FNDC4 Inhibits RANKL-Induced Osteoclast Formation by Suppressing NF- κ B Activation and CXCL10 Expression

Abstract

FNDC4 acts as an anti-inflammatory factor on macrophages and improves mouse model of induced colitis. Considering osteoclast formation is characterized by the activation of inflammation-related pathways, we thus speculated that FNDC4 may play a pivotal role in this process. RT-qPCR analysis was performed to confirm the expression of osteoclast formation related genes in primary murine bone marrow macrophages (BMMs). RANKL-treated BMMs were cultured with FNDC4 to evaluate the effect of FNDC4 on osteoclast differentiation. TRAP staining and bone resorption pits assay were used to assess osteoclast formation and bone resorption, respectively. Luciferase assay and western blotting analysis were conducted to determine whether FNDC4 inhibited osteoclast formation via NF-κB signaling in RAW 264.7 cells. Furthermore, to identify gene signatures in FNDC4-treated BMMs and to use these to elucidate the underlying molecular mechanisms during osteoclast formation, we adopted a bioinformatics approach by downloading the GSE76172 gene expression profiling dataset from the Gene Expression Omnibus (GEO) database. FNDC4 inhibited RANKL-induced osteoclastogenesis and mature osteoclast resorptive function in a dose-dependent manner. Results of NF-κB luciferase assay suggested that FNDC4 could significantly suppress the RANKL-induced NF-κB transcriptional activity. Based on the protein-protein interaction network, CXCL10 was identified as the differentially expressed gene with the highest connectivity degree (degree = 23); it was drastically downregulated in the presence of FNDC4, but supplementation of CXCL10 (10 ng/mL) partially ameliorated the FNDC4-induced inhibition of osteoclast formation. Taken together, we speculated that FNDC4 could suppress osteoclast formation via NF-κB pathway and downregulation of CXCL10.

Figure 1

FNDC4 inhibited osteoclast formation in a dose-dependent manner. BMMs were cultured with different concentrations of FNDC4 (0, 100, 500, and 1000 ng/mL) in the presence of M-CSF (30 ng/mL) and RANKL (100 ng/mL) for 5 days. (a) Representative images of TRAP-stained cells treated with different concentrations of FNDC4 and TRAP-positive multinucleated cells with more than three nuclei were counted as osteoclasts. (b) Quantification of TRAP-positive osteoclasts showed that FNDC4 at 500 ng/mL and 1000 ng/mL could significantly reduce the number of TRAP-stained osteoclasts. (c) CCK-8 analysis was conducted to assess the effect of FNDC4 on the proliferation and viability of BMMs. All experiments were performed in triplicate and results were presented as M ± SD;  ^∗p < 0.05 and ^∗∗p < 0.01 relative to the RANKL group.

Figure 2

FNDC4 suppressed osteoclast function. BMMs were cultured with different concentrations of FNDC4 (0, 100, 500, and 1000 ng/mL) in the presence of M-CSF (30 ng/mL) and RANKL (100 ng/mL) for 7 days in a Corning Osteo Assay Surface plate. (a) Representative images of Osteo Assay plates after the removal of osteoclasts. The bone resorption area in the plates was measured. (b) Quantification of bone resorption area showed that FNDC4 at 500 ng/mL and 1000 ng/mL could significantly reduce the bone resorption area compared with RANKL group. All experiments were performed in triplicate and results were presented as M ± SD. ^∗p < 0.05 and ^∗∗p < 0.01 compared with the RANKL group.

Figure 3

FNDC4 inhibited the expression of osteoclast-related genes (TRAP, CtsK, and NFATc1). BMMs were treated with 1000 ng/mL FNDC4 in the presence of M-CSF (30 ng/mL) and RANKL (100 ng/mL) for 2 days. (a) CtsK. (b) TRAP. (c) NFATc1. FNDC4 could significantly inhibit the RANKL-induced expression of these mRNAs. All experiments were performed in triplicate and results were presented as M ± SD. ^∗p < 0.05 and ^∗∗p < 0.01 compared with the RANKL group.

Figure 4

FNDC4 inhibited RANKL-induced NF-κB activity. (a) BMMs were cultured with M-CSF (30 ng/mL) and RANKL (100 ng/mL) in the presence of FNDC4 (1000 ng/mL) for indicated times. The total cellular lysates were subject to western blotting analysis and immunoblotted with indicated antibodies. (b) The grey level of p-P65 was analyzed by being normalized to total P65. (c) The grey level of p-IκB-α was analyzed by being normalized to β-actin. (d) RAW 264.7 cells were stably transfected with luciferase reporter constructs controlled by NF-κB binding promoter elements. Cells plated in a 48-well plate were pretreated with different concentrations of FNDC4 (100 ng/mL, 500 ng/mL, and 1000 ng/mL) for 1 hour, followed by stimulation with RANKL (100 ng/mL) for another 8 hours; then the luciferase activity was measured. Data were presented as M ± SD. ^∗p < 0.05 and ^∗∗p < 0.01 compared with the RANKL group.

Figure 5

Protein-protein interaction (PPI) network of differentially expressed genes (DEGs). Red nodes and green nodes illustrate upregulated DEGs and downregulated DEGs, respectively. Color depth represents fold-change, with deeper colors indicating greater fold-change. Nodes with higher degree values are depicted with larger sizes.

Figure 6

FNDC4 inhibited the osteoclast formation via downregulation of CXCL10. (a) FNDC4 could significantly suppress the expression of CXCL10 in BMMs. BMMs were treated with M-CSF (30 ng/mL) and RANKL (100 ng/mL) in the presence of absence of 1000 ng/mL FNDC4 for 2 days; then total RNA was collected for qPCR analysis. (b) BMMs were treated with M-CSF (30 ng/mL) and RANKL (100 ng/mL) in the presence of absence of 1000 ng/mL FNDC4 for 2 days; then the medium was collected to detect the concentration of CXCL10 using ELISA. (c) The selective silencing of CXCR3 was confirmed by western blotting; si#3 was selected for later experiments. (d, e) The addition of CXCL10 could attenuate the FNDC4-induced inhibition of osteoclast formation. BMMs were incubated with RANKL (100 ng/mL), M-CSF (30 ng/mL), and FNDC4 (1000 ng/mL) with or without the supplementation of CXCL10 (10 ng/mL) and siRNA against CXCR3 for 5 days. Representative images of TRAP-stained cells and quantitative analysis of TRAP-positive cells were presented for three independent experiments. Data were presented as M ± SD. ^∗p < 0.05 and ^∗∗p < 0.01.