Dual-Function Semaphorin 4D Released by Platelets: Suppression of Osteoblastogenesis and Promotion of Osteoclastogenesis

Abstract

Effects of the antiosteoblastogenesis factor Semaphorin 4D (Sema4D), expressed by thrombin-activated platelets (TPs), on osteoblastogenesis, as well as osteoclastogenesis, were investigated in vitro. Intact platelets released both Sema4D and IGF-1. However, in response to stimulation with thrombin, platelets upregulated the release of Sema4D, but not IGF-1. Anti-Sema4D-neutralizing monoclonal antibody (mAb) upregulated TP-mediated osteoblastogenesis in MC3T3-E1 osteoblast precursors. MC3T3-E1 cells exposed to TPs induced phosphorylation of Akt and ERK further upregulated by the addition of anti-sema4D-mAb, suggesting the suppressive effects of TP-expressing Sema4D on osteoblastogenesis. On the other hand, TPs promoted RANKL-mediated osteoclastogenesis in the primary culture of bone-marrow-derived mononuclear cells (BMMCs). Among the known three receptors of Sema4D, including Plexin B1, Plexin B2 and CD72, little Plexin B2 was detected, and no Plexin B1 was detected, but a high level of CD72 mRNA was detected in RANKL-stimulated BMMCs by qPCR. Both anti-Sema4D-mAb and anti-CD72-mAb suppressed RANKL-induced osteoclast formation and bone resorptive activity, suggesting that Sema4D released by TPs promotes osteoclastogenesis via ligation to a CD72 receptor. This study demonstrated that Sema4D released by TPs suppresses osteogenic activity and promotes osteoclastogenesis, suggesting the novel property of platelets in bone-remodeling processes.

Keywords: bone regeneration; osteoblasts; osteoclasts; platelets; semaphorin 4D.

Figure 1

Sema4D production from TPs. Flow cytometry was performed to detect Sema4D on CD61+ platelets stimulated with and without thrombin (0.5, 1 unit/mL) (A). Sema4D and β-actin were detected in platelets cultured with/without thrombin (1 unit/mL) by Western blotting (B). Sema4D and IGF-1 in supernatant of platelets cultured with and without thrombin (1 unit/mL) were detected by ELISA (C,D). Data represent the mean ± SD of three independent experiments. Values are mean ± S.D. * p < 0.05, ** p < 0.01. ND: No statistically significant difference. +: Applied. −: Not applied.

Figure 2

The effect of Sema4D-neutralized TPs on osteoblastogenesis. MC3T3-E1 cells were cultured with and without recombinant human Sema4D (10 ng/mL). After 7 days, ALP staining was performed, and after 21 says, alizarin red staining was performed (A). MC3T3-E1 cells were cultured with and without platelets (1 × 10⁴ cells) and anti-Sema4D antibody (50 μg/mL) or control IgG (50 μg/mL). After 7 days, ALP staining was performed, and after 21 days, alizarin red staining was performed (B). Data represent the mean ± SD of three independent experiments. Values are mean ± S.D. * p < 0.05, ** p < 0.01. +: Applied. −: Not applied.

Figure 3

The effect of Sema4D and IGF-1 derived from TPs on ERK and Akt phosphorylation in MC3T3-E1 cells. Western blotting was performed to detect total and phosphorylated ERK and Akt in MC3T3-E1 cells cultured with/without ant-Sema4D mAb (50 µg/mL) or control mAb (50 µg/mL) in the presence of TPs (A) and total and phosphorylated ERK and Akt in MC3T3-E1 cells cultured with/without ant-IGF-1 mAb (1 µg/mL) in the presence of TPs (B). ALP activity was measured in TP-mediated MC3T3-E1 cells with and without ERK or Akt inhibitor (C). The mRNA expression of PlexinB1, PlexinB2 and CD72 in MC3T3-E1 cells was evaluated by qPCR (D). Data represent the mean ± SD of three independent experiments. Values are mean ± S.D. * p < 0.05, ** p < 0.01. +: Applied. −: Not applied.

Figure 4

The effect of TPs on osteoclastogenesis and the expression of receptor of Sema4D in BMMCs. BMMCs were cultured in osteoclastogenic medium (Basal medium supplemented with M-CSF/RANKL) with or without TPs (5 × 10³ and 1 × 10⁴ cells). The number of TRAP-positive multinuclear cells (over three nuclei) were counted (A). BMMCs were stimulated with TPs (1 × 10⁴ cells) in the presence or absence of anti-IGF-1 mAb or control mAb. The number of TRAP-positive multinuclear cells (over three nuclei) were counted (B). ELISA was carried out to detect sSema4D in supernatant of M-CSF/RANKL-stimulated BMMCs or thrombin-stimulated platelets (C). The mRNA expression of PlexinB1, PlexinB2 and CD72 in BMMCs stimulated with MCSF/RANKL for 24 h was monitored by qPCR, whose level was then expressed in comparison to the housekeeping gene, Actb (D). The protein expression of PlexinB2 and CD72 was observed through immunofluorescence staining. The images in the first row are indicated as “PlexinB2” while those in the bottom row are indicated as “CD72”. Blue (DAPI): Nuclei, White (Alexa 647): Phalloidin, Red (PE): Plexin B2 or CD72, and Green (FITC): Sema4D. Bar = 10 µm. (E). Data represent the mean ± SD of three independent experiments. Values are mean ± S.D. * p < 0.05, ** p < 0.01. ND: No statistically significant difference. +: Applied. −: Not applied.

Figure 5

The effect of anti-Sema4D mAb and anti-CD72mAb on RANKL-dependent osteoclastogenesis. BMMCs were cultured in osteoclastogenic medium and TPs (1 × 10⁴ cells) with or without anti-Sema4D mAb, anti-Cd72 mAb or control mAb. The number of TRAP-positive multinuclear cells (over three nuclei) was counted (A), and pit formation area was evaluated (B). Data represent the mean ± SD of three independent experiments. Values are mean ± S.D. * p < 0.01. ND: No statistically significant difference. Bar = 50 µm. +: Applied. −: Not applied.