Identification of a human peripheral blood monocyte subset that differentiates into osteoclasts

Abstract

Increased bone resorption mediated by osteoclasts causes various diseases such as osteoporosis and bone erosion in rheumatoid arthritis (RA). Osteoclasts are derived from the monocyte/macrophage lineage, but the precise origin remains unclear. In the present study, we show that the purified CD16- human peripheral blood monocyte subset, but not the CD16+ monocyte subset, differentiates into osteoclast by stimulation with receptor activator of NF-kappaB ligand (RANKL) in combination with macrophage colony-stimulating factor (M-CSF). Integrin-beta3 mRNA and the integrin-alpha(v)beta3 heterodimer were only expressed on CD16- monocytes, when they were stimulated with RANKL + M-CSF. Downregulation of beta3-subunit expression by small interfering RNA targeting beta3 abrogated osteoclastogenesis from the CD16- monocyte subset. In contrast, the CD16+ monocyte subset expressed larger amounts of tumor necrosis factor alpha and IL-6 than the CD16- subset, which was further enhanced by RANKL stimulation. Examination of RA synovial tissue showed accumulation of both CD16+ and CD16- macrophages. Our results suggest that peripheral blood monocytes consist of two functionally heterogeneous subsets with distinct responses to RANKL. Osteoclasts seem to originate from CD16- monocytes, and integrin beta3 is necessary for osteoclastogenesis. Blockade of accumulation and activation of CD16- monocytes could therefore be a beneficial approach as an anti-bone resorptive therapy, especially for RA.

Figure 1

Induction of osteoclasts from human peripheral blood monocytes. (a) Purified CD16⁺ and CD16^- peripheral blood monocytes were cultured with either macrophage colony-stimulating factor (M-CSF) (25 ng/ml) alone or with M-CSF (25 ng/ml) + receptor activator of NF-κB ligand (RANKL) (40 ng/ml) for 7 days and were stained for tartrate-resistant acid phosphatase (TRAP) activity. Original magnification, ×100. (b) The number of TRAP-positive multinucleated cells (MNC) (three or more nuclei) that differentiated from each monocyte subset was counted. (c) Resorbtive activity was assessed by culturing monocytes on plates coated with calcium phosphate films. The cells were treated with M-CSF (25 ng/ml) and RANKL (40 ng/ml) for 7 days. Arrows show resorbed lacunae. Original magnification, ×100. (d) Culture supernatants of CD16⁺ and CD16^- were collected on day 7, and the concentrations of TRAP-5b and MMP-9 were measured with an ELISA. Representative data of more than three independent experiments are shown. Data represent the mean ± standard error of the mean values of duplicate or triplicate wells. *P < 0.01. Scale bars = 100 μm.

Figure 2

Effect of CD16⁺ monocytes on the osteoclastogenesis from CD16^- monocytes. CD16⁺monocytes (0, 1 × 10³, 2.5 × 10³, 5 × 10³ cells/well) were mixed with CD16^- monocytes (5 × 10⁴ cells/well) in 96-well plates, and were cultured for 7 days in the presence of macrophage colony-stimulating factor (M-CSF) + receptor activator of NF-κB ligand (RANKL). The number of tartrate-resistant acid phosphatase (TRAP)-positive multinucleated cells (MNC) induced was counted. Representative data of two independent experiments are shown. Data represent the mean ± standard error of the mean values of quadriplicate wells. N.S., not significant.

Figure 3

Tumor necrosis factor α and IL-6 production by monocyte subsets with stimulation. Purified CD16⁺ and CD16^- peripheral blood monocytes were incubated either with receptor activator of NF-κB ligand (RANKL) (0, 40, 200, 1000 ng/ml) or macrophage colony-stimulating factor (M-CSF) (0, 25, 125 ng/ml) for 24 hours. Tumor necrosis factor alpha (TNFα) and IL-6 concentrations in the culture supernatant were measured by ELISA. Results are representative of more than three independent experiments. Open squares, CD16⁺ monocytes; filled squares, CD16⁺ monocytes. Data are the mean ± standard error of the mean values of duplicate wells. *P < 0.03, no stimulation vs either RANKL or M-CSF stimulation; ^†P < 0.03, CD16⁺ vs the CD16^- monocyte subset.

Figure 4

Differences in expression pattern of molecules related to osteoclastogenesis between CD16⁺ and CD16^- monocyte subsets. (a) Total RNA was extracted from freshly isolated CD16⁺ and CD16^- monocytes or from the cells incubated in either macrophage colony-stimulating factor (M-CSF) (25 ng/ml) alone or M-CSF (25 ng/ml) + receptor activator of NF-κB ligand (RANKL) (40 ng/ml) for 3 days, and semiquantitative RT-PCR analysis was performed. Representative results from three independent experiments are shown. (b) The expression of receptor activator of NF-κB (RANK) and c-fms in unstimulated or stimulated monocytes was analyzed by western blotting. (c) Cell surface expression of c-fms on unstimulated CD16⁺ and CD16^- monocytes was examined by three-color flow cytometry. Gates were set either for CD14⁺ CD16⁺ (left panel) or CD14⁺ CD16⁺ (right panel) monocytes. Histograms show the stained cells with anti-c-fms mAb (solid lines) and isotype-matched control (dotted lines). (d) Purified monocytes were allowed to adhere on plates overnight (unstimulated) or the cells treated with M-CSF (25 ng/ml) + RANKL (40 ng/ml) were examined for the expression of the αvβ3 heterodimer by immunofluorescent staining. Solid arrows indicate mononuclear αvβ3-positive cells. Dotted arrows indicate multinucleated αvβ3-positive cells. Original magnification, ×100. Results are representative of two independent experiments.

Figure 5

Flow cytometric analysis of ERK1/2 and p38 MAPK phosphorylation on monocyte subsets. Purified monocytes were precultured with macrophage colony-stimulating factor (M-CSF) for 3 days, and treated with 40 ng/ml receptor activator of NF-κB ligand (RANKL) for 5 min (pink), 10 min (blue) or 20 min (orange), or were left untreated (light green). The cells were then stained either with phospho-ERK1/2 (T202/Y204) or phospho-p38 MAPK (T180/Y182) after fixation and permeabilization. Isotype controls were shown in dotted line. The data shown are representative of three independent experiments.

Figure 6

Effect of transfection of integrin-β3 siRNA and cyclic RGDfV peptide on osteoclastogenesis from CD16^- monocytes. (a) CD16^- monocytes transfected with either 1 nM control or integrin-β3 siRNA were cultured in macrophage colony-stimulating factor (M-CSF) (25 ng/ml) + receptor activator of NF-κB ligand (RANKL) (40 ng/ml). Forty-eight hours after the transfection, integrin-β3 mRNA levels were examined by semiquantitative RT-PCR. (b) The expression of the αvβ3 heterodimer was examined by immunostaining and the number of αvβ3-positive cells was counted. (c) Seven days after the transfection of siRNAs, the cells were stained for tartrate-resistant acid phosphatase (TRAP) activity, and the number of TRAP-positive multinucleated cells (MNC) was counted. Results are representative of three to five independent experiments. Data are the mean ± SEM values of duplicate wells. *P < 0.01, control-siRNA vs β3-siRNA. (d) CD16^- monocytes were incubated with M-CSF (25 ng/ml) + RANKL (40 ng/ml) for 2 days, followed by the addition of a medium containing either cyclic RGDfV peptide or dimethyl sulfoxide (DMSO). After incubation for a further 5 days, the number of TRAP-positive multinucleated cells (MNC) was counted. Representative results from three independent experiments are shown. Data are the mean ± standard error of the mean values of triplicate wells. Control, without treatment. **P < 0.03, DMSO vs cyclic RGDfV peptide.

Figure 7

Effect of integrin-β3 knockdown on induction of NFATc1 mRNA. CD16^- monocytes transfected with either control or integrin-β3 siRNA were cultured with macrophage colony-stimulating factor (M-CSF) (25 ng/ml) + receptor activator of NF-κB ligand (RANKL) (40 ng/ml). Total RNA was extracted 48 hours post-transfection. Semiquantitative RT-PCR analysis was performed using NFATc1-specific and GAPDH-specific primers. Representative results from four independent experiments are shown.

Figure 8

Double immunofluorescence showing CD16⁺ and CD16^- macrophages in rheumatoid arthritis synovium. Synovial tissue samples from patients with rheumatoid arthritis (RA) were stained with CD68 and CD16. (a) CD68, (b) CD16, and (c) merged (a) with (b). Arrows show CD16⁺ cells. Original magnification, ×400. Representative results from four RA patients are shown. Scale bar = 50 μm.