Title: Myocyte Enhancer Factor 2C and Myogenin Up-regulate Each Other's Expression and Induce the Development of Skeletal Muscle in P19 Cells
Abstract: Two families of transcription factors, myogenic regulatory factors (MRFs) and myocyte enhancer factor 2 (MEF2), function synergistically to regulate myogenesis. In addition to activating structural muscle-specific genes, MRFs and MEF2 activate each other's expression. The MRF, myogenin, can activate MEF2 DNA binding activity when transfected into fibroblasts and, in turn, the myogenin promoter contains essential MEF2 DNA binding elements. To determine which MEF2 is involved in this regulation, P19 cells stably expressing MyoD and myogenin were compared for their ability to activate the expression of MEF2 family members. There was very little cross-activation of MyoD expression by myogenin and vice versa. Myogenin expression, and not MyoD, was found to up-regulate MEF2C expression. MEF2A, -B, and -D expression levels were not up-regulated by overexpression of either MyoD or myogenin. To examine whether MEF2C can differentially regulate MyoD or myogenin expression, P19 cell lines overexpressing MEF2C were analyzed. MEF2C induced myogenesis in P19 cells and up-regulated the expression of myogenin with 25-fold greater efficiency than that of MyoD. Therefore, myogenin and MEF2C participate in a regulatory loop in differentiating stem cells. This positive regulation does not extend to MyoD or the other MEF2 family members. Consequently, MEF2C appears to play a specific role in early events of myogenesis. Two families of transcription factors, myogenic regulatory factors (MRFs) and myocyte enhancer factor 2 (MEF2), function synergistically to regulate myogenesis. In addition to activating structural muscle-specific genes, MRFs and MEF2 activate each other's expression. The MRF, myogenin, can activate MEF2 DNA binding activity when transfected into fibroblasts and, in turn, the myogenin promoter contains essential MEF2 DNA binding elements. To determine which MEF2 is involved in this regulation, P19 cells stably expressing MyoD and myogenin were compared for their ability to activate the expression of MEF2 family members. There was very little cross-activation of MyoD expression by myogenin and vice versa. Myogenin expression, and not MyoD, was found to up-regulate MEF2C expression. MEF2A, -B, and -D expression levels were not up-regulated by overexpression of either MyoD or myogenin. To examine whether MEF2C can differentially regulate MyoD or myogenin expression, P19 cell lines overexpressing MEF2C were analyzed. MEF2C induced myogenesis in P19 cells and up-regulated the expression of myogenin with 25-fold greater efficiency than that of MyoD. Therefore, myogenin and MEF2C participate in a regulatory loop in differentiating stem cells. This positive regulation does not extend to MyoD or the other MEF2 family members. Consequently, MEF2C appears to play a specific role in early events of myogenesis. myocyte enhancer factor 2 myogenic regulatory factor polymerase chain reaction reverse transcription kilobase pair(s) myosin heavy chain Two families of transcription factors, the MEF21 family and the myogenic basic helix-loop-helix family (MRFs), interact to synergistically activate skeletal muscle-specific promoters (1Kaushal S. Schneider J.W. Nadal-Ginard B. Mahdavi V. Science. 1994; 266: 1236-1240Crossref PubMed Scopus (194) Google Scholar, 2Naidu P.S. Ludolph D.C. To R.Q. Hinterberger T.J. Konieczny S.F. Mol. Cell. Biol. 1995; 15: 2707-2718Crossref PubMed Scopus (119) Google Scholar, 3Molkentin J.D. Black B.L. Martin J.F. Olson E.N. Cell. 1995; 83: 1125-1136Abstract Full Text PDF PubMed Scopus (693) Google Scholar). The four vertebrate MEF2 family members, MEF2A-D (4Pollock R. Treisman R. Genes Dev. 1991; 5: 2327-2341Crossref PubMed Scopus (321) Google Scholar, 5Molkentin J.D. Olson E.N. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 9366-9373Crossref PubMed Scopus (372) Google Scholar), contain a conserved MCM1, agamous, deficiens, and serum response factor-box/MEF2 domain at their N termini. This domain mediates protein-protein interactions as well as DNA binding to an AT-rich MEF2 binding site. The four MRFs, MyoD, myogenin, myf-5, and MRF4 (6Davis R.L. Weintraub H. Lassar A.B. Cell. 1987; 51: 987-1000Abstract Full Text PDF PubMed Scopus (2422) Google Scholar, 7Braun T. Buschhausen-Denker G. Bober E. Tannich E. Arnold H.H. EMBO J. 1989; 8: 701-709Crossref PubMed Scopus (650) Google Scholar, 8Edmondson D.G. Olson E.N. Genes Dev. 1989; 3: 628-640Crossref PubMed Scopus (597) Google Scholar, 9Rhodes S.J. Konieczny S.F. Genes Dev. 1989; 3: 2050-2061Crossref PubMed Scopus (560) Google Scholar, 10Wright W.E. Sassoon D.A. Lin V.K. Cell. 1989; 56: 607-617Abstract Full Text PDF PubMed Scopus (924) Google Scholar, 11Braun T. Bober E. Winter B. Rosenthal N. Arnold H.H. EMBO J. 1990; 9: 821-831Crossref PubMed Scopus (353) Google Scholar, 12Miner J.H. Wold B. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 1089-1093Crossref PubMed Scopus (344) Google Scholar), bind to E box sequences in the promoters of skeletal muscle-specific genes and can induce skeletal muscle development when expressed in fibroblasts (13Lassar A.B. Buskin J.N. Lockshon D. Davis R.L. Apone S. Hauschka S.D. Weintraub H. Cell. 1989; 58: 823-831Abstract Full Text PDF PubMed Scopus (566) Google Scholar). A positive feedback loop likely exists between MEF2 and myogenin because myogenin induces MEF2 DNA binding activity in various cell lines (14Cserjesi P. Olson E.N. Mol. Cell. Biol. 1991; 11: 4854-4862Crossref PubMed Scopus (199) Google Scholar), and MEF2 DNA binding sites regulate myogenin expression (15Yee S.P. Rigby P.W. Genes Dev. 1993; 7: 1277-1289Crossref PubMed Scopus (347) Google Scholar, 16Cheng T.C. Wallace M.C. Merlie J.P. Olson E.N. Science. 1993; 261: 215-218Crossref PubMed Scopus (220) Google Scholar) but not MyoD expression (17Goldhamer D.J. Brunk B.P. Faerman A. King A. Shani M. Emerson Jr., C.P. Development. 1995; 121: 637-649Crossref PubMed Google Scholar) during embryogenesis. Insight into the roles of the MRFs has been gained from gene knockout studies in mice. Skeletal muscle development proceeded normally in homozygous null mice missing either MyoD (18Rudnicki M.A. Braun T. Hinuma S. Jaenisch R. Cell. 1992; 71: 383-390Abstract Full Text PDF PubMed Scopus (748) Google Scholar) or myf-5 (19Braun T. Rudnicki M.A. Arnold H.H. Jaenisch R. Cell. 1992; 71: 369-382Abstract Full Text PDF PubMed Scopus (547) Google Scholar), but in double homozygous mice lacking both, myoblasts did not form (20Rudnicki M.A. Schnegelsberg P.N. Stead R.H. Braun T. Arnold H.H. Jaenisch R. Cell. 1993; 75: 1351-1359Abstract Full Text PDF PubMed Scopus (1299) Google Scholar). Mice lacking myogenin produced normal myoblast cells but displayed a marked reduction in secondary myofibers (21Hasty P. Bradley A. Morris J.H. Edmondson D.G. Venuti J.M. Olson E.N. Klein W.H. Nature. 1993; 364: 501-506Crossref PubMed Scopus (1001) Google Scholar, 22Nabeshima Y. Hanaoka K. Hayasaka M. Esumi E. Li S. Nonaka I. Nabeshima Y. Nature. 1993; 364: 532-535Crossref PubMed Scopus (713) Google Scholar, 23Venuti J.M. Morris J.H. Vivian J.L. Olson E.N. Klein W.H. J. Cell Biol. 1995; 128: 563-576Crossref PubMed Scopus (218) Google Scholar). Taken together these results indicate that MyoD and myf-5 play a role in the determination of skeletal muscle and that myogenin plays an essential in vivorole in the terminal differentiation of secondary muscle fibers. An essential role for d-MEF2 in the development of cardiac, skeletal, and smooth muscle has been demonstrated by the deficiency of these tissues in Drosophila lacking the single d-mef2 gene (24Lilly B. Zhao B. Ranganayakulu G. Paterson B.M. Schulz R.A. Olson E.N. Science. 1995; 267: 688-693Crossref PubMed Scopus (419) Google Scholar, 25Bour B.A. O'Brien M.A. Lockwood W.L. Goldstein E.S. Bodmer R. Taghert P.H. Abmayr S.M. Nguyen H.T. Genes Dev. 1995; 9: 730-741Crossref PubMed Scopus (358) Google Scholar, 26Ranganayakulu G. Zhao B. Dokidis A. Molkentin J.D. Olson E.N. Schulz R.A. Dev. Biol. 1995; 171: 169-181Crossref PubMed Scopus (172) Google Scholar). During murine somitogenesis, MEF2C is the first MEF2 factor to be expressed in the somite on day 9, just after expression of myogenin (27Edmondson D.G. Lyons G.E. Martin J.F. Olson E.N. Development. 1994; 120: 1251-1263Crossref PubMed Google Scholar, 28Subramanian S.V. Nadalginard B. Mech. Dev. 1996; 57: 103-112Crossref PubMed Scopus (46) Google Scholar). Mice lacking MEF2C die around embryonic day 10 because of a defect in cardiac looping, whereas mice lacking MEF2B are viable and develop normally. A dominant-negative form of MEF2A inhibits the differentiation of myoblast cells, indicating that the trans-activation function of MEF2 proteins is essential for skeletal muscle-specific gene expression (29Ornatsky O.I. Andreucci J.J. McDermott J.C. J. Biol. Chem. 1997; 272: 33271-33278Abstract Full Text Full Text PDF PubMed Scopus (102) Google Scholar). Although it is believed that MEF2 factors are important for amplifying and maintaining the skeletal myogenic program, the specific MEF2 family member involved and whether MEF2 family members are sufficient to induce myogenesis in mammalian cells is unclear. One study (1Kaushal S. Schneider J.W. Nadal-Ginard B. Mahdavi V. Science. 1994; 266: 1236-1240Crossref PubMed Scopus (194) Google Scholar) demonstrated that MEF2A induced skeletal muscle development in fibroblasts, whereas another study (3Molkentin J.D. Black B.L. Martin J.F. Olson E.N. Cell. 1995; 83: 1125-1136Abstract Full Text PDF PubMed Scopus (693) Google Scholar) found that MEF2A did not, although MRF expression could be activated (24Lilly B. Zhao B. Ranganayakulu G. Paterson B.M. Schulz R.A. Olson E.N. Science. 1995; 267: 688-693Crossref PubMed Scopus (419) Google Scholar). In contrast,d-MEF2 activated the expression of skeletal muscle-specific genes when overexpressed at a particular stage of development inDrosophila ectoderm (30Lin M.H. Bour B.A. Abmayr S.M. Storti R.V. Dev. Biol. 1997; 182: 240-255Crossref PubMed Scopus (43) Google Scholar). P19 cells are an alternative system for studying early embryonic events (31Skerjanc I.S. Trends Cardiovasc. Med. 1999; (in press)PubMed Google Scholar). Upon cellular aggregation, P19 cells undergo differentiation that emulates the biochemical and morphological events of early embryonic development (32McBurney M.W. Int. J. Dev. Biol. 1993; 37: 135-140PubMed Google Scholar, 33Rudnicki M.A. McBurney M.W. Robertson E.J. Teratocarcinomas and Embryonic Stem Cells. A Practical Approach. IRL Press, Oxford1987: 19-49Google Scholar). Aggregation of P19 cells with Me2SO results in the development of cardiac and skeletal muscle along with other mesodermal and endodermal cell types (34Edwards M.K. Harris J.F. McBurney M.W. Mol. Cell. Biol. 1983; 3: 2280-2286Crossref PubMed Scopus (139) Google Scholar). These cardiocytes first appear at day 6 following Me2SO treatment and express cardiac-specific transcription factors, including GATA-4, Nkx2–5, and MEF2C (35Grepin C. Nemer G. Nemer M. Development. 1997; 124: 2387-2395Crossref PubMed Google Scholar, 36Skerjanc I.S. Petropoulos H. Ridgeway A.G. Wilton S. J. Biol. Chem. 1998; 273: 34904-34910Abstract Full Text Full Text PDF PubMed Scopus (124) Google Scholar). P19 cells treated with Me2SO differentiate into skeletal muscle by day 9 of differentiation and express MyoD (37Skerjanc I.S. McBurney M.W. Dev. Biol. 1994; 163: 125-132Crossref PubMed Scopus (35) Google Scholar). The development of skeletal and cardiac muscle in P19 cells is regulated by unknown components in the fetal calf serum (38Wilton S. Skerjanc I.S. In Vitro Cellular and Developmental Biology-Animal. 1999; 35: 175-177Crossref PubMed Scopus (27) Google Scholar). P19 cells aggregated with segments of chick neural tube differentiate into skeletal muscle in the absence of Me2SO, indicating that signals from the neural tube can replace the requirement for Me2SO (39Angello J.C. Stern H.M. Hauschka S.D. Dev. Biol. 1997; 192: 93-98Crossref PubMed Scopus (15) Google Scholar). The approach of overexpressing transcription factors in P19 cells has yielded important information about their function and regulation. For example, P19 cells overexpressing either GATA-4 (35Grepin C. Nemer G. Nemer M. Development. 1997; 124: 2387-2395Crossref PubMed Google Scholar) or MyoD (40Skerjanc I.S. Slack R.S. McBurney M.W. Mol. Cell. Biol. 1994; 1464: 8451-8459Crossref Scopus (70) Google Scholar) differentiate into cardiac muscle or skeletal muscle, respectively, when aggregated in the absence of Me2SO. Using this approach, we have shown recently that MEF2C and Nkx2–5 activate each other's expression and induce cardiomyogenesis in P19 cells aggregated in the absence of Me2SO (36Skerjanc I.S. Petropoulos H. Ridgeway A.G. Wilton S. J. Biol. Chem. 1998; 273: 34904-34910Abstract Full Text Full Text PDF PubMed Scopus (124) Google Scholar). Here we set out to determine if MEF2 can initiate myogenesis in differentiating P19 stem cells and to identify the family members that participate in the MRF/MEF2 regulatory loop. We show that MEF2C can induce skeletal myogenesis when overexpressed in differentiating stem cells and that MEF2C and myogenin, but not MyoD or MEF2A, -B, or -D, participate in a regulatory loop. Therefore, MEF2C appears to play a specific role in early events of myogenesis. Routine culturing of P19 cells and the isolation of P19 cells overexpressing MEF2C, MyoD, or myogenin, termed P19[MEF2C], P19[MyoD], or P19[Mgn] cells, respectively, have been described previously (36Skerjanc I.S. Petropoulos H. Ridgeway A.G. Wilton S. J. Biol. Chem. 1998; 273: 34904-34910Abstract Full Text Full Text PDF PubMed Scopus (124) Google Scholar,40Skerjanc I.S. Slack R.S. McBurney M.W. Mol. Cell. Biol. 1994; 1464: 8451-8459Crossref Scopus (70) Google Scholar). 2A. G. Ridgeway, S. Wilton, and I. S. Skerjanc, submitted for publication. Each of the three high expressing P19[MEF2C] cell lines behaved similarly, and all experiments reported were performed at least twice with at least two of these cell lines with similar results. Differentiation was initiated by plating 5 × 105cells into 60-mm bacterial dishes in the presence of 0.8% Me2SO. Cells were cultured as aggregates for 4 days, plated in tissue culture dishes, and harvested for RNA or fixed for immunofluorescence at the time indicated. The fetal calf serum used in the experiments presented in this manuscript did not support the differentiation of P19 cells into skeletal muscle (38Wilton S. Skerjanc I.S. In Vitro Cellular and Developmental Biology-Animal. 1999; 35: 175-177Crossref PubMed Scopus (27) Google Scholar). P19 and P19[MEF2C] cells were plated on day 4 of differentiation onto gelatin-coated coverslips. For identifying myosin heavy chain (42Bader D. Masaki T. Fischman D.A. J. Cell Biol. 1982; 95: 763-770Crossref PubMed Scopus (794) Google Scholar), cells were fixed in methanol at −20 °C and reacted with antibody as described (40Skerjanc I.S. Slack R.S. McBurney M.W. Mol. Cell. Biol. 1994; 1464: 8451-8459Crossref Scopus (70) Google Scholar). For identifying myogenin, 50 μl of the monoclonal antibody F5D supernatant (43Wright W.E. Dackorytko I.A. Farmer K. Dev. Genet. 1996; 19: 131-138Crossref PubMed Scopus (24) Google Scholar) was incubated overnight at 4 °C with cells that had been fixed for 1 h at room temperature with Lana's fixative (4% paraformaldehyde, 14% v/v saturated picric acid, 125 mmsodium phosphate). After three 5-min washes in phosphate-buffered saline, cells were incubated for 30 min in 50 μl of phosphate-buffered saline with 1 μl of goat anti-mouse IgG(H+L) Cy3-linked antibody (Jackson Immunoresearch Laboratories, West Grove, PA). Immunofluorescence was visualized with a Zeiss Axioskop microscope; images were captured with a Sony 3CCD color video camera, processed using Northern Exposure, Adobe Photoshop, and Corel Draw software, and printed with a dye sublimation phaser 450 Tektronic printer. Total RNA was isolated by the lithium chloride/urea extraction method, and 6 μg were examined by Northern blot analysis as described previously (40Skerjanc I.S. Slack R.S. McBurney M.W. Mol. Cell. Biol. 1994; 1464: 8451-8459Crossref Scopus (70) Google Scholar). The probes used were: a 600-base pair PstI fragment from the human cardiac α-actin last exon (44Rudnicki M.A. Jackowski G. Saggin L. McBurney M.W. Dev. Biol. 1990; 138: 348-358Crossref PubMed Scopus (88) Google Scholar), a 1.8-kb EcoRI fragment from the mouse MyoD cDNA (6Davis R.L. Weintraub H. Lassar A.B. Cell. 1987; 51: 987-1000Abstract Full Text PDF PubMed Scopus (2422) Google Scholar), a 695-base pairEcoRI/PstI fragment from the rat myogenin cDNA (10Wright W.E. Sassoon D.A. Lin V.K. Cell. 1989; 56: 607-617Abstract Full Text PDF PubMed Scopus (924) Google Scholar), a 1.5-kb HindIII/XbaI fragment of MEF2C cDNA (45Martin J.F. Schwarz J.J. Olson E.N. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 5282-5286Crossref PubMed Scopus (218) Google Scholar), a 1.55-kb XhoI/BamHI fragment of MEF2A cDNA (46Yu Y.T. Breitbart R.E. Smoot L.B. Lee Y. Mahdavi V. Nadal-Ginard B. Genes Dev. 1992; 6: 1783-1798Crossref PubMed Scopus (380) Google Scholar), a 1.55-kbXhoI/BamHI fragment of MEF2B cDNA (47Molkentin J.D. Firulli A.B. Black B.L. Martin J.F. Hustad C.M. Copeland N. Jenkins N. Lyons G. Olson E.N. Mol. Cell. Biol. 1996; 16: 3814-3824Crossref PubMed Scopus (100) Google Scholar), and a 1.5-kb XhoI/BamHI fragment of MEF2D cDNA (48Martin J.F. Miano J.M. Hustad C.M. Copeland N.G. Jenkins N.A. Olson E.N. Mol. Cell. Biol. 1994; 14: 1647-1656Crossref PubMed Scopus (185) Google Scholar). Northern blots were visualized with a Molecular Dynamics PhosphorImager SI and quantitated with ImageQuant software. Averages and standard errors (S.E.) were calculated and reported. RT-PCR was performed as described previously (36Skerjanc I.S. Petropoulos H. Ridgeway A.G. Wilton S. J. Biol. Chem. 1998; 273: 34904-34910Abstract Full Text Full Text PDF PubMed Scopus (124) Google Scholar). A titration of the first strand reaction was performed to ensure that the PCR reaction was analyzed in the linear range. Southern blot analysis was performed on RT-PCR products with a probe from the corresponding cDNA. Negative controls included performing RT-PCR experiments in the absence of RNA, of first strand, and of Superscript II RNase H− reverse transcriptase. The following pairs of primers were used: 5′-tccatccacgtcggccaggct-3′ and 5′-gtagggctcaaccacagcagt-3′ for tubulin with an annealing temperature of 61 °C and 5′-gactccccactccccattcacata-3′ and 5′-ggcggcagctttacaaacaacaca-3′ for myogenin (10Wright W.E. Sassoon D.A. Lin V.K. Cell. 1989; 56: 607-617Abstract Full Text PDF PubMed Scopus (924) Google Scholar), with an annealing temperature of 55 °C. Myosin was harvested by the method of Burridge and Bray (49Burridge K. Bray D. J. Mol. Biol. 1975; 99: 1-14Crossref PubMed Scopus (136) Google Scholar) from cultures at various time points as indicated. The second dialysis step was omitted. Total protein concentration of the myosin extract was measured by the Bio-Rad protein assay (Bio-Rad). Samples (5 μg) were separated by electrophoresis on a 7% SDS-polyacrylamide gel, transferred to nitrocellulose, and probed with one of the following mouse monoclonal antibody supernatants: anti-MHC antibody, MF20 (reacts with all forms), at a 1:6000 dilution (42Bader D. Masaki T. Fischman D.A. J. Cell Biol. 1982; 95: 763-770Crossref PubMed Scopus (794) Google Scholar); anti-fast embryonic MHC antibody, 47a, at a 1:500 dilution (50Merrifield P.A. Sutherland W.M. Litvin J. Konigsberg I.R. Dev. Genet. 1989; 10: 372-385Crossref PubMed Scopus (7) Google Scholar); anti-neonatal fast, BF-35 (reacts with IIA and IIB), at a 1:10 dilution (51DeNardi C. Ausoni S. Moretti P. Gorza L. Velleca M. Buckingham M. Schiaffino S. J. Cell Biol. 1993; 123: 823-835Crossref PubMed Scopus (187) Google Scholar); anti-neonatal slow, 10D10, at a 1:10 dilution (50Merrifield P.A. Sutherland W.M. Litvin J. Konigsberg I.R. Dev. Genet. 1989; 10: 372-385Crossref PubMed Scopus (7) Google Scholar); and anti-MHC IIB, IIX, 212F, at a 1:10 dilution (50Merrifield P.A. Sutherland W.M. Litvin J. Konigsberg I.R. Dev. Genet. 1989; 10: 372-385Crossref PubMed Scopus (7) Google Scholar). Anti-mouse IgG, horseradish peroxidase-linked whole antibody (Amersham Pharmacia Biotech) was used as a secondary antibody at a 1:10,000 dilution, and the reaction was visualized using the ECL Western blotting reagent system (Amersham Pharmacia Biotech) and autoradiography. Several studies have shown differences in the abilities of each of the MRFs to transactivate promoters and to activate endogenous genes. However, the functional significance of these results has not been clearly defined (52Asakura A. Fujisawa-Sehara A. Komiya T. Nabeshima Y. Nabeshima Y. Mol. Cell. Biol. 1993; 13: 7153-7162Crossref PubMed Scopus (40) Google Scholar, 53Brennan T.J. Edmondson D.G. Olson E.N. J. Cell Biol. 1990; 110: 929-937Crossref PubMed Scopus (68) Google Scholar, 54Berberich C. Durr I. Koenen M. Witzemann V. Eur. J. Biochem. 1993; 216: 395-404Crossref PubMed Scopus (40) Google Scholar, 55Fujisawa-Sehara A. Nabeshima Y. Komiya T. Uetsuki T. Asakura A. Nabeshima Y. J. Biol. Chem. 1992; 267: 10031-10038Abstract Full Text PDF PubMed Google Scholar, 56Chakraborty T. Olson E.N. Mol. Cell. Biol. 1991; 11: 6103-6108Crossref PubMed Google Scholar, 57Chakraborty T. Brennan T. Olson E. J. Biol. Chem. 1991; 266: 2878-2882Abstract Full Text PDF PubMed Google Scholar, 58Miller J.B. Everitt E.A. Smith T.H. Block N.E. Dominov J.A. Bioessays. 1993; 15: 191-196Crossref PubMed Scopus (39) Google Scholar, 59Olson E.N. Klein W.H. Genes Dev. 1994; 8: 1-8Crossref PubMed Scopus (606) Google Scholar, 60Gerber A.N. Klesert T.R. Bergstrom D.A. Tapscott S.J. Genes Dev. 1997; 11: 436-450Crossref PubMed Scopus (230) Google Scholar). To determine which MEF2 family members are involved in the regulatory loop with the MRFs, P19[MyoD] (40Skerjanc I.S. Slack R.S. McBurney M.W. Mol. Cell. Biol. 1994; 1464: 8451-8459Crossref Scopus (70) Google Scholar) and P19[Mgn] cell lines were differentiated and examined for their ability to activate the expression of MEF2 family members. Northern blots were prepared using RNA from P19[MyoD] and P19[Mgn] cell lines on day 6 of differentiation in the presence of Me2SO. As shown previously (40Skerjanc I.S. Slack R.S. McBurney M.W. Mol. Cell. Biol. 1994; 1464: 8451-8459Crossref Scopus (70) Google Scholar),2 both MyoD and myogenin required cellular aggregation to initiate myogenesis. MyoD and myogenin did not cross-activate each other's expression to high levels when compared with their exogenous expression levels (Fig. 1, A and B). Therefore, P19[MyoD] and P19[Mgn] cells express predominantly one MRF or the other and can be used to study differences in target gene specificity for MyoD and myogenin. MEF2C expression levels (Fig. 1 C) paralleled the levels of myogenin (Fig. 1 A) and not MyoD (Fig. 1 B) expression in these cell lines. The expression levels of MyoD, myogenin, and MEF2C were quantitated by densitometry in five additional P19[MyoD] and nine additional P19[Mgn] cell lines. Linear regression analysis confirmed that the MEF2C expression correlated with myogenin expression (p< 0.0001) and not MyoD. MEF2A, -B, and -D transcripts were expressed at very low levels and were not up-regulated by myogenin or MyoD expression (data not shown). In summary, myogenin, but not MyoD expression, resulted in the activation of MEF2C and not MEF2A, -B, or -D expression. Because P19[MyoD] and P19[Mgn] skeletal myocytes expressed different levels of MyoD, myogenin, and MEF2C, an examination of the differences in muscle-specific structural gene expression was initiated by analysis of the pattern of MHC isoforms/genes expressed. Western blots performed on myosin harvested on days 8, 10, 14, and 17 showed that the same MHC isoforms were present at similar levels in both P19[MyoD] and P19[Mgn] cultures (Fig. 2), including embryonic fast, neonatal fast, neonatal slow, and adult fast IIB and IIX. Although a doublet of MHC neonatal slow (Fig. 2 D) seems to appear in P19[MyoD] cells, this was not a consistent, reproducible finding. Therefore, whereas MyoD and myogenin differentially activate the expression of MEF2C, they activate similar levels of MHC isoform expression. Because MEF2C expression was specifically up-regulated by myogenin, P19[MEF2C] cells (36Skerjanc I.S. Petropoulos H. Ridgeway A.G. Wilton S. J. Biol. Chem. 1998; 273: 34904-34910Abstract Full Text Full Text PDF PubMed Scopus (124) Google Scholar) were examined in terms of their ability to undergo skeletal myogenesis and activate the expression of the MRFs. Previous work has shown that P19[MEF2C] cells differentiate into cardiac muscle when aggregated in the absence of Me2SO (36Skerjanc I.S. Petropoulos H. Ridgeway A.G. Wilton S. J. Biol. Chem. 1998; 273: 34904-34910Abstract Full Text Full Text PDF PubMed Scopus (124) Google Scholar). In this set of experiments, P19[MEF2C] cells were aggregated in the presence of Me2SO under conditions in which control cells do not undergo myogenesis. P19 and P19[MEF2C] cells were aggregated for 4 days with Me2SO and examined on day 9 by phase contrast microscopy and immunofluorescence with antibodies directed against myogenin and myosin heavy chain (Fig. 3). Bipolar skeletal myocytes were visible by phase contrast microscopy on day 9 in P19[MEF2C] cultures (Fig. 3 D) but not in control cultures (Fig. 3 A). In agreement with the cellular morphology, P19[MEF2C] cultures contained cells that expressed myogenin protein in the nuclei of the bipolar skeletal myocytes (Fig. 3 E). In contrast, P19 control cells did not differentiate into skeletal muscle under the serum conditions used for these experiments, shown by the lack of cells expressing myogenin (Fig. 3 B). These results were confirmed by staining for the presence of myosin heavy chain by immunoreaction with MF20 (42Bader D. Masaki T. Fischman D.A. J. Cell Biol. 1982; 95: 763-770Crossref PubMed Scopus (794) Google Scholar). MF20 reacts with myosin heavy chain present in both cardiac and skeletal muscle, and these two cell types can be distinguished by their morphology (37Skerjanc I.S. McBurney M.W. Dev. Biol. 1994; 163: 125-132Crossref PubMed Scopus (35) Google Scholar). As expected, P19[MEF2C] cultures contained bipolar skeletal myocytes (Fig. 3 F), whereas P19 control-transfected cells contained only cardiac myocytes (Fig. 3 C). The bipolar skeletal myocytes present in P19[MEF2C] cultures are morphologically similar to those previously observed in P19 and P19[MyoD] cells (40Skerjanc I.S. Slack R.S. McBurney M.W. Mol. Cell. Biol. 1994; 1464: 8451-8459Crossref Scopus (70) Google Scholar). The presence of MEF2C did not affect the endogenous differentiation of P19[MEF2C] cells into cardiac muscle, and therefore P19[MEF2C] cultures treated with Me2SO contained both cardiac and skeletal muscle on day 9 (data not shown). The conditions required and the extent of differentiation into skeletal muscle were quantitated by counting myogenin and MF20-positive cells under each culture condition. The development of substantial numbers of skeletal myocytes (5–6% of total cells) required MEF2C expression and nine days of differentiation after aggregation with Me2SO (Fig. 4). The number of MF20 positive cells is slightly lower than the number of myogenin-positive cells. The number of myocytes that reacted with MF20 may be an underestimate because only bipolar myocytes were counted. Alternatively, the number could be lower because of a slightly earlier expression of myogenin compared with myosin heavy chain. Northern blot analysis was used to examine the expression of muscle-specific genes. Total RNA from P19 and P19[MEF2C] cultures was harvested on day 6 and day 9 after aggregation with Me2SO. Transfected MEF2C transcripts were expressed in P19[MEF2C] cells (Fig. 5 A, lanes 4–6) but not in P19 control cells (Fig. 5 A, lanes 1–3). Interestingly, the expression of exogenous MEF2C transcripts seemed to result in the up-regulation of endogenous MEF2C expression. The expression levels of MEF2A, MEF2B, and MEF2D were very low and did not appear to change in any of the conditions examined (data not shown). The formation of MEF2C-induced skeletal muscle was demonstrated by the expression of myogenin on day 9 in P19[MEF2C] cells but not in P19 control cells (Fig. 5C, lane 6 compared with 3). Low levels of MyoD were barely detectable in MEF2C-induced muscle (data not shown). Myf-5 and MRF4 expression were not detected under these conditions (data not shown). Aggregation activated cardiac α-actin expression by day 6 in both P19[MEF2C] and P19 cells, indicating the formation of cardiac muscle in both populations (Fig. 5 B,lanes 2 and 5). The level of cardiac α-actin expression appears lower in P19[MEF2C] cultures compared with P19 cultures on day 6 (Fig. 5 B, lanes 2 and5), however this result was variable. Thus, in agreement with the immunofluorescent staining, the Northern blot analysis results support the observation that MEF2C initiates skeletal muscle development. Because MEF2C is known to act synergistically with the MRFs, semi-quantitative RT-PCR was used to determine MRF expression levels under each condition. Substantial expression of myogenin was observed only in day 9 P19[MEF2C] cultures treated with Me2SO (Fig. 6 A), with an average of 130 ± 30-fold over background levels (n = 4). Similarly, increased levels of MyoD were observed only in day 9 P19[MEF2C] cultures treated with Me2SO (Fig. 6 B) with an average of 5 ± 1-fold over background levels (n = 4). Therefore, myogenin expression was activated an average of 26-fold more efficiently than MyoD expression. The low levels of MyoD expression detected in P19 and P19[MEF2C] cultures (Fig. 6 B, lanes 1–5) were not present in the negative control RT-PCR reactions performed (data not shown). No evidence of increased levels of MyoD or myogenin was found in P19 control cells treated with Me2SO (Fig. 6, A andB, compare lanes 1, 3, and5). Therefore, these results show that MEF2C up-regulates the expression of myogenin considerably more efficiently than that of MyoD. Although, the presence of a regulatory loop between the MRFs and MEF2 families has been shown previously (14Cserjesi P. Olson E.N. Mol. Cell. Biol. 1991; 11: 4854-4862Crossref PubMed Scopus (199) Google Scholar, 15Yee S.P. Rigby P.W. Genes Dev. 1993; 7: 1277-1289Crossref PubMed Scopus (347) Google Scholar, 16Cheng T.C. Wallace M.C. Merlie J.P. Olson E.N. Science. 1993; 261: 215-218Crossref PubMed Scopus (220) Google Scholar), the MEF2 family member involved in this regulation was not identified. Our results show that, whereas both MyoD and myogenin induced skeletal myogenesis in aggregated P19 cells, myogenin, and not MyoD, up-regulated the expression of MEF2C during differentiation. Because MEF2A, -B, and -D expression levels remained unchanged, the ability of MEF2C to trigger skeletal muscle was examined. MEF2C initiated skeletal myogenesis in P19 cells aggregated with Me2SO. During this process, myogenin expression was up-regulated to a far greater extent than MyoD. Taken together, our results support the presence of a positive regulatory loop between the MEF2 family and the MRFs and indicate that MEF2C and myogenin, but not MyoD or MEF2A, -B, or -D, participate in this loop in differentiating stem cells. In the murine embryo, MEF2C is expressed shortly after myogenin on day 8.5 (27Edmondson D.G. Lyons G.E. Martin J.F. Olson E.N. Development. 1994; 120: 1251-1263Crossref PubMed Google Scholar). Our finding that myogenin and MEF2C up-regulate each other's expression in P19 cells is consistent with the timing of their expression during embryogenesis. The presence of a positive regulatory loop could enhance the expression of both myogenin and MEF2C in the somites. Similar to MyoD and myogenin, cellular aggregation was required for the full activity of MEF2C. It is likely that proteins expressed as a result of cellular aggregation regulate MEF2C activity. The lack of up-regulation of MRF expression during the cellular aggregation of control cells suggests that MEF2C may not be simply acting synergistically to amplify low levels of these tissue-restricted transcription factors. Our results agree with the finding that ectopic expression of MEF2 in Drosophila epidermis activates the expression of skeletal muscle genes (30Lin M.H. Bour B.A. Abmayr S.M. Storti R.V. Dev. Biol. 1997; 182: 240-255Crossref PubMed Scopus (43) Google Scholar). This activation was found to be stage-dependent, suggesting that a factor is expressed in Drosophila epidermis that can regulate MEF2C function. Factors that regulate MEF2C may be other transcription factors, kinases, or phosphatases. MEF2C can be regulated by phosphorylation events because of interactions with casein kinase-II (61Molkentin J.D. Li L. Olson E.N. J. Biol. Chem. 1996; 271: 17199-17204Abstract Full Text Full Text PDF PubMed Scopus (90) Google Scholar), mitogen-activated protein kinase p38 (62Han J. Jiang Y. Li Z. Kravchenko V.V. Ulevitch R.J. Nature. 1997; 386: 296-299Crossref PubMed Scopus (676) Google Scholar, 63Zetser A. Gredinger E. Bengal E. J. Biol. Chem. 1999; 274: 5193-5200Abstract Full Text Full Text PDF PubMed Scopus (382) Google Scholar, 64Zhao M. New L. Kravchenko V.V. Kato Y. Gram H. di Padova F. Olson E.N. Ulevitch R.J. Han J. Mol. Cell. Biol. 1999; 19: 21-30Crossref PubMed Scopus (373) Google Scholar), and extracellular signal-regulated kinase 5 (65Yang C.C. Ornatsky O.I. McDermott J.C. Cruz T.F. Prody C.A. Nucleic Acids Res. 1998; 26: 4771-4777Crossref PubMed Scopus (141) Google Scholar). These factors are candidates for regulating MEF2C-induced skeletal myogenesis. The difference between the results reported here in P19 cells and the contradictory results from studies in fibroblasts is probably because of the expression of the appropriate regulatory molecules in aggregated P19 cells, which may or may not be expressed in fibroblasts. Because P19[MRF] cells produce abundant skeletal muscle (>30%) and P19[MEF2C] cells produce only moderate amounts of skeletal muscle (<10%), our results suggest that MyoD is regulated by factors expressed in a larger proportion of the total cell population than factors regulating MEF2C. These results agree with the finding that MyoD can readily initiate myogenesis in fibroblasts. The use of serum that does not support Me2SO-induced skeletal myogenesis in P19 cells may contribute to the ability of MEF2C to function in P19 cells. The exact mechanism by which factors in serum regulate P19 cell myogenesis is unknown (38Wilton S. Skerjanc I.S. In Vitro Cellular and Developmental Biology-Animal. 1999; 35: 175-177Crossref PubMed Scopus (27) Google Scholar). Because MEF2C is known to bind MEF2 sites and activate skeletal muscle-specific promoters, it is likely that MEF2C is functioning in a similar manner in P19 cells. However, MEF2C expression may be promoting myogenesis by activating the expression of other factors involved in myogenesis or cooperating with existing factors. Alternatively, it is possible that the overexpression of MEF2C may lead to myogenesis by sequestering negative regulatory elements. The phenotypes of MRF knockout mice clearly indicate that MRFs have distinct biological roles in controlling skeletal muscle development. However, the mechanisms that define these differences are still poorly understood. In fibroblasts, MyoD and myogenin tend to cross-activate each other's expression, making it difficult to determine their specific functions (59Olson E.N. Klein W.H. Genes Dev. 1994; 8: 1-8Crossref PubMed Scopus (606) Google Scholar, 66Arnold T.E. Worrell R.A. Barth J.L. Morris J. Ivarie R. Exp. Cell Res. 1994; 212: 321-328Crossref PubMed Scopus (8) Google Scholar, 67Aurade F. Pinset C. Chafey P. Gros F. Montarras D. Differentiation. 1994; 55: 185-192Crossref PubMed Scopus (47) Google Scholar, 68Braun T. Bober E. Buschhausen-Denker G. Kohtz S. Grzeschik K.H. Arnold H.H. Kotz S. EMBO J. 1989; 8: 3617-3625Crossref PubMed Scopus (252) Google Scholar, 69Thayer M.J. Tapscott S.J. Davis R.L. Wright W.E. Lassar A.B. Weintraub H. Cell. 1989; 58: 241-248Abstract Full Text PDF PubMed Scopus (343) Google Scholar). Nevertheless, several studies have shown differences in the ability of MRFs to transactivate both exogenous (52Asakura A. Fujisawa-Sehara A. Komiya T. Nabeshima Y. Nabeshima Y. Mol. Cell. Biol. 1993; 13: 7153-7162Crossref PubMed Scopus (40) Google Scholar, 54Berberich C. Durr I. Koenen M. Witzemann V. Eur. J. Biochem. 1993; 216: 395-404Crossref PubMed Scopus (40) Google Scholar, 57Chakraborty T. Brennan T. Olson E. J. Biol. Chem. 1991; 266: 2878-2882Abstract Full Text PDF PubMed Google Scholar, 70Yutzey K.E. Rhodes S.J. Konieczny S.F. Mol. Cell. Biol. 1990; 10: 3934-3944Crossref PubMed Scopus (112) Google Scholar) and endogenous promoters (60Gerber A.N. Klesert T.R. Bergstrom D.A. Tapscott S.J. Genes Dev. 1997; 11: 436-450Crossref PubMed Scopus (230) Google Scholar). In the latter experiment, it was shown in fibroblasts that MyoD was 10-fold more efficient than myogenin in activating genes located in regions of transcriptionally silent chromatin. Our results in P19 cells indicate that myogenin was more proficient in activating endogenous MEF2C gene expression than MyoD. The role of chromatin modeling during myogenesis in P19 cells remains to be determined. Our finding that MHC isoforms were not differentially regulated agrees with previous results, which show no correlation between the expression of MRFs and adult muscle transcripts, like MHC, in mouse muscle cell lines (67Aurade F. Pinset C. Chafey P. Gros F. Montarras D. Differentiation. 1994; 55: 185-192Crossref PubMed Scopus (47) Google Scholar, 70Yutzey K.E. Rhodes S.J. Konieczny S.F. Mol. Cell. Biol. 1990; 10: 3934-3944Crossref PubMed Scopus (112) Google Scholar, 71Miller J.B. J. Cell Biol. 1990; 111: 1149-1159Crossref PubMed Scopus (108) Google Scholar). In contrast to the results obtained in tissue culture, myogenin is expressed at high levels in slow muscle fibers, and MyoD is selectively accumulated in adult rat fast muscle fibers (72Hughes S.M. Taylor J.M. Tapscott S.J. Gurley C.M. Carter W.J. Peterson C.A. Development. 1993; 118: 1137-1147Crossref PubMed Google Scholar). Because MRFs act via E-box elements, further experiments are required to determine the mechanism by which some E-boxes are differentially regulated by MyoD and myogenin, whereas other E-boxes are not. It seems likely that the N- and C-terminal regions of the MRFs are involved, as shown previously for the differential activation of exogenous promoters (41Mak K.L. To R.Q. Kong Y. Konieczny S.F. Mol. Cell. Biol. 1992; 12: 4334-4346Crossref PubMed Scopus (34) Google Scholar, 52Asakura A. Fujisawa-Sehara A. Komiya T. Nabeshima Y. Nabeshima Y. Mol. Cell. Biol. 1993; 13: 7153-7162Crossref PubMed Scopus (40) Google Scholar, 56Chakraborty T. Olson E.N. Mol. Cell. Biol. 1991; 11: 6103-6108Crossref PubMed Google Scholar). These domains could interact with other factors that may be involved in determining specificity. In summary, examining the ability of transcription factors to modulate the developmental potential of P19 cells provides a powerful tool for subsequent analysis of the mechanisms involved. Using this system, we have shown that MEF2C and myogenin induce myogenesis and participate in a positive regulatory loop in P19 cells. We thank Judy Ball and Peter Merrifield for reading the manuscript and helpful discussions. We thank Shu-ichi Okamoto and Dmitri Krainc for the PGK-MEF2C construct and Eric Olson and Jeffery Molkentin for MEF2-A, -B, -C, and -D cDNAs.