2025 Volume 72 Issue 2 Pages 205-219
The ferroptosis of osteoblasts has been demonstrated to play a significant role in the development of steroid-induced osteonecrosis of the femoral head (SONFH). Additionally, microRNAs (miRNAs) have been identified as regulators of SONFH progression. However, the precise role of miRNAs in the regulation of osteoblast ferroptosis remains unclear. This study explored the role of exosomal miR-150-3p, derived from bone marrow mesenchymal stem cells (BMSCs), in osteoblast ferroptosis in SONFH. Dexamethasone (DEX) was used to treat osteoblasts to induce ferroptosis. BMSCs exosomes with different levels of miR-150-3p were introduced into a co-culture with the cells. To verify the targeting relationship between growth factor independence 1 (Gfi1) and the miR-150-3p promoter, as well as between miR-150-3p and beta-transducin repeat containing E3 ubiquitin protein ligase (BTRC), respectively, chromatin immunoprecipitation (ChIP), RNA immunoprecipitation (RIP), and dual luciferase assays were employed. It was found that BMSCs-Exos-miR-150-3p mitigated DEX-triggered ferroptosis in osteoblasts. MiR-150-3p directly targeted BTRC, leading to its downregulation in osteoblasts. The BTRC/Nuclear factor erythroid 2-related factor 2 (Nrf2) pathway was involved in the inhibition of DEX-induced osteoblast ferroptosis by BMSCs-Exos-miR-150-3p. Overexpression of BTRC reversed the inhibitory effect of BMSCs-Exos-miR-150-3p. In a SONFH rat model, BMSCs-Exos-miR-150-3p alleviated ferroptosis in osteoblasts through BTRC/Nrf2. In addition, Gfi1 bonded to the miR-150-3p promoter and inhibited its transcription. Gfi1 silencing elevated miR-150-3p levels and improves cell viability of BMSCs. In conclusion, our results suggest that BMSCs-Exos-miR-150-3p alleviates SONFH by suppressing ferroptosis through the regulation of BTRC/Nrf2 and miR-150-3p may be a potential target for SONFH treatment.
Steroid-induced osteonecrosis of the femoral head (SONFH) is a prevalent orthopedic condition that presents with substantial disability and poor prognosis [1]. High doses and long-term steroid use are the causes of SONFH. High doses of glucocorticoids stimulate oxidative stress in osteoblasts and can even lead to cell death [2]. Osteoblasts dominate bone formation less than osteoclasts, resulting in femoral bone loss, collapse, and hip joint dysfunction [3]. At present, total hip arthroplasty is the most effective treatment. However, trauma is large, cost is high, and some patients need to undergo revision surgery. Therefore, identifying novel treatments for SONFH is crucial.
Ferroptosis is a new model of non-apoptotic programmed cell death that arises from iron-dependent lipid peroxidation damage induced by iron dependency [4]. Dysfunctions in iron ion accumulation, lipid metabolism, and intracellular antioxidant components, including glutathione, glutathione peroxidase 4 (GPX4), and solute carrier family 7 member 11 (SLC7A11), cause mitochondrial destruction and lipid peroxidation, resulting in ferroptosis [5]. In addition, acyl-CoA synthetase long-chain family member4 (ACSL4) promotes ferroptosis by promoting the lipid peroxidation reaction [6]. The inhibition of glucocorticoid-induced ferroptosis in bone marrow mesenchymal stem cells (BMSCs) can reduce SONFH [7]. Targeting ferroptosis in osteoblasts is a potential therapeutic approach to halt and manage SONFH progression. However, the mechanisms and targets for regulating ferroptosis in SONFH remain unclear.
BMSCs are multipotent stem cells that exhibit significant therapeutic potential in managing orthopedic conditions such as SONFH [8]. Exosomes (Exos) are a subset of extracellular vesicles measuring 50–150 nm [9]. These vesicles contain DNA, RNA, and proteins. As one of the primary pathways for extracellular signal transduction, they play vital roles in intercellular communication and interactions [10]. BMSCs can secrete abundant exosomes [11], which are able to induce osteoblast proliferation, thereby alleviating osteoporosis [12]. Exosome utilization has emerged as a significant approach for the prevention and treatment of SONFH [13]. MicroRNAs (miRNAs) are small (approximately 22 nt) noncoding RNAs that play crucial roles in modulating gene expression [14]. Literature has revealed that Exosomes carrying overexpressed miR-122-5p promote osteoblast proliferation and attenuate osteonecrosis of the femoral head (ONFH) development [15]. In addition, miR-148a-3p in BMSCs prevents femoral head necrosis by inhibiting SMURF1 [16]. BMSCs-Exos-miR-150-3p demonstrates its capability to enhances osteoblast proliferation and differentiation while suppressing apoptosis in animal models of osteoporosis [17]. However, the effect of BMSCs-Exos-miR-150-3p on SONFH and steroid-induced osteoblast ferroptosis remain unclear.
Growth factor independence 1 (Gfi1) is responsible for encoding a DNA-binding protein with nuclear zinc finger domains [18]. In addition, it functions as a transcriptional repressor of genes associated with hematopoiesis, as well as the self-renewal and quiescence of hematopoietic stem cells [19]. Further, Gfi1 plays a crucial regulatory role in balancing miRNA expression. BMSCs from patients with multiple myeloma and myeloma mouse models showed increased Gfi1 levels [20]. In the BMSCs of patients with myeloma, reducing Gfi1 expression promotes the recovery of osteogenic markers [20]. After inoculation of mice with multiple myeloma cells overexpressing Gfi1, increased bone destruction and an increased number and size of osteoclasts were observed [21]. Therefore, Gfi1 overexpression is associated with an increase in the number of osteoclasts, and an imbalance between osteoblasts and osteoclasts may lead to SONFH. However, there are no reports on the role of Gfi1 in SONFH. In mouse bone marrow cells, Gfi1 regulates the expression of miR-21 and miR-196b [22]. Whether Gfi1 similarly regulates miR-150-3p expression and thus affects SONFH progression needs to be further investigated.
Beta-transducin repeat containing E3 ubiquitin protein ligase (BTRC) is a constituent of E3 ubiquitin ligases that is responsible for substrate recognition [23]. BTRC regulates the expression of numerous proteins through ubiquitination [24]. Nuclear factor erythroid 2-related factor 2 (Nrf2) is degraded by ubiquitination as a substrate for the ubiquitin ligase BTRC [25]. Nrf2 is a redox-sensitive transcription factor and a pivotal modulator of the intracellular antioxidant capacity [26]. Recent studies have shown that after entering the nucleus, Nrf2 promotes the transcription of multiple genes such as SLC7A11 and GPX4 and prevents ferroptosis [27]. In dexamethasone (DEX)-treated osteoblasts, Nrf2 expression is reduced, and Nrf2 activation inhibits DEX-induced osteoblast death [28]. These studies suggest that BTRC acts as an E3 ligase to degrade Nrf2 and plays a role in regulating steroid-induced osteoblast ferroptosis. Bioinformatic analysis revealed targeted binding between miR-150-3p and BTRC.
In this study, we propose that Gfi1 knockdown in BMSCs enhances exosomal miR-150-3p expression, which inhibits Nrf2 ubiquitination and degradation and elevates Nrf2 levels by targeting BTRC, ultimately suppressing osteoblast ferroptosis in SONFH. This study presents novel possibilities for treating SONFH.
BMSCs were purchased from OriCell (RASMX-01001, Guangzhou, China) and cultured in Dulbecco’s modified Eagle medium (DMEM) (Invitrogen, Carlsbad, CA, USA) with 10% fetal bovine serum (FBS; Invitrogen) at 37°C and 5% CO2. Primary osteoblasts were isolated from neonatal Sprague-Dawley rats as previously described [29]. Briefly, the calvarial bone from neonatal rat heads was extracted and rinsed with phosphate-buffered saline (PBS). Following the elimination of soft tissues, the calvaria were sectioned into fragments and placed in a 4 mL digestion solution (0.25% trypsin with 0.02% ethylenediaminetetraacetic acid (EDTA)) at 37°C for 40 min, with periodic gentle agitation every 5 min. Once digested, the calvaria were fragmented and immersed in full-cell culture medium. Four hours post-plating, the medium volume was increased to 8 mL, and static culture conditions for 4 days. Primary osteoblasts that emerged from the bone chips were detached using trypsin and transferred to fresh dishes for subsequent experiments. All cells were cultured in DMEM containing 10% FBS at 37°C and 5% CO2. All animal experiments were conducted with the approval of the Ethics Committee of the Second Xiangya Hospital, Central South University (Approval Number: 2022028). DEX (10 μM, Sigma, St. Louis, MO, USA) for 24 h.
2. Cell transfectionMiR-150-3p mimic (B02001), miR-150-3p inhibitor (B03001), and negative controls were obtained from GenePharma (Shanghai, China). ShRNAs (C02007) (sh-Gfi1 and sh-BTRC) and negative controls were obtained from GenePharma. The Gfi1 or BTRC genes were synthesized and cloned into pcDNA3.1 expression vectors (C05008) (GenePharma was used to construct Gfi1-overexpressing vectors (OE-Gfi1) or BTRC-overexpressing vectors (OE-BTRC)). After reaching 80% confluence, the osteoblasts were transfected with the above constructs using the Lipofectamine 3000 reagent (Invitrogen) for 48 h.
3. Isolation, identification, and internalization of exosomesExosomes were isolated from the BMSC culture medium of BMSCs utilizing an Exo Extraction Kit (4478359, Thermo Fisher Scientific, Carlsbad, CA, USA). Transmission electron microscopy and nanoparticle tracking analysis (NTA) were used to capture morphological images and determine exosome size. In addition, Western blot was used to assess the expression of exosomal markers, including CD63 (MA5-35208, 1:1,000, Invitrogen), CD81 (ab109201, 1:1,000, Abcam, Cambridge, MA, USA), tumor susceptibility gene 101 (TSG101) (ab125011, 1:1,000, Abcam), and heat shock protein 70 (HSP70) (ab181606, 1:1,000, Abcam). For exosome uptake analysis, exosomes were fluorescently labeled with PKH26 membrane dye (Sigma). PKH26-labeled BMSC-derived exosomes were co-cultured with isolated osteoblasts for 24 h. The cells were fixed with 4% paraformaldehyde (PFA), stained with ActinGreenTM 488 and DAPI, and observed under a fluorescence microscope (Olympus, Tokyo, Japan).
4. Cell viability assayThe Cell Counting Kit-8 assay (CCK-8; Dojindo, Kumamoto, Japan) was used to test cell viability. Primary rat osteoblasts were cultured in 96-well plates (6,000 cells per well). After different treatments of DEX and Exos (50 μg/mL) for 24 h, 10 μL CCK-8 was added to each well and incubated at 37°C for 2 h. Absorbance was assessed using a microplate photometer (BioTek, Winooski, VT, USA) at 450 nm.
5. Detection of lipid peroxidationThe BODIPYTM 581/591 C11 kit (#RM02821, Abclonal Technology, Woburn, MA, USA) was used to detect the levels of lipid peroxidation. The osteoblasts were incubated with C11 BODIPY, followed by centrifugation at 500 × g for 5 min. Samples were then assessed using a flow cytometer (BD Biosciences) to measure the green fluorescence intensity in the fluorescein isothiocyanate (FITC) channel, indicating the oxidation of C11 BODIPY. The acquired data were subsequently analyzed using FlowJo v7.6 software.
6. Malondialdehyde (MDA) detectionThe MDA content in femoral head tissues and osteoblasts was measured using an MDA assay kit (#A003-2-2, Jiancheng Bioengineering Institute, Nanjing, China). Protein concentration was determined using the Bradford assay (Beyotime, Shanghai, China). To each well, we added 0.1 mL of the protein sample and 0.2 mL of MDA detection solution. After thoroughly mixing, the mixture was heated in boiling water for 15 min. After cooling, 200 μL of the resulting supernatant was transferred to a 96-well plate. The absorbance was determined at 532 nm with a microplate reader (MultiskanTM FC, Thermo Fisher Scientific).
7. Detection of iron levelThe concentration of Fe2+ in cells was measured using an Intracellular Iron Colorimetric Assay Kit (E1042, Applygen, Beijing, China). The cells were lysed and mixed with 4.5% potassium permanganate solution. Following 1 h of incubation at 60°C, iron ion detector was added. The resulting samples were then transferred to 96-well plates. Absorbance was measured at 550 nm using a microplate reader. After harvesting, rat femoral heads were promptly homogenized using PBS. The supernatant was collected by centrifugation. An Iron Assay Kit (ab83366, Abcam) was used to measure iron levels, following the guidelines provided by the manufacturer.
8. Immunofluorescence stainingFor GPX4 determination, the osteoblasts were plated on coverslips and allowed to adhere overnight. Subsequently, the cells were fixed with 4% PFA for 15 min and rinsed with PBS. The cells were permeabilized with 0.5% Triton X-100 at room temperature for 10 min, followed by blocking with goat serum for 30 min. The cells were then incubated with anti-GPX4 antibody (PA5-102521, 1:200, Invitrogen) at 4°C overnight. Subsequently, the FITC-conjugated goat anti-rabbit IgG secondary antibody (ab6717, 1:1,000, Abcam) was added and incubated at room temperature for 1 h. Finally, the nuclei were stained with DAPI. Images were captured using a fluorescence microscope (Olympus).
9. Alkaline phosphatase (ALP) and alizarin red staining (ARS)For ALP staining, osteoblasts were cultured on coverslips for 7 days. The cells were fixed in 4% PFA for 15 min. After washing twice with PBS, all samples were incubated in a working solution of 5-bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium chloride in the dark for 20 min. ALP staining images were subsequently analyzed using a microscope. For ARS staining, the osteoblasts were allowed to aggregate and form calcium nodules on coverslips. Subsequently, the cells were fixed with 95% ethanol and stained with a 1% solution of Alizarin Red. After sealing, the cells were observed under a microscope.
10. Quantitative real-time PCRTotal RNA was isolated from tissues and cells using TRIzol reagent (Invitrogen) and reverse-transcribed to cDNA using the PrimeScript RT reagent Kit (Takara, Dalian, Liaoning, China). Quantitative real-time PCR was performed using the SYBR Green PCR Core Kit (Applied Biosystems, Foster City, CA, USA) on a QuantStudio 6 Flex Real-Time PCR System (Applied Biosystems). Primer sequences were: miR-150-3p F: 5'-GCCGAGCTGGTACAGGCCT-3', RT: 5'-GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGACTCCCCC-3'; Gfi1 F: 5'-TGAGTGCAGGAGAGTCGAAG-3', R: 5'-CTTGAAAGGCAGCGTGTAGG-3'; BTRC F: 5'-TGCTCCAAAGACCGATCCAT-3', R: 5'-ACCAGCCTGTCTCTGTACTG-3'; U6 F: 5'-CTCGCTTCGGCAGCACA-3', R: 5'-AACGCTTCACGAATTTGCGT-3'; β-actin F: 5'-AGGTCGGAGTCAACGGATTT-3', R: 5'-TGACGGTGCCATGGAATTTG-3'; U6 was regarded as internal reference for miR-150-3p and β-actin as internal reference for Gfi1 and BTRC. Relative expression of genes was calculated using the 2–∆∆Ct method.
11. Western blotThe cells or tissues were lysed in RIPA buffer (Beyotime). Protein was quantified using a BCA kit (Beyotime). Protein extracts were separated using 10% SDS-PAGE and blotted on PVDF membranes (Millipore, Bedford, MA, USA). Membranes were blocked with 5% bovine serum albumin (BSA) and incubated with primary antibodies overnight at 4°C: GPX4 (ab125066, 1:1,000, Abcam), SLC7A11 (PA1-16893, 1:1,000, Invitrogen), ACSL4 (ab155282, 1:1,000, Abcam), BTRC (37-3400, 1:1,000, Invitrogen), and Nrf2 (ab313825, 1:1,000, Abcam), followed by incubation with horseradish peroxidase (HRP)-conjugated secondary antibodies (#7074, 1:1,000, Cell Signaling Technology, CST, Danvers, MA, USA) for 1 h. Membranes were rinsed with phosphate-buffered saline with Tween (PBST) and developed utilizing chemiluminescence (Beyotime). The ImageJ software (NIH, Bethesda, MA, USA) was used for western blot. β-actin was used as internal control.
12. Dual-luciferase reporter assayBinding sequences of miR-150-3p to fragments of wild-type (WT) or mutant (MUT) BTRC were obtained from Promega (Madison, WI, USA). To generate BTRC-WT or BTRC-MUT plasmids, wild-type (WT: 5'-UGUACCA-3') and mutant (MUT: 5'-ACAUGGU-3') sequences were respectively cloned into pmirGLO luciferase reporter vectors (Promega). Osteoblasts were transfected with miR-150-3p mimics (or mimic NC) and either BTRC-WT (or BTRC-MUT) using the Lipofectamine 3000 reagent for 48 h. Luciferase activity was measured using a dual-luciferase reporter system (Promega).
To validate the binding site of Gfi1 within the miR-150-3p promoter region, we amplified and inserted four fragments (pro, pro#1–3) of the miR-150-3p promoter, each containing different combinations of three binding sites (Pro: wild type 1, 2, and 3; Pro#1: wild type 1, mutants 2 and 3; Pro#2: wild type 2, mutants 1 and 3; Pro#3: wild type 3, mutants 1 and 2) into the pGL3-basic vector (Promega) to generate recombinant luciferase reporter plasmids. Cells were seeded into a 24-well plate and incubated for 24 h. Subsequently, cells were co-transfected with miR-150-3p promoter plasmids and either sh-NC or sh-Gfi1 using Lipofectamine 3000 and incubated for 48 h. Luciferase activity was measured using a dual-luciferase reporter system (Promega).
13. Chromatin Immunoprecipitation (ChIP) assayThe EZChIP Chromatin Immunoprecipitation Kit (Sigma) was used for the ChIP assay according to the manufacturer’s instructions. Briefly, cells were crosslinked with 1% formaldehyde for 10 min, and glycine was added for 5 min at room temperature to quench crosslinking. Subsequently, sonication was performed on the cell lysates to generate chromatin fragments. The cells were collected in lysis buffer containing 1% phenylmethanesulfonyl fluoride. The lysates (2%) were used as the input reference. The remaining lysates were incubated with 3 μg of anti-Gfi1 (sc-373960, Santa Cruz Biotechnology, TX, USA) or IgG (#2729, CST) antibodies with rotation at 4°C overnight. Immune complexes were pulled down using ChIP-grade protein G magnetic beads (Roche, Basel, Switzerland). DNA cross-links were reversed using 5 mol/L NaCl and proteinase K at 65°C for 2 h. Finally, the purified DNA was subjected to PCR amplification using miR-150-3p promoter primers.
14. Flow cytometryAn Annexin V-FITC/propidium iodide (PI) Apoptosis Detection Kit (Vazyme Biotech Co., Ltd., Nanjing, China) was used to detect BMSC apoptosis. After cell collection, cells were rinsed with PBS, and resuspended in 96 μL Annexin V binding buffer. Subsequently, cells were stained with 5 μL Annexin V-FITC and 10 μL PI for 10 min in the dark. The rate of apoptosis was evaluated by flow cytometry (Beckman, Florida, USA).
15. RNA immunoprecipitation (RIP)Immunoprecipitation analysis of RNA-binding proteins was performed using the Magna RIP RNA-Binding Protein Immunoprecipitation Kit (Sigma). The cells were then rinsed with cold PBS and lysed in RIP lysis buffer. Either Ago2 antibody (MA5-14861, Invitrogen) or IgG (#2729, CST) was used for immunoprecipitation. RIP lysate was incubated with a magnetic bead binding antibody overnight at 4°C. Proteinase K was used to immunoprecipitate degraded proteins. The bound RNA was separated from the supernatant and the RNA concentration was measured using a NanoDrop spectrophotometer (Thermo Fisher Scientific). Purified RNA was analyzed using RT-qPCR.
16. Co-immunoprecipitation (Co-IP)For Co-IP, cell lysates (1,000 μg) were subjected to overnight incubation at 4°C with antibodies specific to BTRC (37-3400, Invitrogen) or IgG (ab182931, Abcam) at a concentration of 8 μg. Subsequently, 40 μL of protein A + G agarose beads (Beyotime) were added. Mixture was further incubated for 4 h at 4°C. After immunoprecipitation, the immunoprecipitates were washed three times with lysis buffer. Next, they were eluted with the loading buffer. The protein levels of BTRC and Nrf2 were assessed using western blot.
17. SONFH rat modelMale Sprague-Dawley rats (8 weeks old, 200–220 g) were obtained from Vital River (China) and categorized into four groups (n = 6): control, SONFH, SONFH + Exos-mimics-NC, and SONFH + Exos-miR-150-3p-mimics. To induce SONFH, rats were administered weekly intramuscular injections of methylprednisolone (MP, 20 mg/kg/d; Pfizer, NY, USA) for three consecutive days, over a span of three weeks, following the established protocol [30]. Following the same procedure, rats in the control group were injected with saline. Following initial injection, model rats received an additional injection of 200 μL either exosomes or PBS through the tail vein. At the end of the 6-week period, the rats were euthanized to obtain tissue samples. The Animal Research Ethics Committee of the Second Xiangya Hospital, Central South University, approved all experimental procedures.
18. Hematoxylin and eosin (H&E) and immunohistochemistry (IHC) stainingFemoral heads obtained from the rats were fixed with 10% formalin for 24 h and decalcified with 10% EDTA. Subsequently, the tissues were dehydrated using an ethanol gradient, followed by treatment with xylene. Paraffin embedding was performed, and the tissues were then sectioned at a thickness of 5 μm. The samples were stained with hematoxylin and eosin. The images were examined under an Olympus microscope. For IHC staining, slices underwent a series of steps, including deparaffinization, antigen retrieval, blocking with goat serum for 30 min, and incubation with primary antibodies targeting against BTRC (PA5-109459, Invitrogen) at 4°C overnight, and then incubation with HRP-conjugated secondary antibody (#8114, CST). Color development was achieved using diaminobenzidine. Acquisition of section Images were acquired using an Olympus microscope.
19. Statistical analysisData were analyzed using GraphPad Prism 8.0 software. Data were expressed as mean ± standard deviation (SD) and three independent experiments were performed. Student’s t-test was used to assess differences between two groups, whereas one-way analysis of variance (ANOVA) followed by Tukey’s post-hoc test was used for comparisons among multiple groups. Statistical significance was set at p < 0.05.
First, the culture supernatant of rat BMSCs was collected for exosome isolation. Under transmission revealed that BMSC-Exos appeared as rounded or oval-shaped vesicles characterized by a double-layered membrane structure (Fig. 1A). NTA confirmed that the diameter of Exos ranged from 50 to 150 nm (Fig. 1B). Exosomal marker proteins (CD63, CD81, TSG101, and HSP70) were increased in BMSC-derived exosomes but were rarely expressed in BMSCs (Fig. 1C). In addition, confocal microscopy revealed the uptake of PKH26-labeled BMSCs-Exos by rat primary osteoblasts (Fig. 1D). Collectively, we successfully isolated and verified BMSC-Exos, and found that they were taken up by rat primary osteoblasts.
Next, we investigated if BMSCs-Exos-miR-150-3p regulates osteoblast ferroptosis. BMSCs were transfected with miR-150-3p mimics, followed by the isolation of exosomes from the transfected BMSCs. Expression of miR-150-3p in BMSCs or exosomes was upregulated by miR-150-3p mimics but suppressed by the miR-150-3p inhibitor (Fig. 2A). As shown in Fig. 2B, exosomes were extracted from BMSCs, BMSCs transfected with mimic-NC, BMSCs transfected with miR-150-3p mimics, BMSCs transfected with inhibitor-NC, and BMSCs transfected with miR-150-3p inhibitor were extracted respectively. Rat osteoblasts were treated with BMSCs exosomes, co-treated with DEX for 24 h, and then examined for iron death indices and other assays. We observed that the isolated osteoblasts adhered to the wall. They were irregular in shape and cytoplasmically rich. (Fig. 2C). In addition, DEX treatment decreased osteoblast viability, Exos-miR-150-3p-mimics increased the viability of DEX-treated osteoblasts, while the Exos-miR-150-3p-inhibitor did not have the same effect (Fig. 2D). Next, the ferroptosis inducer erastin was used as a positive control. Erastin treatment significantly increased lipid peroxidation in osteoblasts. The level of lipid peroxidation after DEX treatment was similar to that in the erastin group, which was significantly higher than that in the control group. BMSC-derived exosome miR-150-3p inhibited DEX-induced lipid peroxidation, whereas the effect of exosome knockdown on miR-150-3p disappeared (Fig. 2E). Furthermore, miR-150-3p mimics increased the exosome inhibitory effect of DEX-increased MDA and Fe2+ levels, whereas the miR-150-3p inhibitor reversed the exosome inhibitory effect (Fig. 2F, G). Moreover, GPX4 and SLC7A11 protein levels were reduced and ACSL4 levels were increased in DEX-treated osteoblasts, and Exos treatment reversed these changes. Overexpression of miR-150-3p in Exos increased this reversal, whereas the effect of exosomes with miR-150-3p inhibition was greatly attenuated (Fig. 2H). Immunofluorescence staining revealed that, compared to the control group, DEX treatment reduced the level of GPX4 in osteoblasts. GPX4 levels were significantly increased after the intervention with BMSC exosomes overexpressing miR-150-3p, whereas the intervention with BMSC exosomes knocking down miR-150-3p did not change significantly compared to the DEX group (Fig. 2I). In general, these results suggest that DEX treatment could lead to ferroptosis of osteoblasts and BMSCs-Exos-miR-150-3p could alleviate this damage. ALP and ARS showed that DEX treatment inhibited osteoblast differentiation and calcium deposition. BMSC-derived exosomes reversed the inhibitory effect of DEX, miR-150-3p mimics increased this effect, whereas exosomes with miR-150-3p inhibitors showed no significant effect (Fig. 2J, K). Thus, miR-150-3p in BMSC-Exos promotes osteoblast differentiation and calcium deposition in DEX-treated osteoblasts.
Subsequently, we investigated whether Gfi1 regulates miR-150-3p expression in BMSCs. We found that Gfi1 levels were increased in BMSCs transfected with the Gfi1 overexpression vector and were decreased by sh-Gfi1 transfection (Fig. 3A). Furthermore, overexpression of Gfi1 reduced miR-150-3p expression in BMSCs and Exos, whereas knockdown of Gfi1 increased miR-150-3p levels (Fig. 3B). The JASPAR database predicted three binding sites between Gfi1 and the miR-150-3p promoter (Fig. 3C). Subsequently, the ChIP assay verified that the Gfi1 antibody enriched site 3 of the miR-150-3p promoter, but not sites 1 or 2 (Fig. 3D). Furthermore, treatment with sh-Gfi1 increased the luciferase activity of cells driven by Pro or Pro#3, but not by Pro#1 or Pro#2 (Fig. 3E). The viability of BMSCs was inhibited and apoptosis was promoted by overexpressing Gfi1, whereas Gfi1 silencing had the opposite effect (Fig. 3F, G). These data indicate that Gfi1 affects the viability of BMSCs and negatively modulates miR-150-3p expression by directly targeting the miR-150-3p promoter.
We explored the downstream effecter of miR-150-3p. BTRC is predicted to be a downstream target of miR-150-3p. DEX increased the mRNA and protein levels of BTRC in osteoblasts, and BMSC-Exos-miR-150-3p-mimics decreased its expression levels. However, the miR-150-3p inhibitor reversed the exosome inhibitory effect (Fig. 4A, B). Next, the TargetScan database was used to predict the binding site of miR-150-3p in BTRC (Fig. 4C). MiR-150-3p mimics inhibited the luciferase activity of wild-type BTRC, but had no effect on mutant BTRC (Fig. 4D). The RIP assay indicated that compared with IgG, BTRC and miR-150-3p were enriched in anti-Ago2 immunoprecipitation (Fig. 4E). MiR-150-3p overexpression reduced BTRC mRNA and protein expression, whereas miR-150-3p knockdown promoted BTRC expression (Fig. 4F, G). These data clarify that miR-150-3p suppresses BTRC expression in osteoblasts by targeting BTRC.
We further explored the specific mechanisms through which miR-150-3p regulates ferroptosis. The co-IP assay showed that there was a protein-protein interaction between BTRC and Nrf2 in osteoblasts (Fig. 5A). BTRC knockdown increased Nrf2 level, while BTRC overexpression decreased it (Fig. 5B). Next, rat primary osteoblasts were subjected to different treatments: control, DEX, DEX + Exos, DEX + Exos-mimics-NC, DEX + Exos-miR-150-3p-mimics, DEX + Exos-miR-150-3p-mimics + OE-NC, DEX + Exos-miR-150-3p-mimics + OE-BTRC. BTRC upregulation reversed the BMSC-derived exosomal miR-150-3p-increased cell viability of DEX-triggered osteoblasts (Fig. 5C). BTRC overexpression reversed the inhibitory effect of BMSC-derived exosomal miR-150-3p on DEX-induced lipid peroxidation (Fig. 5D). Moreover, BMSCs-Exo-miR-150-3p markedly reduced MDA and Fe2+ levels in DEX-treated osteoblasts, and this reduction was partially ameliorated by BTRC overexpression (Fig. 5E, F). In addition, BTRC overexpression reversed the effects of BMSCs-Exo-miR-150-3p on increased GPX4, SLC7A11, and Nrf2 levels, and decreased ACSL4 and BTRC levels in DEX-treated osteoblasts (Fig. 5G). Immunofluorescence showed that the expression of GPX4 was consistent with the western blot results (Fig. 5H). These data suggested that BMSCs-Exo-miR-150-3p alleviated DEX-induced ferroptosis in osteoblasts by suppressing BTRC and activating Nrf2. However, following BTRC overexpression in osteoblasts, the inhibitory effect of BMSCs-Exo-miR-150-3p on DEX-induced ferroptosis was significantly reduced. Osteoblast differentiation and calcium deposition in DEX-treated cells were significantly promoted by BMSCs-Exo-miR-150-3p, and inhibited by BTRC upregulation (Fig. 5I, J).
To validate the in vivo effect of Exos-miR-150-3p on SONFH, a rat model of SONFH was established using MP. The femoral tissues were collected, and RT-qPCR analysis demonstrated that compared to the control group, miR-150-3p expression in the femoral tissues of SONFH rats was decreased, and BMSC-Exos intervention significantly upregulated miR-150-3p expression in the femoral tissues of SONFH rats (Fig. 6A). HE staining revealed no osteonecrosis in the control group. In the SONFH group, the femoral trabecular bone was missing, the cavity increased, and there were more bone marrow cell fragments. Intervention with BMSC exosomes overexpressing miR-150-3p significantly reduced the pathological damage to the femoral tissue in SONFH rats (Fig. 6B). Furthermore, BMSCs-Exos-miR-150-3p increased the levels of GPX4, SLC7A11, and Nrf2 in the femoral tissues of SONFH rats and decreased the levels of ACSL4 and BTRC (Fig. 6C). IHC staining showed that BMSCs-Exos-miR-150-3p reduced BTRC expression in the femoral tissues of SONFH rats (Fig. 6D). Moreover, MDA and Fe2+ levels were decreased by BMSCs-Exos-miR-150-3p in the femoral tissues of SONFH rats (Fig. 6E, F). Collectively, exosomal miR-150-3p from BMSCs inhibited ferroptosis of femoral tissues through the BTRC/Nrf2 axis in SONFH rats.
SONFH is a common disease in orthopedics, and the use of high-dose steroids is a common cause of femoral head necrosis. In human bones, exosomes promote bone tissue repair by transporting related bioactive substances and promoting osteogenic differentiation, bone mineralization, and angiogenesis, which lays the foundation for the application of exosomes in SONFH treatment [31]. In addition, BMSCs have a strong renewal ability and stem cell characteristics and are utilized to treat various diseases, including SONFH [32, 33]. Notably, miRNAs in exosomes can serve as potential diagnostic biomarkers in SONFH [34, 35]. In this study, we found that Gfi1 downregulation in BMSCs promoted exosomal miR-150-3p expression, which inhibited DEX-induced ferroptosis in osteoblasts by regulating the BTRC/Nrf2 axis (Graphical Abstract), revealing the significant role of BMSCs-Exos-miR-150-3p in preventing SONFH.
Ferroptosis is a type of non-apoptotic programmed cell death triggered by metabolic stress in an iron-dependent manner [36]. Glucocorticoids cause ROS production, iron metabolism disorders, and lipid oxidation reactions that are closely related to ferroptosis [37, 38]. Another study reported that DEX induces ferroptosis through the P53/SLC7A11/GPX4 pathway in SONFH [39]. Similarly, we found that DEX promoted ferroptosis and lipid peroxidation in rat primary osteoblasts.
In addition, miRNAs are involved in the regulation of ferroptosis [40]. For example, the knockdown of the long non-coding RNA ZFAS1 suppresses ferroptosis by increasing miR-150-5p expression and protects against diabetic cardiomyopathy [41]. Here, we discovered that BMSCs-Exos-miR-150-3p inhibited steroid-induced osteoblast ferroptosis and lipid peroxidation damage in vitro and in vivo. Consistent with our results, previous studies have also demonstrated that BMSC-derived exosomal miR-150 suppresses TNF-α-induced osteoblast apoptosis in ONFH [30]. Moreover, BMSCs-Exos-miR-150-3p promote osteoblast proliferation and differentiation during osteoporosis [42]. Therefore, the data from our research confirm the protective function of BMSCs-Exos-miR-150-3p in SONFH.
Gfi1, as a transcriptional repressor, is crucial for regulating cell fate, differentiation, and survival [43]. In addition, Gfi1 has a regulatory effect on miRNAs. For instance, Gfi1 controls myelopoiesis by downregulating miR-21 and miR-196b [22]. However, whether Gfi1 regulates miR-150-3p expression remains unclear. Here, we verified that miR-150-3p is a novel target of Gfi1 and that it negatively regulates miR-150-3p expression in BMSCs. As mentioned previously, Gfi1 expressed in BMSCs plays an important role in the suppression of multiple myeloma-induced osteogenic differentiation suppression [20]. However, the effect of Gfi1 on BMSC viability of BMSCs is unclear. We also demonstrated that Gfi1 silencing in BMSCs increases their viability. These findings suggested that Gfi1 acts upstream of miR-150-3p. Gfi1 knockdown increased cell viability, promoted the transcriptional expression of miR-150-3p, and ultimately upregulated BMSCs-Exos-miR-150-3p level.
MiRNAs regulate the transcription of target mRNAs during several physiological and pathological processes. For instance, miR-27a-3p overexpression in the liver enhances cancer cell proliferation by inhibiting USP46 expression [44]. These results indicate that miR-150-3p directly targets BTRC and reduces its expression in osteoblasts. BTRC, an E3 ubiquitin ligase, exerts regulatory functions in various biological processes by ubiquitinating multiple substrate proteins, including Nrf2 [25]. In addition, Nrf2 activation inhibits ferroptosis. For example, both in vivo and in vitro, Maresin1 exhibits the capacity to activate the Nrf2 pathway, thereby mitigating HG-triggered ferroptosis [45]. Melatonin reduces ferroptosis and enhances the osteogenic potential of mouse embryonic osteoblasts by activating the Nrf2/HO-1 pathway [46]. Moreover, miR-3175 inhibited DEX-induced oxidative damage in human osteoblasts by targeting DCAF1 and activating the Nrf2 signaling pathway [28]. The results of our study indicate that BTRC binds to Nrf2 and negatively regulates its expression. Functionally, the protective effects of BMSCs-Exo-miR-150-3p against DEX-induced ferroptosis in osteoblasts depend on BTRC inhibition and Nrf2 activation.
In summary, our study identified that BMSCs-Exo-miR-150-3p could suppress steroid-induced osteoblast ferroptosis in vitro and in vivo by the BTRC/Nrf2 pathway. However, no direct observations of osteoblast ferroptosis were made in this study. In future studies, we will use electron microscopy to more intuitively detect the effects of exosomal miR-150-3p on DEX-induced ferroptosis in osteoblasts. Our results supported that Gfi1 negatively regulated miR-150-3p expression by directly targeting the miR-150-3p promoter in BMSCs. Thus, this study provides a novel method for utilizing BMSCs-Exo-miR-150-3p for SONFH treatment.
The Animal Research Ethics Committee of The Second Xiangya Hospital, Central South University approved all experimental procedures (Approval Number: 2022028).
Consent for publicationN/A.
Availability of data and materialsThe datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.
Competing interestsThe authors declare that there is no conflict of interest.
FundingThis work was supported in part by the General Project of Hunan Natural Science Foundation (2023JJ30806).
Authors’ contributionsLiwen Zheng: Conceptualization; Methodology; Writing - Original Draft; Funding acquisition; Project administration;
Changjie Zhang: Formal analysis; Supervision; Investigation;
Lele Liao: Data Curation; Writing - Review & Editing
Zhijie Hai: Resources;
Xin Luo: Visualization;
Haoliang Xia: Validation
acyl-CoA synthetase long-chain family member4
ALPalkaline phosphatase
ARSalizarin red staining
ANOVAanalysis of variance
BTRCbeta-transducin repeat containing E3 ubiquitin protein ligase
BMSCsbone marrow mesenchymal stem cells
BSAbovine serum albumin
ChIPchromatin immunoprecipitation
Co-IPco-immunoprecipitation
DEXdexamethasone
DMEMdulbecco’s modified Eagle medium
EDTAethylenediaminetetraacetic acid
Exosexosomes
FBSfetal bovine serum
FITCfluorescein isothiocyanate
GPX4glutathione peroxidase 4
Gfi1growth factor independence 1
HSP70heat shock protein 70
H&Ehematoxylin and eosin
IHCimmunohistochemistry
MDAmalondialdehyde
MPmethylprednisolone
miRNAsmicroRNAs
NTAnanoparticle tracking analysis
Nrf2nuclear factor erythroid 2-related factor 2
ONFHosteonecrosis of the femoral head
PFAparaformaldehyde
PBSphosphate-buffered saline
PBSTphosphate-buffered saline with Tween
PIpropidium iodide
RIPRNA immunoprecipitation
SLC7A11solute carrier family 7 member 11
SDstandard deviation
SONFHsteroid-induced osteonecrosis of the femoral head
TSG101tumor susceptibility gene 101
