Unloading results in rapid loss of TGFβ signaling in articular cartilage: role of loading- induced TGFβ signaling in maintenance of articular chondrocyte phenotype?  W. Madej, A. van Caam, E. Blaney Davidson, P. Buma, P.M. van der Kraan  Osteoarthritis and Cartilage  Volume 24, Issue 10, Pages 1807-1815 (October 2016) DOI: 10.1016/j.joca.2016.05.018 Copyright © 2016 Osteoarthritis Research Society International Terms and Conditions

Fig. 1 Schematic representation of experiments conducted in this study. (A) The effect of unloading on TGF β signaling in articular cartilage was investigated by measuring Smad2P and Smad2/3P responsive genes on indicated time points. (B) Effect of repeated physiological mechanical compression on TGFβ signaling in articular cartilage was investigated by measuring Smad2P and Smad2/3P responsive genes on indicated time points before and 2 h and 48 h after compression for multiple compressions. (C) Effect of loading on GAG content and Col10a1 expression was measured by culturing cartilage explants for 2 weeks ex vivo. Explants were compressed every 48 h the first week and hereafter either cultured for 1 week in serum free medium for Col10a1 or compressed again for GAG measurement. Osteoarthritis and Cartilage 2016 24, 1807-1815DOI: (10.1016/j.joca.2016.05.018) Copyright © 2016 Osteoarthritis Research Society International Terms and Conditions

Fig. 2 Loss of active Smad2/3 signaling in unloaded cartilage. (A) Phosphorylated Smad2 levels in adult cartilage after unloading, as detected by IHC. Representative individual shown. (B) Relative gene expression of the TGFβ-responsive genes: Alk5, Smad7 and Pai1 24 h and 48 h after unloading compared to fresh samples. Tukey box plot, + = mean, N = 7, *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001. (C) The effect of the ALK4/5/7 blocker SB-505124 on relative gene expression of bAlk5, bSmad7 and bPai1 24 h and 48 after unloading, compared to fresh samples. DMSO was used as vehicle control. Tukey box plot, + = mean, N = 7, N.S. = not significant, *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001. (D) Relative gene expression of Aggrecan or Smad3 24 h and 48 h after unloading. Tukey box plot, + = mean, N = 7. Osteoarthritis and Cartilage 2016 24, 1807-1815DOI: (10.1016/j.joca.2016.05.018) Copyright © 2016 Osteoarthritis Research Society International Terms and Conditions

Fig. 3 Repeatable restoration of active Smad2/3 signaling by dynamic mechanical compression. (A) Induction of Smad2 phosphorylation by dynamic mechanical compression (3 MPa) of adult articular cartilage, as detected by IHC. (B) Relative gene expression of the TGFβ-responsive genes: Alk5, Smad7 and Pai in response to repeated dynamic mechanical compression and unloading. Tukey box plot, + = mean, N = 4, *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001. n.d. = not detected. (C) Relative gene expression of Smad7 in response to repeated dynamic mechanical compression and unloading in the presence of the ALK4/5/7 inhibitor; SB-505124. DMSO was used as vehicle control. Tukey box plot, N = 4, + = mean, N.S. = not significant, *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001. (D) Scheme reflecting the effects of dynamic compression on TGFβ-signaling in the absence or presence of SB-505124. (E) Relative gene expression of Alk1 in (repeatedly) dynamically compressed (3 MPa) cartilage compared to unloaded controls 48 h after compression. Tukey box plot, + = mean, N = 8, ***P ≤ 0.001. Osteoarthritis and Cartilage 2016 24, 1807-1815DOI: (10.1016/j.joca.2016.05.018) Copyright © 2016 Osteoarthritis Research Society International Terms and Conditions

Fig. 4 Dynamic mechanical compression protects against chondrocyte hypertrophy. (A) The sulfated GAG content (w/w) of adult articular cartilage, cultured for 2 weeks ex vivo either with or without 10% FCS, 10 ng/ml TGFβ1 or subjected to dynamic mechanical compression (3 MPa). Tukey box plot, + = mean, N = 8, ***P ≤ 0.001. (B) GAG content (w/w) of articular cartilage cultured for 2 weeks ex vivo in serum free medium, or in the presence of either: 10% FCS, 20 ng/ml IGF1, 10 ng/ml TGFβ1, or a combination of IGF1 and TGFβ1. Tukey box plot, N = 9, *P ≤ 0.05, ***P ≤ 0.001. (C) Expression of the hypertrophy marker Col10a1 after 2 weeks ex vivo culture in serum free medium or in the presence of 10% FCS, 1 or 10 ng/ml TGFβ1, or 50 ng/ml Activin A. Tukey box plot, + = mean, N = 4 (for Activin A, N = 2), ***P ≤ 0.001. (D) The effect of repeated dynamic mechanical compression (3 MPa) on Col10a1 expression after 2 weeks ex vivo culture. Tukey box plot, + = mean, N = 6. **P ≤ 0.01, ***P ≤ 0.001. Osteoarthritis and Cartilage 2016 24, 1807-1815DOI: (10.1016/j.joca.2016.05.018) Copyright © 2016 Osteoarthritis Research Society International Terms and Conditions

Fig. 5 Proposed hypothetical model for compression-mediated protection of articular cartilage integrity. Compression of articular cartilage leads to release of active TGFβ from the ECM. This active TGFβ signals via ALK5, resulting in phosphorylation of Smad2/3. Subsequently, phosphorylated Smad2/3 inhibits hypertrophic differentiation of articular chondrocytes, as characterized by Col10a1. Additionally, phosphorylated Smad2/3 activates a positive feedback loop by not only inducing expression of ALK5 and lowering expression of ALK1, but also by inducing the expression of inactive TGFβ1, which after production will bind to the ECM, returning the system to its original state. Osteoarthritis and Cartilage 2016 24, 1807-1815DOI: (10.1016/j.joca.2016.05.018) Copyright © 2016 Osteoarthritis Research Society International Terms and Conditions