help button home button Endocrine Society Molecular Endocrinology ENDO 08 Sessions Library
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Molecular Endocrinology, doi:10.1210/me.2007-0199
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
21/11/2805    most recent
Author Manuscript (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dong, Y.-F.
Right arrow Articles by Drissi, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Dong, Y.-F.
Right arrow Articles by Drissi, H.
Right arrowPubmed/NCBI databases
*Gene*GEO Profiles
*HomoloGene*Nucleotide
*Protein*UniGene
*Substance via MeSH
Molecular Endocrinology 21 (11): 2805-2820
Copyright © 2007 by The Endocrine Society

Transforming Growth Factor-ß and Wnt Signals Regulate Chondrocyte Differentiation through Twist1 in a Stage-Specific Manner

Yu-Feng Dong, Do Y. Soung, Yoon Chang, Motomi Enomoto-Iwamoto, Mark Paris, Regis J. O’Keefe, Edward M. Schwarz and Hicham Drissi

Center for Musculoskeletal Research (Y.-F.D., D.Y.S., Y.C., R.J.O., E.M.S., H.D.), University of Rochester Medical Center, Rochester, New York 14642; Department of Orthopaedic Surgery (M.E.-I.), Jefferson Medical College, Thomas Jefferson University, Philadelphia, Pennsylvania 19107; and Vaccinex, Inc. (M.P.), Rochester, New York 14620

Address all correspondence and requests for reprints to: Hicham Drissi, Ph.D., Center for Musculoskeletal Research, University of Rochester, 601 Elmwood Avenue, Box 665, Rochester, New York 14642. E-mail: Hicham_Drissi{at}urmc.rochester.edu.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
We investigated the molecular mechanisms underlying the transition between immature and mature chondrocytes downstream of TGF-ß and canonical Wnt signals. We used two developmentally distinct chondrocyte models isolated from the caudal portion of embryonic chick sternum or chick growth plates. Lower sternal chondrocytes exhibited immature phenotypic features, whereas growth plate-extracted cells displayed a hypertrophic phenotype. TGF-ß significantly induced ß-catenin in immature chondrocytes, whereas it repressed it in mature chondrocytes. TGF-ß further enhanced canonical Wnt-mediated transactivation of the Topflash reporter expression in lower sternal chondrocytes. However, it inhibited Topflash activity in a time-dependent manner in growth plate chondrocytes. Our immunoprecipitation experiments showed that TGF-ß induced Sma- and Mad-related protein 3 interaction with T-cell factor 4 in immature chondrocytes, whereas it inhibited this interaction in mature chondrocytes. Similar results were observed by chromatin immunoprecipitation showing that TGF-ß differentially shifts T-cell factor 4 occupancy on the Runx2 promoter in lower sternal chondrocytes vs. growth plate chondrocytes. To further determine the molecular switch between immature and hypertrophic chondrocytes, we assessed the expression and regulation of Twist1 and Runx2 in both cell models upon treatment with TGF-ß and Wnt3a. We show that Runx2 and Twist1 are differentially regulated during chondrocyte maturation. Furthermore, whereas TGF-ß induced Twist1 in mature chondrocytes, it inhibited Runx2 expression in these cells. Opposite effects were observed upon Wnt3a treatment, which predominates over TGF-ß effects on these cells. Finally, overexpression of chick Twist1 in mature chondrocytes dramatically inhibited their hypertrophy. Together, our findings show that Twist1 may be an important regulator of chondrocyte progression toward terminal maturation in response to TGF-ß and canonical Wnt signaling.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
CHICK CHONDROCYTES PROVIDE unique models for studying cell maturation processes in a stage-specific manner. Sternum-derived chondrocytes and more mature cells isolated from the growth plate have been previously characterized. Chick caudal (lower) sternal chondrocytes, which are derived from a region of the embryonic chick sternum that never mineralizes and remains as permanent hyaline cartilage, are undifferentiated, immature chondrocytes that do not spontaneously proceed toward a terminal differentiation stage (1, 2, 3). Chick embryonic cephalic (upper) sternal chondrocytes, which participate in mineralization of the sternum, progressively display a terminal differentiation phenotype characterized by synthesis of alkaline phosphatase and type X collagen (1, 4, 5, 6, 7). Furthermore, whereas progression of upper sternal chondrocytes toward hypertrophy can be accelerated by inducers of chondrocyte maturation such as bone morphogenetic protein 2 (BMP-2), ascorbic acid, and retinoic acid (8, 9, 10, 11), it is delayed by TGF-ß (12, 13). Growth plate chondrocytes derived from the tibia of 4- to 6-wk-old chicks are more advanced in their maturation process. These cells are hypertrophic in their majority at the time of harvest (14, 15, 16, 17) and TGF-ß was shown to induce a reversal of hypertrophy in culture (18, 19, 20, 21, 22).

Among the master local and systemic regulators of chondrocyte differentiation and cartilage maturation in vivo are the TGF-ß superfamily of growth factors (23). TGF-ß transduces its signaling by binding the membrane-anchored type II receptor leading to dimerization with the type I receptor followed by activation of its Ser/Thr kinase activity (24, 25). Subsequently, Smads 2 and 3 are phosphorylated, form complexes with the common cofactor Smad4, and translocate into the nucleus to participate in transcriptional regulation (26). Once this pathway is activated, TGF-ß/Smad3 signals inhibit chondrocyte maturation evidenced by the blockade of hypertrophy hallmarks such as the expression of type X collagen and alkaline phosphatase (27, 28). Consistent with this, mice overexpressing a dominant-negative form of the type II TGF-ß receptor have accelerated chondrocyte maturation in the growth plate and articular chondrocyte hypertrophy coupled with an osteoarthritis-like form of cartilage erosion (29). A similar growth plate and articular cartilage phenotype is present in mice lacking Smad3 (30). Furthermore, we have previously shown that TGF-ß is able to accelerate chondrogenesis in the chick limb, whereas it also delays chondrocyte terminal differentiation in vivo (31).

Wnt secreted proteins have been recently identified as another family of potent regulators of skeletal development, endochondral ossification, and mesenchymal stem cell commitment to chondrogenic and osteogenic phenotypes (32, 33, 34, 35). Canonical and noncanonical pathways have been described as downstream mediators of Wnt family members in skeletal tissues (36). Upstream of their canonical signaling, Wnt proteins bind to the frizzled receptors and coreceptors low-density lipoprotein receptor-related protein 5/6, to induce phosphorylation of disheveled protein, which inhibits glycogen synthase kinase 3 from phosphorylating ß-catenin (37). Unphosphorylated ß-catenin accumulates in the cytoplasm and is translocated to the nucleus, where it binds with transcription factors T-cell factor (TCF)/lymphoid-enhancing factor to activate or repress downstream osteoblast and chondrocyte target gene expression (38, 39).

The mammalian homolog of the Drosophila gene twist, Twist1, belongs to a small family of basic helix-loop-helix transcription factors that includes Dermo1/Twist2, Paraxis, Scleraxis, and most recently Hand1 and Hand2 (40, 41, 42, 43, 44). These proteins recognize a consensus sequence named E-box (C A C/T C/A T G), and seem to be phylogenetically conserved among various species (45, 46, 47). In addition to its implication in craniosynostosis both in humans (48, 49) and mice (50, 51), Twist1 also plays a key role in mediating cell patterning during limb morphogenesis and development through cooperative and antagonistic effects on sonic hedgehog and fibroblast growth factor signaling (52, 53). Together these observations suggest a role for Twist1 not only during intramembranous but also endochondral skeletal development in vivo. Furthermore, the implication of various other signaling pathways including TGF-ß superfamily, fibroblast growth factors, or Wnts during limb morphogenesis upstream and downstream of Twist1 is not yet well defined.

Several studies have reported that a cross talk between Wnt and TGF-ß signaling cooperatively or antagonistically regulates specific target genes in various tissues (54, 55, 56) including chondrocytes (57, 58). However, the mechanisms underlying this switch between cooperative and antagonistic effects are not yet understood. Furthermore, Wnt signaling appears to exert different effects on cartilage maturation depending on the stage of chondrocyte differentiation. Thus, gain and loss of ß-catenin in vivo shows a differential effect on endochondral ossification. ß-Catenin and Lef transgenic mice have severe skeletal defects, particularly in the appendicular skeleton. Growth plates are disorganized, lack maturing chondrocytes expressing Indian hedgehog and type X collagen, and fail to undergo endochondral ossification (59). Conditionally deleting ß-catenin in limb and head mesenchyme shows that osteoblast precursors lacking ß-catenin are blocked during differentiation and alternatively develop into chondrocytes (60). We have previously reported that TGF-ß inhibits canonical Wnt signaling during chick upper sternal chondrocyte maturation (61). Furthermore, we have shown that the skeletal development-related gene Runx2 is a downstream target gene for canonical Wnt signaling mediating chick chondrocyte maturation in vitro (62). In addition to its role in promoting chondrocyte terminal maturation, low levels of Runx2 were also shown to be required to maintain the undifferentiated phenotype of immature chick chondrocytes (63, 64). In osteoblasts, Twist1 was shown to inhibit their differentiation through its inhibitory effects on Runx2 activity (65). Furthermore, a more recent study showed an inhibitory effect of Twist1 on chondrogenesis downstream of Wnts in ATDC-5 cells (66), whereas Hinoi et al. (67) described a Twist1-dependent inhibitory effect of Runx2 on mouse perichondrial immature cell proliferation, thereby preventing their ability to hypertrophy. Thus, the mechanisms behind the transition between immature and mature chondrocytes under various growth factors remain unclear. We hypothesized that a differential regulation of Twist1 and Runx2 may mediate stage-specific cooperative or antagonistic effects of TGF-ß and Wnt signaling during maturation of chick chondrocytes. Our study first demonstrates an interaction between Smad3 and TCF-4 by immunoprecipitation (IP) and chromatin immunoprecipitation (ChIP) assay in two differentially mature chick chondrocyte models: lower sternal, and tibial growth plate chondrocytes. Furthermore, we show an inverse relationship between Runx2 and Twist1 expression and regulation by TGF-ß and canonical Wnt signaling in chick chondrocytes. Finally, our forced expression of Twist1 abrogates chondrocyte terminal maturation in chick growth plate chondrocytes. Together, our findings suggest a stage-specific regulation of Twist1 and Runx2 serving as molecular switches between immature and hypertrophic chondrocytes in response to TGF-ß and canonical Wnt signaling.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Characterization of Two Developmental Chondrocyte Differentiation Models
We compared phenotypic gene expression of developmentally distinct populations of chondrocytes in our differentiation culture media. Thus, we assessed the expression levels of early (Sox9) and late (alkaline phosphatase and type X collagen) chondrocyte markers along with Runx2 by real time RT-PCR in lower sternal, and growth plate chondrocytes treated with TGF-ß for 3, 6, 12, 24, 48, and 72 h. Sox9 expression is progressively up-regulated in the immature lower sternal chondrocytes between 24 and 72 h in the control cultures. Additionally, TGF-ß treatments significantly induced Sox9 gene expression in these cells at 24, 48, and 72 h (Fig. 1AGo). In the mature growth plate chondrocytes, Sox9 mRNA levels were still high in the early time points and were gradually down-regulated with time. This reduction was further inhibited by TGF-ß treatment at 3, 12, 24, 48, and 72 h (Fig. 1BGo). We assessed mRNA expression of a later chondrocyte phenotypic marker, alkaline phosphatase. Figure 1CGo shows that in immature lower sternal chondrocytes, alkaline phosphatase transcripts are up-regulated between 6 and 24 h in culture and to a higher extent at 48 and 72 h. Furthermore, TGF-ß treatment significantly inhibited this expression between 6 and 72 h of culture. Although alkaline phosphatase levels were higher at 3 h in the hypertrophic growth plate chondrocytes, this expression was progressively enhanced with time to be maximal at 48 h and thereafter inhibited at 72 h in the controls. TGF-ß inhibited this expression at all time points between 3 and 72 h of treatment (Fig. 1DGo). Consistent with the established role of Runx2 in mediating terminal maturation, we observed a progressive up-regulation of Runx2 mRNA via real time RT-PCR in these chondrocyte models. The pattern of this up-regulation mirrored that of the alkaline phosphatase gene expression. Figure 1EGo shows that TGF-ß only inhibited Runx2 transcripts at 48 and 72 h after treatment by 2-fold in lower sternal chondrocytes. Suppression of Runx2 transcripts in growth plate chondrocytes, however, was observed as early as 3 and 6 h after treatment (7- and 6-fold, respectively), and to a lesser extent at 12 and 24 h (2- and 3-fold, respectively) (Fig. 1FGo). Terminally mature growth plate chondrocytes at 48 and 72 h in which type X collagen expression, as seen in Fig. 1HGo, is maximal and no longer regulated in the controls, exhibited only a slight inhibition of Runx2 mRNA upon TGF-ß treatment (Fig. 1FGo). Finally, a progressive induction of type X collagen mRNA expression from the immature lower sternal chondrocytes to the hypertrophic growth plate chondrocytes in the controls was also observed (Fig. 1Go, G and H). The most robust expression of this late chondrocyte phenotypic gene is, as expected, in the growth plate-derived cells (Fig. 1HGo). TGF-ß significantly inhibited type X collagen expression in lower sternal chondrocytes at 48 and 72 h by 1.5- and 2-fold, respectively (Fig. 1GGo). As expected, the growth plate chondrocytes, being a more mature population, exhibited suppression of type X collagen at all time points starting after only 3 h of treatment with TGF-ß (Fig. 1HGo). Together, these results validate the developmental potential of these cells to serve as differential models for chondrocyte maturation.


Figure 1
View larger version (44K):
[in this window]
[in a new window]

  		Fig. 1. Lower Sternal and Growth Plate Chondrocytes Display Distinct Stages of Differentiation

Chondrocytes were plated in six-well plates at a density of 5 x 105 cells per well and incubated in the presence and absence of TGF-ß for 3–72 h. Total RNAs were extracted from the cultures at the indicated time points (3, 6, 12, 24, 48, and 72 h). mRNA levels of Sox9 (A and B), alkaline phosphatase (C and D), Runx2 (E and F), and type X collagen (G and H) in lower sternal (A, C, E, and G) and growth plate chondrocytes (B, D, F, and H) were measured by real time RT-PCR. Values are expressed as means ± SE, normalized to GAPDH. Asterisks (*) denote statistical significance from the corresponding control at each time point (P < 0.05). GPC, Growth plate chondrocytes; LSC, lower sternal chondrocytes.

 
TGF-ß Enhances ß-Catenin Expression in Immature Chondrocytes but Inhibits Its Expression in Mature Chondrocytes
We have previously reported that TGF-ß modulates canonical and noncanonical Wnt gene expression in chick upper sternal chondrocytes (68). Here we examined the time-dependent effects of TGF-ß on Wnt signaling in various stages of chondrocyte differentiation, extracted total RNA from chick lower sternal and growth plate chondrocytes treated with TGF-ß, and performed real-time RT-PCR using chick specific ß-catenin primers. TGF-ß significantly up-regulated ß-catenin transcripts in chick lower sternal chondrocytes by 3- to 5-fold after 12, 24, and 48 h of treatment whereas it only induced ß-catenin mRNA levels by 2-fold thereafter (Fig. 2AGo). Interestingly, TGF-ß progressively inhibited ß-catenin expression in the hypertrophic growth plate chondrocytes between 12 and 72 h of treatment (2- to 3-fold) (Fig. 2BGo). These results suggest that TGF-ß differentially regulates ß-catenin expression at various stages of chondrocyte maturation.


Figure 2
View larger version (33K):
[in this window]
[in a new window]

  		Fig. 2. TGF-ß Exerts Distinct Effects on ß-Catenin Expression in Chondrocytes at Different Stages of Maturation

Chick lower sternal and growth plate chondrocytes were plated in six-well plates at a density of 5 x 105 cells per well for gene expression experiments and 10-cm dishes at a density of 3 x 106 for protein expression analysis. Cells were incubated in the presence and absence of TGF-ß for 12–72 h. Total mRNAs and proteins were extracted from the cultures at the indicated time points (12, 24, 48, and 72 h). ß-Catenin mRNA (A and B) and protein (C and D) expression levels were measured in lower sternal (A and C) and growth plate chondrocytes (B and D) by real-time RT-PCR and Western blot, respectively. Values are expressed as means ± SE, normalized to GAPDH. Asterisks (*) denote statistical significance from the corresponding control at each time points (P < 0.05). GPC, Growth plate chondrocytes; LSC, lower sternal chondrocytes.

 
We also performed Western blot analyses to examine the time-dependant effects of TGF-ß on protein levels of ß-catenin in these chondrocyte models. Similar to its gene expression patterns, protein levels of ß-catenin were up-regulated by the TGF-ß in immature lower sternal chondrocytes between 12 and 72 h by at least 3-fold (Fig. 2CGo). Finally, ß-catenin levels were inhibited by TGF-ß as early as 12 h after treatment of growth plate chondrocytes and throughout all time points (Fig. 2DGo). This protein inhibition in mature chondrocytes mimics that observed at the mRNA level and supports a stage-specific regulation of chondrocyte hypertrophy by TGF-ß through its regulation of canonical Wnt signaling.

TGF-ß Regulates Canonical Wnt Signaling in Chick Chondrocytes in a Stage-Specific Manner
To substantiate the functional relevance of this differential TGF-ß-mediated regulation of the canonical Wnt pathway, we performed reporter studies aiming to assess the effects of TGF-ß on TCF-4 and ß-catenin-induced activation of the Topflash reporter in both chick chondrocyte models. Figure 3AGo shows that TGF-ß alone slightly increased Topflash promoter activity at 12, 24, and 36 h after transfection. Furthermore, TCF-4 and ß-catenin overexpression significantly increased Topflash activity in lower sternal chondrocytes at 12, 24, and 36 h by 7-, 4-, and 3-fold, respectively, compared with the control. This induction was significantly enhanced by TGF-ß by 80% at 12 h and 60% at 24 h, whereas no significant effect was observed 48 h after TGF-ß treatment (Fig. 3AGo). Similarly, TCF-4 or ß-catenin alone induced basal Topflash activity in growth plate chondrocytes at 12, 24, and 48 h by 8-, 5-, and 3-fold, respectively (Fig. 3BGo). However, this enhanced activity of Topflash by the combined treatment of TCF-4 and ß-catenin in these mature chondrocytes was inhibited by TGF-ß in a time-dependent manner at 24 and 36 h (2-fold) (Fig. 3BGo). These results confirm that TGF-ß exerts a differential effect on Wnt signaling through its enhancement in immature chondrocytes and its inhibition in hypertrophic chondrocytes.


Figure 3
View larger version (20K):
[in this window]
[in a new window]

  		Fig. 3. TGF-ß Has a Maturation Stage-Dependent Effect on ß-Catenin and TCF-4-Induced Signaling Detected by the Topflash Reporter

Chick lower sternal and growth plate chondrocytes were plated at a density of 2 x 105 cells per well in 12-well plates and cultured overnight. Cells were cotransfected with 500 ng of Topflash with or without 500 ng of ß-catenin cDNA and 500 ng of TCF-4 cDNA. Two hours after transfection, cultures were maintained in regular media with or without TGF-ß for 12, 24, and 36 h. Topflash reporter activity in lysed cells was measured using the dual luciferase assay system. pGL3 basic luciferase reporter vector was used as empty vector control and 20 ng of pRL vector was used as the internal control to normalize for transfection efficacy. Values are expressed as means ± SE. Asterisks (*) denote statistical significance from the corresponding no TGF-ß control at each time point (P < 0.05). GPC, Growth plate chondrocytes; LSC, lower sternal chondrocytes.

 
TCF-4 Recruitment to the Runx2 Promoter Is Dependent on the Stage of Chondrocyte Maturation
We have previously established Runx2 as a downstream target gene for canonical Wnt signaling in chondrocytes and further characterized the TCF/Lef binding element on the Runx2 promoter (69). We have therefore assessed the recruitment of TCF-4 to this transcriptionally active site via ChIP analyses. Sonicated chromatin containing approximately 500-bp fragments of DNA, subjected to IP using specific antibodies for TCF-4 followed by PCR of a fragment of the Runx2 promoter between nucleotide –284 and nucleotide –34, which contains the previously characterized TCF/Lef binding site. We examined the occupancy of this TCF/Lef binding region on the Runx2 promoter in chondrocytes at distinct developmental stages. ChIP with anti-TCF-4 antibody shows protein-DNA complex formation with the Runx2 promoter in both chick chondrocytes. However, TCF-4 interaction with the chromatin was stronger in the mature growth plate chondrocytes in comparison with the lower sternal chondrocytes (Fig. 4AGo). We further examined the effects of TGF-ß on TCF4 and Smad3 intertaction via IP of nuclear proteins extracted from immature lower sternal vs. hypertrophic chick growth plate chondrocytes. Basal interactions between p-Smad3 and TCF-4 were detected in both immature and mature chondrocytes, although this interaction was higher in the more mature growth plate chondrocytes. However, TGF-ß treatment enhanced association of p-Smad3 with TCF-4 in lower sternal chondrocytes whereas this treatment inhibited the interaction in growth plate chondrocytes (Fig. 4BGo).


Figure 4
View larger version (34K):
[in this window]
[in a new window]

  		Fig. 4. The Effect of TGF-ß on Binding of TCF-4-Containing Complexes to Target DNA that Is Dependent on the Maturational Stage of the Cells

A, Chick lower (LSC) and upper sternal (USC), and growth plate chondrocytes (GPC) were immunoprecipitated with antirabbit TCF-4 antibody. DNA present in the immunoprecipitated complexes was used as a template for PCR with primers which cover the region within 300 bp downstream of the TCF/Lef site (–97 to –92 bp) on the Runx2 promoter. B, LSC and GPC were plated and incubated in the presence and absence of TGF-ß for 2 h. Total proteins were extracted and immunoprecipated with antibodies specific for target protein (TCF-4). Complexes with TCF-4 and p-Smad3 were detected by Western blot analysis with p-Smad3 antibody. C, LSC and GPC were treated with or without TGF-ß for 2 h, and ChIP assay were performed using the same the region of the TCF/Lef site (–97 to –92 bp) on the Runx2 promoter. Asterisks (*) denote statistical significance from control (P < 0.05). Neg, Negative.

 
To determine the in vivo effect of TGF-ß treatment on chromatin occupancy by TCF-4 molecule, lower sternal and growth plate chondrocytes were treated for 2 h with TGF-ß and ChIP assays performed. Figure 4CGo shows that TGF-ß enhanced in vivo complex formation between the TCF/Lef response element and TCF-4 in immature lower sternal chondrocytes. However, when mature growth plate chondrocytes were used, TGF-ß inhibited this complex formation with TCF-4 in comparison with untreated controls. As expected, the IgG negative control did not show any band whereas the input shows protein-DNA complex formation. Together, our results provide the first evidence that chondrocyte maturational status influences TGF-ß effects on canonical Wnt-mediated chondrogenesis and chondrocyte hypertrophy.

Twist1 and Runx2 Transcripts Are Developmentally Regulated by TGF-ß and Canonical Wnt in Chick Chondrocytes
Our previous studies showed that Runx2 is a potent inducer of chick chondrocyte maturation downstream of BMP-2 (70) and canonical Wnt signaling (71). However, Runx2 was shown to mediate the onset of chondrogenesis. Furthermore, Twist1 plays a critical role in promoting limb development in vivo (72, 73). To better understand the molecular switch between immature and mature chondrocytes, we first assessed Runx2 and Twist1 mRNA levels in immature lower sternal chondrocytes and mature growth plate chondrocytes after 2 and 4 d in culture. Figure 5AGo shows that whereas Runx2 levels are increased in both cell models with time, significantly higher endogenous levels of Runx2 are present in growth plate mature chondrocyte in comparison with the immature lower sternal chondrocytes (15- to 30-fold higher). On the other hand, Twist1 endogenous levels are higher in lower sternal chondrocytes at d 2 and are down-regulated after 4 d in culture (3-fold inhibition). However, mature growth plate chondrocytes exhibited dramatically lower endogenous levels of Twist1 in comparison with the immature lower sternal chondrocytes and are also decreased with time. These results suggest a differential role for Runx2 and Twist1 in mediating chondrocyte maturation.


Figure 5
View larger version (23K):
[in this window]
[in a new window]

  		Fig. 5. Endogenous Levels of Runx2 and Twist1 at Distinct Stages of Chondrocytes

Lower sternal and growth plate chondrocytes were cultured for 2 and 4 d. Total RNAs were extracted from the cultures at the indicated time points (2 and 4 d). mRNA levels of Runx2 (A) and Twist1 (B) were measured by real time RT-PCR. Values are expressed as means ± SE, normalized to GAPDH. Asterisks (*) denote statistical significance from lower sternal chondrocytes at each time point (P < 0.05). GPC, Growth plate chondrocytes; LSC, lower sternal chondrocytes.

 
To further establish the effects of TGF-ß and canonical Wnt signaling on these key transcription factors in immature vs. mature chondrocytes, we performed real time RT-PCR on RNAs extracted from both cell models treated with either TGF-ß or Wnt3a. Figure 6AGo shows that whereas TGF-ß slightly inhibited Twist1 expression, Wnt3a mildly induced its mRNA levels in lower sternal chondrocytes. As expected, Runx2 levels were significantly inhibited by TGF-ß and induced by Wnt3a (Fig. 6BGo). However, in the mature growth plate chondrocytes TGF-ß dramatically induced Twist1 expression whereas Wnt3a inhibited it (Fig. 6CGo). Interestingly, TGF-ß and Wnt3a treatments had the opposite effects on Runx2 transcripts (Fig. 6DGo). Together, these results suggest a possible switch between Twist1 and Runx2 expression may be required for the transition from the immature to the hypertrophic phenotype upon TGF-ß and canonical Wnt signals.


Figure 6
View larger version (22K):
[in this window]
[in a new window]

  		Fig. 6. mRNA Levels of Twist1 and Runx2 Are Differentially Medicated by TGF-ß and Wnt3a in Chick Chondrocytes

Chondrocytes were plated in six-well plates at a density of 5 x 105 cells per well and incubated in either TGF-ß or Wnt3a for 24 h to extract total RNAs. mRNA levels of Twist1 (A and C) and Runx2 (B and D) in lower sternal (A and B) and growth plate chondrocytes (C and D) were measured by real time RT-PCR. Values are expressed as means ± SE, normalized to GAPDH. Asterisks (*) denote statistical significance (P < 0.05) relative to the corresponding control. GPC, Growth plate chondrocytes; LSC, lower sternal chondrocytes.

 
Twist1 Inhibits Chondrocyte Hypertrophy
Because the levels of Twist1 are very low in growth plate chondrocytes in comparison with immature chondrocytes, we used these growth plate chondrocytes to perform Twist1 gain of function, aiming to determine its role in regulating chondrocyte hypertrophy. We first subcloned the chick Twist1 into replication-competent avian sarcoma (RCAS) and performed viral overexpression studies in these mature cells. Figure 7AGo shows a strong and progressive overexpression of Twist1, 6 and 8 d after infection. Twist1 overexpression resulted in a significant repression of Runx2 transcripts at d 6 (2.5-fold) and to a higher extent at d 8 (5-fold) (Fig. 7BGo). This forced expression is also concomitant with a strong and progressive inhibition of alkaline phosphatase (Fig. 7CGo) and type X collagen (Fig. 7DGo) expression. In addition to its strong inhibitory effects on hypertrophy makers’ expression, Twist1 dramatically inhibited alkaline phosphatase activity in Twist1-infected cells compared with empty RCAS or uninfected controls (Fig. 7EGo). These results indicate that Twist1 is a strong inhibitor of chondrocyte maturation, possibly and partially through its repressive effects on Runx2.


Figure 7
View larger version (58K):
[in this window]
[in a new window]

  		Fig. 7. Twist1 Prevents Chondrocyte Maturation

Chondrocytes were plated in six-well plates at a density of 5 x 105 cells per well and infected Twist1 RCAS. Total RNAs were extracted from the cultures of 6 and 8 d after infection. mRNA levels of Twist1 (A), Runx2 (B), alkaline phosphatase (C), and type X collagen (D) in growth plate chondrocytes were measured by real-time RT-PCR. Alkaline phosphatase staining was conducted using one-step nitro blue tetrazolium chloride/5-bromo-4-chloro-3-indolyhosphate P-toluidine salt 6 d after infection (E). Empty RCAS and no treatment were used as negative controls. Values are expressed as means ± SE, normalized to GAPDH. Asterisks (*) denote statistical significance from the corresponding control at each time point (P < 0.05).

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
The molecular mechanisms downstream of the growth and systemic factors that regulate chondrogenesis and chondrocyte differentiation are still poorly understood. Thus, cross talk between the various pathways that mediate cartilage development downstream of these factors remains a growing area of investigation. We and others have previously reported antagonistic effects between TGF-ß and BMP-2 on chondrocyte hypertrophy through both Smad and non-Smad pathways (74, 75, 76, 77, 78). Additionally, BMP-2 also potentiates retinoic acid-accelerated chondrocyte maturation through cooperative effects between Smad1 and retinoic acid-induced Runx2 (79). On the other hand, PTHrP delays chondrocyte maturation partially through inhibition of Runx2 expression through a protein kinase A pathway (80, 81). TGF-ß, however, is known to exert stage-specific effects on chondrogenesis and chondrocyte maturation. Thus, although it induces chondrocyte cell growth during early chondrogenesis, it delays maturation and hypertrophy (82, 83).

Chick chondrocytes isolated from sternae have been previously established as distinct maturation models in vitro. Whereas the lower sternal chondrocytes exhibit immature features and do not spontaneously undergo hypertrophy, more advanced upper sternal chondrocytes are capable of spontaneously reaching terminal maturation (1, 84, 85, 86, 87). In contrast, cells isolated from chick growth plates are hypertrophic shortly after plating (88, 89, 90, 91, 92, 93, 94). We took advantage of these developmentally distinct cell models to investigate stage-specific effects of TGF-ß on their maturation. We found that lower sternal chondrocytes cultured in the presence of ascorbic acid can express type X collagen. Furthermore, we observed differentially precocious expression of basal type X collagen gene expression from growth plate to lower sternal chondrocytes. TGF-ß inhibition of type X collagen was also differentially increased from lower sternal to growth plate chondrocytes. Our results demonstrate the validity of these developmental models for our subsequent studies. Similarly, the levels of Runx2 were significantly higher in more mature chondrocytes than in immature cells, and TGF-ß strongly inhibited this master transcription factor in hypertrophic chondrocytes in comparison with cells that do not express high levels of type X collagen. We and others have previously demonstrated the correlation between Runx2 expression and the induction of chick sternal chondrocyte hypertrophy (95, 96, 97). However, this is the first demonstration of a developmental regulation of Runx2 by TGF-ß that depends on the maturation stage of these cells.

In the present study we examined the mechanisms of TGF-ß on ß-catenin expression and signaling in the immature vs. mature chick chondrocyte models described above. Our findings show a biphasic effect of TGF-ß on ß-catenin both at the transcriptional and posttranscriptional level, and canonical Wnt signaling, during chondrocyte maturation. In immature chondrocytes, TGF-ß enhances canonical Wnt signaling, which subsequently inhibits early stages of chondrogenesis. In hypertrophic chondrocytes, TGF-ß inhibits ß-catenin expression and results in delayed chondrocyte maturation. Because we did not observe a strong activation of Topflash activity without overexpressing ß-catenin and TCF-4, it is possible that TGF-ß by itself only weakly enhances Wnt signaling and activity in immature chondrocytes. Additionally, canonical Wnt signaling did not result in dedifferentiation of sternal chondrocytes in the presence of TGF-ß as evidenced by sustained Sox9 up-regulation. We speculate that a minimal level of canonical Wnt signaling could be required for immature chondrocyte proliferation or survival and also may help maintain Runx2 expression at a threshold level, to counteract TGF-ß-inhibitory effects on this important transcription factor. Although previous studies described a cross-link between TGF-ß and Wnt signaling, the molecular mechanisms underlying this cross talk seem to be tissue dependent (55, 98, 99, 100, 101, 102, 103, 104, 105). Cooperative effects between TGF-ß and Wnt pathways were observed during human mesenchymal cell differentiation into chondrocytes (106, 107). Furthermore, TGF-ß was shown to have a cooperative induction of the canonical Wnt signaling in murine embryonic maxillary mesenchymal cells (108). This TGF-ß effect was also mediated by Smad3 in these cells and appears to enhance the effects of Wnt3a on cell growth inhibition (109). Our experiments further elucidate functional interactions between downstream signaling pathways of Wnt and TGF-ß within the context of a stage-specific regulation of chondrogenesis and chondrocyte maturation. Interactions between Smads and TCF/Lef proteins were also previously described in other cell models. Thus, the interactions of Smads 2, 3, and 4 with Lef1 synergistically activate the Xenopus gene Xtwn on its promoter (110), whereas Lei et al. (111) previously demonstrated the requirement for cAMP response element binding protein (CREB)-binding protein/p300 for a direct interaction between Smads 3 and 4, TCF-4, and ß-catenin in gastric adenocarcimoma cells. In our chick chondrocyte models we observed a direct interaction with Smad3 but not with Smad2 (data not shown). However, we do not exclude an interaction between Wnt downstream effectors and non-Smad TGF-ß pathway mediators such as activating transcription factor 2. TCF-4 expression was previously shown to be limited to mesenchymal cells surrounding the differentiating cartilages of bones with highest expression at the distal region of mouse embryonic d 13.5 limb buds, which may be responsible for the regulation of chondrogenesis (112). We provide more complete mechanistic information regarding interactions between Smad3 and TCF-4 in addition to ß-catenin to complement the work previously done in various tissues but more importantly in chondrocytes. We focused on TCF-4 in our ChIP assay experiments because of the abundance of the interactive signals we observed in our IP results.

We wanted to determine whether this interaction may have any functional significance in vivo. Because we had previously established that Runx2 is a canonical Wnt downstream target gene (113), we assessed its promoter occupancy by TCF-4 in our two chondrocyte models and examined whether Smad3 may be recruited to this complex. Our ChIP analyses demonstrated a stronger interaction of TCF-4 with the Runx2 promoter in mature growth plate chondrocytes compared with immature lower sternal chondrocytes. This is in agreement with the established role of canonical Wnt signaling in mediating chondrocyte maturation (114, 115, 116, 117). When we assessed the effects of TGF-ß on this interaction in the immature vs. hypertrophic cells, we observed that TGF-ß enhances the interaction between TCF-4 and Smad3 in lower sternal chondrocytes whereas it inhibited it in the growth plate chondrocytes in vitro through IP analyses. TCF-4 recruitment to the Runx2 promoter was also differentially regulated by TGF-ß in immature vs. mature chondrocytes in vivo by ChIP analysis. However, there is no direct evidence that TCF-4 recruits Smad3 proteins to a TCF/Lef binding site on the Runx2 promoter, and the changes in protein binding upon TGF-ß treatments may also result in the abundance of TCF/Lef and ß-catenin proteins as suggested by our gene and protein expression results. Nonetheless, because TGF-ß enhanced the recruitment of TCF-4 to the Runx2 promoter in immature chondrocytes and it inhibited this complex formation in hypertrophic chondrocytes, we speculate that at the later stages of chondrocyte maturation, canonical Wnt signaling may also be regulated by TGF-ß signaling through possible recruitment of Smad3 to alleviate the effects of TGF-ß on delaying chondrocyte maturation and promote their differentiation through activation of Runx2 gene expression. One could not exclude that the close proximity of a Runx binding site to the TCF/Lef site on the Runx2 promoter may contribute to a higher level of complexity in this responsiveness, which can not be addressed via ChIP assays. Future studies will unravel the respective roles of these sites in the responsiveness to TGF-ß and canonical Wnt signals.

Our in vitro data clearly demonstrate that whereas chick Twist1 is not highly expressed in mature chondrocytes it is abundant in immature cells. Most im-portantly, TGF-ß and canonical Wnt signals have op-posite effects on Twist1 during chondrocyte hypertrophy. Our data complement the previously described effect of Twist1 by Reinhold et al. (118) in the ATDC-5 chondrogenic cell line (isolated from mouse teratocarcinoma), in which they show that Twsit1 inhibits chondrogenesis through their assessment of the early chondrogenic marker type II collagen and aggrecan gene expression. In our study, we demonstrate inhibitory effects of Twist1 on the chondrocyte maturation marker alkaline phosphatase and the marker for chondrocyte hypertrophy type X collagen. We elected to perform our gain of function studies in the mature growth plate chondrocytes because immature chondrocytes express very high endogenous levels of Twist1 whereas mature chondrocytes do not. This is also in agreement with in vivo observations by Hinoi et al. (119) showing that murine Twist1 expression may be below detectable levels in mature chondrocytes by in situ hybridization whereas immature perichondrial cells exhibit high levels of Twist1. We further demonstrate a correlation between Twist1 and Runx2 levels through inhibition of Runx2 transcripts by Twist1 in mature chondrocytes. This novel finding may explain that in order for a chondrocyte to transition into hypertrophy, minimum levels of Runx2 need to be reached. We have previously demonstrated that Runx2 mediates chick chondrocyte hypertrophy (120, 121), and this is downstream of canonical Wnt signaling (122). We also show that TGF-ß is a potent inhibitor of chondrocyte maturation through inhibition of canonical Wnt-mediated induction of Runx2 in chick chondrocytes (123). Here we describe novel developmental interplay between TGF-ß and canonical Wnt signaling through regulation of Twist1 and Runx2. In our model presented in Fig. 8Go, we propose that high levels of Twist1 and low levels of Runx2 are required to promote and maintain chondrogenic differentiation in immature chondrocytes. High levels of Twist1 could block precautious expression of hypertrophic phenotype in response to Wnt canonical signaling, and low levels of Runx2 would be important for keeping the cells from dedifferentiation, as suggested previously (63, 124). The transition from immature to hypertrophic chondrocytes can only occur when Twist1 level decreases and Runx2 increases at the necessary levels for terminal maturation. We believe that Twist1 functions to maintain an immature phenotype under the control of TGF-ß whereas its inhibition by canonical Wnt signaling may be required for the transition from immature to hypertrophic chondrocytes.


Figure 8
View larger version (13K):
[in this window]
[in a new window]

  		Fig. 8. Model Describing the Transition from Immature to Mature Chondrocytes

Immature cells express high levels of Twist1 and low levels of Runx2. Whereas canonical Wnt signaling maintains the levels of Twist1 in immature chondrocytes, it dramatically inhibits Twist1 expression in hypertrophic chondrocytes and induced Runx2 levels in these cells. TGF-ß has opposite effects on Twist1 expression but also exerts stage-specific effects on canonical Wnt signaling, adding a higher level of complexity to the regulation of the transition from immature to mature chondrocytes.

 

   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Chondrocyte Isolation and Cell Culture
Chondrocytes were isolated from 15-d-old embryos lower sternal, and 4- to 6-wk-old chick growth plates by digestion in Hanks’ balanced salt solution containing 0.05% collagenase D (Sigma, St. Louis, MO) and 0.25% trypsin (Sigma) in the presence of 5% CO2 for 4 h at 37 C as previously described . Chondrocytes were resuspended in DMEM containing 10% NuSerum IV (Collaborative Biomedical Products, Bedford, MA) supplemented with 50 U penicillin/streptomycin and 2 mM L-glutamine. Cells were then plated in 10-cm plates at a density of 3 x 106 cells per plate. After primary culture for 7 d, chondrocytes were harvested and secondary cultures were placed in 12-well plates at a density of 2 x 105 cells per well for luciferase experiments in DMEM supplemented with 50 ng/ml of ascorbic acid. Cells were also seeded in the same media in six-well plates at a density of 5 x 105 cells per well for our gene expression experiments and 10-cm dishes at a density of 3 x 106 for protein extraction. TGF-ß1 (5 ng/ml; R&D Systems, Minneapolis, MN) were added to the media for various time points before cell harvesting for protein and RNA extraction in our treated conditions. To assess endogenous expression of Twist1 and Runx2, we performed experiments in chick lower-sternal and growth plate chondrocytes. Cells were seeded into 6-cm plates at a density of 1 x 106 cells per plate in DMEM supplemented with 10% NuSerum IV, 50 U penicillin/streptomycin, and 2 mM L-glutamine and in the absence of ascorbic acid. Cells were maintained for 2 and 4 d and media changed every other day.

RNA Extraction and Real-Time RT-PCR
Total RNA was extracted from respective lower sternal and growth plate chondrocytes using the RNAeasy kit (QIAGEN, Valencia, CA) following the manufacturer’s recommendations. Total RNA (1 µg) was reverse transcribed using Advantage RT-for-PCR kit (CLONTECH Laboratories, Mountain View, CA). Freshly reverse transcribed cDNA (1 µl) was used for real-time RT-PCR with chicken-specific primers for ß-catenin, Runx2, Twist1, and chondrocyte phenotypic genes as shown in Table 1Go. DNA synthesis was monitored via the fluorescent dye SYBR Green I (SYBR Green PCR Master Mix, Applied Biosystems, Foster City, CA) using the RotorGene real-time DNA amplification system (Corbett Research, San Francisco, CA). The following protocol was used; a 95 C denaturation step for 15 min followed by 45 cycles with denaturation for 20 sec at 95 C, annealing for 30 sec at 56 C, and extension for 30 sec at 72 C. Gene expression was normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH). PCR products were subjected to a melting curve analysis, and the data were analyzed and quantified with the RotorGene analysis software (Corbett Research).


View this table:
[in this window]
[in a new window]

  		Table 1. Chicken-Specific Primers Were Used for Real Time RT-PCR

 
Western Blot Analysis
The respective lower sternal and growth plate chondrocytes were plated at the density of 3 x 106 cells in 10-cm dishes and cultured overnight in the described above media. Total proteins were isolated as previously described from each chondrocyte model after 12, 24, 48, and 72 h treatment with TGF-ß. Protein samples (30 µg) were subsequently separated on 10% sodium dodecyl sulfate-polyacrylamide and transferred to a polyvinylidine difluoride membrane. Specific proteins were detected using the mouse monoclonal primary antibodies for stabilized ß-catenin (1:1500) and then further probed with rabbit antimouse secondary antibody (1:3000) (Santa Cruz Biotechnology, Inc., Santa Cruz, CA). Anti-ß-actin antibody (Sigma) was used as a control for equal protein loading. Immunoblots was detected using Supersignal west femto maximum sensitivity substrate (Pierce Chemical Co., Rockford, IL).

Transfection and Luciferase Assay
For our transfection experiments, chick lower sternal and growth plate chondrocytes were plated at a density of 2 x 105 cells per well in 12-well plates and cultured overnight. Cells were transfected with 500 ng of Topflash luciferase reporter vector (kindly provided by Dr. Jennifer Westendorf, University of Minnesota) 500 ng of ß-catenin cDNA, and 500 ng of TCF-4 cDNA (a gift from Dr. Ben Alman, Hospital for Sick Children, Toronto, Canada) using the Superfect transfection reagent (QIAGEN) as suggested by the manufacturer’s protocol. Two hours after transfection, cultures were maintained in regular media with TGF-ß for 12, 24, and 36 h. Cells were lysed using the passive lysis buffer (Promega Corp., Madison, WI), and luciferase activity in the lysed cells was measured using the dual Luciferase assay system (Promega) and Optocomp luminometer (MGM Instruments, Hamden, CT). pGL3 basic luciferase reporter vector was used as empty vector control and 20 ng of pRL vector (Promega) carrying the Renilla uniformis gene was used as the internal control to normalize for transfection efficacy.

Immunoprecipitation
The respective lower sternal and growth plate chondrocytes were treated with TGF-ß. After 2 h of treatment cells were washed once with phosphate-buffered saline, lysed for 30 min in lysis buffer (50 mM Tris-HCl, pH 7.5; 150 mM NaCl; 1% Nonidet P-40) containing protease inhibitors (10 µg/ml leupeptin, 10 µg/ml pepstatin A, and 10 µg/ml aprotinin) and phosphatase inhibitors (1 mM NaF and 1 mM Na3VO4), and clarified by centrifugation at 4 C for 15 min. Total protein (200 µg) was precleared with Protein G-Sepharose 4 Fast Flow (Amersham Pharmacia Biotech, Piscataway, NJ) for 1 h at 4 C and then immunopreicipitated with 5 µg of antibody specific for TCF-4. Immune complexes were purified using Protein G-Sepharose beads overnight at 4 C and washed three times with the lysis buffer. After which, ther immune complexes were detected using Western blot analysis.

ChIP Assay
Soluble chromatin was extracted according to the manufacturer’s instructions (Upstate Biotechnology, Inc., Lake Placid, NY). Briefly, cells fixed with formaldehyde were resuspended with sodium dodecyl sulfate lysis buffer supplemented with protease inhibitor and then sonicated for 15 sec. The supernatant from centrifugation were precleared with salmon sperm DNA/Protein A Agarose-50% slurry and then purified by centrifugation. The precleared 2 ml of supernatant solution was incubated with antirabbit TCF-4 antibody overnight at 4 C with constant agitation. Immunoprecipitate complexes were collected after salmon sperm DNA/Protein A Agarose-50% slurry treatment for 1 h at 4 C with rotation. DNA released from immunoprecipitate complex using proteinase K was recovered by phenol/chloroform extraction, precipitated by ethanol, and resuspended with 30 µl Tris-EDTA. DNA solution (2 µl) was used as a template for PCR with primers that were designed to cover the region within 300 bp downstream of the TCF/Lef site (–97 to –92 bp) of Runx2 promoter: sense: 5'-GGGAGGAAGGGAGAGAGAGA-3'; antisense: 5'-TTTTTGCAAGCACTATTACTGGA-3'.

Cloning of cTwist-1 and RCAS-Mediated Infections
For our gain of function studies, full-length chick Twist1 cDNA was generated by PCR and subcloned into pCRII TOPO vector (Invitrogen, Carlsbad, CA). After verification by sequencing, Twist1 cDNA was inserted into the NsiI site of the RCAS BP (A) retroviral vector. Chick embryonic fibroblasts DF-1 (ATCC, Manassas, VA) grown in DMEM containing 10% fetal bovine serum, 0.2% fetal chick serum, and 50 U penicillin/streptomycin were transfected with the replication competent avian sarcoma (RCAS) retrovirus vector, which contains the Twist1 cDNA. RCAS Twist1 Virus generation was performed as previously described. Briefly, transfected DF-1 cells were passaged three times and plated in 100-mm dishes. Upon confluence, media was changed to DMEM containing 10% NuSerum IV and 50 U penicillin/streptomycin. Viral supernatants were collected at 24-h intervals for 3 d. Growth plate chondrocytes were incubated for 48 h with fresh viral supernatant mixed in a 1:1 ratio with plating medium. Virus-containing media were removed 48 h after infection and fresh media were added. Media were replenished at 48-h intervals.

Alkaline Phosphatase Staining
Chick growth plate chondrocytes were seeded in six-well plates at a density of 5 x 105 cells per well. Cells were maintained for 2 d in DMEM supplemented with 10% NuSerum IV (Collaborative Biomedical Products, Bedford, MA) and 50 U penicillin/streptomycin and 2 mM L-glutamine, and then infected with RCAS-Twist1 retrovirus. Empty RCAS virus was also used as control. Virus-containing media were removed 48 h after infection and fresh media were added. Media were changed every other day for 6 d before cell staining. For our alkaline phosphatase staining, cell were washed twice with ice-cold PBS and then fixed in 10% formaldehyde per well for 10 min. Alkaline phosphatase staining was then performed through incubation of fixed cells with nitro blue tetrazolium chloride/5-bromo-4-chloro-3-indolyhosphate P-toluidine salt (Pierce) for 30 min. Stained plates were photographed and data were presented as blue alkaline phosphatase staining.

Statistical Analysis
Each of the experiments was repeated at least three times. Differences between the treatments were compared using a t test. Statistical significance was considered when the P < 0.05.


   ACKNOWLEDGMENTS
 
We thank Dr. Jennifer Westendorf (University of Minnesota, Minneapolis, MN) and Dr. Ben Alman (Hospital for Sick Children, Toronto, Canada) for providing us with reagents used in our study.


   FOOTNOTES
 
Disclosure Statement: All authors have no conflicts of interest.

First Published Online August 7, 2007

Abbreviations: BMP-2, Bone morphogenetic protein 2; ChIP, chromatin immunoprecipitation; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; IP, immunoprecipitation; RCAS, replication-competent avian sarcoma; Smad, Sma- and Mad-related protein; TCF, T-cell factor.

Received for publication April 20, 2007. Accepted for publication August 2, 2007.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Gerstenfeld LC, Kelly CM, Von Deck M, Lian JB 1990 Comparative morphological and biochemical analysis of hypertrophic, non-hypertrophic and 1,25(OH)2D3 treated non-hypertrophic chondrocytes. Connect Tissue Res 24:29–39[Medline]
  2. Iwamoto M, Shimazu A, Pacifici M 1995 Regulation of chondrocyte maturation by fibroblast growth factor-2 and parathyroid hormone. J Orthop Res 13:838–845[CrossRef][Medline]
  3. Reynolds SD, Johnston C, Leboy PS, O’Keefe RJ, Puzas JE, Rosier RN, Reynolds PR 1996 Identification and characterization of a unique chondrocyte gene involved in transition to hypertrophy. Exp Cell Res 226:197–207[CrossRef][Medline]
  4. Iwamoto M, Shimazu A, Pacifici M 1995 Regulation of chondrocyte maturation by fibroblast growth factor-2 and parathyroid hormone. J Orthop Res 13:838–845[CrossRef][Medline]
  5. Reynolds SD, Johnston C, Leboy PS, O’Keefe RJ, Puzas JE, Rosier RN, Reynolds PR 1996 Identification and characterization of a unique chondrocyte gene involved in transition to hypertrophy. Exp Cell Res 226:197–207[CrossRef][Medline]
  6. Leboy PS, Sullivan TA, Nooreyazdan M, Venezian RA 1997 Rapid chondrocyte maturation by serum-free culture with BMP-2 and ascorbic acid. J Cell Biochem 66:394–403[CrossRef][Medline]
  7. Venezian R, Shenker BJ, Datar S, Leboy PS 1998 Modulation of chondrocyte proliferation by ascorbic acid and BMP-2. J Cell Physiol 174:331–341[CrossRef][Medline]
  8. Leboy PS, Sullivan TA, Nooreyazdan M, Venezian RA 1997 Rapid chondrocyte maturation by serum-free culture with BMP-2 and ascorbic acid. J Cell Biochem 66:394–403[CrossRef][Medline]
  9. Venezian R, Shenker BJ, Datar S, Leboy PS 1998 Modulation of chondrocyte proliferation by ascorbic acid and BMP-2. J Cell Physiol 174:331–341[CrossRef][Medline]
  10. Grimsrud CD, Romano PR, D’Souza M, Puzas JE, Schwarz EM, Reynolds PR, Roiser RN, O’Keefe RJ 2001 BMP signaling stimulates chondrocyte maturation and the expression of Indian hedgehog. J Orthop Res 19:18–25[CrossRef][Medline]
  11. Li X, Schwarz EM, Zuscik MJ, Rosier RN, Ionescu AM, Puzas JE, Drissi H, Sheu TJ, O’Keefe RJ 2003 Retinoic acid stimulates chondrocyte differentiation and enhances bone morphogenetic protein effects through induction of Smad1 and Smad5. Endocrinology 144:2514–2523[Abstract/Free Full Text]
  12. Bohme K, Winterhalter KH, Bruckner P 1995 Terminal differentiation of chondrocytes in culture is a spontaneous process and is arrested by transforming growth factor-ß2 and basic fibroblast growth factor in synergy. Exp Cell Res 216:191–198[CrossRef][Medline]
  13. Ionescu AM, Schwarz EM, Zuscik MJ, Drissi H, Puzas JE, Rosier RN, O’Keefe RJ 2003 ATF-2 cooperates with Smad3 to mediate TGF-ß effects on chondrocyte maturation. Exp Cell Res 288:198–207[CrossRef][Medline]
  14. O’Keefe RJ, Puzas JE, Brand JS, Rosier RN 1988 Effects of transforming growth factor-beta on matrix synthesis by chick growth plate chondrocytes. Endocrinology 122:2953–2961[Abstract]
  15. O’Keefe RJ, Crabb ID, Puzas JE, Rosier RN 1989 Countercurrent centrifugal elutriation. High-resolution method for the separation of growth-plate chondrocytes. J Bone Joint Surg Am 71:607–620[Abstract/Free Full Text]
  16. Grimsrud CD, Rosier RN, Puzas JE, Reynolds PR, Reynolds SD, Hicks DG, O’Keefe RJ 1998 Bone morphogenetic protein-7 in growth-plate chondrocytes: regulation by retinoic acid is dependent on the stage of chondrocyte maturation. J Orthop Res 16:247–255[CrossRef][Medline]
  17. Kirsch T, Nah HD, Shapiro IM, Pacifici M 1997 Regulated production of mineralization-competent matrix vesicles in hypertrophic chondrocytes. J Cell Biol 137:1149–1160[Abstract/Free Full Text]
  18. O’Keefe RJ, Puzas JE, Brand JS, Rosier RN 1988 Effects of transforming growth factor-ß on matrix synthesis by chick growth plate chondrocytes. Endocrinology 122:2953–2961[Abstract]
  19. O’Keefe RJ, Puzas JE, Brand JS, Rosier RN 1988 Effect of transforming growth factor-ß on DNA synthesis by growth plate chondrocytes: modulation by factors present in serum. Calcif Tissue Int 43:352–358[Medline]
  20. Rosier RN, O’Keefe RJ, Crabb ID, Puzas JE 1989 Transforming growth factor ß: an autocrine regulator of chondrocytes. Connect Tissue Res 20:295–301[Medline]
  21. Farquharson C, Law AS, Seawright E, Burt DW, Whitehead CC 1996 The expression of transforming growth factor-ß by cultured chick growth plate chondrocytes: differential regulation by 1,25-dihydroxyvitamin D3. J Endocrinol 149:277–285[Abstract/Free Full Text]
  22. Wing-Hoi C, Kwong-Man L, Kwok-Pui F, Pauline LP, Kwok-Sui L 2001 TGF-ß1 is the factor secreted by proliferative chondrocytes to inhibit neo-angiogenesis. J Cell Biochem 81:79–88[CrossRef]
  23. Frenkel SR, Saadeh PB, Mehrara BJ, Chin GS, Steinbrech DS, Brent B, Gittes GK, Longaker MT 2000 Transforming growth factor ß superfamily members: role in cartilage modeling. Plast Reconstr Surg 105:980–990[CrossRef][Medline]
  24. Massague J, Blain SW, Lo RS 2000 TGFß signaling in growth control, cancer, and heritable disorders. Cell 103:295–309[CrossRef][Medline]
  25. Massague J, Chen YG 2000 Controlling TGF-ß signaling. Genes Dev 14:627–644[Free Full Text]
  26. Miyazono K 2000 TGF-ß signaling by Smad proteins. Cytokine Growth Factor Rev 11:15–22[CrossRef][Medline]
  27. Ballock RT, Heydemann A, Wakefield LM, Flanders KC, Roberts AB, Sporn MB 1993 TGF-ß 1 prevents hypertrophy of epiphyseal chondrocytes: regulation of gene expression for cartilage matrix proteins and metalloproteases. Dev Biol 158:414–429[CrossRef][Medline]
  28. Ferguson CM, Schwarz EM, Reynolds PR, Puzas JE, Rosier RN, O’Keefe RJ 2000 Smad2 and 3 mediate transforming growth factor-ß1-induced inhibition of chondrocyte maturation. Endocrinology 141:4728–4735[Abstract/Free Full Text]
  29. Serra R, Johnson M, Filvaroff EH, LaBorde J, Sheehan DM, Derynck R, Moses HL 1997 Expression of a truncated, kinase-defective TGF-ß type II receptor in mouse skeletal tissue promotes terminal chondrocyte differentiation and osteoarthritis. J Cell Biol 139:541–552[Abstract/Free Full Text]
  30. Yang X, Cao X 2001 Smad interactors in bone morphogenetic protein signaling. Methods Mol Biol 177:163–178[Medline]
  31. Ferguson CM, Schwarz EM, Reynolds PR, Puzas JE, Rosier RN, O’Keefe RJ 2000 Smad2 and 3 mediate transforming growth factor-ß1-induced inhibition of chondrocyte maturation. Endocrinology 141:4728–4735[Abstract/Free Full Text]
  32. Yang Y 2003 Wnts and wing: Wnt signaling in vertebrate limb development and musculoskeletal morphogenesis. Birth Defects Res C Embryo Today 69:305–317[CrossRef][Medline]
  33. Hill TP, Spater D, Taketo MM, Birchmeier W, Hartmann C 2005 Canonical Wnt/ß-catenin signaling prevents osteoblasts from differentiating into chondrocytes. Dev Cell 8:727–738[CrossRef][Medline]
  34. Day TF, Guo X, Garrett-Beal L, Yang Y 2005 Wnt/ß-catenin signaling in mesenchymal progenitors controls osteoblast and chondrocyte differentiation during vertebrate skeletogenesis. Dev Cell 8:739–750[CrossRef][Medline]
  35. Tamamura Y, Otani T, Kanatani N, Koyama E, Kitagaki J, Komori T, Yamada Y, Costantini F, Wakisaka S, Pacifici M, Iwamoto M, Enomoto-Iwamoto M 2005 Developmental regulation of Wnt/ß-catenin signals is required for growth plate assembly, cartilage integrity, and endochondral ossification. J Biol Chem 280:19185–19195[Abstract/Free Full Text]
  36. Yates KE, Shortkroff S, Reish RG 2005 Wnt influence on chondrocyte differentiation and cartilage function. DNA Cell Biol 24:446–457[CrossRef][Medline]
  37. Klein PS, Melton DA 1996 A molecular mechanism for the effect of lithium on development. Proc Natl Acad Sci USA 93:8455–8459[Abstract/Free Full Text]
  38. Nusse R 1999 WNT targets. Repression and activation. Trends Genet 15:1–3[Medline]
  39. Huelsken J, Birchmeier W 2001 New aspects of Wnt signaling pathways in higher vertebrates. Curr Opin Genet Dev 11:547–553[CrossRef][Medline]
  40. Cserjesi P, Brown D, Lyons GE, Olson EN 1995 Expression of the novel basic helix-loop-helix gene eHAND in neural crest derivatives and extraembryonic membranes during mouse development. Dev Biol 170:664–678[CrossRef][Medline]
  41. Burgess R, Cserjesi P, Ligon KL, Olson EN 1995 Paraxis: a basic helix-loop-helix protein expressed in paraxial mesoderm and developing somites. Dev Biol 168:296–306[CrossRef][Medline]
  42. Li L, Cserjesi P, Olson EN 1995 Dermo-1: a novel twist-related bHLH protein expressed in the developing dermis. Dev Biol 172:280–292[CrossRef][Medline]
  43. Cross JC, Anson-Cartwright L, Scott IC 2002 Transcription factors underlying the development and endocrine functions of the placenta. Recent Prog Horm Res 57:221–234[Abstract/Free Full Text]
  44. Hollenberg SM, Sternglanz R, Cheng PF, Weintraub H 1995 Identification of a new family of tissue-specific basic helix-loop-helix proteins with a two-hybrid system. Mol Cell Biol 15:3813–3822[Abstract]
  45. Cai J, Jabs EW 2005 A twisted hand: bHLH protein phosphorylation and dimerization regulate limb development. Bioessays 27:1102–1106[CrossRef][Medline]
  46. Firulli AB 2003 A HANDful of questions: the molecular biology of the heart and neural crest derivatives (HAND)-subclass of basic helix-loop-helix transcription factors. Gene 312:27–40[CrossRef][Medline]
  47. Castanon I, Baylies MK 2002 A Twist in fate: evolutionary comparison of Twist structure and function. Gene 287:11–22[CrossRef][Medline]
  48. el G, V, Le Merrer M, Perrin-Schmitt F, Lajeunie E, Benit P, Renier D, Bourgeois P, Bolcato-Bellemin AL, Munnich A, Bonaventure J 1997 Mutations of the TWIST gene in the Saethre-Chotzen syndrome. Nat Genet 15:42–46[CrossRef][Medline]
  49. Howard TD, Paznekas WA, Green ED, Chiang LC, Ma N, Ortiz de Luna RI, Garcia DC, Gonzalez-Ramos M, Kline AD, Jabs EW 1997 Mutations in TWIST, a basic helix-loop-helix transcription factor, in Saethre-Chotzen syndrome. Nat Genet 15:36–41[CrossRef][Medline]
  50. Chen ZF, Behringer RR 1995 Twist is required in head mesenchyme for cranial neural tube morphogenesis. Genes Dev 9:686–699[Abstract/Free Full Text]
  51. Bourgeois P, Bolcato-Bellemin AL, Danse JM, Bloch-Zupan A, Yoshiba K, Stoetzel C, Perrin-Schmitt F 1998 The variable expressivity and incomplete penetrance of the twist-null heterozygous mouse phenotype resemble those of human Saethre-Chotzen syndrome. Hum Mol Genet 7:945–957[Abstract/Free Full Text]
  52. O’Rourke MP, Soo K, Behringer RR, Hui CC, Tam PP 2002 Twist plays an essential role in FGF and SHH signal transduction during mouse limb development. Dev Biol 248:143–156[CrossRef][Medline]
  53. Tavares AT, Izpisuja-Belmonte JC, Rodriguez-Leon J 2001 Developmental expression of chick twist and its regulation during limb patterning. Int J Dev Biol 45:707–713[Medline]
  54. Labbe E, Letamendia A, Attisano L 2000 Association of Smads with lymphoid enhancer binding factor 1/T cell-specific factor mediates cooperative signaling by the transforming growth factor-ß and wnt pathways. Proc Natl Acad Sci USA 97:8358–8363[Abstract/Free Full Text]
  55. Letamendia A, Labbe E, Attisano L 2001 Transcriptional regulation by Smads: crosstalk between the TGF-ß and Wnt pathways. J Bone Joint Surg Am 83-A (Suppl 1):S31