TRAF family member-associated NF-κB activator (TANK) induced by RANKL negatively regulates osteoclasts survival and function

Abstract

Osteoclasts are the principle bone-resorbing cells. Precise control of balanced osteoclast activity is indispensable for bone homeostasis. Osteoclast activation mediated by RANK-TRAF6 axis has been clearly identified. However, a negative regulation-machinery in osteoclast remains unclear. TRAF family member-associated NF-κB activator (TANK) is induced by about 10 folds during osteoclastogenesis, according to a genome-wide analysis of gene expression before and after osteoclast maturation, and confirmed by western blot and quantitative RT-PCR. Bone marrow macrophages (BMMs) transduced with lentivirus carrying tank-shRNA were induced to form osteoclast in the presence of RANKL and M-CSF. Tank expression was downregulated by 90% by Tank-shRNA, which is confirmed by western blot. Compared with wild-type (WT) cells, osteoclastogenesis of Tank-silenced BMMs was increased, according to tartrate-resistant acid phosphatase (TRAP) stain on day 5 and day 7. Number of bone resorption pits by Tank-silenced osteoclasts was increased by 176% compared with WT cells, as shown by wheat germ agglutinin (WGA) stain and scanning electronic microscope (SEM) analysis. Survival rate of Tank-silenced mature osteoclast is also increased. However, acid production of Tank-knockdown cells was not changed compared with control cells. IκBα phosphorylation is increased in tank-silenced cells, indicating that TANK may negatively regulate NF-κB activity in osteoclast. In conclusion, Tank, whose expression is increased during osteoclastogenesis, inhibits osteoclast formation, activity and survival, by regulating NF-κB activity and c-FLIP expression. Tank enrolls itself in a negative feedback loop in bone resorption. These results may provide means for therapeutic intervention in diseases of excessive bone resorption.

Keywords: NF-κB; Osteoclast.; RANKL; TANK.

Figure 1

Expression of TANK is induced by RANKL. (A) Induction folds of expression level of TANK and other osteoclast marker genes by RANKL, based on GeneChip data. (B) Time course analysis of TANK transcription level by semi-quantitative RT-PCR in RANKL and M-CSF stimulated bone marrow macrophages (BMMs) on Day(D)0, D1, D2, D3, D4 and D5. (C) Western blot analysis of time-dependent TANK protein expression level in RANKL and M-CSF stimulated bone marrow macrophages (BMMs) on Day(D)0, D1, D2, D3, D4 and D5. (D) Quantification of western blot results in B. Data is expressed as mean ± sd (n=5).

Figure 2

Expression of TANK in vivo. (A) Immunofluorescence staining of frozen femur sections from 10 day old mice shows that tank expression (green) and Cathepsin K expression (red) are partially merged. (B) Analysis of tank transcription levels in different tissues by semi-quantitative RT-PCR. (C) Quantification of RT-PCR results in B. (D,E) Immunohistochemistry staining of paraffin sections from femur of 1 month old mice shows TANK is expressed in both bone (marked by arrow) and bone marrow (arrow). And high expression can be detected on the surface of trabecular bone (marked by star).

Figure 3

TANK knockdown in BMM promotes osteoclast differentiation. (A) Western blot analysis of BMMs (mock), BMMs infected with lentivirus carrying scrambled shRNA (control), BMMs which were infected by lentivirus carrying shRNA targeting tank expression (tank-shRNA-1 and tank-shRNA-2) and treated with M-CSF to confirm the knockdown of TANK. (B) Quantification of TANK protein expression level in osteoclasts with and without viral infection. (C, D) TRAP staining of mature osteoclasts from the mock group, the control-shRNA group, the tank-shRNA-1 group, and the tank-shRNA-2 group on Day 5 (upper row) and on Day 7 (bottom row). (E) Quantification of the number of TRAP+ multiple nucleic (≥3 nuclei) cells per random view on days 5 and 10. (F) Detection of IκBα phosphorylation in BMM induced by RANKL by western blot. 1, mock; 2, control; 3, shRNA-1; 4, shRNA-2. Data is expressed as mean ± sd (n=5).

Figure 4

TANK knockdown in BMM doesn't influence osteoclast acidification but promotes bone resorption. (A) Acridine orange staining of osteoclasts from mock group, control-shRNA group, and tank-shRNA-1 group. (B, C) Osteoclasts from mock, control-shRNA, and tank-shRNA-1 groups were cultured on cattle bone slides and analyzed for bone resorption activity through scanning electron microscopy (SEM) (B) and wheat germ agglutinin (WGA) staining (C). (D) Quantification of bone resorption area per view area in B. Data is expressed as mean ± sd (n=15). *p≤0.05.

Figure 5

TANK knock-down promotes osteoclast survival through promoting Nf-κB signaling. (A) TRAP staining of osteoclasts from the mock group, the control-shRNA group, and the tank-shRNA group on Day 5 and Day 10. (B) Mature osteoclasts from the mock group, the control-shRNA group, and the tank-shRNA group were cytokine and serum starved for 8hrs. mRNA was derived for analysis of pro-survival genes c-FLIP, Bc1-x1, and c-IAP expression by RT-PCR. (C-D) Mature osteoclasts from the mock group, the control-shRNA group, and the tank-shRNA group were cytokine and serum starved for 8hrs and apoptosis cells (red nuclei) were detected by TUNEL staining. (E) quantification of C&D. Data is expressed as mean ± sd (n=5). *p≤0.05.