Suppression of autophagy by FIP200 deletion leads to osteopenia in mice through the inhibition of osteoblast terminal differentiation

Abstract

Autophagy is a conserved lysosomal degradation process that has important roles in both normal human physiology and disease. However, the function of autophagy in bone homeostasis is not well understood. Here, we report that autophagy is activated during osteoblast differentiation. Ablation of focal adhesion kinase family interacting protein of 200 kD (FIP200), an essential component of mammalian autophagy, led to multiple autophagic defects in osteoblasts including aberrantly increased p62 expression, deficient LC3-II conversion, defective autophagy flux, absence of GFP-LC3 puncta in FIP200-null osteoblasts expressing transgenic GFP-LC3, and absence of autophagosome-like structures by electron microscope examination. Osteoblast-specific deletion of FIP200 led to osteopenia in mice. Histomorphometric analysis revealed that the osteopenia was the result of cell-autonomous effects of FIP200 deletion on osteoblasts. FIP200 deletion led to defective osteoblast terminal differentiation in both primary bone marrow and calvarial osteoblasts in vitro. Interestingly, both proliferation and differentiation were not adversely affected by FIP200 deletion in early cultures. However, FIP200 deletion led to defective osteoblast nodule formation after initial proliferation and differentiation. Furthermore, treatment with autophagy inhibitors recapitulated the effects of FIP200 deletion on osteoblast differentiation. Taken together, these data identify FIP200 as an important regulator of bone development and reveal a novel role of autophagy in osteoblast function through its positive role in supporting osteoblast nodule formation and differentiation.

Keywords: AUTOPHAGY; BONE DEVELOPMENT; DIFFERENTIATION; MOUSE; OSTEOBLAST.

Figure 1. Autophagy is activated during osteoblast differentiation

(A) Lysates isolated from primary calvarial osteoblasts at indicated different differentiation stages with or without 2h 30mM NH₄Cl treatment were analyzed by Western blotting using anti-LC3 (top) or anti-vinculin (bottom) antibodies. Graph on the right shows the quantification of LC3-II expression. *p<0.05, n=3. (B) Primary calvarial osteoblasts were isolated from GFP-LC3 transgenic mice, fixed and analyzed under fluorescent microscope at indicated day after plating. (C) Primary calvarial osteoblasts isolated from neonatal GFP-LC3 transgenic mice were cultured in complete or starvation medium for 2h or starvation medium for 2h followed by 2h complete medium at day 2 cultures before having reached confluence. The cells were then fixed, stained with Dapi and subject to fluorescent microscope analysis. (D) Bone marrow cells were isolated from 6-8 weeks old GFP-LC3 transgenic mice and subject to osteogenic differentiation till day 11. D1 showed the isolated osteoblast like cells in the culture. D2 showed the slightly condensed osteoblast-like cell nodule. D3 showed the highly condensed osteoblast-like cell nodule. The data shown were the representatives of at least 3 independent experiments.

Figure 2. FIP200 null osteoblasts were autophagy deficient

(A) Primary calvarial osteoblasts were isolated from control or Osx-CKO neonatal mice and cultured in complete medium for 3 days. Cell lysates were subjected to immunoblot analysis with indicated antibodies. Graph on the right shows the quantification of p62 expression. (B) Primary calvarial osteoblasts were cultured in complete medium or starvation medium (serum and amino acid free EBSS) for up to 60 minutes. Cell lysates were subjected to immunoblot analysis with indicated antibodies. (C) Control and FIP200 null primary calvarial osteoblasts were cultured in the complete or starvation medium for 3h with or without 100 μM Chloroquine. The cell lysates were subjected to immunoblot analysis with indicated antibodies. Graph on the right shows the quantification of LC3-II expression with starvation and Chloroquine treatment. (D) Control and FIP200 null primary calvarial osteoblasts isolated from mice expressing transgenic GFP-LC3 were cultured in complete medium or starvation medium for 120 minutes and then were observed directly with a fluorescence microscope. (E) Transmission electron microscopic image of control and FIP200 null primary calvarial osteoblasts. Arrows point to autophagosome-like structure,arrow heads point to autolysosome-like structure, and asterisks indicate mitochondria. (F-H) Primary calvarial osteoblasts were isolated from neonatal control or Osx-CKO mice. Osteoblasts were cultured in complete or starvation medium for 6h and then fixed and subjected to EM analysis and quantification. (F) Mitochondria size. ^#p<0.001, n=312-476 per group. (G) Mitochondria area per cytoplasmic area. *p<0.01, ^#p<0.001, n=37-46 per group. (H) Mitochondria number. ^#p<0.001, n=37-46 per group. For F-H, the data were presented as mean ± SD. Western blotting and immunofluorescence data shown were the representative of 3 independent experiments. All experiments were performed before cells had reached confluence.

Figure 3. FIP200 deletion in osteoblasts led to osteopenia and decreased bone strength

(A-G) Trabecular and cortical parameters were determined by MicroCT for the femurs from one month (1M) to six month (6M) old Osx-CKO female mice: (A) Bone volume/tissue volume (BV/TV); (B) Trabebular number (TbN); (C) Trabecular thickness (TbTh); (D) Trabebular spacing (TbSp); (E) Cortical bone thickness; (F) Cortical bone outer perimeter; (G) Cortical bone inner perimeter. (H) Calvarial bone thickness was directly measured with caliper. (I-N) Four-point bending test with femurs: (I) Yield load; (J) Ultimate load; (K) Stiffness; (L) Elastic energy; (M) Plastic energy; and (N) Energy to failure. For each group, n=6-9, *p<0.05. Data are mean + SD.

Figure 4. FIP200 deletion in osteoblasts led to decreased bone formation

(A-F) Static histomorphometry for the femurs of one month old female Osx-CKO and control mice: (A) Trabecular bone area/tissue area (BA/TA); (B) Trabecular bone number (TbN); (C) Osteoblast number per bone surface (NOb/BS); (D) Osteoblast surface per bone surface (ObS/BS); (E) Osteoclast number per bone surface (NOc/BS). (F) Osteoclast surface per bone surface (OcS/BS); (G-J) Dynamic histomorphometry for the femurs of one month old female Osx-CKO and control mice: (G) mineral deposition rate; (H) double labeling surface per bone surface (dLS/BS); (I) decreased bone formation rate (BFR/BS), and (J) Representative calcein and xylenol orange double labeling image (5 days apart between two labelings). *p<0.05, n=5-7 per group.

Figure 5. FIP200 deletion compromised osteoblast terminal differentiation

(A-C) Bone marrow cells were collected from 6-8 weeks old control or Osx-CKO mice and subject to osteoblast differentiation: (A) Representative Alizarin red staining image at the end of 21 days osteogenic culture. (B) Quantified calcium concentration for the samples in (A). (C) Osteoblast differentiation marker expression in cultures shown in (A). (D-F) Bone marrow cells from FIP200^F/F mice were infected with adenovirus encoding Cre (Ade-Cre) or Laz (Ade-Laz) after 7 days osteogenic culture: (D) Representative Alizarin red staining image at the end of 21 days culture. (E) Quantified calcium concentration for the samples in (D). (F) Osteoblast differentiation marker expression in the cultures shown in (D). (G-L) Primary calvarial osteoblast cultures: (G) Early osteoblast differentiation was evaluated by Alkaline phosphatase staining at day 7 culture. (H) Early osteoblast differentiation marker (Alkaline phosphatase mRNA) was determined by quantitative PCR. (I) Representative Alizarin red staining image at the end of 21 days osteogenic culture. (J) Quantified calcium concentration for the samples in (I). (K) Osteoblast differentiation marker expression in cultures shown in (I). Data are the representatives of 3 independent experiments with triplicates for each experiment. *p<0.05, n=3 per group.

Figure 6. FIP200 deficiency compromised the osteoblast nodule formation

(A, B) Immunostaining with anti-Ki67 antibody was performed in the primary osteoblasts isolated from neonatal calvaria of Osx-CKO and control mice. (A) Representative fluorescent images. (B) Quantitative data of the Ki67 positive osteoblasts. (C, D) Primary calvarial osteoblasts were isolated from neonatal control or Osx-CKO mice and cultured in osteogenic medium for 3 weeks: (C) Cell numbers were counted at indicated time point. (D) Representative images showing the nodule formation at the end of 21 days culture. Arrow heads indicated the big nodules in control culture and arrows indicated the small nodules in Osx-CKO culture. (E-H) Bone marrow cells were collected from 6-8 weeks old control or Osx-CKO mice and subject to osteoblast differentiation: (E) Alkaline phosphatase staining at indicated time points. (F) Quantitative Alkaline phosphatase positive (ALP⁺) osteoblastic colony numbers at indicated time points as shown in (E). (G) Quantitative AP⁺ osteoblastic colony size at indicated time points as shown in (E). (H) Quantitative ALP⁺ area at indicated time points as shown in (E).

Figure 7. Inhibition of autophagy led to compromised osteoblast differentiation

(A-C) Bone marrow cells were isolated from 6-8 weeks old C57/BL6 mice and subject to osteogenic culture. (A) Diagram of the inhibitor treatment (2mM 3-MA or 15μM Chloroquine) scheme (late treatment) and the representative results of Alkaline phosphatase staining and Alizarin red staining results. (B) Bone marrow cells were treated with autophagy inhibitors at indicated dose starting from day 7 for two days and cell numbers were counted at day 9. (C) Diagram of the inhibitor treatment scheme (early treatment) and the representative Alkaline phosphatase staining result. (D) Primary calvarial osteoblasts were isolated from neonatal C57/BL6 mice and subject to osteogenic culture. Diagram shows the inhibitor treatment (2mM 3MA or 5μM Chloroquine) scheme and images show the representative Alkaline phosphatase staining and Alizarin red staining results. (E) Primary calvarial osteoblasts were isolated from neonatal control or Osx-Cre mice and cultured in osteogenic medium. For the control cells, one group was treated with 2 mM 3MA and the other group was treated with 5 μM Chloroquine. Phase contrast images were taken at indicated time points (D12 and D19 culture). Arrows point to the big nodule in control culture. Arrow heads point to the small nodule in autophagy inhibitors treated culture. Scale bar=0.25mm