3780 Research Article The Znf219 regulates chondrocyte differentiation by assembling a transcription factory with Sox9

Yoko Takigawa1,2,*, Kenji Hata1,*, Shuji Muramatsu3, Katsuhiko Amano1, Koichiro Ono1, Makoto Wakabayashi1,3, Akio Matsuda3, Kenji Takada2, Riko Nishimura1,‡ and Toshiyuki Yoneda1 1Osaka University Graduate School of Dentistry, Molecular and Cellular Biochemistry, 1-8 Suita, Osaka 565-0871, Japan 2Osaka University Graduate School of Dentistry, Orthodontics and Dentofacial Orthopedics, 1-8 Suita, Osaka 565-0871, Japan 3Asahi-KASEI Pharma, 632-1 Mifuku, Izunokuni, Shizuoka 410-2321, Japan *These authors contributed equally to this work ‡Author for correspondence ([email protected])

Accepted 19 July 2010 Journal of Cell Science 123, 3780-3788 © 2010. Published by The Company of Biologists Ltd doi:10.1242/jcs.071373

Summary Sox9 is an essential transcription factor for chondrogenesis by regulating the expression of chondrogenic . However, its regulatory mechanism is not fully understood. To address this, we attempted to identify the transcriptional partners of Sox9 by screening the cDNA library of the chondrogenic cell line ATDC5 using the collagen 21 (Col21) promoter fused to a luciferase reporter gene. One of the positive clones encoded the Znf219 gene. Whole mount in situ hybridization experiments indicated that Znf219 mRNA was specifically expressed in the developing limb buds where Col21 and Sox9 were strongly expressed. Znf219 markedly enhanced the transcriptional activity of Sox9 on the Col2a1 gene promoter. In addition, Znf219 is physically associated with Sox9 and is colocalized with Sox9 in the nucleus. We also found that overexpression of Znf219 profoundly increased Sox9-induced mRNA expression of Col2a1, aggrecan and Col11a2. Consistently, knockdown of Znf219 decreased the Sox9-induced mRNA expression of these genes. Furthermore, a dominant-negative mutant Znf219 inhibited Bmp2-induced chondrocyte differentiation. Our results suggest that Znf219 plays an important role in the regulation of chondrocyte differentiation as a transcriptional partner of Sox9.

Key words: Sox9, Chondrogenesis, Transcription factor

Introduction 1994; Morais da Silva et al., 1996; Wunderle et al., 1998; McDowall

Journal of Cell Science Chondrogenesis is an important biological event involved in tissue et al., 1999; Giordano et al., 2001; Sock et al., 2003). Sox9 mutant patterning, skeletal development and endochondral ossification mice or mice lacking Sox9 function show impaired endochondral (Shum and Nuckolls, 2002; Kronenberg, 2003; Kobayashi and bone formation (Bi et al., 1999; Akiyama et al., 2002; Akiyama et Kronenberg, 2005). Chondrogenesis starts from the condensation al., 2007). These findings demonstrate that Sox9 plays essential of multipotent mesenchymal cells and then the cells proliferate and roles in chondrogenesis, particularly in the early stage (Lefebvre et differentiate into chondrocytes. Proliferative chondrocytes secrete al., 1996; Lefebvre et al., 1997; Lefebvre et al., 1998; Lefebvre et chondrogenic matrices, including collagen 21 (Col21), collagen al., 2001; Akiyama, 2008). 112 (Col112) and aggrecan, and further differentiate into Sox9 has been proposed to form a large molecular complex that hypertrophic chondrocytes (de Crombrugghe et al., 2000; de regulates several processes of including chromatin Crombrugghe et al., 2001; Kronenberg, 2006). Cartilage tissues remodeling, histone modification, transcription and splicing of the are finally replaced by bone tissues at the end of chondrogenesis target genes during chondrogenesis (Maniatis and Reed, 2002; (Provot and Schipani, 2005). Conversely, some permanent Carrozza et al., 2003; Kitagawa et al., 2003; Kameoka et al., cartilages, including articular cartilage, retain chondrocytes and 2004). Sox5 and Sox6 have been shown to regulate the expression are never replaced by bone tissues. Transcription factors such as of Sox9 target genes as transcriptional partners of Sox9 (Lefebvre the Sox9 family (Sox9, Sox5 and Sox6) and the Runx2 family et al., 1998; Akiyama et al., 2002; Smits et al., 2001; Smits et al., (Runx2 and Runx3) play crucial roles in these complex processes 2004; Ikeda et al., 2004); however, they have no molecular (Lefebvre et al., 1998; Ueta et al., 2001; Akiyama et al., 2002; constitutions to control the nuclear events required for gene Akiyama et al., 2004a; Yoshida et al., 2004). expression. Transcriptional coactivators PGC-1 and p300 might The transcription factor Sox9, which contains an SRY-related regulate chondrogenesis by potentiating Sox9-dependent high-mobility group box, promotes the differentiation of transcription on chromatinized DNA through the acetylation of mesenchymal cells into chondrocytes and induces the expression histones (Tsuda et al., 2003; Furumatsu et al., 2004; Furumatsu et of Col21 (Lefebvre et al., 1996; Lefebvre et al., 1997), Col112 al., 2005; Kawakami et al., 2005). Recently, p54nrb has been (Bridgewater et al., 1998) and aggrecan (Sekiya et al., 2000) in the identified as a member of the Sox9-assembled transcriptional chondrocytes (Bi et al., 1999; Bi et al., 2001; Akiyama et al., 2002; factory and p54nrb links to Sox9-dependent transcription to control Akiyama et al., 2004b; Akiyama et al., 2007). In human, splicing of the target genes of Sox9 (Kameoka et al., 2004; Hata heterozygous mutations of the Sox9 gene result in campomelic et al., 2008). These reports have contributed to the identification of dysplasia characterized by severe chondrodysplasia (Wagner et al., transcriptional components for Sox9 that are implicated in Regulation of chondrogenesis by Znf219 3781

chondrogenesis. However, not all transcriptional partners of Sox9 gene, which has not been previously reported as being related to have been identified yet. Thus, we believe that identification of the chondrogenesis. Znf219, which was initially isolated from a transcriptional factory members associated with Sox9 will lead to human testis cDNA library, contains nine C2H2-type a breakthrough in our understanding of the mechanisms of Sox9- domains (Sakai et al., 2000). To examine whether Znf219 is regulated chondrogenesis. involved in chondrocyte differentiation, we first determined the To understand the molecular basis by which Sox9 controls expression profiles of Znf219 in chondrocytes by performing chondrogenesis, we have attempted to identify the transcriptional reverse-transcription polymerase chain reaction analysis (RT- partners of Sox9. We isolated the Znf219 gene, which contains PCR). This analysis indicated that Znf219 mRNA is expressed in nine zinc finger domains, as a member of the Sox9-assembled primary chondrocytes (Fig. 1A,B). We also observed that Znf219 transcriptional factory and determined whether Znf219 functionally is highly expressed in the heart, pancreas and testis, where Sox9 and physically associates with Sox9 and regulates chondrocyte plays an important role (De Santa Barbara et al., 1998; Stolt et differentiation. Our data indicate that Znf219 plays an important al., 2003; Chaboissier et al., 2004; Akiyama et al., 2004a; Lynn role in regulation of chondrocyte differentiation as a member of et al., 2007) (Fig. 1B). Furthermore, whole mount in situ Sox9-assembed transcriptional factory. hybridization using E11.5 mouse embryos demonstrated that Znf219 is specifically expressed in developing limb buds, where Results Col2a1 and Sox9 are strongly expressed (Fig. 1C). In addition, Identification of the Znf219 gene by mammalian immunocytochemical analyses indicate that Znf219 is expression cloning and expression of the Znf219 gene predominantly expressed in proliferating chondrocytes as Sox9 associated with chondrogenesis (Fig. 1D). We then examined whether Znf219 expression is To identify the components of the Sox9 transcription factory, we associated with chondrocyte differentiation. Znf219 expression screened a cDNA library of the chondrogenic cell line ATDC5 was very marginal in a murine mesenchymal cell line, C3H10T1/2 (Shukunami et al., 1996) using a Col2a1 gene promoter luciferase (Fig. 1E). However, Znf219 expression was markedly upregulated reporter construct (Muramatsu et al., 2007; Hata et al., 2008). We in C3H10T1/2 cells infected with Sox5, Sox6 and Sox9 identified approximately 50 positive clones that activate Col2a1 adenoviruses, which promote chondrocyte differentiation (Fig. gene promoter activity, including Sox5 and Sox6 (Yano et al., 1E). Conversely, Znf219 overexpression did not affect Sox9 2005; Jin et al., 2006; Glass and Karsenty, 2007; Muramatsu et expression (Fig. 1F). Consistent with Fig. 1E, Znf219 expression al., 2007; Hata et al., 2008). We have also isolated the Znf219 is associated with chondrocyte differentiation of ATDC5 cells

Fig. 1. Expression of Znf219 in chondrocytes. (A)Expression of Znf219 in primary chondrocytes. The expression of Znf219, Col2a1 and -actin in primary chondrocytes was examined by reverse transcription-polymerase chain reaction (RT- Journal of Cell Science PCR). (B)Expression of Znf219 in chondrocytes in adults. The tissue distribution of Znf219 in adults was examined by real- time PCR. (C)Expression of Znf219 in developing limb buds. The expression of Znf219 in E11.5 mice was examined by whole mount in situ hybridization. (D)Paraffin sections of E15.5 mouse tibia were subjected to immunocytochemical analyses using anti-Sox9 and Znf219 antibody. (E)Expression of Znf219 during chondrocyte differentiation induced by Sox5, Sox6 and Sox9. C3H10T1/2 cells were infected with adenoviruses carrying Sox5, Sox6 and Sox9 at 20 multiplicity of infection and cultured for 4 days. Expression of Znf219, Col2a1, aggrecan and -actin after infection of the adenoviruses was examined by RT-PCR. (F)Effect of Znf219 on Sox9 expression. C3H10T1/2 cells were infected with control (Cont) or Znf219 adenoviruses at 20 multiplicity of infection and cultured for 4 days. Total RNA was isolated from the cells and subjected to real-time PCR. (G)ATDC5 cells were cultured for 3, 6, 10 or 14 days in the presence of insulin (10g/ml). Total RNA was isolated from the cells and subjected to real-time PCR using Col2a1 and Znf219 primers. 3782 Journal of Cell Science 123 (21)

(Fig. 1G). These results indicate that expression of Znf219 is Znf219 stimulates the function of Sox9 by forming a associated with chondrocyte differentiation and suggest that transcriptional complex with Sox9 Znf219 is involved in the regulation of chondrogenesis. We next examined whether Znf219 activates the transcriptional activity of Sox9 using a Col2a1 gene promoter luciferase construct containing four repeats of the Sox9 binding element (Muramatsu et al., 2007; Hata et al., 2008). Znf219 itself clearly stimulated Col2a1 gene promoter activity (Fig. 2A). In addition, Venus-tagged Znf219 was localized in the nucleus (Fig. 2B), suggesting that Znf219 functions as a transcriptional regulator. We also found that Znf219 markedly upregulated the transcriptional activity of Sox9 on the Col2a1 gene promoter (Fig. 2A). By contrast, a dominant-negative Sox9 markedly inhibited the effect of Znf219 on the Col2a1 gene promoter in C3H10T1/2 cells, where Sox9 is endogenously expressed (Fig. 2C). However, overexpression of Znf219 had no effect on the transcriptional activity of Runx2, an essential transcription factor for osteogenesis (Komori et al., 1997; Ducy et al., 1997), and PPAR, a crucial nuclear for adipogenesis (Tontonoz et al., 1994; Wu et al., 1995) (Fig. 2D,E). These results suggest that Znf219 specifically regulates the transcriptional activity of Sox9. Consistent with these results, Venus-tagged Znf219 was colocalized well with DsRed-tagged Sox9 in the nuclei, but was not colocalized with Ds- Red-tagged Runx2 (Fig. 3). Interestingly, Znf219 forms smaller granules in the nuclei when Znf219 is present with Sox9 (Fig. 3B). Moreover, co-immunoprecipitation experiments showed that Znf219 is physically associated with Sox9 (Fig. 4A). To further understand the molecular interaction between Znf219 and Sox9, we performed an in vitro binding assay using a series of deletion mutants of Znf219 (Fig. 4B). As shown in Fig. 4C, the region including the fifth and sixth zinc finger domains is required for the association with Sox9. Conversely, an in vitro binding assay using Sox9 mutants (Fig. 4D) Journal of Cell Science

Fig. 2. Importance of association of Znf219 with Sox9 in the upregulation of Col2a1 gene promoter activity. (A)The costimulatory effect of Znf219 and Sox9 on Col2a1 promoter activity was studied. C3H10T1/2 cells were transfected with a Col2a1 gene promoter luciferase construct together with empty vector (Cont), Znf219 expression vector, Sox9 expression vector or both expression vectors as indicated. After 48 hours, the luciferase activities of the cell lysates were measured. Data represent mean ± s.d. (n3) (*P<0.01 vs Cont; **P<0.01 vs Sox9). The left panel is shown as part of the right panel (Cont and Znf219) in a different scale. (B)Znf219 is localized in the nucleus. C3H10T1/2 cells transfected with Venus or Venus-tagged-Znf219 were visualized under a fluorescence microscope. Nuclei of cells were stained with DAPI. (C)Inhibition of Znf219-stimulated Col2a1 promoter activity by dominant-negative Sox9. C3H10T1/2 cells were transfected with Col2a1 luciferase construct, together with empty vector (Cont), Znf219 expression vector, dominant-negative Sox9 (DN-Sox9) expression vector, or both expression vectors as indicated and incubated for 48 hours. The luciferase activities of the cell lysates were then measured. Data represent mean ± s.d. (n3) (*P<0.01 vs Znf219). (D)No upregulation of Runx2 activity by Znf219 was seen. C3H10T1/2 cells were transfected with 63OSE luciferase construct Fig. 3. Colocalization of Znf219 and Sox9 in the nucleus. (A)Localization together with empty vector (Cont), Znf219 expression vector, Runx2 of Znf219 in the nucleus. C3H10T1/2 cells transfected with Venus-tagged- expression vector or both expression vectors as indicated and the luciferase Znf219 (Znf219-Venus) was monitored under a fluorescence microscope. activities of the cell lysates were measured. Data represent mean ± s.d. (n3). (B)Colocalization of Znf219 and Sox9 in the nucleus. C3H10T1/2 cells (E)No upregulation of PPAR activity by Znf219 was seen. C3H10T1/2 cells transfected with Venus-tagged-Znf219 (Znf219-Venus) and DsRed-tagged- were transfected with PPRE luciferase construct together with empty vector Sox9 (DsRed-Sox9) were monitored under a fluorescence microscope. (Cont), Znf219 expression vector, PPAR expression vector or both expression (C)Znf219 is not localized with Runx2. C3H10T1/2 cells transfected with vectors as indicated and the luciferase activities of the cell lysates were Venus-tagged Znf219 (Znf219-Venus) and DsRed-tagged-Runx2 (DsRed- measured. Data represent mean ± s.d. (n3). Runx2) were monitored under a fluorescence microscope. Regulation of chondrogenesis by Znf219 3783

Fig. 4. Physical association of Znf219 with Sox9. (A)Physical association of Znf219 with Sox9. BOSC23 cells were transfected with -tagged- Znf219 expression vector, Flag-tagged-Sox9 expression vector, or both and incubated for 48 hours. The lysates of these cells were immunoprecipitated with anti-Flag antibody and immunoblotted with anti-Myc antibody (top). The expression levels of Myc-tagged-Znf219 and Flag- tagged-Sox9 were monitored by immunoblotting with anti-Myc (middle) or anti-Flag (bottom) antibody. (B)Schematic diagram of the mutants of Znf219. 1, 2, 3, 4 and 5 are mutants of Flag-tagged-Znf219. (C)Analysis of the binding domain of Znf219 with Sox9. The cell lysates expressing mutants of Znf219 were precipitated with His-tagged Sox9 and then the precipitates were determined by immunoblotting with anti-Flag antibody. (D)Schematic diagram of the mutants of Sox9. (E)Analysis of the binding region of Sox9 with Znf219. The cell lysates expressing wild-type or mutants of Sox9 were precipitated with Myc-tagged Znf219 and then the precipitates were determined by immunoblotting with anti-HA antibody.

indicated that the C-terminal region of Sox9 is required for the chondrocyte differentiation of C3H10T1/2 cells (Shea et al., 2003), binding to Znf219 (Fig. 4E). Taken together, Znf219 forms a as determined by Alcian Blue staining (Fig. 6G). In addition, the transcriptional factory with Sox9, thereby upregulating the function mutant of Znf219 inhibited chondrocyte differentiation of limb

Journal of Cell Science of Sox9, and might be involved in chondrocyte differentiation regulated by Sox9.

Functional role of Znf219 in chondrogenic differentiation To examine the functional role of Znf219 in chondrocyte differentiation, we overexpressed Znf219 in C3H10T1/2 mesenchymal cells using an adenovirus gene delivery system. Znf219 overexpression dramatically enhanced Sox9-induced mRNA expression of Col2a1, aggrecan and Col11a2 mRNA (Fig. 5A–C). By contrast, Znf219 had no effect on Col1a2 expression (Fig. 5D). These results suggest that Znf219 regulates chondrocyte differentiation of mesenchymal cells through forming a transcription factory with Sox9. To confirm this, we attempted to examine the role of Znf219 in chondrocyte differentiation using a dominant- negative strategy. As shown in Fig. 6A, a mutant of Znf219 (5) inhibited the transcriptional activity of wild-type Znf219 for the Col2a1 gene promoter. The mutant also suppressed the transcriptional activity of Sox9 on the Col2a1 gene promoter (Fig. 6B). In addition, the Venus-tagged Znf219 5 mutant was colocalized well with DsRed-tagged Sox9 in the nucleus (Fig. 6C). As shown in Fig. 4C, the binding assay showed a physical interaction between the Znf219 mutant and Sox9. These results Fig. 5. Regulation of Sox9-dependent gene expression by Znf219. suggested that the Znf219 5 mutant functions as a dominant- (A–D)C3H10T1/2 cells were infected with control (Cont), Sox9 adenovirus, negative mutant of Znf219. Importantly, we found that Znf219 adenovirus or both adenoviruses and then cultured for 7 days. Total overexpression of the Znf219 mutant inhibited Sox9-induced RNA of the cells was determined by real-time PCR for Col2a1 (A), aggrecan Col2a1, aggrecan and Col11a2 expression (Fig. 6D–F). Moreover, (B), Col11a2 (C) and Col1a2 (D). Data represent mean ± s.d. (n3; *P<0.01 overexpression of the Znf219 mutant inhibited Bmp2-induced vs Sox9 by one-way analysis of variance). 3784 Journal of Cell Science 123 (21)

Fig. 6. Inhibition of chondrocyte differentiation by a mutant of Znf219. (A,B)The Znf219 mutant (Znf2195) failed to transactivate the transcriptional activity of wild-type Znf219 (Znf219WT) and Sox9. C3H10T1/2 cells were transfected with constructs as indicated and the luciferase activities of the cell lysates were measured. Data represent mean ± s.d. (n3; *P<0.01 vs Znf219WT by one-way analysis of variance). (C)Colocalization of a mutant of Znf219 (Znf2195) with Sox9. C3H10T1/2 cells transfected with DsRed- tagged Sox9 (DsRed-Sox9) and the mutant of Venus-tagged Znf219 (Znf2195-Venus) were monitored under a fluorescence microscope. (D–F)Inhibition of Sox9-induced chondrogenic gene expression by Znf2195. C3H10T1/2 cells infected with control (Cont), Znf2195 adenovirus, Sox9 adenovirus or both adenoviruses were incubated for 7 days. Total RNA of the cells was determined by real-time PCR for Col2a1 (D), aggrecan (E) and Col11a2 (F). Data represent mean ± s.d. (n3; *: P<0.01 vs Sox9 by one-way analysis of variance). (G)Inhibition of Bmp2-induced chondrocyte differentiation in C3H10T1/2 cells by Znf2195. C3H10T1/2 cells infected with control (Cont) or Znf2195 adenovirus were cultured in the presence or absence of Bmp2 (100g/ml) for 10 days and subjected to Alcian Blue staining. (H)Inhibition of Bmp2-induced chondrocyte differentiation in mouse limb bud cells by Znf2195. Mouse limb bud cells were infected with control or Znf2195 adenovirus and then subjected to micromass culture in the presence or absence of Bmp2 (100g/ml). After 4 days, the cells were stained with Alcian Blue. (I)Inhibition of Bmp2-induced calcification in primary chondrocytes by Znf2195. Primary chondrocytes cultured with or without Bmp2 (100g/ml) were infected with control (Cont) or Znf2195 adenoviruses, cultured for 7 days and stained with Alizarin Red. (J)Inhibition of Bmp2-induced calcification in mouse limb bud cells by Znf2195. Mouse limb bud cells were infected with control or Znf2195 adenovirus and then subjected to micromass culture in the presence or absence of Bmp2 (100g/ml). After 10 days, the cells were stained with Alizarin Red. (K)No effect of Znf2195 on Bmp2-induced osteoblast differentiation in C3H10T1/2 cells was seen. C3H10T1/2 cells infected with control (Cont) or Znf2195 adenovirus were cultured in the presence or absence of Bmp2 (100g/ml) for 7 days and subjected to alkaline phosphatase staining. (L)No effect of Journal of Cell Science Znf2195 on Bmp2-induced osteoblast differentiation in C2C12 cells was seen. C2C12 cells infected with control (Cont) or Znf2195 adenovirus were cultured in the presence or absence of Bmp2 (100g/ml) for 4 days and subjected to alkaline phosphatase staining.

bud cells in a micromass culture system (Fig. 6H). Overexpression domain of Znf219 for the Col2a1 gene. A previous study reported of the Znf219 mutant inhibited calcification of mouse primary that Znf219 regulates the target genes through the GGGGG motif chondrocytes and mouse embryo limb bud cells (Fig. 6I,J). By (Sakai et al., 2003). We found that two copies of GGGGG motifs contrast, overexpression of the Znf219 mutant did not affect are present in the Col2a1 gene promoter region (Fig. 8A). To osteoblast differentiation of C3H10T1/2 and C2C12 cells (Fig. address whether Znf219 regulates the Col2a1 gene expression 6K,L). These results indicate that the mutant of Znf219 specifically through these GGGGG motifs, we generated mutants of the Col2a1 inhibits chondrocyte differentiation. To further verify these results, gene promoter luciferase construct (Fig. 8B). The mutations in the we generated synthetic microRNA for Znf219 and examined the Col2a1 gene suppressed the transcriptional activity of Znf219 (Fig. effect of knockdown of Znf219. We could clearly knockdown 8C). These results suggested that Znf219 regulates Col2a1 Znf219 after the introduction of microRNA for Znf219 by an expression through these GGGGG motifs. adenovirus system (Fig. 7A,B). Conversely, knockdown of Znf219 To confirm whether Znf219 directly binds to these regions of by microRNA system had no significant effects on endogenous the Col2a1 promoter, we performed a pull-down assay using an Sox9 expression (Fig. 7B). Knockdown of Znf219 inhibited Sox9- oligonucleotide probe containing the GGGGG motifs. We found induced expression of Col2a1, aggrecan Col11a2, Sox5 and Sox6 that Znf219 directly bound to the GGGGG motifs and this (Fig. 7C–G). These results support the importance of Znf219 in association was abolished by an unlabelled competitive probe (Fig. chondrocyte differentiation. 8D). We have revealed that Znf219 functions as a transcriptional activator for the Col2a1 promoter by recognizing two copies of the Znf219 regulates Col2a1 expression through two copies of GGGGG motif. Collectively, our results suggest that Znf219 GGGGG motifs functions as a molecular component of the transcription factory To understand the molecular mechanism by which Znf219 regulates formed by Sox9 in the promotion of chondrocyte differentiation chondrocyte differentiation, we attempt to identify the binding and plays an important role in chondrocyte differentiation. Regulation of chondrogenesis by Znf219 3785

Fig. 7. Inhibition of Sox9-dependent gene expression by knockdown of Znf219. (A)Knockdown of Znf219 was achieved with synthetic microRNA for Znf219. Adenovirus containing microRNA for Znf219 (miZnf219) was infected into C3H10T1/2 cells and the total RNA of the cells was determined by real-time RT-PCR. (B)Effect of knockdown of Znf219 on Sox9 expression. Adenovirus containing miZnf219 was infected into ATDC5 cells, and the total RNA of the cells was determined by real-time RT-PCR using Znf219 (left) and Sox9 (right) probes. (C–G)C3H10T1/2 cells infected with control (Cont), Sox9 adenovirus or miZnf219 adenovirus were incubated for 4 days. Total RNA of the cells was determined by real-time PCR for Col2a1 (B), aggrecan (C) Col11a2 (D), Sox5 (E) and Sox6 (F). Data represent mean ± s.d. (n3; *P<0.01 vs Sox9 by one-way analysis of variance).

Discussion Znf219-deficient mice will contribute to the better understanding of Transcription factors sequentially and harmoniously regulate the Znf219 in endochondral ossification. It has been shown that Sox9 expression of target genes by forming a large complex with the plays crucial roles in the development of the heart and testis partner molecules. To better understand the process of Sox9- (Akiyama et al., 2004a; Wagner et al., 1994). Because we observed regulated chondrogenesis, we attempted to identify the transcriptional partner of Sox9. As a result of performing the mammalian expression cloning, we have identified Znf219 as an interacting partner of Sox9. We showed that Znf219 not only

Journal of Cell Science physically, but also functionally, associates with Sox9 during chondrogenesis. First, the overexpression of Znf219 stimulates the transcriptional activity of Sox9 and its chondrogenic action. Second, Znf219 colocalized well with Sox9 and formed much smaller granules in the nucleus when Sox9 was present. The size of nuclear speckle has been shown to be inversely correlated with the state of transcription of the cells (Lamond and Spector, 2003); therefore, it is probable that interaction of Znf219 with Sox9 stimulates its transcriptional activity. Third, knockdown or inhibition of Znf219 markedly suppressed chondrocyte differentiation. Collectively, Znf219 is an important transcriptional factory member of Sox9 during chondrocyte differentiation. We showed that Znf219 had no effect on osteoblast differentiation from mesenchymal cells. In addition, Znf219 does not interact and Fig. 8. Znf219 regulates Col2a1 expression through GGGGG motifs. colocalize with Runx2. Moreover, Znf219 overexpression had little (A)Sequence of the Col2a1 promoter region. Two copies of GGGGG motifs effect on expression of Col1a2, an osteoblastic marker. Therefore, are present in the Col2a1 gene promoter (underlined). (B)Mutations in Znf219 seems to selectively regulate chondrocyte differentiation of luciferase constructs of the Col2a1 gene promoter. (C)Mutations in the undifferentiated mesenchymal cells. To support this, Znf219 was Col2a1 gene suppress the transcriptional activity of Znf219 and Sox9. specifically expressed in the developing limb buds of the mouse C3H10T1/2 cells were transfected with Col2a1 luciferase construct or mutants embryo and the proliferative chondrocytes of mouse growth plate. of the construct together with empty vector (Cont), Sox9 expression vector, Furthermore, expression of Znf219 is upregulated during Znf219 expression vector or both expression vectors as indicated and chondrocyte differentiation, probably through Sox9 family members. incubated for 48 hours. The luciferase activities of the cell lysates were then measured. (D)Znf219 binds to the GGGGG motif in the Col2a1 promoter. Conversely, Znf219 does not appear to function as upstream of Lysates of BOSC23 transfected with empty or Flag-Znf219 vector were Sox9 as overexpression or knockdown of Znf219 did not affect incubated with biotinylated oligonucleotide containing a putative Znf219 Sox9 expression. Because the physiological role of Znf219 in vivo binding element, the GGGGG motif, of the Col2a1 gene promoter. After is still unclear, it is necessary to perform the experiments using an precipitation with streptavidin beads, associated with the appropriate in vivo model. Therefore, we believe that analyses of oligonucleotide were determined by immunoblotting with anti-Flag antibody. 3786 Journal of Cell Science 123 (21)

that Znf219 was also highly expressed in hearts and testis, Znf219 anterior and posterior limb buds of E11.5 mouse embryos were pooled and digested with DMEM containing 0.1% collagenase type II (collagenase from Clostridium might be involved in development of heart and testis in association histolyticum, type II, Sigma) and 0.1% trypsin (trypsin, type I, from bovine pancreas, with Sox9. Znf219-deficent mice would be a good model to Sigma) at 37°C. The cells were dissociated by pipetting and then centrifuged at investigate the role for Znf219 in these organs. 300 g for 5 minutes. The pellet was resuspended in DMEM with 10% FBS at a ϫ 7 5 It has been reported that Znf219 recognizes the GGGGG motif density of 2 10 cells/ml. A total of 2ϫ10 cells/20 l of cell suspension was placed in a 12-well plate at 37°C in a humidified 5% CO2 incubator for 2 hours then 2 ml as the DNA binding element and functions as a transcriptional of culture medium was added to the wells. Twelve hours later, the cells were treated repressor for the Hmgn1 gene promoter activity (Sakai et al., 2003). with infection of adenoviruses and administration of Bmp2 (100 g/ml) and the We also showed that Znf219 binds to a GGGGG motif present in culture medium was replaced every other day thereafter. Cell differentiation was assessed by Alcian Blue and Alizarin Red staining. the Col2a1 gene promoter. Interestingly, potential Znf219 binding sites were found in the promoter region of the aggrecan and Col11a2 Isolation and culture of mouse primary chondrocytes gene promoters as well as the Col2a1 gene. In contrast to a previous Primary chondrocytes were isolated from the ribs of 4-week-old DDY mice and repetitively digested with 0.2% collagenase at 37°C. The fractioned cells were publication (Sakai et al., 2003), our data indicate that Znf219 filtered by 40 m nylon mesh (BO Bioscience), collected by centrifugation and functions as a transcriptional activator for the Col2a1 gene promoter. used as primary chondrocytes, except for the cells from the first digestion It is therefore probable that Znf219 has a function as a transcriptional (Gartland et al., 2005). The cell culture was performed in DMEM supplemented regulator to control not only negatively, but also positively, according with 10% bovine serum albumin, 0.1 mg/ml ascorbic acid, 5 mM - glycerophosphate (Sigma) and 10% FBS at 37°C in a humidified 5% CO2 to the binding molecule. It has been shown that Hmgn1 binds to the incubator. Infection of adenoviruses and administration of Bmp2 (100 g/ml) Sox9 gene and inhibits Sox9 expression and chondrogenesis were performed after 2 days. (Furusawa et al., 2006). Together, it is hypothesized that Znf219 Whole mount in situ hybridization negatively regulates chondrogenesis through the upregulation of C57BL/6 mouse embryos (at E11.5) were fixed in 4% paraformaldehyde in PBS Hmgn1. However, we did not observe that Znf219 affects Hmgn1 overnight at 4°C, treated with 6% hydrogen peroxide in PBS for 30 minutes, washed expression in C3H10T1/2 cells (data not shown). Thus, it is unlikely with PBS, incubated in 20 mg/ml proteinase K in PBS for 15 minutes, post-fixed with 4% paraformaldehyde and 0.2% glutaraldehyde in PBS for 20 minutes, washed that Znf219 forms a negative-feedback loop with Hmgn1. twice for 5 minutes in PBS, washed twice for 5 minutes in hybridization solution The Znf219 mutant that lacks the last zinc finger domain shows (50% formamide, 5ϫSSC, pH 7.0, 2% blocking-reagent, 50 g/ml yeast tRNA, the strongest dominant-negative effect on wild-type Znf219. This 0.1% Triton X-100, 0.1% CHAPS, 5 mM EDTA, 50 g/ml heparin), incubated at 70°C in hybridization solution for 1 hour and hybridized overnight with a gene- mutant still associates with Sox9 and retains the ability to bind to specific RNA probe. Post-hybridization washes were done in Solution I (50% the Col2a1 gene promoter region. As expected, the mutant markedly formamide, 5ϫSSC, pH 4.5, 0.1% Triton X-100, 0.1% CHAPS), Solution II (50% inhibited Bmp2-induced chondrocyte differentiation in C3H10T1/2 formamide, 2ϫSSC, pH 7.0) and KTBTx (10 mM KCl, 150 mM NaCl, 50 mM Tris- ϫ cells and mouse limb bud cells. It is therefore probable that the last HCl, pH 7.0, 0.1% Triton X-100, 10 KTB). Embryos were pre-blocked with 1.5% blocking reagent and KTBTx for 1 hour. After pre-blocking of anti-DIG in the zinc finger domain of Znf219 is necessary for chondrogenesis to mouse embryo, embryos were incubated overnight at room temperature. Whole-day recruit important transcriptional partners of Sox9 and Znf219. The post-antibody washes were done using KTBTx. Visualization began with a brief concept is compatible with the structural feature that Znf219 contains rinse in NTMTx (0.1 M NaCl, 0.1 M Tris-HCl, pH 9.5, 50 mM MgCl2, 0.1% Triton X-100) and continued according to the manufacturer’s instructions (NBT/BCIP). nine sets of zinc finger domains but does not contain other domains When the reaction was judged complete, embryos were post-fixed and photographed involved in transcription, chromatin remodeling and histone using a microscope. modification. Thus, Znf219 might function as an adaptor protein

Journal of Cell Science Immunocytochemical analyses that links Sox9 to transcriptional events. Several studies have Paraffin sections of E15.5 mouse tibia were prepared and immunostained with demonstrated that transcription factors exert their activities through polyclonal rabbit antibodies: -Znf219 antibody (Abcam, ab71279) at 1:500 dilution, docking with co-regulatory factors (Kornberg and Lorch, 1999; and -Sox9 (Santa Cruz Biotechnology, H-90) at 1:500 dilution. Immunoreactivity was visualized with a biotinylated anti-rabbit IgG secondary antibody using the ABC Dallas et al., 2000; Emerson, 2002; Wilsker et al., 2004). Although Vectastain Kit (Vector) and the peroxidase substrate DAB Kit (Vector) according to the precise biochemical role of Znf219 in chondrogenesis is the manufacturer’s protocol. unknown, identification of molecules associated with the last zinc finger domain of Znf219 might allow further understanding of the Constructs and transfection Venus cDNA, a variant of green fluorescent protein, was a kind gift from Dr molecular mechanisms by which Sox9 regulates chondrogenesis. Atsushi Miyawaki (RIKEN). Mutations of Znf219 were generated by subcloning In conclusion, our data provide evidence that Znf219 functions of the corresponding polymerase chain reaction (PCR) products into pcDNA3 as a molecular component of the transcription factory formed by tagged with Flag, Myc and Venus. Deletion mutants of Znf219 were generated by subcloning the corresponding PCR products into pcDNA3 tagged with a Flag Sox9 during the promotion of chondrocyte differentiation. These epitope in the N-terminus. HA-tagged Sox9 and dominant-negative Sox9 findings contribute to a better understanding of chondrogenesis (aa21–234) cDNA were kindly provided by Dr Vincent Harley (Prince Henry’s conducted by Sox9. Institute of Medical Research, Australia). HA-tagged HMG mutant Sox9 (aa181–510) was used as described previously (Hata et al., 2008). Transfection of C3H10T1/2 cells was carried out using FuGENE6 (Roche, Indianapolis, IN) Materials and Methods according to the manufacturer’s protocol. Cells and reagents A murine chondrogenic cell line (ATDC5), a murine mesenchymal cell line Generation of adenovirus (C3H10T1/2) and a human kidney cell line (BOSC23) were obtained from the The recombinant adenoviruses carrying a wild-type of Sox9, Sox5 and Sox6 or a RIKEN Cell Bank (Ibaragi, Japan) and cultured in a 1:1 mixture of Dulbecco’s mutant form of Znf219 were constructed by homologous recombination between the Modified Eagle’s Medium (DMEM) and Ham’s F-12 medium (Sigma) containing expression cosmid cassette (pAxCAwt) and the parental virus genome in 293 cells 10% fetal bovine serum (FBS). C2C12, CHO-K1 and COS7 cell lines were purchased using an adenovirus construction kit (Takara, Tokyo, Japan), as previously described from the RIKEN Cell Bank and cultured in alpha-modified minimal essential medium (Hata et al., 2008; Amano et al., 2009). The viruses showed no proliferative activity (-MEM) containing 10% FBS. These cell cultures were performed at 37°C in a owing to a lack of E1A–E1B. Infection of recombinant adenoviruses into C3H10T1/2 humidified 5% CO2 incubator. Recombinant Bmp2 was obtained from the conditioned cells, C2C12 cells, mouse limb bud cells or mouse primary chondrocytes was medium of CHO-K1 cells infected with Bmp2 adenovirus. The activity of Bmp2 was performed by incubation with the adenoviruses at a multiplicity of infection (moi) determined by a comparison with human recombinant Bmp2. of 20, except where specifically indicated.

Micromass culture of mouse limb bud cells Knockdown of Znf219 Limb bud cells were isolated from embryonic day (E) 11.5 DDY mouse embryos Znf219 microRNA was obtained using the BLOCK-iT Pol II miR RNAi Expression and mass cultured as described previously (Carlberg et al., 2001). Briefly, the System (Invitrogen). The target sequence was: 5Ј-TGCTGTTCATGTGG TT - Regulation of chondrogenesis by Znf219 3787

CTTAAGGAACGTTTTGGCCACTGACTGACGTTCCTTAAACCACATGAA-3Ј. wild-type, dominant-negative or HMG mutant Sox9 for 2 hours. Proteins associated The microRNA for the Znf219 construct was infected into C3H10T1/2 cells using with Znf219-Myc were determined by immunoblotting with anti-HA antibody. an adenovirus system. Oligonucleotide precipitation assay Immunoprecipitation and western blotting Cells were lysed in lysis buffer [20 mM HEPES (pH 7.4), 150 mM NaCl, 1 mM The cells were washed three times with ice-cold PBS and solubilized in lysis buffer EGTA, 1.5 mM MgCl2, 10% glycerol, 1% Triton X-100, 10 g/ml leupeptin, 1 mM [20 mM HEPES (pH 7.4), 150 mM NaCl, 1 mM EGTA, 1.5 mM MgCl2, 10% PMSF, 0.2 mM sodium orthovanadate]. The lysates pre-incubated with streptavidin- glycerol, 1% Triton X-100, 10 g/ml of aprotinin, 10 g/ml of leupeptin, 1 mM agarose beads were incubated with 1 g of a biotinylated double-stranded phenylmethylsulfonyl fluoride (PMSF), 0.2 mM sodium orthovanadate]. The lysates oligonucleotide probe that contained two repeats of the Znf219 binding element were then centrifuged for 15 minutes at 4°C at 16,000 g and with antibodies for 16 present in the Col2a1 gene promoter (sense primer: 5Ј-CCAGCTGGGGGCAG- hours at 4°C, followed by immunoprecipitation with protein G-agarose (Santa Cruz GGGGCGGCCCGCG-3Ј; anti-sense primer: 5Ј-CGCGGGCCGCCCCCTGCCCC- Biotechnology, Inc., Santa Cruz, CA, USA). Immunoprecipitates were washed five CAGCTG-3Ј) and 1 g of poly(dI-dC)poly(dI-dC) for 16 hours. DNA-bound proteins times with lysis buffer and boiled in sodium dodecyl sulfate (SDS) sample buffer were collected with streptavidin dynabeads (Dynabeads My One Streptavidin T1, containing 0.5 M -mercaptoethanol for 5 minutes. The supernatants were recovered Invitrogen) for 1 hour, washed with the lysis buffer, separated on an SDS- as immunoprecipitate samples. These samples were separated by SDS-polyacrylamide polyacrylamide gel and identified by western blotting. gel electrophoresis (PAGE), transferred to nitrocellulose membranes, immunoblotted with corresponding antibodies and visualized with horseradish peroxidase coupled Alcian Blue staining to anti-mouse rabbit IgG antibodies (Jackson ImmunoResearch Laboratories, West The cells were rinsed twice with PBS, fixed in 4% buffered paraformaldehyde and Grove, PA, USA) with enhancement by Electrochemiluminescence (ECL) Advance stained with 1% Alcian Blue in 5% acetic acid for 10 minutes. Western Blotting Detection Kits (Amersham). Alizarin Red staining Luciferase reporter gene assay The cells were cultured in the presence of ascorbic acid (100 g/ml) and - The luciferase reporter construct driven by the Col2a1 gene promoter was co- glycerophosphate (5 mM), rinsed twice with PBS, fixed in 4% buffered transfected with the thymidine kinase-renilla luciferase construct (Promega) into paraformaldehyde and stained with 1% Alizarin Red solution for 5 minutes. C3H10T1/2 cells. Two days after transfection, the cells were lysed and luciferase activity was determined using specific substrates in a luminometer (Promega) Alkaline phosphatase staining according to the manufacturer’s protocol. Transfection efficiency was normalized by The cells were washed with PBS, fixed with 3.7% formaldehyde and stained with a determining the activity of renilla luciferase. mixture of 330 g/ml Nitro Blue Tetrazolium, 165 g/ml bromochloroindoyl phosphate, 100 mM NaCl, 5 mM MgCl2 and 100 mM Tris (pH 9.5). Reverse-transcriptase PCR and real-time PCR Total RNA was isolated from cells and treated with DNase using a commercial total Fluorchrome staining RNA isolation system (NucleoSpin RNAII, MACHEREY-NAGLE, Germany). After C3H10T1/2 cells were transfected with Venus-Znf219, DsRed-Sox9, or DsRed- denaturation of total RNA at 70°C for 10 minutes, cDNA was synthesized using an Runx2. Cultured cells were washed three times with ice-cold PBS and fixed with oligo-dT primer and reverse transcriptase (Prime Script 1st Strand cDNA Synthesis 10% paraformaldehyde in PBS for 10 minutes. After incubation for 20 minutes with Kit, Takara, Japan). Amplifications were performed using the specific primers for - 0.2% Triton X-100 in PBS, the cells were blocked with PBS containing 1% bovine actin (sense primer: 5Ј-TTCGAGCAGGAGATGGCCAC-3Ј; anti-sense primer: 5Ј- serum albumin for 1 hour, incubated with 0.1 mg/ml DAPI in PBS containing 1% TCTGCATCCTGTCAGCAAT-3Ј), Col2a1 (sense primer: 5Ј-TATGGAAGCCCT- bovine serum albumin, then washed with 0.2% Triton X-100 in PBS and visualized CATCTTGC-3Ј; anti-sense primer: 5Ј-TCACCTCTGGGTCCTTG TTC-3Ј), aggrecan by a fluorescence microscope (Zeiss). (sense primer: 5Ј-TCCTCTCCGGTGGCAAAGAAGTTG-3Ј; anti-sense primer: 5Ј- CCAAGTTCCAGGGTCACTGTTACCG-3Ј) or Znf219 (sense primer: 5Ј-AGTAC- Statistical analysis GACCCAAATGGTGAGG-3Ј; anti-sense primer: 5Ј-GGTGCAGGGAAGTCAGA- The data were statistically analyzed by Student’s t-test or analysis of variance. Data GAG-3Ј). Amplification products were loaded onto agarose gel and stained with represent mean ± s.d. ethidium bromide. Real-time PCR was performed by the Taqman PCR protocol using the ABI 7300 real-time PCR system (Applied Biosystems, Branchburg, NJ, USA). This work was supported in part by the Ministry of Education, Taqman primers and probes used for amplification were as follows: -actin (probe: Journal of Cell Science  CCTGGCTGCCTCAACACCTCAACCC-TAMURA; sense primer: 5Ј-TTAATT - Science, Sports and Culture Grants-in-Aid for Scientific Research TCTGAA TGGCCC AGGTCT-3Ј; anti-sense primer: 5Ј-ATTGGTCTCAAGTCAGT- (T.Y., K.H. and R.N.), the 21st Century COE Program (T.Y. and R.N.), GTACAGG-3Ј), Col2a1 (probe: TGAGGTTGCCAGCCGCTTCGTCCA-TAMURA; The Uehara Memorial Foundation (R.N.) and The Astellas Foundation sense primer: 5Ј-CCTCCGTCTACTGTCCACTGAG-3Ј; anti-sense primer: 5Ј- for Research on Metabolic Disorders (R.N.). TGGAGCCCTG GATGAGCAAG-3Ј), aggrecan (probe: ACCGTCTCTCCGCATC- Ј Ј CACCCAGG-TAMURA; sense primer: 5 -TCACTGTTACCGCCACTTTCC-3 ; References anti-sense primer: 5Ј-TGCTGCTCAGATGTGACTGC-3Ј), Col11a2 (probe: CCCCA- Ј Akiyama, H. (2008). Control of chondrogenesis by the transcription factor Sox9. Mod. GAAGCACCCCTCCCCGTC-TAMURA; sense primer: 5 -GATGCCTCCT - Rheumatol. 18, 213-219. Ј Ј Ј TCGCAGACCT-3 ; anti-sense primer: 5 -CCTAGCCCATGAAGCAGACAG-3 ), Akiyama, H., Chaboissier, M. C., Martin, J. F., Schedl, A. and de Crombrugghe, B. Znf219 (probe: TGTCCCTCCAACCAGCAGCCCGTT-TAMURA; sense primer: 5Ј- (2002). The transcription factor Sox9 has essential roles in successive steps of the GGGACCCAGGCGAGATCC-3Ј; anti-sense primer: 5Ј-CACGGCCCGCTTCA- chondrocyte differentiation pathway and is required for expression of Sox5 and Sox6. GATC-3Ј), Sox5 (probe: ACGTTGGGCAACAAGCACAGATCCCC-TAMURA; Genes Dev. 16, 2813-2828. sense primer: 5Ј-AGGCAGGAAATGCGACAGTAC-3Ј; anti-sense primer: 5Ј-CTCG- Akiyama, H., Chaboissier, M. C., Behringer, R. R., Rowitch, D. H., Schedl, A., GAGGGCAGGTGAGG-3Ј), Sox6 (probe: TCAGCAGCAGCGTTCACGAGCAGC- Epstein, J. A. and de Crombrugghe, B. (2004a). Essential role of Sox9 in the pathway TAMURA; sense primer: 5Ј-TACCCACAGCTCCCCTGAAG-3Ј; anti-sense primer: that controls formation of cardiac valves and septa. Proc. Natl. Acad. Sci. USA 101, 5Ј-CTCACCTTCAGTGGCAAGAGC-3Ј), Col11a2 (probe: AGTCTCCTCCCTTG- 6502-6507. GCCCCTCCCC-TAMURA; sense primer: 5Ј-CAGAGCAGTGTGCAATATGATCC- Akiyama, H., Lyons, J. P., Mori-Akiyama, Y., Yang, X., Zhang, R., Zhang, Z., Deng, J. 3Ј; anti-sense primer: 5Ј-ATGTCCACAACAGGTGTCAGG-3Ј) and Sox9 (probe: M., Taketo, M. M., Nakamura, T., Behringer, R. R. et al. (2004b). Interactions between CCGCCCATCACCCGCTCGCAATAC-TAMURA; sense primer: 5Ј-CCTTCAAC- Sox9 and -catenin control chondrocyte differentiation. Genes Dev. 18, 1072-1087. CTTCCTCACTACAGC-3Ј; anti-sense primer: 5Ј-GGTGGAGTAGAGCCCTGAGC- Akiyama, H., Stadler, H. S., Martin, J. F., Ishii, T. M., Beachy, P. A., Nakamura, T. and 3). Expression levels of mRNA were normalized by -actin mRNA expression. de Crombrugghe, B. (2007). Misexpression of Sox9 in mouse limb bud mesenchyme induces polydactyly and rescues hypodactyly mice. Matrix Biol. 26, 224-233. In vitro binding assay using recombinant His-tagged Sox9 protein and Amano, K., Hata, K., Sugita, A., Takigawa, Y., Ono, K., Wakabayashi, M., Kogo, M., Nishimura, R. and Yoneda, T. (2009). Sox9 family members negatively regulate Znf219 maturation and calcification of chondrocytes through up-regulation of parathyroid The amplification product containing the coding region of Sox9 was subcloned into hormone-related protein. Mol. Biol. Cell 20, 4541-4551. pCold1 vector (Takara). The construct was transformed into the BL21 E. coli strain Bi, W., Deng, J. M., Zhang, Z., Behringer, R. R. and de Crombrugghe, B. (1999). Sox9 and then His-tagged Sox9 protein was generated and purified with Ni-NTA magnetic is required for cartilage formation. Nat. Genet. 22, 85-89. agarose beads (QIAGEN) according to the manufacturer’s protocol. The purified Bi, W., Huang, W., Whitworth, D. J., Deng, J. M., Zhang, Z., Behringer, R. R. and de His-tagged Sox9 protein immobilized with the beads was used for the binding assay Crombrugghe, B. (2001). Haploinsufficiency of Sox9 results in defective cartilage with Znf219. The association of Sox9 with mutants of Flag-tagged Znf219 was primordia and premature skeletal mineralization. Proc. Natl. Acad. Sci. USA 98, 6698- determined by immunoblotting with anti-Flag antibody, following precipitation of 6703. cell lysates with the recombinant His-tagged Sox9 protein. To determine the Bridgewater, L. C., Lefebvre, V. and de Crombrugghe, B. (1998). Chondrocyte-specific association of Znf219 with mutant Sox9 in vitro, Znf219-Myc protein immobilized enhancer elements in the Col11a2 gene resemble the Col2a1 tissue-specific enhancer. on Dyna Protein G beads were incubated with cell lysates expressing HA-tagged J. Biol. Chem. 273, 14998-15006. 3788 Journal of Cell Science 123 (21)

Carlberg, A. L., Pucci, B., Rallapalli, R., Tuan, R. S. and Hall, D. J. (2001). Efficient Lefebvre, V., Li, P. and de Crombrugghe, B. (1998). A new long form of Sox5 (L-Sox5), chondrogenic differentiation of mesenchymal cells in micromass culture by retroviral Sox6 and Sox9 are coexpressed in chondrogenesis and cooperatively activate the type gene transfer of BMP-2. Differentiation 67, 128-138. II collagen gene. EMBO J. 17, 5718-5733. Carrozza, M. J., Utley, R. T., Workman, J. L. and Cote, J. (2003). The diverse functions Lefebvre, V., Behringer, R. R. and de Crombrugghe, B. (2001). L-Sox5, Sox6, and of histone acetyltransferase complexes. Trends Genet. 19, 321-329. Sox9 control essential steps of the chondrocyte differentiation pathway. Osteoarthr. Chaboissier, M. C., Kobayashi, A., Vidal, V. I., Lutzkendorf, S., van de Kant, H. J., Cartil. 9 Suppl. A, S69-S75. Wegner, M., de Rooij, D. G., Behringer, R. R. and Schedl, A. (2004). Functional Lynn, F. C., Smith, S. B., Wilson, M. E., Yang, K. Y., Nekrep, N. and German, M. S. analysis of Sox8 and Sox9 during sex determination in the mouse. Development 131, (2007). Sox9 coordinates a transcriptional network in pancreatic progenitor cells. Proc. 1891-1901. Natl. Acad. Sci. USA 104, 10500-10505. Dallas, P. B., Pacchione, S., Wilsker, D., Bowrin, V., Kobayashi, R. and Moran, E. Maniatis, T. and Reed, R. (2002). An extensive network of coupling among gene (2000). The human SWI-SNF complex protein p270 is an ARID family member with expression machines. Nature 416, 499-506. non-sequence-specific DNA binding activity. Mol. Cell. Biol. 20, 3137-3146. McDowall, S., Argentaro, A., Ranganathan, S., Weller, P., Mertin, S., Mansour, S., de Crombrugghe, B., Lefebvre, V., Behringer, R. R., Bi, W., Murakami, S. and Tolmie, J. and Harley, V. (1999). Functional and structural studies of wild type SOX9 Huang, W. (2000). Transcriptional mechanisms of chondrocyte differentiation. Matrix and mutations causing campomelic dysplasia. J. Biol. Chem. 274, 24023-24030. Biol. 19, 389-394. Morais da Silva, S., Hacker, A., Harley, V., Goodfellow, P., Swain, A. and Lovell- de Crombrugghe, B., Lefebvre, V. and Nakashima, K. (2001). Regulatory mechanisms Badge, R. (1996). Sox9 expression during gonadal development implies a conserved in the pathways of cartilage and bone formation. Cell Biol. 13, 721-727. role for the gene in testis differentiation in mammals and birds. Nat. Genet. 14, 62-68. De Santa Barbara, P., Bonneaud, N., Boizet, B., Desclozeaux, M., Moniot, B., Sudbeck, Muramatsu, S., Wakabayashi, M., Ohno, T., Amano, K., Ooishi, R., Sugahara, T., P., Scherer, G., Poulat, F. and Berta, P. (1998). Direct interaction of SRY-related Shiojiri, S., Tashiro, K., Suzuki, Y., Nishimura, R. et al. (2007). Functional gene protein SOX9 and regulates transcription of the human anti- screening system identified TRPV4 as a regulator of chondrogenic differentiation. J. Mullerian hormone gene. Mol. Cell. Biol. 18, 6653-6665. Biol. Chem. 282, 32158-32167. Ducy, P., Zhang, R., Geoffroy, V., Ridall, A. L. and Karsenty, G. (1997). Osf2/Cbfa1: Provot, S. and Schipani, E. (2005). Molecular mechanisms of endochondral bone a transcriptional activator of osteoblast differentiation. Cell 89, 747-754. development. Biochem. Biophys. Res. Commun. 328, 658-665. Emerson, B. M. (2002). Specificity of gene regulation. Cell 109, 267-270. Sakai, T., Toyoda, A., Hashimoto, K. and Maeda, H. (2000). Isolation and Furumatsu, T., Tsuda, M., Taniguchi, N., Tajima, Y. and Asahara, H. (2004). Smad3 characterization of a novel zinc finger gene, ZNF219, and mapping to the human induces chondrogenesis through the activation of SOX9 via CREB-binding protein/p300 14q11 region. DNA Res. 7, 137-141. recruitment. J. Biol. Chem. 280, 8343-8350. Sakai, T., Hino, K., Wada, S. and Maeda, H. (2003). Identification of the DNA binding Furumatsu, T., Tsuda, M., Yoshida, K., Taniguchi, N., Ito, T., Hashimoto, M., Ito, T. specificity of the human ZNF219 protein and its function as a transcriptional repressor. and Asahara, H. (2005). Sox9 and p300 cooperatively regulate chromatin-mediated DNA Res. 10, 155-165. transcription. J. Biol. Chem. 280, 35203-35208. Sekiya, I., Tsuji, K., Koopman, P., Watanabe, H., Yamada, Y., Shinomiya, K., Nifuji, Furusawa, T., Lim, J. H., Catez, F., Birger, Y., Mackem, S. and Bustin, M. (2006). A. and Noda, M. (2000). SOX9 enhances aggrecan gene promoter/enhancer activity Down-regulation of nucleosomal binding protein HMGN1 expression during and is up-regulated by retinoic acid in a cartilage-derived cell line, TC6. J. Biol. Chem. embryogenesis modulates Sox9 expression in chondrocytes. Mol. Cell. Biol. 26, 592-604. 275, 10738-10744. Gartland, A., Mechler, J., Mason-Savas, A., MacKay, C. A., Mailhot, G., Marks, S. C. Shea, C. M., Edgar, C. M., Einhorn, T. A. and Gerstenfeld, L. C. (2003). BMP and Odgren, P. R. (2005). In vitro chondrocyte differentiation using costochondral treatment of C3H10T1/2 mesenchymal stem cells induces both chondrogenesis and chondrocytes as a source of primary rat chondrocyte cultures: an improved isolation and osteogenesis. J. Cell. Biochem. 90, 1112-1127. cryopreservation method. Bone 37, 530-544. Shukunami, C., Shigeno, C., Atsumi, T., Ishizeki, K., Suzuki, F. and Hiraki, Y. (1996). Giordano, J., Prior, H. M., Bamforth, J. S. and Walter, M. A. (2001). Genetic study of Chondrogenic differentiation of clonal mouse embryonic cell line ATDC5 in vitro: SOX9 in a case of campomelic dysplasia. Am. J. Med. Genet. 98, 176-181. differentiation-dependent gene expression of parathyroid hormone (PTH)/PTH-related Glass, D. A., II and Karsenty, G. (2007). In vivo analysis of Wnt signaling in bone. peptide receptor. J. Cell Biol. 133, 457-468. Endocrinology 148, 2630-2634. Shum, L. and Nuckolls, G. (2002). The life cycle of chondrocytes in the developing Hata, K., Nishimura, R., Muramatsu, S., Matsuda, A., Matsubara, T., Amano, K., skeleton. Arthritis Res. 4, 94-106. Ikeda, F., Harley, V. R. and Yoneda, T. (2008). Paraspeckle protein p54nrb links Sox9- Smits, P., Li, P., Mandel, J., Zhang, Z., Deng, J. M., Behringer, R. R., de Crombrugghe, mediated transcription with RNA processing during chondrogenesis in mice. J. Clin. B. and Lefebvre, V. (2001). The transcription factors L-Sox5 and Sox6 are essential Invest. 118, 3098-3108. for cartilage formation. Dev. Cell. 1, 277-290. Ikeda, T., Kamekura, S., Mabuchi, A., Kou, I., Seki, S., Takato, T., Nakamura, K., Smits, P., Dy, P., Mitra, S. and Lefebvre, V. (2004). Sox5 and Sox6 are needed to Kawaguchi, H., Ikegawa, S. and Chung, U. I. (2004). The combination of SOX5, develop and maintain source, columnar, and hypertrophic chondrocytes in the cartilage

Journal of Cell Science SOX6, and SOX9 (the SOX trio) provides signals sufficient for induction of permanent growth plate. J. Cell Biol. 164, 747-758. cartilage. Arthritis Rheum. 50, 3561-3573. Sock, E., Pagon, R. A., Keymolen, K., Lissens, W., Wegner, M. and Scherer, G. (2003). Jin, E. J., Lee, S. Y., Choi, Y. A., Jung, J. C., Bang, O. S. and Kang, S. S. (2006). BMP- Loss of DNA-dependent dimerization of the transcription factor SOX9 as a cause for 2-enhanced chondrogenesis involves p38 MAPK-mediated down-regulation of Wnt-7a campomelic dysplasia. Hum. Mol. Genet. 12, 1439-1447. pathway. Mol Cells 22, 353-359. Stolt, C. C., Lommes, P., Sock, E., Chaboissier, M. C., Schedl, A. and Wegner, M. Kameoka, S., Duque, P. and Konarska, M. M. (2004). p54 (nrb) associates with the 5 (2003). The Sox9 transcription factor determines glial fate choice in the developing splice site within large transcription/splicing complexes. EMBO J. 23, 1782-1791. spinal cord. Genes Dev. 17, 1677-1689. Kawakami, Y., Tsuda, M., Takahashi, S., Taniguchi, N., Esteban, C. R., Zemmyo, M., Tontonoz, P., Hu, E. and Spiegelman, B. M. (1994). Stimulation of adipogenesis in Furumatsu, T., Lotz, M., Belmonte, J. C. and Asahara, H. (2005). Transcriptional fibroblasts by PPAR gamma 2, a lipid-activated transcription factor. Cell 79, 1147-1156. coactivator PGC-1alpha regulates chondrogenesis via association with Sox9. Proc. Tsuda, M., Takahashi, S., Takahashi, Y. and Asahara, H. (2003). Transcriptional co- Natl. Acad. Sci. USA 102, 2414-2419. activators CREB-binding protein and p300 regulate chondrocyte-specific gene expression Kitagawa, H., Fujiki, R., Yoshimura, K., Mezaki, Y., Uematsu, Y., Matsui, D., Ogawa, via association with Sox9. J. Biol. Chem. 278, 27224-27229. S., Unno, K., Okubo, M., Tokita, A. et al. (2003). The chromatin-remodeling complex Ueta, C., Iwamoto, M., Kanatani, N., Yoshida, C., Liu, Y., Enomoto-Iwamoto, M., WINAC targets a to promoters and is impaired in Williams syndrome. Ohmori, T., Enomoto, H., Nakata, K., Takada, K. et al. (2001). Skeletal malformations Cell 113, 905-917. caused by overexpression of Cbfa1 or its dominant negative form in chondrocytes. J. Kobayashi, T. and Kronenberg, H. (2005). Minireview: transcriptional regulation in Cell Biol. 153, 87-100. development of bone. Endocrinology 146, 1012-1017. Wagner, T., Wirth, J., Meyer, J., Zabel, B., Held, M., Zimmer, J., Pasantes, J., Komori, T., Yagi, H., Nomura, S., Yamaguchi, A., Sasaki, K., Deguchi, K., Shimizu, Bricarelli, F. D., Keutel, J., Hustert, E. et al. (1994). Autosomal sex reversal and Y., Bronson, R. T., Gao, Y. H., Inada, M. et al. (1997). Targeted disruption of Cbfa1 campomelic dysplasia are caused by mutations in and around the SRY-related gene results in a complete lack of bone formation owing to maturational arrest of osteoblasts. SOX9. Cell 79, 1111-1120. Cell 89, 755-764. Wilsker, D., Patsialou, A., Zumbrun, S. D., Kim, S., Chen, Y., Dallas, P. B. and Moran, Kornberg, R. D. and Lorch, Y. (1999). Twenty-five years of the nucleosome, fundamental E. (2004). The DNA-binding properties of the ARID-containing subunits of yeast and particle of the eukaryote chromosome. Cell 98, 285-294. mammalian SWI/SNF complexes. Nucleic Acids Res. 32, 1345-1353. Kronenberg, H. M. (2003). Developmental regulation of the growth plate. Nature 423, Wu, Z., Xie, Y., Bucher, N. L. and Farmer, S. R. (1995). Conditional ectopic expression 332-336. of C/EBP beta in NIH-3T3 cells induces PPAR gamma and stimulates adipogenesis. Kronenberg, H. M. (2006). PTHrP and skeletal development. Ann. NY Acad. Sci. 1068, Genes Dev. 9, 2350-2363. 1-13. Wunderle, V. M., Critcher, R., Hastie, N., Goodfellow, P. N. and Schedl, A. (1998). Lamond, A. I. and Spector, D. L. (2003). Nuclear speckles: a model for nuclear organelles. Deletion of long-range regulatory elements upstream of SOX9 causes campomelic Nat. Rev. Mol. Cell Biol. 4, 605-612. dysplasia. Proc. Natl. Acad. Sci. USA 95, 10649-10654. Lefebvre, V., Zhou, G., Mukhopadhyay, K., Smith, C. N., Zhang, Z., Eberspaecher, Yano, F., Kugimiya, F., Ohba, S., Ikeda, T., Chikuda, H., Ogasawara, T., Ogata, N., H., Zhou, X., Sinha, S., Maity, S. N. and de Crombrugghe, B. (1996). A18-base-pair Takato, T., Nakamura, K., Kawaguchi, H. et al. (2005). The canonical Wnt signaling sequence in the mouse proalpha1(II) collagen gene is sufficient for expression in pathway promotes chondrocyte differentiation in a Sox9-dependent manner. Biochem. cartilage and binds nuclear proteins that are selectively expressed in chondrocytes. Mol. Biophys. Res. Commun. 333, 1300-1308. Cell. Biol. 16, 4512-4523. Yoshida, C. A., Yamamoto, H., Fujita, T., Furuichi, T., Ito, K., Yamana, K., Zanma, Lefebvre, V., Huang, W., Harley, V. R., Goodfellow, P. N. and de Crombrugghe, B. A., Takada, K., Ito, Y. and Komori, T. (2004). Runx2 and Runx3 are essential for (1997). SOX9 is a potent activator of the chondrocyte-specific enhancer of the pro chondrocyte maturation, and Runx2 regulates limb growth through induction of Indian alpha1 (II) collagen gene. Mol. Cell. Biol. 17, 2336-2346. hedgehog. Genes Dev. 18, 952-963.