GLUL mediates FOXO3 O-GlcNAcylation to regulate the osteogenic differentiation of BMSCs and senile osteoporosis

Abstract

The abnormal osteogenic differentiation of bone marrow mesenchymal stem cells (BMSCs) is an important cause of senile osteoporosis (SOP). Glutamine synthetase (GLUL) is a key enzyme in glutamine biosynthesis; however, its functional role in SOP remains unclear. Here, we found that GLUL expression was downregulated in the BMSCs of SOP patients. Mice with BMSC-specific Glul-knockout (KO) exhibited dysplasia of the skull and phalanges and osteoporosis due to disordered osteogenic differentiation. Mechanistically, GLUL competitively bound to the Tripartite Motif Containing 25 (TRIM25) SPRY subunit, reduced the ubiquitin-mediated degradation of UDP-N-acetylglucosamine pyrophosphorylase 1 (UAP1) and increased the synthesis of uridine 5-diphosphate N-acetylglucosamine (UDP-GlcNAc), thereby regulating the O-linked β-N-acetylglucosamine modification (O-GlcNAcylation) of serine 296 residues and increasing Forkhead Box O3 (FOXO3) stability to reduce oxidative stress. Moreover, blocking the O-GlcNAcylation of FOXO3 at Ser296 inhibited osteogenic differentiation. Finally, GLUL supplementation specifically in BMSCs slowed bone loss in SOP model mice. Overall, our study suggests that GLUL plays an important role in regulating osteogenic differentiation and bone development, which may have implications for SOP treatment. Schematic illustration of the molecular mechanism by which GLUL mediates FOXO3 O-GlcNAcylation to regulate the osteogenic differentiation of BMSCs and senile osteoporosis. The graphical abstract was created by figdraw2.0.

Schematic illustration of the molecular mechanism by which GLUL mediates FOXO3 O-GlcNAcylation to regulate the osteogenic differentiation of BMSCs and senile osteoporosis. The graphical abstract was created by figdraw2.0.

Fig. 1. GLUL plays an important role in the osteogenic differentiation of BMSCs and the progression of SOP.

A Heatmap of the metabolic and transcriptional changes that occur during the osteogenic differentiation of human mesenchymal stem cells according to the GSE191136 dataset. B Western blotting analysis of GLUL expression during the osteogenic differentiation of BMSCs. C Immunohistochemical staining analysis of GLUL expression in mouse femur sections at different times after birth. D Heatmap of the effects of SOP on human mesenchymal stem cells according to the GSE35956 dataset. E Immunohistochemical staining analysis of GLUL expression in femur sections from young and old samp6 mice (n = 3). F Quantification of GLUL expression in bone samples from patients with SOP. G Western blotting analysis of GLUL expression in BMSCs from patients in the SOP group (n  = 6) and the control group (n = 7). ** P < 0.01 vs. other groups; *P < 0.05 vs. other groups.

Fig. 2. Mesenchymal stem cell-specific Glul deficiency leads to abnormal bone formation.

A Construction strategies for BMSC-specific Glul-KO (Glul^Prx1) mice. B Representative images of Alizarin Red-Alcian Blue double staining of tissues from Glul^Prx1 and Glul^fl/fl mice(n = 5). C Representative images of Alizarin Red-Alcian Blue double staining of the skull, upper limb, and lower limb from Glul^Prx1 and Glul^fl/fl mice (n = 5). D Quantitative analysis of Alizarin Red-Alcian Blue double staining. E Representative images of ARS and ALP staining of BMSCs from Glul^Prx1 and Glul^fl/fl mice. F Quantitative analysis of ARS and ALP staining. G Representative images of ARS and ALP staining of BMSCs from Glul^Prx1 and Glul^fl/fl mice after treatment with glutamine. H Quantitative analysis of ARS and ALP staining. I Representative micro-CT images of trabecular bone from the femoral metaphysis of Glul^Prx1 and Glul^fl/fl mice (n = 5). J Quantitative analysis of cancellous bone volume (BV/TV, %), trabecular thickness (Tb.Th). K Quantitative analysis of trabecular number (Tb.N), and trabecular separation (Tb.Sp). L Representative images of VON KOSSA staining (n = 3). M Calcein double-label staining image. N Quantitative analysis of the MAR and BFR/BS in Glul^Prx1 and Glul^fl/fl mice (n = 3). **P < 0.01 vs. other groups; *P < 0.05 vs. other groups.

Fig. 3. GLUL regulates HBP metabolism and affects UDP-GlcNAc generation.

A Heatmap of metabolomic assays of the BMSCs from Glul^Prx1 and Glul^fl/fl mice (n = 6). B Schematic diagram of the hexosamine biosynthetic pathway. C Heatmap of the proteomics analysis of BMSCs from Glul^Prx1 and Glul^fl/fl mice (n = 3). D Western blotting analysis of O-GlcNAcylation levels in BMSCs from Glul^Prx1 and Glul^fl/fl mice at different osteogenic differentiation timepoints. The numbers on the line graph represent the P -values comparing the relative protein expression levels between the two groups. E Western blotting analysis of UAP1 expression in BMSCs from Glul^Prx1 and Glul^fl/fl mice at different osteogenic differentiation timepoints. The numbers on the line graph represent the P -values comparing the relative UAP1 expression levels between the two groups. F Western blotting analysis of UAP1 expression in BMSCs from patients in the SOP group (n = 6) and the control group (n = 7). G Western blotting analysis of OGT levels in BMSCs from Glul^Prx1 and Glul^fl/fl mice at different osteogenic differentiation timepoints. H Representative images of ARS and ALP staining of BMSCs from Glul^Prx1 and Glul^fl/fl mice after transfection with the UAP1 overexpression or control plasmid. **P < 0.01 vs. other groups; *P < 0.05 vs. other groups.

Fig. 4. GLUL competently binds to the TRIM25-SPRY subunit and UAP1.

A Molecular docking image of the TRIM25 protein and the GLUL/UAP1 protein. B The results of the Biacore assay to assess the binding strength of GLUL/UAP1 to TRIM25. C Representative co-IP images of TRIM25 with GLUL/UAP1 in HEK-293T cells. D Representative images of bimolecular fluorescence complementation experiments between TRIM25 carrying YFP (aa 1–154) and GLUL/UAP1 carrying YFP (aa 155–238). E Representative images of live-cell imaging after cotransfection of GFP-TRIM25, RFP-GLUL, and BFP-UAP1 into HEK293T cells. F Representative images of immunofluorescence staining for TRIM25, GLUL and UAP1 in BMSCs. G Schematic diagram of TRIM25 truncation. H Representative images of co-IP experiments conducted with Myc-UAP1, HA-GLUL, and Flag-TRIM25 full-length and truncated samples. **P < 0.01 vs. other groups; *P < 0.05 vs. other groups.

Fig. 5. GLUL reduces the K48 ubiquitin connectivity level of TRM25 to UAP1.

A Representative Western blotting images of HEK-293T cells cotransfected with the TRIM25 overexpression vector and UAP1 in a dose-dependent manner. B Representative image of a Western blotting experiment showing UAP1 expression levels after cell transfection with si-TRIM25. C Representative image of a Western blotting experiment showing UAP1 expression levels after cell transfection with si-TRIM25. D Representative images of Western blotting analysis of proteins that were extracted from HEK293T cells that were transfected with the Myc-UAP1 and Flag-TRIM25 overexpression plasmids and treated with MG132. E Representative images of Western blotting analysis of proteins that were HEK293T cells that were transfected with the Myc-UAP1 and Flag-TRIM25 overexpression plasmids and treated with MG132 and chloroquine. F A Myc-UAP1 overexpression plasmid was transfected into BMSCs, and osteogenic differentiation was induced. Proteins were extracted from the cells at different timepoints, and representative images were obtained after co-IP experiments with anti-UAP1 antibodies. G The V5 Ub overexpression plasmid was transfected into HEK293T cells, which were then cotransfected with Flag-TRIM25, Flag-TRIM25+Myc-UAP1 + HA-GLUL, or Myc-UAP1 + HA-GLUL overexpression plasmids. Subsequent Western blot images and quantitative analysis revealed the  K48-linked ubiquitination of UAP1 in different cell groups. **P < 0.01 vs. other groups; *P < 0.05 vs. other groups.

Fig. 6. O-GlcNAcylation of FOXO3-Ser296, which is regulated by GLUL, promotes the osteogenic differentiation of BMSCs.

A Volcano plots of BMSCs from Glul^Prx1 (n = 3) and Glul^fl/fl (n = 3) mice subjected to 4D DIA O-GlcNAcylation profiling analysis. B Mass spectrometry image of the O-GlcNAcylation modification of the FOXO3 Ser296 residue. C The results of the co-IP experiment on the effect of Glul knockout on the level of O-GlcNAcylation modification of FOXO3 in BMSCs. D Representative images of target protein decay at different time points in BMSCs from Glul^Prx1 and Glul^fl/fl mice. E Representative images of target protein attenuation at different time points were obtained by transfecting wild-type or S296V-mutant FOXO3 plasmids into Foxo3-KO BMSCs. F Representative images of ARS and ALP staining of Foxo3-KO BMSCs after transfection with wild-type or S296V-mutant FOXO3 plasmids. G Quantitative analysis of ARS and ALP staining. H Representative images of FOXO3-Ser297 O-GlcNAcylation levels as determined by Western blotting analysis of BMSCs from patients in the SOP and control groups. I Western blotting analysis of the expression of osteogenesis-related genes in Foxo3-KO BMSCs after transfection with wild-type or S296V-mutant FOXO3 plasmids. J Representative images of ROS detection. **P < 0.01 vs. other groups; *P < 0.05 vs. other groups.

Fig. 7. Supplementation of GLUL can alleviate bone loss in SOP model mice.

A Schematic diagram of in vivo experiment. B Representative micro-CT images of trabecular bone from the femoral metaphysis of mice(n = 4). C Quantitative analysis of BV/TV, Tb.Th, Tb.N and Tb.Sp. D ELISA detection results of serum P1NP levels in each group of mice (n = 4). E Immunohistochemical staining of osteogenesis-related genes in mice femur sections. (n = 3). F Calcein double-label staining image and quantitative analysis of the MAR and BFR/BS in each group of mice (n = 3). G Analysis of Osteoblast Density (N.O/BS) and Osteoblast Surface Area (O.S/BS) Data. H Representative images of ARS and ALP staining of BMSCs from mice of each group. I Quantitative analysis of ARS and ALP staining. J Representative images of VON KOSSA staining (n = 3). K Western blot analysis of the expression of osteogenesis-related genes in BMSCs from mice of each group. L Quantitative data analysis of Western blot results. **P < 0.01 vs. other groups; *P < 0.05 vs. other groups.