Notch is expressed and activated during osteoclast differentiation. (A) BMM cells were treated with 50 ng/ml of RANKL for the indicated times. Total RNA was isolated from BMMs, and expression levels of Notch1 to -4, Hes-1, RANK, cathepsin K, and GAPDH (glyceraldehyde-3-phosphate dehydrogenase) mRNA were measured by RT-PCR. (B) RANKL activates Notch signaling in BMMs. Total cell lysates from RANKL-treated BMMs (50 ng/ml) were immunoblotted with anti-Notch1 or anti-Notch1 ICD (left) or anti-Notch2 (right) antibodies. The images of anti-Notch1 and anti-Notch1 ICD were taken from the same membrane by reblotting. (C) Immunostaining of Notch2 protein in BMMs stimulated with RANKL (50 ng/ml) (top). The middle panels show nuclei stained with DAPI. The lower panels show merged images of Notch2 and DAPI staining. Scale bar = 50 μm. (D) Active form of Notch2 translocates to nuclei. BMMs were treated with or without RANKL (50 ng/ml) for 30 min, and then cytoplasm (C), membrane (M), and nuclear (N) extracts were prepared and immunoblotted with anti-Notch2 antibodies. As fractionation controls we used anti-IκBα antibodies for cytosol, anti-RANK antibodies for the membrane, and anti-HDAC1 antibodies for the nuclear fraction. Similar results were obtained in three independent experiments.

GSI suppresses RANKL-induced osteoclastogenesis. (A) Mouse bone marrow cells were treated with M-CSF (50 ng/ml) for 3 days in the presence of GSI and further cultured in the presence of 50 ng/ml RANKL together with GSI for 3 more days. (B) Mouse bone marrow cells were treated with M-CSF for 3 days in the presence of GSI. The culture medium was removed, and cells were washed with phosphate-buffered saline twice and were further cultured in the presence of 50 ng/ml RANKL for 3 days. (C) Mouse bone marrow cells were treated with M-CSF (50 ng/ml) for 3 days and further cultured in the presence of RANKL (50 ng/ml) together with GSI for 3 days. Cells were fixed and stained for TRAP. TRAP⁺ MNCs were counted as osteoclasts. Scale bar = 100 μm. Data shown are the numbers of TRAP⁺ MNCs per culture well (values are means ± SEM, n = 3). Hes-1 and GAPDH mRNA expression levels were determined by RT-PCR.

Inhibition of RANKL-induced osteoclastogenesis by silencing Notch2 mRNA. Data for RAW cells with stable expression of tetracycline-regulated expression of shRNAs (Tet-on) that target Notch1 (RAW-teton-shNotch1) (A) and Notch2 (RAW-teton-shNotch2) (B) are shown. RAW-teton-shNotch1 and RAW-teton-shNotch2 cells were treated with RANKL (20 ng/ml) for 3 days. TRAP⁺ MNCs were counted as osteoclasts. Scale bar = 100 μm. The graph values are means ± SEM (n = 3). Both RAW-teton-shNotch1 and RAW-teton-shNotch2 cells were treated with or without tetracycline for 3 days. Total cell lysates were immunoblotted with anti-Notch1 or anti-Notch1 ICD (A, right) or anti-Notch2 (B, right) antibodies. β-Actin is shown as a loading control. The images of anti-Notch1 or anti-Notch1 ICD were taken from same membrane by reblotting.

The active form of Notch2 positively regulates RANKL-induced osteoclastogenesis. BMMs were transduced with N1IC^ΔOP, N2IC^ΔOP, or empty pMX-IRES-EGFP retroviral vectors. (A) Images of GFP⁺ MNCs from BMM cells induced by RANKL stimulation. Scale bar = 100 μm. (B) Total cell lysates were immunoblotted with anti-Notch1, -Notch2, and -myc antibodies or anti-β-actin as a loading control. (C) Transduced BMMs were treated with RANKL (0, 25, or 50 ng/ml) for 3 days. TRAP⁺ MNCs were counted as osteoclasts. The values are means ± SEM (n = 3).

Expression of Notch ligands in osteoblasts and osteoclast precursors. (A) Expression of Notch ligand-related genes in mice treated with RANKL (50 ng/ml). Total RNA was isolated from osteoblasts or BMMs, and expression levels of indicated genes were measured by RT-PCR. (B) BMMs were treated with RANKL (50 ng/ml) in the presence of soluble FLAG-fused Jagged1 (Jagged-1 FL) (0.5, 1, or 2 μg/ml) for 3 days. (C) RAW-teton-shNotch2 cells were treated with RANKL (20 ng/ml) in the presence of either the soluble form or immobilized Jagged1 for 3 days with or without tetracycline. TRAP⁺ MNCs were counted as osteoclasts. Values are means ± SEM (n = 3).

Molecular mechanisms by which Notch modulates RANKL-induced osteoclastogenesis. (A) Expression of NFATc1 induced by RANKL in mouse BMMs in the presence of soluble Jagged1, transduced Notch2, shRNA for Notch2, or GSI. Cells were treated with RANKL (50 ng/ml) for the indicated times. Total cell lysates were immunoblotted with anti-NFATc1 antibody. IB, immunoblotting. (B) Schematic representation of the NFATc1 P1 promoter. The sequences of the NF-κB binding site and the overlapping putative RBPJκ binding site are shown. (C) Chromatin from RANKL-treated BMMs was precipitated using the indicated antibodies. The NFATc1 promoter was amplified by PCR from precipitated DNA and from 1% of the input to monitor the amount of chromatin used. (D) Interaction of Notch2 and p65 in BMMs upon RANKL stimulation. Nuclear extracts from RANKL-treated BMMs were immunoprecipitated (IP) with anti-Notch2 antibodies and processed for immunoblotting with anti-p65 antibodies. The membrane was stripped and reblotted with anti-Notch2 antibodies to determine the amount of precipitated proteins. (E) RAW cells were transfected with the NFATc1 p1 promoter and the indicated plasmids and assayed for luciferase activity after 24 h. (F) Luciferase activity of NFATc1 p1 when cotransfected with the indicated plasmids in RAW cells. (G) ChIP assay with anti-p65, anti-Notch2, anti-RBPJκ, or control immunoglobulin G from RAW-teton-shNotch2 cells treated with RANKL with or without tetracycline. Analysis of the NFATc1 promoter was performed by PCR. (H) Chromatin treated with RANKL from RAW cells with knocked down RBPJκ expression was precipitated with the indicated antibodies. The NFATc1 promoter was amplified from precipitated DNA and 1% of the input by PCR. Similar results were obtained in three independent experiments.