Zoledronic acid promotes osteoclasts ferroptosis by inhibiting FBXO9-mediated p53 ubiquitination and degradation

Cell culture

RAW264.7 cells were purchased from the ATCC (TIB-71; Manassas, VA, USA) and cultured in alpha-Modified Eagle’s Medium (α-MEM; Gibco, Jenks, OK, USA) with 100 U/ml penicillin, and 100 μg/ml streptomycin at 37 °C in a humidified atmosphere containing 5% CO₂ and 95% air. To BMDMs, bone marrow cells were purchased from the ATCC (CRL-2420; Manassas, VA, USA) and cultured in RPMI 1640 medium (Thermo Fisher Scientific, Waltham, MA, USA) for 6 days with 100 U/ml penicillin, 100 μg/ml streptomycin, 10% FBS, and 10 ng/mL recombinant mouse macrophage colony-stimulating factor (PeproTech, Cranbury, NJ, USA). For osteoclast formation assay.RAW264.7 cells were seeded in 12-well plate s (1 × 10⁴ cells/well) supplemented with 50 ng/ml RANKL (R&D Systems) for 6 days. BMDMs (1 × 10⁴ cells/well) were cultured in the presence of M-CSF (10 ng/mL) and RANKL (50 ng/mL) for 6 days.

TRAP staining

TRAP histochemical staining was performed to confirm the osteoclast as previously described (Takahashi et al., 2003), by using an acid phosphatase, leukocyte (TRAP) kit (Sigma-Aldrich, St. Louis, MO, USA). Brifely, 1 × 10⁵ BMDMs or RAW264.7 induced osteoclasts were then fixed in 10% neutral-buffered formalin (NBF) solution for 20 min, then the NBF solution was replaced with TRAP staining solution and 0.1% Fast Red AS TR salt at RT. After 45 min cells were washed with 1× PBS for 3 times and imaged.

Measurement of cell viability

Cell viability of BMDMs and RAW264.7 was assessed by using a CCK-8 and FDA assay (Dojindo, Kumamoto, Japan). For CCK-8 assay, BMDMs and RAW264.7 cells (1 × 10⁴ cells per well) were seeded in 96-well plates for 24 h, then cells were treated with different doses of ZA (5, 10, and 50µM) for 48 h. To analyze the cells death of osteoclast, BMDMs, and RAW264.7 (5 × 10³ cells per well) were treated with 10 μM of ZVAD-FMK, 2 μM of Fer-1, or 10 μM of necrostatin-1, for 48 h with or without ZA (50 μM). Then, a total 10 µL of CCK-8 reagent was added to each well additional 4 h at 37 °C with 5% CO₂, and the absorbance at 450 nm of each well was assayed using a microplate reader (BioTek Instruments). For FDA assay, after culture with different dose ZA (5, 10, and 50 µM) for 48 h, BMDMs and RAW264.7 cells were treated with 10 μl of FDA solution (5 mg/mL; Invitrogen, Waltham, CA, USA) at 37 °C with 5% CO₂ for 20 min, then cells images were obtained using a fluorescence microscope (Olympus Corporation, Tokyo, Japan).

Fe²⁺ concentration

The concentration of Ferrous iron (Fe²⁺) in BMDMs and RAW264.7 cells in the presence or absence of ZA (50 μM) was assessed using an iron assay kit (MAK025; Sigma-Aldrich, St. Louis, MO, USA) as the manufacturer’s instructions. Briefly, cell samples were incubated with 10 μL of iron reducer for 30 min at RT, then 100 μL iron probe was added to trigger the reaction; thus, the absorbance was measured at 593 nm.

Lipid reactive oxygen species (ROS) assay

Lipid ROS level in BMDMs and RAW264.7 cells in the presence or absence of ZA (50 μM) was assayed using C11-BODIPY (Invitrogen, Waltham, MA, USA), a fluorescent-labelled oxidation sensitive probe. In brief, BMDMs and RAW264.7 cells were seeded in 24-well plates (5 × 10⁵/well) and treated with ZA (50 μM) with or without FBXO9 overexpression for 48 h, then BMDMs and RAW264.7 cells were cultured with C11-BODIPY probe with a final concentration of 1 μM in at 37 °C with 5% CO₂ for 30 min, then the Lipid ROS levels were assayed using flow cytometer.

MDA and GSH content

MDA in BMDMs and RAW264.7 cells was analyzed using a lipid peroxidation assay kit (ab118970; Abcam, Cambridge, England) in the presence or absence of ZA (50 μM). GSH content in BMDMs and RAW264.7 cells was assayed using a Glutathione Assay Kit (ab65322; Abcam, Cambridge, England) according to the standard protocol. Briefly, cell supernatant, 5, 5′-dithio-bis 2-nitrobenzoic acid solution and the reagents of kits were mixed together and incubated at RT for 10 min, then NADPH was added into this system to trigger the reaction. The absorbance of 5-thio-2-nitrobenzoic acid was detected at 412 nm.

Transient transfection of FBXO9 or si-FBXO9

The recombinant plasmids pcDNA-FBXO9 containing FBXO9 cDNA were sub-cloned into pcDNA3.1 vector via EcoRV/HindIII sites. To overexpress FBXO9, the pcDNA-FBXO9 was transfected into BMDMs cells using Lipofectamine 2000. FBXO9 Knockdown and transfection were performed according to the manufacturer’s instructions. Briefly, cells were transfected with 10 nM of si-FBXO9 RNA (sense AUCAGAAUGACAAUCUUCCUCU, antisense GGAAGAUUGUCAUUCUGAUGCU) or si-Control RNA (sense CAGUCGCGUUUGCGACUGGUU, antisense CCAGUCG-CAAACGCGACUGUU), and the cell was induced by M-CSF (10 ng/mL) and RANKL (50 ng/mL) for further experiment.

Quantitative real-time PCR (qPCR)

After treatment with 50 μM of ZA for 48 h, total RNA was isolated with Trizol reagent (Sigma-Aldrich, , St. Louis, MO, USA) as instructed by the manufacturer. Reverse transcriptional PCR was carried out using the SMART PCR cDNA Synthesis Kits (Clontech, Mountain View, CA, USA ). qPCR was carried out on ABI 7500 RealTime PCR System (Applied Biosystem, Waltham, MA, USA) with powerup SYBR green Mix (Thermo Fisher, Waltham, MA, USA). The fold changes of RNA transcripts were calculated by the 2^−ΔΔCt method and the 18s was used as a reference gene. The qPCR primer pairs in Table 1.

Western blotting analysis

BMDMs-induced osteoclast (1 × 10⁵ cells/well in 12-well plates) were treated with or without ZA (50 μM) for 48 h, the total protein was isolated using RIPA Buffer (Solarbio, Beijing, China), and total protein concentration was quantified by BCA protein assay kit (Beyotime). A total of 20 μg of protein were separated by 8% SDS-PAGE and transferred to PVDF membranes (Merck Millipore, Billerica, MA, USA). After blocking with 5% bovine serum albumin or 5% nonfat milk, the membranes were incubated with anti-FBXO9 (1:1,000, PA5-25475; Thermo Fisher Scientific, Waltham, MA, USA), p53 (1.0 µg/mL, MA5-14067; Thermo Fisher Scientific, Waltham, MA, USA), ubiquitin (1:2,000, ab134953; Abcam, Cambridge, England), and GAPDH (1:5,000, MA1-16757; Thermo Fisher Scientific, Waltham, MA, USA) overnight at 4 °C. Then the membranes were treated with HRP-conjugated anti-mouse or rabbit secondary antibody (1:5,000) for 1 h at room temperature.

Co-immunoprecipitation

BMDMs-induced osteoclast were lysed using NP40 buffer (10 mM Tris-HCl at pH 8.0, 140 mM NaCl, 1.5 mM MgCl2, 0.5% Nonidet-P40, 20 mM dithiothreitol, 500 U/mL RNAsin, and 0.5% [w/v] deoxycholate), cell lysates were incubated with FBXO9 (2 µg/ml; PA5-25475) or p53 (2 µg/ml; MA5-14067) antibody for 4 h, then Protein A/G beads (Thermo Fisher Scientific, Waltham, MA, USA) were added to the IP reactions and left rotating overnight at 4 °C, then beads were washed by PBS containing protein inhibitors for three times, then the immunoprecipitates were analyzed using western blotting with FBXO9 antibody or p53 antibody.

Statistical analysis

All the data are shown as the mean ± standard error of mean (SEM) from three independent experiments. Statistical analysis was carried out using SPSS 19.0 software (IBM Corp., New York, USA). The significance between two groups was analyzed using one-way ANOVA followed by Tukey-Kramer multiple comparisons test or unpaired Student’s t-test. p < 0.05 was considered to indicate a statistically significant difference.

ZA treatment facilitated the ferroptosis of osteoclasts

To investigate the function of ZA on osteoclasts, RAW264.7 cells and bone marrow-derived macrophages (BMDMs) were pre-treated with RANKL for 6 days followed by ZA treatment in different concentrations (5, 10, and 50 uM) for 48 h, the cell model was confirmed by TRAP staining (Figs. 1A and 1B). Cell viability was analyzed by CCK-8 assay. As shown in Figs. 1C and 1D, ZA treatment suppressed cell viability in a dose-dependent manner. The results from FDA staining also showed that the effect of ZA on promoting cell viability of osteoclasts (Figs. 1E and 1F). Impressively, the CCK-8 assay results showed that cell death of osteoclasts induced by ZA treatment was obviously blocked by ferrostatin-1 (Fer-1, a specific inhibitor of ferroptosis) but not necrostatin-1 (a specific inhibitor of necroptosis) and ZVAD-FMK (a specific inhibitor of apoptosis) (Figs. 1G and 1H).

To define the role of ZA in the ferroptosis of osteoclasts, the ferroptosis signaling was evaluated in osteoclasts after ZA treatment. As shown in Figs. 2A–2C and 2F–2H, ZA treatment markedly increased the Fe²⁺ level, MDA content, and ROS level in a dose-dependent manner in osteoclasts, differentiated from RAW264.7 cells and BMDMs, suggesting the promotion of ferroptosis signaling in osteoclasts treated by ZA. Besides, ZA treatment also suppressed the levels of Gpx4 and GSH in a dose-dependent manner in osteoclasts (Figs. 2D, 2E, 2I, and 2J). These results demonstrate that ZA treatment facilitates the ferroptosis of osteoclasts.

FBXO9 was downregulated in osteoclasts after ZA treatment

To investigate the mechanism underlying ZA-induced osteoclasts ferroptosis, the differentially expressed genes (DEGs) of osteoclasts induced by bisphosphonates alendronate- and risedronate-treatment were obtained from GSE63009, and the common DEGs were identified by venn diagram analysis. As shown in Fig. 3A, 18 common genes were identified (CFAP53, COL14A1, ARSJ, ABCA9, CXorf57, GPR22, STXBP5L, MSANTD4, RRP15, UGT1A2, IRF4, TFAP2D, TRHDE, ASMT, CAPS, COMMD10, VSTM4, FBXO9). The level of these 18 genes was evaluated in osteoclasts treated with or without ZA using qPCR analysis. Figures 3B and 3C showed that only FBXO9 was significantly decreased in osteoclasts after treatment with ZA. Similar to the qPCR results, the results from western blotting showed that the expression of FBXO9 was obviously decreased in osteoclasts after treatment with ZA. These results indicate that the FBXO9 was downregulated by ZA treatment (Fig. 3D).

FBXO9 inhibition facilitated the ferroptosis of osteoclasts

To investigate the function of FBXO9 on osteoclasts, the expression of FBXO9 was down-regulated by si-FBXO9 in BMDMs-differentiated osteoclasts. As shown in Figs. 4A and 4B, the expression of FBXO9 was significantly decreased by si-FBXO9. Cell viability of osteoclasts was significantly decreased in the FBXO9 knockdown group compared with the control group (Fig. 4C). The results from FDA staining also showed the effect of FBXO9 on inhibiting cell viability (Figs. 4D and 4E).

To investigate the role of FBXO9 in ferroptosis of osteoclasts, the ferroptosis signaling was evaluated in osteoclasts differentiated from BMDMs. As shown in Figs. 4F–4J, FBXO9 knockdown significantly increased the Fe²⁺ level, MDA content, ROS level and decreased the GPX4 level, GSH content in osteoclasts. These results suggested that FBXO9 inhibition facilitates the ferroptosis of osteoclasts.

ZA treatment facilitated the ferroptosis of osteoclasts by suppressing FBXO9

To explore whether FBXO9 mediated the function of ZA in regulating the ferroptosis of osteoclasts, the osteoclasts differentiated from BMDMs were treated by ZA in the presence or absence of FBXO9. As shown in Figs. 5A and 5B, the qPCR and western blotting analysis results showed that the expressions of FBXO9 were decreased by ZA treatment and restored by FBXO9 overexpression. CCK8 results showed that cell viability of osteoclasts was decreased by ZA treatment, but these effects were blocked by FBXO9 overexpression (Fig. 5C). Consistently, the FDA staining also showed the inhibition of ZA on osteoclasts cell viability was restored by FBXO9 overexpression (Figs. 5D and 5E). Besides, the Fe²⁺ level, MDA content, and ROS level were obviously increased and the GPX4 level, GSH content was significantly decreased by ZA treatment, while these effects were blocked by FBXO9 overexpression (Figs. 5F–5J). These results suggested that ZA treatment facilitates the ferroptosis of osteoclasts by suppressing FBXO9.

FBXO9 inhibition facilitated the ferroptosis of osteoclasts by blocking the ubiquitin mediated-proteasome degradation of p53

Previous studies showed that FBXO9, an E3 ubiquitin ligase, mediated protein stability through ubiquitin mediated-proteasome degradation (Liu et al., 2020). Given that dysregulated ubiquitination has been widely reported to be involved in many diseases by regulating cell ferroptosis (Yang et al., 2020; Zhang et al., 2020), the target of FBXO9 was predicted by ubibrowser (http://ubibrowser.ncpsb.org.cn/ubibrowser/). Figure 6A showed the 20 potential target genes that interacted with FBXO9. Interestingly, among these genes, the p53 gene is an important regulator of ferroptosis. We next explored whether FBXO9 decreases p53 protein level by promoting its ubiquitination-mediated degradation. Knockdown of FBXO9 in osteoclasts did not change the p53 mRNA level (Fig. 6B). Fascinatingly, the protein level of p53 was significantly increased after FBXO9 inhibition (Fig. 6C), suggesting that FBXO9 decreased p53 expression possibly by the ubiquitin-proteasome-mediated degradation. Next, a reciprocal Co-IP assay was performed to confirm whether FBXO9 directly interacts with p53. As shown in Fig. 6D, a positive p53 signal was observed in the protein complex pulled down by FBXO9 antibody. Meanwhile, FBXO9 was also detected in the co-immunoprecipitation complex pulled-down by p53 antibody. Next, cycloheximide assay (CHX) was performed to detect the protein stability of p53 in osteoclasts transfected with si-FBXO9. As shown in Fig. 6E, the protein stability of p53 was obviously increased in the FBXO9 knockdown osteoclasts. Then the p53 ubiquitination was assessed through IP with FBXO9 antibody and subsequent western blotting with ubiquitin antibody. Figure 6F showed that FBXO9 knockdown obviously decreased p53 ubiquitination in osteoclasts (Fig. 6F). In conclusion, these results indicated that FBXO9 directly interacts with p53 and promotes its degradation.