Sfrp4 repression of the Ror2/Jnk cascade in osteoclasts protects cortical bone from excessive endosteal resorption

Abstract

Loss-of-function mutations in the Wnt inhibitor secreted frizzled receptor protein 4 (SFRP4) cause Pyle's disease (OMIM 265900), a rare skeletal disorder characterized by wide metaphyses, significant thinning of cortical bone, and fragility fractures. In mice, we have shown that the cortical thinning seen in the absence of Sfrp4 is associated with decreased periosteal and endosteal bone formation and increased endocortical resorption. While the increase in Rankl/Opg in cortical bone of mice lacking Sfrp4 suggests an osteoblast-dependent effect on endocortical osteoclast (OC) activity, whether Sfrp4 can cell-autonomously affect OCs is not known. We found that Sfrp4 is expressed during bone marrow macrophage OC differentiation and that Sfrp4 significantly suppresses the ability of early and late OC precursors to respond to Rankl-induced OC differentiation. Sfrp4 deletion in OCs resulted in activation of canonical Wnt/β-catenin and noncanonical Wnt/Ror2/Jnk signaling cascades. However, while inhibition of canonical Wnt/β-catenin signaling did not alter the effect of Sfrp4 on OCgenesis, blocking the noncanonical Wnt/Ror2/Jnk cascade markedly suppressed its regulation of OC differentiation in vitro. Importantly, we report that deletion of Ror2 exclusively in OCs (CtskCreRor2^fl/fl ) in Sfrp4 null mice significantly reversed the increased number of endosteal OCs seen in these mice and reduced their cortical thinning. Altogether, these data show autocrine and paracrine effects of Sfrp4 in regulating OCgenesis and demonstrate that the increase in endosteal OCs seen in Sfrp4^-/- mice is a consequence of noncanonical Wnt/Ror2/Jnk signaling activation in OCs overriding the negative effect that activation of canonical Wnt/β-catenin signaling has on OCgenesis.

Keywords: Pyle’s disease; Ror2/Jnk; Sfrp4; Wnt signaling; endocortical remodeling.

Fig. 1.

Sfrp4 cell-autonomously regulates OC differentiation and activity. (A) Sfrp4 expression during Rankl-induced OCgenesis of wt BMMs (n = 5). (B and C) TRAP staining (B) and quantification (C) in wt and Sfrp4^−/− cultures (n = 9). (D) OC-specific gene expression in wt and Sfrp4^−/− OCs (n = 5). (E and F) Pit assay: representative images of dentin seeded with wt or Sfrp4^−/− cells (E) and quantification (F) of bone resorption area (n = 5). All data are mean ± SEM; open circles, wt; black triangles, Sfrp4^−/− cells. *P < 0.05, **P < 0.005, ***P < 0.0001. Student’s t test vs. day (D)0 or vs. wt cells. (Scale bars, 100 μm.)

Fig. 2.

sFrp4 treatment prevents early and late progenitor cells from developing into mature OCs. (A and B) TRAP staining (A) and quantification (B) in wt OCs treated w/wo increasing doses of sFrp4 (n = 9). (C) OC-specific gene expression in OCs treated w/wo sFrp4 (10 μg/mL) (n = 5). All data are mean ± SEM; open circles, vehicle; black circles, sFrp4-treated cells. **P < 0.005, ***P < 0.0001. Student’s t test vs. vehicle-treated cells. (Scale bars, 100 μm.)

Fig. 3.

Effects of BMM-secreted and OB-secreted Sfrp4 on OCgenesis. (A) Sfrp4 gene expression in wt OCs and cOBs (n = 3 to 5). (B and C) TRAP staining (B) and quantification (C) in mixed-and-matched cocultures of BMMs and cOBs. All data are mean ± SEM. Two-way ANOVA followed by Tukey’s test. a < 0.0001 vs. cOBwt/BMMwt, b < 0.0001 vs. cOBwt/BMMSfrp4^−/−, c < 0.0001 vs. cOBSfrp4^−/−/BMMwt (n = 6).

Fig. 4.

Sfrp4 regulates both Wnt/β-catenin canonical and Wnt/Jnk noncanonical cascades in OCs. p-Jnk, total Jnk, and active β-catenin levels in wt and Sfrp4^−/− BMMs. All data are mean ± SEM; open circles, wt; black triangles, Sfrp4^−/− cells. **P < 0.005, ***P < 0.0001. Student’s t test vs. wt (n = 5 to 6).

Fig. 5.

Sfrp4 regulation of OCgenesis is independent of canonical Wnt/β-catenin signaling. (A) Active β-catenin, p-Jnk, and total Jnk levels in Sfrp4^−/− OCs in the presence or absence of XAV939. Data are mean ± SEM; *P < 0.05, **P < 0.005 vs. untreated cells, ^#P < 0.05 vs. 1 μM XAV939. Student’s t test (n = 3). (B and C) TRAP staining (B) and quantification (C) in Sfrp4^−/− OCs treated w/wo XAV939. All data are mean ± SEM; ***P < 0.001 vs. untreated cells, ^#P < 0.05 vs. 1 μM XAV939. Student’s t test (n = 9). (D) Lrp6, active β-catenin, p-Jnk, and total Jnk levels in Tam-treated and untreated (control) Cre-ER^T2;Lrp5/6^fl/fl BMMs. Data are mean ± SEM; ***P < 0.001 vs. control. Student’s t test (n = 6). (E and F) TRAP staining (E) and quantification (F) in Tam-treated and control Cre-ER^T2;Lrp5/6^fl/fl OCs in the presence of 10 μg/mL sFrp4 (black squares) or vehicle (open squares) (n = 5). All data are expressed as mean ± SEM; a < 0.0001 vs. control+vehicle, b < 0.0001 vs. Tam-treated+vehicle. Two-way ANOVA followed by Tukey’s test (n = 5). (G) Sfrp4 efficacy in regulating TRAP⁺ MNCs in Cre-ER^T2;Lrp5/6^fl/fl OCs. Data are mean ± SEM (n = 5). (Scale bars, 100 μm.)

Fig. 6.

Sfrp4 regulates osteoclastogenesis via the noncanonical Wnt/Ror2/Jnk cascade. (A) p-Jnk, total Jnk, and active β-catenin levels in Sfrp4^−/− OCs w/wo SP600. Values are expressed as mean ± SEM; *P < 0.05, **P < 0.005 vs. untreated cells, ^#P < 0.05 vs. 2.5 μM SP600. Student’s t test (n = 3). (B and C) TRAP staining (B) and quantification (C) in Sfrp4^−/− OCs treated w/wo SP600 (n = 9). Data are mean ± SEM; ***P < 0.001 vs. untreated cells, ^###P < 0.05 vs. 2.5 μM SP600. Student’s t test. (D) Ror2, p-Jnk, total Jnk, and active β-catenin levels in Tam-treated and untreated (control) Cre-ER^T2;Ror2^fl/fl BMMs (n = 5 to 6). Data are mean ± SEM; ***P < 0.001 vs. control cells. Student’s t test. (E and F) TRAP staining (E) and quantification (F) in Tam-treated and control Cre-ER^T2;Ror2^fl/fl OCs in the presence of 10 μg/mL sFrp4 (black squares) or vehicle (open squares) (n = 5). Data are mean ± SEM; a < 0.0001 vs. control+vehicle. Two-way ANOVA followed by Tukey’s test. (G) Sfrp4 efficacy in regulating TRAP⁺ MNCs in Cre-ER^T2;Ror2^fl/fl OCs. Data are mean ± SEM; ***P < 0.001 vs. control cells. (Scale bars, 100 μm.)

Fig. 7.

Sfrp4 suppresses Wnt5-dependent induction of OCgenesis via the noncanonical Wnt/Ror2/Jnk cascade. (A and B) TRAP staining (A) and quantification (B) in wt Rankl-induced OCs treated w/wo Wnt5a (100 ng/mL) or sFrp4 (10 μg/mL) (n = 5). Data are mean ± SEM; ***P < 0.01 vs. untreated cells, ^###P < 0.001 vs. Wnt5a-treated cells. Student’s t test. (C) p-Jnk and total Jnk levels in wt BMMs w/wo Wnt5a or sFrp4 (n = 4). (Scale bars, 100 μm.)

Fig. 8.

OC-specific deletion of Ror2 in Sfrp4^−/− mice protects their cortical bone from excessive endosteal resorption. Skeletal phenotype of cortical bone of 5-wk-old male mice. (A) Representative μCT images. (Scale bars, 2 μm.) (B) Quantification of cortical bone parameters by μCT. Ct.Ar/Tt.Ar, bone area fraction; Ct.Th, cortical thickness. (C) Representative Von Kossa staining images. (Scale bars, 250 μm.) (D) Histomorphometric analysis. Ct.Th, cortical thickness; En.N.Oc/BS, endosteal osteoclst number/bone surface. All data are mean ± SEM (n = 4 to 7); ^aP < 0.005 compared with Ror2^fl/fl;Sfrp4^+/+ mice, ^bP < 0.005 compared with Ror2_OC;Sfrp4^+/+ mice, ^cP < 0.005 compared with Ror2^fl/fl;Sfrp4^−/− mice. Two-way ANOVA followed by Tukey’s test. (E) Working model. See text.