Ubiquitin-specific protease 53 promotes osteogenic differentiation of human bone marrow-derived mesenchymal stem cells

Abstract

The ubiquitin protease pathway plays important role in human bone marrow-derived mesenchymal stem cell (hBMSC) differentiation, including osteogenesis. However, the function of deubiquitinating enzymes in osteogenic differentiation of hBMSCs remains poorly understood. In this study, we aimed to investigate the role of ubiquitin-specific protease 53 (USP53) in the osteogenic differentiation of hBMSCs. Based on re-analysis of the Gene Expression Omnibus database, USP53 was selected as a positive regulator of osteogenic differentiation in hBMSCs. Overexpression of USP53 by lentivirus enhanced osteogenesis in hBMSCs, whereas knockdown of USP53 by lentivirus inhibited osteogenesis in hBMSCs. In addition, USP53 overexpression increased the level of active β-catenin and enhanced the osteogenic differentiation of hBMSCs. This effect was reversed by the Wnt/β-catenin inhibitor DKK1. Mass spectrometry showed that USP53 interacted with F-box only protein 31 (FBXO31) to promote proteasomal degradation of β-catenin. Inhibition of the osteogenic differentiation of hBMSCs by FBXO31 was partially rescued by USP53 overexpression. Animal studies showed that hBMSCs with USP53 overexpression significantly promoted bone regeneration in mice with calvarial defects. These results suggested that USP53 may be a target for gene therapy for bone regeneration.

Fig. 1. Upregulation of USP53 during the osteogenic differentiation of hBMSCs.

a Relative mRNA expression levels of USP53 during the osteogenic differentiation of hBMSCs. HPRT was used for normalization. b Protein levels of USP53 during the osteogenic differentiation of hBMSCs. Heat shock protein 90 (HSP90) was used as a loading control. USP53 protein expression was quantified using ImageJ. Day 0: undifferentiated hBMSCs. Results are means ± SDs. ns: not significant, *P < 0.05, ***P < 0.001 by one-way ANOVA. n = 3.

Fig. 2. USP53 knockdown inhibits the osteogenic differentiation of hBMSCs in vitro.

a Alkaline phosphatase staining was performed in shMock- or shUSP53-1-, 2-infected hBMSCs in osteogenic medium for 3 days. Scale bars, 200 μm. b Quantification of ALP activity. c Alizarin red S staining was performed in shMock- or shUSP53-1-, 2-infected hBMSCs in osteogenic medium for 12 days. Scale bars, 200 μm. d Quantification of Alizarin red S staining. e qRT-PCR analysis of osteogenesis-related genes on day 3 after osteogenic differentiation. f Immunoblot analysis of osteogenesis-related genes on day 5 after osteogenic differentiation. Data were quantified using ImageJ. Results are means ± SDs. ns: not significant, **P < 0.01, ***P < 0.001, ****P < 0.0001 by one-way ANOVA. n = 3.

Fig. 3. USP53 overexpression promotes the osteogenic differentiation of hBMSCs in vitro.

a Alkaline phosphatase staining was performed in control or USP53-infected hBMSCs in osteogenic medium for 3 days. Scale bars, 200 μm. b Quantification of ALP activity. c Alizarin red S staining was performed in osteogenic medium for 12 days. Scale bars, 200 μm. d Quantification of Alizarin red S staining. e qRT-PCR analysis of osteogenesis-related genes on day 3 after osteogenic differentiation. f Immunoblot analysis of osteogenesis-related genes on day 5 after osteogenic differentiation. Data were quantified using ImageJ. Results are means ± SDs. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 by unpaired two-tailed Student’s t-test. n = 3.

Fig. 4. USP53 regulates osteogenic differentiation in hBMSCs through Wnt/β-catenin signaling.

a, b Expression levels of Wnt/β-catenin signaling pathway mediators were investigated by immunoblotting in USP53-overexpressing (a) or -knockdown (b) hBMSCs in osteogenic medium for 7 days. Data were quantified using ImageJ. c Immunoblot analysis of Wnt3a-induced osteogenic marker expression. Data were quantified using ImageJ. d Immunoblot analysis of DKK1-induced osteogenic marker expression. Data were quantified using ImageJ. e Immunoblot analysis of β-catenin-linked polyubiquitination with overexpression of USP53. β-Catenin ubiquitination (top panel) and protein expression assays (bottom panel) were evaluated. f Immunoblot analysis of β-catenin-linked polyubiquitination with siUSP53. β-Catenin ubiquitination (top panel) and protein expression assays (bottom panel) were evaluated. g hBMSCs were transfected with Flag-β-catenin overexpression plasmid and control or Flag-USP53 and then treated with cycloheximide (100 μg/mL) and Wnt3a. Immunoblots with active β-catenin protein at the indicated time points. n = 3.

Fig. 5. Identification of USP53 binding proteins and effects of FBXO31 on β-catenin degradation through the SCF complex.

a Instantblue staining of a co-IP mixture separated by SDS-PAGE. The indicated band was extracted for analysis. b hBMSCs were cotransfected with FLAG-β-catenin with either myc-FBXO31 or myc-FBXO31ΔF for 48 h. Transfected cells were incubated with 10 μM MG132 for 6 h, and whole cells were lysed and subjected to IP with anti-myc antibodies. Immunoprecipitates and input protein extracts (Pre-IP) were resolved in SDS-PAGE. c hBMSCs were transfected with pCMV-myc or myc-FBXO31 for 48 h, and whole cell lysates were immunoblotted. d hBMSCs were transfected with pCMV-myc or myc-FBXO31ΔF for 48 h, and whole-cell lysates were immunoblotted. e hBMSCs were transfected with the pCMV-myc or myc-FBXO31 for 48 h. Transfected cells were then incubated with or without 10 μM MG132 for 6 h. Cell lysates were immunoblotted. f hBMSCs were transfected with pCMV-myc, myc-FBXO31, and myc-FBXO31ΔF for 48 h. Cells were harvested and lysed, and whole-cell protein extracts were immunoblotted. g β-Catenin transcriptional activity was measured on day 5 after induction of osteogenesis by TOP/FOP luciferase assays. h HEK293 cells were transfected with the indicated plasmids for 48 h, treated with 10 μM MG132 for 6 h, lysed, subjected to IP with anti-HA antibodies, and immunoblotted. i hBMSCs were transfected with negative control or FBXO31 siRNA for 5 days in osteogenic induction medium. Immunoblot analysis was performed, and data were quantified using ImageJ. j hBMSCs were transfected with empty vector, FBXO31, or GFP-USP53 and then cultured with osteogenic induction medium for 5 days. Immunoblot analysis was performed, and data were quantified using ImageJ. Results are means ± SDs. ns: not significant, **P < 0.01 by one-way ANOVA. n = 3.

Fig. 6. hBMSCs with AAV2-USP53 improved bone formation in vivo.

a Experimental design of the mouse calvarial defect model. b Critical-size calvarial defects (5 mm in diameter) in mice were treated with AAV2-control-infected hBMSCs or AAV2-USP53-infected hBMSCs in fibrin matrix with PBS. Eight weeks after surgery, bone regeneration was measured by μCT. c Relative quantification of μCT analysis. d Histomorphometric analysis of calvarial defects in mice. Arrows indicate the distance between double calcein labeling. Scale bars, 20 μm. Relative histomorphometric quantification of the mineral apposition rate (MAR) is shown (right). e Representative images of M&T staining of calvarial bone sections in mice. Scale bars, 200 μm. *New bone. f Immunohistochemistry analysis using an antibody against human vimentin and IgG control of calvarial bone sections in mice. Scale bars, 20 μm. g Immunohistochemistry analysis of USP53 (phycoerythrin [PE]; red fluorescence) and OCN (fluorescein isothiocyanate [FITC]; green fluorescence) with DAPI counterstaining of the calvarial defects in mice (left). Quantification of IHC analysis (right). Scale bars, 50 μm. Results are presented as means ± SDs. ns: not significant, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 by one-way ANOVA. n = 5 per group.

Fig. 7. USP53 levels decrease in an osteoporosis mouse model.

a Sham or ovariectomy (OVX) surgery was performed on 6-week-old female mice. Two months later, femoral trabecular bone mass was assessed by μCT. Scale bars, 50 μm. b Serum levels of the c-terminal telopeptide (CTX)-1 in sham and OVX mice. c Immunohistochemistry analysis of USP53 (phycoerythrin [PE]; red fluorescence) and OCN (fluorescein isothiocyanate [FITC]; green fluorescence) with DAPI counterstaining in sham-operated and OVX mice (left). Quantification of IHC analysis (right). Scale bars, 20 μm. Results are means ± SDs. **P < 0.01 by an unpaired two-tailed Student’s t-test. n = 5–10 per group. d Diagram showing the molecular mechanism through which FBXO31 and USP53 regulate the Wnt signaling mediator, β-catenin, during the osteogenic differentiation of hBMSCs.