BMPR1A Is Necessary for Chondrogenesis and Osteogenesis

RESEARCH ARTICLE BMPR1A is necessary for chondrogenesis and osteogenesis, whereas BMPR1B prevents hypertrophic differentiation Tanja Mang1,2, Kerstin Kleinschmidt-Doerr1, Frank Ploeger3, Andreas Schoenemann4, Sven Lindemann1 and Anne Gigout1,*

ABSTRACT essential for osteogenesis and bone formation during this process BMP2 stimulates bone formation and signals preferably through BMP (Bandyopadhyay et al., 2006; McBride et al., 2014; Yang et al., 2013). – receptor (BMPR) 1A, whereas GDF5 is a cartilage inducer and signals Similarly, during bone fracture healing where a similar mechanism – preferably through BMPR1B. Consequently, BMPR1A and BMPR1B are takes place conditional deletion of Bmp2 in mesenchymal believed to be involved in bone and cartilage formation, respectively. progenitors or osteoprogenitors prevents fracture healing (Mi et al., However, their function is not yet fully clarified. In this study, GDF5 2013; Tsuji et al., 2006). In vitro, BMP2 provokes an induction of mutants with a decreased affinity for BMPR1A were generated. These alkaline phosphatase (ALP) activity, osteocalcin expression and matrix mutants, and wild-type GDF5 and BMP2, were tested for their ability to mineralization in pluripotent mesenchymal progenitor cells (Cheng induce dimerization of BMPR1A or BMPR1B with BMPR2, and for their et al., 2003), and also stimulates chondrogenesis or adipogenesis (Date chondrogenic, hypertrophic and osteogenic properties in chondrocytes, et al., 2004). Finally, BMP2 has been shown to promote bone repair in in the multipotent mesenchymal precursor cell line C3H10T1/2 and the animal models (Kleinschmidt et al., 2013; Wulsten et al., 2011) and in human osteosarcoma cell line Saos-2. Mutants with the lowest potency a clinical setting (Ronga et al., 2013). for inducing BMPR1A–BMPR2 dimerization exhibited minimal Growth differentiation factor 5 (GDF5; also named BMP14 or chondrogenic and osteogenic activities, indicating that BMPR1A is CDMP1) is also expressed in the developing limb at sites where joint necessary for chondrogenic and osteogenic differentiation. BMP2, GDF5 cavitation occurs, and null mutation of Gdf5 in mice disrupts the and the GDF5 R399E mutant stimulated expression of chondrogenic and formation of 30% of the synovial joints (Storm and Kingsley, 1996). hypertrophy markers in C3H10T1/2 cells and chondrocytes. However, Several gene expression or deletion studies strongly indicate that GDF5 R399E, which induces the dimerization of BMPR1B and BMPR2 GDF5 (together with GDF6 and GDF7) provides a specialized more potently than GDF5 or BMP2, displayed reduced hypertrophic function for joint morphogenesis (Settle et al., 2003; Storm and activity. Therefore, we postulate that stronger BMPR1B signaling, Kingsley, 1999). In vitro, GDF5 promotes chondrogenic compared to BMPR1A signaling, prevents chondrocyte hypertrophy differentiation and chondrocyte hypertrophy (Coleman et al., 2013; and acts as a cartilage stabilizer during joint morphogenesis. Erlacher et al., 1998). Subcutaneous implantation of collagen or collagen/hyaluronate loaded with GDF5 in the rat results in cartilage This article has an associated First Person interview with the first author formation followed by bone formation (Erlacher et al., 1998; Spiro of the paper. et al., 2000). In bone defect models, GDF5 caused delayed tissue mineralization compared to BMP2 and the presence of cartilage tissue KEY WORDS: GDF5, BMP2, M1673, Chondrogenesis, Hypertrophy, in the defect (Kleinschmidt et al., 2013; Wulsten et al., 2011). Finally, Osteogenesis intra-articular injections of GDF5 stimulated cartilage repair in a rat osteoarthritis model (Parrish et al., 2017). In conclusion, both BMP2 INTRODUCTION and GDF5 promote cartilage and bone formation, but BMP2 more Bone morphogenetic proteins (BMPs) play a role in a multitude of strongly promotes osteogenesis and bone formation than does GDF5. processes during embryonic development, including skeletal BMP2 and GDF5 elicit their effects through two types of serine/ development. Most skeletal elements are formed by endochondral threonine kinase transmembrane receptors; type I and type II ossification, which is initiated by mesenchymal cell condensation. receptors. There are three type II and seven type I receptors that Under the influence of several BMPs, these cells differentiate into interact with BMPs, and the association of a type I with a type II chondrocytes that produce cartilaginous tissue, and further differentiate receptor is required for the formation of an active signaling complex. into hypertrophic chondrocytes. Afterward, vascularization takes place BMP2 and GDF5 are both known to interact with all three type II concomitantly with replacement of cartilage into bone, driven by receptors (BMPR2, ACVR1 and ACVR2) together with the type I osteoblasts and osteoclasts (Katagiri and Watabe, 2016; Salazar et al., receptors BMPR1A or BMPR1B. However, BMP2 and GDF5 2016). Several conditional deletion studies have shown that BMP2 is preferentially interact with BMPR1A and BMPR1B, respectively (Heinecke et al., 2009; Nishitoh et al., 1996). In accordance with the 1Osteoarthritis Research, Merck KGaA, 64293 Darmstadt, Germany. 2Institute for role of BMPs in skeletal development, the receptors Bmpr1a and Organic Chemistry and Biochemistry, Technische Universität, 64289 Darmstadt, Bmpr1b are expressed in the developing limbs (Baur et al., 2000; Germany. 3Biopharm GmbH, 69214 Eppelheim, Germany. 4Discovery Dewulf et al., 1995; Zou et al., 1997). Several studies have aimed to Pharmacology, Merck KGaA, 64293 Darmstadt, Germany. decipher the role of BMPR1A and BMPR1B in skeletal development. *Author for correspondence ([email protected]) In vivo, the use of loss-of-function or gain-of-function mutations revealed some redundancy in the role of these two BMP type I T.M., 0000-0002-6497-3034; A.G., 0000-0003-1399-7957 receptors (Kobayashi et al., 2005; Yoon et al., 2005). In vitro, Handling Editor: John Heath contradictory results have been published. For instance, it has been

Received 26 March 2020; Accepted 21 July 2020 shown that constitutively active forms of BMPR1A and BMPR1B Journal of Cell Science

1 RESEARCH ARTICLE Journal of Cell Science (2020) 133, jcs246934. doi:10.1242/jcs.246934

activate chondrogenesis in ADTC5 cells (Fujii et al., 1999), whereas BMP2, GDF5, W417R and W417F. Similar to the Kd values, the forced expression of BMPR1A, but not BMPR1B, stimulates ratio of EC50 values were calculated. The same selectivity profile osteogenesis and chondrogenesis in C3H10T1/2 cells (Kaps et al., was obtained; BMP2 was more selective for BMPR1A than 2004). BMPR1B and the selectivity was shifted toward BMPR1B with In conclusion, the respective role of BMPR1A and BMPR1B in GDF5, R399E, W417F and W417R. chondrogenesis and osteogenesis is not yet fully clear. In this study, we tackled this question from a different angle. Although genetic BMPR expression, Nog sensitivity and Nog expression manipulations are extremely helpful, the results should be interpreted To ensure that the cells selected to study chondrogenesis and with caution. Silencing the expression of one protein can be osteogenesis are relevant, we verified that they express the receptors compensated by the expression of other similar proteins (Rossi we intend to study. The C3H10T1/2 cells, Saos-2 cells and porcine et al., 2015), while overexpression induces non-physiological protein chondrocytes all express Bmpr1a/BMPR1A, Bmpr1b/BMPR1B and expression levels and might generate results that do not accurately Bmpr2/BMPR2 (Fig. S1). In addition, because the proteins we use reflect the biology. As an alternative, we generated several mutants of in this work derive from the human GDF5 sequence and were tested GDF5 that differentially activate BMPR1A or BMPR1B. These on human receptors for the affinity or dimerization measurements, mutants were tested together with wild-type GDF5 and BMP2 for we also evaluated the percentage identity between human, porcine their chondrogenic, hypertrophic and osteogenic properties in the or mouse BMPRs (see Table S1). For all sequences, a high identity multipotent mesenchymal precursor cell line C3H10T1/2, primary was found (>96%) and variations within the sequence are limited to chondrocytes and the human osteosarcoma cell line Saos-2. regions that are not involved into ligand binding (Kotzsch et al., 2009; Weber et al., 2007; Mace et al., 2006). RESULTS Finally, we also evaluated the noggin (Nog) sensitivity of the Binding affinity of BMP2, GDF5 and the GDF5 mutants as GDF5 mutants. Nog is an inhibitor of BMPs that can be produced by measured by surface plasmon resonance chondrocytes or C3H10T1/2 cells (Kameda et al., 2000; Zehentner The binding affinities of BMP2, GDF5 and the three GDF5 mutants et al., 2002). It was previously shown that Noggin inhibits GDF5, for BMPR2, BMPR1A and BMPR1B were measured by surface and that specific mutations of GDF5 can suppress Noggin plasmon resonance (SPR) (Table 1). The affinity measurements of inhibition. (Seemann et al., 2009). We found cells that stimulated BMP2, GDF5, R399E and W417F with BMPR2 resulted in Kd with GDF5, R399E and BMP2 were strongly sensitive to Nog values in the 2-digit nM range, while no binding of W417R could be treatment but those stimulated with W417F and W417R were less detected. The Kd values obtained for BMPR1B were in the 2- to affected by Nog (Fig. S1). We also observed that GDF5 treatment 3-digit pM range. GDF5 showed the strongest affinity for BMPR1B, caused increased Nog expression in C3H10T1/2 cells but treatment followed by BMP2, R399E, W417F and W417R. Regarding of R399E, W417R or W417F did not. BMPR1A, BMP2 showed the strongest affinity, followed by GDF5 To summarize, the chosen cells express all three receptors, and and the GDF5 mutants, which displayed strongly reduced affinities porcine and murine BMPRs present a high identity to the human for BMPR1A; the affinity of R399E and W417F for BMPR1A was BMPRs. However, we do not know whether mouse or porcine 6 and 41 times lower compared to GDF5 and no binding of W417R BMPRs behave in a similar manner to the human receptor, which is to BMPR1A could be detected. The ratios of the Kd BMPR1A to Kd a limitation of the present study. In addition, all mutants were found for BMPR1B were calculated to describe the binding preferences of to cause different sensitivities to Nog to that of GDF5, and they did the compounds to BMPR1A or BMPR1B. As previously described, not stimulate Nog expression while GDF5 did. compared to BMP2, GDF5 has an increased binding preference for BMPR1B over BMPR1A (Heinecke et al., 2009). Here, the GDF5 Chondrogenesis of C3H10T1/2 mutants R399E and W417F showed a strongly increased selectivity C3H10T1/2 cells were cultured as pellets in a chondrogenic for BMPR1B in comparison to GDF5. medium in the presence of 300 ng/ml of BMP2, GDF5, R399E, W417R or W417F. After 14, 21 or 28 days, samples were harvested Dimerization of BMPRs for the analysis of chondrogenic and hypertrophic marker BMP2, GDF5 and the GDF5 mutants were further evaluated in a expression (Fig. 2). cellular assay to determine their ability to dimerize BMPR2 with In the absence of compound (control), none of the analyzed genes BMPR1A or BMPR1B (Fig. 1). All compounds were found to be were modulated at day 14 in comparison to day 0. At later time active in both assays. The corresponding EC50 Kd values were points it was impossible to isolate RNA, indicating a low level of calculated (Table 2). BMP2 was the most potent compound for the cellular activity. Similarly, no RNA could be isolated from the cells dimerization of BMPR2 with BMPR1A, followed by R399E, cultured with W417R (data not shown). GDF5, W417F and W417R. Regarding the dimerization of BMPR2 With GDF5 treatment, the expression of Acan, Col2a1 and with BMPR1B, R399E was the most potent compound, followed by Sox9 were increased, which indicates that GDF5 promotes

Table 1. Dissociation constants evaluated by SPR and KDBMPR1A/KDBMPR1B ratios

Kd BMPR1A: Kd BMPR2 (pM) Kd BMPR1A (pM) Kd BMPR1B (pM) Kd BMPR1B BMP2 10,400 119 72 1.7 GDF5 26,500 2080 42 50 R399E 23,000 12,400 110 113 W417F 15,000 85,900 115 747 W417R Not determined Not determined 273 Not determined

The ratio of the Kd for BMPR1A to the Kd for BMPR1B (Kd BMPR1A:Kd BMPR1B) describes the preference of one compound for BRMP1B over BMPR1A. No binding to BMPR2 or BMPR1A could be measured for W417R. Journal of Cell Science

2 RESEARCH ARTICLE Journal of Cell Science (2020) 133, jcs246934. doi:10.1242/jcs.246934

Fig. 1. Cellular assay for BMPR dimerization. U2OS cells expressing BMPR2 and BMPR1A or BMPR1B were used. A fluorescent signal is generated upon binding and dimerization of BMPR2 with BMPR1A or BMPR1B, and was recorded after treatment with indicated concentration of each molecule. RFU, relative fluorescence unit. Journal of Cell Science

3 RESEARCH ARTICLE Journal of Cell Science (2020) 133, jcs246934. doi:10.1242/jcs.246934

Table 2. EC50 values calculated from a functional cellular assay chondrogenic with a limited hypertrophic effect, while BMP2 showed weaker chondrogenic but a stronger hypertrophy effect. EC50 for BMPR1a– EC50 for BMPR1b– BMPR2 (ng/ml) BMPR2 (ng/ml) Ratio Chondrogenic effect on primary chondrocytes BMP2 16.6 52 0.3 GDF5 73 72 1 The chondrogenic and hypertrophic activity of the different R399E 21.5 12 1.8 compounds was evaluated in primary chondrocytes. Cells were W417F 117 88 2 cultured in 3D for 28 days with 300 ng/ml GDF5, R399E, W417F, W417R 1200 79 15 W417R or BMP2. Similar to the C3H10T1/2 cells, chondrogenic and

The ratio column describes the ratio of the EC50 value for BMPR1a–BMPR2 to hypertrophic markers were evaluated (Fig. 3). that for BMPR1b–BMPR2 and describes the preference of one compound for GDF5, R399E and BMP2 all stimulated at least two chondrogenic the binding and activation of BMPR2 together with BMPR1B over BMPR1A. markers, and all increased COL2A1 expression. In addition, GDF5 and BMP2 stimulated ACAN expression while R399E and BMP2 chondrogenesis. Col10a1 was also increased in the presence of showed elevated SOX9 expression. W417F and W417R showed GDF5 but Runx2 and ALP activity were not. Similarly, R399E no significant effects on these parameters. BMP2 was the only increased Acan and Col2a1 expression (albeit both to a lesser compound to significantly increase COL10A1 expression. RUNX2 extent than GDF5) and Sox9 expression. In contrast to GDF5, in was slightly increased by all compounds (only reaching significance the presence of R399E the expression of Col10a1 remained low; for R399E, but no significant difference between R399E and GDF5 Runx2 was also not modulated at early time points and ALP was observed). Finally, ALP activity was increased by all compounds activity was not detectable. W417F had only a weak effect on all except W417R. GDF5 and BMP2 showed the greatest ALP activity parameters. Finally, BMP2 increased Acan expression but not increase, followed by R399E and W417F. These results indicate that Col2a1, while all hypertrophy markers – Col10a1, Runx2 GDF5, R399E and BMP2 exert a chondrogenic effect in primary expression and ALP activity – were stimulated by BMP2. From chondrocytes. In addition, GDF5 and R399E stimulated hypertrophy these results it appears that both GDF5 and R399E are but to a lesser extent than BMP2.

Fig. 2. Chondrogenesis in C3H10T1/2 cells. C3H10T1/2 cell pellets were cultured for 28 days with 300 ng/ml of the different compounds in the chondrogenic medium (n=3 parallel cell cultures). Gene expression for several chondrogenesis and hypertrophy markers were analyzed, as well as levels of ALP in the medium. Data represent the means±s.e.m. +P<0.05, ++P<0.01, +++P<0.001 or ++++P<0.0001 indicates difference from day 0 and *P<0.05, **P<0.01, ***P<0.001 or

****P<0.0001 indicates difference from GDF5 for the same time point. nd, not determined (no RNA could be isolated). Journal of Cell Science

4 RESEARCH ARTICLE Journal of Cell Science (2020) 133, jcs246934. doi:10.1242/jcs.246934

Fig. 3. Stimulation of chondrogenic and hypertrophic markers in primary chondrocytes. Primary porcine chondrocytes were cultured in a scaffold-free 3D culture for 28 days with 300 ng/ml of the different compounds (n=3 parallel cell cultures). Gene expression for several chondrogenesis and hypertrophy markers were analyzed, as well as levels of ALP in the medium. Data represent the means±s.e.m. +P<0.05, ++P<0.01, +++P<0.001 or ++++P<0.0001 indicates difference from control and *P<0.05, **P<0.01, ***P<0.001 or ****P<0.0001 indicates difference from GDF5.

R399E exerted the same effects on primary chondrocytes as EC50 values showed that BMP2 binds preferentially to BMPR1A over GDF5, except for ALP activity, which was significantly lower with BMPR1B, followed by GDF5, R399E, W417F and W417R. Of note, R399E, and both W417F and W417R presented reduced the three GDF5 mutants showed a strongly reduced binding affinity for chondrogenic and hypertrophic capabilities in comparison to GDF5. BMPR1A in comparison to GDF5, while the affinity for BMPR1B was less affected. When comparing the obtained Kd and EC50 values Osteogenesis for BMPR1A or BMPR1B, there were some discrepancies. For C3H10T1/2 cells were cultured in monolayer in an osteogenic instance, R399E had a lower affinity for BMPR1A but was more medium for 28 days with 300 ng/ml GDF5, R399E, W417F, W417R potent at inducing dimerization of BMPR1A–BMPR2thanGDF5.In or BMP2 (Fig. 4A). Both GDF5 and BMP2 increased three of the six addition, no binding of W417R with either BMPR2 or BMPR1A osteogenic markers tested; Spp1 and Bglap expression were elevated could be detected by SPR, while W417R was found to induce in the presence of GDF5, Bglap expression and ALP activity were dimerization of these two receptors in the dimerization assay. These increased by BMP2 and both showed positive Alizarin Red staining differences can be explained by the fact that SPR evaluates binding to after 28 days. R399E, W417F and W417R showed no osteogenic one receptor subtype only and does not take into account that the activity. oligomerization of type I and type II receptor is required for signaling. The culture of Saos-2 cells in the presence of the different In contrast, the dimerization assay might better encompass the compounds was conducted to further evaluate the osteogenic capacity complexity of a ternary receptor–ligand complex (Heinecke et al., of the compounds. Only BMP2 increased ALP levels. In addition, the 2009). Consequently, we have chosen to use the results from the strongest Alizarin Red staining was obtained in the presence of BMP2 dimerization assay to subsequently interpret the roles of BMR1A (Fig. 4B). and BMPR1B in chondrogenic, hypertrophic and osteogenic differentiation. DISCUSSION BMP receptor binding and dimerization profile of BMP2, GDF5 GDF5 is a stronger chondroinductor, whereas BMP2 is a and the GDF5 mutants stronger osteoinductor GDF5, BMP2 and the GDF5 mutants were evaluated for their binding Chondrogenesis and hypertrophic differentiation of the multipotent affinities to BMPR2, BMPR1A and BMPR1B (by SPR) and their mesenchymal precursor cells, C3H10T1/2, was stimulated by ability to induce dimerization of BMPR2 with BMPR1A or BMPR1B GDF5. BMP2 stimulated chondrogenesis in C3H10T1/2 cells to a in a cellular dimerization assay. The resulting affinities (Kd values), lesser extent than GDF5, and increased Acan and Sox9 expression potencies (EC50 values) and calculated ratios (Kd for BMPR1A to Kd but not Col2a1 expression. However, the hypertrophic capacity of for BMPR1B; EC50 for BMPR1a–BMPR2toEC50forBMPR1b– BMP2 was stronger than GDF5; it increased Col10a1 more strongly BMPR2) enabled comparison of the compounds for their selectivity than GDF5 and stimulated Runx2 expression and ALP activity for BMPR1B or BMPR1A. As expected, both the ratios for the Kd and while GDF5 did not. In mature chondrocytes, both GDF5 and Journal of Cell Science

5 RESEARCH ARTICLE Journal of Cell Science (2020) 133, jcs246934. doi:10.1242/jcs.246934

Fig. 4. Osteogenesis in C3H10T1/2 cells and mineralization of Saos-2 cells. (A) C3H10T1/2 cells were cultured in a monolayer for 28 days with 300 ng/ml of the different compounds in the osteogenic medium (n=5 parallel cell cultures). Gene expression of several osteogenic markers were analyzed. ALP levels were measured in the medium (only detectable in the presence of BMP2) and Ca2+ deposition was revealed by Alizarin Red staining. (B) Saos-2 cells were cultured in a monolayer for 14 days with 300 ng/ml of the different proteins (n=6 parallel cell cultures). ALP levels were measured in the medium and Ca2+ deposition was revealed by Alizarin Red staining. Data represent the means±s.e.m. +P<0.05,++P<0.01, +++P<0.001 or ++++P<0.0001 indicates difference from control for the same time point and *P<0.05, **P<0.01, ***P<0.001 or ****P<0.0001 indicates difference from GDF5 for the same time point.

BMP2 had a similar impact on chondrogenic markers, but only chondrogenesis and osteogenesis. Similarly, Kaps et al. (2004) BMP2 stimulated COL10A1 expression. Finally, GDF5 and BMP2 observed that, in C3H10T1/2 cells, the forced expression of a stimulated osteogenesis in C3H10T1/2 cells, but only BMP2 dominant negative Bmpr1a strongly affected the initiation of stimulated ALP activity and increased Alizarin Red staining in the chondrogenesis and osteogenesis, while the expression of a osteoblast-like Saos-2 cells. These results confirm that BMP2 and dominant negative Bmpr1b did not. In addition, Jing et al. (2013, GDF5 can both promote chondrogenesis and osteogenesis, but that 2017) observed that a conditional Bmpr1a knockout in the cartilage GDF5 appears to favor chondrogenesis while BMP2 favors of mice resulted in impaired postnatal chondrogenesis, reduced osteogenesis. This is in accordance with the results from other cartilage matrix in the growth plate and the arrest of long bone growth. studies (Kleinschmidt et al., 2013; Wulsten et al., 2011). The ratio of BMPR1A to BMPR1B signaling is a determinant BMPR1A is necessary for chondrogenesis and osteogenesis for hypertrophic differentiation W417R was as potent as GDF5 in the BMPR2–BMPR1B R399E showed a similar profile to GDF5 on chondrogenesis in dimerization assay but possessed the lowest potency for inducing C3H10T1/2 cells and the expression of chondrogenic markers in BMPR2–BMPR1A dimerization (16 times lower than GDF5). It was primary chondrocytes, but had little or no effect on hypertrophy also the only mutant that had no effect on C3H10T1/2 marker expression. In contrast, W417F and W417R had little/no effect chondrogenesis or osteogenesis, and no impact on the expression on both chondrogenesis and hypertrophy in C3H10T1/2 cells and of chondrogenic, hypertrophy and osteogenic markers in primary primary chondrocytes. To better understand how selectivity for chondrocytes or Saos-2 cells, respectively. This reduced activity BMPR1A or BMPR1B influences the chondrogenic and hypertrophic cannot be due to an effect mediated by Nog as W417R was found to capacities of the compounds, Col2a1/COL2A1 (as a chondrogenesis be poorly inhibited by Noggin (Fig. S1B). These results indicate that marker) and Col10a1/COL10A1 (as a hypertrophy marker) expression

BMPR1A, but not BMPR1B, signaling is necessary for were plotted against the EC50 values obtained in the dimerization Journal of Cell Science

6 RESEARCH ARTICLE Journal of Cell Science (2020) 133, jcs246934. doi:10.1242/jcs.246934

assay. No correlation between the EC50 values for BMPR2–BMPR1A well as several BMPs (Salazar et al., 2016), are involved at this stage or BMPR2–BMPR1B dimerization and Col2a1/COL2A1 or Col10a1/ explaining the redundancy between BMPR1A and BMPR1B (Yoon COL10A1 expression in C3H10T1/2 cells or chondrocytes was et al., 2005) observed during early chondrogenesis. observed (data not shown). The ratio of the EC50 values To further understand the role of BMPR1A and BMPR1B it (EC50BMPR1a–BMPR2:EC50BMPR1b–BMPR2) were then plotted against would be of interest to evaluate how their expression level might the ratio of Col2a1/COL2A1 to Col10a1/COL10A1 (Fig. 5). The latter change during chondrogenesis and osteogenesis and how BMPs was used as an indicator of the intensity of chondrogenesis versus with different receptor affinities influence their expression. We hypertrophy stimulation. A correlation was observed between the could not observe any influence of any of the tested compound on EC50 ratio and the ratio of Col2a1:Col10a1 or COL2A1:COL10A1 Bmpr1a and Bmpr2 during chondrogenesis and osteogenesis in expression. This indicates that, in contrast to BMPR1A, stronger C3H10T1/2 cells (data not shown). In our experimental setting BMPR1B signaling favors chondrogenesis over hypertrophy. As a Bmpr1b expression was too low to be detected. result, R399E, for example, induced chondrogenesis similarly to GDF5 but exerted a low hypertrophic effect. R399E for cartilage repair We observed that GDF5 and the GDF5 mutants are differently GDF5 was previously shown to promote cartilage repair in an inhibited by Noggin, and that the GDF5 mutant did not influence Nog osteoarthritis rat model and was proposed as a possible therapy for expression in C3H10T1/2 cells while GDF5 did. In addition, others osteoarthritis (Parrish et al., 2017). However, in the same study the have demonstrated that BMP2 is inhibited by Nog and increases Nog unwanted formation of osteophytes was observed in GDF5-treated expression similarly to GDF5 (Seemann et al., 2009). These animals, which could be the result of the hypertrophic and osteogenic differences can influence the activity of the compounds, but we properties of GDF5. Interestingly, our results demonstrate that R399E postulate that Nog would similarly inhibit chondrogenesis, and can promote chondrogenesis of mesenchymal cells, stimulate hypertrophy and would not affect the Col2a1:Col10a1 or COL2A1: cartilage matrix molecule expression in primary chondrocytes and COL10A1 ratios. simultaneously be less hypertrophic and osteogenic than GDF5. Consequently, R399E might be a better therapeutic candidate than BMPR1A initiates chondrogenesis and osteogenesis, and GDF5 for cartilage repair. BMPR1B influences chondrocyte differentiation In conclusion, we characterized the chondrogenic, hypertrophic In this study, instead of using genetic manipulation to elucidate the and osteogenic profile of BMP2, GDF5 and three GDF5 mutants role of BMPR1A and BMPR1B in chondrogenesis and that have different affinities for BMPR1A and BMPR1B in vitro. osteogenesis, we used a different approach. We engineered GDF5 We conclude that BMPR1A signaling is necessary for mutants that bind to these receptors with different affinities and chondrogenesis and osteogenesis, and that BMPR1B does not evaluated their effects on the chondrogenesis and osteogenesis of play a role in osteogenesis. In addition, we postulate that stronger C3H10T1/2 cells, the expression of chondrocyte and hypertrophy BMPR1B signaling compared to BMPR1A signaling could prevent markers in primary chondrocytes and the expression of osteogenic chondrocyte hypertrophy and might act as a cartilage stabilizer markers in Saos-2 cells. Our results confirm that BMPR1A is during joint morphogenesis. Finally, because of its chondrogenic necessary for the initiation of chondrogenesis and osteogenesis. On properties and low hypertrophic and osteogenic activities, we the other hand, BMPR1B does not seem to play a role in suggest that R399E is a promising growth factor for cartilage repair. osteogenesis; we propose that BMPR1B acts as an inhibitor of chondrocyte hypertrophy and that the relative strength of BMPR1A MATERIALS AND METHODS versus BMPR1B signaling determines cellular fate. This role of GDF5, GDF5 mutants and BMP2 BMPR1B is in accordance with its expression pattern in cartilage DNA coding for the mature parts of human GDF5 proteins was been isolated and areas prefiguring cartilage during skeletal development (Yi from human ROB-C26 osteoprogenitor cells (Yamaguchi and Kahn, 1991) et al., 2000; Zou et al., 1997). Therefore, we suggest that BMPR1B via RT-PCR and subsequently ligated into prokaryotic plasmid vectors. signaling might prevent hypertrophy and stabilize the cartilage Single mutations were introduced in residues in the binding site through which GDF5 binds to BMPR1A and BMPR1B (Schreuder et al., 2005) via surface during joint morphogenesis. Bmpr1b is also highly site-directed mutagenesis and the proteins expressed in E. coli, isolated from expressed during early chondrogenesis in mesenchymal cells inclusion bodies, renatured and purified, as described elsewhere (Ploeger before and during condensation of progenitor cells (Yi et al., and Wagner, 2017, United State Patent US9,718,697B2). The amino acid 2000; Yoon et al., 2005) and might play a different role at this stage. substitution consisted of replacing a hydrophobic amino acid with a Indeed, it currently appears that both BMPR1A and BMPR1B, as hydrophilic/polar amino acid and vice versa, or replacing an amino acid with

Fig. 5. Ratio of collagens versus ratio of potencies. For the C3H10T1/2 cells and the chondrocyte culture, the ratio of Col2a1/ COL2A1 to Col10a1/COL10A1 expression (at day 14 for the C3H10T1/2 cells) was calculated and plotted against the ratio of EC50s obtained in the BMPR2–BMPR1A and BMPR2–BMPR1B assays. A linear regression was performed and the obtained R2 values and P values are indicated on the graph. W417R was not used for the analysis because it had no effect. Journal of Cell Science

7 RESEARCH ARTICLE Journal of Cell Science (2020) 133, jcs246934. doi:10.1242/jcs.246934 a different steric demand or a different charge. The resulting mutants (about Chondrocyte culture 20) were profiled in several cell lines for their bioactivity and tested for their Porcine chondrocytes were isolated from the femoral head of an ∼1-year-old binding affinities to BMPR1A and BMPR1B (Ploeger and Wagner, 2017, pig provided by a local slaughterhouse. The cartilage was firstly digested for United State Patent US9,718,867B2). The selection of the mutants used in 45 min with 0.25% collagenase (collagenase NBG4, Serva, Heidelberg, this study was based on their increased preference to bind BMPR1B over Germany) and subsequently overnight with 0.1% collagenase. The resulting BMPR1A compared to GDF5. The selected mutants are R399E (also named cell suspension was filtered and washed, and 106 cells/well were incubated in M1673 in other studies), W417F and W417R where the number refers to the a low-binding 96-well plate in high-glucose DMEM with 10% FCS position of the residue in the GDF5 wild-type sequence (accession number (Biochrom, Berlin, Germany), 0.4 mM proline, 50 µg/ml ascorbate-2- AAH32495) and the first and last letter correspond to the original amino acid phosphate and 1% 100× penicillin/streptomycin. After 1 week, the resulting and the mutated amino acid, respectively. 3D constructs were transferred to a 24-well plate and cultured for an additional BMP2 was from R&D Systems (cat. no. 355-BM-500/CF, Minneapolis, 4 weeks in the same medium supplemented with 300 ng/ml of GDF5, R399E, MN, USA). All proteins were resuspended in 10 mM HCl. W417F, W417R or BMP2. The medium was changed twice a week. At the end of the culture, the cell constructs were homogenized in RLT buffer (from Receptor affinity measurement the RNeasy Mini Kit, Qiagen, Hilden, Germany) and processed for gene A Biacore4000 surface plasmon resonance (SPR) instrument (GE Healthcare, expression analysis. Three experiments were realized (only one with BMP2) Chicago, IL) was used. A total of 400 to 900 resonance units (RU) of the Fc- and for each experiment three (n=3) cultures were run in parallel. The results chimera of the ectodomain of the receptors (from R&D Systems; BMPR1A, of one experiment are shown. cat. no. 2406-BR-100; BMPR1B, cat. no. 505-BR-100; BMPR2, cat. no. 811- BR-100) were immobilized on a Protein A-coated CM-5 sensorchip. GDF5, Saos-2 cell culture BMP2 or the GDF5 mutants were injected (ten steps of a 1-to-2 dilution Saos-2 cells (ATCC HTP-85) were seeded at 50,000 cells/well in a 24-well series, the highest tested concentration was 200 nM) for 300 s at 30 µl/min. plate with a differentiation medium composed of high-glucose DMEM, Sensorgrams were processed with the Biacore 4000 Evaluation Software 10% FBS, 50 µg/ml ascorbate-2-phosphate, 10 mM β-glycerol phosphate, version 1.1 (GE Healthcare). In detail, non-specific interactions of the test 10−8 M dexamethasone and 1% 100× penicillin/streptomycin supplemented samples to the reference surfaces (Protein A), as well as buffer injections with 300 ng/ml of GDF5, R399E, W417F, W417R or BMP2. Alizarin Red (double referenced), were subtracted from the recorded binding data. staining was performed after 14 days of culture and the ALP concentration Processed data were analyzed using a simple 1:1 interaction model was measured in the medium at days 4, 7, 11 and 14. The medium was including a term for mass transport using numerical integration and changed twice a week. Three experiments were realized and for each nonlinear curve fitting to determine kinetic rate constants (kon and koff )and experiment three to six (n=3–6) cultures were run in parallel. The results of the dissociation constants (Kd). Each experiment was performed two times. one experiment are shown.

BMPR dimerization assay Alizarin Red staining U2OS cells expressing BMPR2 together with BMPR1A (cat. no. 93-1006E3) Cells were stained directly in the culture plate. Cells were first washed with or BMPR1B (cat. no. 93-1053C3) were acquired from Discoverx (Fermont, PBS and then fixed with 4% paraformaldehyde (w/v) for one hour at room CA). This cell assay utilizes enzyme fragment complementation technology, temperature. The cells were washed twice with ddH2O before being stained where two fragments of the galactosidase are unified in a functional enzyme with 400 µl/well of 0.5% (w/v) Alizarin Red for 20 min at room temperature. – upon receptor dimerization and generate a luminescent signal. A dose Finally, cells were washed with ddH2O until the water became clear and the response curve was established for each ligand and tested according to the cells were covered with PBS before being photographed. recommendation of the manufacturer. The EC50 values were calculated with GraphPad Prism software (Version 7.0). Two experiments were performed ALP measurement – (without duplicate for the BMPR2 BMPR1b dimerization assay and in For ALP measurement, 100 µl of medium was mixed with 100 µl – duplicate for the BMPR2 BMPR1a dimerization assay) and the obtained p-nitrophenolphosphate (10 mg/ml). After 1 h of incubation, the EC50 values were averaged. absorbance was read at 405 nm using 490 nm as a reference wavelength and compared to that of a nitrophenol standard curve. C3H10T1/2 culture C3H10T1/2 clone 8 cells were obtained from ATCC (cat. no. CCL226) and Gene expression used at passage 8. To initiate chondrogenesis, 106 cells/microtube were RLT buffer (from the RNeasy Mini Kit, Qiagen, Hilden, Germany) was added centrifuged (500 g for 5 min) to form a pellet, and these pellets were cultured to the cells. For the pellets and chondrocyte 3D constructs, a proteinase K in chondrogenic medium, composed of high-glucose DMEM (Gibco, (Qiagen) digestion was performed for 10 min at 55°C before proceeding with Thermo Fisher Scientific, Waltham, MA, USA), 50 µg/ml ascorbate-2- the RNA isolation. RNA isolation was then performed with the RNeasy Mini phosphate (Sigma-Aldrich, St Louis, MO), 0.4 mM proline (Sigma- Kit according to the recommendations of the manufacturer. mRNA Aldrich), 1×10−7 M dexamethasone (Sigma-Aldrich), 0.1% (v/v) ITS concentration and quality were analyzed using an Agilent Bioanalyzer (insulin, transferrin and sodium selenite supplement; Sigma-Aldrich) and (Model 2100, Agilent, Santa Clara, CA) with an Agilent RNA 6000 Nano 1% (v/v) 100× penicillin/streptomycin (Gibco). Chip G2938-80023. To stimulate osteogenesis, cells were seeded at 100,000 cells/well in a 24-well Reverse transcription was performed using SuperScript III First-Strand plate and cultured for 2 days in high-glucose DMEM with 10% fetal bovine Synthesis SuperMix (Invitrogen, Carlsbad, CA), followed by an RNase H serum (FBS) and 1% penicillin/streptomycin. Osteogenesis was stimulated by treatment. Quantitative PCR was performed using the SYBR Green Jumpstart further culturing the cells in osteogenic medium, composed of high-glucose Taq Ready Mix (Sigma-Aldrich) with 200 nM of the reverse and forward DMEM, ascorbate-2-phosphate, 10 mM β-glycerol phosphate (Sigma-Aldrich), primers (Eurofins Genomics, Ebersberg, Germany; see Table S2) and a 1×10−8 M dexamethasone and 1% (v/v) 100× penicillin/streptomycin. Mx3000P Thermocycler (Agilent). Data were acquired and analyzed with the For both chondrogenesis and osteogenesis, cells were treated with 300 ng/ml software MxPro v4.10 (Agilent). GDF5, R399E, W417R, W417F or BMP2 or 12.5 µM HCl (negative control). For each gene, the cycle threshold (Ct) was determined and the relative Twice a week the medium was changed, and the compounds were freshly expression was calculated according to the following formula: added. At day 0, 7, 14, 21 and 28, cells were harvested for gene expression, histology analysis (for chondrogenesis only) or Alizarin Red staining (for ctHKG ¼ ðctHKGctGOIÞ ¼ 2 ; osteogenesis only). In addition, the medium was also harvested and analyzed relative expression 2 ctGOI for ALP. Two experiments were realized and for each experiment three (n=3) or 2 five (n=5) cultures were run in parallel for chondrogenesis and osteogenesis, where HKG represent the housekeeping gene (Ef1a for C3H10T1/2 cells respectively. The results of one experiment are shown. and RLP13A for porcine chondrocytes) and GOI the gene of interest. Journal of Cell Science

8 RESEARCH ARTICLE Journal of Cell Science (2020) 133, jcs246934. doi:10.1242/jcs.246934

Statistical analysis Kameda, T., Koike, C., Saitoh, K., Kuroiwa, A. and Iba, H. (2000). Analysis of Data was statistically analyzed with GraphPad Prism software (Version 7.0). cartilage maturation using micromass cultures of primary chondrocytes. Dev. One-way analysis of variance (ANOVA) was used with Dunnett’s post-hoc Growth Differ. 42, 229-236. doi:10.1046/j.1440-169x.2000.00508.x Kaps, C., Hoffmann, A., Zilberman, Y., Pelled, G., Häupl, T., Sittinger, M., test to correct for multiple comparisons. Statistical analyses were realized on Burmester, G., Gazit, D. and Gross, G. (2004). Distinct roles of BMP receptors replicates from parallel cultures. Type IA and IB in osteo-/chondrogenic differentiation in mesenchymal progenitors (C3H10T1/2). Biofactors 20, 71-84. doi:10.1002/biof.5520200202 Acknowledgements Katagiri, T. and Watabe, T. (2016). Bone morphogenetic proteins. Cold Spring We would like to thank Claudia Arras and Yvonne Wilhelm, who contributed to this work. Harb. Perspect. Biol. 8, a021899. doi:10.1101/cshperspect.a021899 Kleinschmidt, K., Ploeger, F., Nickel, J., Glockenmeier, J., Kunz, P. and Richter, W. (2013). Enhanced reconstruction of long bone architecture by a growth factor Competing interests mutant combining positive features of GDF-5 and BMP-2. Biomaterials 34, T.M., K.K.-D., A.S., S.L. and A.G. were all employees of Merck KGaA, and F.P. of 5926-5936. doi:10.1016/j.biomaterials.2013.04.029 Biopharm GmbH at the time of the study. F.P. is the inventor of several patents Kobayashi, T., Lyons, K. M., McMahon, A. P. and Kronenberg, H. M. (2005). BMP describing mutants of GDF5, including the mutant used in this work. signaling stimulates cellular differentiation at multiple steps during cartilage development. Proc. Natl. Acad. Sci. USA 102, 18023-18027. doi:10.1073/pnas.0503617102 Author contributions Kotzsch, A., Nickel, J., Seher, A., Sebald, W. and Müller, T. D. (2009). Crystal Conceptualization: K.K.-D., A.G.; Methodology: T.M., F.P., A.S., S.L., A.G.; Formal structure analysis reveals a spring-loaded latch as molecular mechanism for GDF- analysis: T.M., A.S., A.G.; Investigation: T.M., A.S.; Resources: F.P.; Writing - 5-type I receptor specificity. EMBO J. 28, 937-947. doi:10.1038/emboj.2009.37 original draft: A.G.; Writing - review & editing: T.M., K.K.-D., F.P., A.S., S.L.; Mace, P. D., Cutfield, J. F. and Cutfield, S. M. (2006). High resolution structures of Supervision: S.L., A.G.; Project administration: K.K.-D. the bone morphogenetic protein type II receptor in two crystal forms: implications for ligand binding. Biochem. Biophys. Res. Commun. 351, 831-838. doi:10.1016/j. bbrc.2006.10.109 Funding McBride, S. H., McKenzie, J. A., Bedrick, B. S., Kuhlmann, P., Pasteris, J. D., This research received no specific grant from any funding agency in the public, Rosen, V. and Silva, M. J. (2014). Long bone structure and strength depend on commercial or not-for-profit sectors. BMP2 from osteoblasts and osteocytes, but not vascular endothelial cells. PLoS ONE 9, e96862. doi:10.1371/journal.pone.0096862 Supplementary information Mi, M., Jin, H., Wang, B., Yukata, K., Sheu, T.-J., Ke, Q. H., Tong, P., Im, H.-J., Xiao, Supplementary information available online at G. and Chen, D. (2013). Chondrocyte BMP2 signaling plays an essential role in https://jcs.biologists.org/lookup/doi/10.1242/jcs.246934.supplemental bone fracture healing. Gene 512, 211-218. doi:10.1016/j.gene.2012.09.130 Nishitoh, H., Ichijo, H., Kimura, M., Matsumoto, T., Makishima, F., Yamaguchi, Peer review history A., Yamashita, H., Enomoto, S. and Miyazono, K. (1996). Identification of type I The peer review history is available online at and type II serine/threonine kinase receptors for growth/differentiation factor-5. https://jcs.biologists.org/lookup/doi/10.1242/jcs.246934.reviewer-comments.pdf J. Biol. Chem. 271, 21345-21352. doi:10.1074/jbc.271.35.21345 Parrish, W. R., Byers, B. A., Su, D., Geesin, J., Herzberg, U., Wadsworth, S., Bendele, A. and Story, B. (2017). Intra-articular therapy with recombinant human References GDF5 arrests disease progression and stimulates cartilage repair in the rat medial Bandyopadhyay, A., Tsuji, K., Cox, K., Harfe, B. D., Rosen, V. and Tabin, C. J. meniscus transection (MMT) model of osteoarthritis. Osteoarthr. Cartil. 25, (2006). Genetic analysis of the roles of BMP2, BMP4, and BMP7 in limb patterning 554-560. doi:10.1016/j.joca.2016.11.002 and skeletogenesis. PLoS Genet. 2, e216. doi:10.1371/journal.pgen.0020216 Ronga, M., Fagetti, A., Canton, G., Paiusco, E., Surace, M. F. and Cherubino, P. Baur, S. T., Mai, J. J. and Dymecki, S. M. (2000). Combinatorial signaling through (2013). Clinical applications of growth factors in bone injuries: experience with BMP receptor IB and GDF5: shaping of the distal mouse limb and the genetics of BMPs. Injury 44, S34-S39. doi:10.1016/S0020-1383(13)70008-1 distal limb diversity. Development 127, 605-619. Rossi, A., Kontarakis, Z., Gerri, C., Nolte, H., Hölper, S., Krüger, M. and Stainier, Cheng, H., Jiang, W., Phillips, F. M., Haydon, R. C., Peng, Y., Zhou, L., Luu, H. H., D. Y. R. (2015). Genetic compensation induced by deleterious mutations but not An, N., Breyer, B., Vanichakarn, P. et al. (2003). Osteogenic activity of the gene knockdowns. Nature 524, 230-233. doi:10.1038/nature14580 fourteen types of human bone morphogenetic proteins (BMPS). J. Bone Joint Salazar, V. S., Gamer, L. W. and Rosen, V. (2016). BMP signalling in skeletal Surg. Am. 85, 1544-1552. doi:10.2106/00004623-200308000-00017 development, disease and repair. Nat. Rev. Endocrinol. 12, 203-221. doi:10.1038/ Coleman, C. M., Vaughan, E. E., Browe, D. C., Mooney, E., Howard, L. and Barry, nrendo.2016.12 F. (2013). Growth differentiation factor-5 enhances in vitro mesenchymal stromal Schreuder, H., Liesum, A., Pohl, J., Kruse, M. and Koyama, M. (2005). Crystal cell chondrogenesis and hypertrophy. Stem Cells Dev. 22, 1968-1976. doi:10. structure of recombinant human growth and differentiation factor 5: evidence for interaction of the type I and type II receptor-binding sites. Biochem. Biophys. Res. 1089/scd.2012.0282 Commun. 329, 1076-1086. doi:10.1016/j.bbrc.2005.02.078 Date, T., Doiguchi, Y., Nobuta, M. and Shindo, H. (2004). Bone morphogenetic Seemann, P., Brehm, A., König, J., Reissner, C., Stricker, S., Kuss, P., Haupt, J., protein-2 induces differentiation of multipotent C3H10T1/2 cells into osteoblasts, Renninger, S., Nickel, J., Sebald, W. et al. (2009). Mutations in GDF5 reveal a chondrocytes, and adipocytes in vivo and in vitro. J. Orthop. Sci. 9, 503-508. key residue mediating BMP inhibition by NOGGIN. PLoS Genet. 5, e1000747. doi:10.1007/s00776-004-0815-2 doi:10.1371/journal.pgen.1000747 ́ Dewulf, N., Verschueren, K., Lonnoy, O., Moren, A., Grimsby, S., Vande Settle, S. H., Rountree, R. B., Sinha, A., Thacker, A., Higgins, K. and Kingsley, Spiegle, K., Miyazono, K., Huylebroeck, D. and Ten Dijke, P. (1995). Distinct D. M. (2003). Multiple joint and skeletal patterning defects caused by single and spatial and temporal expression patterns of two type I receptors for bone double mutations in the mouse Gdf6 and Gdf5 genes. Dev. Biol. 254, 116-130. morphogenetic proteins during mouse embryogenesis. Endocrinology 136, doi:10.1016/S0012-1606(02)00022-2 2652-2663. doi:10.1210/endo.136.6.7750489 Spiro, R. C., Liu, L.-S., Heidaran, M. A., Thompson, A. Y., Ng, C. K., Pohl, J. and Erlacher, L., McCartney, J., Piek, E., ten Dijke, P., Yanagishita, M., Oppermann, Poser, J. W. (2000). Inductive activity of recombinant human growth and H. and Luyten, F. P. (1998). Cartilage-derived morphogenetic proteins and differentiation factor-5. Biochem. Soc. Trans. 28, 362-368. doi:10.1042/bst0280362 osteogenic protein-1 differentially regulate osteogenesis. J. Bone Miner. Res. 13, Storm, E. E. and Kingsley, D. M. (1996). Joint patterning defects caused by single 383-392. doi:10.1359/jbmr.1998.13.3.383 and double mutations in members of the bone morphogenetic protein (BMP) Fujii, M., Takeda, K., Imamura, T., Aoki, H., Sampath, T. K., Enomoto, S., Kawabata, family. Development 122, 3969-3979. M., Kato, M., Ichijo, H. and Miyazono, K. (1999). Roles of bone morphogenetic Storm, E. E. and Kingsley, D. M. (1999). GDF5 coordinates bone and joint formation protein type I receptors and smad proteins in osteoblast and chondroblast during digit development. Dev. Biol. 209, 11-27. doi:10.1006/dbio.1999.9241 Tsuji, K., Bandyopadhyay, A., Harfe, B. D., Cox, K., Kakar, S., Gerstenfeld, L., differentiation. Mol. Biol. Cell 10, 3801-3813. doi:10.1091/mbc.10.11.3801 Einhorn, T., Tabin, C. J. and Rosen, V. (2006). BMP2 activity, although Heinecke, K., Seher, A., Schmitz, W., Mueller, T. D., Sebald, W. and Nickel, J. dispensable for bone formation, is required for the initiation of fracture healing. (2009). Receptor oligomerization and beyond: a case study in bone Nat. Genet. 38, 1424-1429. doi:10.1038/ng1916 morphogenetic proteins. BMC Biol. 7, 59. doi:10.1186/1741-7007-7-59 Weber, D., Kotzsch, A., Nickel, J., Harth, S., Seher, A., Mueller, U., Sebald, W. Jing, J., Ren, Y., Zong, Z., Liu, C., Kamiya, N., Mishina, Y., Liu, Y., Zhou, X. and and Mueller, T. (2007). A silent H-bond can be mutationally activated for high- Feng, J. Q. (2013). BMP receptor 1A determines the cell fate of the postnatal affinity interaction of BMP-2 and activin type IIB receptor. BMC Struct. Biol. 7,6. growth plate. Int. J. Biol. Sci. 9, 895-906. doi:10.7150/ijbs.7508 doi:/10.1186/1472-6807-7-6 Jing, Y., Jing, J., Ye, L., Liu, X., Harris, S. E., Hinton, R. J. and Feng, J. Q. (2017). Wulsten, D., Glatt, V., Ellinghaus, A., Schmidt-Ble, K., Petersen, A., Schell, H., Chondrogenesis and osteogenesis are one continuous developmental and Lienau, J., Sebald, W., Plöger, F., Seemann, P. et al. (2011). Time kinetics of lineage defined biological process. Sci. Rep. 7, 10020. doi:10.1038/s41598- bone defect healing in response to BMP-2 and GDF-5 characterised by in vivo

017-10048-z biomechanics. Eur. Cell Mater. 21, 177-192. doi:10.22203/eCM.v021a14 Journal of Cell Science

9 RESEARCH ARTICLE Journal of Cell Science (2020) 133, jcs246934. doi:10.1242/jcs.246934

Yamaguchi, A. and Kahn, A. J. (1991). Clonal osteogenic cell lines express Yoon, B. S., Ovchinnikov, D. A., Yoshii, I., Mishina, Y., Behringer, R. R. and myogenic and adipocytic developmental potential. Calcif. Tissue Int. 49, 221-225. Lyons, K. M. (2005). Bmpr1a and Bmpr1b have overlapping functions and are doi:10.1007/BF02556122 essential for chondrogenesis in vivo. Proc. Natl. Acad. Sci. USA 102, 5062-5067. Yang, W., Guo, D., Harris, M. A., Cui, Y., Gluhak-Heinrich, J., Wu, J., Chen, X.-D., doi:10.1073/pnas.0500031102 Zehentner, B. K., Haussmann, A. and Burtscher, H. (2002). The bone Skinner, C., Nyman, J. S., Edwards, J. R. et al. (2013). Bmp2 in osteoblasts of morphogenetic protein antagonist Noggin is regulated by Sox9 during periosteum and trabecular bone links bone formation to vascularization and endochondral differentiation. Dev. Growth Differ. 44, 1-9. doi:10.1046/j.1440- mesenchymal stem cells. J. Cell Sci. 126, 4085-4098. doi:10.1242/jcs.118596 169x.2002.00616.x Yi, S. E., Daluiski, A., Pederson, R., Rosen, V. and Lyons, K. M. (2000). The type I Zou, H., Wieser, R., Massagué, J. and Niswander, L. (1997). Distinct roles of type I BMP receptor BMPRIB is required for chondrogenesis in the mouse limb. bone morphogenetic protein receptors in the formation and differentiation of Development 127, 621-630. cartilage. Genes Dev. 11, 2191-2203. doi:10.1101/gad.11.17.2191 Journal of Cell Science