Fig 1: Defective TCA cycle upon GLS inhibition in BMM plays different roles between aging-impaired osteogenic differentiation and RANKL-induced osteoclast differentiation. BMSC from aged (18-month-old) male mice and BMM from 8-week-old male mice were isolated. (A) MPC overexpression via transduction of adenovirus overexpressing wildtype MPC1 and MPC2 (oeMPC) in BMSC and that in BMM were examined by western blots, representative images and quantitative analyses were shown, n = 5/group respectively. (B-C) BMSC treated with vehicle (DMSO) accompanied by empty vector (EV), vehicle accompanied by MPC overexpression, CB-839 (2 mM, the same below) accompanied by EV, and CB-839 accompanied by MPC overexpression respectively (B), or BMM with these treatments (C) were cultured in normal growth medium for up to 5 days for cell proliferation ability evaluation by CCK8 assay, n = 3/group respectively. (D-E) BMSC with the same treatments as that in 7B were cultured in osteogenic differentiation medium for 18 days for alizarin red staining (representative micrographs are shown, scale bar represents 100 mm) (D), BMM with the same treatments as that in 7C were cultured with M-CSF and RANKL for 5 days for TRAP staining (representative micrographs are shown, scale bar represents 100 mm) (E). (F-G) Quantitative analysis of alizarin red staining in 7D (F) and that of the number of osteoclasts in 7E (G), n = 5/group respectively. (H-I) BMSC treated with vehicle accompanied by EV, CB-839 accompanied by EV, and CB-839 accompanied by MPC overexpression respectively, were cultured in osteogenic differentiation medium for 14 days and then extracellular oxygen consumption rate (OCR) was assayed and calculated (H), BMM with these treatments were cultured with M-CSF and RANKL for 5 days and then OCR was assayed and calculated (I), n = 3/group respectively. (J-K) BMSC treated with scrambled shRNAs (SC) accompanied by EV, SC accompanied by MPC overexpression, shGLS1 accompanied by EV, and shGLS1 accompanied by MPC overexpression respectively (J), or BMM with these treatments (K) were cultured in normal growth medium for up to 5 days for cell proliferation ability evaluation, n = 3/group respectively. (L-M) BMSC with the same treatments as that in 7J were cultured in osteogenic differentiation medium for 18 days for alizarin red staining (representative micrographs are shown, scale bar represents 100 mm) (L), BMM with the same treatments as that in 7K were cultured with M-CSF and RANKL for 5 days for TRAP staining (representative micrographs are shown, scale bar represents 100 mm) (M). (N-O) Quantitative analysis of alizarin red staining in 7L (N) and that of the number of osteoclasts in 7M (O), n = 5/group respectively. (P-Q) BMSC treated with SC accompanied by EV, shGLS1 accompanied by EV, and shGLS1 accompanied by MPC overexpression respectively were cultured in osteogenic differentiation medium for 14 days and then extracellular oxygen consumption rate (OCR) was assayed and calculated (P), BMM with these treatments were cultured with M-CSF and RANKL for 5 days and then OCR was assayed and calculated, n = 3/group respectively (Q). (R) BMM were firstly induced to differentiate into pre-osteoclasts and then digested and seeded on bone slices, treated with vehicle accompanied by EV, CB-839 accompanied by EV, and CB-839 accompanied by MPC overexpression respectively, for osteoclastic resorption activity assay by DAB staining, representative micrographs are shown and scale bar represents 100 mm; total bone resorption area and resorption area per osteoclast were calculated, n = 5/group respectively. Lines and error bars represent mean S.D.; P values in 7B, 7C, 7H, 7I, 7J, 7K, 7P, 7Q were determined by Kruskal-Wallis test with Dunn’s multiple comparisons test, P values in 7A by unpaired two-tailed Student’s t test and the rest by one-way ANOVA with Sidak’s multiple comparisons test (* P < 0.05, ** P < 0.01, *** P < 0.001, and **** P < 0.0001; “ns” not significant).
Fig 2: Inhibiting GLS via CB-839 aggravates age-related bone loss in male mice while prevents OVX-induced bone loss in female mice. (A-D) Distal femurs from 6-month-old and 18-month-old male groups treated with CB-839 (200 mg/kg body weight) or the same volume of vehicle twice daily were analyzed. Representative distal femoral μCT images (above is the longitudinal view, below is the axial view of metaphysis), scale bars represent 1 mm respectively (A); quantitative analysis of trabecular bone parameters including bone volume fraction (BV/TV), trabecular spacing (Tb.Sp), trabecular number (Tb.N) and trabecular thickness (Tb.Th), n = 6/group respectively (B); representative midshaft femoral μCT images (axial view), scale bar represents 1 mm (C); quantitative analysis of cortical bone parameters including cortical thickness (Ct.Th), cortical area (Ct.Ar), relative cortical-area-to-total-area ratio (Ct.Ar/Tt.Ar) and total area (Tt.Ar), n = 6/group respectively (D). (E-H) Distal femurs from sham-operated and ovariectomized groups treated with CB-839 (200mg/kg body weight) or same volume of vehicle twice daily were analyzed. Representative distal femoral μCT images (above is the longitudinal view, below is the axial view of metaphysis), scale bars represent 1 mm respectively (E); quantitative analysis of trabecular bone parameters including bone volume fraction (BV/TV), trabecular spacing (Tb.Sp), trabecular number (Tb.N) and trabecular thickness (Tb.Th), n = 7/group respectively (F); representative midshaft femoral μCT images (axial view), scale bar represents 1 mm (G); quantitative analysis of cortical bone parameters including cortical thickness (Ct.Th), cortical area (Ct.Ar), relative cortical-area-to-total-area ratio (Ct.Ar/Tt.Ar) and total area (Tt.Ar), n = 7/group respectively (H). Lines and error bars represent mean S.D.; all P values were determined by two-way ANOVA with Tukey’s multiple comparisons test (* P < 0.05, ** P < 0.01, *** P < 0.001, and **** P < 0.0001; “ns” not significant).
Fig 3: Defective metabolic pathways upon GLS inhibition in aged BMSC play different roles between aging-impaired osteogenic differentiation and RANKL-induced osteoclast differentiation. BMSC from aged (18-month-old) male mice and BMM from 8-week-old male mice were isolated. (A-B) BMSC were cultured in normal growth medium in presence of vehicle (DMSO), CB-839 (2 mM, the same below), CB-839 accompanied with GSH-EE (1 mM, the same below), CB-839 accompanied with 1× non-essential amino acids (NEAA), and CB-839 accompanied with 1× nucleotide mixture (nucleo) respectively, for up to 5 days for cell proliferation ability evaluation by CCK8 assay (n = 3/group) (A), or cultured in osteogenic differentiation medium in presence of these treatments for 18 days for alizarin red staining (representative micrographs are shown, scale bar represents 100 mm) (B). (C-D) BMM were cultured in normal growth medium in presence of the same treatments as that in 6A for up to 5 days for cell proliferation ability evaluation (n = 3/group) (C) or cultured with M-CSF and RANKL in presence of these treatments for 5 days for TRAP staining (representative micrographs are shown, scale bar represents 100 mm) (D). (E-F) Quantitative analysis of the alizarin red staining in 6B (E) and that of the number of osteoclasts in 6D (F), n = 5/group respectively. (G) GLS1 knockdown via adenovirus transduction of shRNAs (shGLS1) in BMSC and that in BMM were examined by western blots, representative images and quantitative analyses were shown, n=5/group respectively. (H-I) BMSC were treated with scrambled shRNAs (SC), shGLS1, shGLS1 accompanied with GSH-EE, shGLS1 accompanied with 1× NEAA, and shGLS1 accompanied with 1× nucleo respectively, cultured in normal growth medium for up to 5 days for cell proliferation ability evaluation (n = 3/group) (H), or cultured in osteogenic differentiation medium for 18 days for alizarin red staining (representative micrographs are shown, scale bar represents 100 mm) (I). (J-K) BMM with the same treatments as that in 6H were cultured in normal growth medium for up to 5 days for cell proliferation ability evaluation (n = 3/group) (J) or cultured with M-CSF and RANKL for 5 days for TRAP staining (representative micrographs are shown, scale bar represents 100 mm) (K). (L-M) Quantitative analysis of the alizarin red staining in 6I (L) and that of the number of osteoclasts in 6K (M), n = 5/group respectively. (N) BMM were firstly induced to differentiate into pre-osteoclasts and then digested and seeded on bone slices, treated with vehicle, CB-839, and CB-839 accompanied with 1× NEAA respectively, for osteoclastic resorption activity assay by DAB staining, representative micrographs are shown, and scale bar represents 100 mm; total bone resorption area and resorption area per osteoclast were calculated, n = 5/group respectively. Lines and error bars represent mean S.D.; P values in 6A, 6C, 6H, 6J were determined by Kruskal-Wallis test with Dunn’s multiple comparisons test, P values in 6G by unpaired two-tailed Student’s t test and the rest by one-way ANOVA with Sidak’s multiple comparisons test (* P < 0.05, ** P < 0.01, *** P < 0.001, and **** P < 0.0001; “ns” not significant).
Fig 4: miR-125a-5p directly targets 3’UTR of GLS in hMSCs. (A) Prediction of the miR-125a-5p/GLS 3’UTR association from starBase. (B) Negative correlation between GLS mRNA and miR-125a-5p was detected in hMSCs of OP patients. (C) hMSCs were transfected with control miRNA or miR-125a-5p, protein expression of GLS was detected by Western blot. (D) Luciferase assay was conducted in hMSCs through co-transfection of control miRNA or miR-125a-5p with WT- or Mut- GLS 3’UTR. **, p < 0.01; ***, p < 0.001
Fig 5: circFNDC3B and miR-125a-5p reversely regulate glutamine metabolism of hMSCs. (A) circFNDC3B was silenced in hMSCs, the glutamine uptake and (B) GLS activity were examined. (C) miR-125a-5p was overexpressed in hMSCs, the glutamine uptake and (D) GLS activity were examined. *, p < 0.05
Supplier Page from Abcam for Glutaminase (GLS) Activity Assay Kit (Fluorometric)