Functional redundancy of type II BMP receptor and type IIB activin receptor in BMP2-induced osteoblast differentiation

Abstract

Signaling pathways for bone morphogenetic proteins (BMPs) are important in osteoblast differentiation. Although the precise function of type I BMP receptors in mediating BMP signaling for osteoblast differentiation and bone formation has been characterized previously, the role of type II BMP receptors in osteoblasts is to be well clarified. In this study, we investigated the role of type II BMP receptor (BMPR-II) and type IIB activin receptor (ActR-IIB) in BMP2-induced osteoblast differentiation. While osteoblastic 2T3 cells expressed BMPR-II and ActR-IIB, loss-of-function studies, using dominant negative receptors and siRNAs, showed that BMPR-II and ActR-IIB compensated each other functionally in mediating BMP2 signaling and BMP2-induced osteoblast differentiation. This was evidenced by two findings. First, unless there was loss of function of both type II receptors, isolated disruption of either BMPR-II or ActR-IIB did not remove BMP2 activity. Second, in cells with loss of function of both receptors, restoration of function of either BMPR-II or ActR-IIB by transfection of the wild-type forms, restored BMP2 activity. These findings suggest a functional redundancy between BMPR-II and ActR-IIB in osteoblast differentiation. Results from experiments to test the effects of transforming growth factor β (TGF-β), activin, and fibroblast growth factor (FGF) on osteoblast proliferation and differentiation suggest that inhibition of receptor signaling by double-blockage of BMPR-II and ActR-IIB is BMP-signaling specific. The observed functional redundancy of type II BMP receptors in osteoblasts is novel information about the BMP signaling pathway essential for initiating osteoblast differentiation.

Fig. 1

Expression of type II BMP receptors in osteoblastic 2T3 cells. A: Expression of endogenous BMPR-II, ActR-II, and ActR-IIB detected by RT-PCR. B: Molecular structures of BMPR-II and ActRIIB. Top: Wild-type consists of the extracellular ligand-binding domain (gray), transmembrane domain (black), and intracellular kinase domain (blank). Bottom: Dominantnegative C' terminal, intracellular domain, truncated forms, dnBMPR-II, and dnActR-IIB. C, D: Expression of dnBMPR-II and dnActR-IIB. Expression plasmids of (1) the dominantnegative mutantsor (2) their empty vectors were stably transfected into 2T3 cells. Expression of these mutant receptors and endogenous receptors was determined by RT-PCR (C) and Western blot (D) with anti-HA and anti-Flag antibodies for dnBMPR-II and dnActR-IIB.

Fig. 2

Effects of dnBMPR-II and dnActR-IIB on BMP signaling. A: Smad phosphorylation. 2T3 cells, transfected with empty vector, dnBMPR-II, ActR-IIB, or dnBMPR-II/dnActR-IIB expression vector, were treated with BMP2 (100 ng/ml) for 6 h. Levels of phosphorylated BMP-specific Smads were determined by Western blot using an anti-phospho-Smad1/5/8 antibody, with normalization to non-phosphorylated Smad1 and β-actin. B: Smad1 nuclear translocation. The mutant receptor-carrying cells were co-transfected with Flag-Smad1 plasmid and treated with BMP2 for 6 h. After fixation, cells were probed with FITC-conjugated anti-Flag antibody. Immunofluorescence of Smad1 was recorded. C: BMP/Smad signaling reporter assay. Cells were transfected with 12×SBELuc reporter construct and treated with BMP2 for 24 h. Relative reporter luciferase activity was determined and normalized by β-galactosidase activity. *P < 0.01 versus vehicle controls.

Fig. 3

Effects of dnBMPR-II and dnActR-IIB on osteoblast differentiation. A: Mineralized matrix formation. Osteoblastic 2T3 cells expressing dominantnegative receptors were treated with BMP2 (100 ng/ml) in an osteogenic culture medium for 2 weeks. Mineralized matrix formation was determined by combined von Kossa and van Gieson staining. B: ALP activity. Cells were treated with BMP2 for 2 days. ALP activity of cell lysates was measured with cell protein normalization. *P < 0.01 versus vehicle controls. C: Expression of osteoblast marker genes. At 4 or 8 days after BMP2 treatment, total RNA was extracted from these cells. mRNA levels of Runx2 and OCN were determined by Northern blotting with GAPDH normalization.

Fig. 4

Effects of dnBMPR-II/dnActR-IIB double blockage on osteoblast proliferation. A, C: Cell proliferation assay. 2T3 cells transfected with vector or dnBMPR-II/dnActR-IIB double mutants were treated with (A) TGF-β1 (0–0.06 ng/ml), (C) aFGF (0–0.4 ng/ml), or their vehicle for 40 h. Cell proliferation was determined by adding an assay reagent (CellTiter, Promega, Madison, WI) and reading absorbance at 570 nm. B: TGF-β signaling reporter assay. Both control and mutant cells were transfected with TGF-β signaling reporter 3TP-Lux and incubated with TGF-β1 (0.04 ng/ml) for 36 h. The reporter luciferase activity was determined by luminescence with β-gal normalization. *P < 0.05 versus vehicle control.

Fig. 5

Restoration of dnBMPR-II/dnActR-IIB-impaired BMP signaling and osteoblast differentiation. Osteoblastic 2T3 cells, in which BMP signaling was blocked by expression of dnBMPR-II/dnActR-IIB double mutants, were transfected with wild-type BMPR-II, wild-type ActR-IIB, or an empty vector. After BMP2 treatment, (A) the BMP signaling reporter (12×SBE-Luc) activity and (B) ALP activity in the cell lysates were determined as described in Figures 1 and 2. *P < 0.01 versus vehicle control.

Fig. 6

Effects of knockdown of BMPR-II and ActR-IIB on BMP signaling and osteoblast differentiation. A: siRNA knockdown. Wild-type 2T3 cells were transfected with targeting siRNAs for BMPR-II and/or ActR-IIB, or control scrambled RNA, at 20 nM for 24 h and mRNA levels of endogenous BMPR-II and ActR-IIB were examined by PCR. B, E: BMP signaling reporter assay. Osteoblastic 2T3 (B) and C2C12 (E) cells were co-transfected with siRNAs and 12SBE-Lucreporter, and then treated with BMP2 for 48 h. The reporter luciferase activity was measured with β-gal normalization. *P < 0.05 versus vehicle control. ^#P < 0.05 versus BMP2 treatment in other groups. C: ALP activity assay. Osteoblastic 2T3 cells transfected with siRNAs were incubated with BMP2 for 48 h, followed by measurement of ALP activity as described in Figure 3. *P < 0.05 versus vehicle control. D: Osteocalcin expression. 2T3 cells transfected with siRNAs were treated with BMP2 for 24 h. Osteocalcin mRNA levels were determined by real time PCR with 18S rRNA normalization. *P < 0.05 versus vehicle control.

Fig. 7

Effects of activin on osteoblast differentiation. A: ALP activity. Wild-type 2T3 cells were treated with BMP2 or activin A for 48 h. ALP activity of cell lysates was determined with normalization to cell protein. B: Mineralized matrix formation. Wild-type 2T3 cells were cultured in osteogenic media in the presence of BMP2 (0.5–50 ng/ml) or activin A (1–100 ng/ml) for 2 weeks. Areas of mineralized bone nodules in the culture were quantitated with von Kossa stain.