BMP-9 induces both smad1/5/8 and Smad2/3 phosphorylation, and specific response genes, in chondrocytes

Abstract

Background: Osteoarthritis is characterized by loss of articular cartilage. The main cartilage degrading enzyme is Matrix metalloproteinase 13 (MMP13). Expression of MMP13 is positively correlated with Activin receptor-like kinase 1 (ALK1) expression. Moreover, terminal differentiation of chondrocytes is induced by ALK1 via phosphorylation of Smad1/5/8 (Smad1/5/8p). BMP-9, also known as GDF-2, has recently been characterized as a potent, high-affinity ALK1 ligand, that is present in serum in high amounts. We wanted to further investigate intracellular BMP-9 signaling in chondrocytes to clarify its role in cartilage maintenance and degradation. Purpose: To study Smad1/5/8 and Smad2/3 phosphorylation kinetics in response to addition of exogenous BMP-9, to investigate BMP-9’s potency on chondrocytes, and to investigate downstream signaling responses on mRNA level. Methods: Both primary bovine chondrocytes, isolated from the metacarpophalangeal joint of 2 year old animals, as well as the murine H4 chondrocyte cell line were used in this study. Chondrocytes were cultured to near confluency, and subsequently incubated with BMP-9. After short term stimulation (1 h), Smad phosphorylation was analyzed by specific Smad2/3p and Smad1/5/8p staining of Western blots. After long term stimulation (24-48 h), expression of PAI-1, a downstream marker of Smad2/3 signaling, and ID-1, a downstream marker of Smad1/5/8 signaling, were analyzed by quantitative real time PCR (qPCR). Results: In both the murine cell line, as well as the primary bovine chondrocytes, addition of BMP-9 resulted in phosphorylation of Smad 1/5/8, that was maximal at very low concentrations (100 pg / ml). Unexpectedly, Smad2/3p was also observed at this concentration. Further investigations showed that BMP-9-induced Smad2/3p was maximal after 1 h, at the physiological relevant concentration of 5 ng/ml. Exposure of chondrocytes to BMP9 not only resulted in both Smad1/5/8p and Smad2/3p but also in transcription of pathway specific response genes, as observed by a significant increase in PAI-1 and ID-1 mRNA levels after both 24 and 48h. In both cell types, collagen type 2A1 and aggrecan mRNA levels were also up regulated by BMP9. In the described experiments, no significant changes in MMP13 expression were observed after 48 h of BMP-9 stimulation. Conclusion: In chondrocytes, BMP9 is capable of inducing both Smad1/5/8p and Smad2/3p. By doing so it can up regulate both PAI-1 and ID-1 simultaneously. BMP-9 is very potent in inducing Smad phosphorylation, as very low concentrations (pg/ml) are sufficient for signal induction. The induction of Smad2/3p and subsequent PAI-1 activity sets BMP-9 apart from other BMPs, like BMP-2 and BMP-6, that only induce Smad1/5/8 phosphorylation. The lack of MMP13 expression after BMP-9 stimulation can be explained by the induced Smad2/3p expression, as this potently blocks MMP13 transcription. Based on these results we propose that exposure of young cartilage to BMP-9 does not result in MMP13 expression due to its ability to induce Smad2/3p. However, we previously reported that in ageing cartilage the ability to phosphorylate Smad2/3 is reduced, and therefore BMP-9 exposure will most likely result in MMP13 production in old cartilage.