trps1 plays a pivotal role downstream of gdf5 signaling in promoting chondrogenesis and apoptosis of...

Post on 16-Nov-2023

0 Views

Category:

Documents

0 Downloads

Preview:

Click to see full reader

TRANSCRIPT

© 2008 The Authors

Genes to Cells (2008)

13

, 355–363

Journal compilation © 2008 by the Molecular Biology Society of Japan/Blackwell Publishing Ltd.

355

DOI: 10.1111/j.1365-2443.2008.01170.x

Blackwell Publishing IncMalden, USAGTCGenes to Cells1356-95971365-2443© 2008 The AuthorsJournal compilation © 2008 by the Molecular Biology Society of Japan/Blackwell Publishing Ltd.XXXOriginal ArticlesTrps1 plays a pivotal role downstream of Gdf5S Itoh et al.

Trps1 plays a pivotal role downstream of Gdf5 signaling in promoting chondrogenesis and apoptosis of ATDC5 cells

Shunji Itoh

1

, Seiji Kanno

2

, Zhibo Gai

1

, Hiroki Suemoto

1

, Motohisa Kawakatsu

1

, Hiroyuki Tanishima

1

, Yoshifumi Morimoto

1

, Katsuhiro Nishioka

1

, Ikuji Hatamura

1

, Munehito Yoshida

2

and Yasuteru Muragaki

1,

*

1

First Department of Pathology and

2

Department of Orthopedic Surgery, Wakayama Medical University, Wakayama, Japan

Tricho-rhino-phalangeal syndrome (TRPS) is an autosomal dominant skeletal disorder caused bymutations of

TRPS1

. Based on the similar expression patterns of Trps1 and Gdf5, we hypothesizeda possible functional interaction between these two molecules. Using a chondrogenic cell line(ATDC5), we investigated the association of Gdf5-mediated signaling pathways with Trps1 andthe phenotypic changes of ATDC5 cells due to over-expression or suppression of Trps1. Treatmentof cells with Gdf5 enhanced Trps1 protein levels and phosphorylation of p38 mitogen-activatedprotein kinase (MAPK) in a dose-dependent manner. Nuclear translocation of Trps1 was alsoinduced by Gdf5. These effects were blocked by a dominant negative form of activin-linked kinase6 (dn-Alk6) and by SB203580, an inhibitor of the p38 MAPK pathway. Conversely, Gdf5 expressionwas suppressed by the over-expression of Trps1. Trps1-overexpressing ATDC5 (O/E) cells differ-entiated into chondrocytes more quickly than mock-infected control cells, whereas cells trans-fected with dn-Alk6 showed slower differentiation. On the other hand, O/E cells showed anincrease of apoptosis along with the up-regulation of cleaved caspase 3 and down-regulation ofBcl-2, whereas dn-Alk6 cells showed suppression of apoptosis. In conclusion, Trps1 acts downstreamof the Gdf5 signaling pathway and promotes the differentiation and apoptosis of ATDC5 cells.

Introduction

Tricho-rhino-phalangeal syndrome (TRPS) is an autosomaldominant hereditary disorder characterized by craniofacialand skeletal anomalies that is caused by mutations in

TRPS1

, which encodes a novel transcription factor ofthe GATA family (Momeni

et al

. 2000). Molecular analysishas revealed that TRPS1 contains nine zinc finger motifs,including one GATA-type DNA-binding domain, whiletwo at the carboxyl terminus show homology with theIkaros family of lymphoid transcriptional factors (Chrousos& Kino 2005). It has been demonstrated that Trps1 is atranscriptional repressor (Malik

et al

. 2001), which interactswith the dynein light chain and RING finger protein 4(Rnf4) in the nucleus (Kaiser

et al

. 2003b).Trps1 is mainly expressed in the joints and in the limb

growth plate cartilages during late embryogenesis (Kunath

et al

. 2002). Mice with Trps1 deficiency die of respiratoryfailure soon after birth and show changes such as skeletaldeformities that resemble those seen in human TRPSpatients (Malik

et al

. 2002). These findings suggest thatTrps1 acts as a transcriptional regulator during the processof cartilage formation. However, the molecules that lieupstream and downstream of Trps1 are largely unknown.

During development of the cartilage and joints, numer-ous cytokines and growth factors may have a role. Thetransforming growth factor-

β

(TGF-

β

) superfamily ofproteins, especially TGF-

β

itself and the bone morpho-genetic proteins (BMPs), play a central role in the processof cartilage formation. The BMP subfamily consists of atleast 20 proteins that can be placed into distinct subgroupsbased on their amino acid sequences. During chondro-genesis, BMP2 and BMP4 (a subgroup of BMPs)stimulate the proliferation of chondrocytes and counteractthe effects of FGFs (Yoon

et al

. 2006), while BMP7promotes the differentiation of chondrocytes and the accu-mulation of matrix molecules (Asahina

et al

. 1996). Growth

Communicated by

: Kohei Miyazono*

Correspondence

: Email: ymuragak@wakayama-med.ac.jp

S Itoh

et al.

Genes to Cells (2008)

13

, 355–363

© 2008 The AuthorsJournal compilation © 2008 by the Molecular Biology Society of Japan/Blackwell Publishing Ltd.

356

and differentiation factors (GDFs) 5, 6 and 7 form theother subgroup of BMPs, and play an important role injoint formation (Ducy & Karsenty 2000). Although thefunctions of these proteins are broadly known, the precisemolecular mechanisms involved are not fully understood.

A human genetic skeletal disease caused by mutationsof

GDF5

(Holder-Espinasse

et al

. 2004; Dawson

et al

. 2006;Wang

et al

. 2006) has similar phenotypic features (likebrachydactyly) to those of TRPS. In addition, brachypo-dism (bp) caused by a Gdf5-mutation leads to limbshortening and abnormalities of joint formation in mice(Storm

et al

. 1994; Storm & Kingsley 1996), in whichTrps1 is strongly expressed during a development. Thesefindings raise the possibility that GDF5 and TRPS1could be linked in the processes of joint and cartilageformation.

ATDC5 cells are a well-characterized chondrogeniccell line derived from a mouse teratocarcinoma thatmimics the multistep process of chondrocyte differenti-ation in culture. Several studies have demonstrated thatthe TGF-

β

superfamily of proteins induces chondrogenesisin cultured ATDC5 cells (Shukunami

et al

. 1998; Naka-mura

et al

. 1999). In addition, GDF5 has been shown topromote chondrocyte differentiation and activation ofp38 MAPK in these cells (Nakamura

et al

. 1999).In the present study, we examined the functional

interaction between Gdf5 and Trps1 using ATDC5 cells.Our results indicated that Gdf5 and Trps1 mutually reg-ulate their expression and that Trps1 acts downstream ofGdf5 during the differentiation and apoptosis of culturedATDC5 cells.

Results

Trps1 and Gdf5 regulate each other via feedback

Immunohistochemical analysis showed that Trps1 wasstrongly expressed by chondrocytes in the phalangealjoints of E15.5 embryos (Fig. 1A,B). We found that theimmunohistochemical localization of Trps1 was similarto the pattern of Gdf5 expression on E15.5 (Fig. 1C,D),as reported elsewhere (Storm & Kingsley 1996). Thus,we hypothesized that Trps1 might interact with Gdf5. Totest this hypothesis, we examined possible interactionsbetween Trps1 expression and Gdf5 signaling using theATDC5 chondrogenic cell line.

Expression of Trps1 by ATDC5 cells was not observedin the absence of insulin, while it gradually increasedin the presence of insulin (Fig. 2A). Because Gdf5 inducesthe chondrogenic differentiation of ATDC5 cells (Nakamura

et al

. 1999), we assumed that Gdf5 signaling could alsoinduce the expression of Trps1. In fact, we confirmed thatGdf5 induced the expression of Trps1 in a dose-dependentmanner (Fig. 2B). Phosphorylation of p38 mitogen-activated protein kinase (MAPK) increased in parallelwith the up-regulation of Trps1 (Fig. 2B). Conversely,Trps1 over-expression significantly suppressed Gdf5 expres-sion compared with that in control ATDC5 cells (Fig. 2C).In addition, treatment with Gdf5 led to translocation ofTrps1 into the nucleus (Fig. 2D). Strong positive immuno-staining for Trps1 was seen in the nuclei of cells at 1 hafter treatment with Gdf5, while there was no nuclearstaining for Trps1 in the absence of Gdf5 (Fig. 2D).

Figure 1 Expression of Trps1 in thedeveloping limb. Immunohistochemicallocalization of Trps1 in the forelimb of anE15.5 embryo (A, B). In situ hybridizationof Gdf5 (C, D) in the forelimb of an E15.5embryo. Scale bars indicate 500 μm in Aand C, and 100 μm in B and D.

Trps1 plays a pivotal role downstream of Gdf5

© 2008 The Authors

Genes to Cells (2008)

13

, 355–363

Journal compilation © 2008 by the Molecular Biology Society of Japan/Blackwell Publishing Ltd.

357

Gdf5 signaling regulates Trps1 via activin-linked kinase 6 (Alk6) and p38 MAPK

Since Gdf5 utilizes Alk6 as its receptor, we examinedwhether or not up-regulation of Trps1 by Gdf5 occursvia Alk6. When we transfected ATDC5 cells with anexpression vector bearing a dominant negative form ofAlk6 (dn-Alk6), the expression of Trps1 was stronglysuppressed at the protein levels although the suppressionof the mRNA levels was not so prominent (Fig. 3A).p38 MAPK is reported to be activated by the TGF-

β

superfamily of proteins, including BMPs (Moriguchi

et al

. 1996). To test whether p38 MAPK mediates theinteraction of Gdf5 with Trps1, we examined phospho-rylation of p38 MAPK by dn-Alk6 as well as expressionof Trps1 after inhibition of p38 MAPK. In cells trans-fected with dn-Alk6, phosphorylation of p38 MAPKwas suppressed, while no significant change was seen inmock-transfected control cells (Fig. 3B). In addition,when we treated ATDC5 cells with a p38 MAPK inhib-itor (SB203580), Trps1 protein levels were dramaticallysuppressed (Fig. 3C).

Gdf5–Trps1 signaling promotes chondrogenic differentiation of ATDC5 cells

To investigate the role of Gdf5–Trps1 signaling in chon-drogenesis, we examined the expression of chondrogenicmarkers (type II and type X collagen) and Trps1, as wellas Alcian Blue staining. Cartilage-like nodules stained byAlcian Blue were larger and grew faster in cultures ofTrps1-overexpressing and Gdf5-treated cells comparedwith cultures of control cells (Fig. 4A). Conversely, in

cultures of cells transfected with dn-Alk6, the cartilage-like nodules were fewer and grew more slowly comparedwith cultures of control cells (Fig. 4A). In addition, theexpression of chondrogenic marker proteins (type IIand type X collagen) was up-regulated earlier in Trps1-overexpressing cells compared with control cells, whereas

Figure 2 Expression of Trps1 in ATDC5cells. (A) Expression of Trps1 mRNA inATDC5 cells detected by Northern blotting.ATDC5 cells were treated with insulin andthen incubated for 3 days (3d), 1 week(1w), 2 weeks (2w) and 3 weeks (3w). (B)Western blot analysis of Trps1 and phospho-p38 MAPK expression in ATDC5 cells.ATDC5 cells were treated with 0, 10 or100 ng/mL Gdf5 for 3 days. (C) RT-PCRanalysis of Gdf5 mRNA expression by Trps1-overexpressing ATDC5 cells (Trps1) andmock-transfected control cells (control). Cellswere treated with insulin for 3 days (3d), 1week (1w) or 2 weeks (2w), and then wereharvested. (D) Trps1 nuclear translocationassay in ATDC5 cells. Cells were treatedwith 0 (−) or 100 ng/mL Gdf5 and thenwere stained with an anti-Trps1 antibody.

Figure 3 Inhibition of Trps1 expression by dn-Alk6 and SB203580.(A) Western blot analysis (top) and RT-PCR (bottom) of Trps1expression in dn-Alk6 transfected ATDC5 cells (dn-Alk6) andmock-transfected control cells (pcDNA3.1) after 3 days (3d), 1week (1w) or 2 weeks (2w) of treatment with insulin. (B) Westernblot analysis of phospho-p38 MAPK expression in dn-Alk6-transfected (dn-Alk6) and mock-transfected control cells (pcDNA3.1).(C) Western blot analysis of Trps1 expression. ATDC5 cells weretreated with (+) or without (–) SB203580.

S Itoh

et al.

Genes to Cells (2008)

13

, 355–363

© 2008 The AuthorsJournal compilation © 2008 by the Molecular Biology Society of Japan/Blackwell Publishing Ltd.

358

their expression was delayed in dn-Alk6-transfected cellscompared with control cells (Fig. 4B). Treatment withGdf5 also up-regulated the expression of chondrogenicmarker proteins in control ATDC5 cells, but not indn-Alk6-transfected cells (Fig. 4C). Trps1 mRNA levelswere not significantly enhanced by Gdf5 compared withTrps1 protein levels (Figs 2B and 4C). These resultssuggested that chondrogenic differentiation of ATDC5cells was accelerated by Gdf5–Trps1 signaling.

Gdf5–Trps1 signaling promotes apoptosis of ATDC5 cells

In the growth plates, hypertrophic chondrocytes die ofapoptosis during the final phase of cartilage development(Gibson

et al

. 1995). Based on the fact that ATDC5 cellsundergo apoptotic death after chondrogenic differentiation(Mushtaq

et al

. 2002), we examined the expression of pro-teins associated with apoptosis in Trps1-overexpressingand dn-Alk6-transfected ATDC5 cells. Among these pro-teins, expression of Bcl-2 mRNA was significantly decreasedby the over-expression of Trps1, whereas transfection withdn-Alk6 resulted in elevated Bcl-2 expression (Fig. 5A).

Next, we induced the apoptotic death of ATDC5 cellsby addition of an agonistic anti-Fas antibody ( Jo2) andexamined differences in the number of apoptotic cellsbetween Trps1-overexpressing, Alk6-transfected and mock-infected control cells. TUNEL staining showed enhance-ment of apoptosis in Trps1-overexpressing cells (Fig. 5B).In contrast, few apoptotic cells were observed afterdn-Alk6-transfection (Fig. 5B). Furthermore, Westernblot analysis showed that the level of activated caspase3was increased in Trps1-overexpressing cells and decreasedin dn-Alk6-transfected cells (Fig. 5C).

Discussion

Gdf5 is a member of the BMP family of proteins that hasa role in the process of endochondral bone formation(Settle

et al

. 2003). It has been reported that Gdf5 isexpressed in the joints of the digits (Merino

et al

. 1999)and that

GDF5

mutations cause abnormalities of thephalanges (Holder-Espinasse

et al

. 2004; Dawson

et al

. 2006;Wang

et al

. 2006). Based on the patterns of expressionfor Trps1 and Gdf5 in the developing digits, we assumedthat Trps1 could be associated with Gdf5 signaling. In

Figure 4 Chondrogenic differentiation of ATDC5 cells. (A) Alcian Blue staining of Trps1-overexpressing (Trps1), dn-Alk6 transfected(dn-Alk6), mock-transfected control (Control) and 100 ng/mL Gdf5-treated (Gdf5) ATDC5 cells. Cells were incubated for 10 days inmedium containing insulin, fixed with 4% paraformaldehyde, and stained with 1% Alcian Blue. Then the Alcian blue dye was extractedwith guanidine–HCl and measured at 630 nm (right panel). (B) RT-PCR analysis (top) and Western blot analysis (bottom) of theexpression of type II (Col2) and type X (Col10) collagens and Trps1. Total RNA and proteins were extracted from Trps1-overexpressing(Trps1), dn-Alk6-transfected (dn-Alk6) and mock-transfected control ATDC5 cells (control) after 3 days (3d), 1 week (1w) or 2 weeks(2w) of treatment with insulin. (C) RT-PCR analysis of type II (Col2) and type X (Col10) collagen and Trps1 expression. Total RNAwas extracted from dn-Alk6-transfected (dn-Alk6), and mock-transfected control ATDC5 cells (control) after 3 days (3d), 1 week (1w)or 2 weeks (2w) of treatment with insulin and 100 ng/mL Gdf5.

Trps1 plays a pivotal role downstream of Gdf5

© 2008 The Authors

Genes to Cells (2008)

13

, 355–363

Journal compilation © 2008 by the Molecular Biology Society of Japan/Blackwell Publishing Ltd.

359

this study, we demonstrated that Trps1 acts downstreamof the Gdf5 signaling pathway, and that Trps1 and Gdf5mutually regulate their expression in a feedback manner.These findings are in good agreement with the fact thatmutations of both

GDF5

and

TRPS1

cause brachydactyly.However, the other effects of

GDF5

and

TRPS1

muta-tions are quite different. A possible explanation for thiswould be that other downstream signaling pathways existin addition to that of Trps1. For example, Barx2 has beenreported to act downstream of Gdf5 (Meech

et al

. 2005).It would be interesting to further examine the associ-ations between Trps1 and Barx2 in cartilage.

The regulation of Gdf5 expression by Trps1 is poorlyunderstood. At least four GATA motifs have been reportedto exist in the 3.5 kb 5

-flanking region of the humanGDF5 gene (Sugiura

et al

. 1999). In the mouse Gdf5 gene,multiple GATA motifs also exist in 5

-flanking region(Mouse genome informatics: 95688). Given that Trps1 isa GATA-type transcriptional repressor, it is possible thatit binds directly to these GATA motifs in the promoterregion of Gdf5 and thus regulates its expression.

Noggin is an antagonist of BMPs including Gdf5 thatmay influence Gdf5–Trps1 signaling. It has been reportedthat Noggin is expressed during joint formation andcontributes to the regulation of Gdf5 expression (Brunet

et al

. 1998). Although it antagonizes Gdf5 itself, Noggininduces Gdf5 expression by inhibiting Bmp7 duringjoint formation (Merino

et al

. 1999). Like Trps1-deficient

mice or bp mice, Noggin-deficient mice lack Gdf5expression and display fusions between the phalangesand carpal bones (Brunet

et al

. 1998; Suemoto &Muragaki, unpublished data). These findings suggest thatTrps1–Gdf5 signaling is regulated by Noggin.

Although it has been reported that Trps1 interactswith LC8a and RNF4, and localizes in the nucleus(Kaiser

et al

. 2003a,b), we demonstrated in this studythat Trps1 underwent nuclear translocation in ATDC5cells after treatment with Gdf5. Since Gdf5 and Trps1 areco-expressed at virtually the same time in ATDC5 cells,Gdf5 signaling occurs in these cells and Trps1 is alwaysdetected in the nuclei by immunohistochemistry. Toexclude the effect of endogenous Gdf5, we used ATDC5cells after 1 day of insulin treatment for the nuclear trans-location assay. Nuclear translocation of Trps1 was com-pletely blocked by treatment with SB203580 (data notshown). Thus, we assume that the mechanism of nucleartranslocation could involve the phosphorylation of cer-tain amino acid residues of Trps1.

Although Gdf5 was previously shown to activate p38MAPK in ATDC5 cells (Nakamura

et al

. 1999), themolecular events downstream of p38 MAPK were notunderstood. In this study, we demonstrated for the firsttime that Trps1 expression is regulated via Alk6 and p38MAPK in ATDC5 cells. When cells were transfectedwith a dominant negative form of Alk6, phosphorylationof p38 MAPK and Trps1 protein levels were dramatically

Figure 5 Promotion of apoptosis inATDC5 cells by Trps1. (A) Semi-quantitativeRT-PCR for Bcl-2. Total RNA wasextracted from Trps1-overexpressing (Trps1),dn-Alk6-transfected (dn-Alk6), and mock-transfected control (control) ATDC5cells after 1 week of insulin treatment. (B)TUNEL staining of Trps1-overexpressing(Trps1), dn-Alk6-transfected (dn-Alk6),and mock-transfected control ATDC5 cells(control) treated with an anti-Fas antibody.The number of TUNEL positive cells wascounted. DAPI (blue) was used for nuclearstaining. (C) Western blot analysis of cleavedcaspase 3 expression in Trps1-overexpressing(Trps1), dn-Alk6-transfected (dn-Alk6),and mock-transfected control ATDC5 cells(control).

S Itoh

et al.

Genes to Cells (2008)

13

, 355–363

© 2008 The AuthorsJournal compilation © 2008 by the Molecular Biology Society of Japan/Blackwell Publishing Ltd.

360

suppressed. In addition, treatment of ATDC5 cells withSB203580 (an inhibitor of p38 MAPK) led to a substantialreduction of Trps1 protein levels. These results indicatethat Trps1 expression in ATDC5 cells is induced via Alk6and p38 MAPK.

It is interesting to note that Trps1 protein but notmRNA levels are strongly enhanced by Gdf5 treatmentand are strongly affected by dnALK6 over-expression.For some reason, the fluctuation quantity of Trps1mRNA levels was little. This would support the notionthat p38 MAPK stabilizes proteins by protecting fromproteasomal degradation (Zimmermann

et al

. 2001; Brook

et al

. 2006). p38 MAPK, which is up-regulated by Gdf5and is down-regulated by dnAlk6, would be essential formaintaining Trps1 protein levels in ATDC5 cells.

Increased phosphorylation of Smad1/5 and ERK1/2was also observed after treatment with Gdf5 (data notshown). Since addition of SB203580 resulted in a dramaticreduction of Trps1 protein levels, p38 MAPK is consid-ered to be involved in the main pathway for Gdf5–Trps1signaling.

Previous studies have shown that Gdf5 stimulateschondrogenic differentiation in ATDC5 cells (Nakamura

et al

. 1999). In this study, we also found that over-expressionof Trps1 promoted chondrogenesis of ATDC5 cells. Thisis in good agreement with the concept that Trps1 actsdownstream of Gdf5, but whether Gdf5 and Trps1 actsynergistically or antagonistically in chondrogenesis ofATDC5 cells still remains unclear. In contrast, whenATDC5 cells were transfected with dn-Alk6, fewercartilage-like nodules were formed compared withcultures of control cells. This could be explained by thefact that Trps1 expression is reduced by dn-Alk6. Takentogether, it is conceivable that chondrogenic differenti-ation of ATDC5 cells is promoted in proportion to thelevel of Trps1 expression.

Since Trps1 is a transcriptional repressor (Malik

et al

.2001), it is difficult to consider that it regulates theexpression of types II and X collagen. One possibility isthat Trps1 might negatively regulate a transcriptionalrepressor of the type II and type X collagen genes. It hasbeen reported that the other transcription factor down-stream of Gdf5, Barx2, binds to the intronic regulatoryregion of the Col2a1 gene during chondrogenesis (Meech

et al. 2005). We found that Trps1 binds to two GATAbinding sites in the promoter of the Stat3 gene andrepresses the expression of Stat3 (Suemoto et al. 2007). Itis not yet clear how this effect on Stat3 influences chon-drogenesis. Further study is needed to explain how Trps1regulates the expression of cartilage-specific collagens.

We also demonstrated that the other effect of Trps1 onATDC5 cells is the promotion of apoptosis and that

Trps1 suppressed the expression of anti-apoptoticBcl-2. During growth plate development, hypertrophicchondrocytes die by apoptosis and are replaced byosteoblasts. Given that Trps1 is reportedly expressed byprehypertrophic chondrocytes (Kunath et al. 2002), itcould promote apoptosis of hypertrophic chondrocytesby suppressing Bcl-2 expression.

Gdf5 is considered to be involved in joint formation,but the precise molecular mechanisms remain unclear.Previous studies have shown that Gdf5 induces apoptosisof interzone cells during phalangeal joint development(Storm & Kingsley 1996; Merino et al. 1999). In thepresent study, we showed that Trps1 promotes the apop-tosis of ATDC5 cells. In parallel with these results, wealso found that the phalangeal joints of Trps1-deficientnewborn mice were not completely formed becauseinterzone cells did not die of apoptosis (Suemoto &Muragaki, unpublished data). Some of the interzonecells, which are expected to disappear completely due toapoptosis, persisted in the joint spaces of the newbornmice. These findings suggest that apoptosis of interzonecells could be induced by Gdf5–Trps1 signaling.

It was reported that Pthrp inhibits apoptosis in carti-lage (Yamanaka et al. 2003) and in lung tumors (Hastingset al. 2004). Because the pattern of Pthrp expression issimilar to that of Gdf5 (Kobayashi et al. 2005) and becauseTrps1 is a transcriptional repressor, it is conceivable thatGdf5–Trps1 signaling could suppress Pthrp expressionand thus induce apoptosis. In fact, we found that Trps1suppressed the transcription of Pthrp by binding totwo GATA binding sites in the promoter (Nishioka &Muragaki, unpublished data).

In summary, we identified Trps1 as a regulator ofchondrogenesis and apoptosis in ATDC5 cells, which isa downstream target of the Gdf5 signaling pathway.Further studies are needed to identify target(s) of Trps1and the details of Gdf5–Trps1 signaling during chondro-genesis and joint formation.

Experimental proceduresCell culture

ATDC5 cells were obtained from the RIKEN cell bank (Tsukuba,Japan). Cells were maintained in D-MEM/F12 medium (GibcoBRL, Grand Island, NY) supplemented with 5% FBS (MBL,Nagoya, Japan), 10 μg/mL transferrin (Sigma, St Louis, MO) and3 × 10–8 m sodium selenite (Sigma). The culture medium wasrenewed every second day. Cells were plated into plastic dishes or24-well plates at a density of 1 × 104/cm2. After reaching semi-confluence, the cells were treated with insulin (Sigma) at a con-centration of 10 μg/mL and cultured for the indicated periods.

Trps1 plays a pivotal role downstream of Gdf5

© 2008 The Authors Genes to Cells (2008) 13, 355–363Journal compilation © 2008 by the Molecular Biology Society of Japan/Blackwell Publishing Ltd.

361

Construction of expression vectors

DNA fragments for the entire coding region of Trps1 and Alk6were amplified by PCR using the following primer sets: 5′-gagggtgttcctgacgatta-3′ and 5′-tggataaggcaggctctatg-3′ for Trps1 and5′-catgctcttacgaagctctg-3′ and 5′-gtccacaagtatctgacgtc-3′ for Alk6. PCR products were cloned into the pCR II vector (Invitrogen,Carlsbad, CA) and sequenced. It is known that the K231R muta-tion of Alk6 has a dominant negative effect (Fujii et al. 1999).To introduce this mutation into Alk6 cDNA, a QuickChangeSite-Directed Mutagenesis kit (Stratagene, La Jolla, CA) was used.The full-length Trps1 and mutated Alk6 cDNAs were subclonedinto a mammalian expression vector, pcDNA3.1(+) (Invitrogen)and transfected into ATDC5 cells by using a Gene Pulser (BioRad,Hercules, CA).

Production of anti-Trps1 antibodies

cDNA for the carboxy-terminal region of Trps1 was amplified byPCR using a forward primer (5′-ggaagccccttagagag-3′) and a reverseprimer tagged with the PstI site (5′-gtgctaactgcagaggttttactctttagg-3′).The PCR product was cloned into the pQE41 vector (Qiagen,Venlo, the Netherlands) for protein expression. The peptides thusproduced were then purified and used to immunize rabbits. Afterthe fourth dose, antiserum was prepared and purified by using aHiTrap (Amersham, Uppsala, Sweden) conjugated with the antigen.

Immunohistochemistry

Forelimbs removed from E15.5 and E16.5 embryos were fixedovernight in phosphate-buffered saline (PBS) with 4% parafor-maldehyde and then embedded in paraffin. Sections were cut at5 mm. Immunostaining was performed using the dextran polymerconjugate two-step visualization system (Dako Envision System;DAKO, Carpinteria, CA) and the specimens were observed bylight microscopy (Nikon BW40).

In situ hybridization

In situ hybridization was performed using digoxigenin-labeledriboprobes (Roche, Basel, Switzerland). The mouse Gdf5 probewas generated from a 493-bp RT-PCR product (bases 1872–2160of the mouse Gdf5 mRNA, accession no.: NM_008109) and wascloned into pCRII-TOPO (Invitrogen). Hybridization was per-formed overnight at 50 °C, and washed were performed at 42 °C.Sections were then incubated for 1 h with alkaline phosphatase-labeled anti-digoxigenin antibody (1 : 1000, Roche) at RTfollowed by color reactions with BM purple substrate (Roche).Sections were finally counterstained with methyl green andmounted with Permount.

Detection of nuclear translocation

Confluent cells at 1 day after insulin treatment were incubatedwith or without Gdf5 (100 ng/mL) for 1 h. Then the cells were

fixed with cold acetone and reacted with an anti-Trps1 antibodyovernight at 4 °C, followed by reaction with a Cy3-labeled anti-rabbit IgG secondary antibody. Nuclei were stained with DAPI.

Immunoblotting

Cells were washed with PBS and lysed in the sample buffer(62.5 mm Tris–HCl, pH 6.8, 2% SDS, 10% glycerol and 5%β-mercaptoethanol). Then the cell lysate was subjected to SDS-polyacrylamide gel electrophoresis, followed by immunoblottingwith antibodies for β-actin (Santa Cruz, Santa Cruz, CA), cleavedcaspase 3 (Cell Signaling, Danvers, MA) and phosphorylated p38MAPK (Cell Signaling).

Northern blot analysis

Total RNA was extracted from cells using TRIzol (Invitrogen),separated on 1% formaldehyde-agarose gel, transferred to HybondN+ membranes (Amersham) and hybridized with a radiolabeledcDNA probe for nucleotides 2925–3854 of Trps1 cDNA. Thenthe membranes were washed in 0.1× SSC/0.1%SDS, and exposedto an X-ray film.

Reverse transcription-polymerase chain reaction (RT-PCR) and real-time quantitative PCR

Using 1 μg of total RNA, cDNAs were synthesized by the Super-Script III system (Invitrogen). Amplification was performed witha GeneAmp PCR system 9600 (Perkin Elmer Life Science,Boston, MA) using AmpliTaq Gold DNA polymerase (AppliedBiosystems. Foster City, CA). Specific primers were designed fromthe sequences available in GenBank: 5′-ccatgtttgtgatgggtgtg-3′and 5′-taggccatgaggtccaccac-3′ for Gapdh; 5′-cttgccaagacctgaaactc-3′and 5′-caccaaattcctgttcagcc-3′ for collagen α1(II); 5′-acttcctgtcaagctcatcc-3′ and 5′-tcctgcatgtttcctagatg-3′ for collagen α1(X);5′-caaatctcaggcctgagtga-3′ and 5′-gtgaagagctgatatcctgcag-3′ forTrps1; and 5′-aatgccagggccaagggaag-3′ and 5′-gacagcttcagttgggcaac-3′ for Gdf5. The PCR products were separated by electrophoresison 1%–2% agarose gel and the gels were stained with ethidiumbromide. The PCR product of Gapdh cDNA was used to ascertainthat an equivalent amount of cDNA was obtained from each sample.

Quantitative real time-PCR was performed with SYBR GREENPCR Master Mix (Applied Biosystems) and the products weremeasured by using a 7500 Real Time PCR System (AppliedBiosystems). The following primers were designed with PrimerEx-press Software (Applied Biosystems): 5′-tgagcaagagaggccctatc-3′and 5′-aggcccctcctgttattatg-3′ for Gapdh, and 5′-cgccgctgcctttttg-3′and 5′-gagaatgtcaatccgtaggaatcc-3′ for Bcl-2. Each sample was measuredin triplicate, and the level of expression of the target gene was nor-malized for a Gapdh standard curve run in duplicate on the same plate.

Analysis of chondrogenesis

The extent of chondrogenesis was evaluated by staining withAlcian Blue. Cells were washed with PBS, fixed with 4%

S Itoh et al.

Genes to Cells (2008) 13, 355–363 © 2008 The AuthorsJournal compilation © 2008 by the Molecular Biology Society of Japan/Blackwell Publishing Ltd.

362

paraformaldehyde, stained with 1% Alcian Blue 8 GX (Wako PureChemicals, Tokyo, Japan) in 3% acetic acid and rinsed with dis-tilled water. For quantitative analysis, cells stained with Alcian Bluewere extracted with 6 m guanidine–HCl and the optical densitywas measured at 630 nm by a spectrophotometer.

TUNNEL staining

For examination of apoptosis induced by an agonistic anti-Fasantibody, we used the terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling (TUNEL) method with anApopTag Fluorescein In Situ Apoptosis Detection Kit (Chemi-con, Temecula, CA). Cells were cultured in differentiation mediumfor 1 week and then treated with an agonistic anti-Fas antibody(0.1 μg/mL) and cycloheximide (5 μm) for 10 h. Next, the cellswere washed twice with PBS, fixed with 1% paraformaldehydeand stained. Subsequently, the number of TUNEL-positive cellswas counted in randomly chosen fields under a fluorescencemicroscope.

AcknowledgementsThis work was supported in part by a Research Grant on PriorityAreas from Wakayama Medical University (to Y.M.).

ReferencesAsahina, I., Sampath, T.K. & Hauschka, P.V. (1996) Human oste-

ogenic protein-1 induces chondroblastic, osteoblastic, and/oradipocytic differentiation of clonal murine target cells. Exp. CellRes. 222, 38–47.

Brook, M., Tchen, C.R., Santalucia, T., McIlrath, J., Arthur, J.S.,Saklatvala, J. & Clark, A.R. (2006) Posttranslational regulationof tristetraprolin subcellular localization and protein stability byp38 mitogen-activated protein kinase and extracellular signal-regulated kinase pathways. Mol. Cell. Biol. 26, 2408–2418.

Brunet, L.J., McMahon, J.A., McMahon, A.P. & Harland, R.M.(1998) Noggin, cartilage morphogenesis, and joint formationin the mammalian skeleton. Science 280, 1455–1457.

Chrousos, G.P. & Kino, T. (2005) Ikaros transcription factors:flying between stress and inflammation. J. Clin. Invest. 115,844–848.

Dawson, K., Seeman, P., Sebald, E., King, L., Edwards, M.,Williams, J. 3rd, Mundlos, S. & Krakow, D. (2006) GDF5 is asecond locus for multiple-synostosis syndrome. Am. J. Hum.Genet. 78, 708–712.

Ducy, P. & Karsenty, G. (2000) The family of bone morphogeneticproteins. Kidney Int. 57, 2207–2214.

Fujii, M., Takeda, K., Imamura, T., Aoki, H., Sampath, T.K.,Enomoto, S., Kawabata, M., Kato, M., Ichijo, H. & Miyazono,K. (1999) Roles of bone morphogenetic protein type I receptorsand Smad proteins in osteoblast and chondroblast differentiation.Mol. Biol. Cell 10, 3801–3813.

Gibson, G.J., Kohler, W.J. & Schaffler, M.B. (1995) Chondrocyteapoptosis in endochondral ossification of chick sterna. Dev.Dyn. 203, 468–476.

Hastings, R.H., Araiza, F., Burton, D.W., Bedley, M. & Deftos,L.J. (2004) Parathyroid hormone-related protein regulatesapoptosis in lung cancer cells through protein kinase A. Am. J.Physiol. 287, C1616–C1622.

Holder-Espinasse, M., Escande, F., Mayrargue, E., Dieux-Coeslier, A., Fron, D., Doual-Bisser, A., Boute-Benejean, O.,Robert, Y., Porchet, N. & Manouvrier-Hanu, S. (2004) Angelshaped phalangeal dysplasia, hip dysplasia, and positional teethabnormalities are part of the brachydactyly C spectrum associatedwith CDMP-1 mutations. J. Med. Genet. 41, e78.

Kaiser, F.J., Moroy, T., Chang, G.T., Horsthemke, B. & Ludecke,H.J. (2003a) The RING finger protein RNF4, a co-regulatorof transcription, interacts with the TRPS1 transcription factor.J. Biol. Chem. 278, 38780–38785.

Kaiser, F.J., Tavassoli, K., Van den Bemd, G.J., Chang, G.T.,Horsthemke, B., Möröy, T. & Lüdecke, H.J. (2003b) Nuclearinteraction of the dynein light chain LC8a with the TRPS1transcription factor suppresses the transcriptional repressionactivity of TRPS1. Hum. Mol. Genet. 12, 1349–1358.

Kobayashi, T., Soegiarto, D.W., Yang, Y., Lanske, B., Schipani, E.,McMahon, A.P. & Kronenberg, H.M. (2005) Indian hedgehogstimulates periarticular chondrocyte differentiation to regulategrowth plate length independently of PTHrP. J. Clin. Invest.115, 1734–1742.

Kunath, M., Ludecke, H.J. & Vortkamp, A. (2002) Expression ofTrps1 during mouse embryonic development. Gene Expr.Patterns 2, 119–122.

Malik, T.H., Shoichet, S.A., Latham, P., Kroll, T.G., Peters, L.L.& Shivdasani, R.A. (2001) Transcriptional repression anddevelopmental functions of the atypical vertebrate GATAprotein TRPS1. EMBO J. 20, 1715–1725.

Malik, T.H., Von Stechow, D., Bronson, R.T. & Shivdasani, R.A.(2002) Deletion of the GATA domain of TRPS1 causes anabsence of facial hair and provides new insights into the bonedisorder in inherited tricho-rhino-phalangeal syndromes. Mol.Cell. Biol. 22, 8592–8600.

Meech, R., Edelman, D.B., Jones, F.S. & Makarenkova, H.P.(2005) The homeobox transcription factor Barx2 regulateschondrogenesis during limb development. Development 132,2135–2146.

Merino, R., Macias, D., Ganan, Y., Economides, A.N., Wang, X.,Wu, Q., Stahl, N., Sampath, K.T., Varona, P. & Hurle, J.M.(1999) Expression and function of Gdf-5 during digit skele-togenesis in the embryonic chick leg bud. Dev. Biol. 206, 33–45.

Momeni, P., Glockner, G., Schmidt, O., von Holtum, D.,Albrecht, B., Gillessen-Kaesbach, G., Hennekam, R., Meinecke,P., Zabel, B., Rosenthal, A., Horsthemke, B. & Lüdecke, H.J.(2000) Mutations in a new gene, encoding a zinc-fingerprotein, cause tricho-rhino-phalangeal syndrome type I. Nat.Genet. 24, 71–74.

Moriguchi, T., Kuroyanagi, N., Yamaguchi, K., Gotoh, Y.,Irie, K., Kano, T., Shirakabe, K., Muro, Y., Shibuya, H.,Matsumoto, K., Nishida, E. & Hagiwara, M. (1996) A novelkinase cascade mediated by mitogen-activated protein kinasekinase 6 and MKK3. J. Biol. Chem. 271, 13675–13679.

Mushtaq, T., Farquharson, C., Seawright, E. & Ahmed, S.F.(2002) Glucocorticoid effects on chondrogenesis, differentiation

Trps1 plays a pivotal role downstream of Gdf5

© 2008 The Authors Genes to Cells (2008) 13, 355–363Journal compilation © 2008 by the Molecular Biology Society of Japan/Blackwell Publishing Ltd.

363

and apoptosis in the murine ATDC5 chondrocyte cell line.J. Endocrinol. 175, 705–713.

Nakamura, K., Shirai, T., Morishita, S., Uchida, S., Saeki-Miura,K. & Makishima, F. (1999) p38 mitogen-activated proteinkinase functionally contributes to chondrogenesis induced bygrowth/differentiation factor-5 in ATDC5 cells. Exp. Cell Res.250, 351–363.

Settle, S.H., Jr., Rountree, R.B., Sinha, A., Thacker, A., Higgins,K. & Kingsley, D.M. (2003) Multiple joint and skeletal pattern-ing defects caused by single and double mutations in the mouseGdf6 and Gdf5 genes. Dev. Biol. 254, 116–130.

Shukunami, C., Ohta, Y., Sakuda, M. & Hiraki, Y. (1998)Sequential progression of the differentiation program by bonemorphogenetic protein-2 in chondrogenic cell line ATDC5.Exp. Cell Res. 241, 1–11.

Storm, E.E., Huynh, T.V., Copeland, N.G., Jenkins, N.A.,Kingsley, D.M. & Lee, S.J. (1994) Limb alterations in brachy-podism mice due to mutations in a new member of the TGFβ-superfamily. Nature 368, 639–643.

Storm, E.E. & Kingsley, D.M. (1996) Joint patterning defectscaused by single and double mutations in members of the bonemorphogenetic protein (BMP) family. Development 122, 3969–3979.

Suemoto, H., Muragaki, Y., Nishioka, K., Sato, M., Ooshima, A.,

Itoh, S., Hatamura, I., Ozaki, M., Braun, A., Gustafsson, E. &Fässler, R. (2007) Trps1 regulates proliferation and apoptosis ofchondrocytes through Stat3 signaling. Dev. Biol. 312, 572–581.

Sugiura, T., Hotten, G. & Kawai, S. (1999) Minimal promotercomponents of the human growth/differentiation factor-5 gene.Biochem. Biophys. Res. Commun. 263, 707–713.

Wang, X., Xiao, F., Yang, Q., Liang, B., Tang, Z., Jiang, L.,Zhu, Q., Chang, W., Jiang, J., Jiang, C., Ren, X., Liu, J.Y.,Wang, Q.K. & Liu, M. (2006) A novel mutation in GDF5causes autosomal dominant symphalangism in two Chinesefamilies. Am. J. Med. Genet. A 140, 1846–1853.

Yamanaka, Y., Tanaka, H., Koike, M., Nishimura, R. & Seino, Y.(2003) PTHrP rescues ATDC5 cells from apoptosis induced byFGF receptor 3 mutation. J. Bone Miner Res. 18, 1395–1403.

Yoon, B.S., Pogue, R., Ovchinnikov, D.A., Yoshii, I., Mishina, Y.,Behringer, R.R. & Lyons, K.M. (2006) BMPs regulate multipleaspects of growth-plate chondrogenesis through opposingactions on FGF pathways. Development 133, 4667–4678.

Zimmermann, J., Lamerant, N., Grossenbacher, R. & Furst, P.(2001) Proteasome-and p38-dependent regulation of ERK3expression. J. Biol. Chem. 276, 10759–10766.

Received: 25 September 2007 Accepted: 31 December 2007

top related