Transgenic mice with a C57BL/6 background and carrying the heterozygous Fgfr3^𝑎𝑐ℎ transgene were obtained from Shanghai Model Organisms (China). They were housed in SPF conditions with free access to water and food. The mean bodyweight of mice at 8 weeks of age was approximately 20 g. Mouse genotypes were ascertained by qPCR. Wild-type C57BL/6 mice were used as controls. For all procedures involving tissue harvesting, mice were first anesthetized using isoflurane (2% for induction, 1.5% for maintenance) delivered via a precision vaporizer (2V22105103, RWD Life Science Co., Ltd., Shenzhen, Guangdong, China), ensuring a consistent and humane depth of anesthesia. Following anesthesia, mice were euthanized by cervical dislocation to minimize suffering and ensure ethical compliance. For treatment, lentivirus with METTL16 overexpression vectors (pCDH-CMV vectors; GenePharma, Shanghai, China) was intraperitoneally injected into newborn mice for 20 days. The tibiae from mice were collected and processed into paraffin-embedded samples or stored in liquid nitrogen for subsequent experiments. All animal procedures were approved by the Institutional Animal Care and Use Committee (IACUC) of Animal Care and Use Committee of Pediatric Orthopaedic Hospital, Honghui Hospital, Xi’an Jiaotong University (No. KT2023-01-05-01) and conducted in accordance with the relevant guidelines and regulations.

Chondrocytes were prepared from the tibia of newborn mice [17] and cultured in DMEM medium with 10% fetal bovine serum (FBS) in a 37 °C incubator. Cell transfection was conducted using Lipofectamine 2000 (11668027, Invitrogen™, Carlsbad, CA, USA) for 48 h. Chondrocytes were stimulated with IL-1$\beta$ (10 µg/mL) for 6 h to induce cell death and apoptosis.

All primary chondrocytes were freshly isolated and used at early passage. STR authentication is not applicable to primary cells.

Primary cells were observed daily under a phase-contrast microscope. The cells exhibited typical morphology consistent with human nucleus pulposus/endplate chondrocyte characteristics, including a polygonal or round shape, relatively large nuclei, and a pericellular lacuna-like structure. No fibroblast-like overgrowth or abnormal morphological changes were observed within passages P2–P4, which were used for subsequent experiments.

The micro-architecture parameters of the tibia from mice were analyzed by assessing the bone mineral density (BMD), bone volume to total tissue volume (BV/TV), trabecular number (TbN), and trabecular thickness (TbTh).

For histological examination, tibial tissues were preserved in 10% formalin, demineralized using 0.5 M EDTA, embedded in paraffin, and then cut into 5 µm sections. The histomorphology was observed by staining with hematoxylin and eosin (H&E; C0105S, Beyotime, Shanghai, China). The expression of METTL16 was assessed by immunohistochemical (IHC) staining using anti-METTL16 antibody (17676, Cell Signaling Technology, Danvers, MA, USA, 1:1000) and visualized by DAB (Beyotime, China) as the substrate.

The samples were stained with toluidine blue (89640, Sigma, St. Louis, MO, USA). This stains cartilage and osteoblasts a dark blue color against a light blue background.

Bone samples were stained with safranin O solution, Weigert hematoxylin solution, and solid green staining solution (Sigma, USA). This results in chondrocyte cytoplasm presenting as red, and chondrocyte nuclei presenting as gray black.

Cells induced by IL-1$\beta$ and transfected with METTL16 overexpression vectors were seeded into 96-well plates and cultured for 24 h. At the final time point, CCK-8 reagent (C0005, TargetMol, Shanghai, China) was added to each well and reacted for 2 h. The absorbance values at 450 nm were then measured, and the relative cell viability was calculated.

Chondrocytes were digested into a single cell suspension and seeded into 6-well plates at a density of 10,000 cells per well. After incubation for 10 days, the colonies were stained with crystal violet (Sigma, USA) for 20 min, and the images were recorded with a camera.

The concentration of malondialdehyde (MDA), which is the final product of lipid peroxidation, was determined using a lipid peroxidation assay kit (Doojindo, Kumamoto, Japan). Additionally, total iron and Fe²⁺ levels were evaluated with an iron assay kit (ab83366, Abcam, Woburn, MA, USA), as per the instructions provided by the manufacturer.

Cell lysates were obtained using RIPA buffer (89900, Invitrogen™, Carlsbad, CA, USA). Following loading and separation on an SDS-PAGE gel, the proteins were transferred to PVDF membranes and blocked with 5% skim milk. The membranes were then incubated overnight at 4 °C with primary antibodies against METTL16 and GPX4 (1:1000, ab252420; ab125066, Abcam, USA). They were subsequently treated with HRP-conjugated anti-rabbit secondary antibodies (1:1000, ab6721, Abcam, USA), and protein bands were visualized using an ECL chemiluminescence substrate (WBKLS0500-2, Millipore, Burlington, MA, USA).

Tissues and chondrocytes were treated with Trizol reagent to isolate total RNA. This was converted into cDNA using the PrimeScript RT reagent kit (RR037A, Takara, Kyoto, Japan). Real-time qPCR was performed with the SYBR Green qPCR Master Mix (Thermo, Waltham, MA, USA). Gene expression levels were measured relative to the $\beta$-actin gene as an endogenous control.

RNAs were isolated from cells and incubated with protein G beads (70024, CST, Danvers, MA, USA) that had previously been reacted with anti-m⁶A monoclonal antibody at 4 °C for 12 h in rotation. After reaction at 4 °C for 6 h, the beads were eluted and washed, and RNA was extracted using Trizol. The level of METTL16 was measured by qPCR assay.

Statistical analyses were conducted using GraphPad Prism 7.0 software (GraphPad Software, Inc,San Diego, CA, USA). The paired Student’s t-test was utilized to compare two groups. For analyzing differences among multiple groups, either a one-way Analysis of Variance with Tukey’s post hoc test or a Mann–Whitney U test was employed. The criterion for statistical significance was defined as p 0.05.

We first examined the in vivo effects of METTL16 on ACH using transgenic mice. The expression of METTL16 was suppressed in the proximal tibia of ACH mice compared with the control, but was recovered by treatment for METTL16 overexpression (Fig. 1A). Further analysis of the proximal tibia revealed an increased number of chondrocytes and cartilage, as well as an organized tissue structure after METTL16 overexpression compared with ACH mice (Fig. 1A). Treatment with METTL16 overexpression vectors also increased the BMD, BV/TV, TbN and TbTh compared to ACH (Fig. 1B). Notably, the expression of METTL16 and GPX4 were suppressed in the tibia tissues of ACH mice, while METTL16 overexpression recovered the levels of both (Fig. 1C). This observation suggests a potential correlation between the two genes and participation in ferroptosis during ACH. We next investigated the level of ferroptosis in the tibia of mice. The elevated levels of total iron and Fe²⁺ in the ACH group compared with the control group indicated an increase in cell ferroptosis. This was suppressed upon METTL16 overexpression (Fig. 1D). Perls’ Prussian blue staining demonstrated minimal iron deposition in control tibiae, whereas ACH mice exhibited prominent deposits concentrated in the growth plate and trabecular regions. Quantitative image analysis confirmed a significantly larger Perls’-positive area in ACH mice compared with controls, which was significantly reduced by METTL16 overexpression (Fig. 1E). In line with this, IHC for 4-Hydroxynonenal (4-HNE) showed markedly elevated lipid peroxidation in ACH tibiae, with intense cytoplasmic and extracellular 4-HNE staining. Overexpression of METTL16 substantially decreased this 4-HNE positivity (Fig. 1F). These histological findings, together with increased tissue iron and Fe²⁺ levels and decreased GPX4 expression in ACH, indicate enhanced ferroptosis in ACH that can be alleviated by restoring METTL16 expression. To further delineate the therapeutic effects of METTL16 on bone microstructure and cartilage morphology, we performed micro-computed tomography (CT) reconstruction of the proximal tibiae (Fig. 1G). CT analysis revealed that ACH mice exhibited severe trabecular bone deterioration, characterized by fragmented, disconnected trabeculae with reduced spatial continuity. However, METTL16 overexpression substantially restored the integrity of the trabecular network.

Fig. 1.

METTL16 promotes bone chondrogenesis and suppresses ferroptosis in the ACH mouse model. Fgfr3^𝑎𝑐ℎ transgenic mice were treated with METTL16 lentivirus, and the proximal tibia tissues were collected for analysis. (A) The morphology of cartilage, chondrocyte, and tissues was examined by toluidine blue, safranin O, and H&E staining, while the expression of METTL16 was examined by IHC. Scale bar = 100 µm. (B) Results for bone mineral density (BMD), bone structure via the bone volume to total tissue volume ratio (BV/TV), trabecular number (TbN), and trabecular thickness (TbTh). (C) RNA levels of METTL16 and GPX4, as measured by qPCR assay. (D) The levels of total iron and Fe²⁺ were measured to assess ferroptosis. (E) Representative Perls’ Prussian blue staining of tibial sections showing iron deposition (blue). Quantitative analysis of Perls-positive area. Scale bar = 100 µm. (F) Representative IHC staining for 4-HNE showing the levels of lipid peroxidation in tibiae. Quantitative analysis of 4-HNE–positive signal intensity. Scale bar = 100 µm. (G) CT reconstruction for observing the microstructure of bone cartilage in ACH mice. Scale bar = 100 µm. **p 0.01, *p 0.05. ACH, Achondroplasia; H&E, hematoxylin and eosin; IHC, immunohistochemical; GPX4, glutathione peroxidase 4; qPCR, quantitative real-time PCR; 4-HNE, 4-Hydroxynonenal; CT, computed tomography.

To further examine whether ferroptosis is a major form of regulated cell death in chondrocytes and to confirm the protective role of METTL16, we employed specific inhibitors of ferroptosis (Ferrostatin-1) and apoptosis (Z-VAD-FMK) in a model of IL-1$\beta$-induced chondrocyte injury. The experimental groups comprised the control, IL-1$\beta$, IL-1$\beta$ + METTL16 overexpression (OE), IL-1$\beta$ + Ferrostatin-1, and IL-1$\beta$ + Z-VAD-FMK. Cell viability, as determined by CCK-8 assay, was markedly reduced after IL-1$\beta$ stimulation compared to the control group, indicating severe cellular injury. Both METTL16 OE and Ferrostatin-1 treatment significantly restored cell viability, whereas Z-VAD-FMK exerted only a minor protective effect (Fig. 2A). Lipid peroxidation, assessed using C11-BODIPY fluorescence, was significantly higher in IL-1$\beta$-treated chondrocytes, confirming enhanced oxidative stress. The elevated C11-BODIPY oxidation ratio was markedly decreased by METTL16 OE and Ferrostatin-1, but not by Z-VAD-FMK (Fig. 2B), suggesting a ferroptosis-dominant phenotype. In accordance with this, the GSH level was significantly reduced, and GPX4 protein expression was downregulated in the IL-1$\beta$ group. Both METTL16 OE and Ferrostatin-1 markedly restored intracellular GSH and GPX4 levels, whereas Z-VAD-FMK showed negligible effects (Fig. 2C,D). In addition, Annexin V/PI staining demonstrated that IL-1$\beta$ treatment induced a significant increase in Annexin V⁺/PI⁺ cells, indicating cell death. METTL16 OE and Ferrostatin-1 significantly reduced the proportion of Annexin V⁺/PI⁺ cells. LDH release, which is an indicator of membrane damage, was markedly elevated in IL-1$\beta$-treated cells. The increase in LDH was significantly attenuated by METTL16 OE and Ferrostatin-1, but remained largely unchanged with Z-VAD-FMK (Fig. 2E,F). Together, these findings demonstrate that IL-1$\beta$ induces ferroptosis-like cell death in chondrocytes, characterized by lipid peroxidation, GSH depletion, and GPX4 downregulation. Both METTL16 overexpression and the ferroptosis inhibitor Ferrostatin-1 effectively alleviated these effects, confirming that METTL16 protects chondrocytes primarily by inhibiting ferroptosis rather than apoptosis. To further clarify the specificity of ferroptosis in IL-1$\beta$-induced chondrocyte death, we evaluated the contribution of autophagy using the pharmacological inhibitor 3-methyladenine (3-MA, 5 mM). IL-1$\beta$ stimulation was found to induce autophagy, as evidenced by increased LC3B-II and decreased p62 levels. However, pretreatment with 3-MA effectively blocked LC3B-II accumulation but failed to restore cell viability (CCK-8 assay) or reduce ferroptosis markers (Fe²⁺, MDA, GSH) (Supplementary Fig. 1A–C). METTL16 overexpression retained its protective effects in the presence of 3-MA. These data indicate that although autophagy is activated in IL-1$\beta$-stimulated chondrocytes, it does not represent the major form of regulated cell death, supporting ferroptosis as the dominant pathway.

Fig. 2.

METTL16 protects chondrocytes from IL-1β-induced injury primarily by inhibiting ferroptosis. (A) Cell viability assessed by CCK-8 assay in control, IL-1$\beta$, IL-1$\beta$ + METTL16 OE, IL-1$\beta$ + Ferrostatin-1, and IL-1$\beta$ + Z-VAD-FMK groups. (B) Lipid peroxidation detected by C11-BODIPY fluorescence imaging and quantitative analysis. Scale bar = 50 µm. (C) Intracellular glutathione (GSH) levels in each group. (D) Western blot analysis and quantification of GPX4 protein expression. (E) Annexin V/PI staining showing the percentage of apoptotic and necrotic cells. (F) Lactate dehydrogenase (LDH) release indicates cell membrane integrity. Data are presented as the mean $\pm$ SD (n = 3). ***p 0.001, **p 0.01, *p 0.05. CCK-8, cell counting kit 8; IL-1$\beta$, interleukin-1 beta.

Subsequently, we investigated the effects of METTL16 on chondrocytes. Cells from primary culture were treated with IL-1$\beta$ to reduce viability and induce ferroptosis. As shown in Fig. 3A, treatment with IL-1$\beta$ decreased the RNA and protein levels of METTL16 and GPX4, respectively, consistent with results from the in vivo experiments. Results from the CCK-8 and colony formation assays showed that overexpression of METTL16 could rescue the cell viability and proliferation of IL-1$\beta$-induced primary chondrocytes (Fig. 3B,C). Moreover, the status of several ferroptosis biomarkers, including elevated iron, Fe²⁺, and MDA, and decreased GSH level, was significantly reversed by METTL16 overexpression (Fig. 3D). We next examined key protein markers to further investigate the mechanism by which METTL16 regulates ferroptosis. Western blot analysis revealed that IL-1$\beta$ stimulation significantly upregulated the pro-ferroptotic proteins ACSL4 and PTGS2, while downregulating the iron storage protein ferritin. Notably, METTL16 overexpression effectively reversed these aberrant protein expressions (Fig. 3E). Furthermore, overexpression of METTL16 significantly inhibited the reactive oxygen burst induced by IL-1$\beta$ (Fig. 3F) and restored mitochondrial ultrastructure (Fig. 3G). Transmission electron microscopy revealed that IL-1$\beta$ treatment caused marked mitochondrial swelling, loss or fragmentation of cristae, and increased electron density, all of which are characteristic features of ferroptosis. Quantitative analysis showed a significant increase in the proportion of swollen mitochondria and cristae-deficient mitochondria in the IL-1$\beta$ group, which was markedly reduced upon METTL16 overexpression.

Fig. 3.

METTL16 enhances chondrocyte proliferation and suppresses cell ferroptosis. Primary chondrocytes were stimulated with IL-1$\beta$ and transfected with METTL16 overexpression vectors. (A) The RNA and protein levels of METTL16 and GPX4 were measured by qPCR and Western blot assays. (B) Cell viability was measured by the CCK-8 assay. (C) Cell proliferation was detected by colony formation. Scale bar = 100 µm. (D) The levels of iron, Fe²⁺, MDA, and GSH were measured to assess ferroptosis. (E) Protein levels of ACSL4, Ferritin, and PTGS2, as measured by Western blot assay. (F) TEM analysis of mitochondrial ultrastructure in IL-1$\beta$-stimulated chondrocytes. Scale bar = 200 µm. (G) ROS levels as detected by the reagent kit. ***p 0.001, **p 0.01, *p 0.05. MDA, malondialdehyde; ACSL4, Acyl-CoA synthetase long-chain family member 4; PTGS2, prostaglandin-endoperoxide synthase 2; TEM, transmission electron microscopy; ROS, reactive oxygen species.

To investigate how METTL16 influences GPX4 expression, we assessed the m⁶A methylation levels of GPX4 mRNA. The reduction of METTL16 levels using siRNAs was found to markedly decrease m⁶A modification of GPX4 mRNA (Fig. 4A). Moreover, depletion of METTL16 decreased the protein level of GPX4 (Fig. 4B). We also utilized CHX to block protein synthesis. The protein level of GPX4 was notably decreased 6 h after CHX treatment, and depletion of METTL16 induced faster degradation of GPX4 (Fig. 4C). These results indicate that knockdown of METTL16 can reduce the stability of GPX4 mRNA.

Fig. 4.

METTL16 regulates m⁶A modification and degradation of GPX4 mRNA. (A) The level of m⁶A modification of GPX4 mRNA was measured by the m⁶A mRNA IP assay. (B) The protein level of GPX4 in chondrocytes was measured by Western blotting. (C) Chondrocytes were first transfected with siMETTL16-1 or siMETTL16-2 and then treated with CHX, an inhibitor of protein synthesis. The protein level of GPX4 at 0, 3, 6, and 12 h after CHX treatment was assessed by Western blot assay. **p 0.01, *p 0.05.

To further validate the involvement of ferroptosis in ACH pathology and clarify the therapeutic potential of METTL16, ACH mice were treated with METTL16 overexpression (OE) vectors, the ferroptosis inhibitor Ferrostatin-1 (Fer-1, 1 mg/kg i.p.), or Caspase inhibitor/Necrostatin-1 (Casp/Nec-1). Normal mice served as controls. Proximal tibiae and vertebrae were analyzed histologically and biochemically to assess ferroptosis and inflammation. H&E staining revealed that ACH mice exhibited a disorganized growth plate, reduced cartilage thickness, and decreased chondrocyte density compared with control animals. These pathological alterations were markedly attenuated by METTL16 OE and Fer-1 treatment, whereas Casp/Nec-1 intervention resulted in minimal improvement. Perls’ Prussian blue staining revealed extensive iron accumulation in the trabecular and growth plate regions of ACH tibiae, which was significantly reduced in METTL16 OE- and Fer-1-treated mice (Fig. 5A). IHC for 4-HNE showed pronounced lipid peroxidation in ACH mice, reflected by intense cytoplasmic and extracellular staining, which was substantially alleviated by METTL16 OE and Fer-1 (Fig. 5B). Biochemical assays further supported these observations, with MDA and Fe²⁺/total iron levels being significantly elevated, while GSH content was reduced in ACH mice. Both METTL16 OE and Fer-1 normalized these parameters, whereas Casp had little effect (Fig. 5C). Western blot analysis revealed marked downregulation of GPX4 and SLC7A11, and upregulation of the pro-ferroptotic proteins ACSL4 and transferrin receptor (TfR1) in ACH tissues. METTL16 OE and Fer-1 effectively restored GPX4, SLC7A11 and ferritin (FTH1) expression, while reducing ACSL4 and TfR1 levels (Fig. 5D). Furthermore, serum and tissue cytokine analyses showed that TNF-$\alpha$ and IL-6 levels were markedly increased in ACH mice, while METTL16 OE and Fer-1 significantly reduced these inflammatory markers (Fig. 5E). Collectively, these data demonstrate that METTL16 overexpression and ferroptosis inhibition (Fer-1) effectively alleviate iron accumulation, lipid peroxidation, mitochondrial damage, and inflammation in ACH, whereas inhibition of apoptosis or necroptosis (Casp/Nec-1) exerts only limited effects, confirming that ferroptosis is the dominant form of regulated cell death in ACH.

Fig. 5.

METTL16 overexpression and ferroptosis inhibition alleviate ferroptosis and cartilage degeneration in ACH mice. (A) Perls’ Prussian blue staining of proximal tibia sections showing iron accumulation in the trabecular and growth plate regions of ACH mice. Iron deposition was significantly reduced in groups with METTL16 overexpression (OE) and Ferrostatin-1 (Fer-1) treatment, whereas Caspase inhibitor (Casp) and Necrostatin-1 (Nec-1) treatment had minimal effects. Scale bar = 100 µm. (B) Immunohistochemistry for 4-HNE showed lipid peroxidation in ACH tibiae. Intense cytoplasmic and extracellular 4-HNE staining was observed in ACH mice, which was substantially reduced in METTL16 OE and Fer-1-treated groups. Scale bar = 100 µm. (C) Biochemical assays were used to measure MDA (malondialdehyde) and the Fe²⁺/total iron ratio, as well as GSH (glutathione) content. Both METTL16 OE and Fer-1 restored MDA, Fe²⁺/total iron levels, and GSH content to near normal levels, whereas Casp/Nec-1 treatment had little effect. (D) Western blot analysis of ferroptosis-related proteins in ACH tibiae. GPX4, SLC7A11, and ferritin (FTH1) were downregulated in ACH mice and restored by METTL16 OE and Fer-1, whereas ACSL4 and transferrin receptor (TfR1) were upregulated in ACH mice and reduced by METTL16 OE and Fer-1 treatment. (E) Serum and tissue inflammatory cytokine levels (TNF-$\alpha$ and IL-6) were measured by ELISA. METTL16 OE and Fer-1 significantly reduced the elevated levels of TNF-$\alpha$ and IL-6 in ACH mice, while Casp/Nec-1 had minimal effects. ***p 0.001, **p 0.01.

We further investigated the regulatory role of METTL16/GPX4 in ACH development. The results from H&E staining and bone structure parameters indicated that suppression of GPX4 abolished METTL16-promoted cartilage growth and bone formation (Fig. 6A,B). The results from in vitro experiments also revealed that GPX4 knockdown reversed the METTL-16-elevated viability and proliferation of chondrocytes (Fig. 6C,D). The decreased level of GPX4 protein in chondrocytes following siGPX4 transfection was verified by Western blotting (Fig. 6E).

Fig. 6.

METTL16 improves chondrogenesis through the regulation of GPX4 expression. (A) H&E staining of proximal tibia tissue. Scale bar = 200 µm. (B) Results for bone mineral density (BMD), bone structure as analyzed by measuring the bone volume to total tissue volume ratio (BV/TV), trabecular number (TbN), and trabecular thickness (TbTh). (C–E) Primary chondrocytes were stimulated with IL-1$\beta$ and transfected with METTL16 overexpression vectors and siGPX4. (C) Cell viability was measured by the CCK-8 assay. (D) Cell proliferation was detected by colony formation. Scale bar = 200 µm. The GPX4 protein level was detected by Western blotting. (E) GPX4 protein level as assessed by Western blotting. **p 0.01, *p 0.05.