Involvement of MAPK signaling molecules and Runx2 in the NELL1-induced osteoblastic differentiation

Abstract

NELL1 is an extracellular protein inducing osteogenic differentiation and bone formation of osteoblastic cells. To elucidate the intracellular signaling cascade evoked by NELL1, we have shown that NELL1 protein transiently activates the MAPK signaling cascade, induces the phosphorylation of Runx2, and promotes the rapid intracellular accumulation of Tyr-phosphorylated proteins. Unlike BMP2, NELL1 protein does not activate the Smad signaling cascade. These findings suggest that upon binding to a specific receptor NELL1 transduces an osteogenic signal through activation of certain Tyr-kinases associated with the Ras-MAPK cascade, and finally leads to the osteogenic differentiation.

Fig. 1

Schematic structure of hNELL1 and sequence comparison with laminin G family proteins. Protein motifs were searched in the SMART database accessible at the SMART website (http://smart.embl-heidelberg.de/) and shown by boxes for the laminin G domain (LG), von Willebrand factor C domains (V), EGF-like domains (E), and Ca²⁺-binding EGF-like domains (E^*). Signal peptide (aa 1–19) is indicated by a gray bar. Partial sequence alignment of laminin G domains was carried out by the CLUSTAL W protocol using a BLOSUM62 amino acid substitution matrix available at the DNA Data Bank of Japan website (http://clustalw.ddbj.nig.ac.jp/). hNELL1, human NELL1; hLAMA3, human laminin α3 chain; mLAMA3, mouse laminin α3 chain; hCOLA1, human collagen 11 α3 chain; and hTSP1, human thrombospondin-1. Residue numbers of the core regions of laminin G domains are indicated in the margins. Strand assignments for β-strands (C–J) are according to Hohenester et al. [25]. Conserved amino acid residues are shaded. Glu residues presumed as the binding site for integrin α6β1 are indicated in boldface.

Fig. 2

Induction of osteomarkers by hNELL1-VH. RFC cells were treated with various concentrations (or 100 ng/ml in B and C) of the purified hNELL1-VH protein or 100 ng/ml hBMP2 protein. (A) Intracellular ALP activities in the 3-day cultured cells were measured. (B) Amounts of OPN secreted into the 6-day cultured medium were measured. (C) Amounts of OCN secreted into the 9-day cultured medium were measured. Each value represents the means ± S.E.M. of at least four times measurements. (*) indicate that the data are significantly different from the control (P < 0.005, t-test).

Fig. 3

Activation of MAPK by hNELL1-VH. RFC cells (A), C3H10T1/2 cells (B), and Saos-2 cells (C) were treated with 100 ng/ml hNELL1-VH protein (derived from 125 μl medium), 100 ng/ml hBMP2 protein, or mock sample (derived from 2 ml medium) for the indicated times. Cell lysates (15 μg protein) were analyzed by Western blotting either with anti-phospho-ERK1/2 (p-ERK1/2), anti-phospho-JNK1/2/3 (p-JNK1/2/3), and anti-phospho-p38α/β/γ (p-p38) antibodies. Anti-ERK1/2 (ERK1/2), anti-JNK1/2/3 (JNK1/2/3), anti-p38α/β/γ (p38), and anti-β-actin (actin) antibodies were used as internal controls. All experiments were repeated more than three times, and representative data are shown.

Fig. 4

Effects of Ras siRNA on OPN expression. C3H10T1/2 cells were transfected with Ras siRNA on 1st day, and then treated with 100 ng/ml hNELL1-VH protein for 15 min on 2nd day. (A) Cell lysates (15 g protein) were analyzed by Western blotting as described in Fig. 3. (B) Cells were cultured for 6 days in the presence of hNELL1-VH. Amounts of OPN secreted into the culture medium were measured by ELISA on 7th day. Each value represents the means ± S.E.M. of triplicate measurements. (*) indicate that the data are significantly different from the control (DMSO, P < 0.005, t-test).

Fig. 5

Phosphorylation of Smad proteins. RFC (A) and Saos-2 (B) cells were treated with 100 ng/ml hNELL1-VH or 100 ng/ml hBMP2 protein for the indicated times. Cell lysates (15 μg protein) were analyzed by Western blotting with anti-phospho-Smad1/5/8 (p-Smad1/5/8), anti-Smad1/5/8 (Smad1/5/8), and anti-actin (actin) antibodies. All experiments were repeated at least three times, and representative data are shown. (C) C3H10T1/2 cells were treated with 100 ng/ml hNELL1-VH or 100 ng/ml hBMP2 protein at 37 °C for 1 h. Intracellular localization of Smad4 was analyzed by the immunocytochemical method using anti-Smad4 and Alexa-546-conjugated anti-mouse IgG antibodies (bar, 50 μm). (D) RFC cells transformed with the luciferase gene under the Id1-985 or mutated Id1 (Id1-985MutB) promoter were treated with 10 or 100 ng/ml hNELL1-VH or 100 ng/ml hBMP2 protein for 24 h. Smad-dependent transcriptional activity was determined as intracellular luciferase activity. Each value represents the means ± S.E.M. of four times measurements. (*) indicates that the data are significantly different from the control (P < 0.005, t-test).

Fig. 6

Phosphorylation of Runx2. C3H10T1/2 cells were transfected with Ras siRNA on 1st day, and then treated with 100 ng/ml hNELL1-VH protein for 15 min on 2nd day. The anti-Runx2 immunoprecipitates were analyzed by Western blotting using an anti-phosphoserine antibody (anti-pSer) (upper panel) or an anti-Runx2 antibody (anti-Runx2) (lower panel). All experiments were repeated more than three times, and representative data are shown.

Fig. 7

Detection of Tyr-phosphorylated proteins interacting with hNELL1-VH. RFC cells were treated with 100 ng/ml hNELL1-VH protein for 0, 2, and 5 min. Cell lysates were subjected to immunoprecipitation with either mouse IgG (A) or an anti-V5 tag antibody (B) and analyzed by Western blotting using an anti-phosphotyrosine antibody 4G10. Protein G released from the affinity resin was detected as non-specific bands (white asterisks). Tyr-phosphorylated proteins are indicated by arrowheads.