The Role of Ferroptosis Induced by Iron Overload in Osteoblast and Osteoclast Function

ABSTRACT

Iron overload disrupts bone homeostasis by suppressing osteoblast survival and mineralization, while promoting osteoclastogenesis. As a programmed cell death driven by iron‐dependent lipid peroxidation and glutathione peroxidase 4 (GPX4) downregulation. This study investigated the role of ferroptosis in bone cells under iron overload. Exposure to ferric ammonium citrate (FAC) led to decreased osteoblast viability, downregulation of GPX4 and nuclear factor erythroid 2‐related factor 2 (Nrf2) protein expression and osteoblastic markers. It also increased ROS production and lipid peroxidation in osteoblasts. In addition, intracellular reduced glutathione (GSH) levels showed a decreasing trend in osteoblasts exposed to FAC. Ferrostatin‐1 (Fer‐1), a ferroptosis inhibitor, significantly restored osteoblast viability and reduced lipid peroxidation, confirming the involvement of ferroptosis in iron‐induced cytotoxicity. Additionally, iron overload increased the receptor activator of nuclear factor‐κB ligand (RANKL)/osteoprotegerin (OPG) ratio, thereby promoting osteoclast differentiation. Interestingly, Fer‐1 effectively attenuated these osteoclastogenic effects. On the other hand, FAC increased RAW264.7 osteoclast precursor cells viability, ROS production and lipid peroxidation; while, Fer‐1 suppressed lipid peroxidation without affecting cell viability and ROS levels. Furthermore, we examined the role of cystine, an essential precursor for GSH synthesis and critical antioxidant cofactor for GPX4, in bone cells under iron‐overload conditions. Cystine enhanced osteoblast viability and reduced osteoclast viability under iron overload, but it did not exhibit a synergistic effect with Fer‐1. These findings highlight ferroptosis as an underlying mechanism for iron‐induced osteoporosis and may represent a promising therapeutic strategy for managing iron overload‐associated bone disorders.

Abbreviations

ACSL4

acyl‐CoA synthetase long‐chain family member 4

ALP

alkaline phosphatase

ARE

antioxidant response element

FAC

ferric ammonium citrate

Fer‐1

ferrostatin‐1

GPX4

glutathione peroxidase 4

GSH

reduced glutathione

MAPK

mitogen‐activated protein kinase

NAC

N‐acetylcysteine

NF‐κB

nuclear factor kappa B

Nrf2

nuclear factor erythroid 2‐related factor 2

PUFAs

polyunsaturated fatty acids

RANKL

receptor activator of nuclear factor‐κB ligand

ROS

reactive oxygen species

TRAP

tartrate‐resistant acid phosphatase

Wnt

wingless‐related integration site

1. Introduction

Iron is an essential element that plays a crucial role in various biological functions, but excess iron can lead to detrimental effects. About 40%–50% of patients with iron overload, including those with hemochromatosis and thalassemia, develop osteopenia and osteoporosis (Rossi et al. 2014; Valenti et al. 2009). Previous studies conducted both in vitro and in vivo have shown the relationship between iron overload and bone pathology, such as reduced bone mass and alteration of bone microarchitecture (Jeney 2017; Mitchell 2012). While bone homeostasis is tightly regulated by the interaction of bone‐forming osteoblasts and bone‐resorbing osteoclasts (Bolamperti et al. 2022), iron overload disrupted this balance by interfering with both osteoblast and osteoclast viability and activity.

Our previous studies demonstrated that iron directly inhibited osteoblast survival, differentiation, and mineralization by generating reactive oxygen species (ROS), triggering apoptotic cell death, and cell cycle arrest (Lertsuwan et al. 2018; Lertsuwan et al. 2020). Similarly, iron‐overloaded mouse model also exhibited the defected bone microstructure with trabecular and cortical bone thinning accompanied by decreased bone formation but increased bone resorption. These mice also have increased ROS levels in the bone marrow and elevated concentration of pro‐inflammatory cytokines, i.e., tumor necrosis factor‐α (TNF‐α) and interleukin‐6 (IL‐6), which are associated with iron overload intensity (Tsay et al. 2010). Overall, iron overload causes an imbalance between bone formation and resorption, leading to impaired bone homeostasis and osteoporosis.

Ferroptosis is a programmed cell death characterized by iron‐dependent lipid peroxidation. Since an intracellular iron accumulation triggers the production of ROS through the Fenton reaction. Consequently, these free radicals interact with polyunsaturated fatty acids (PUFAs) present in cellular membranes, leading to the initiation of lipid peroxidation. This mechanism disrupts the constancy of the cell membrane lipid bilayer and triggers morphological changes and cell death (P. Liu et al. 2022). This mechanism also plays an important role in various pathological conditions, including cancer, kidney damage, and neurological diseases (Stockwell et al. 2017). Since osteoporosis and osteoblast cell death were observed in iron overload conditions both in vitro and in vivo, it is interesting to determine whether ferroptosis is involved in these phenomena.

The decrease in glutathione peroxidase 4 (GPX4) level and activity has been used as one of the hallmarks of ferroptosis. GPX4 suppresses ferroptotic cell death by utilizing the essential cofactor glutathione (GSH) to detoxify lipid peroxidation (Yang et al. 2014). The biosynthesis of GSH requires the uptake of cystine, which is subsequently converted to cysteine used for GSH synthesis (Forcina and Dixon 2019). Previous studies have shown that depletion of cystine reduces GSH synthesis and GPX4 expression (Yu and Long 2016; Zhang et al. 2021). Notably, cystine could promote both the synthesis of GSH and protein expression of GPX4, partly via Rag‐mTORC1‐4EBP signaling (Zhang et al. 2021), suggesting the protective roles of cystine against ferroptotic cell death.

A few studies have demonstrated the involvement of ferroptosis in bone homeostasis. Its involvement in osteoblast cell death and bone formation suppression under iron overload was reported in mouse osteoblast precursors MC3T3‐E1 cells and iron overload mice (Jiang et al. 2022; Xiong et al. 2022). However, this phenomenon still needs to be confirmed and whether this mechanism also involves in osteoblast differentiation and function suppression under iron overload is not known. In this study, the involvement of ferroptosis in osteoblast cell survival is confirmed, and its effects on osteoblast differentiation and function are further elaborated in both osteoblast precursors (MC3T3‐E1 cells) and a mature osteoblast model (UMR‐106 cells). MC3T3‐E1 is a widely used pre‐osteoblast cell line that expresses key osteogenic markers such as alkaline phosphatase and osteocalcin, and exhibits ALP activity, proliferation, and mineral deposition abilities comparable to those of primary osteoblasts, with similar growth kinetics and calcium accumulation (Czekanska et al. 2014). On the other hand, UMR‐106 cells are a highly differentiated osteogenic cell line exhibiting mature osteoblast characteristics and parathyroid hormone (PTH) responsiveness similar to normal osteoblasts in vivo (Teerapornpuntakit et al. 2016).

While iron overload was shown to negatively affect bone formation, the opposite effect was reported for bone degradation when higher bone resorption activity was found in iron overload mice (Thongchote et al. 2014; Tsay et al. 2010). Interestingly, whether ferroptosis also involve in this process is not clear. Therefore, RAW 264.7 mouse macrophage‐like cells were used as an osteoclast precursor line to investigate the involvement of ferroptosis in osteoclast differentiation under iron overload.

Accordingly, ferroptosis inhibitor ferrostatin‐1 (Fer‐1) and/or GSH precursor (cystine) were used to investigate the effects of ferroptosis inhibition on osteoblast and osteoclast cell viability and function under iron overload. Taken together, this study illustrated the involvement of ferroptosis in osteoblast and osteoclast viability and function under iron overload. This would provide crucial information on the mechanism underlying iron overload‐induced osteoporosis as well as potential therapeutic targets for this condition.

2. Materials and Methods

2.1. Cell Culture

Rat osteoblast‐like osteosarcoma cell line UMR‐106 (ATCC, VA, USA) and mouse macrophage‐like cell line RAW 264.7 (ATCC) were used to represent osteoblast and osteoclast precursor cell lines as verified in (Collin‐Osdoby et al. 2003; Forrest et al. 1985). They were maintained in Dulbecco's Modified Eagle Medium (DMEM; Gibco, TX, USA). Pre‐osteoblastic MC3T3‐E1 cell (ATCC) was grown in alpha Minimum Essential Medium (α‐MEM) with ribonucleosides, deoxyribonucleosides, 2 mM l‐glutamine, and 1 mM sodium pyruvate, but without ascorbic acid (Gibco). Both culture media were supplemented with 10% (v/v) fetal bovine serum (FBS; Gibco) and 1% (v/v) penicillin‐streptomycin (Gibco). Cells were incubated at 37°C under a humidified atmosphere containing 5% CO₂ and subcultured according to the ATCC's protocol.

2.2. Chemicals

Ferric ammonium citrate (FAC), ferrostatin‐1 (Fer‐1) and primary β‐actin antibody (A2066) were purchased from (Sigma‐Aldrich, MO, USA). Antibodies against GPX4 (ab125066) was purchased from Abcam (Cambridge, UK). Antibodies against nuclear factor erythroid 2‐related factor 2 (Nrf2, MA5‐38583) was purchased from Thermo Fisher Scientific (MA, USA). Antibodies against rabbit IgG‐HRP (ab2099233) conjugated was purchased from Cell Signaling Technology Inc (MA, USA).

2.3. Gelatin Methacrylate (GelMA) Synthesis

GelMA synthesis protocol was adapted from Shirahama et al. 2016 (Shirahama et al. 2016) and Loessner et al. 2016 (Loessner et al. 2016). In brief, type A porcine skin gelatin (Sigma‐ Aldrich) was dissolved in 0.1 M carbonate–bicarbonate buffer to a final concentration of 10% w/v. Then, methacrylic anhydride (Sigma‐Aldrich) was added at a ratio of 0.1 mL per 1 g of gelatin. After the reaction, the mixture was centrifuged, and the supernatant was collected. The pH was then adjusted to 7.4. The solution was diluted with deionized water and dialyzed against deionized water using 10 kDa MWCO SnakeSkin Dialysis Tubing (Thermo, MA, USA). Following dialysis, the solution was frozen at –80°C and lyophilized until dry.

2.4. MC3T3‐E1 Cell Culture on GelMA Hydrogel

GelMA synthesized in our laboratory was dissolved in 2‐hydroxy‐4′‐(2‐hydroxyethoxy)‐2‐methylpropiophenone (Sigma‐Aldrich) prepared in phosphate‐buffered saline (PBS) to final concentrations of 10% (w/v) and 0.5% (w/v), respectively. A volume of 500 µL of the solution was added to each well of a 24‐well plate. Crosslinking was initiated by exposing the solution to UV light at 355 nm for 15 min. The resulting hydrogel was allowed to cool at room temperature for 10 min. Then, PBS was added to each well and incubated at 37°C with 5% CO₂ for 15 min. The PBS was then replaced with complete α‐MEM, and the hydrogel was incubated overnight. On the following day, MC3T3‐E1 cells were seeded directly onto the gel at a density of 5.0 × 10³ cells/well.

2.5. MTT Cell Viability Assay

Cell viability was determined using the tetrazolium‐based colorimetric assay (MTT assay) as described in (Lertsuwan et al. 2020). After the treatment, the medium was discarded, and MTT (Invitrogen, CA, USA) was added to achieve a final concentration of 1 mg/mL. The cells were incubated for 3 h at 37°C before the medium was discarded. Then, DMSO (Millipore, MA, USA) was added to dissolve the formazan crystal. The absorbance was measured at 595 nm by a microplate reader (Multiskan EX, Thermo Fisher Scientific, MA, USA).

2.6. CCK‐8 Cell Viability Assay

Cell viability was assessed using Cell Counting Kit‐8 (CCK‐8) (Dojindo Laboratories, Kumamoto, Japan). After treatment, a 1:10 dilution of CCK‐8 solution prepared in complete culture medium was added to each well of a 24‐well plate. The plate was incubated at 37°C with 5% CO₂ for 3 h. For MC3T3‐E1 cells grown on GelMA hydrogel, the supernatant was transferred to a 96‐well plate prior to optical density (OD) measurement at 450 nm using a microplate reader (Fisher Scientific). Percent cell viability was calculated using the following equation:

$$\mathrm{Cell}\mathrm{viability}(\%)=\frac{\mathit{OD}\left(\mathrm{experiment}\right)-\mathit{OD}\left(\mathrm{Blank}\right)}{\mathit{OD}\left(\mathrm{control}\right)-\mathit{OD}\left(\mathrm{Blank}\right)}\times 100$$

2.7. Lipid Peroxidation Assay

The Image‐iT Lipid Peroxidation Kit (ThermoFisher) was used to measure lipid peroxidation according to the manufacturer's instructions. Stained cells positive for lipid peroxidation were analyzed by flow cytometer (model FACSCanto; BD Biosciences, NJ, USA), and data was analyzed using FACSDiva version 6.1.3 software (BD Biosciences).

2.8. Cellular ROS Detection Assay

Cellular ROS was assessed using the 2′,7′‐dichlorofuorescin diacetate (DCFDA) Cellular ROS Detection Assay Kit (Abcam) according to the manufacturer's instructions. In brief, cells were seeded into a 96‐black well plate with clear bottom at 2.5 × 10⁴ cells/well overnight before stained with 25 μM DCFDA for 45 min at 37°C. Then, cells were washed with 1× buffer and treated with treatments for 6 h. A microplate reader (Spark 10 M, Tecan, Switzerland) was used to measure the fluorescence signal at the excitation and emission wavelengths of 485 and 535 nm, respectively.

2.9. Western Blot Analysis

After cell collection and washing with cold PBS, proteins were extracted from the cells using a modified radioimmunoassay precipitation (RIPA) buffer containing 10% protease inhibitor (Roche, USA). Cell lysates were centrifuged at 12,000 rpm at 4°C for 20 min. The supernatants were collected to measure protein concentration using BCA protein assay kit (Thermo scientific). Proteins (25 μg) were separated on 15% polyacrylamide gel. Proteins were then transferred onto nitrocellulose membrane. After that, membranes were incubated with specific antibodies for GPX4 (ab125066, Abcam), Nrf2 (MA5‐38583, Thermo Fisher Scientific), and β‐actin (A2066, Sigma‐Aldrich) diluted in blocking solution overnight at 4°C. Then, membranes were incubated with anti‐rabbit IgG secondary antibody (ab2099233, Cell Signaling). The membrane was visualized by chemiluminescence (ECL; Millipore, USA) on the Azure 600 imaging system (Azure Biosystems, CA, USA). Band intensity was quantified by using Image J software (National Institutes of Health, MD, USA).

2.10. Intracellular GSH Assay

Intracellular reduced GSH levels were measured using the GSH/GSSG Ratio Detection Assay Kit (ab138881; Abcam) according to the manufacturer's instructions. MC3T3‐E1 cells were seeded in 6‐well plates and allowed to attach overnight before treatment with 200 µM FAC for 18 h. Then, cells were washed twice with PBS and lysed using Mammalian Lysis Buffer (ab179835; Abcam). Cell lysates were processed for selective GSH detection, and fluorescence intensity was measured at excitation/emission wavelengths of 490/520 nm using a microplate reader (Multiskan EX; Thermo Fisher Scientific). GSH concentrations were calculated from a standard curve, normalized to total protein content determined by the BCA assay, and expressed as µM GSH/mg protein.

2.11. Quantitative Real‐Time PCR (qRT‐PCR)

RNA was extracted from the cells using TRIzol reagent (Invitrogen) according to the manufacturer's instructions. RNA was measured with NanoDrop‐2000c spectrophotometer (Thermo Fisher Scientific) at 260 and 280 nm. An acceptable range for the ratio between these two measurements was between 1.8 and 2.0. Then, RNA (1 μg) was converted to cDNA using the iScript cDNA synthesis kit (Bio‐Rad, CA, USA) with the thermal cycler (model MyCycler; Bio‐Rad). The primers used in this study are shown in Table 1. qRT‐PCR was operated by Bio‐Rad MiniOpticon using iTaq Universal SYBR Green Supermix (Bio‐Rad) according to the manufacturer's recommendation.

Table 1.

Rattus norvegicus primers used in this experiment.

Gene	Accession no.	Primer (forward/reverse)	Annealing temperature (°C)	Product size
Osteoblast‐specific marker genes
Alkaline phosphatase (ALP)	NM_013059	5′–AGAACTACATCCCCCACG–3′ 5′–CAGGCACAGTGGTCAAGGT–3′	58.00	144 bp
Collagen type I alpha 1 chain (Col1a1)	NM_053304.1	5’–CAGTCGATTCACCTACAGCAC–3’ 5’–GGGATGGAGGGAGTTTACACG–3’	59.00	194 bp
Osteoclastogenic genes
Receptor activator of NF‐κB ligand (RANKL)	NM_057149	5′–TCGCTCTGTTCCTGTACT–3′ 5′–AGTGCTTCTGTGTCTTCG–3′	47.00	145 bp
Osteoprotegerin (OPG)	NM_012870	5′–ATTGGCTGAGTGTTCTGGT–3′ 5′–CTGGTCTCTGTTTTGATGC–3′	50.00	140 bp
Interleukin 1β (IL‐1β)	NM_031512	5′–TCAAGCAGAGCACAGACCTGT–3′ 5′–TGAGAGACCTGACTTGGCAGA–3′	56.00	197 bp
Interleukin 6 (IL‐6)	NM_012589	5′–GCAAGAGACTTCCAGCCAGT–3′ 5′–AGCCTCCGACTTGTGAAGTG–3′	55.00	145 bp
Housekeeping gene
Hypoxanthine phosphoribosyltransferase 1 (Hprt1)	NM_012583	5′–GGCCAGACTTTGTTGGATTTG–3′ 5′–CTTTCGCTGATGACACAAACAT–3′	53.00	109 bp

2.12. Tartrate‐Resistant Acid Phosphatase (TRAP) Staining

RAW 264.7 cells were plated on 24‐well plates overnight before treated with 10 ng/mL RANKL (R&D Systems, MN, USA) with treatment for 4 days. The culture medium was changed every other day. Cells were then washed with PBS and fixed with 10% formalin neutral buffer. Each well was rinsed with deionized water before adding TRAP chromogenic substrates (Cosmo Bio, Tokyo, Japan) and incubated at 37°C for 60 min. Cells were washed with deionized water to stop the reaction. The number of trap‐positive multinucleated osteoclasts containing at least 3 nuclei was visualized and manually counted under an inverted microscope (model MBL3200, A. KRÜSS Optronic GmbH, Hamburg, Germany).

2.13. Alizarin Red Staining

UMR‐106 cells were treated with medium supplemented with 50 mM β‐glycerophosphate (Sigma‐Aldrich) and 50 μg/mL ascorbic acid (Sigma‐Aldrich) to induce mineralization for 6 days. The media was changed every other day. After that, the media was removed and gently washed with PBS. Cells were fixed with 70% cold ethanol for 1 h at 4°C, and then gently washed with deionized water 3 times. Calcium deposition was stained by 40 mM alizarin red S (Sigma‐Aldrich). Cells were washed with PBS 5 times and inspected under a light microscope (Nikon, USA). The pictures were taken to analyze the total area of red nodule formation using the Image J software (National Institutes of Health).

2.14. Statistical Analysis

Statistical analysis for multiple comparisons was performed by one‐way analysis of variance (ANOVA). The difference between pairs of means was analyzed by Tukey's test. The level of significance for statistical tests is at p ≤ 0.05. Data were analyzed by GraphPad Prism 9 (GraphPad Software Inc., CA, USA).

3. Results

3.1. Ferroptosis Inhibitor Could Rescue Osteoblast Cell Death but Not the Total Cellular ROS Under Iron Overload

Previous studies indicated that FAC could activate the apoptosis pathway in osteoblasts, as shown by the elevated levels of cleaved caspase 3 and 7 (Lertsuwan et al. 2018; Tian et al. 2016). Therefore, this study aims to investigate whether FAC also triggers another form of cell death known as ferroptosis. The potent and specific ferroptosis inhibitor Fer‐1 was used to examine the involvement of ferroptosis in iron‐treated osteoblasts. According to our and other published data, FAC at 200 µM at 72 h induced significantly increased intracellular iron to represent iron overload as well as caused considerable effects on osteoblast characteristics and gene expression (Jiang et al. 2022; Lertsuwan et al. 2018; Xia et al. 2019). Following 72 h treatment under complete growth media without Fer‐1, UMR‐106 osteoblast viability was significantly decreased by FAC treatment in a concentration‐dependent manner as compared to control. FAC treatment at 30, 100, and 200 µM decreased osteoblast viability to 62.28%, 39.95%, and 28.23%, respectively. Conversely, in the presence of Fer‐1, cell viability increased from 62.28% to 93.78%, 39.95% to 95.79%, and 28.23% to 99.64% in cells treated with 30, 100, and 200 µM of FAC, respectively (Figure 1A). Similar results could be seen in another osteoblast cell line (MC3T3‐E1) confirmed the protective effect of Fer‐1 on cell viability under iron overload (200 µM), with the viability percentage increasing approximately threefold as compared to FAC treatment alone (Figure 1B). These findings indicated that Fer‐1 can suppress cell death in osteoblasts treated with FAC, suggesting that iron overload also induced osteoblast cell death through the ferroptosis pathway. Effects of FAC and Fer‐1 on MC3T3‐E1 pre‐osteoblast cell viability cultured on GelMA hydrogel, which serves to mimic the interaction between cells and extracellular matrix (ECM) environment. The results demonstrated that FAC reduced MC3T3‐E1 cell viability; while, Fer‐1 significantly rescued cell viability under iron overload condition. This confirmed the protective role of ferroptosis inhibition in maintaining osteoblast survival under the interaction with extracellular matrix (Figure 1C).

Figure 1.

Cell viability of UMR‐106 cells (A; n = 3) and MC3T3‐E1 cells (B; n = 4 with 3 technical repeats each) measured by the MTT assay, and MC3T3‐E1 cells cultured on GelMA hydrogels (C; n = 4) measured by the CCK‐8 assay, following 72‐h exposure to FAC and Fer‐1; Cell viability of UMR‐106 (D) and MC3T3‐E1 cells (E) after exposure to FAC and NAC for 72 h (n = 4 with 3 technical repeats each); Cellular ROS production of UMR‐106 cells after exposure to FAC and Fer‐1 for 6 h (F; n = 4 with 3 technical repeats each). The results were expressed as mean ± SEM from at least three independent biological experiments. *p < 0.05, **p < 0.01, ***p < 0.001 analyzed by One‐way ANOVA with Tukey's multiple comparison test.

Furthermore, previous experiments indicated that iron‐induced osteoblast cell death was associated to ROS production (Lertsuwan et al. 2020). N‐acetyl‐l‐cysteine, a potent antioxidant, was used to determine osteoblasts cell viability under iron‐overload conditions. Figure 1D–E demonstrated that in the presence of iron overload, NAC treatment was unable to prevent osteoblastic cell death in the UMR‐106 or MC3T3‐E1 cell lines. As NAC did not significantly restore cell viability under iron‐overload conditions, its effects on cellular ROS and lipid peroxidation were not further investigated. Cellular ROS production in osteoblasts treated with FAC was also measured to investigate whether the protective effect of Fer‐1 was related to cellular ROS level. The results revealed a significant increase in cellular ROS levels in FAC‐treated osteoblasts as compared to control group (Figure 1F). Interestingly, Fer‐1 treatment did not alter ROS levels in osteoblasts under iron overload, suggesting that Fer‐1 could not attenuate iron‐induced total ROS production in osteoblasts. These results suggested that the protective effect of Fer‐1 on osteoblasts was not associated with the regulation of cellular ROS production under iron‐overload conditions.

3.2. Iron Overload Suppressed GPX4 and Nrf2 Expression While Increased Lipid Peroxidation in Osteoblast, Which Was Attenuated by Ferroptosis Inhibitor

It has been shown that inactivation of lipid‐repairing enzyme of GPX4 led to increased intracellular lipid peroxides, resulting in ferroptosis. Thus, GPX4 downregulation is considered one of the hallmarks for ferroptosis (Forcina and Dixon 2019). In the present study, the expression of GPX4 in UMR‐106 cells upon FAC exposure was, therefore, investigated. As shown in Figure 2A, FAC treatment significantly reduced GPX4 protein level in osteoblasts as compared to control.

Figure 2.

Representative image of GPX4 protein expression and the quantification (A; n = 4) in UMR‐106 cells exposed to FAC for 72 h; Nrf2 protein expression and the quantification (B; n = 3) in MC3T3‐E1 cells exposed to FAC for 72 h; Intracellular GSH in MC3T3‐E1 cells exposed to FAC for 18 h (C; n = 3); Lipid peroxidation (D; n = 4) in UMR‐106 cells exposed to FAC or Fer‐1 for 72 h. The results were expressed as mean ± SEM from at least three independent biological experiments. *p < 0.05, ***p < 0.001 analyzed by One‐way ANOVA with Tukey's multiple comparison test.

As Nrf2 plays a critical role in regulating antioxidant defense, iron metabolism, and lipid homeostasis associated with ferroptosis (Shakya et al. 2023; Yan et al. 2023), Nrf2 protein expression was evaluated under iron overload. FAC treatment markedly decreased Nrf2 protein expression in osteoblasts compared with the control group (Figure 2B). In parallel, intracellular GSH levels showed a decreasing trend in osteoblasts under iron‐overload (Figure 2C; p = 0.0576), suggesting a potential impairment of redox balance and reduced GSH availability, which may in turn compromise GPX4‐dependent detoxification of lipid peroxides.

Furthermore, we examined the effects of Fer‐1 on lipid peroxidation in osteoblast under iron overload. As shown in Figure 2D, FAC exposure led to increased lipid peroxidation in UMR‐106 cell by more than 50% as compared to control. Then, Fer‐1 significantly attenuated lipid peroxidation in osteoblast under iron overload. These results indicated that FAC increased lipid peroxidation and reduced GPX4 levels in osteoblasts, whereas Fer‐1 could effectively recover these effects. These results indicated that FAC induced ferroptosis in osteoblast UMR‐106 cells.

3.3. Iron Overload Reduced the Expression of Alkaline Phosphatase (Czekanska et al.), Collagen Type I Alpha 1 (Col1a1) and Osteoblast Mineralization

ALP and Col1a1 are the key osteoblastic proteins expressed in differentiated osteoblast (Ponzetti and Rucci 2021). Therefore, effects of FAC and Fer‐1 on ALP and Col1a1 expression in UMR‐106 cells were investigated. As shown in Figure 3A,B, FAC notably reduced ALP expression, but Fer‐1 could increase ALP expression. Under iron overload, Fer‐1 significantly restored the expression of ALP. On the other hand, although the expression of Col1a1 was not significantly affected, a positive trend in Col1a1 expression was observed upon Fer‐1 exposure.

Figure 3.

Effect of FAC and Fer‐1 on the expression of osteoblastic genes including alkaline phosphatase (A) and collagen type I alpha 1 (Col1a1) (B) after 72 h of treatment (n = 3); representative image of osteoblast mineralization (C), and the quantification (D) in UMR‐106 cells (n = 7); cell viability of UMR‐106 cells exposed to FAC and Fer‐1 in osteogenic media for 72 h (E; n = 3 with two technical replicates each). The results were expressed as mean ± SEM from at least three independent biological experiments. *p < 0.05, **p < 0.01, ***p < 0.001 analyzed by One‐way ANOVA with Tukey's multiple comparison test.

Further, osteoblast cells were induced to differentiate and exposed to FAC and Fer‐1 for 6 days. Subsequently, alizarin red staining was used to determine calcium deposition as an indicator of bone mineralization. The results illustrated in Figure 3C,D revealed a significant reduction in the mineralization of osteoblast nodules following FAC treatments. However, Fer‐1 treatment did not restore mineralization under iron overload conditions. Cell viability of UMR‐106 cells under osteogenic induction was also performed to confirm the effects of FAC and Fer‐1 on osteoblast viability under iron overload. Our findings demonstrate that osteoblasts exposed to FAC under osteogenic conditions experience a significant reduction in viability, indicating that iron overload adversely affects cell viability during osteogenesis under osteogenic inducing media (Figure 3E). Treatment with the combination of FAC and Fer‐1 (FF) also restores osteoblast viability to a level similar to the control, with no significant difference from either the FAC or Fer‐1 group.

3.4. Iron Overload Increased RANKL/OPG Expression Ratio in Osteoblast UMR‐106 Cells, But This Alteration Was Attenuated by Ferroptosis Inhibitor

Osteoblasts play a crucial role in influencing the formation and differentiation of osteoclasts through the release of osteoclast regulatory factors, such as receptor activator of RANKL, and OPG. Therefore, we investigated the effects of FAC and Fer‐1 on the expression of osteoclastogenic genes under iron overload by qRT‐PCR. The results revealed that FAC increased RANKL/OPG ratio in osteoblast, which reflected a shift of balance toward bone resorption. Conversely, Fer‐1 significantly decreased RANKL/OPG ratio in osteoblast under iron overload (Figure 4A–C). While the expression level of IL‐1β and IL‐6 was not affected by FAC and Fer‐1, the expression of IL‐1β was significantly decreased in UMR‐106 cells exposed to Fer‐1 under iron overload (Figure 4D,E). Taken together, these findings suggested that iron overload disrupted the osteoblast‐derived osteoclastogenic factors, which could increase osteoclast differentiation. On the other hand, a ferroptosis inhibitor was able to attenuate these deleterious effects.

Figure 4.

Effect of FAC and Fer‐1 on the expression of osteoclastogenic genes in UMR‐106 cells including RANKL (A), OPG (B), RANKL/OPG ratio (C), IL‐1β (D), and IL6 (E) after 72 h of treatment. The results were expressed as mean ± SEM from four independent biological experiments. *p < 0.05, **p < 0.01 analyzed by One‐way ANOVA with Tukey's multiple comparison test.

3.5. Iron Overload Enhanced the Viability and Lipid Peroxidation of RAW264.7 Cells, While, Ferroptosis Inhibitor Mitigated Lipid Peroxidation But Did Not Affect ROS Levels

Iron overload was reported to accelerate bone resorption (Wang et al. 2018). To investigate the effect of FAC and Fer‐1 on osteoclast viability and differentiation, the progenitor cells for macrophage‐monocyte and osteoclast linage, RAW264.7 cells, were utilized as osteoclast precursor cells in this study and others (Ishii et al. 2006; Kikuta et al. 2013; Kong et al. 2019). As shown in Figure 5A, RAW264.7 cell viability significantly increased with FAC in a dose‐dependent manner; while, Fer‐1 did not significantly alter RAW264.7 cell viability as compared to the FAC‐treated group. Furthermore, iron overload stimulated lipid peroxidation and ROS production in RAW264.7 cells. Fer‐1 alleviated lipid peroxidation in RAW264.7 cells under iron overload but not cellular ROS levels (Figure 5B,C). In addition, Fer‐1 treatment increased GPX4 protein expression, whereas FAC did not significantly affect its expression (Figure 5D). Taken together, our findings indicated that iron markedly promoted cell viability and lipid peroxidation in the progenitor cells for macrophage‐monocyte and osteoclast linage, and ferroptosis inhibitor could recover these effects.

Figure 5.

Effect of FAC and Fer‐1 on osteoclast cell viability (A; n = 4 with three technical replicates each), lipid peroxidation (B; n = 4), ROS production (C; n = 4 with three technical replicates each), and GPX4 protein expression (D; n = 4). The results were expressed as mean ± SEM from four independent biological experiments. *p < 0.05, **p < 0.01, ***p < 0.001 analyzed by One‐way ANOVA with Tukey's multiple comparison test.

3.6. Iron Overload Enhanced Osteoclast Differentiation in RAW 264.7 Cells, Which Was Mitigated by the Ferroptosis Inhibitor

In this study, the effect of Fer‐1 on osteoclast differentiation was investigated in RAW 264.7 cells under conditions of iron overload. Upon treatment with RANKL to induce osteoclastogenesis, it was observed that FAC stimulated osteoclast formation, as indicated by an increase in TRAP‐positive multinucleated cells with at least three nuclei (Figure 6A,B). Interestingly, Fer‐1 reduced the number of TRAP‐positive cells, regardless of the presence of iron. These results suggested that the ferroptosis inhibitor suppressed osteoclast differentiation under both normal and iron‐overload conditions.

Figure 6.

Effect of FAC and Fer‐1 on osteoclast differentiation of RAW264.7 cells. Representative image of TRAP staining from RAW264.7 cells treated with RANKL, alongside untreated controls, FAC‐treated, Fer‐1‐treated, and co‐treated with FAC and Fer‐1 (A), and the quantification (B). The results were expressed as mean ± SEM from four independent biological experiments. *p < 0.05, **p < 0.01, ***p < 0.001 analyzed by One‐way ANOVA with Tukey's multiple comparison test.

3.7. Cystine Enhanced Osteoblast Viability and Inhibited Osteoclast Viability Under Iron Overload, But It Did Not Exhibit a Synergistic Effect With Fer‐1

The inactivation of GPX4 is one of the hallmarks for ferroptosis. GPX4 plays a critical role in suppressing ferroptosis by utilizing the essential cofactor GSH to detoxify lipid peroxidation (Yang et al. 2014). The biosynthesis of GSH requires the uptake of cystine, which is subsequently converted to cysteine used for GSH synthesis (Forcina and Dixon 2019). Previous studies have demonstrated that cystine depletion leads to a reduction in GSH synthesis and GPX4 expression (Yu and Long 2016; Zhang et al. 2021). Therefore, this study aimed to investigate the effects of cystine on osteoblast and osteoclast progenitor cell viability under iron overload. Our data showed that 200 μM was selected as the highest concentration that has no effect on the viability of both osteoblasts and osteoclasts (Figure 7A,B). As indicated in Figure 7C, FAC reduced UMR‐106 cell viability in a dose‐dependent manner, and cystine rescued osteoblast cell viability under iron overload. While both Fer‐1 and cystine rescued osteoblast viability under iron overload, the co‐treatment of cystine and Fer‐1 did not exhibit a synergistic effect on osteoblast cell viability (Figure 7D). Moreover, FAC induced RAW264.7 cell viability as cystine suppressed this effect under iron overload (Figure 7E). This suggested that cystine probably had a dual effect, promoting osteoblast viability while suppressing osteoclast viability under iron‐overload conditions.

Figure 7.

Effect of cystine on UMR‐106 (A) and RAW264.7 (B) cell viability after 72 h of treatment. Effect of FAC, cystine and Fer‐1 on UMR‐106 (C and D) and RAW264.7 (E) cell viability after 72 h of treatment. The results were expressed as mean ± SEM from at least three independent biological replicates with three technical replicates each. *p < 0.05, **p < 0.01, ***p < 0.001 analyzed by One‐way ANOVA with Tukey's multiple comparison test.

4. Discussion

Iron overload has been shown to perturb bone homeostasis by promoting bone resorption by osteoclasts while suppressing bone formation by osteoblasts (Piriyakhuntorn et al. 2020). Previous studies from our laboratory showed that iron overload induced osteoblast cell death and disrupted osteoblastic gene expression; however, the mechanisms behind this phenomenon are not clear. Previous studies have shown that excessive iron accumulation leads to osteoblast cell death through several cell death pathways (Che et al. 2021; Jiang et al. 2022; Tian et al. 2020). Ferroptosis is an iron‐dependent regulated cell death, which is driven by iron‐induced oxidative stress. Excess iron promotes ROS generated from several sources, including Fenton reactions, mitochondrial ROS production, and membrane‐associated ROS generated by the NADPH oxidase (NOX) family (Ru et al. 2024). This leads to the accumulation of lipid‐associated radicals and lipid peroxides, resulting in cell death (Jiang et al. 2022). As a key process that triggers ferroptosis, lipid peroxidation governed by several enzymes including acyl‐CoA synthetase long chain family member 4 (ACSL4), lysophosphatidylcholine acyltransferase 3 (LPCAT3), arachidonic acid lipoxygenases (ALOXs) and cytochrome P450 oxidoreductase (J. Liu et al. 2022; Mahachoklertwattana et al. 2003). Membrane phospholipids are abundant in PUFAs, which are highly susceptible to oxidative damage. ROS‐mediated oxidation of PUFAs leads to a chain reaction of lipid peroxidation, compromising phospholipid integrity and triggering ferroptosis (Su et al. 2019). On the other hand, oxidative stress disrupts the Xc⁻–GSH–GPX4 antioxidant axis, a critical defense against lipid peroxidation. Suppression of SLC7A11‐mediated cystine uptake reduces intracellular GSH synthesis, leading to impaired GPX4 activity and accumulation of lipid peroxides (Ru et al. 2024). Consistently, iron overload has been shown to decrease GPX4 and SLC7A11 expression in osteoblasts (Jiang et al. 2022). GPX4 is an enzyme that utilizes GSH as a cofactor to reduce toxic lipid peroxides into non‐reactive lipid alcohols. By eliminating intracellular lipid ROS, GPX4 prevents ferroptosis. Therefore, its downregulation results in the onset of ferroptosis and has been used as a ferroptosis marker (Endale et al. 2023). Ferroptosis is also indirectly regulated by another key transcription factor in oxidative response, Nrf2. In physiological conditions, Nrf2 undergoes ubiquitin‐dependent protein degradation mediated by Kelch‐like ECH associated protein 1 (Keap1). Under moderate oxidative stress conditions, a conformational change of Keap1 results in Nrf2 stabilization. Nrf2 can then function as a transcription factor that recognizes the antioxidant response element (Borriello et al. 2016) leading to the expression of antioxidant responsive proteins including GPX4 (Yang et al. 2025). Interestingly, iron overload can impair Nrf2 activity by reducing its DNA‐binding ability, thereby suppressing the transcription of these antioxidant genes and facilitating ferroptosis (G. H. Chen et al. 2022). Consistently, ferroptosis in primary osteoblasts has been associated with dysregulation of the Nrf2/HO‐1 signaling pathway (Xiang et al. 2024). Accordingly, Nrf2 downregulation is another hallmark of ferroptosis as also indicated in this study (Yan et al. 2023).

Previous study suggested that ferroptosis plays a significant role in iron‐induced osteoblast dysfunction and death in pre‐osteoblast MC3T3‐E1 and in vivo (Jiang et al. 2022). In our study, we demonstrated that iron exposure suppressed osteoblast cell viability, but this phenomenon was recovered by ferroptosis inhibitor Fer‐1 in mature osteoblast, UMR‐106 cells. Iron overload also downregulated GPX4 expression in osteoblasts, leading to ROS accumulation, lipid peroxidation and eventually cell death via ferroptosis. These findings aligned with previous research demonstrating that iron overload reduced osteoblast viability and GSH levels while increasing ROS production, lipid peroxidation, and the expression of ferroptosis‐related proteins (Jiang et al. 2022). Additionally, in vivo evidence further supports the role of ferroptosis in iron‐induced bone loss, as iron overload triggered osteoblast ferroptosis and contributed to osteoporosis in iron dextran‐induced osteoporosis mouse model (Jiang et al. 2022). Interestingly, this study further demonstrated that Fer‐1 effectively rescued osteoblast from iron‐induced cell death by reducing lipid peroxidation independently of total ROS level. These findings suggested that ferroptosis played a crucial role in iron‐induced osteoblast cell death. Consistent with our findings, previous studies have shown that Fer‐1 protected against ferroptosis cell death in various cell types by decreasing lipid ROS production in neurons (Li et al. 2017) and chondrocytes (Yao et al. 2021) as well as by inhibiting lipid peroxidation in endothelial cells (Bai et al. 2020) and osteoblasts (Jiang et al. 2022). Previous studies have demonstrated that Fer‐1 (0.5–20 μM) enhances MC3T3‐E1 cell viability over 2–6 days of culture, yet shows no significant difference compared with the control group within the first 72 h of treatment (Jiang et al. 2022; Valanezhad et al. 2021). Together with our results, these studies further support the notion that Fer‐1 at 10 μM does not exert cytotoxic effects on osteoblasts in vitro. Regarding the in vivo study, Fer‐1 is predominantly administered via intraperitoneal injection in rodent models, with most studies reporting no overt toxicity at doses of 1–10 mg/kg over study durations ranging from acute exposure to repeated daily administration for up to several weeks (X. Liu et al. 2023; Y. Liu et al. 2023; Scarpellini et al. 2023; Yu et al. 2026; Zhang et al. 2026).

N‐acetyl‐l‐cysteine is a widely utilized antioxidant that protects against ROS by scavenging oxidants, such as nitric oxide and nitrogen dioxide, and serving as a precursor for GSH synthesis. Then, GSH acts as an antioxidant and serves as a substrate for several antioxidant enzymes (Aldini et al. 2018; Zhitkovich 2019). Several in vitro and in vivo studies have shown that NAC effectively prevents or mitigates oxidative stress under various conditions, such as iron overload (de Andrade et al. 2015; Dodd et al. 2008; Tian et al. 2020; Tian et al. 2016). Previous studies have suggested that the therapeutic efficacy of NAC may be limited in iron‐overload condition because antioxidants alone do not directly target excess iron. Thus, limited benefits of NAC have been reported in hemoglobinopathy‐associated iron overload, including sickle cell disease and thalassemia, and iron‐specific chelators are generally used for treatment (Kontoghiorghes and Kontoghiorghe 2019). Moreover, oral NAC administration aggravated acute iron intoxication and mortality in rats by lowering glutathione‐related enzyme activity but increasing serum iron level and hepatic damage (Abu‐Kishk et al. 2010). Notably, this study found that NAC could not restore osteoblast cell viability under iron overload, suggesting that iron‐induced osteoblast cell death is driven primarily by lipid oxidation rather than by general ROS. However, future experiments on the effect of NAC on lipid peroxidation in osteoblasts under iron overload will be beneficial for confirming this mechanism. In support of this concept, Miotto et al. reported that Fer‐1 effectively inhibited lipid peroxidation by scavenging initial alkoxyl radicals produced in liposomes (Miotto et al. 2020). Likewise, a previous study found that Fer‐1 inhibited lipid peroxidation but did not suppress mitochondrial ROS formation (Skouta et al. 2014). These findings emphasize the critical role of ferroptosis in iron‐induced osteoblast cell death and the potential of Fer‐1 as a targeted therapeutic strategy.

Osteoblast differentiation involves the expression of specific molecular markers contributing to osteoblast maturation and bone mineralization. Among these, ALP and Col1 are widely recognized as key indicators of osteoblast activity. ALP plays a critical role in bone mineralization by regulating inorganic phosphate availability and promoting hydroxyapatite formation for bone matrix mineralization (Vimalraj 2020). Meanwhile, Col1 is a major component of the bone extracellular matrix, providing structural integrity and serving as a scaffold for mineral deposition (Selvaraj et al. 2024). Consistent with previous studies reported that iron overload suppressed osteoblast mineralization and downregulated key osteoblastic genes, including ALP and Col1 (Jiang et al. 2022; Lertsuwan et al. 2020; Zhao et al. 2012), the present study demonstrated that iron overload decreased ALP and Col1a1 expression in osteoblast, as well as inhibited osteoblast mineralization. These results confirmed the detrimental effects of excess iron on osteoblast function and bone formation capacity. Furthermore, iron overload suppresses Wnt signaling through Smad and MAPK pathways. Iron dose‐dependently downregulates Wnt pathway proteins and inhibits canonical Wnt signaling, whereas Wnt agonists or ferroptosis inhibitors can reverse this effect and restore osteoblast differentiation by reducing ROS and lipid peroxidation (Luo et al. 2022). Consistently, elevated ROS levels impair osteoblast differentiation by disrupting multiple osteogenic signaling pathways, including Wnt/β‐catenin, PI3K/AKT, MAPK, and Hedgehog, ultimately leading to reduced osteoblast formation (Yang et al. 2025). Iron overload also modulates signaling pathways involved in cell survival and differentiation. Iron accumulation activates ERK and p38 signaling, increases ferritin expression, and reduces RUNX2 levels in mesenchymal stem cells, thereby decreasing cellular activity and osteogenic differentiation (Yang et al. 2017). While Fer‐1 recovered the expression of ALP but partially rescued the expression of Col1a1, it did not fully restore osteoblast mineralization under iron overload during 6 day‐experiment. Interestingly, a previous study reported that Fer‐1 significantly increased calcium nodule formation in MC3T3‐E1 cells after 30 days of incubation (Valanezhad et al. 2021). These findings suggested that prolonged Fer‐1 exposure may confer beneficial effects on bone mineralization; however, whether the long‐term exposure could recover bone mineralization under iron overload needs to be further investigated. In addition, mineralization is also influenced by the interaction between osteoblasts and Col1. Previous studies showed that collagen accelerated osteoblast and BMSC mineralization. While Vitamin D3 increased ALP and osteocalcin expression, it failed to restore mineralization in the absence of Col1 (Elango et al. 2019; Lynch et al. 1995). Col1 stimulated mineralization via the interaction with Discoidin Domain Receptor‐2 (DDR‐2) though MAPK‐p38 signaling pathway (Elango et al. 2019; Rodríguez‐Carballo et al. 2016). While these mechanisms have been reported to partially affected by ferroptosis, it is not the downstream signaling pathways from lipid peroxidation, which is the direct target of Fer‐1 (Miotto et al. 2020; X. Wang et al. 2023). This corresponds to our results showing the decreased trend of Col1 expression in pre‐osteoblasts exposed to FAC, and its expression was not recovered by Fer‐1 treatment.

Additionally, previous studies reported that FAC caused a dose‐dependent decrease in the calcium content within the ECM under iron‐overload (Balogh et al. 2016; Zhang et al. 2026). Mechanistically, excess iron could interfere with hydroxyapatite formation by interacting with phosphate and calcium ions. Fe³⁺ has a high affinity for phosphate and can form iron‐phosphate complexes, which may hinder hydroxyapatite nucleation and crystal growth (Guggenbuhl et al. 2008). Moreover, iron accumulation has been detected within calcified matrices, suggesting that excess iron may compete with calcium during mineral deposition and lower Ca–phosphate accumulation (Borriello et al. 2016). Consistently, iron deposition in bone is associated with impaired matrix maturation and defective mineralization in β‐thalassemia patients (Mahachoklertwattana et al. 2003). Accordingly, even though Fer‐1 restored ALP expression in pre‐osteoblasts under iron overload, the mineralization process was not recovered, potentially due to the decreased Col1 production and the chemical disturbance of calcium deposition.

Bone remodeling is regulated by the interaction of osteoblasts and osteoclasts through the RANKL/OPG signaling pathway (Kim et al. 2020). Under physiological conditions, osteoblasts regulate osteoclast differentiation and activity by secreting both RANKL and OPG. RANKL binds to its receptor—known as RANK—on osteoclast precursors, thereby stimulating their differentiation into mature multinucleated osteoclasts and promoting bone resorption. On the other hand, OPG functions as a decoy receptor for RANKL, competitively binding to and sequestering RANKL to prevent excessive osteoclast differentiation and bone resorption. Thus, the RANKL/OPG ratio serves as a critical determinant of bone cell activity with an increased RANKL/OPG ratio favors bone loss via increased osteoclastogenesis (Boyce and Xing 2007). This study demonstrated that iron overload elevated the RANKL/OPG ratio in osteoblasts, indicating a shift towards osteoclastogenesis and bone resorption. Conversely, Fer‐1 significantly lowered RANKL/OPG ratio, thereby indicating the reduction in bone resorption under iron overload. Similarly, oxidative stress activates MAPK signaling pathways, including p38 and JNK, which are known to enhance RANKL expression and osteoclast differentiation (Danks et al. 2016). On the other hand, Fer‐1 scavenges lipid radicals and suppresses lipid peroxidation, thereby inhibiting activation of the p38/JNK/MAPK pathway and reducing osteoclastogenic signaling (Xu et al. 2025). By inhibiting of these pathways, Fer‐1 may reduce RANKL expression and thereby lower the RANKL/OPG ratio under iron‐overload conditions.

In addition to the key regulators mentioned above, pro‐inflammatory cytokines secreted by osteoblast, IL‐1β and IL‐6 have been shown to promote osteoclast formation (Zhou et al. 2022). These cytokines upregulate RANKL expression and promote osteoclastogenesis in a synergistic fashion (Kwan Tat et al. 2004; Ruscitti et al. 2015). Moreover, previous studies indicated that IL‐1β also inhibited osteoblast differentiation, leading to a reduction in bone formation (Ruscitti et al. 2015). Our results showed that iron overload did not significantly increase the expression levels of IL‐1β and IL‐6. However, Fer‐1 significantly reduced IL‐1β expression in UMR‐106 cells suggesting the potential of Fer‐1 as an anti‐inflammatory agent in osteoblasts under iron overload. Ferroptosis and inflammatory gene expression were intricately linked through the activation of key inflammatory signaling pathways such as JAK‐STAT, NF‐κB, inflammasome, cGAS‐STING, and MAPK (Chen et al. 2023). Disruptions in redox homeostasis and elevated lipid peroxidation stimulated various inflammatory responses. In turn, the release of proinflammatory cytokines exacerbated intracellular oxidative stress and lipid peroxidation (Chen et al. 2021). Experimental evidence demonstrated that ferroptosis induced the release of proinflammatory cytokines in smooth muscle cells via NF‐κB activation (Y. Chen et al. 2022), whereas ferroptosis inhibitors exert anti‐inflammatory effects by reducing the expression of inflammatory cytokines, including IL‐1β, TNF‐α, and ICAM‐1, in contusion spinal cord injuries (Zhang et al. 2019). Thus, the downregulation of osteoinflammatory gene expression could be another pathway through which ferroptosis inhibition contributes to the reduction of inflammation‐induced bone resorption under iron‐overload conditions.

In addition to the indirect effects of iron overload on osteoclastogenesis via the manipulation of osteoblast gene expression, the direct effects of iron overload on osteoclast cells were elucidated in this study. Osteoclasts are specialized bone‐resorbing cells derived from the monocyte/macrophage hematopoietic lineage. Increased osteoclast viability and activity enhance bone matrix degradation, leading to pathological bone resorption and the development of osteoporosis (Boyle et al. 2003). ROS serve as critical regulators of osteoclast biology, exhibiting a dual function in osteoclast regulation. Under physiological conditions, ROS functions as essential signaling molecules that facilitate osteoclast differentiation and activation. However, excessive ROS generation from oxidative stress has been shown to enhance osteoclast survival and hyperactivity, thereby exacerbating bone resorption and contributing to osteoporosis (Agidigbi and Kim 2019). This study demonstrated that iron overload significantly increased ROS production and lipid peroxidation in osteoclasts while also enhancing osteoclast viability. These findings are consistent with previous research showing that iron promotes osteoclastogenesis and bone resorption by increasing ROS production and upregulating osteoclast‐specific genes (Jia et al. 2012; Wang et al. 2018). Notably, Fer‐1 did not affect osteoclast viability under iron overload, which aligns with its inability to modulate ROS levels. This data indicated that iron overload‐induced osteoclast viability is independent from ferroptosis. The differential effects of FAC on GPX4 expression between RAW264.7 cells and osteoblasts are likely attributable to the distinct biological functions and redox regulatory mechanisms of osteoclast‐lineage cells and osteoblasts under iron overload (Balogh et al. 2018; Jeney 2017; Tsay et al. 2010; Zhen Wang et al. 2022). RAW264.7 cells belong to the monocyte/macrophage lineage and serve as osteoclast precursors. Macrophages possess efficient iron‐handling mechanisms, including ferritin‐mediated iron storage and ferroportin‐dependent export, which help buffer intracellular iron and maintain redox homeostasis against oxidative damage induced by iron overload (Gan et al. 2017). Moreover, Previous studies have indicated that the Nrf2‐mediated antioxidant response is actively regulated in osteoclasts both in vitro and in vivo (L. Wang et al. 2023). Under iron overload, the activation of Nrf2 has been reported to facilitate osteoclast differentiation by enhancing antioxidant capacity and reducing intracellular iron levels (Zhang et al. 2023). Because the osteoclast linage possesses robust defense mechanisms against iron‐induced oxidative stress, such as rapid ferritin induction and activation of the Nrf2‐GPX4 antioxidant pathway, exposure to FAC may not sufficiently overwhelm their antioxidant capacity to induce the GPX4 downregulation typically seen during ferroptosis. In contrast, osteoblasts lack these specialized iron‐buffering systems, rendering them highly susceptible to iron‐overload‐induced lipid peroxidation and subsequent GPX4 depletion. However, Fer‐1 still attenuated iron overload‐induced lipid peroxidation and enhanced GPX4 protein expression, an anti‐ferroptosis protein in osteoclasts. These results confirmed the ability of Fer‐1 to modulate ferroptosis in osteoclasts, and suggest that removing basal lipid ROS may trigger a compensatory or protective upregulation of antioxidant enzymes in these specialized cells. Previous studies have demonstrated that ferroptosis plays a role in RANKL‐induced osteoclast differentiation in human primary macrophages. Notably, Fer‐1 treatment effectively mitigated this effect (Ni et al. 2021). Furthermore, a recent study by Xu et al. also reported that Fer‐1 suppressed osteoclast differentiation from mouse bone marrow macrophages by reducing lipid peroxidation levels regulated by ACSL4, which is mediated through the p38/JNK/MAPK signaling pathway (Xu et al. 2025). These findings are consistent with our results, which demonstrated that iron overload significantly enhanced osteoclast differentiation. Importantly, our findings demonstrate that Fer‐1 markedly attenuates the FAC‐induced osteoclast differentiation. The observed effects of Fer‐1 suggested its potential role in mitigating iron overload‐induced bone loss. These findings suggest that osteoclasts maintain viability under iron‐overload conditions through adaptive redox systems, allowing ROS to function as differentiation signals rather than inducing cell death. In contrast, Fer‐1 reduces lipid peroxidation–derived ROS, thereby disrupting ROS‐dependent signaling and suppressing osteoclast differentiation without affecting cell viability.

Besides the use of synthetic ferroptosis inhibitor Fer‐1, cystine plays a critical role in maintaining cellular redox homeostasis, and its depletion has been shown to induce ferroptotic cell death (Kang et al. 2021; Shi et al. 2021). Cystine starvation triggered ferroptosis in tumor cells by depleting intracellular GSH and downregulating GPX4 protein levels through mTORC1 signaling (Zhang et al. 2021). In vivo studies also demonstrated that cystine depletion induced ferroptosis in pancreatic tumors in mice (Badgley et al. 2020). Remarkably, cystine has been reported to inhibit ferroptosis through a GPX4‐dependent mechanism rather than solely by contributing to GSH synthesis (Xia et al. 2024). Cystine uptake via the cystine/glutamate antiporter System Xc⁻ is essential for maintaining cellular redox homeostasis and survival in bone cells. After uptake, cystine is reduced to cysteine by TXNRD1 and subsequently utilized for GSH synthesis (Xiao et al. 2025). Previous studies demonstrated that cystine deprivation induces cell death, whereas cells capable of endogenous cysteine synthesis are more resistant to ferroptosis (Hayano et al. 2016). Consistently, erastin, a classical ferroptosis inducer, inhibits System Xc⁻, leading to reduced cystine uptake, GSH depletion, and ferroptotic cell death (Jiang et al. 2021). Together, these findings highlight the critical role of cystine in protecting bone cells against ferroptosis‐associated oxidative damage. Accumulating evidence further supports the biological relevance of cystine in skeletal tissues, as cystine has been reported to promote the proliferation of primary mouse chondrocytes (Sasama et al. 2024). Moreover, cystine‐rich peptides isolated from rat bone marrow was shown to play an essential role for maintaining protein structural stability and regulating biological activity (Belcourt et al. 1992).

In the present study, we found that cystine rescued osteoblast cell viability while suppressing osteoclast cell viability under iron overload. Based on previous evidence in osteoclast precursors, the suppression of osteoclast viability by cystine was likely associated with the excessive accumulation of intracellular cystine. Due to its extremely low solubility, imported cystine tended to accumulate within cells, resulting in increased disulfide stress. This condition potentially promoted the formation of aberrant disulfide bonds, particularly within actin cytoskeletal proteins, thereby leading to cytoskeletal dysfunction and cell death (Zhong et al. 2023). Furthermore, 200 µM cystine has been shown to activate mTORC1 signaling, which promotes GPX4 protein synthesis via the Rag–mTORC1–4EBP axis, thereby suppressing ferroptotic cell death. Conversely, cystine deprivation (0–5 µM) markedly reduces GPX4 protein levels and sensitizes cells to ferroptosis (Zhang et al. 2021). Interestingly, co‐treatment with cystine and ferroptosis inhibitor Fer‐1 did not further enhance their effects on osteoblast viability, suggesting that both agents may exert their cytoprotective effects through overlapping or converging antioxidant pathways (Scarpellini et al. 2023). Alternatively, it is possible that the activation of a single pathway is sufficient to counteract iron overload‐induced oxidative stress and preserve osteoblast survival. This dual effect of cystine suggested a potential therapeutic application in mitigating iron overload‐induced bone loss from its recovery effects of osteoblast cell survival and its negative effect on osteoclast viability under iron‐overload conditions. Nevertheless, future studies are required to explore the molecular pathways and potential applications in preventing bone loss associated with iron overload. There are some limitations in our research. Although the immortalized cell lines used in this study are the well‐established and reproducible models for mechanistic studies as shown in numerous studies (Czekanska et al. 2014; Hwang and Horton 2019; Jiang et al. 2022; Quarles et al. 1992), they cannot fully recapitulate the biological characteristics of primary bone cells or the complexity of the in vivo microenvironment. This study has provided robust evidence for FAC‐induced ferroptosis in osteoblasts through multiple approaches including (1) direct quantification of lipid peroxidation, (2) measurement of both GSH depletion and GPX4 downregulation reflecting the impairment of Xc⁻–GSH–GPX4 axis, (3) reduced Nrf2 expression reflecting disrupted antioxidant defense mechanism, and (4) functional rescue with fer‐1 indicating that iron‐induced osteoblast cell death, at least partly, relying on ferroptosis, further studies on additional markers including ACLS4 and SLC7A11 could further strengthen the mechanistic characterization of ferroptosis. Further cross‐validation through genetic interventions, such as overexpression or knockdown of GPX4 and SLC7A11, or by using other types of ferroptosis inhibitors could also be beneficial for mechanistic validation. However, our findings support the involvement of ferroptosis in iron overload‐induced osteoblast cell death, although this does not exclude the possible contribution of other regulated cell death pathways, including apoptosis. Future studies will therefore focus on validating our findings using primary osteoblasts and osteoclasts, as well as conducting in vivo animal experiments to further confirm the role of ferroptosis in iron overload‐induced bone remodeling impairment under physiological conditions.

5. Conclusion

Our study highlighted how iron overload disrupted bone homeostasis by inducing ferroptosis in osteoblasts and promoting osteoclast differentiation. Treatment with the ferroptosis inhibitor Fer‐1 effectively mitigated lipid peroxidation, enhanced osteoblast viability, and suppressed osteoclastogenesis. Furthermore, our results show that Fer‐1 markedly suppresses osteoclast differentiation under iron‐overload conditions without affecting osteoclast viability, suggesting that ferroptosis inhibition regulates osteoclastogenic signaling rather than pathways governing cell survival. Our findings also suggest potential therapeutic implications of targeting ferroptosis pathways in iron overload‐induced bone disorders (Figure 8). However, the in vitro models used in the present study did not fully replicate the complexity of in vivo conditions, where intricate cell‐cell interactions and systemic factors may influence bone cell viability and activity. Further studies are thus required to validate these findings in animal models and to explore the therapeutic applications of ferroptosis inhibitors in bone diseases associated with iron overload.

Figure 8.

A schematic diagram summarizing the major findings of this study, illustrating the proposed mechanisms, and highlighting the potential therapeutic implications of targeting ferroptosis pathways in iron overload‐induced bone disorders.

Author Contributions

Supagarn Sooksawanwit: conceptualization, experimental design, data curation, formal analysis, methodology, data criticism and conclusion, visualization, writing – original draft, writing – review and editing. Kornkamon Lertsuwan: research design, conceptualization, experimental design, data curation, formal analysis, funding acquisition, methodology, project administration, resources, supervision, data criticism and conclusion, writing – original draft, writing – review and editing, corresponding author. Natnicha Tannop, Ploypapus Lapatkeattisukul, Thanayuth Jenpichitkulchai, Tueanjai Khunluck, Siriwan Punvilai: data curation, formal analysis, methodology, writing – original draft. Nittiyaporn Sotapong: data curation, formal analysis, methodology. Jarinthorn Teerapornpuntakit: data curation, formal analysis, visualization, writing – review and editing. Narattaphol Charoenphandhu: conceptualization, experimental design/research design, formal analysis/data analysis, data criticism, discussion and conclusion, funding acquisition, resources, supervision, writing – review and editing.

Ethics Statement

This study did not involve experiments with human participants or animals. All experiments were performed using cell lines, that is, rat osteoblast‐like osteosarcoma cell line UMR‐106 (ATCC, VA, USA), mouse macrophage‐like cell line RAW 264.7 (ATCC), and pre‐osteoblastic MC3T3‐E1 cell (ATCC).

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.