Angiopoietin-like 8 governs osteoblast-adipocyte lineage commitment during skeletal aging

Figures

• 

  Figure 1. ANGPTL8 expression increases with age in BM during skeletal aging. (A) Oil Red O and H&E staining of human BM samples from individuals at different ages. Scale bars: 200 μm. (B) ANGPTL8 levels in BM supernatants (n = 14) from different individuals. (C) Bioinformatics analysis of single-cell RNA sequencing of BM cells expressing ANGPTL8 from old (24 months) WT male mice. (D) Age-associated ANGPTL8 mRNA expression in hBMMSCs (n = 33). (E and F) ANGPTL8 levels in BM supernatants (E) and serum (F) from 3-, 12-, and 24-month-old WT male mice (n = 6). (G–J) mRNA expression of Angptl8 (G), corresponding ANGPTL8 levels in supernatants (H), and Western blot analysis for ANGPTL8 (I and J) in 18-month-old male mice BMMSCs from different cell passages (n = 3). (K) Representative images showing immunofluorescent staining of ANGPTL8 (green) in passage 5 (P5) BMMSCs from 18-month-old male mice. Scale bar: 50 μm. (L) qRT-PCR analysis of the mRNA expression of Angptl8, Pparg, and Fabp4 during the differentiation of BMMSCs from 3-month-old WT male mice into adipocytes (n = 3). (M and N) mRNA expression of ANGPTL8 (M) and ANGPTL8 levels (N) in supernatants from hBMMSCs from different passages (n =…

• 

  Figure 2. ANGPTL8 modulates cell-fate choice of MSCs between adipocytes and osteoblasts. (A) Cell differentiation was assessed 14 days after adipogenic induction by Oil Red O staining in BMMSCs transfected with LV-GFP or LV-ANGPTL8. Scale bar: 50 μm. (B) Quantification of Oil Red O staining in A. n = 3. (C) Cell differentiation was assessed 14 days after osteogenic induction by alizarin red staining in BMMSCs transfected with lentiviral LV-GFP or LV-ANGPTL8. Scale bar: 200 μm. (D) Quantification of calcium mineralization based on alizarin red staining in C. n = 3. (E) qRT-PCR analysis of the mRNA expression of Angptl8, Pparg, Fabp4, Fasn, Ldlr, Runx2, Sp7, p16, and p21 in ANGPTL8-overexpressing BMMSCs. (F) Western blot analysis of ANGPTL8, PPARγ, Fabp4, Fasn, Ldlr, Runx2, Sp7, p16, and p21 in BMMSCs after transfection with LV-GFP or LV-ANGPTL8. (G) Representative images showing immunofluorescent staining of PPARγ in BMMSCs treated with ANGPTL8. Scale bar: 50 μm. (H) The adipogenic potential of BMMSCs after transfected with siANGPTL8 was assessed 14 days after induction of differentiation by Oil Red O staining. Scale bar: 50 μm. (I) Quantification of Oil Eed O staining in H. n = 3. (J) The…

• 

  Figure 3. Angptl8-Nestin-creTg mice display lower bone mass and higher fat accumulation in BM. (A) Representative microCT images of femurs from 1-, 3-, and 18-month-old WT and Angptl8-Nestin-creTg male mice. Scale bar: 1 mm. (B–F) Quantitative microCT analysis of the femurs from 1-, 3-, and 18-month-old WT and Angptl8-Nestin-creTg male mice. n = 6–8. BMD, bone mineral density; BV/TV, bone volume per tissue volume; Ct.Th, cortical bone thickness; Tb.N, trabecular number; Tb.Sp, trabecular separation. (G and H) Three-point bending measurement of tibia maximum load (G) and stiffness (H) from 18-month-old WT and Angptl8-Nestin-creTg male mice. n = 5. (I) H&E staining for BMAT in the femurs of 18-month-old WT and Angptl8-Nestin-creTg male mice. Scale bars: 100 μm. (J and K) Quantification of area (J) and number of adipocytes (K) in the femurs of 18-month-old WT and Angptl8-Nestin-creTg male mice. n = 3. (L) Representative images of calcein double labeling of trabecular bone from 18-month-old WT and Angptl8-Nestin-creTg male mice. Scale bar: 50 μm. (M and N) Quantification of the mineral apposition rate (MAR) and bone formation rate (BFR) based on calcein double labeling from 18-month-old WT and…

• 

  Figure 4. Angptl8–/– mice exhibit higher bone mass and lower BMAT accumulation. (A) Representative microCT images of femurs from 1-, 3-, and 18-month-old WT and Angptl8–/– male mice. Scale bar: 1 mm. (B–F) Quantitative microCT analysis of the femurs from 1-, 3-, and 18-month-old WT and Angptl8–/– male mice. n = 6–8. BMD, bone mineral density; BV/TV, bone volume per tissue volume; Ct.Th, cortical bone thickness; Tb.N, trabecular number; Tb.Sp, trabecular separation. (G and H) Three-point bending measurement of tibia maximum load (G) and stiffness (H) from 18-month-old WT and Angptl8–/– male mice. n = 5. (I) H&E staining for BMAT in the femurs of 18-month-old WT and Angptl8–/– male mice. Scale bars: 100 μm. (J and K) Quantification of number (J) and area of adipocytes (K) in the femurs of 18-month-old WT and Angptl8–/– male mice. n = 3. (L) Representative images of calcein double labeling of trabecular bone from 18-month-old WT and Angptl8–/– male mice. Scale bar: 50 μm. (M and N) Quantification of the mineral apposition rate (MAR) and bone formation rate (BFR) based on calcein double labeling from 18-month-old WT and Angptl8–/– male mice. n = 5. Data are mean ± SEM. *P < 0.05; **P < 0.01 by…

• 

  Figure 5. ANGPTL8 regulates MSC differentiation via the Wnt/β-catenin signaling pathway. (A) Pathway enrichment analysis in BMMSCs treated with LV-ANGPTL8. (B) qRT-PCR analysis of the mRNA expression of Wnt3a, Fzd9, Axin2, and Ctnnb1 after BMMSC transfection with LV-GFP and LV-ANGPTL8. n = 3. (C) qRT-PCR analysis of the mRNA expression of Wnt3a, Fzd9, Axin2, and Ctnnb1 after BMMSC transfection with siANGPTL8. n = 3. (D) ANGPTL8 blunted Wnt3a-induced phosphorylation of Gsk3β and prevented the subsequent accumulation of β-catenin in BMMSCs. n = 3. (E) Western blotting of PPARγ, Fabp4, Runx2, and Sp7 from BMMSCs treated with Angptl8 and Wnt3a. (F) Wnt3a rescued Oil Red O and alizarin red staining of lineage differentiation bias caused by overexpression of ANGPTL8 in BMMSCs. n = 3. Scale bars: 100 μm. (G) Quantification of Oil Red O staining in F. n = 3. (H) Quantification of calcium mineralization based on alizarin red staining in F. n = 3. (I) Immunohistochemical staining of β-catenin in the femurs of WT, Angptl8-Nestin-creTg, and Angptl8–/– mice. Scale bar: 50 μm. (J) Western blotting of p-Smad3, Smad3, p-Gsk3β, Gsk3β, and β-catenin from BMMSCs treated with TGF-β1 and Angptl8-ASO. Data are…

• 

  Figure 6. PPARγ inhibition partially inhibits ANGPTL8 expression and rescues the phenotype of ANGPTL8 overexpression in MSCs. (A) Representative images of Oil Red O staining and alizarin red staining in mouse BMMSCs after transfection with LV-GFP and LV-ANGPTL8 treated with vehicle or a PPARγ inhibitor (GW9662). Scale bars: 100 μm. (B) Quantification of Oil Red O staining in A. n = 3. (C) Quantification of calcium mineralization based on alizarin red staining in A. n = 3. (D) qRT-PCR analyses of the mRNA expression of Angptl8 in GW9662-treated BMMSCs. (E–G) qRT-PCR analyses of the mRNA expression of Pparg, Cebpα, and Fabp4 in GW9662-treated BMMSCs under adipogenic conditions. (H–K) qRT-PCR analyses of the mRNA expression of Alp, Bglap, Runx2, and Sp7 in GW9662-treated BMMSCs under osteogenic conditions. n = 3 biologically independent BMMSC samples. Data are mean ± SEM. *P < 0.05; **P <0.01; ***P < 0.001; ****P < 0.0001 by 2-way ANOVA followed by Tukey’s multiple-comparison test.

• 

  Figure 7. Administration of a PPARγ inhibitor partially rescues the phenotype of Angptl8-Nestin-creTg mice. (A) Weight of the GW9662-treated WT and Angptl8-Nestin-creTg male mice at 18 months. n = 7–8. (B) H&E staining the femur of the GW9662-treated WT and Angptl8-Nestin-creTg male mice at 18 months. Scale bars: 100 μm. (C and D) Quantification of number (C) and area (D) of adipocytes in the GW9662-treated WT and Angptl8-Nestin-creTg male mice at 18 months. n = 3. (E) Representative microCT images of distal femurs and midshaft cortical bone from the GW9662-treated WT and Angptl8-Nestin-creTg male mice at 18 months. Scale bar: 1 mm. (F–I) Quantitative microCT analyses of the distal end of the femurs from the GW9662-treated WT and Angptl8-Nestin-creTg male mice at 18 months. n = 7–8. BMD, bone mineral density; BV/TV, bone volume per tissue volume; Ct.Th, cortical bone thickness; Tb.N, trabecular number. (J) Representative osteocalcin-positive cell images of distal femurs from the GW9662-treated WT and Angptl8-Nestin-creTg male mice at 18 months. Scale bars: 50 μm. (K) Quantification of osteocalcin+ cells on the bone surface (number of osteocalcin+ cells per bone perimeter, N.Ocn+/B.Pm).…

• 

  Figure 8. Administration of Angptl8-ASO partially rescues the phenotype of aging mice. (A) qRT-PCR analyses of the mRNA expression of Angptl8 in the BMMSCs transfected with control-ASO and Angptl8-ASO. n = 3. (B) Experimental scheme to demonstrate that Angptl8-ASO partially rescues the phenotype of aging male mice. (C) H&E staining of the femurs of the control-ASO and Angptl8-ASO male mice at 18 months. Scale bar: 100 μm. n = 8. (D and E) Quantification of number (D) and area (E) of adipocytes in the control-ASO and Angptl8-ASO male mice at 18 months. n = 3. (F) Representative microCT images of distal femurs and midshaft cortical bone from the control-ASO and Angptl8-ASO male mice at 18 months. Scale bar: 1 mm. n = 8. (G–J) Quantitative microCT analyses of the distal end of the femurs from the control-ASO and Angptl8-ASO male mice at 18 months. n = 8. BMD, bone mineral density; BV/TV, bone volume per tissue volume; Ct.Th, cortical bone thickness; Tb.Th, trabecular thickness. (K) Representative osteocalcin+ cell images of distal femurs from the control-ASO and Angptl8-ASO male mice at 18 months. Scale bar: 50 μm. n = 8. (L) Representative β-Gal staining images of distal femurs from the…