Deletion of BMP receptor type IB decreased bone mass in association with compromised osteoblastic differentiation of bone marrow mesenchymal progenitors

Abstract

We previously found that disruption of two type I BMP receptors, Bmpr1a and Acvr1, respectively, in an osteoblast-specific manner, increased bone mass in mice. BMPR1B, another BMP type I receptor, is also capable of binding to BMP ligands and transduce BMP signaling. However, little is known about the function of BMPR1B in bone. In this study, we investigated the bone phenotype in Bmpr1b null mice and the impacts of loss of Bmpr1b on osteoblasts and osteoclasts. We found that deletion of Bmpr1b resulted in osteopenia in 8-week-old male mice, and the phenotype was transient and gender specific. The decreased bone mass was neither due to the changes in osteoblastic bone formation activity nor osteoclastic bone resorption activity in vivo. In vitro differentiation of Bmpr1b null osteoclasts was increased but resorption activity was decreased. Calvarial pre-osteoblasts from Bmpr1b mutant showed comparable differentiation capability in vitro, while they showed increased BMP-SMAD signaling in culture. Different from calvarial pre-osteoblasts, Bmpr1b mutant bone marrow mesenchymal progenitors showed compromised differentiation in vitro, which may be a reason for the osteopenic phenotype in the mutant mice. In conclusion, our results suggested that BMPR1B plays distinct roles from BMPR1A and ACVR1 in maintaining bone mass and transducing BMP signaling.

Figure 1. Bmpr1b deletion leads to osteopenia in 8-week-old male mice, but does not change osteoblastic bone formation activity or osteoclastic bone resorption activity in vivo.

(A) Representative micro-CT images of trabecular compartment of distal femur and cortical compartment of middle femur from 8-week-old males. (B–F) Trabecular parameters were determined by micro-CT for the femora from 8-week-old male mice: (B) trabecular bone volume fraction (bone volume/tissue volume, BV/TV); (C) trabecular number (Tb. N); (D) trabecular thickness (Tb.Th); (E) trabecular separation (Tb. Sp); (F) tissue mineral density (TMD); (G–K) Cortical parameters: (G) cortical porosity; (H) inner perimeter; (I) outer perimeter; (J) cortical thickness; (K) TMD. For each group, n = 8. (L–R) Static histomorphometry for the trabecular compartment of distal femora of 8-week-old male mice (for each group, n = 6): (L) Representative H&E stained sections; (M) trabecular bone area/tissue area (BA/TA); (N) trabecular bone number (Tb. N); (O) trabecular bone thickness (Tb. Th); (P) osteoblast number per bone surface (N. Ob/BS); (Q) osteoclast number per bone surface (N. Oc/BS); (R) eroded surface per bone surface (ES/BS); (S) Representative calcein labeling images (5 days apart between two labels). (T–V) Dynamic histomorphometry for the femora of 8-week-old male mice (for each group, n = 8): (T) mineral apposition rate (MAR); (U) mineralized surface per bone surface (MS/BS); (V) bone formation rate (BFR/BS). *p < 0.05; **p < 0.01. WT: wild type; KO: knockout.

Figure 2. In vitro differentiation of osteoclasts from KO mice is increased but resorption activity is decreased.

(A) Osteoclast precursors were treated with 30 ng/ml M-CSF for 6 days, and the number of cells was counted. (B) Osteoclasts precursors were prompted differentiation with gradient concentration of RANKL for 6 days. Cells were stained with TRAP, and the number of osteoclasts (B) as well as the number of nuclei per osteoclasts (C) were counted. (C) Representative images of osteoclast differentiation from WT and KO mice. (D) Osteoclasts precursors were differentiated with 30 ng/ml M-CSF and 50 ng/ml RANKL. On day 6, M-CSF and RANKL were removed from the culture, and number of osteoclasts was counted at indicated time points. The numbers of osteoclasts at each time point was normalized to that at time point 0. (E) Osteoclast precursors were cultured on coated surface and treated with 30 ng/ml M-CSF and 50 ng/ml RANKL. After 6 days, the area of resorption was measured and the number of osteoclasts on coated surface was counted. Resorption area was normalized by osteoclast number. Representative images of resorption by WT and KO osteoclasts. *p < 0.05; **p < 0.01. WT: wild type; KO: knockout.

Figure 3. Deletion of Bmpr1b transduces BMP signaling pathway distinctly in calvarial pre-osteoblasts.

(A) Cultured calvarial pre-osteoblasts from WT and KO were serum starved overnight, followed by rhBMP-2 stimulation for the indicated times. Cells were lysed and immunoblotted with P-SMAD1/5/9, P-ERK1/2 and P-P38. GAPDH, ERK1/2 and P38 were used as loading controls. (B) Quantification of the blotting of P-SMAD1/5/9, P-ERK1/2 and P-P38. (C) Cultured calvarial pre-osteoblasts from Acvr1 fx/+ :Ubi Cre-ER^TM (+)/(−):R26R/+ (con) and Acvr1 fx/-:Ubi Cre-ER^TM (+)/(−):R26R/+ (mut) were treated with TM and sorted as mentioned in materials and methods section. Cell lysate was immunoblotted as mentioned above. (D) Cultured calvarial pre-osteoblasts from Bmpr1a fx/+ :Ubi Cre-ER^TM (+)/(−):R26R/+ (con) and Bmpr1a fx/-:Ubi Cre-ER^TM (+)/(−):R26R/+ (mut) were treated with TM. Cell lysate was immunoblotted as mentioned above. (E) Cells were fixed and immunostained with P-SMAD1/5/9 and SMAD. (F) The mean density of P-SMAD1/5/9 signal in the nucleus and the signal ratio of nucleus and cytoplasm of total SMAD were quantified. *p < 0.05; **p < 0.01. WT: wild type; KO: knockout; con: control (fx/+ :Ubi Cre-ER(+)/(−):R26R/+); mut: mutant (fx/-:Ubi Cre-ER(+)/(−):R26R/+). For western blot results, the gels were run, and transferred to the membrane under the same experimental conditions, and the blots within comparisons were exposed at the same conditions. The blots were cropped. Full-length blots are presented in Supplementary Figure S13.

Figure 4. Osteoblastic differentiation of bone marrow mesenchymal progenitors is compromised in KO mice.

Bone marrow mesenchymal progenitors were isolated from femora and tibiae of 8-week-old mice and subjected to osteoblast differentiation. Cells were treated with or without BMP-2. (A) Alkaline phosphatase (ALP) staining and mean density of the staining after 7 days of differentiation. (B) ALP staining and mean density of the staining after 14 days of differentiation. (C) Alizarin red staining and quantification of the staining after 21 days of differentiation. After staining, 300 μl/well of 10% CPC was used to dissolve bound alizarin red S. The dissolved solution was diluted 10 times with 10% CPC before measurement. *p < 0.05; **p < 0.01. WT: wild type; KO: knockout.

Figure 5. Expression of osteoblast marker genes was decreased in KO bone marrow mesenchymal progenitors.

Bone marrow cells were cultured in osteogenic medium for the indicated days. Total mRNA was isolated, and quantitative RT-PCR was performed. Expression of Runx2 (A), Sp7 (B), Alp1 (C), Ocn (D), Dmp1 (E), Ibsp (F) and Col1a2 (G) was calculated. *p < 0.05; **p < 0.01. WT: wild type; KO: knockout.

Figure 6

BMP signaling was decreased in bone marrow mesenchymal progenitors deficient for Bmpr1b (A,B), Bmpr1a (C,D) and Acvr1 (E,F), respectively. (A,C,E) Bone marrow mesenchymal progenitors were serum starved overnight, followed by rhBMP-2 stimulation for the indicated times. Cells were lysed and immunoblotted with P-SMAD1/5/9, P-ERK1/2 and P-P38. GAPDH, ERK1/2 and P38 were used as loading controls. The gels were run, and transferred to the membrane under the same experimental conditions and the blots within comparisons were exposed at the same conditions. The blots were cropped. Full-length blots are presented in Supplementary Fig. S13. (B,D,F) Quantification of the blotting of P-SMAD1/5/9, P-ERK1/2 and P-P38. (G) Expression pattern of Bmpr1a, Acvr1 and Bmpr1b in bone marrow mesenchymal progenitors, calvaria and cartilage. (H) A working model showing distinct roles of BMP type I receptors during osteoblastic differentiation from bone marrow mesenchymal progenitors. Our previous findings demonstrated that BMPR1A and ACVR1 in osteoblasts positively regulate osteoblast activity and osteoclast activity. Our current study found that BMPR1B positively regulate early osteoblastic differentiation of bone marrow mesenchymal progenitors, but not later stage. However, we still do not know how BMPR1A and ACVR1 regulate BMSCs, due to lack of mouse model targeting BMPR1A and ACVR1 in BMSCs. *p < 0.05; **p < 0.01. WT: wild type; KO: knockout; con: control (fx/+ :Ubi Cre-ER(+)/(−):R26R/+); mut: mutant (fx/-:Ubi Cre-ER(+)/(−):R26R/+); Het: heterozygous (fx/−:Ubi Cre-ER(−)/(−):R26R/+); BM: bone marrow mesenchymal.