and designed the study. XW produced pictures and drafted the manuscript. XW, XF, and FL conducted the literature review. XF, FL, and QC prepared the table. All authors read the manuscript and approved the order of authors’ presentation.