Rab11A Functions as a Negative Regulator of Osteoclastogenesis through Dictating Lysosome-Induced Proteolysis of c-fms and RANK Surface Receptors

Abstract

Osteoclast differentiation and activity are controlled by two essential cytokines, macrophage colony-stimulating factor (M-CSF) and the receptor activator of nuclear factor-κB ligand (RANKL). Rab11A GTPase, belonging to Rab11 subfamily representing the largest branch of Ras superfamily of small GTPases, has been identified as one of the crucial regulators of cell surface receptor recycling. Nevertheless, the regulatory role of Rab11A in osteoclast differentiation has been completely unknown. In this study, we found that Rab11A was strongly upregulated at a late stage of osteoclast differentiation derived from bone marrow-derived macrophages (BMMs) or RAW-D murine osteoclast precursor cells. Rab11A silencing promoted osteoclast formation and significantly increased the surface levels of c-fms and receptor activator of nuclear factor-κB (RANK) while its overexpression attenuated osteoclast formation and the surface levels of c-fms and RANK. Using immunocytochemical staining for tracking Rab11A vesicular localization, we observed that Rab11A was localized in early and late endosomes, but not lysosomes. Intriguingly, Rab11A overexpression caused the enhancement of fluorescent intensity and size-based enlargement of early endosomes. Besides, Rab11A overexpression promoted lysosomal activity via elevating the endogenous levels of a specific lysosomal protein, LAMP1, and two key lysosomal enzymes, cathepsins B and D in osteoclasts. More importantly, inhibition of the lysosomal activity by chloroquine, we found that the endogenous levels of c-fms and RANK proteins were enhanced in osteoclasts. From these observations, we suggest a novel function of Rab11A as a negative regulator of osteoclastogenesis mainly through (i) abolishing the surface abundance of c-fms and RANK receptors, and (ii) upregulating lysosomal activity, subsequently augmenting the degradation of c-fms and RANK receptors, probably via the axis of early endosomes-late endosomes-lysosomes in osteoclasts.

Keywords: NFATc-1; RANK; Rab11A; c-fms; osteoclast; vesicular transport.

Figure 1

Rab11A upregulation at a late stage of osteoclast differentiation. (A,B) RAW-D cells (A) or BMMs (B) were treated with RANKL over the indicated time course. Total expression levels of c-Fos, NFATc-1, Rab11A, and GAPDH used as a loading control were evaluated by WB. (C,D) Quantitative analyses of Western blot for c-fms and RANK, NFATc1, in RAW-D cells (C) or BMMs (D). GAPDH was used as an internal control. * p < 0.05, (E,F) TRAP staining was carried out to assess the formation of mature osteoclasts differentiated from RAW-D cells (E) or from BMMs (F) upon RANKL stimulation over a time course. Arrowheads indicated the mature osteoclasts. Scale bars: 200 μm. Data shown were the representative of three independent experiments.

Figure 2

The effects of Rab11A silencing on osteoclast differentiation. (A) RAW-D cells were transfected with nontargeting (Ctrl) or Rab11A-specific siRNA for 24 h without RANKL stimulation. The knockdown efficacy of Rab11A mRNA was analyzed by qRT-PCR. (B) RAW-D cells were transfected with Ctrl si or Rab11A si for 24 h, followed by RANKL stimulation for 3 days. The knockdown efficacy of Rab11A mRNA levels was analyzed by RT-qPCR. (C,D) Upper: The endogenous level of Rab11A protein was assessed by WB on day 0 (C) and day 3 (D). Lower: Column scatter plotting to compare Rab11A protein level on day 0 (C) and day 3 (D). (E) TRAP staining of Ctrl si or Rab11A si-treated osteoclasts. The cells were treated with Ctrl si or Rab11A si 24 h, followed by RANKL stimulation for 3 days. An arrowhead indicated the mature osteoclasts. Scale bars: 200 μm. (F,G) The number of TRAP-positive osteoclasts with 3–10 nuclei (F), or with more than 10 nuclei (G) per viewing field was counted. * p < 0.05, ** p < 0.01, n = 3. Data are the representative of three independent experiments.

Figure 3

The effects of Rab11A silencing on BMM-derived osteoclast differentiation. (A) Upper: BMM cells were transfected with nontargeting siRNA (Ctrl si) or Rab11A-specific siRNA (Rab11A si) for 24 h without RANKL stimulation. Lower: Column scatter plotting to compare Rab11A protein level. (B) Upper: BMM cells were transfected with Ctrl siRNA or Rab11A siRNA for 24 h, followed by RANKL stimulation for 3 days. The endogenous level of Rab11A protein was evaluated by Western blotting. Lower: Column scatter plotting to compare Rab11A protein level. (C) TRAP staining of osteoclast transfected Ctrl si or Rab11A si. Arrowheads showed large osteoclasts. Bars: 200 μm. (D) Images of the bone resorption area of BMM-derived osteoclasts transfected with Ctrl si or Rab11A si. Bars: 100 μm. (E) The resorption area was determined using Image J software. The data are represented as mean ± SD of values from three independent experiments. * p < 0.05, compared to control cells.

Figure 4

The effect of Rab11A overexpression on osteoclastogenesis. (A) GFP or GFP-Rab11A expression was determined in RAW-D cells transduced with either retrovirus vector encoding GFP or GFP-tagged Rab11A (GFP-Rab11A). An “asterisk” indicated the GFP-Rab11A band. (B) TRAP staining of GFP and GFP-Rab11A-expressing osteoclasts derived from RAW-D cells. Arrowheads showed mature osteoclasts. Scale bars: 200 μm. (C,D) The number of TRAP-positive multinucleated osteoclasts with 3–10 nuclei or more than 10 nuclei per viewing field was counted. ** p < 0.01. (E) Images of the bone resorption area of RAW-D-derived osteoclasts expressing GFP-alone or GFP- Rab11A. Scale bars: 100 μm. (F) The resorption area was determined using Image J software. The data are represented as mean ± SD of values from three independent experiments.

Figure 5

Subcellular localization of Rab11A in RAW-D cells and RAW-D cell-derived osteoclasts expressing GFP-Rab11A (green). (A,B) RAW-D cells (A) or osteoclasts following RANKL stimulation for 3 days (B) were seeded on cover glasses, permeabilized with 0.2% Triton X-100 diluted in PBS, subsequently reacted with one of the antibodies against Rab5, Rab7, or LAMP1 (red, as indicated). DNA was stained with DAPI (blue). Arrowheads indicated the positive region of GFP-Rab11A and each organelles-specific markers. Scale bars: 5 μm. The images shown were the representative of three independent experiments.

Figure 6

The effects of Rab11A silencing on cell surface levels of c-fms and RANK receptors in RAW-D cells stimulated with RANKL (300 ng/mL). (A) The endogenous levels of c-fms, RANK, and NFATc1 in Ctrl or Rab11A si-treated RAW-D cells stimulated with RANKL for 0 (left panels) or 3 days (right panels). (B) The endogenous levels of c-fms, RANK and NFATc1 in Ctrl or Rab11A si-treated BMMs stimulated with RANKL for 0 (left panels) or 3 days (right panels). (C,D) Quantitative analyses of Western blot for c-fms, RANK, NFATc1, in RAW-D cells (C) or BMMs (D). GAPDH was used as an internal control. * p < 0.05. (E) Upper: The biotinylated fractions were subjected to immunoblotting with anti-mouse c-fms and anti-mouse RANK antibodies, and whole cell lysates (WCLs) were subjected to immunoblotting with anti-GAPDH antibody as a loading control in RAW-D cells following RANKL stimulation for 3 days. Lower: Quantitative analyses of Western blot for c-fms, RANK. * p < 0.05. (F) Flow cytometric analyses of TfR using Ctrl or Rab11A siRNA-transfected RAW-D cells, followed by RANKL stimulation for 2 days. Data are representative of three independent experiments.

Figure 7

The effects of Rab11A overexpression on cell surface levels of c-fms and RANK receptors in RAW-D cells upon RANKL stimulation. (A) The endogenous levels of c-fms, RANK, and NFATc1 in RAW-D cells expressing GFP (control) or GFP- Rab11A following RANKL stimulation for 0 or 3 days. (B) Quantitative analyses of Western blot for c-fms, RANK, NFATc1. GAPDH was used as an internal control. * p < 0.05. (C) Upper: The biotinylated fractions were subjected to immunoblotting with anti-mouse c-fms and anti-mouse RANK antibodies, and WCLs were subjected to immunoblotting with GAPDH-HRP antibody as a loading control in RAW-D cells expressing GFP (control) or GFP-Rab11A. Lower: Quantitative analyses of Western blot for c-fms, RANK. (D) Flow cytometric analyses of TfR were done using RAW-D cells expressing GFP (control) or GFP-Rab11A. Data are the representative of three independent experiments.

Figure 8

The lysosomal function on c-fms and RANK protein degradation ameliorated by Rab11A overexpression in RAW-D cell-derived osteoclasts. (A) The endogenous levels of LAMP1 (a specific lysosomal receptor), and Cathepsins B and D (two major lysosomal enzymes) in GFP or GFP-Rab11A-expressing RAW-D cells following RANKL stimulation for 0 (left panel) or 3 days (right panel) were analyzed by WB with anti-rat LAMP1, anti-rabbit Cathepsins B and D, and GAPDH-HRP (loading control). (B) Quantitative analyses of Western blot for LAMP1, Cathepsin B, Cathepsin D. GAPDH was used as an internal control. * p < 0.05. (C) After 3 days of RANKL (300 ng/mL) stimulation, RAW-D cell-derived osteoclasts were treated simultaneously with 20 μg/mL CHX and with or without 10 μM CLQ for 3 h. c-fms and RANK protein levels were assessed by WB with anti-mouse c-fms, anti-mouse RANK, and GAPDH-HRP (loading control) antibodies. (D) Quantitative analyses of Western blot for c-fms and RANK. GAPDH was used as an internal control. * p < 0.05. (E) Cell viability was assessed by the cellular ATP content measurement using the CTG Assay system. After stimulated with RANKL (300 ng/mL) for 3 days, RAW-D cells were subsequently added with 20 μg/mL CHX and with or without 10 μM CLQ for 3 h. The values were the average of triplicate determinations with the S.D indicated by error bars. n.s; no significant.