BACKGROUND

Both Wnt signaling and TGF-β signaling have been implicated in the regulation of the phenotype of many cell types including chondrocytes, the only cell type present in the articular cartilage. A changed chondrocyte phenotype, resulting in chondrocyte hypertrophy, is one of the main hallmarks of osteoarthritis. TGF-β signaling via activin-like kinase (ALK)5, resulting in Smad 2/3 phosphorylation, inhibits chondrocyte hypertrophy. In contrast, TGF-β signaling via ALK1, leading to Smad 1/5/8 phosphorylation, has been shown to induce chondrocyte hypertrophy. In this study, we investigated the capability of Wnt3a and WISP1, a protein downstream in canonical Wnt signaling, to skew TGF-β signaling in chondrocytes from the protective Smad 2/3 towards the Smad 1/5/8 pathway.

RESULTS

Stimulation with Wnt3a, either alone or in combination with its downstream protein WISP1, decreased TGF-β-induced C-terminal phosphorylation of Smad 2/3. In addition, both Wnt3a and WISP1 increased Smad 1/5/8 phosphorylation at the C-terminal domain in both murine and human chondrocytes. DKK-1, a selective inhibitor of canonical Wnt signaling, abolished these effects. TGF-β signaling via Smad 2/3, measured by the functional CAGA12-Luc reporter construct activity, was decreased by stimulation with Wnt3a in accordance with the decrease in Smad 2/3 phosphorylation found on Western blot. Furthermore, in vivo overexpression of the canonical Wnt8a decreased Smad 2/3 phosphorylation and increased Smad 1/5/8 phosphorylation.

CONCLUSIONS

Our data show that canonical Wnt signaling is able to skew TGF-β signaling towards dominant signaling via the ALK1/Smad 1/5/8 pathway, which reportedly leads to chondrocyte hypertrophy. In this way canonical Wnts and WISP1, which we found to be increased during experimental osteoarthritis, may contribute to osteoarthritis pathology.