Title: Dlx5 Specifically Regulates Runx2 Type II Expression by Binding to Homeodomain-response Elements in the Runx2 Distal Promoter
Abstract: Two major isoforms of the Runx2 gene are expressed by alternative promoter usage: Runx2 type I (Runx2-I) is derived from the proximal promoter (P2), and Runx2 type II (Runx2-II) is produced by the distal promoter (P1). Our previous results indicate that Dlx5 mediates BMP-2-induced Runx2 expression and osteoblast differentiation (Lee, M.-H., Kim, Y-J., Kim, H-J., Park, H-D., Kang, A-R., Kyung, H.-M., Sung, J-H., Wozney, J. M., Kim, H-J., and Ryoo, H-M. (2003) J. Biol. Chem. 278, 34387-34394). However, little is known of the molecular mechanisms by which Dlx5 up-regulates Runx2 expression in BMP-2 signaling. Here, Runx2-II expression was found to be specifically stimulated by BMP-2 treatment or by Dlx5 overexpression. In addition, BMP-2, Dlx5, and Runx2-II were found to be expressed in osteogenic fronts and parietal bones of the developing cranial vault and Runx2-I and Msx2 in the sutural mesenchyme. Furthermore, Runx2 P1 promoter activity was strongly stimulated by Dlx5 overexpression, whereas Runx2 P2 promoter activity was not. Runx2 P1 promoter deletion analysis indicated that the Dlx5-specific response is due to sequences between -756 and -342 bp of the P1 promoter, where three Dlx5-response elements are located. Dlx5 responsiveness to these elements was confirmed by gel mobility shift assay and site-directed mutagenesis. Moreover, Msx2 specifically suppressed the Runx2 P1 promoter, and the responsible region overlaps with that recognized by Dlx5. In summary, Dlx5 specifically transactivates the Runx2 P1 promoter, and its action on the P1 promoter is antagonized by Msx2. Two major isoforms of the Runx2 gene are expressed by alternative promoter usage: Runx2 type I (Runx2-I) is derived from the proximal promoter (P2), and Runx2 type II (Runx2-II) is produced by the distal promoter (P1). Our previous results indicate that Dlx5 mediates BMP-2-induced Runx2 expression and osteoblast differentiation (Lee, M.-H., Kim, Y-J., Kim, H-J., Park, H-D., Kang, A-R., Kyung, H.-M., Sung, J-H., Wozney, J. M., Kim, H-J., and Ryoo, H-M. (2003) J. Biol. Chem. 278, 34387-34394). However, little is known of the molecular mechanisms by which Dlx5 up-regulates Runx2 expression in BMP-2 signaling. Here, Runx2-II expression was found to be specifically stimulated by BMP-2 treatment or by Dlx5 overexpression. In addition, BMP-2, Dlx5, and Runx2-II were found to be expressed in osteogenic fronts and parietal bones of the developing cranial vault and Runx2-I and Msx2 in the sutural mesenchyme. Furthermore, Runx2 P1 promoter activity was strongly stimulated by Dlx5 overexpression, whereas Runx2 P2 promoter activity was not. Runx2 P1 promoter deletion analysis indicated that the Dlx5-specific response is due to sequences between -756 and -342 bp of the P1 promoter, where three Dlx5-response elements are located. Dlx5 responsiveness to these elements was confirmed by gel mobility shift assay and site-directed mutagenesis. Moreover, Msx2 specifically suppressed the Runx2 P1 promoter, and the responsible region overlaps with that recognized by Dlx5. In summary, Dlx5 specifically transactivates the Runx2 P1 promoter, and its action on the P1 promoter is antagonized by Msx2. The Runt-related transcription factor Runx2 plays an essential role in osteoblast differentiation and bone mineralization (1Ducy P. Zhang R. Geoffroy V. Ridall A.L. Karsenty G. Cell. 1997; 89: 747-754Abstract Full Text Full Text PDF PubMed Scopus (3578) Google Scholar, 2Komori T. Yagi H. Nomura S. Yamaguchi A. Sasaki K. Deguchi K. Shimizu Y. Bronson R.T. Gao Y.H. Inada M. Sato M. Okamoto R. Kitamura Y. Yoshiki S. Kishimoto T. Cell. 1997; 89: 755-764Abstract Full Text Full Text PDF PubMed Scopus (3569) Google Scholar). Two major isoforms are expressed from the mouse Runx2 locus, and these isoforms are generated by different promoter usage. Runx2 type I (Runx2-I), 2The abbreviations used are: Runx2-IRunx2 type IRunx2-IIRunx2 type IITGF-β1transforming growth factor-β1BMPsbone morphogenetic proteinsASantisenseHDREshomeodomain-response elementsEMSAelectrophoretic mobility shift assay. 2The abbreviations used are: Runx2-IRunx2 type IRunx2-IIRunx2 type IITGF-β1transforming growth factor-β1BMPsbone morphogenetic proteinsASantisenseHDREshomeodomain-response elementsEMSAelectrophoretic mobility shift assay. referred to as the Cbfa1/p56 isoform or PEBP2αA, is a 513-amino acid protein that starts with the amino acid sequence MRIPV (3Ogawa E. Maruyama M. Kagoshima H. Inuzuka M. Lu J. Satake M. Shigesada K. Ito Y. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 6859-6863Crossref PubMed Scopus (560) Google Scholar) and is derived from the proximal P2 promoter of the gene (4Xiao Z.S. Liu S.G. Hinson T.K. Quarles L.D. J. Cell. Biochem. 2001; 82: 647-659Crossref PubMed Scopus (72) Google Scholar). More recently, upstream exons of the Runx2 gene that potentially encode the N termini of Runx2 isoforms expressed in osteoblasts have been identified (5Geoffroy V. Corral D.A. Zhou L. Lee B. Karsenty G. Mamm. Genome. 1998; 9: 54-57Crossref PubMed Scopus (84) Google Scholar, 6Thirunavukkarasu K. Mahajan M. McLarren K.W. Stifani S. Karsenty G. Mol. Cell. Biol. 1998; 18: 4197-4208Crossref PubMed Google Scholar). These upstream exons contain a 5′-untranslated region and encode the N-terminal 19 amino acids of Runx2 type II (Runx2-II; also referred to as Cbfa1/p57 and OSF2), which starts with the sequence MASNSL (7Stewart M. Terry A. Hu M. O'Hara M. Blyth K. Baxter E. Cameron E. Onions D.E. Neil J.C. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 8646-8651Crossref PubMed Scopus (194) Google Scholar). This isoform is expressed from the P1 or “bone-related” upstream promoter (8Drissi H. Luc Q. Shakoori R. Chuva De Sousa Lopes S. Choi J.-Y. Terry A. Hu M. Jones S. Neil J.C. Lian J.B. Stein J.L. Van Wijnen A.J. Stein G.S. J. Cell. Physiol. 2000; 184: 341-350Crossref PubMed Scopus (234) Google Scholar), and its expression is predominant in osteoblasts (9Banerjee C. Javed A. Choi J.-Y. Green J. Rosen V. Van Wijnen A.J. Stein J.L. Lian J.B. Stein G.S. Endocrinology. 2001; 142: 4026-4039Crossref PubMed Scopus (161) Google Scholar). The alternative promoter usage strongly implies that the expression pattern of each isoform differs temporally and/or spatially. Indeed, they exhibit distinct expression patterns during bone development (10Park M.H. Shin H.I. Choi J.-Y. Nam S.H. Kim Y.-J. Kim H.-J. Ryoo H.-M. J. Bone Miner. Res. 2001; 16: 885-892Crossref PubMed Scopus (72) Google Scholar, 11Choi K.-Y. Lee S.W. Park M.H. Bae Y.C. Shin H.I. Nam S. Kim Y.-J. Kim H.-J. Ryoo H.-M. Exp. Mol. Med. 2002; 34: 426-433Crossref PubMed Scopus (46) Google Scholar). Thus, it is natural to assume that these two promoters differently respond to different extracellular signals or their downstream transcription factors because these promoters have distinct transcription factor-binding sites. Runx2 type I Runx2 type II transforming growth factor-β1 bone morphogenetic proteins antisense homeodomain-response elements electrophoretic mobility shift assay. Runx2 type I Runx2 type II transforming growth factor-β1 bone morphogenetic proteins antisense homeodomain-response elements electrophoretic mobility shift assay. Runx2 plays a central role in the BMP-2-induced trans-differentiation of C2C12 cells at an early restriction point by diverting them from the myogenic pathway to the osteogenic pathway (12Lee M.-H. Javed A. Kim H.-J. Shin H.I. Gutierrez S. Choi J.-Y. Rosen V. Stein J.L. Van Wijnen A.J. Stein G.S. Lian J.B. Ryoo H.-M. J. Cell. Biochem. 1999; 73: 114-125Crossref PubMed Scopus (238) Google Scholar, 13Lee K.S. Kim H.-J. Li Q.L. Chi X.-Z. Ueta C. Komori T. Wozney J.M. Kim E.G. Choi J.-Y. Ryoo H.-M. Bae S.-C. Mol. Cell. Biol. 2000; 20: 8783-8792Crossref PubMed Scopus (748) Google Scholar). We found that the homeobox gene Dlx5 is an upstream target of BMP-2 signaling and that it plays a pivotal role in stimulating the downstream osteogenic master transcription factor Runx2. In turn, Runx2 acts simultaneously or sequentially to induce the expression of bone-specific genes that represent BMP-2-induced osteogenic trans-differentiation. In addition, it has also been suggested that Dlx5 is a critical target of the inhibitory action of transforming growth factor-β1 (TGF-β1) on BMP-2-induced osteoblast trans-differentiation (14Lee M.-H. Kim Y.-J. Kim H.-J. Park H.-D. Kang A.-R. Kyung H.-M. Sung J.-H. Wozney J.M. Kim H.-J. Ryoo H.-M. J. Biol. Chem. 2003; 278: 34387-34394Abstract Full Text Full Text PDF PubMed Scopus (349) Google Scholar). Several lines of in vivo evidence indicate that the Dlx and Msx families of homeodomain proteins include regulatory factors that preferentially support skeletal tissue differentiation. Among them, Dlx5 is a bone-inducing transcription factor that is expressed in the later stages of osteoblast differentiation (15Ryoo H.-M. Hoffmann H.M. Beumer T. Frenkel B. Towler D.A. Stein G.S. Stein J.L. Van Wijnen A.J. Lian J.B. Mol. Endocrinol. 1997; 11: 1681-1694Crossref PubMed Scopus (217) Google Scholar). Forced expression of Dlx5 in cultured cells leads to osteocalcin expression and a fully mineralized matrix (14Lee M.-H. Kim Y.-J. Kim H.-J. Park H.-D. Kang A.-R. Kyung H.-M. Sung J.-H. Wozney J.M. Kim H.-J. Ryoo H.-M. J. Biol. Chem. 2003; 278: 34387-34394Abstract Full Text Full Text PDF PubMed Scopus (349) Google Scholar, 16Miyama K. Yamada G. Yamamoto T.S. Takagi C. Miyado K. Sakai M. Ueno N. Shibuya H. Dev. Biol. 1999; 208: 123-133Crossref PubMed Scopus (172) Google Scholar, 17Tadic T. Dodig M. Erceg I. Marijanovic I. Mina M. Kalajzic Z. Velonis D. Kronenberg M.S. Kosher R.A. Ferrari D. Lichtler A.C. J. Bone Miner. Res. 2002; 17: 1008-1014Crossref PubMed Scopus (67) Google Scholar). These results strongly suggest that Dlx5 plays important roles in the development of mineralized tissues. Another homeodomain protein that plays critical roles in bone formation and osteoblast differentiation (Msx2) is also induced by bone morphogenetic proteins (BMPs) (18Harada S. Rodan G.A. Nature. 2003; 423: 349-355Crossref PubMed Scopus (1095) Google Scholar). Several in vitro studies have shown that Msx2 negatively regulates the transcription of osteoblast-specific genes such as osteocalcin (15Ryoo H.-M. Hoffmann H.M. Beumer T. Frenkel B. Towler D.A. Stein G.S. Stein J.L. Van Wijnen A.J. Lian J.B. Mol. Endocrinol. 1997; 11: 1681-1694Crossref PubMed Scopus (217) Google Scholar, 19Towler D.A. Rutledge S.J. Rodan G.A. Mol. Endocrinol. 1994; 8: 1484-1493Crossref PubMed Scopus (122) Google Scholar). This was further supported by findings that the Dlx5 and Msx2 proteins bind the promoters of osteoblast markers and appear to have opposing transcriptional properties because Msx2 functions as a transcriptional repressor (20Newberry E.P. Latifi T. Towler D.A. Biochemistry. 1998; 37: 16360-16368Crossref PubMed Scopus (122) Google Scholar, 21Shirakabe K. Terasawa K. Miyama K. Shibuya H. Nishida E. Genes Cells. 2001; 6: 851-856Crossref PubMed Scopus (155) Google Scholar, 22Barnes G.L. Javed A. Waller S.M. Kamal M.H. Hebert K.E. Hassan M.Q. Bellahcene A. Van Wijnen A.J. Young M.F. Lian J.B. Stein G.S. Gerstenfeld L.C. Cancer Res. 2003; 63: 2631-2637PubMed Google Scholar), whereas Dlx5 is an activator of osteoblast marker gene expression (15Ryoo H.-M. Hoffmann H.M. Beumer T. Frenkel B. Towler D.A. Stein G.S. Stein J.L. Van Wijnen A.J. Lian J.B. Mol. Endocrinol. 1997; 11: 1681-1694Crossref PubMed Scopus (217) Google Scholar, 23Benson M.D. Bargeon J.L. Xiao G. Thomas P.E. Kim A. Cui Y. Franceschi R.T. J. Biol. Chem. 2000; 275: 13907-13917Abstract Full Text Full Text PDF PubMed Scopus (81) Google Scholar, 24Dodig M. Kronenberg M.S. Bedalov A. Kream B.E. Gronowicz G. Clark S.H. Mack K. Liu Y.H. Maxon R. Pan Z.Z. Upholt W.B. Rowe D.W. Lichtler A.C. J. Biol. Chem. 1996; 271: 16422-16429Abstract Full Text Full Text PDF PubMed Scopus (85) Google Scholar, 25Kim Y.-J. Lee M.-H. Wozney J.M. Cho J.-Y. Ryoo H.-M. J. Biol. Chem. 2004; 279: 50773-50780Abstract Full Text Full Text PDF PubMed Scopus (160) Google Scholar). Dlx5 is the key mediator of BMP-induced Runx2 expression (14Lee M.-H. Kim Y.-J. Kim H.-J. Park H.-D. Kang A.-R. Kyung H.-M. Sung J.-H. Wozney J.M. Kim H.-J. Ryoo H.-M. J. Biol. Chem. 2003; 278: 34387-34394Abstract Full Text Full Text PDF PubMed Scopus (349) Google Scholar), but little is known about how it regulates Runx2 expression in response to BMP-2 signaling. Therefore, the first aim of this study was to assess which promoter responds to BMP signaling because Runx2 has two major isoforms whose expression is regulated by alternative promoters. In addition, as Dlx5 acts downstream of BMP-2, we undertook to define cis-acting elements in the BMP-2-responsive Runx2 promoter that interact with Dlx5. Finally, as Msx2 is thought to inhibit the Dlx5-induced expression of osteoblast marker genes (20Newberry E.P. Latifi T. Towler D.A. Biochemistry. 1998; 37: 16360-16368Crossref PubMed Scopus (122) Google Scholar, 21Shirakabe K. Terasawa K. Miyama K. Shibuya H. Nishida E. Genes Cells. 2001; 6: 851-856Crossref PubMed Scopus (155) Google Scholar), we also explored the relationship of these two transcription factors with respect to the Runx2 promoter. Here, we demonstrate that Dlx5 up-regulates Runx2 (more specifically, Runx2-II) expression and that the effect of Dlx5 on the Runx2 P1 promoter is antagonized by Msx2. Materials—Bioactive recombinant human BMP-2 was from Wyeth Pharmaceuticals (Cambridge, MA). Recombinant human TGF-β1 was purchased from R&D Systems (Minneapolis, MN). Dulbecco's modified Eagle's medium, α-minimal essential medium, fetal bovine serum, and Lipofectamine PLUS™ reagent were from Invitrogen. Tissue culture plasticware was from Corning (Corning, NY); the Megaprime DNA labeling system was from Amersham Biosciences; DNA midi-prep kits were from Qiagen Inc.; and ExpressHyb hybridization solution was from Clontech. The Zeta-Probe blotting membrane was from Bio-Rad, and the Dual-Luciferase reporter assay kit, Taq polymerase, Pfu polymerase, dNTP mixture, pGL3-Basic vector, and TnT coupled reticulocyte lysate system were from Promega Corp. Anti-FLAG monoclonal antibody M2, HEPES, Nonidet P-40, EDTA, dithiothreitol, phenylmethylsulfonyl fluoride, glycerol, and poly(dI-dC) were purchased from Sigma. Cell Culture and Northern Blot Analysis—Mouse myogenic C2C12 cells, osteoblast-like MC3T3-E1 cells, and the rat osteosarcoma cell line ROS17/2.8 were maintained as described previously (12Lee M.-H. Javed A. Kim H.-J. Shin H.I. Gutierrez S. Choi J.-Y. Rosen V. Stein J.L. Van Wijnen A.J. Stein G.S. Lian J.B. Ryoo H.-M. J. Cell. Biochem. 1999; 73: 114-125Crossref PubMed Scopus (238) Google Scholar). C2C12 cells stably transfected with Dlx5, antisense (AS) Dlx5 (Dlx5-AS), Msx2, and antisense Msx2 (Msx2-AS) constructs were established and maintained as described previously (14Lee M.-H. Kim Y.-J. Kim H.-J. Park H.-D. Kang A.-R. Kyung H.-M. Sung J.-H. Wozney J.M. Kim H.-J. Ryoo H.-M. J. Biol. Chem. 2003; 278: 34387-34394Abstract Full Text Full Text PDF PubMed Scopus (349) Google Scholar). The Runx2-/- calvarial cell line H1-127-21-2 was maintained as described previously (13Lee K.S. Kim H.-J. Li Q.L. Chi X.-Z. Ueta C. Komori T. Wozney J.M. Kim E.G. Choi J.-Y. Ryoo H.-M. Bae S.-C. Mol. Cell. Biol. 2000; 20: 8783-8792Crossref PubMed Scopus (748) Google Scholar). Chinese hamster ovary cells was maintained in α-minimal essential medium in the presence of 10% fetal bovine serum. C2C12 cells were plated at a density of 1 × 106 cells/100-mm culture dish. To examine the effects of BMP-2 or TGF-β1 on cell differentiation, the cells were cultured for the indicated periods with or without the indicated factors in medium with 5% fetal bovine serum. All cells were harvested with phosphate-buffered saline by scraping with a rubber policeman at 4 °C. Total RNA was extracted from the cells, and Northern blot analysis for Runx2 was performed as described previously (14Lee M.-H. Kim Y.-J. Kim H.-J. Park H.-D. Kang A.-R. Kyung H.-M. Sung J.-H. Wozney J.M. Kim H.-J. Ryoo H.-M. J. Biol. Chem. 2003; 278: 34387-34394Abstract Full Text Full Text PDF PubMed Scopus (349) Google Scholar). Reverse Transcription-PCR and Quantitative Real-time PCR—Conventional reverse transcription-PCR for murine Runx2 was performed with the primers described previously (26Lee M.-H. Kwon T.G. Park H.S. Wozney J.M. Ryoo H.-M. Biochem. Biophys. Res. Commun. 2003; 309: 689-694Crossref PubMed Scopus (326) Google Scholar). Quantitative real-time PCR was carried out using SYBR Green fluorescence dye on a LightCycler machine (Roche Applied Science, Mannheim, Germany) as described previously (27Choi K.-Y. Kim H.-J. Lee M.-H. Kwon T.G. Nah H.D. Furuichi T. Komori T. Nam S.H. Kim Y.-J. Kim H.-J. Ryoo H.-M. Dev. Dyn. 2005; 233: 115-121Crossref PubMed Scopus (89) Google Scholar). PCR primers for mouse Runx2 (forward, 5′-CCAGAATGATGGTGTTGACG-3′; and reverse, 5′-GGTTGCAAGATCATGACTAGGG-3′) and mouse glyceraldehyde-3-phosphate dehydrogenase (26Lee M.-H. Kwon T.G. Park H.S. Wozney J.M. Ryoo H.-M. Biochem. Biophys. Res. Commun. 2003; 309: 689-694Crossref PubMed Scopus (326) Google Scholar) were synthesized by TaKaRa Korea (Seoul, Korea). All samples were run in triplicate, and the relative levels of Runx2 mRNA were normalized to those of glyceraldehyde-3-phosphate dehydrogenase. DNA Constructs—The construction of the Dlx5, Dlx5-AS, Msx2, and Msx2-AS expression vectors and the establishment of each stable cell line have been described previously (14Lee M.-H. Kim Y.-J. Kim H.-J. Park H.-D. Kang A.-R. Kyung H.-M. Sung J.-H. Wozney J.M. Kim H.-J. Ryoo H.-M. J. Biol. Chem. 2003; 278: 34387-34394Abstract Full Text Full Text PDF PubMed Scopus (349) Google Scholar, 25Kim Y.-J. Lee M.-H. Wozney J.M. Cho J.-Y. Ryoo H.-M. J. Biol. Chem. 2004; 279: 50773-50780Abstract Full Text Full Text PDF PubMed Scopus (160) Google Scholar). The mouse Runx2 P1 promoter (-2782 to +112 bp) and the Runx2 P2 promoter (-4056 to +246 bp) were generated based on GenBank™ accession numbers AF155360 (from 1784 to 4677 bp) and AF155361 (from 1 to 4303 bp), respectively. The Runx2 P1 and Runx2 P2 promoter deletion constructs were generated by serial deletion from the 5′-end of the promoter with mung bean nuclease, and the fragments were ligated into the KpnI/BglII and SmaI sites, respectively, of the pGL3-Basic vector. The Runx2 P1 promoter deletion constructs P1-342 (-342 to +112 bp), P1-756 (-756 to +112 bp), P1-1664 (-1664 to +112 bp), and P1-2782 (-2782 to +112 bp) were cloned into the pGL3-Basic vector, as were the Runx2 P2 promoter deletion constructs P2-857 (-857 to +246 bp), P2-1843 (-1843 to +246 bp), P2-2648 (-2648 to +246 bp), and P2-4056 (-4056 to +246 bp). The Runx2 P1 promoter deletion constructs P1-458 (-458 to +112 bp; containing the Dlx5 D3-binding site) and P1-610 (-610 to +112 bp; containing the Dlx5 D2/D3-binding sites) were constructed using P1-756 as a template. The forward primers (bearing a KpnI restriction site for the P1-458 and P1-610 constructs) and the reverse primer (bearing a StuI restriction site) are listed in TABLE ONE. Each PCR product was digested with KpnI and StuI and subcloned into the KpnI/StuI-digested Runx2 P1-756 vector. Myc-tagged Dlx5 and FLAG-tagged Msx2 were subcloned into pcDNA3.1.TABLE ONEPrimer sequences for construction of Runx2 promoter deletion mutants and site-directed mutagenesisNameOligonucleotide sequenceSequence locationbpP1-6105′-GCggtaccGCCACACACTCAGTTGAGAC-3′ (forward)−610 to −591P1-4585′-CAggtaccTGCTCTCCAGAGGCTTAACC-3′ (forward)−458 to −439Reverse5′-CTTGTGGTAaggcctTCCTG-3′−331 to −350MTD15′-GATACAATCCCAAGATGCGAgcgggTGCAAAGCAGCACTGTTGCTC-3′ (forward)−661 to −616MTD25′-CAATTTTGCTCACTTTTCCATAGACAccgtcgccAAGGAAAGGGAGGAGGGGTAG- 3′ (forward)−591 to −537MTD35′-CAAATCCTCATGAGTCACAAAAgcggcAAAGCTATAACCTTCTGAATG-3′ (forward)−399 to −352GL25′-CTTTATGTTTTTGGCGTCTTCCA-3′ (reverse)RV35′-CTAGCAAAATAGGCTGTCCC-3′ (forward) Open table in a new tab Site-directed Mutagenesis of Dlx5-binding Sites—To produce constructs that bear mutations of each binding site alone, two sites, or all three sites in the three putative homeodomain-response elements (HDREs) of the Runx2 P1-756 construct, a site-directed mutagenic two-step PCR strategy was followed. For the first set of PCRs, three mutant primers designated MTD1 (-661 to -616 bp), MTD2 (-591 to -537 bp), and MTD3 (-399 to -352 bp) (TABLE ONE) were used as the forward primers, and the GL2 primer (Promega Corp.) was used as the reverse primer to create Runx2-MTD1, Runx2-MTD2, and Runx2-MTD3, respectively, using the Runx2 P1-756 construct as a template. In the second round of PCR, the RV3 primer (Promega Corp.) was used as the forward primer, and PCR products from the first round were used as the reverse primers with P1-756 as a template. The cycling parameters for PCR were as follows: 94 °C for 60 s, 48 °C for 1 min (first round) or 44 °C for 1 min (second round), and 72 °C for 2 min for 35 cycles, followed by 72 °C for 10 min. PCR products with mutated binding sites were digested with KpnI/PstI, and the resulting 868-bp fragments were ligated to pGL3-Basic. Derivatives with two or three mutated sites were constructed using combinations of each mutant construct described above. All constructs were confirmed by sequencing (Macrogen, Seoul). In Situ Hybridization—Probe preparation (BMP-2, Dlx5, Msx2, Runx2-I, and Runx2-II), tissue preparation, and in situ hybridization procedures were as described previously (10Park M.H. Shin H.I. Choi J.-Y. Nam S.H. Kim Y.-J. Kim H.-J. Ryoo H.-M. J. Bone Miner. Res. 2001; 16: 885-892Crossref PubMed Scopus (72) Google Scholar). Calvariae of ICR mice (embryonic day 16) were prepared as described previously (28Kim H.-J. Rice D.P. Kettunen P.J. Thesleff I. Development (Camb.). 1998; 125: 1241-1251PubMed Google Scholar). Sections were stained with hematoxylin and eosin to assess the developing calvaria. The 240-bp murine BMP-2 fragment in pGEM3 (Promega Corp.) was digested with HindIII or EcoRI. The 800-bp rat Dlx5 fragment in pCR2 (Invitrogen) was digested with BamHI or XbaI. The 850-bp murine Msx2 fragment in pSP72 (Invitrogen) was digested with HindIII or BglII. In all three cases, antisense and sense riboprobes were produced by T7 and SP6 RNA polymerases, respectively. Runx2-I- and Runx2-II-specific riboprobe preparation and in situ hybridization were as described previously (10Park M.H. Shin H.I. Choi J.-Y. Nam S.H. Kim Y.-J. Kim H.-J. Ryoo H.-M. J. Bone Miner. Res. 2001; 16: 885-892Crossref PubMed Scopus (72) Google Scholar, 28Kim H.-J. Rice D.P. Kettunen P.J. Thesleff I. Development (Camb.). 1998; 125: 1241-1251PubMed Google Scholar). Transfection and Luciferase Assay—C2C12, ROS17/2.8, and Runx2-/- cells were plated in 6-well plates at a density of 1 × 105 cells/well. After overnight culture, cells were transfected with Lipofectamine PLUS™ according to the manufacturer's instructions. Each transfection assay was performed with 0.5 μg of the Dlx5 or Msx2 expression vector or pcDNA3 and 0.5 μg of the Runx2 promoter-luciferase reporter vector. All plasmid DNA was prepared using a DNA midi-prep kit. Three hours after transfection, the medium was changed, and the cells were cultured for the indicated periods. The cells were then harvested, and luciferase activity was determined with a Dual-Luciferase reporter assay kit. The results presented are representative data from at least three independent experiments with triplicate samples. Electrophoretic Mobility Shift Assay (EMSA)—The oligonucleotide sequences of three putative homeodomain-binding sites between -342 and -756 bp from the transcription start site of the Runx2 P1 promoter were synthesized (Takara Bio Inc., Shiga, Japan) (see Fig. 4D). The mutant D3 oligonucleotides were synthesized by substituting the ATTA sequence with the underlined sequences in Fig. 4D. These double-stranded DNA probes were end-labeled with [α-32P]dCTP using Klenow enzyme. The Dlx5 and Msx2 proteins were produced by in vitro transcription and translation using the TnT coupled reticulocyte lysate system. Nuclear extracts from C2C12 or Chinese hamster ovary cells that were transiently transfected with Dlx5 for 24 h were prepared as described previously (29Kim H.-J. Kim J.-H. Bae S.-C. Choi J.-Y. Kim H.-J. Ryoo H.-M. J. Biol. Chem. 2003; 278: 319-326Abstract Full Text Full Text PDF PubMed Scopus (209) Google Scholar). The Dlx5 or Msx2 protein was incubated with labeled double-stranded DNA probes in the absence or presence of a 50-, 100-, or 500-fold molar excess of unlabeled competitor for 20 min at room temperature. For the supershift assay, Dlx5 or Msx2 was preincubated with anti-Dlx5 rabbit polyclonal antibody (14Lee M.-H. Kim Y.-J. Kim H.-J. Park H.-D. Kang A.-R. Kyung H.-M. Sung J.-H. Wozney J.M. Kim H.-J. Ryoo H.-M. J. Biol. Chem. 2003; 278: 34387-34394Abstract Full Text Full Text PDF PubMed Scopus (349) Google Scholar) or anti-FLAG monoclonal antibody M2, respectively, for 20 min at room temperature before incubation with the labeled probe. Protein-DNA complexes were separated at 4 °C on a 6% polyacrylamide gel containing 0.5× Tris borate/EDTA. Chromatin Immunoprecipitation Assays—Chromatin immunoprecipitation assays were performed as described in detail previously (30Shen J. Montecino M. Lian J.B. Stein G.S. Van Wijnen A.J. Stein J.L. J. Biol. Chem. 2002; 277: 20284-20292Abstract Full Text Full Text PDF PubMed Scopus (64) Google Scholar, 31Shen J. Hovhannisyan H. Lian J.B. Montecino M.A. Stein G.S. Stein J.L. Van Wijnen A.J. Mol. Endocrinol. 2003; 17: 743-756Crossref PubMed Scopus (85) Google Scholar). C2C12 cells were transiently transfected with the Myc-Dlx5 or FLAG-Msx2 construct for 24 h. PCR primer pairs were generated to detect DNA segments located between nucleotides -715 and -431 of the mouse Runx2 distal promoter (nucleotide +1 is the mRNA cap site): forward, 5′-AACAGAAGGAAGCAGCCACC; and reverse, 5′-CCACACTCCTGTAAGGTTAAGC. Runx2-II Is Specifically Stimulated by BMP-2 and Dlx5—Our previous study showed that both BMP-2 and TGF-β1 stimulate Runx2 expression in C2C12 cells (12Lee M.-H. Javed A. Kim H.-J. Shin H.I. Gutierrez S. Choi J.-Y. Rosen V. Stein J.L. Van Wijnen A.J. Stein G.S. Lian J.B. Ryoo H.-M. J. Cell. Biochem. 1999; 73: 114-125Crossref PubMed Scopus (238) Google Scholar). In this study, we found that BMP-2 treatment stimulated the expression of Runx2-II (5.4 kb) (Fig. 1A, arrow) and Runx2-I (6.0 kb) (arrowhead) transcripts, whereas TGF-β1 treatment stimulated mainly Runx2-I expression. Runx2-II expression was more strongly stimulated by BMP-2 treatment compared with Runx2-I expression. In addition, overexpression of Dlx5, which mediates Runx2 expression in response to BMP-2 signaling (14Lee M.-H. Kim Y.-J. Kim H.-J. Park H.-D. Kang A.-R. Kyung H.-M. Sung J.-H. Wozney J.M. Kim H.-J. Ryoo H.-M. J. Biol. Chem. 2003; 278: 34387-34394Abstract Full Text Full Text PDF PubMed Scopus (349) Google Scholar), also specifically stimulated Runx2-II transcription (Fig. 1B, lanes 5 and 6). Moreover, in cells stably transfected with Dlx5-AS (14Lee M.-H. Kim Y.-J. Kim H.-J. Park H.-D. Kang A.-R. Kyung H.-M. Sung J.-H. Wozney J.M. Kim H.-J. Ryoo H.-M. J. Biol. Chem. 2003; 278: 34387-34394Abstract Full Text Full Text PDF PubMed Scopus (349) Google Scholar), the antisense intervention of Dlx5 specifically blocked the BMP-2-stimulated expression of Runx2-II, but not Runx2-I (Fig. 1B, lanes 1-4). We previously demonstrated that these two Runx2 isoforms are differentially expressed during intramembranous bone formation in mouse cranial and axial bone development (10Park M.H. Shin H.I. Choi J.-Y. Nam S.H. Kim Y.-J. Kim H.-J. Ryoo H.-M. J. Bone Miner. Res. 2001; 16: 885-892Crossref PubMed Scopus (72) Google Scholar, 11Choi K.-Y. Lee S.W. Park M.H. Bae Y.C. Shin H.I. Nam S. Kim Y.-J. Kim H.-J. Ryoo H.-M. Exp. Mol. Med. 2002; 34: 426-433Crossref PubMed Scopus (46) Google Scholar). As Runx2-II expression was specifically induced by BMP-2 treatment or by overexpression of its immediate downstream target gene Dlx5, we examined the expression of BMP-2, Dlx5, and Runx2 in the developing mouse calvaria. The pattern of Dlx5 expression matched those of BMP-2 and Runx2-II (Fig. 2, A-C). All three transcripts were strongly expressed in osteogenic fronts and parietal bones, but not in poorly differentiated sutural mesenchymal cells, whereas Runx2-I and Msx2 were strongly expressed in the latter cells (Fig. 2, E and F). Dlx5 Specifically Enhances Runx2 P1 Promoter Activity—Because expression of the two major isoforms of Runx2 is differentially regulated by two different promoters and because both BMP-2 treatment and Dlx5 overexpression are more closely associated with Runx2-II, the P1 promoter product, we analyzed the responsiveness of the Runx2 P1 and P2 promoters to Dlx5 overexpression. For this purpose, we prepared a series of 5′-deletion constructs as illustrated in Fig. 3A. The basal promoter activity of P1 promoter deletion constructs was notably stronger than that of P2 promoter deletion constructs (Fig. 3B). P1 promoter deletion analysis indicated a strong increase in reporter activity between P1-342 and P1-756 in both non-osteogenic (C2C12) and osteogenic (ROS17/2.8) cells. Interestingly, the stimulatory activity of Dlx5 disappeared for longer constructs in these cells (P1-1664 and P1-2782) (Fig. 3, B and C). In contrast, the stimulatory effect of Dlx5 was determined in P1-756 and was still observed for longer constructs (P1-1664 and P1-2782) in Runx2-/- cells (Fig. 3C). Taken together, these results suggest that the Dlx5-response element in the P1 promoter is located between -756 and -342 bp from the Runx2-II transcription start site. Identification of Dlx5-response Elements in the Runx2 P1 Promoter—Insilico analysis of the Runx2 P1 promoter indicated that it contains three putative H