| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 282, Issue 24, 17665-17675, June 15, 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1(I) Collagen Gene in Tendon Fibroblasts*
12
3
From the
INSERM U652 and Paris-Descartes University, 75006 Paris, France, CNRS UMR 7622 and Pierre and Marie Curie University, 75005 Paris, France, the ¶Department of Molecular Pharmacology, Medical Research Institute, Tokyo Medical and Dental University, 101-0062 Tokyo, Japan, the ||Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, Boston, Massachusetts 02115, and the **Department of Pathology and Laboratory Medicine, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts 02114
Received for publication, October 30, 2006 , and in revised form, March 29, 2007.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1 chains and one 2 chain that are coiled around each other in an uninterrupted triple helix. In the extracellular space, type I collagen molecules self-assemble into highly organized fibrils and then into fibers. The high tensile strength of these fibers is essential to provide mechanical strength to tissues and in particular to bone, tendon, and ligament. This is exemplified by the dramatic consequences of genetic diseases resulting from mutations in one of the genes encoding type I collagen, such as osteogenesis imperfecta or Ehlers-Danlos syndromes type VIIA and VIIB (1). The hallmark of osteogenesis imperfecta is brittle bones, but the disease can also involve other tissues rich in type I collagen, such as ligaments, tendons, fascia, sclerae, and teeth. Ehlers-Danlos syndromes are characterized by skin hyperextensibility, vascular fragility, and increased ligament elasticity responsible for joint hypermobility.
Although type I collagen is widely distributed throughout the body, it is produced by a discrete subset of cells, including fibroblasts, osteoblasts, and odontoblasts. Studies performed using transgenic mice harboring various segments of the mouse, rat, or human type I collagen promoters cloned upstream of a reporter gene have shown that the cell-specific expression of type I collagen genes is based on a modular organization of separate cell-specific cis-acting elements (2–5). In particular, the segment of the mouse pro-1(I) collagen (COL1a1) promoter located downstream of -2.3 kb induces high level expression of reporter genes in osteoblasts of transgenic mice, whereas the segment of COL1a1 located between -3.2 and -2.3 kb drives high level expression of the lacZ reporter gene specifically in tendon, ligament, and fascia fibroblasts (4, 6–8). Analysis of this latter segment has shown that two short cis-acting elements, named TSE14 and TSE2, are necessary for activation of the COL1a1 promoter in tendon fibroblasts (8). TSE2 is located between -2,363 and -2,316 bp. It contains a consensus E-box (CACGTG) that is located at -2,325 bp, and it binds a tendon-specific nuclear protein (8). TSE1 is located between -2,845 and -2,741 bp, and its 3'-half is necessary for expression of the lacZ reporter gene in tendons of transgenic mice (8).
Molecular mechanisms responsible for the differentiation of mesenchymal cells into tendon fibroblasts are still largely unknown (9). Few DNA-binding proteins have a pattern of expression restricted to developing tendons, which suggests that they could regulate tendon cell fate. The Eya1, Eya2, and Six2 genes are expressed in limb tendons, with Eya1 and Six2 being largely restricted to the flexor tendons and Eya2 to the extensor tendons (10, 11). Scleraxis (SCX), which is a member of the basic helix-loop-helix (bHLH) family of transcription factors, is specifically expressed in tendons and ligaments, where it can be detected from early progenitor cells to mature fibroblasts (12–14). As most bHLH transcription factors that have a restricted pattern of expression, SCX heterodimerizes with the E2A gene products and binds E-boxes as SCX/E12 or SCX/E47 heterodimers (12, 15, 16). Recently, it has been shown that overexpression of SCX in tendon fibroblasts up-regulates the expression of the tenomodulin gene (17). This latter gene encodes a transmembrane protein that is expressed selectively in tendons, ligaments, and epimysium of skeletal muscle and can be considered as a phenotypic marker of tendon fibroblasts (18). However, tenomodulin may not be a direct target of SCX. It has also been shown that a biologically active variant of Smad8 promotes the differentiation of C3H10T1/2 mesenchymal cells into tendon fibroblasts in the presence of bone morphogenetic protein 2, indicating that this Smad pathway is involved in the differentiation of mesenchymal cells into tendon fibroblasts (19). In particular, it induced the expression of the SCX, EphA4, and COL1a1 genes. However, the genes that are direct targets of Smad8 have not been identified.
The NFATc family of transcription factors is composed of five members, NFATc1/NFAT2/NFATc, NFATc2/NFAT1/NFATp, NFATc3/NFAT/NFATx, NFATc4/NFAT3, and the atypical member NFAT5. There is a tight regulation of the subcellular localization of these proteins through the action of calcineurin, a calmodulin-dependent serine/threonine phosphatase. Under basal conditions, NFATc proteins are localized in the cytoplasm in a hyperphosphorylated latent form. Sustained increase in intracellular calcium concentration leads to calcineurin activation, and subsequently NFATc dephosphorylation and translocation into the nucleus where they bind to cis-acting elements (20). Although originally described in immune cells, members of the NFATc family are expressed in many cell types outside of the immune system and are involved in processes such as heart valve development, chondrogenesis, skeletal muscle growth, and angiogenesis (21, 22).
In this study, we demonstrate that SCX binds TSE2, preferentially as a SCX/E47 heterodimer, and activates the COL1a1 proximal promoter in transient transfection experiments. We also show that DNA-binding proteins of the NFATc family are present in tendon fibroblasts, bind TSE1, and transactivate the COL1a1 proximal promoter in transient transfection experiments. Conversely, inhibition of the nuclear translocation of these proteins strongly decreases the expression of COL1a1 in tendon fibroblastic cells. Taken together, these data suggest that SCX and NFATc activate the COL1a1 gene in tendon fibroblasts, and are part of a regulatory network involved in differentiation of mesenchymal cells into tendon fibroblasts.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
GGATCC) by site-directed mutagenesis using the Quick-Change site-directed mutagenesis kit (Stratagene) and following the manufacturer's instructions. The mouse cDNA encoding SCX (a gift from E. Olson, Southwestern University, Dallas, TX) was cloned in the pcDNA3.1 expression vector (Invitrogen), which contains an Xpress tag upstream of the multiple cloning site, to generate pcDNA-SCX. The mouse cDNAs encoding E12 and E47 (a gift from C. Murre, University of California at San Diego, La Jolla, CA) were cloned into the pCI expression vector (Promega), generating pCI-E12 and pCI-E47, respectively. The expression vectors pREP-NFATc1, pREP-NFATc3, and pREP-NFATc4 were given by C. Wilson (University of Washington, Seattle). pGL3 control vector (Promega), containing the SV40 promoter/enhancer cloned upstream of the luciferase reporter gene, was used as efficiency control for transfection experiments.
Cell Cultures—NIH/3T3 fibroblasts and COS-7 cells were obtained from ATCC-LGC Promochem and cultured in Dulbecco's modified Eagle's medium (Invitrogen) supplemented with 10% fetal bovine serum, at 37 °C in a 5% CO2 atmosphere. C2 myoblastic cells, a gift of H. Alameddine (INSERM U582, Paris), were grown in Dulbecco's modified Eagle's medium supplemented with glutamax-I, 4.5 g/liter glucose, and 20% fetal bovine serum, at 37 °C in a 5% CO2 atmosphere. TT-D6 tendon fibroblasts were cultured in -minimum Eagle's medium supplemented with 1% penicillin/streptomycin and 0.5% fetal bovine serum, at 33 °C in a 5% CO2 atmosphere (23).
Gel Shift Assays—For gel shift assays, we used nuclear extracts from TT-D6 cells or from COS-7 cells transfected with pcDNA, pCI, pcDNA-SCX, pCI-E47 or pcDNA-SCX, and pCI-E47. The nuclear extracts were prepared as described previously but were not dialyzed (24). The probes were end-labeled by filling in with [-32P]dCTP using the Klenow fragment of the Escherichia coli DNA polymerase I (Invitrogen). The following double-stranded synthetic oligonucleotides were used as probes or competitors: TSE1 (5'-GGTTGGGAAAATTTGGAGAGAGACAGAACTCAGAGCT-3'); TSE2 (5'-GGGAGTAGAGAGTGGCAGAGGAGGTCCTGAGCTGCAGCTCCATGCCACGTGTAAAGG-3'); 
TSE1 (5'-GGTTGATTTGGAGAGAGACAGAACTCAGAGCT-3'); AgE (5'-GATCTCGGCTAGCATCTGACGCAGGTGCCGAGGGA-3'); IL-4wt (5'-GTACATTGGAAAATTTTATTACAC-3'); and IL-4mut (5'-TACATTCCTTAATTTTATTACAC-3').
To study the binding of SCX to TSE2, the TSE2 probe was incubated with 5 µg of nuclear extracts from transfected COS-7 cells for 30 min at room temperature in a reaction mixture containing 50 mM Tris-HCl, 50 mM NaCl, 10 µM ZnSO4, 1 mM DTT, in the presence of 1 µg of poly(dI-dC)·poly(dI-dC) (Sigma). The complexes were fractionated on 4% polyacrylamide gels in 1x TBE.
Because SCX binds the AgE sequence of the aggrecan promoter, we used it as a competitor (25). For competition experiments, a 200-fold molar excess of unlabeled double-stranded oligonucleotides were added to the reaction mixture. In some cases, nuclear extracts from COS-7 cells transfected with pcDNA-SCX and pCI-E47 were preincubated with 2 µg of anti-Xpress antibody (Invitrogen) for 30 min at room temperature, before addition of labeled probe.
To analyze the binding of NFATc proteins to TSE1, the TSE1 probe was incubated with 5 µg of nuclear extracts from TT-D6 cells, for 30 min at room temperature in a reaction mixture containing 50 mM KCl, 25 mM HEPES, 0.05 mM EDTA, 5% glycerol, 1 mM DTT, 100 ng of bovine serum albumin and in the presence of 1 µg of poly(dI-dC)·poly(dI-dC) (Sigma). Because NFATc proteins bind the IL-4wt sequence (26), we used it as a competitor. The same sequence but with a mutated GGAAA motif (IL-4mut) was used as control. For competition experiments, different amounts of unlabeled double-stranded oligonucleotides were added to the reaction mixture. The DNA-protein complexes were resolved by electrophoresis on 6% polyacrylamide gels in 1x TBE.
Transfection Experiments—Transfections were carried out using Lipofectamine Plus reagent (Invitrogen) and following the manufacturer's instructions.
For stable transfection experiments, NIH/3T3 fibroblasts were cotransfected in triplicate with linearized pJ320 plasmid and linearized plasmid expressing the neomycin resistance gene (pSVneo, Promega). Starting 48 h later, pools of resistant clones (NIH/3T3-pJ320 clones) were selected by treatment with G418 (Invitrogen).
For transient transfection experiments, cells were cotransfected with various amounts of expression vectors, 100 ng of pGL3 control vector, and 900 ng of a lacZ reporter construct, except for the NIH/3T3-pJ320 cells that already harbored a reporter construct. Forty eight hours after transfection, cells were lysed in 0.1 M potassium phosphate buffer, pH 7.8, 1 mM DTT, and the 
-galactosidase and luciferase activities were assessed using the -galactosidase reporter gene assay (Roche Diagnostics) and the luciferase assay system (Promega), respectively. For cells transiently transfected with a lacZ reporter construct, -galactosidase activities were normalized to luciferase activities to compensate for variations in transfection efficiency. For the stably transfected NIH/3T3-pJ320 cells, -galactosidase activities were normalized to the amount of protein. All transient transfection experiments were done in triplicate and repeated at least three times.
RT-PCR Assays—To prepare total RNA, cells were lysed in TRIzol (Invitrogen), whereas tissue samples were homogenized in the presence of TRIzol. Total RNA was then isolated following the manufacturer's instructions and treated with DNase I (Invitrogen). It was reverse-transcribed using random primers (Invitrogen) and SuperScript II reverse transcriptase (Invitrogen). As a control, an RT reaction was always performed after omitting the reverse transcriptase. All RT reactions were tested for DNA contamination.
PCRs specific for each NFATc were carried out for 30 cycles, using the primers listed in Table 1. To monitor the efficiency of the RT-PCRs, PCR with primers corresponding to the constitutively expressed -actin gene (Table 1) were systematically performed.
|
-actin primers, and 0.4 µM of COL1a1 or COL1a2 primers (Table 1). Amplification of -actin was used for normalization. Normalized data are expressed as a percentage of COL1a1 or COL1a2 mRNA expression in untreated cells. In some experiments, 70–80% confluent TT-D6 or NIH/3T3 cells were treated with different amounts of either cyclosporin A (CsA, Sandimmun®, Novartis) or INCA-6 (27) before harvesting RNA. When INCA-6 was used, the medium containing INCA-6 was replaced by fresh cell culture medium after 1 h. Total RNA was extracted after 72 h when cells were treated with CsA or after 48 h when they were treated with INCA-6 and used for real time RT-PCR assays. All experiments were done in triplicate and repeated at least three times.
Western Blot Experiments—For Western blot analyses, we used nuclear extracts from TT-D6 cells and NIH/3T3 fibroblasts, which were prepared as described above, and total cell lysates from COS-7 cells transfected with an expression vector encoding NFATc1, NFATc3, or NFATc4 as positive controls. Briefly, COS-7 cells were transfected with pREP-NFATc1, pREP-NFATc3, or pREP-NFATc4 as described above. After 48 h, the cells were washed with cold PBS and lysed in buffer containing 50 mM Tris-Cl, pH 8.0, 150 mM NaCl, 0.5% sodium deoxycholate, 1% Nonidet P-40, 0.1% SDS, supplemented with protease inhibitors. Then 25 µg of nuclear proteins or 10 µg of cell lysates were fractionated on a 7.5% SDS-polyacrylamide gel, electrotransferred onto a nitrocellulose membrane (HybondTM-ECLTM, Amersham Biosciences), incubated with a monoclonal antibody directed against the mouse NFATc1 (sc-7294, Santa Cruz Biotechnology), NFATc3 (sc-8405, Santa Cruz Biotechnology), or NFATc4 (sc-13036, Santa Cruz Biotechnology) protein and then with a peroxidase-labeled secondary antibody. Specific immune complexes were visualized by chemiluminescence using the Western lightning chemiluminescence reagent plus (PerkinElmer Life Sciences).
In Situ Hybridization Experiments—E15.5 mouse embryos were fixed in 4% paraformaldehyde and processed for in situ hybridization to wax sections as described previously (28, 29). The SCX probe was used as described previously (29). The probes for NFATc1, NFATc3, and NFATc4 originate from the UNIGENE data base. Probes were transcribed in the presence of digoxigenin coupled to uridine. The digoxigenin-labeled probes were revealed by immunohistochemistry using nitro blue tetrazolium chloride/5-bromo-4-chloro-3-indodyl phosphate (28). Differentiated muscle cells were detected after in situ hybridization using the MF20 monoclonal antibody against sarcomeric myosin heavy chain (Developmental Hybridoma Bank), a horseradish peroxidase-conjugated secondary antibody, and diaminobenzidine as substrate (28).
Statistical Analyses—Results are presented as mean ± S.D. Statistical analyses were performed by one-way ANOVA followed by post hoc tests, or by unpaired Student's t test, as appropriate using Statview 5.0 software (SAS, Cary, NC). p values <0.05 were considered as significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
Transfection experiments were then repeated using pJ320 instead of p4TSE2mp. pJ320 contains a 3.2-kb segment of the mouse COL1a1 proximal promoter cloned upstream of the lacZ reporter gene, and it is expressed at high levels in tendons and ligaments of transgenic mice (4, 8). Overexpression of SCX plus E47 increased the transcriptional activity of pJ320 up to 10-fold, and this transactivating effect was dose-dependent (Fig. 2B). By contrast, transfection of pCI-E47 alone had less transactivating effects on pJ320, and transfection of pcDNA-SCX alone had no transactivating effects (Fig. 2B). Overexpression of SCX plus E12 also had almost no transactivating effects on the activity of pJ320 (data not shown). Interestingly, in C2 myoblasts, a mesenchymal cell line that does not express type I collagen, cotransfection of pcDNA-SCX plus pCI-E47 very slightly enhanced the expression of pJ320 (Fig. 2C). This suggests that SCX/E47 heterodimers need to cooperate with other nuclear proteins that are present in NIH/3T3 cells but not in C2 cells.
Because the 3.2-kb segment of the COL1a1 proximal promoter contains nine consensus E-boxes (CANNTG), transfection experiments were repeated using a reporter construct in which the E-box of TSE2 had been mutated (pJ320-MTSE2). The mutation of this E-box decreased by about 30% the transcriptional activation induced by transfection of 200 ng of pcDNA-SCX plus 200 ng of pCI-E47 (Fig. 2D). Taken together, these results indicate that SCX can transactivate reporter constructs containing TSE2, and these transactivating effects appear to be mostly mediated by the binding of SCX/E47 heterodimers to the E-box of TSE2.
SCX/E47 Heterodimers Need to Cooperate with Other Nuclear Factors to Transactivate the COL1a1 Proximal Promoter—Transgenic experiments have shown that TSE2 needs to cooperate with another tendon-specific cis-acting element, named TSE1, to drive reporter gene expression in tendon fibroblasts (8). This led us to test the effects of overexpression of SCX and E47 on pJ320, in stable transfection experiments. pcDNA-SCX and pCI-E47 expression vectors were cotransfected in pools of NIH/3T3 cells that had previously been stably transfected with pJ320. Unlike what had been observed in transient transfection experiments, overexpression of SCX plus E47 did not increase the activity of the lacZ reporter gene (Fig. 2E). Thus, as shown in transgenic experiments, SCX needs to cooperate with other DNA-binding elements to exert its transactivating effects on the COL1a1 promoter, when it is stably integrated into the genome.
NFATc Transcription Factors Are Expressed in Tendon Fibroblasts—Bioinformatic analysis of the 3'-half of TSE1 revealed the presence of few potential binding sites for transcription factors but a motif at -2,776 bp, which corresponds to the core sequence of a potential NFATc-binding site (GGAAA) (30); this motif is conserved between species (data not shown). Interestingly, NFATc DNA-binding proteins have been shown to play a role in the development of heart valves, which share common features with tendons such as the expression of tenascin (31–33). They have also recently been shown to cooperate with Osterix to enhance the expression of COL1a1 in osteoblasts (34). We thus decided to study the expression of the NFATc genes in tendon fibroblasts and the ability of the corresponding proteins to bind TSE1 and transactivate the COL1a1 promoter.
We first performed RT-PCR experiments to examine the presence of mRNAs encoding the different NFATc proteins in TT-D6 cells. This cell line has been derived from Achilles tendons of transgenic mice harboring a temperature-sensitive mutant of SV40 large T antigen, and it has characteristics of tendon fibroblasts (23). Spleen mRNA was used as positive control, because NFATc1–3 mRNAs, and to a lesser extent NFATc4 mRNA, are present in this organ. As shown in Fig. 3A, mRNAs corresponding to NFATc1, NFATc3, and NFATc4, but not NFATc2, were detected in TT-D6 cells. Western blot experiments performed using total cell lysates from TT-D6 cells showed that the NFATc1, NFATc3, and NFATc4 proteins were expressed in these cells but at low levels (data not shown). The activity of the NFATc proteins is mainly regulated by their subcellular localization, and only the hypophosphorylated NFATc proteins present in the nucleus can bind DNA and activate transcription. To determine whether the NFATc proteins present in TT-D6 cells were present in the nucleus, we performed Western blot experiments using nuclear extracts from TT-D6 cells. Cellular extracts from COS-7 cells transfected with expression vectors encoding NFATc1, NFATc3, or NFATc4 were used as positive controls. NFATc4 was detected in nuclear extracts from TT-D6 cells, whereas NFATc1 and NFATc3 proteins could not be detected even after prolonged exposure (Fig. 3B and data not shown). These data suggest that only the NFATc4 protein is activated and translocated into the nucleus in TT-D6 fibroblasts. Interestingly, C2 cells did not express NFATc4 mRNA (data not shown).
|
|
NFATc Proteins Bind TSE1 and Transactivate the COL1a1 Proximal Promoter—To test the ability of NFATc proteins present in TT-D6 cells to bind TSE1, we performed gel shift assays. Incubation of nuclear extracts from TT-D6 cells with an end-labeled TSE1 probe induced the appearance of a shifted complex (Fig. 4, lane 1). This complex was competed by an excess of unlabeled TSE1 probe (Fig. 4, lane 2) but not by a mutated probe deleted of the GGAAA motif (Fig. 4, lane 3). Similarly, it was competed by a double-stranded oligonucleotide corresponding to a well described NFATc-binding site in the mouse IL-4 promoter but not by an oligonucleotide in which this binding site had been mutated (Fig. 4, lanes 4–9). Taken together, these results indicate that TSE1 binds NFATc proteins that are present in tendon fibroblasts.
|
NFATc Proteins Regulate the Expression of Type I Collagen Genes in Tendon Fibroblasts—To further explore the ability of NFATc proteins to regulate the transcription of COL1a1, we used the calcineurin inhibitor CsA. By blocking the phosphatase activity of calcineurin, CsA inhibits the dephosphorylation of NFATc and thus its nuclear translocation (20). The expression of COL1a1 mRNA in TT-D6 cells treated with various amounts of CsA was analyzed by real time RT-PCR. As shown in Fig. 6A, treatment with CsA induced a dramatic and dose-dependent decrease in the levels of COL1a1 mRNA; this reduction reached about 90% when TT-D6 cells were incubated with 20 µg/ml (i.e. 16.6 nmol/ml) of CsA (Fig. 6A). Because there is a coordinated regulation of COL1a1 and COL1a2, we analyzed the ability of NFATc proteins to regulate the expression of COL1a2. Treatment of TT-D6 cells with CsA induced a strong and dose-dependent inhibition of the expression of the pro-2(I) collagen gene (data not shown).
|
2(I) collagen gene (data not shown). Thus, NFATc transcription factors appear to be potent regulators of the expression of COL1a1 and COL1a2 in tendon fibroblastic cells. NFATc Proteins Regulate Type I Collagen Gene Expression in Cells Different from Tendon Fibroblasts—Because NFATc proteins are expressed in a large number of cell types, we decided to analyze the expression of the NFATc genes in a fibroblastic cell line different from TT-D6 cells and the effects of treatment with CsA and INCA-6 on the expression of COL1a1 mRNA in this cell line. We chose to use NIH/3T3 cells, which is a well known fibroblastic cell line (35). RT-PCR experiments showed the presence of NFATc1, NFATc3, and NFATc4 mRNAs in NIH/3T3 fibroblasts (Fig. 7A). However, Western blot analyses only demonstrated the presence of NFATc3 and NFATc4 proteins in total cell lysates and nuclear extracts from NIH/3T3 cells (data not shown and Fig. 7B). Real time RT-PCR experiments showed that treatment of NIH/3T3 cells with CsA or INCA-6 induced an important and dose-dependent inhibition of the expression of the COL1a1 and COL1a2 mRNA (Fig. 7, C and D, and data not shown). Thus, NFATc proteins are expressed in fibroblasts different from tendon fibroblasts, and they also appear to regulate the expression of type I collagen genes in these cells.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Because tendon fibroblasts appear to contain NFATc4, we studied the ability of NFATc proteins to bind TSE1 and transactivate the COL1a1 gene. Three arguments suggest that NFATc proteins regulate the activity of COL1a1 in tendon fibroblasts through their binding to TSE1. First, gel shift assays showed that TSE1 can bind NFATc proteins that are present in nuclear extracts from TT-D6 tendon fibroblastic cells. Incubation of a TSE1 probe with nuclear extracts from TT-D6 cells induced the appearance of a single retarded complex that was competed by a double-stranded oligonucleotide corresponding to a consensus NFATc-binding site but not by a mutated one. This complex was also not competed by a TSE1 double-stranded oligonucleotide in which the consensus binding site for NFATc had been deleted. Second, in transient transfection experiments, overexpression of NFATc had transactivating effects on a reporter construct containing four copies of TSE1 cloned upstream of the COL1a1 minimal promoter. Third, inhibition of the nuclear translocation of NFATc proteins by two unrelated pharmacologic molecules, CsA and INCA-6, induced a dramatic and dose-dependent inhibition of the expression of COL1a1 in TT-D6 tendon fibroblastic cells. CsA is a pharmacologic agent that has been widely used to inhibit the nuclear translocation of NFATc transcription factors and thus their transactivating effects. However, it inhibits the enzymatic activity of calcineurin toward all its physiological substrates and can thus block all signaling downstream of calcineurin (38). By contrast, INCA-6 is a newly developed small organic molecule that specifically blocks targeting of calcineurin to NFATc by inhibiting the interaction between these two proteins and that does not prevent the dephosphorylation of other substrates (27). Interestingly, Koga et al. (34) have recently demonstrated that NFATc proteins regulate the activity of the COL1a1 gene in osteoblasts. In particular, they have shown the following: 1) that overexpression of NFAT proteins slightly but significantly enhances the activity of the COL1a1 proximal promoter in osteoblastic cells, and 2) that treatment of osteoblasts with FK506, a pharmacologic agent that is closely related to CsA and also inhibits the dephosphorylation of NFATc proteins, strongly reduces the mRNA expression of COL1a1 (34). Thus, NFATc could have similar transactivating effects on COL1a1 in osteoblasts and tendon fibroblasts. In addition, Koga et al. (34) have shown that NFATc1 interacts with Osterix, a transcription factor that plays a key role in bone development and osteoblast differentiation (39), to activate the COL1a1 promoter. Hence, NFATc proteins appear to cooperate with Osterix to activate the COL1a1 gene in osteoblasts and with SCX to activate this gene in tendon fibroblasts. Analyses of knock-out mice have also shown that NFATc1 is indispensable for early development of heart valves (31, 32, 40), but its role in the differentiation of heart valve fibroblasts remains unknown. An attractive hypothesis would be that NFATc1 also cooperates with another transcription factor, such as Sox9 (41), to induce the differentiation of heart valve fibroblasts and expression of COL1a1. Because our data suggest that NFATc proteins are also involved in activation of the COL1a1 gene in NIH/3T3 fibroblastic cells, they could even have a more general role in activation of the collagen genes in fibroblastic cells.
In conclusion, our data show that NFATc proteins and SCX activate the COL1a1 gene in tendon fibroblasts through their binding to TSE1 and TSE2, respectively. These two transcription factors are probably part of a larger transcriptional network that induces the differentiation of mesenchymal cells into tendon fibroblasts.
|
FOOTNOTES |
|---|
1 Recipient of a fellowship from the French Ministry of Research and from the Fondation pour la Recherche Médicale.
2 Present address: Dept. of Medical Biochemistry and Molecular Biology, University of Oulu, 90014 Oulu, Finland.
3 To whom correspondence should be addressed: Cordeliers Biomedical Institute, INSERM U652, 15 Rue de l'École de Médecine, 75006 Paris, France. Tel.: 33 1 44 41 37 10/19; Fax: 33 1 44 41 37 17; E-mail: jerome.rossert{at}egp.aphp.fr.
4 The abbreviations used are: TSE1, tendon-specific element 1; SCX, scleraxis; AgE, E-box-containing sequence of the aggrecan promoter; bHLH, basic helix-loop-helix; CsA, cyclosporin A; LacZ, E. coli -galactosidase; TSE2, tendon-specific element 2; RT, reverse transcription; ANOVA, analysis of variance; DTT, dithiothreitol; E15.5, embryonic day 15.5.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
