Long noncoding RNA Gm31629 promotes bone regeneration by maintaining bone marrow mesenchymal stem cells activity

Animals and treaments

Gm31629 knockout (Gm31629-KO) mice were obtained from Cyagen Biosciences as previously reported (Xiao et al., 2020). We purchased 3-month-old and 12-month-old C57BL/6J male mice from the Laboratory Animal Center of Central South University (Changsha, China). The model of bone regeneration was established as described before (Chen et al., 2019; Fukuda et al., 2013; Yang et al., 2020). Briefly, after anesthesia, the anterior medial approach was used to expose the distal femoral. Then, a 25-guage needle was used to drill a hole at the distal femur along the long axis of the femur and a 0.6 mm diameter Kirschner wire was used to ablate trabecular bone of distal femur. This was minimally invasive injury and we made great effort to reduce the sufferings of the mice. One week after the injury, the mice were euthanized by cervical dislocation after anesthesia to collect the bone samples.

For TF2A administration, mice were treated with TF2A or vehicle by gavage at a dosage of 8 mg/kg every day for three weeks before the establishment of bone regeneration model. After that, TF2A treatment continued for one week before the mice were euthanized. All the mice in this study were healthy and C57BL/6 background and kept in the Experimental Animal Research Center of Central South University with specific pathogen-free standard. The mice were housed in individual ventilated cage with six mice per cage. The mice were kept in room temperature with 12 h light-dark cycle and had free access to food and water. No animal was excluded from the experiments. Xiangya Hospital of Central South University Ethics Committee (Changsha, Hunan, China) approved this research (2019030350). All animal experiments conformed to all ethical requirements relating to animal research.

BMSCs isolation, culture and senescence assays

BMSCs were isolated as previously described (Li et al., 2015; Yu et al., 2018). The isolated BMSCs were cultured with α-MEM supplemented with 15% FBS, 100 U/mL penicillin and 100 µg/mL streptomycin in a humidified atmosphere of 5% CO2 at 37 °C to reach 80% confluence. Then the first-passage BMSCs were harvested and seeded in culture dishes for enrichment of cell populations. When the second-passage reach confluence after 1–2 week, they were subcultured. Only third-passage BMSCs were applied to perform further study unless specified otherwise.

The senescent BMSCs were stained by a senescence β-galactosidase staining Kit (Solarbio Science & Technology) according to the manufacturer’s instructions. Briefly, after washing with PBS, the cells were fixed with 4% paraformaldehyde for 15 min at room temperature. Then the cells were stained with working solution overnight at 37 °C. Five different fields were randomly selected under a microscope to count the SA-βGal-positive (blue cells) and the percentage of SA-βGal-positive were calculated.

Cell transfection

For Gm31629 overexpression, the adenovirus particles expressing mouse Gm31629 were purchased from OBiO Technology Corporation (Shanghai, China). For Yb-1 overexpression, pcDNA3.1-mYb-1 was purchased from Sino Biological Inc (Beijing, China). PcDNA3.1-mYb-1 and negative control were transfected into BMCSs with lipofectamine 2000 (Invitrogen, Thermo Fisher Scientific, Waltham, MA) by a standard method.

Wound healing assay

BMSCs were seeded in 6-well plates at a density of 1 ×10⁶ cells per well for each group. A linear wound was made using a sterile 200 µl pipette tip to scratch across the confluent cell layer. Images of wound healing were observed at 0 h and 24 h and the migration rate was calculated using ImageJ software (National Institutes of Health, Bethesda, MD, USA).

CCK8 assay

CCK8 assay was used to evaluate the growth of BMSCs according to the manufacture’s protocol (MedChemExpress, LLC). BMSCs were seeded in 96-well plates at a density of 5,000 cells per well for each group. Then, we added 10 µl of CCK-8 solution into each well and incubated the plate at 37 °C for 2 h. At last, the OD value of each well was examined at 450 nm using a spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA).

Brdu staining assay

Brdu staining assay was conducted using standard methods. Briefly, BMSCs were seeded in 24-well plates at a density of 1 × 10⁵ cells per well for each group and incubated with 10 µM Brdu labeling solution for 24 h in a cell incubator. After that, the cells were fixed with 4% paraformaldehyde for 20 min. After permeabilization using 0.2% triton, the cells were incubated with 3% BSA for blocking. Then, the cells were incubated with the primary (Cell Signaling Technology, Danvers, MA; 5292, 1:400) and secondary antibody (Invitrogen, Thermo Fisher Scientific, Waltham, MA; A21202, 1:500), and counterstained DAPI.

Osteogenic differentiation assay

BMSCs were seeded in 6-well plates at a density of 5 × 10⁵ cells per well for each group and cultured with osteogenic induction conditional medium (10 mM β-glycerol phosphate, 50 µM ascorbate-2-phosphate, and 0.1 µM dexamethasone) for three weeks. We changed the osteogenic medium every other day. To assess the mineralization of cell matrix, 2% Alizarin Red S (Cyagen Biosciences, Santa Clara, CA) was used to stain the cell matrix. Alizarin Red S was destained with cetyl-pyridinium chloride solution and the OD value was quantified by spectrophotometry at 562 nm.

Osteoclast differentiation assay

We perform osteoclasts differentiation assay as described before (Yang et al., 2020). Briefly, bone marrow was flushed out of bone marrow cavity of mice. Isolated bone marrow cells were cultured with complete media for 14 h. Then the unattached cells were collected and treated with α-MEM containing 10% FBS, 30 ng/mL M-CSF (R&D Systems Inc., Minneapolis, MN), 100 µg/mL streptomycin, and 100 U/mL penicillin for 72 h to gain pure monocytes and macrophages. After that, the monocytes and macrophages were cultured with osteoclastic induction medium (30 ng/mL M-CSF, 60 ng/mL RANKL) for 1 week. Osteoclasts were stained with TRAP staining kit (Sigma-Aldrich, St Louis, MO) according to manufacturer’s instructions.

RT-qPCR analysis

Extraction of total RNA was performed with Trizol (Invitrogen, Thermo Fisher Scientific, Waltham, MA) following standard methods and reverse transcription was conducted using 1 µg total RNA. RT-qPCR was conducted in duplicate using SYBR Premix Ex Taq II (Takara). We normalized the Ct value of Gm31629 to that of Gapdh and calculated ΔCt value (ΔCt=Ct_(Gm31629)- Ct_(Gapdh)) for both 3-moth-old and 12-moth-old group. Then we normalized the ΔCt value of 12-moth-old group to the ΔCt value of 3-moth-old group and calculated ΔΔCt value (ΔΔCt = Δ Ct_{(12-moth-old)}−Δ Ct_(3-moth-old)). The relative gene expression was calculated using the 2^−ΔΔCT method. All experiments were repeated three times. The primer sequences are listed in Table S1.

Western blot

Western blotting was performed as previously described (Liu et al., 2021b; Peng et al., 2019). Total cell proteins were separated by SDS-PAGE and blotted on PVDF membranes (Millipore, Sigma, Burlington, MA). After blocking with 5% milk, the membranes were incubated with specific antibodies to YB-1 (Cell Signaling Technology, Danvers, MA; 4202, 1:1000), p16^INK4A (Sigma-Aldrich, St Louis, MO, SAB4500072, 1:1000) and GAPDH (Proteintech, Rosemont, IL, USA; 10494-1-AP, 1:5000). Blots were visualized using an ECL Kit (Thermo Fisher Scientific, Waltham, MA; 32,106).

RNA pull-down assay and RNA immunoprecipitation

We performed RNA pull down assay as previously reported (Xiao et al., 2020). Briefly, biotin-labeled full-length Gm31629 and antisense Gm31629 were incubated with nuclear lysate of BMSCs for 1 h. After that, the streptavidin agarose beads (Invitrogen, Thermo Fisher Scientific, Waltham, MA) were added and incubated at 25 °C for another 1 h. After washing with cold NT2 buffer, the pulled-down proteins were used for western blot analysis. RNA immunoprecipitation was performed using a Magna RIP RNA-Binding Protein Immunoprecipitation Kit (Millipore, Sigma, Burlington, MA; 17-700) following the manufacturer’s instructions. The precipitated RNA was extracted, reversed transcribed and analyzed by RT-qPCR. All experiments were repeated three times. The primer sequences are listed in Table S1.

Chromatin immunoprecipitation (ChIP) assay

ChIP assay was performed with SimpleChip Kit (9003; Cell Signaling Technology, Danvers, MA) following the manufacturer’s instructions as previously described (Yang et al., 2019). Briefly, chromatin was crossed-linked (1 % formaldehyde, 10 min) and sheared to 100- to 500-bp fragments by sonication. The relevant protein-DNA complex was immunoprecipitated by YB-1 antibody (Santa Cruz Biotechnology, Dallas, TX; sc-398146) or IgG control. The ChIP DNA was used to perform standard PCR or RT-qPCR. All experiments were repeated three times. The primer sequences are listed in Table S1.

µCT analysis

µCT scanning was performed using a high-resolution micro-CT (SCANCO Medical AG, VIVACT 80; Wangen-Brüttisellen, Switzerland) with a resolution of 12 µm per pixel at 55 kV and 145 µA. We reconstructed a 3D model and analyzed the structure indices as previous reported (Li et al., 2021; Yang et al., 2019; Yang et al., 2020). Trabecular bone volume (Tb. BV/TV) in the regeneration region was calculated.

Histochemistry and immunohistochemistry

Histochemical and immunohistochemical staining were conducted as previously described (Cai et al., 2022; Yang et al., 2020). Femora were collected and fixed with 4% paraformaldehyde for 1 day at 4 °C. Then we decalcified the bones with 10% EDTA and embedded them in paraffin. For histochemistry, 4 micrometer-thick slides were subjected to HE and TRAP staining according to a standard protocol. For immunocytochemistry, after antigen retrieval the samples were incubated with primary antibodies against osteocalcin (Takara M173) at 4 °C overnight and Horseradish peroxidase-streptavidin detection system (Dako Agilent, Santa Clara, CA) was used to detect immuno-activity.

Statistics analysis

Statistical analysis was performed using GraphPad Prism software 8.0. Data are expressed as the mean ± standard deviation (sd). Unpaired Student’s t test was applied to compare two groups. One-way ANOVA was employed while comparing multiple groups. The difference was considered to be statistically significant at p < 0.05. In order to avoid the type II error, we used G*Power 3.1 to perform the statistical power analysis and the minimum power required in this study was set at 0.8. All the samples were randomly assigned and no blinding was used.

BMSCs undergo senescence during aging with reduced ability of bone regeneration

To study the characteristics of senescent BMSCs, we compared the function and phenotype of BMSCs from middle-aged (12-month old) mice with that from young mice (3-month old) in vitro. The Brdu staining assay and CCK8 assay revealed that the proliferation ability of BMSCs from middle-aged mice was significantly reduced in comparison with BMSCs from young mice (Figs. 1A–1C). The wound healing assay indicated that the migration ability of BMSCs from middle-aged mice was markedly declined in comparison with BMSCs from young mice (Figs. 1D, 1E). As expected, there were more SA-βGal-positive BMSCs in the middle-aged group than in the young groups (Figs. 1F, 1G). BMSCs from middle-aged mice also showed decreased osteogenic differentiation ability compared to that of young mice (Figs. 1H, 1I). These results indicated that an aging phenotype of BMSCs presented in the middle-aged mice, resulting in significant impaired function of BMSCs.

To investigate the change of bone regeneration ability during aging, a bone regeneration model was established by surgical ablation of trabecular bone in distal femur. As expected, the bone volume in the regeneration area of middle-aged mice was lower than that of young mice at 7 days after ablation (Figs. 2A–2D). The number of osteocalcin positive (ocn⁺) osteoblasts in the bone regeneration area of middle-aged mice was less than that of young mice at 7 days after ablation (Figs. 2E, 2F). The number of tartrate-resistant acid phosphatase positive (TRAP⁺) osteoclasts in the bone regeneration area of middle-aged mice was also less than that of young mice at 7 days after ablation (Figs. 2G, 2H). Altogether, these results indicated that middle-aged mice had reduced bone regeneration ability compared to young mice.

Gm31629 regulates the senescence of BMSCs

Gm31629 expression decreased significantly in BMSCs of middle-aged mice compared with that of young mice as analyzed by RT-qPCR (Fig. 3A). To study the function of Gm31629 in the regulation of BMSCs senescence, we isolated BMSCs from 3-month-old Gm31629 knockout (Gm31629-KO) mice and wild type (WT) mice. The proliferation and migration ability of BMSCs isolated from Gm31629-KO mice were markedly reduced in comparison with that of BMSCs isolated from WT mice (Figs. 3B–3F). In addition, there were more SA-βGal-positive BMSCs in Gm31629-KO group than in WT group (Figs. 3G, 3H). BMSCs of Gm31629-KO mice also showed decreased osteogenic differentiation capacity compared to that of WT group (Figs. 3I, 3J). These data revealed that BMSCs from Gm31629-KO mice showed an aging phenotype with significant impaired function of BMSCs.

Gm31629 knockout mice show impaired bone regeneration ability

To further study the function of Gm31629 in bone regeneration, we established the bone regeneration model in Gm31629-KO mice and WT mice at 3-month-old by surgical ablation of trabecular bone in distal femur (Fig. 4A). We found that the bone volume in bone regeneration region of Gm31629-KO mice was significantly lower than that of WT controls at 7 days after ablation (Figs. 4B–4D). The number of ocn⁺ osteoblasts was markedly reduced in bone regeneration region of Gm31629-KO mice in comparison with that of WT controls at 7 days after ablation (Figs. 4E, 4F). There is no significant difference of TRAP⁺ osteoclasts in bone regeneration region of Gm31629-KO mice in comparison with that of WT controls 7 days after ablation (Figs. 4G, 4H). These data showed that the bone regeneration ability of Gm31629-KO mice was lower than that of WT mice.

Gm31629 regulates BMSCs senescence through YB-1/P16^INK4A pathway

Our previous study demonstrated that Gm31629 directly interacted with YB-1 and increased the protein level of YB-1 by preventing its degradation, further reducing the expression of p16^INK4A and suppressing the senescence of htNSCs (Xiao et al., 2020). YB-1 is a DNA/RNA-binding protein (Lyabin, Eliseeva & Ovchinnikov, 2014), and has been reported to bind to the promoter region of p16^INK4A and inhibit the expression of p16^INK4A (Kotake et al., 2013; Xiao et al., 2020), a maker of cellular senescence (Ogrodnik et al., 2019; Omori et al., 2020). Furthermore, several studies have demonstrated that the expression of p16^INK4A is much higher in BMCSs of older mice than in young controls (Hu et al., 2022; Li et al., 2017). To verify the interaction between YB-1 and Gm31629 in BMSCs, the RNA pull-down assay was repeated and the binding of Gm31629 to YB-1 in BMSCs was confirmed (Fig. 5A). RNA immunoprecipitation assay further confirmed that Gm31629 could bind to YB-1 in BMSCs (Fig. 5B). In addition, Gm31629 knockout markedly reduced YB-1 protein level and Gm31629 overexpression significantly increased YB-1 protein level (Figs. 5C, 5D). To confirm that Gm31629 increases YB-1 protein level by preventing the degradation of YB-1 in BMSCs, we inhibited protein synthesis in BMSCs with cycloheximide (CHX) and found that Gm31629 knockout accelerated the degradation of YB-1 (Fig. 5E). These results indicated that Gm31629 could prevent the degradation of YB-1 in BMSCs. Moreover, ChIP-PCR assays showed that YB-1 could directly bind to the promoter of p16^INK4A in BMSCs (Fig. 5F, 5G). Then, we observed that Gm31629 knockout not only reduced YB-1 protein level, but also increased the expression of p16^INK4A (Fig. 5H). However, the overexpression of YB-1 in Gm31629 knockout BMSCs rescued the reduced YB1 protein level and reduced the expression of p16^INK4A (Fig. 5H). Moreover, the overexpression of YB-1 rescued the increased level of senescence in BMSCs derived from Gm31629- KO mice (Figs. 5I, 5J). In addition, osteogenic differentiation assay revealed that YB-1 overexpression rescued the reduced osteogenic differentiation of BMSCs derived from Gm31629- KO mice (Figs. 5K, 5L). Thus, these results indicated Gm31629 could stabilized YB1 protein and inhibited the expression of p16^INK4A, a possible mechanism for Gm31629 in regulating the senescence of BMSCs.

TF2A treatment in vitro alleviates the senescence of BMSCs

In the previous study, we also identified a natural compound, TF2A, which mimics the ability of Gm31629 to increased YB-1 protein level, reducing the senescence of htNSCs (Xiao et al., 2020). As expected, treatment of TF2A could mimic the ability of Gm31629 to increase YB-1 protein level and reduce the expression of p16^INK4A in BMSCs (Fig. 6A). We then found that TF2A attenuated the senescence of BMSCs and promoted osteogenesis of BMSCs (Figs. 6B–6E). TF2A did not affect osteoclast differentiation, as evaluated by TRAP staining (Figs. 6F, 6G). These data demonstrated that TF2A could mimic the activity of Gm31629 to increase the protein level of YB-1, thus alleviating the senescence of BMSCs.

TF2A treatment promotes bone regeneration in middle-aged mice

To investigate whether treatment of TF2A could promote bone regeneration in middle-aged mice, 12-month-old C57BL/6J mice were orally treated with TF2A at a dosage of 8 mg/kg every day or with vehicle for three weeks. There weeks after TF2A or vehicle treatment, the mice were performed with surgical ablation of trabecular bone in the right femur, and continued the treatment with TF2A for one week (Fig. 7A). Administration of TF2A had no obvious influence on the weight of the mice (Fig. 7B). Treatment with TF2A promoted bone regeneration in 12-month-old mice in comparison with the control group (Figs. 7C–7E). The number of ocn+ osteoblasts in bone regeneration region was also increased after TF2A administration, which indicated increased bone formation (Figs. 7F, 7G). There was no significant difference of TRAP+ osteoclasts in bone regeneration region of TF2A treated group compared with that of vehicle treated group (Figs. 7H, 7I). These data indicated that TF2A could promote bone regeneration in middle-aged mice.