University of Calgary PRISM: University of Calgary's Digital Repository
Graduate Studies Legacy Theses
2000 Regulation of skeletal muscle differentiation by ATBF1, a multiple homeodomain-zinc finger transcription factor
Berry, Fred Brandon
Berry, F. B. (2000). Regulation of skeletal muscle differentiation by ATBF1, a multiple homeodomain-zinc finger transcription factor (Unpublished doctoral thesis). University of Calgary, Calgary, AB. doi:10.11575/PRISM/23689 http://hdl.handle.net/1880/39899 doctoral thesis
University of Calgary graduate students retain copyright ownership and moral rights for their thesis. You may use this material in any way that is permitted by the Copyright Act or through licensing that has been assigned to the document. For uses that are not allowable under copyright legislation or licensing, you are required to seek permission. Downloaded from PRISM: https://prism.ucalgary.ca NOTE TO USERS
Page(s) not included in the original manuscript and are unavailable from the author or university. The manuscript was microfilmed as received.
This reproduction is the best copy available.
THE UNIVERSITY OF CALGARY
Regulation of Skeletal MuscIe Differentiation by ATBF1,
a Multiple Homeodomain-zinc Finger Transcription Factor
Fred Brandon Berry
A DISSERTATION SUBMITTED TO THlE
FACC?TYOF GRADUATE STUDIES
IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
DEPARTMENT OF BIOCHEMISTRY AND MOLECULAR BIOLOGY
CALGARY, ALBERTA
APRIL, 2000
O Fred Brandon Berry 2000 National Library Bibliotheque nationale I+IofCanada du Canada Acquisitions and Acquisitions et Bibliographic Services services bibliographiques 395 Wetlington Street 395. rue Wellington Ottawa ON KIA ON4 Ottawa ON KIA ON4 Canada Canada Your fib Worm rdfarmce
Our fiIe Ndrro fel(Irenc8
The author has granted a non- L'auteur a accorde me licence non exclusive licence allowing the exclusive permeftant a la National IArary of Canada to Bibliotheque nationale du Canada de reproduce, loan, distribute or sell reproduire, prster, distribuer ou copies of this thesis in microform, vendre des copies de cette these sous paper or electronic formats. la forme de microfiche/fih, de reproduction sur papier ou sur format electronique.
The author retains ownership of the L'auteur conserve la propriiete du copyright in this thesis. Neither the droit d'auteur qui protkge cette these. thesis nor substantial extracts from it Ni la these ni des extraits substantiels may be printed or otherwise de celle-ci ne doivent Etre knprimks reproduced without the author's ou autrement reproduits sans son permission. autorisation. Abstract
The ATBFl gene encodes two protein isoforms, the 404 kDa ATBF1-A, possessing four homeodomains and 23 zinc fmgers, and the 306 kDa ATBF1-B, lacking a 920-amino acid N-terminal region of ATBFI-A. We found that ATBF1-A was expressed in proliferating CZC 12 myoblasts but its expression levels decreased upon induction of myogenic differentiation in low serum medium. This down- regulation of ATBF la is one of the requirements for muscle differentiation. Cells constitutively expressing ATBF1 -A fail to differentiate as characterized by aberrant expression patterns of myogenic regulatory transcription factors, cell cycle regulatory proteins, and muscle structural proteins. In addition, these cells were characterized by high levels of expression of Id3 arid cyclin D 1, known to inhibit
C2C 12 dBerentiation. In contrast, transfection of C2C 12 cells with the ATBF 1-B isoform resulted in an acceleration of myogenic differentiation, as indicated by an earlier onset of MHC expression and the formation of a higher percentage of multinucleated myotubes. In 10T1/2 cells, transfection of ATBF 1-A prevented
MyoD-dependent myogenic conversion, whereas ATBF 1B enhanced this process.
In addition this study identified the mouse Mm4 gene as a target of ATBF 1 regulation. A region of ATBFl-A containing homeodomain 4 interacted with an
AT-rich element overlapping the El E-box of the MRF4 promoter and displaced myogenin from this El site. To inhibit the MyoD-dependent activation of the
MRF4 promoter by ATBF 1-A, the binding to these sites was not sufficient, but two
iii segments of the ATBF1-A-specific N-terminal region were also required. We show that the ATBF1-A-specific regions possess transcription repressor activity. These results show that the ATBFl isoforms regulate myogenesis in an opposite manner.
The down-regulation of ATBF 1-A expression occurs in response to the removal of growth factors from the mediun and is a prerequisite to initiate terminal differentiation of C2C 12 cells. ATBF 1-A-expressing cells remain undifferentiated but enter a reversible quiescence in response to DM, which is a feature common to muscle satellite and reserve cells. Therefore, ATBFl-A may divert myoblasts from entering terminal differentiation and may promote &se cells into adopting a reserve cell identity. Acknowledgments
I wish to express my gratitude to my supervisor. Dr. Taiki Tarnaoki for his support, patience and encouragement throughout the duration of this project. I thank him for the opportunity to undertake this research. allowing me the fieedom to explore the topics in which I was interested, and for stressing the foundations for good scientific research. I would also like to extend my appreciation to the members of my supervisory committee. Dr. Gilbert Schultz, Dr. Frans van der
Hoorn and Dr. Henk Zwien. for their helpfd comments and suggestions.
I would like to thank my lab members, both past and present. Dr. Yutaka
Miura. Dongping Ma. Dr. Koichiro Mihara, Dr. Petr Kaspar, and Dr. Kazuo
Fushimi for their helpful discussions and assistance. 1 thank Howard Chen for his technical assistance. I am grateful for the many friends that I have made throughout the duration of my stay in Calgary. In particular I thank Robbie
Loewith and Chris Howlett for their practical applications of yeast. I extend my thanks to my family for their continuous support. Finally to Misbah Qureshi, I am indebted to her for her companionship and encouragement.
I would like to acknowledge financial support provided through Faculty of
Graduate Studies Research Scholarships and from grants awarded to my supervisor from the National Cancer Institute of Canada. Table of Contents
Approvalpage ...... ii
Abstract ...... iii
AcknowIedgments ...... v
Table of Contents ...... vi
ListofFigures ...... x
ListofTables ...... xii
List of Abbreviations ...... xiii
1. Introduction ...... 1
1.1 ZFHproteins regulate cellular development and differentiation...... 2
1.1.1 Class I ZFH proteins ...... 2
1.1.2 Class II ZFHproteins ...... 4
1 -2ATBF 1 is a zinc-finger homeodomain transcription factor ...... 5
1 .2.1 ATBF 1 structural-function relationships ...... 6
1.2.2 Comparison between ATBF1-A and B isoforms...... 10
1.2.3 The homeodomains ...... 11
1.2.4 The zinc-fmgers ...... 13
1.3 Potential functions for ATBF 1 in embryonic development...... 14
1.4 Regulation of vertebrate skeletal muscle differentiation...... 17
1.4.1 The myogenic differentiation program ...... 18 1.4.2 Induction of MRF expression ...... 20
1.4.3 Molecular mechanisms of MRF activity...... 22
1.4.4 Integration of muscle differentiation and cell cycle regulation...... 26 1.5 Rationale for current research ...... 29
2 . Materials and Methods ...... 32
2. 1Plasmids...... 32
2.2 Cell Cultures and Stable Transfections...... 32
2 -3 Southern Blotting ...... 35
2.4 mRNA expression analysis ...... 37
2.4.1 RNA Isolation...... 37
2.4.2 RNase protection assays ...... 38
2.4.3 Northern Blotting ...... 39
2.5 Irnrnunocytochemistry...... 40
2 -6 BrdU incorporation ...... 41
2.7 Western Blotting...... 42
2.8 Myogenic conversion assays ...... 44
2 -9 Luciferase reporter assays ...... 44
2.10 Mammalian one-and two-hybrid assays ...... 45
2.1 1 DNA-protein interactions...... 46
2.1 1.1 Nuclear Extract Isolation ...... 46
2.1 1.2 Probe preparation...... 47 2.11.3 In vitro translation of proteins ...... 48
2.1 1.4 Bacterially expressed proteins...... 48
2.11 -5 ElecQophoretic Mobility Shift Assays (EMSAs) ...... 49
2.1 1.6 W cross-linking of proteins to nucleic acids...... 49
2.12 Immunoprecipitation and His-tag pull-down assays ...... 50
2.13 Tibialis anterior regeneration ...... 52
3.1 Expression of ATBFI -A during myogenic differentiation...... 53
3.2 ATBF 1-A inhibits myogenic differentiation of C2C 12 cells ...... 57
3 -3 ATBF 1-A expressing C2C12 cells undergo growth arrest in response toDM...... 72
3.4 The ATBF1-B isoform enhances myogenic differentiation of C2C12 cells ...... 72
3-5 MyoD induced trans-differentiation of 10T1/2 fibroblasts to myotubes. is inhibited by ATBFl-A and enhanced by ATBF l.B ...... 80
3.6 Activation of the MRF4 promoter by MyoD is inhibited by ATBF1.A ...... 83
3 -7 An AT-rich motif is necessary for ATBF 1-A-mediated repression .... 86
3.8 ATBFI-A horneodomain 4 binds to the AT-rich element of the MRF4 promoter ...... 91 3.9 The AT-rich MEF2 element represents an additional ATBF1-A binding site...... 109
3.10 The ATBF 1 -A amino-terminus is necessary but not sufficient to inhibit MyoDactiviq ...... 112
3.1 1 ATBF 1-A is a transcriptional repressor...... 116 3.12 A tethered MyoD-E47 protein is resistant to ATBF1-A inhibition of MRF4 promoter ...... 117
3.13 The myogenic differentiation stimuli promotes the down-regulation of ATBF1-A expression in C2C12 ...... 130
4.1 Ectopic ATBF 1-A expression inhibits myogenic differentiation .... 141
4.1.1 ATBFl-A perturbs expression of positive and negative regulators of muscle differentiation...... 141
4.1.2 Mechanisms of MRF4 promoter inhibition...... 143
4.1.3 DNA binding properties of ATEiF1.A ...... 145
4.1.4 ATBF 1-A-specific functional domains...... 153
4.1.5 Possible mechanisms of ATBF I -A-mediated transcriptional repression...... 154
4.1.6 A comparison with myogenic inhibitory transcription factors ...... 158
4.1.7 Differential activity of ATBF 1 isofoms...... 159
4.2 ATBFl-A alters cell cycle progression during C2C12 rnyogenic differentiation and promotes the reserve cell fate...... 161
4.3 Regulation of ATBFl-A expression by growth promoting stimuli... 163
4.4 Future Directions ...... 170
4.4.1 Expression of ATBFl-A during myogenic differentiation in vivo ...... 170
4.4.2 Identification of ATBF 1-A targets of transcriptional regulation ...... 173 4.5Summary ...... 174 5.References...... 177 List of Figures
Figure 1. Potential functional domains for ATBF 1.
Figure 2. Expression of ATBF 1-A mRNA in differentiated and undifferentiated muscle tissues.
Figure 3. Conformation of ATBF 1-A stable transfected C2C 12 cell lines.
Figure 4. Constitutive expression of ATBF 1-A inhibits myogenic differentiation of C2C 12 cells.
Figure 5. Alteration in the expression patterns of myogenic and cell cycle regulatory factors in ATBF 1-A transfected cells.
Figure 6. Western blot analysis of ATBF1 -A protein in differentiating C2C12 cells.
Figure 7. BrdU incorporation profdes of differentiating and serum- stimulated C2 C 12 cells.
Figure 3. Enhanced myogenic differentiation in ATBF1-B transfected C2C12 cells.
Figure 9. Early onset of MHC expression detected in ATBFI-B- transfected cells.
Figure 10. Normal expression patterns of differentiation markers in ATBF 1-B expressing cells.
Figure 11. ATBF 1 regulates the MyoD-mediated myogenic conversion of 10T1/2 cells.
Figure 12. ATBF 1-A inhibits MyoD-dependent transcriptional activation of the MRF4 promoter.
Figure 13. The E-box element is resistant to ATBF 1-A repression.
Figure 14. The fourth homeodomain of ATBFI-A binds to the AT richE-box element of the MRF4 promoter. Figure 15. W-crosslinking of HD4 binding complexes to the MRF4 A/E probe.
Figure 16. A muscle differentiation-specific complex binds the MRF4 A/E probe.
Figure 17. HD4 binding to the AIE probe disrupts myogenin-E-box interactions.
Figure 18. Displacement of myogenin-E-box interactions is specifc to HD4 of ATBF1-A.
Figure 19. HD4 binds to the AT-rich portion of the MRF4 promoter.
Figure 20. The AT-rich MEF2 element may represents a novel target for HD4 binding.
Figure 2 1. The ATBF1-A N-terminal and DNA binding domain is necessary inhibition of MyoD- induced activation of MRF4 promoter.
Figure 22. ATBF1-A contains a transcription repressor domain.
Figure 23. Forced MyoD-E47 dirners can overcome ATBF1-A-mediated inhibition of MRF4 promoter.
Figure 24. Analysis of MyoD-ATBF 1-A interactions by mammalian two-hybrid assays.
Figure 25. ATBF1-Adoes not interact withMyoD.
Figure 26. The down-regulation of ATBF 1-A expression in C2C12 cells occurs independent of muscle differentiation.
Figure 27. Serum growth factors inhibit C2C 12 myogenic differentiation and modulate ATBFI-A expression patterns.
Figure 28. Induction of ATBFl-A mRNA expression following a crush injury to the tibialis anterior muscle. List of Tables
Table 1. Classification of ZFH proteins. 3
Table 2. Comparison of ATBFl-A binding element with AT-rich motifs. 147 List of Abbreviations
AFP alpha-fetoprotein
AT-rich Adenosine and thymidine rich bFGF Basic fibrobIast growth factor p-Gal Beta-galactosidase bHLH Basic helix-loop-helix
BMP Bone morphogenetic protein
BrdU 5Bromo-deoxpridine
CBP Creb-binding protein
Cdk Cyclin dependent kinase
CMV Cytomegalovirus
CNS Central nervous system
DB DNA-binding domain
DM Differentiation medium
DMEM Dulbecco 's modified Eagle medium dpc Days post-coitum EMSA Electrophoretic mobility shift assay
GAPDH Glyceraldehyde-3-phosphate dehydrogenase
GM Growth medium
GST Glutathione-S-transferase
HA Hemaggl u tinin HAT Histone acelyltransferase
HD Homeodomain
HLH Helix-loo p-helix
IgG Immunoglobulin G LPA Lysophosphatidic acid luc Luciferase
MAP kinase Mitogen activated protein kinase
MASH-1 Mammalian achaete-scute homolog- 1
MCK Muscle creatine kinase
MHC Myosin heavy chain
MRF Myogenic regulatory factors
Muscle precursor cell
Nuclear corepressor pgk Phosphoglycerate kinase
PRB Retinoblastoma protein RA Retinoic acid sm Sonic hedgehog
SV40 Simian virus 40
TAD Transactivation domain
TGF-P Transforming growth factor-beta tk Thymidine kinase
ZFH Zinc finger and homeodomain
xiv 1. Introduction.
One of the central questions key to the formation of complex body plans is
how do the diverse cell types that compose tissues and organs, originate from a single cell, the fertilized egg? At the onset of fertilization, this cell has the capacity
to form every cell type. As the embryo develops, the ultimate fate of these cells
becomes more and more restricted with each cell division. Many of the decisions
that limit developmental capacity of these cells are driven by changes in gene
expression. Thus transcriptional regulatory proteins have been demonstrated to be key components in the generation of cell diversity.
Mutations in genes encoding homeodomains and zinc finger proteins can have severe developmental implications and the importance of these classes of transcription factors in patterning the metazoan body plan is well documented
@erg, 1990; Biggin and Tjian, 1989; Hollemann et al., 1996; Hunt and Krurnlauf,
1992 ; Krempler and Brenig, 1999; McGinnis and Krurnlauf, 1992). Recently, a novel family of transcription factors containing both homeodomains and zinc- fingers has been identified (Hashimoto et al., 1992). Although these zinc finger- homeodomain (ZFH) proteins have not been extensively studied, emerging data is demonstrating that they may play important roles in controlling developmental processes in both vertebrates and invertebrates (Funahashi et al., 1993; keda et al..
1998; Kostich and Sanes, 1995; Lai et d.,1991; Lai et al., 1993; Lundell and
Hirsh, 1992; Miura et al., 1995: Morinaga et al., 1991; Postigo and Dean, 1997). 2
1.1 ZFH proteins regulate cellular development and differentiation.
ZFH proteins have been idenmed in a number of species including humans, mice, chickens and Drosophila. The organization of these motifs along the protein is conserved between species and fall into two classes based on the number and position of the homeodomain(s). Class I proteins contain a single, centrally positioned homeodomain, flanked by zinc fmgers at the N- and C-termini. In
Class I1 proteins. multiple homeodomains (3 to 5) are located at the C-terminal. with the zinc-fingers predominately clustered at the N-terminal. Although these proteins share features in their structural organization, there is variation in the number of the zinc fmgers and homeodomain present in these proteins (Table 1).
1.1.1 CIass I ZFH proteins
Many of ZFH proteins are developmentally expressed and are involved in the differentiation of cell types derived from the neuroectoderm and mesoderm lineages. In Drosophila. ZFH-1 is expressed in mesoderm of early embryos and later in adult muscle precursor cells and motor neurons. Alterations in &-I expression in gain-of-function and loss-of-function mutants Leads to defects in the developing CNS and musculature. respectively (Lai et al.. 199 1; Lai et al.. 1993;
Postigo et al., 1999). The vertebrate factors, GEFI . ZEB and AREB6 are orthologs of Drosophila zfh-I, as they share extensive sequence similarity as well Table 1 : Classification of ZFH proteins.
Zinc Homeo- Size Species Reference fingers domains (ma)
ZFH- 1 Drosophila 9 1 117 (Fortinietal.,1991) SEF 1 Chicken 7 1 123 (Funahashietal., 1993) AFEB6/ Human 7 1 126 (Genetta et at,, 1994; ZEB Watanabe et al., 1993) Nil-2 -a Human 3 1 84 (Williams et al., 199 1) mSEF 1 Mouse 7 1 123 (Genetta et al., 1994)
Class 11 ZFH-2 Drosophila 16 3 332 Fortini et d.,199 1) chATBF 1* Chicken N/A N/A N/A (Kaspar et al., 1999)
ATBF 1-B Human 17 4 306 (Morinaga et al.. 1991) ATBF 1 Mouse 23 4 406 &do et al., 1996) ZFH-4 Mouse 22 4 390 (Kostich and Sanes, 1995) (N. Sakata, T. Tamaoki, and M, Hori, personal communication) ZHX- 1 Mouse 2 5 96 (Barthelemy et al., 1996) *A full length clones for the chicken homolog of ATBFl has not been identified. Its existence is based on expression studies. 4 as conservation in the positioning of the zinc-fmger and homeodomain motifs. In fact, mouse ZEB and Drosophila zfh-1 are functionally interchangeable in vivo
(Postigo et al.. 1999). In addition they display similar expression patterns. GEFI is predominately expressed in the developing chicken in tissues derived from mesoderm and neuroectoderm (Funahashi et al., 1993). while expression of
GEF 1/ZEB/AREBG is prominent in muscle tissues as well (Funahashi et al., 1993 ;
Postigo and Dean. 1997; Watanabe et al., 1993). Moreover, GEFl/ZEB/AREBG are transcriptional repressor proteins that regulate muscle. and B-and T-cell differentiation in vih-o and in vivo (Genetta et al.. 1994; Higashi et al.. 1997;
Postigo and Dean, 1999; Postigo and Dean. 1997; Sekido et al., 1994: Takagi et al., 1998; Williams et al., 199 1).
1.1.2 Class 11 ZFH proteins.
Vertebrate ZFH proteins containing multiple homeodomains may also play important roles in development. However there is less known about these factors compared to class I members. Class I1 proteins may differ from Class I in that multiple members of the second class may exist within a single species. Three distinct mouse genes encoding ZFH proteins have been documented: ATBFI.
ZFH-4 and 2.-I (Barthelemy et al., 1996; Ido et al.. 1996; Kostich and Sanes.
1995). ATBFI (AT-rich biding factor) was the first member identified (Morinaga et al., 1991). It is as transcriptional repressor of the human a-fetoprotein (AFP) 5
gene and may represent an ortholog of the DrosophiIa zfh-2gene (Morinaga et al.,
199 1; Yasuda et al., 1994). Subsequent orthologs have been detected in the mouse
and chicken genomes, (Ido et al.. 1996; Kaspar et al., 1999; Miura et al., 1995;
Morinaga et al.. 199 1). Expression of Drosophila zfh-2 is limited to the CNS and
may regulate the expression of enzymes required for catecholamine biosynthesis
(La.et al.. 199 1; Lundell and Hirsh. 1992). In the mouse, ATBFl is also expressed
in the developing CNS (Watanabe et al., 1996) and may regulate neural
differentiation events (discussed below). A partial clone encoding ZFH-4 was
initially isolated containing 2 homeodomains and 2 zinc-fingers (Kostich and Sanes,
1995). Like many ZFH proteins. ZFH-4 is expressed in developing muscle and
CNS, and may play a role in the formation of these tissues (Kostich and Sanes,
1995). Recently the full length ZFH-4 clone has been identified and it encodes a protein predicted to contain 22 zinc fmgers and 4 homeodomains that shares extensive similarities with ATBFl (N. Sakata, T. Tarnaoki, and M. Hori, personal communication). Thus ATBFI and ZFH-4 may represent paralogous genes.
ZHX- 1 possesses five horneodomains and only two zinc fingers (Barthelemy et al.,
1996). It is expressed in a wide range of tissues in the adult mouse, but is preferentially expressed in the brain. Factors related to ZFH-4 have not been detected in humans or chickens.
1.2 ATBFl is a zinc-finger homeodornain transcription factor.
The ATBF1 transcription factor was originally identified as a regulator of 6 the human AFP gene (hdorinaga et al., 1991). A single ATBFl gene encodes two protein isoforms termed ATBFl-A and -B, that contain multiple-homeodomains and zinc fmger motifs. The larger ATBFl-A isofom is a 404 kDa protein consisting of 23 zinc fmgen and four homeodomains (Miura et al., 1995).
ATBF1 -B is a 306 kDa protein which carries the same 4 homeodomains but 5 less zinc finger motifs due to the absence of 920 amino acid residues at the N-terminus
(Miura et al., 1995; Morinaga et al., 1991). Apart from this unique N-terminus, the remainder of the two proteins are identical in their amino acid composition.
Throughout this report ATBF1 will be used to refer to both protein isoforms.
1-2.1 ATBF 1 structure-function relationships.
The potential DNA-binding motifs of ATBF1 are arranged in a co-linear manner dong the length of the protein with the zinc fingers primarily localized at the N-terminal and the four homeodomains (IFID 1-4) located at the C-terminal region (Figure 1). The DNA binding activity of ATBF1 to the AFP enhancer has been attributed to HD4 (Yasuda et al., 1994). Prior to this study, the only known target for ATBFl regulation was the AT-rich enhancer of the AFP gene (Morinaga etal., 1991; Yasudaetal., 1994). ATBFl-B binds to this region andrepresses transcriptional activation from the AFP promoter. Only the ATBF1-B isoform was identified at that time and effect of the ATBF1-A isofom on this promoter is not known. Since other ATBFl targets have yet to be discovered, it is unknown whether the remaining horneodomains or zinc fmgers bind to DNA. The Figure 1. Potential bctional domains for ATBFl.
ATBFl possesses several functional domains based on the predicted amino acid sequence. The four ATBFl homeodornains (rectangles) are positioned near the
C-terminal region while the majority of the zinc fingers (black ovals) are clustered near the N-terminal region. The position and length of ATBFl-A and -B are indicated by the thick bars. In addition, there are several regions rich in acidic amino acids (Ac) , proline (E') , or glutamine (G) residues that may be involved with transcriptional regulation. DEAD and DEW elements correspond to regions of homology to RNA and DNA helicases, respectively. The ATP binding site of
ATBF 1 is located N-terminal to the fourth homeodomain (adapted from Muira et al., 1995). ATP Binding Site I
DEAH DEAD
ATBFI-A
Homeodomain @ Glutamine-rich region t
1 Zinc finger Proline-rich region (J Acidic amino acid residues 9
homeodomains of DrosopMa ZFH-2 display differential DNA-binding activities. as HD2 and HD3 binds to the Ddc and Opsin promoters respectively (Fortini et al.. 1991 ; Lundell and Hinh, 1992). It is possible, however, that not all of these
DNA-binding motifs of ATBFl are functional. In the SEFl/ZEB/AREBG proteins, the zinc finger domains bind to DNA, but the HD does not (Ikeda and Kawakami,
1995).
In addition to the potential DNA binding motifs, there are several other potential functional domains located in ATBFl-A and B proteins. These include regions rich in acidic amino acids, as well as domains rich in proline or glutarnine residues that may play a role in transcriptional regulation (Han and Manley, 1993;
Madden et al., 1993; Mailly et d., 1996; Mitchell and Tjian, 1989; Miura et al.,
1995; Morinaga et al., 1991; Venot et al., 1998). To date, all data demonstrate that ATBFl functions as a transcriptional repressor protein; however the regions required for this activity have not been determined (Kataoka et al., 2000; Yasuda et al., 1994). Finally, an ATP-binding motif and lysine residues involved in ATP- binding, have been localized to a region N-terminal to HD4 (Kamps et al., 1984;
Lai et al., 1991). Recently this region has been demonstrated to possess DNA- dependent ATPase activity. Although the ~i~canceof this activity is not known,
ATPase activity may be involved in chromatin remodeling as well as providing the necessary energy required for gene transcription events (Kmgston and Narlikar.
1999)- 12.2 Comparison between ATBFl -A and B isoforms.
The two ATBF 1 isoforms arise through alternative splicing and the use of two independently regulated promoters (Miura et d., 1995). At the 5' region of the ATBFl gene, there are four exons that will form the basis of the 5' regions of
ATBFl-A and B mRNAs. Exons 1 and 2 contain the 5' noncoding sequence of
ATBF 1B and -A, respectively. Exon 3 contains the remaining ATBF 1-A noncoding region plus the ATBF1-A protein coding region. Exon 4 contains the start of the
ATBF 1-B coding region. Therefore the splicing of exons 2,3 and 4 wiU generate
ATBFI -A rnRNA, while the splicing of exons 1 and 4 will produce ATBFI B transcripts. The fiiand remaining exons are incorporated into both isoforms.
Thus the unique ATBF1-A N-terminus arises from information encoded in exon 3.
The 920 amino acid extension of the ATBF1-A N-terminus contains several putative functional domains which may result in these two isoforms possessing different activities (Miura et al., 1995). Five zinc fingers are Located in this ATBF 1-
A-specific region, as well as two acidic domains, a region rich in serine and threonine, and a region rich in proline residues. In addition, two sequences similar to the DEAD and DEAH consensus motifs of RNA and DNA helicases, respectively, are located in this region. This, together with the ATPase activity demonstrated for ATBF 1 -A, suggests that this protein may function as a helicase ; however such activity has yet to be determined. 1.2-3The homeodomains.
The homeodomain is an evolutionarily conserved 60 amino acid sequence that forms a DNA-binding domain composed of three a-helices (Callaerts et al.,
1997; Gehring et al., 1994; Gehring et al., 1994). Helix1 and Helix2 are separated by a 7 amino acid Ioop, while Helix2 and Helix3 are separated by three residues and form a helix-turn-helix motif. The homeodomain is defmed by consensus amino acid sequence derived from a compilation of homeodomain sequences (Gehring et al., 1994). Comparison between the four ATBF1 homeodomains reveal differences in their sequence identities ranging from 30 to
43% (Morinaga et al., 199 1). The sequence of HD 1 differs from that of the prototype homeodomain consensus in that it contains a Leu substitution for a highly conserved Arg residue at position 53 of the homeodomain. It is not known what effect this substitution has on HD1 function. Nevertheless, this residue is involved in DNA-binding (Gehring et al., 1994) and the change from a positively charged Arg residue to a non-polar Leu residue may alter the DNA binding specificity or affinity of HD 1.
The crystal structure for several homeodomains bound to DNA has been solved (Billeter et al., 1990; Billeter et al., 1993; Gehring et d., 1994; Klemm et al., 1994; Li et al., 1995; Otting et al., 1990; Passner et al., 1999; Qian et al.,
1994; Qian et al., 1992)- The data from these studies indicate that different homeodomains may interact with their appropriate DNA target in a similar manner. Helix 3 is the recognition helix and makes base-specific contacts with the major groove of the DNA target while the N-terminal arm contacts the minor groove. In ATBF 1, HD 1 differs from HD2-4 in the amino acid residue present at position 50 (R vs Q) in helix 3. The residue at this position in the homeodomain lies within the recognition helix and can determine the DNA binding specificity
(Gehring et al.. 1994). suggesting that HD 1 may have a different DNA-binding sequence preference compared to the other regions.
There is a greater degree of sequence similarity between the respective homeodomains of ATBFl orthologs than there is between homeodomains within the same protein. For example, there is a 77% identity between the fvst homeodomain in human ATBFl and the first horneodomain of DrosophiIa ZFH-2, including conservation of the atypical amino acid substitutions (Fortini et al., 199 1;
Hashimoto et al.. 1992). However. there is less than 38% similarity between HDl and HD2.3 and 4 of human ATBFl (Morinaga et al.. 199 1). This similarity between the corresponding homeodomains increases as the evolutionary origins of the species become closer. Mouse and human ATBFl share a 100% amino acid identity between the four corresponding HDs (Ido et al., 1996; Miura et al., 1995;
Morinaga et al., 199 1). The amino acid sequence similarities in the ATBFlEFH-2 horneodomains between proteins of different species is greater than that between the homeodomains within a protein, suggesting that this family may have originated from a common archetype protein. Furthermore, the multiple homeodomains may have arisen by ancient gene duplication events that occurred prior to the invertebrate-vertebrate divergence. However, as Drosophila ZFH-2 13 contains only 3 HDs, the fourth HD detected in ATBFl may not have arisen from a gene duplication event after the vertebratelinvertebrate divergence, since the similarity of HD4 between the remaining three HDs is low. Rather, this fourth homeodomain may have been lost over the course of Drosophila evolution (Lundell and Hirsh, 1992).
1.2.4 The zinc-fmgers.
ATBF1-A and B contain 23 and 17 zinc fmgers, respectively, that belong to the C2H2 class of zinc finger motifs. In this arrangement the binding of the zinc ion is co-ordinated by two Cys and two His, as compared the other predominate class (C2C2) in which the zinc ion is chelated by four Cys residues. However, the
18h zinc finger of ATBF1-A and zinc fmger 13 of ATBF 1-B, contain an atypical
'CCHC' sequence (Morinaga et al., 1991). Thls unusual motif is conserved in human AREB6, chicken 6EF. mouse ATBF 1, and ZFH-4 as well as Drosophila
ZFH-1 and 2 . The si@icance of this motif is not known, but it has been detected in retroviral proteins (Berg, 1990; Berg, 1990). The large number of zinc finger motifs may increase the DNA-binding specificity for these proteins by contacting several sites on the promoter simultaneously. Alternatively, by interacting with different promoter elements, the activity of these proteins may shift from transcriptional repressors to activators. For example, AREBG interacts with two consensus sequences, CACCTGT and GTTTC/G (Ikeda and Kawakarni.
1995), and promoters harbouring both sequences are subject to transcriptional 14 repression by AREBG, while those containing the fmt consensus are activated by
AREBG. It is suggested that binding to multiple elements evokes conformational changes in the protein that may influence its traqscriptional regulatory activity
(Ikeda and Kawakarni. 1995). However, as mentioned above, the DNA-binding activity for the ATBFl zinc finger elements has not been identifed.
Apart from their DNA-binding properties, zinc-fmger motifs have been demonstrated to mediate protein-protein interactions. For example the murine zinc fmger transcription factor GATA4 interacts, through its zinc-finger motifs, with the Nkx2.5 homeodomain protein to cooperatively activate transcription from the atrial natriuretic factor promoter (Lee et al.. 1998). In addition, the LIM- homeodomain proteins contain a zinc-binding structure &IM domain) that facilitates interactions with other proteins. This property of these proteins is of interest since ATBF 1 homeodomains display the greatest homology with LIM homeodomains (Morinaga et al., 1991). However the ATBFl zinc-fmger motifs and the LIM domains are quite different structurally. Nevertheless the ATBFl zinc-fingers may act in a similar manner as the LIM domains in mediating interactions with other proteins such as co-activators or co-repressors, and the nature of these interactions may influence the transcriptional regulalatory activity of ATBF I.
1.3 Potential functions for ATBFl in embryonic development.
Many of the ZFH transcription factor are preferentidy expressed in the 15 developing CNS and skeletal musculature, suggesting they have important roles in the development and differentiation of the neuroectoderm and mesoderm lineages.
Consistent with the proposed developmental functions, mouse ATBFl mRNA is detected at higher levels in embryonic tissues compared to adult structures (Ido et al., 1996; Watanabe et al., 1996). Although ATBFl was initially isolated from a hepatoma cell line. ATBFI transcripts are detected in a wide range of non-hepatic mouse tissues including brain, lung, heart, spleen and kidney &do et al., 1996). In fact, little ATBFl mRNA was detected in the adult and embryonic mouse liver.
The AT-rich enhancer of the human MP gene, to which ATBFl binds, is not conserved in the mouse AFP gene (Godbout et al., 1988; Godbout and Tilghrnan,
1988), suggesting that ATBFl functions beyond the regulation of AFP expression.
Comparison of the two ATBFl isofoms reveals that in all cell types examined, the
ATBFl-A mRNA is expressed at higher levels compared to that of the ATBF1-B isoform (Miura et al.. 1995; Watanabe et al., 1996). However, since hnctional differences between the two ATBFl isoforrns have not been examined, the sigruficance of this differential expression is unknown.
ATBFl -A is preferentially expressed in the developing mouse nervous system. ATBFl-A rnRNA is detected as early as 13 days post coitum (dpc) in the embryonic mouse brain (Watanabe et al.. 1996) and its expression levels subsequently decline throughout embryonic and postnatal development, reaching levels barely detectable in the adult brain. In situ hybridization analysis of a 15 dpc mouse brain reveals that ATBFl -A mRNA is predominantly localized to mid- and 16 hind-brain structures including the superior and inferior coIIiculi, thalamus,
hypothalamus, pons and medulla oblongata (Watanabe et al., 1996). This
expression pattern persists until birth (postnatal day 1). ATBF 1-A mRNA levels
then decline, and by postnatal day 2 1 ATBFl-A rnRNA is detected at low levels in
the inferior colliculus and thalamus (Watanabe et al., 1996) . The abundance of
ATBFl-A expression in the developing mouse brain suggests ATBF1-A protein may
play an important role in the development of midbrain regions.
The notion of ATBFl-A regulating neural development is further substantiated by its expression following retinoic acid @A) -induced neural differentiation of PI9 embryonal carcinoma cells (Ido et al., 1994; Miura et al.,
1995). P 19 cells are a pleuripotent cell line that can differentiate into derivatives of the three germ layers. mesoderm. endodem and ectoderm, when subjected to the appropriate stimuli (McBurney, 1993; McBurney et al., 1982; Rudnicki and
McBurney, 1987). Treatment of these cells with micromolar concentrations of RA will induce neuronal differentiation which is accompanied by expression of neuron- specific genes, such as neurofilament proteins, and by morphological changes. such as neurite outgrowth, characteristic of neuronal morphology (Rudnicki and
McBurney, 1987). ATBF 1-A mRNA is not detected in undifferentiated P 19 cells; however its expression is elevated in response to RA-induced differentiation (Ido et al., 1994; Miura et al., 1995). Expression of the ATBF1-B isoform is also induced in these cells, but to a lesser extent than ATBF 1-A. This activation is specific to neuronal induction rather than RA-treatment itself since F9 cells stimulated to 17 differentiate into endoderm cells by RA is not accompanied by ATBFl-A activation
(Miura et al.. 1995). Furthermore. P 19 cells stably transfected with antisense-
ATBF 1 fail to differentiate as indicated by reduced expression levels of neuronal- specific genes such as MASH1 and NF-L and an absence of neurite projections when treated with RA (Ma. M.Sc. Thesis 1997). Dimethyl sulfoxide @MSO)- induced muscle differentiation of P 19 cells is also accompanied by an increase in
ATBFI-A mRNA. but to a lesser extent than that observed for neuronal differentiation. The expression patterns of ATBFl-A in differentiating P 19 cells suggests a role for ATBF1 -A in the regulation of neuronal and possibly myogenic differentiation in vitro.
1.4 Regulation of vertebrate skeletal muscle differentiation.
Skeletal muscle development and differentiation is a multi-step progression that involves the commitment of paraxial mesoderm cells to the myogenic lineage, followed by the initiation of the differentiation program. the exit from the cell cycle and fmally expression of muscle structural genes and the assembly of the contractile apparatus. In vertebrates. essentially all skeletal muscle arise from condensations of presomitic mesoderm that form spherical structures known as sornites that lie on either side of the neural tube (Hogan et al., 1994). Over time. the ordered structure of these sornites will begin to delaminate as it begins the maturation process to form the sclerotome, myotome and dermatome. Cells from the sclerotome will ultimately form the ribs and vertebrae, while cells from the 18 dermatome will form the dennis of the skin. The myotome will sibdivide into epaxial mesoderm which will differentiate into the muscles of the back and hypaxial mesoderm which will give rise to the muscles of the limbs and body wall.
The processes that form skeletal muscles is probably one of the best understood differentiation events as extensive information is known about the molecular mechanisms that initiate and execute the differentiation program.
1-4.1 The mvogenic differentiation promam.
The differentiation events that form muscle cells in vivo are faithfully recapitulated in a number of cell lines that differentiate in vitro. From studies in these systems, a core network of transcription factors have been identified that regulate skeletal muscle differentiation. The MyoD family of muscle regulatory factors (MFWs) includes MyoD (Davis et al., 1987). myf-5 (Braun et al.. 1989). myogenin (Wright et al.. 1989) and MRF4 (Miner and Wold. 1990; Rhodes and
Konieczny, 1989), and are basic helix-loop-helix (bl3LE-I) transcription factors.
MyoD was initially identified by its ability to convert fibroblasts to myogenic cells that express skeletal muscle-specific myosin proteins (Davis et al., 1987).
Although each factor exhibits temporal and spatial differences in their expression patterns, all four of the MyoD-MRFs are expressed in the developing sornites to suggest they act as regulators of myogenic differentiation in avo. Myf-5 is the frst member to be expressed with transcripts appearing in the sornite by 8 dpc. and its expression is down-regulated by 14 dpc (Ott et al., 1991). Expression of 19 myogenin immediately follows that of myf-5 with myogenin mRNA detected by
8.5 dpc, followed by the transient appearance of MRF4 expression at 10 and 11 dpc, before reemerging at 16 dpc and persisting into the adult muscles (Sassoon et al., 1989; Lyons and Buckingham 1992). MyoD is expressed in rostra1 myotome at
10.5 dpc (Sassoon et al.. 1989). These expression patterns indicate that the MyoD members are expressed at the right time and place to regulate myogenic differentiation.
The MyoD-MRFs act in a hierarchal manner with each protein involved in a distinct stage of the myogenic program. First, MyoD and myf-5 are expressed to act in the determination of the myogenic lineage, followed by myogenin and MRF4 which execute terminal myogenic differentiation (reviewed in Arnold and Braun,
1996). In cultured muscle cells, MyoD and myf-5 mRNA are detected in the undifferentiated muscle precursor cells (myoblasts), while myogenin and MRF4 expression is not present until the cells have initiated the differentiation program
(Miller, 1990). Furthermore mice lacking both MyoD and myf-5 genes fail to form both myoblasts and myofibres (Rudnicki et al., 1993), while mice lacking myogenin form myoblasts, yet these cells are unable to differentiate into a mature muscle fibre (Hasty et al., 1993). MRF4 deficient mice display modest axial muscle defects (Patapoutian et al., 1995; Zhang et al., 1995). Surprisingly, mice lacking either MyoD or myf-5 develop an essentially normal skeletal musculature, indicating that these factors may compensate for one another (Braun et al., 1992;
Rudnicki et al., 1992). The effect of these mutations reveals that MyoD and myf-5 20 are important for the generation of myoblast cells while myogenin is necessary for the differentiation of myoblasts to the mature muscle fibre.
Although MyoD and myf-5 can compensate for the absence of the other, it is thought that these proteins are not performing redundant roles in muscle differentiation, but rather each factor may regulate the development of a specific muscle population (Ordahl and Williams, 1998). Within the somite, myf-5 and
MyoD are expressed in regions of the dermamyotome that will eventually form epaxial muscles and hypaxial muscles, respectively (Cossu et al., 1996; Smith et al.,
1994). Mice lacking MyoD or myf-5 develop essentially normal muscles.
However, a detailed examination of these mutant mice revealed that myf-54- display a delay in the formation of the muscles of the back and MyoD -/- mice exhibited delayed development of the limb muscles (Kablar et al.. 1997). Together these data suggest that MyoD and myf-5 control the specification of distinct muscle lineages.
1.4.2 Induction of MRF expression
Since MyoD-MRFs mark the fxst muscle-specific factors to be expressed in myogenic progenitor cells, there is great interest in determining how these factors are initially induced. Data from co-cultures of embryonic tissue explants experiments demonstrate that a combination of Sonic Hedgehog (SHH) signals from the notochord and Wnt signals from the neural tube, as well as basic fibroblast growth factor (bFGF) and transforming growth factor (TGF)-P signals 21
originating from the dorsal neural tube. can induce MyoD and myf-5 expression in
the developing somites (Borycki et al.. 1999; Miinsterberg et aI., 1995;
Miinsterbergand Lassar, 1995; Spence et al., 1996; Stem et al., 1995; Stem et al..
1997; Tajbakhsh et al., 1998). How these signals lead to the muscle-specific
transcriptional activation of the MyoD-family is unknown. The Pax-3 transcription
factor has been implicated as a positive regulator of MyoD expression and
myogenesis. Splotch mice carry a Pax3 mutation and display limb muscle defects
(Bo ber et al.. 1994). Furthermore splotch~myf-5double mutant mice Lack all body
muscles and fail to activate MyoD expression, whereas MyoD is expressed when
either gene alone is deleted, and ectopic Pax3 expression can induce MyoD
expression in non-myogenic tissues such as the neural tube (Maroto et al., 1997;
Tajbakhsh et al., 1997). The potential activation of MyoD by Pax-3 is interesting
since Pax3 expression in the developing spinal cord can be induced by SHH signals
originating from the notochord (Goulding et al.. 1993). However. there is no
evidence demonstrating direct activation of MyoD by Pax-3.
Several transcriptional regulatory regions upstream of the MyoD-MRF genes
have been characterized. Two regulatory regions have been identified in the
mouse MyoD gene, one 20 kb upstream and the other 5 kb upstream of the
transcription start site. that can individually direct muscle specific expression
(Asakura et al., 1995; Goldhamer et al., 1995; Kablar et al., 1999; Kucharczuk et al., 1999). The regulatory regions upstream of the myf-5 may prove difficult to identQ since this gene lies only 8 kb downstream of the MRF4 gene (Olson et al., 22 1996). Distal regulatov elements 50 kb upstream of the myf-5 gene can direct expression in the sornites and developing Limbs patapoutian et al., 1993;
Zweigerdt et al., 1997). The potential trans-acting factors which may regulate
MyoD and Myf-5 expression remain elusive. The complexity observed for the transcriptional regulation of MyoD and myf-5 may be required to ensure the correct spatiotemporal regulation as these proteins specify the myogenic lineage and are likely to be regulated by non-myogenic factors. Muscle-specific transcriptional activation of the rnyogenin gene in vitro and in vivo is directed by a small proximal promoter region containing two E-box elements as well as an
MEF2 binding sites (Cheng et al., 1993; Edrnondson et al., 1992). MRF4 expression is also controlled by a proximal promoter that it similar to rnyogenin in structure as it contains two E-box elements in addition to an MEF2 site (Black et al., 1995; Naidu et al., 1995). The differentiation class MRFs (myogenin and
MRF4) promoters may not need to be under as stringent a regulatory control as the determination class MRFs (MyoD and myf-5). since transcriptional activation of myogenin and MRF4 can be mediated by the upstream MyoD-MRFs.
1.4.3 Molecular mechanisms of MRF activitv.
The MyoD-regulatory proteins are a class of bHLH transcription factors.
They can directly activate transcription of genes that are expressed in a muscle- specific manner. The MyoD-MRF proteins form heterodimers with ubiquitously expressed bHLH proteins, known as E proteins, to bind to a consensus E-box sequence (CANNTG) found in the regulatory regions of these muscle-specific genes. Many of the promoter elements for the MyoD-MRF genes contain E-boxes themselves, and are subjected to activation from upstream MyoD-MRFs as well as positive-feedback regulation from those downstream.
The basic helix-loop-helix domain of MyoD-MRFs contains important structural features required for activity of these proteins. Based on the MyoD-
DNA crystal structure, the helix-loop-helix domain mediates protein-dirnerization with other HLH proteins, while the basic region interacts directly with the E-box
DNA element (Ma et al., 1994). Thus, inhibitory HLH proteins such as Id can dimerize with other HLH domains, but can not bind DNA since they lack the basic domain. N-terminal transactivation domains (TAD) have been identified for MyoD and MRF4, while myf-5 and myogenin contain both C- and N-terminal TADs.
There are three conserved amino acids (Ala, Thr, and Lys) adjacent to the basic domain of MRFs that are responsible for their muscle-specific activity. Mutating these residues can inactivate the myogenic activity of MyoD, and the transfer of these residues to non-myogenic El2 proteins or the neurogenic MASH-1 protein can transform these proteins into a myogenic factor capable of converting fibroblasts to muscle cells (Davis and Weintraub, 1992; Dezan et al., 1999). The mechanisms of how the MRFs are influencing transcriptiond activation of their target genes is beginning to be understood.
Recent studies have demonstrated that MyoD may influence changes in chromatin architecture necessary for transcriptional activation. Gerber et al., 24
(1997) demonstmte that MyoD can remodel chromatin in the myogenin promoter and activate transcription at this previously dent locus. The two regions responsible for this remodelling activity are distinct from the previously identified
TADs. This remodelling activity may be dependent upon histoone acetyltransferase
(HAT) activity. The degree of acetylation of lysine residues in the tails of histone proteins is related to transcription activity: hyperacetylated histones are often associated with transcriptionally active chromatin, while hypoacetylated his tones are associated with transcriptionally silent chromatin (Hebbes et al., 1988). MyoD can interact directly with p300/CBP and recruit the HAT PCAF. which is necessary for MyoD-dependent transcriptional activation and myogenesis (Puri et d.,1997;
Puri et al., 1997; Sartorelli et al., 1997). This interaction serves two functions:
PCAF provides acetyltransferase activity needed to remodel the chromatin to a transcriptionally permissive state, and pSOO/CBP serves as a co-activator uniting the MyoD-protein complex with the basal transcriptional machinery (Puri et al.,
1997). In addition to its interactions with co-activator proteins, MyoD can also associate with the nuclear co-repressor N-CoR (Bailey et al., 1999). This interaction is somewhat surprising since MyoD-mediated transcriptional inhibition has not been documented. Like their co-activator counterparts, the co-repressor proteins recruit histone deacetylase complexes that can remodel surrounding chromatin from a transcriptionally active to an inhibitory state (Kuo and Allis,
1998). N-CoR is expressed in undifferentiated myoblasts and is down-regulated upon induction of muscle differentiation in vitro (Bailey et al.. 1999). These 25 investigators suggest that the association of a co-repressor with MyoD in the undifferentiated cell prevents MyoD from activating its targets (Bailey et al.. 1999).
Therefore, myogenic differentiation may be regulated by the differential association of MyoD with co-activator/repressor complexes. The formation of these higher order complexes may be influenced by the cellular environment. For example, a factor present in mitogen-rich growth media may prevent the association of MyoD with a co-activator complex, and prevent premature activation of muscle differentiation in growth promoting conditions.
Interactions between the MyoD-MRFs and the myocyte enhancer factor
WF)-2 proteins are also important in the regulation of muscle-specific gene expression Black and Olson. 1998; Olson et al.. 1995). The MEF2 proteins which belong to the MADS-box family of transcription factors, are expressed at the onset of myogenic differentiation and bind to AT rich elements often positioned in close proximity to E-boxes in the control regions of muscle-specific genes (Gossett et al.,
1989; Wright et al.. 1991). On their own. MEF2 proteins can not activate the myogenic differentiation program. However, these proteins can physically interact with MyoD-MRFs and activate transcription of muscle-specfic promoters in a synergistic manner (Molkentin et al.. 1995; Molkentin and Olson, 1996). This activation occurs with MEF2 mutants that fail to bind DNA suggesting that the interaction between MEFZ with MyoD/E proteins bound to DNA are essential for enhanced myogenic differentiation (Black et al.. 1998; Black and Olson, 1998;
Molkentin et al., 1995). It is becoming increasingly clear that the MyoD-MRFs are 26 not acting alone in their myogenic regulatory activities. Interactions with accessory factors such as MEFZ and co-activator proteins to form higher order regulatory complexes as described above, are required for the activation of the differentiation program.
1.4.4 Integration of muscle diflerentiation and cell cvcle regulation.
Cell differentiation and cell proliferation are mutually exclusive events as skeletal muscle differentiation is marked by a loss of proliferative capacity (Nadal-
Ginard, 1978). Often this growth arrest is marked by an entry into a quiescent Go phase of the cell cycle, the absence of DNA synthesis, and a decrease in cyclin- dependent kinase (Cdk) activity through an association of Cdk-inhibitory proteins such as p2 lC'PINml(reviewed in Walsh, 1997). There is a growing body of evidence to suggest that the MyoD-MRFs are not only involved in the regulation of the myogenic differentiation program but may also promote a withdrawal from the cell cycle. For example, cell lines that are refractory to myogenic conversion by MyoD will cease to proliferate and enter a growth arrest in response to MyoD expression
(Crescenzi et al., 1990; Sorrentino et al., 1990; Weintraub et al., 1989). In addition, MyoD can activate expression of the cyclin-dependent kinase (Cdk) inhibitor p2 lCml~AF1in muscle cells which, in turn, hactivates cyclin-Cdk complexes and induces growth arrest upon myogenic differentiation (Guo et al.,
1995; Halevy et al., 1995; Parker et al., 1995; Puri et al., 1997). However, the relationship between MyoD and p21Cm1PNAF1expression is not entirely clear since p2 lCLP1"NAFLmRNA is detected in myogenic precursor cells from M~OD-~mice
Parker et al.. 1995; Sabourin et al., 1999). It is possible that other MyoD-MRFs such as myf-5 may compensate. This inhibition of cell proliferation by MyoD may act to commit cells to the differentiated state once the differentiation program has been initiated.
One role of p21C1P1"YAF'in muscle differentiation may be its inactivation of cyclinD 1/Cdk4 complex preventing phosphorylation of the retinoblastoma protein
(pRB). The phosphorylation status of pRB influences its activity and can regulate cell cycle progression. In its active, hypophosphorylated form, it can inhibit the
E2F family of transcription factors which are required for the GI-S progression of the cell cycle. This restraint on the cell cycle is relieved through the phosphorylation of pRB by cyclinD 1/Cdk4. The activity of pRB is instrumental in the maintenance of the postmitotic state of the differentiated muscle cell.
Fibroblasts and myoblasts isolated from mice deficient in pRB can form myotubes that express markers of differentiation such as myogenin; however, these cells are still capable of DNA synthesis when stimulated with serum (Novitch et al.. 1996:
Schneider et al., 1994). MyoD can bind to the hypophosphorylated form of pRB and may act to keep pRB in its active state (Gu et al.. 1993). Based on these examples. MyoD may play multiple roles in regulating the exit from the cell cycle at the onset of differentiation.
Although MyoD is expressed in myoblasts, these cells continue to proliferate and remain undifferentiated. Therefore the environment of the proliferating cell 28
can negatively regulate MyoD activity- The addition of serum or growth factors
such as bFGF or TGF-P to differentiation media can block the terminal
differentiation of muscle cells and inhibit the DNA binding activity of MyoD
(Vaidya et aI.. 1989). The inhibitory actions of serum may be partially mediated
through cyclin D 1, since its expression can be regulated by growth factors (Rao and
Kohtz, 1995). Furthermore ectopic cyclin D 1 expression in myoblasts inhibits
muscle differentiation and can prevent MyoD-transactivation of muscle specific
promoters (Rao et al., 1994; Rao and Kohtz, 1995; Skapek et al., 1995).
Interestingly, this inhibition was independent of pRB phosphorylation by cyclin
D l/cdk-4, suggesting that cyclin D 1 is acting in a manner different from its stereotypical role. Recently Zhang et al., (1999a; 1999b) have demonstrated that an interaction between MyoD and cdk-4 inhibits the DNA-binding ability of MyoD and prevents the myogenic conversion of 10T1/2 cells as well as cdk-4 phosphorylation of pRB. A suggested role for cyclin D 1 in this inhibitory mechanism is the recruitment of cdk4 to the nucleus and not the formation of an active cyclin D l/cdk-4 complex since MyoD is not a substrate for cdk-4 phosphorylation (Skapek et al., 1996; P. Zhang et al., 1999).
An additional strategy to inhibit MyoD activity in proliferating cells may involve the regulation of its stability though phosphorylation. The abundance of
MyoD protein is regulated throughout the cell cycle with highest levels detected in
G, that rapidly decline at the G,/S boundary, and essentially no MyoD is detected from S to M phase (Kitunarm et al., 1998). The phosphorylation of nuclear 29
MyoD by cdk-1 or 2 inhibits its transactivation of E-box promoters and promotes its export from the nucleus and its degradation (Kitwnann et al., 1999; Song et al..
1998). Conversely. the addition of Cdk inhibitors such as p2 lC'P1"NAF1.p27 and p57 promotes the accumulation of a stable, hypophosphorylated MyoD protein
(Reynaud et al.. 1999). These results suggest a scenario where in proliferating myoblasts, the activity of MyoD is restricted by cyclins and Cdks and the signal to initiate differentiation is the removal of this constraint though the inactivation or down-regulation of these factors. However, since the induction of the myogenic program, as defmed by the activation of myogenin expression. occurs prior to the exit from the cell cycle. MyoD is likely active before the down-regulation of the inhibitory cyclins and Cdks (Andres and Walsh. 1996). It is the activation of
MyoD that may induce the expression of p21C"'"KAF1and initiate growth arrest
(Guo et al.. 1995; Halevy et al., 1995). Therefore other factors may be involved that inhibit MyoD activity in the undifferentiated cell, and it is the down-regulation of these factors that initiates the differentiation program.
1.5 Rationale for current research.
Although ATBFl has been identified as a negative regulator of the AFP gene, very little is known about the ATBFl protein isoforms and their functions. The expression patterns for ATBFl-A mRNA in the embryonic mouse suggest it may participate in the development of the central nervous system . Extensive work in our laboratory has been aimed at determining a role for ATBF1-A in neuronal 30
differentiation, but its role in other tissues is not known.
Like other members of the ZFH family. ATBF 1-A may also participate in
the differentiation of a variety of cell lineages apart from neuronal cells such as
muscle, T-cells and B-cells (Kostich and Sanes, 1995; Lai et al., 1993; Postigo and
Dean, 1997; Sekido et al., 1994). Preliminary work in our laboratory has
demonstrated that ATBF 1A mRNA was expressed in undifferentiated C2C 12
myoblast cells, and these expression levels were sigmf~cantlyreduced in
differentiated muscle cells (Y. Miura and T. Tarnaoki, unpublished observations).
We were particularly interested in whether ATBF 1-A regulated myogenic
differentiation since muscle and neuronal differentiation programs share many
common regulatory strategies. For example, both programs are regulated by a
hierarchy of bHLH transcription factors that act in the determination and
differentiation events (Kageyarna and Nakanishi, 1997; Yun and Wold, 1996).
Members of MEF2 transcription factor farnily are expressed in the CNS and these
proteins interact with neuronal-bHLH proteins to enhance transcriptional
activation in a similar manner observed in muscle cells (Black et al., 1996; Lyons et
al., 1995; Molkentin et al., 1995). In addition, the inductive signals that initiate
determination and differentiation of the myogenic lineage, such as SHH and Wnts, also provide inductive cues for the developing CNS . Based upon the similarities
of the two differentiation programs, it is not unlikely that ATBF1-A may
participate to some capacity in the differentiation of these cell types. Therefore, we were interested whether this down-regulation of ATBF 1-A expression was important for the differentiation process, specifically, whether constitutive expression in C2C12 cells would inhibit their terminal differentiation.
To examine this possibility, stable C2C 12 cell-lines expressing ATBF 1-A and
B under the control of the constitutively-active mouse pgk promoter or CMV promoter were generated. The ability of these cells to differentiate was monitored and assayed by the expression of differentiation marker such as MymD-MRFs, cell cycle regulatory proteins and muscle structural genes. The expression patterns of these genes were to be compared with control cells and any factors chisplaying altered expression profdes were analysed for potential targets of ATBF1 regulation.
Such experiments may lead to a better understanding of ATBFl functions and to the identification of targets of ATBF 1transcriptional regulation. In addition data generated from these experiment may provide insight to functions of ATBFl that may be applicable to other biological processes including neuronal differentiation.
Specific goals for this work
1. Establish C2C12 cell lines permanently expressing ATBFI-A and ATBF 1-B. 2. Determine whether constitutive ATBF1-A expression inhibits skeletal muscle differentiation. 3. Compare expression patterns of myogenic regulatory factors between differentiating C2C 12 cells and ATBF1-A expressing cells to identify possible targets of ATBF1-A regulation, including MRF4. 4. Determine whether ATBF1-A and B have functionally equivalent activities in C2C12 cells. 5. Define specific ATBF 1 functional domains through the analysis of ATBF 1- deletion mutations. 6. Idenw factors which regulate ATBFl-A mRNA expression in muscle cells. 7. Assign a role for ATBF1-A in undifferentiated myoblasts vs. differentiated myotubes. 2. Materials and Methods
2.1 Plasmids.
pPOP-A, ATBF 1A expression vector, and pPOP-AS, antisense ATBF 1
expression vector, were constructed by cloning the full-length human ATBF 1-A
cDNA into pPOP in the normal and reverse orientation, respectively Ma, M.Sc.
Thesis 1997). ATBF1-B expression vectors, pPOP-B and pcA'TBF 1. were
constructed by cloning the full-length human ATBFl-B cDNA into pPOP and
pcDNAl (Invitrogen), respectively (Ma, M.Sc. Thesis 1997; Yasuda et al., 1994).
An HA-epitope tagged ATBF 1-A expression vector, pHA-ATBF 1-A, was
constructed by isolating a SalI-Not1 fragment fdlength ATBFI-A cDNA from
pHBSME (Katakao et al.. 2000) and Ligatation, in frame, into SalI-Not1 sites of
pCI-HA (a gft from Dr. D. Young, University of Calgary). The HA-tagged ATBF1-
A cDNA was digested with NheI and Not1 and subcloned into the corresponding
sites in the pcDNA3 expression vector to generate pc3HA-ATBF1-A. pPOP.
pgkMyoD (mouse MyoD expression vector) and p65-7, (rat myogenin cDNA) were
kind @ of Dr. M-W-McBurney (University of Ottawa). pcDNA3.1-Id3 (human
Id3 expression vector) and pId3-luc (human Id3 promoter-luciferase reporter) was
kindly provided by Dr. B.A. Christy (University of Texas), pECE-MyoD* -E47
(tethered MyoD-E47 construct) by Dr. B. Wold (Cal Tech), pRCCycD (human cyclin D 1 cDNA) by Dr. K Riabowal (University of Calgary) and pMRF4-CAT and pDNA-MEF2C by Dr. E.N. Olson (University of Texas. pBX15 1 and pMD 14 33 ATBFl cDNAs used to prepare riboprobes for RNase protection assays of mouse and human ATBF 1 mRNA respectively, have been described previously (Miura et al., 1995).
2.2 Cell Cultures and Stable Transfections.
C2C 12 myoblasts and 10T112 fibroblasts were routinely maintained in high glucose, Dulbecco's modified Eagle medium @MEW supplemented with 10% fetal bovine serum (Growth Media, GM). To induce rnyogenic differentiation of
C2C 12 cells, subconfluent cultures were switched from GM to differentiation media (DM) consisting of high glucose DMEM supplemented with 2% horse serum. Terminal differentiated multinucleated myotubes were normally harvested after 4 days of differentiation. Undifferentiated C2C12 cells were harvested as the semi-confluent cultures, prior to the addition of DM.
The growth and differentiation of C2C 12 cells was manipulated by the addition of pharmacological agents. To inhibit myogenesis, C2C 12 cells were induced to differentiate as described above, however in addition to 2% horse serum, the DM contained either 10 FM okadaic acid, 300 ng/mL recombinant human Bone Morphogenetic Protein 2 (BMP: Genetics Institute), 50 pM
PD098059 (New England Biolabs). 10 pg/rnL Lysophosphatidic acid (LPA; Sigma).
10 ng/mL bFGF (Gibco BRL) or 10 ngfd TGF-P (1. Okadaic acid and PD098059 were dissolved in DMSO, while the bFGF. LPA and TGF-P were dissolved in PBS. 34
BMP was prepared in DMEM. Differentiation media containing these vehicles were included as controls. Media was replaced with fresh inhibitor every 12 hours.
After 24 hours cells were either fured and processed for myosin heavy chain (MHC) im~unocytochemistryor RNA was harvested (see below).
C2C12 cells were transfected using the high efficiency calcium phosphate method (Chen and Okayarna. 1987). AU transfections employed high quality, supercoiled plasmid DNA, purifed through two rounds of CsCl- ultracentrifitgation. Cells were subcultured 5x104cells/mL prior to the day of transfection. Twenty micrograms of pPOP-A or pcATBF1, pPOP or pcDNAl expression vectors along with 4 pg of the neomycin resistance vector, pCI-NEO
(Strategene), were added to 500 yl0.25 M CaCI,. An equal volume of 2x BES- buffered saline (50 rnM N,N-bis (2-hydroxyethyl)-2-aminoethanesul acid, 280 mM NaCl, 1.5 mM Na,HPO,, pH 6.95) was added and the solution was allowed to stand at room temperature for 20 minutes. The calcium phosphate-DNA mixture was added dropwise onto a 100 rnrn culture plate. The cells were incubated for 18 hours in a 35°C. 3% CO, incubator. The following day the cells were twice washed with PBS, refed with fresh GM and incubated in a 37"C, 5%
CO, incubator. Once the cells approached confluence they were split 1: 15 in fresh
GM containing G4 18 and incubated for 2 weeks until single colonies were formed.
Fresh media containing G4 18 was added every 4 days. Individual G4 18-resistant colonies were isolated using pieces of sterile Whatman filter paper soaked in trypsin to physically remove the colony to a chamber of a 24 well culture plate. 35
Twenty four and eight colonies were isolated and propagated from the ATBF 1-A and -B transfections, respectively. The integration of the transfected ATBF 1 expression vectors was monitored by Southern blotting and mRNA expression was verified by RNase protection assays.
For expression of ATBFl-A in COS cells. 2x106 ceWml were transfected with 18 mg of pc3-HA-ATBFl-A DNA by the BES-CaPO, method as describe above. Twenty-four hours after transfection. the medium was replace with fresh medium containing G4 18 (800pg/rnl). CeLls were harvested 48 hours later in a buffer consisting of 40 mM Tris-HC1, pH 8.0; 2.5 mM EDTA; 2.5 mM EGTA;
250 mM NaCl; 1% NP-40; and 0.1% CHAPS. COS cell extracts were provided and analysed for ATBF1-A expression by Dr. K Fushirni (University of Calgary),
2.3 Southern Blotting.
Genomic DNA was isolated from C2C 12 cells by the rapid extraction method of Laird et al., 1991. Media was removed from a 60 mm culture dish and replaced with 2 ml of lysis buffer (100 rnM TrisHCl pH 8.5, 5 mM EDTA. 0.2%
SDS,200 mM NaC1. 100 ~g/rnlProteinase K). The plates were incubated for 3-4 hours at 37°C. After digestion, the cell lysate was transferred into a 15 p1 centrifuge tube containing 2 ml isopropanol. The tubes were inverted several times, and the precipitated DNA was removed from the isopropanol solution with a clean micropipette tip and transferred to a microcentrifuge tube containing 500 36 ml70% ethanol. The tubes were quickly centrifuged to pellet the DNA. The DNA pellet was resuspended in TE (10 rnM Tris and 1 mM EDTA, pH 8.0) and stored at -20°C.
Five micrograms of genomic DNA was digested with 5 units each of
HindIII and EcoRI for 4 to 6 hours at 37OC. To terminate the reaction, samples were heated to 65OC for 10 minutes. Digested DNA was electrophoretically resolved on a 0.7% agarose-TBE gel run overnight at 50 V at 4°C. Following electrophoresis, the gels were stained with ethidiurn bromide, visualized. photographed and placed in a 0.25 M HCl solution for 30 minutes. The gel was rinsed with distilled water and incubated in a 0.4 M NaOH denaturation solution for 20 min. DNA was transferred from the gel to Hybond N+ charged nylon membrane (Amersharn) by capillary action in 0.4 M NaOH overnight. Following transfer, membranes were placed between sheets of Whatman filter paper and dried at 70°C for 2 hours.
Prior to hybridization, membranes were rehydrated with 5X SSPE (1X
SSPE= 0.18 M NaCI, 10 mM NaH,PO,, at 65OC for 10 minutes. This solution was replaced with hybridization buffer consisting of 5X SSPE, 5X Denhardt's solution
(0.1 % [wlv] Ficoll, 0.1 % [wlv] polyvinylpyrrolidone. 1 mg/ml bovine serum albumin), 0.5% SDS and 100 mg/ml sheared. salmon sperm DNA, and incubated for 2 hours at 65°C. The prehybridization buffer was removed and replaced with fresh hybridization buffer containing '*P labelled probe at concentration of 2 X lo6cprn/rnl. The probe used to detect both human and mouse ATBFl DNA 37 sequences was a 907 bp human cDNA. termed the I probe, corresponding to a sequence highly conserved between human and mouse coding regions. Following hybridization, the membranes were f~st,washed twice in 2X SSPE and 0.1% SDS at room temperature for 15 minutes, proceeded by two subsequent washes in 0.1X
SSPE and 0.1% SDS at 65OC for 30 minutes. The membranes were briefly air- dried. covered in plastic wrap and exposed to Kodak Biomax X-ray film with an intenslfylng screen at -70°C
2.4 mRNA expression analysis.
2.4.1 RNA Isolation.
Total RNA was isolated from cell cultures and mouse muscle tissue by the acid-phenol procedure (Chornczynski and Sacchi, 1987). Briefly, cells were washed several times with PBS and to each 100 rnm plate, 1 ml of denaturation solution consisting of 4 M guanidium thiocyanate, 25 rnM sodium citrate (pH 7.0). 0.1 M
2-mercaptoethathol, 0.5% N-laurylsarkosine, was added. Cells were scraped and collected into centrifuge tubes containing 0.1 ml of 2 M sodium acetate, pH 4.0.
Homogenates were extracted with an equal volume of water-saturated phenol, arid
1/5 volume of 49: 1 chloroform/isoarnyl alcohol. The suspensions were incubated on ice for 15 minutes centrifuged at 10,000x g for 20 minutes at 4OC. The aqueous phase was collected and precipitated with an equal volume of isopropanol.
The RNA pellet was dissolved in denaturation solution, precipitated a second time with isopropanol and washed with 70% ethanol. The final RNA pellet was resuspended in formarnide and stored at -70°C.
2.4.2 RNase protection assavs
Three probes were utilized in the RNase protection assays: a human ATBFl probe. MD14: a mouse ATBFl probe. BX151 and a mouse GAPDH probe, mGAP.
To generate the antisense BX15 1 and mGAP riboprobes, the vectors was linearized with BarnHI, and transcribed with T3 polymerase, whereas. the MD14 ribropobe was generated by linearization with BamHI followed by transcription with T7 polymerase.
Single stranded RNA probes were transcribed with 3Z~-~~~.Linearized plasmid (1 pg) was added to a reaction mixture consisting of 40 rnM Tris-C1 (pH
7.5). 10 rnM MgCl,. 2 mM spemidine, 50 mM NaCl. 1 unit/@ RNase inhibitor.
50 pM ATP, UTP, GTP. 12 pM unlabeIled CTP, 10 units of RNA polymerase (T7 or T3) and 50 pCi 32~-~~~.The mixture was incubated at 37OC for 2 hours. To terminate the reaction and to digest the template DNA, 10 units of RNase-free
DNase I was added, and then further incubated for 15 minutes, The labelled riboprobe was extracted initially with an equal volume of phenol:CHCl,:IAA followed by a subsequent extraction with an equal volume of CHC1,:IA.A. The extracted reaction mixture was then run through a Sephadex G50 spun column and ethanol precipitated. The final purifed riboprobe was resuspended in EWA hybridization buffer (80% formarnide; 40 rnM PIPES, pH 6.4; 400 rnM NaCI; 1 rnM EDTA) at a fmd concentration of 5x10~cpm/ml.
Five micrograms of total RNA was hybridized to 5x10' cpm of labelled riboprobe overnight at 45OC. Ribonuclease digestion buffer containing 40 pg/ml ribonuclease A and 2 pg/d ribonuclease T1 in 10 mM Tris. pH 7.5. 300 rnM
NaCl and 5 mM EDTA was added to each reaction and incubated for 1 hour at
30°C. Following Wase digestion. SDS (0.5%) and Proteinase K (125 pg/rnl) were added and the samples were further incubated for 30 minutes at 3'7°C. Reaction mixtures were then extracted with an equal volume of phenol:CHCl,:IAA and ethanol precipitated. The protected riboprobe samples were resolved on a 5% denaturing polyacrylamide/urea gel and autoradiographed with an intenslfylng screen overnight.
2 -4.3Northern Blottin~i.
For Northern analysis. 10 pg of total RNA was fractionated on a 1.5% denaturing-agarose gel (6% formaldehyde; 40 mM MOPS. pH.7.0; 10 rnM sodium acetate; 1 rnM EDTA, pH 8.0). transferred to Hybond N nylon membrane
(Arnersham Pharmacia Biotech) in 10 X SSC (1X SSC= 150 mM NaCl. 15 mM sodium citrate, pH 7.0) by capillary action. The membranes were exposed to UV light for 2 minutes and air dried prior to processing.
Membranes were rehydrated in 5X SSPE and incubated at 42°C in a solution of 50% formamide. 5X SSPE, 10X Denhardt's solution, 1% SDS and 100 pgml 40 sheared. salmon sperm DNA. Probe hybridization was carried out in the above solution supplemented with106 cprn/ml 32 P-dCTP labelled, denatured, cDNA probes. The following plasrnids were used to generate the appropriate probes: myogenin- a 695-bp EcoRI-Pst I fragment of p65-7; MyoD- a 2.2-kb Barn HI fragment of pgkMyoD, Id3- a 550-bp EcoFU fragment of pcDNA3.1-Id3 ;cyclinD 1- a 1-1-kbHind HI-Not3 fragment of pRCCycD; p2l-a 900 bp SacII-SpeI fragement of pMWAF 1. Radiolabelled probes were synthesized using random primer method as described by the manufacturer (T7 quickprime. Arnersham Pharmacia Biotech).
Following hybridization, membranes were washed twice at room temperature in
1X SSPE and 0.1% SDS for 15 minutes followed by two subsequent 30 minute washes in 0.1X SSPE and 0.1% SDS at 65°C. Membranes were exposed to X-ray film at -70°C. To reuse membranes for hybridization with several different probes, the existing probe was stripped from the membrane, by placing it in a boiling solution of 0.1% SDS and allowed to cool to room temperature before hybridization with the next probe.
2.5 Immunocytochemistry.
Cells were grown on 6- or 24-well plates, washed three times with phosphate buffered saline (PBS) and then fuced in a 70% ethanol, 10% formaldehyde, 5% acetic acid solution at -20 OC for 10 minutes. The cells were rehydrated with PBS and incubated for 2 hours at 37 "C with a monoclonal antibody against MHC (MF-20;Developmental Studies Hybridoma Bank (Bader et 41 al., 1982)). at a final concentration of 5 pgMin PBS supplemented with 5% goat serum. The cells were then washed twice in PBS and incubated for 1 hour at 37 OC with peroxidase-coupled anti-mouse IgG (1 500; Santa Cruz). After two washes with PBS, the cells were incubated with 0.3% diaminobenzidine and 0.15% Hz 02 in PBS for 10 minutes at room temperature- The cells were washed with PBS for 5 minutes, stained with 0.05% Cresyl Violet and dehydrated with ethanol.
2.6 BrdU incorporation.
C2C12 and A6 cells were seeded at a density of 1x10~ceWmL on sterile, acid washed coverslips. When cells had reached semi-confluence, they were induced to differentiate as described above. At specific time points the DM was removed and replaced with DM or GM containing 10 pM BrdU and incubated for
4 hours. The cells were washed with PBS, fixed with methanol and then blocked with 10% normal goat serum diluted with PBS. The cells were then treated with 2
N HC1 and 0.5% Triton X-100 for 30 minutes followed by extensive washes with
0.1 M sodium tetraborate (pH 8.5) and PBS. BrdU antibody diluted 1500 was applied to each coverslip and incubated for 1 hour at 37OC. Following primary antibody treatment the cells were incubated with FITC-conjugated goat-anti-mouse
IgG (1 :25). Nuclei were counterstained with DAPI (1 mg/mL) and mounted with
Gelvatol onto microscope slides. At least 900 nuclei were scored for BrdU incorporation at each time point. 2.7 Western Blotting.
Total protein lysates were obtained from five confluent 100 rnrn plates of
C2C 12 and C 12A6 cells cultured in both growth and differentiation media. Plates were fmt washed twice with ice-cold PBS, then cells were scraped into PBS and pooled in 50 rnl centrifuge tubes. Cells were pelleted by centrifugation at 4OC for
5 rnin at 2000 rpm. The pellet was resuspended in 0.5 ml of NP 40 lysis buffer (20 rnM Tris pH 8.0, 150 mM NaCl, 1 mM EDTA. 1% NP40, 10% glycerol. 10 pg/rnl aprotinin, IrnM PMSF. 5 pg/ml leupeptin. 5 mM Benzamidine, 2 pg/rnl antipain.
50 pM AEBSF, 1 pg/d pepstatin. 0.1% 2-rnercaptoethanol. 0.5 mM DTT and 20 units/rnl Urinostatin) . transferred to a microcentrifuge tube and gently rocked for
20 min at 4OC. To remove cellular debris, lysates were centrifuged at 4OC for 30 min. at 15,000 rpm in a microcentrifuge. The supernatant was removed and stored in 50 pl diquouots at -70°C. Protein concentrations were determined using the
Biorad protein assay.
Proteins were separated electrophoretically using the discontinuous buffer system of Laernrnli (Laemrnli, 1970). Samples were solubilized by boiling for 3 minutes in 50 mM Tris-Cl pH 7.5, 2% SDS, 150 rnM 2-mercaptoethanol. 10% glycerol and 0.1 rnM EDTA (SDS sample buffer). Proteins (30 pg) were separated on 242% SDS-polyacrylamide gradient mini-gels at constant current of 70 mA.
After separation, proteins were transferred to Hybond C+ supported nitrocellulose membranes (Amersharn) in a solution of 25 mM Tris, 0.19 M glycine, 20% 43 methanol and 0.125% SDS at 4OC for 90 minutes at 150 mA. Following transfer. membranes were stained with 0.5% Ponceau S, destained with water and washed four times in 1X TBST for 5 minutes (1X TBST=lO rnM Tris, 250 rnM NaCl and
0.05% Tween-20, pH 7.5). Blots were then transferred to blocking solution consisting of 0.5% skim milk in TEST and incubated for 30 rnin at room temperature. Blocking solution was then removed and replaced with primary antibody diluted in blocking solution and incubated overnight at 4°C. The following primary antibody dilutions were employed a-D 1 (rabbit polyclonal, a gift from Dr. T. Hasirnoto-Tarnaoki) . 1:2500; a-Desmin and a-tubulin (mouse monoclonals, Arnersham) , 1:2000; and a-p27 and a-cdk4 (rabbit polyclonals.
Santa Cruz) , 1: 500. The primary antibody solution was removed and membranes were washed three times for 10 minutes each in 1X TBST. Following washing, membranes were incubated for 2 hours in appropriate horseradish peroxidase - conjugated secondary antibody (anti-mouse-, or anti-rabbit IgG) diluted 1:5000 in blocking solution. Finally, membranes were washed six times in 1X TBST for 5 minutes. Protein-antibody complexes were detected by enhanced chemilurninescence @hamacia Arnersham) and visualized by exposure to Kodak
OMAT X-ray film. Western blot data are representative of at least 3 replicate experiments. 2.8 Myogenic conversion assays.
The conversion of 10T112 fibroblasts to the myo blast lineage was performed by MyoD transfection. 10T 112 cells were transiently trans fected with Superfect transfection reagent according to the manufacturer's protocol (Qiagen). In a 60 rnrn culture plate, cells were transfected 1 pg PGK-MyoD expression vector, along with 1 to 5 pg of either pPOP-A, pPOP-B or pPOP-AS. Quantities of DNA were equilibrated to 6 pg with empty pPOP expression vector. To induce differentiation, transfected cells were wansferred to DM and incubated for 4 days.
The extent of myogenic differentiation was monitored by MHC immunocytochernistry as described above. Transfections were performed in triplicate and the number of MHC positive cells were averaged from 5 random fieIds of view per plate. Values obtained for the cells transfected with MyoD alone were set at 100%.
2.9 Lucserase reporter assays.
To generate the MRF4-luc reporter, the plasmid pMRF4-CAT (Black et aI..
1995) was digested with Xba I and Hind 111 to liberate the mouse MRF4 promoter element, end-fded with Klenow, and subcloned into pUBT-Luc luciferase reporter plasmid. 2x1o5 C3H 10T112 cells in 35 rnm plates were transfected with 0.5 pg pMRF4-Luc, 0.5 pg pCH 110. a P-galactosidase expression vector (Pharmacia) . 0.5 pg of pPOP and 0.5 pg of the appropriated ATBFl expression vector with 45
Superfect as described above. Cells were transferred to DM for 2 days and
harvested in reporter lysis buffer (Promega). Protein concentrations were
determined by Biorad Proteh assay and samples were processed immediately.
Luciferase activity was determined fkom 10 pg of total cellular lysate using the
Luciferase Assay System (Promega) and detected on a Monolight 20 10
luminometer (Analytical Luminescence Laboratory). P-galactosidase activity was
detected by incubating 10 pg of total cellular Iysate in a reaction mixture of 100
mM sodium phosphate buffer. pH 7.3, 1 mM MgCl,, 50 rnM 2-mercaptoethanol,
and 665 pg/d o-nitrophenyl-b-D-galactopyranoside for 3 hours at 37' C. The
colourmetric reaction was stopped by the addition of 1 M Tris base and absorbance
read at 420 m. The transfection effeciencies were normalized by dividing
luciferase activity by P-galactosidase activity.
In addition to the MRF4-luc reporter, other promoter-reporter constructs
were examined. 4Rtk-luc contains four copies of the muscle creatine kinase E box
element upstream of the herpes simplex virus thymidine kinase promoter
(Weintraub et al., 1990). Again, these plasmids were transfected in the manner
described for MRF4-luc, above with the exact DNA concentrations listed in the figure legends.
2.10 Mammalian one-and two-hybrid assays.
The yeast GAL4 DNA binding domain @B) was fused in-frame to the 46 amino-terminus of human ATBF 1-A and B proteins to generate the vectors GAT1 and GAT8 respectively. G4tk-luc contains four GAL4 binding sites upstream of the thymidine kinase promoter, while GCluc contains the same four GAL4 binding sites upstream of the El b TATA box. Expression vectors corresponding to MyoD-
VP16 and ID-GAL4, fusion proteins were components of the Matchmaker
Mammalian Two-hybrid System (Promega). For mammalian two hybrid assays.
CZC12 cells were transfected with 0.5 pg G4tk-luc reporter. 0.5 pg of each
GAL$,, fusion protein. 0.5 pg pCH110 (B-gal). and 0.5 pg of either pMyoD-VP16 or an empty pPOP expression vector. Cell extracts were prepared 48 hours after transfection and luciferase activity was analysed as described above.
2.1 1 DNA-protein interactions.
2.1 1.1 Nuclear Extract Isolation.
Nuclei were isolated from C2C 12 cells and 10T1/2 fibroblasts essentially as described by Lassar et al.. (199 1). Briefly, confluent 100 mrn plates were washed three times with ice-cold TBS. Cells were lysed by adding 1.5 rnl cell lysis buffer consisting of 20 mM HEPES (pH 7.6). 20% glycerol. 10 rnM NaCl, 1.5 rnM
MgCl, 0.2 mM EDTA, 0.1% Triton X-100, 1 rnM DTT, 1 mM PMSF, 10 pg/rnl leupeptin. 10 pg/d pepstatin A. and 100 pg/ml aprotenin to each plate. Cells were scraped. collected and pelleted by centrifugation for 5 minutes at 2.000 rpm
4OC. Each pellet was resuspended in nuclear lysis buffer (as above with 0.5 M 47
NaCl) at a concentration of 2.5 x107 nucleilrnl, transferred to a microcentrifuge tube and rocked for 1 hour at 4' C. Samples were then centrifuged at 10,000 rpm for 10 minutes at 4" C. The supernatants were removed, "ed and stored at -
70" C. Protein concentrations were determined using the Biorad protein assay.
2.1 1.2 Probe preparation.
The probes used for gel mobility shift assays were oligonucleotides corresponding to the sequence 5'-ACGTTAATTAAATGCCATCTGGGTG-3' derived from the mouse MRF4 promoter containing an E-box (underlined) and an
AT-rich sequence (bold) @lack et al., 1995). A 4 bp, 5' overhang was created to facilitate labelling with Klenow polymerase. This probe, termed AIE,was further subdivided into two smaller probes, El, containing the E-box sequence and AT, containing the AT-rich element. Single stranded oligonucleotides were annealed in a buffer consisting of 10 rnM Tris-HC1 (pH 73, 20 mM NaCl and 50 mM MgC1,.
Double stranded oligonucleotide DNA (50 ng) was labelled with 1 unit of Klenow enzyme, 500 pMeach dATP. dTTP, dGTP and 20 pCi 3ZPdCTPfor 30 minutes at
37OC. Unincorporated radiolabelled nucleotides were removed using the Qiaquick
Nucleotide Removal Kit (Qiagen). The labelled probe was diluted to 100 pgpl
(30,000 cpdpl). A commercially available oligonucleotide probe corresponding to an MEF2 consensus site (5'-GATCGCTCTAAAAATAACCCTGTCG-3')was purchased from Santa CNZ Biotechnology. To label this probe, 40 ng (2 pmol) of 48
DNA was end-labelled with 20 pCi 32~-~-ATPand T4 polynucleotide kinase for 1 hour at 37°C.
2-11.3 h vitro translation of proteins.
A coupled h-anscription-translation reticulocyte Lysate system was used to produce in vitro translated proteins as described by the manufacturer (Promega).
Plasrnids, pcDNA-MEFZC and pcDM-2, each containing the T7 promoter, were used to produce recombinant MEF2C protein as well as a 200 kDa portion of the
C-terminus of ATBF 1, respectively. A pardel set of reactions was performed for each plasmid. One reaction contained 20 pCi 35S-methioninein the amino acid mixture to venfy the production of the correct protein, while a second reaction contained no radiolabelled amino acid. This second, cold reaction product (3 p1) was subsequently used in gel mobility shift assays.
2.1 1.4 Bacteriallv emressed proteins.
Recombinant proteins corresponding to specfic regions of ATBF 1-A and B were produced using the pET30 bacterial expression and HIS-bind purification systems (Novagen). NH, a 6 1 kDa protein corresponding to the N-terminus of
ATBFl-A and HD4, a 60 kDa protein corresponding to the fourth homeodomain of ATBF 1-A and B were generously supplied by Dr. Koichiro Mihara (University of
Calgary) . 2.1 1.5 Electro~horeticMobilitv Shift Assavs (EMSAs)-
For gel mobility shift assays, nuclear extracts (10 pg) , bacterially expressed proteins (5 ng) or in vitro translated proteins (3 pJ) were incubated in a 20 PI binding reaction containing 20 mM HEPES, pH 7.6; 5% glycerol; 50 rnM NaCl;
1.5 mM MgC1,; and 1 pg poly(d1-dC)-poly(d1-dC)for 10 minutes at room temperature. Probe was added at a concentration of 100 pg/20 pl reaction (30,
000 cpm) and the reaction was incubated for a further 20 minutes. DNA-protein interactions were resolved on a 5% polyacrylamide gel at 30 rnA in buffer of 50 rnM Tris, 50 mM boric acid, and 1 mM EDTA. Gels were dried on a Biorad Gel drying apparatus and exposed to X-ray fhat -70°C.
To iden* protein components present in DNA-protein complexes. antibodies were added to the binding reaction prior to the addition of the probe
DNA. The following antibodies were used for super-shift assays: D4, a rabbit polyclonal raised against HD4 of ATBF1; F5D, a-myogenin hybridoma supernatant
(Wright et al., 1991) ; MF-20, a-MHC hybridoma supernatant; C20, MyoD polyclonal; and C 19, c-myc polyclonal antibodies (both purchase from Santa Crux
Bioctechnology) . Typical volumes of antibodies added for hybridoma supernatants and for polyclonal antibodies were 5 pl and 1 pl, respectively.
2.1 1.6 W cross-linking: of proteins to nucleic acids.
DNA-protein binding reactions required for W cross-linking were performed using bacterially expressed proteins as described above with the following modifications: 20 ng of protein was incubated with 5 ng (3x10' cpm) of probe DNA. The binding reaction was then placed 5 crn under a hand-held UV- transillurninator and exposed to a wavelength of 254 nrn for 1 hour at room temperature. An equal volume of SDS sample buffer (section 2.7) was then added to the mixture and boded for 5 minutes. Samples were run on a 12% SDS polyacrylamide gel, dried and exposed to X-ray fh
2.1 2 Immunoprecipitation and His-tag pull-down assays.
For irnmunoprecipitation experiments, ceU extracts were isolated from 100 rnm plates. Briefly, cells were washed in PBS, harvested in RIPA buffer (RIPA buffer= lX PBS, 0.5% NP-40, 1% sodium deoxjcholate, 0.1 % SDS ) containing protease inhibitors as described above in the Western Blotting section, and collected into fresh microcentrifuge tubes. The cellular debris was pelleted by centrifugation at 10,000 x g for 10 minutes at 4OC. The supernatant was transferred to a fresh tube. Polyclonal antibodies (1 p1) raised against ATBF 1-HD 1,
ATBF 1-HD4. MyoD , or a-actin were individually added to the lysates and incubated at 4OC for 1 hour with gentle rocking. Protein-A Sepharose, resuspended in PBS (20 pl), was added to the lysate-antibody mixtures and incubated overnight at 4OC with gentle rocking. The immunoprecipitates were collected by centrifugation at 1000 x g for 5 minutes at 4OC. The pellets were washed four times in fresh RIPA buffer. Following each wash, the lysates were subjected to centrifugation at 1000 x g for 5 minutes at 4OC. After the fmal wash, 20 pl of SDS-
PAGE loading buffer was added, and the samples were boiled for 3 min, and loaded onto a 10% SDS-polyacrylamide gel. The proteins were transferred to nitrocellulose membranes and visualized for MyoD immunoreactivity as described above.
In vitro protein-protein interactions were analysed through a modified Ni2'- chelation chromatography (Novagen). The serum response factor protein (SRF) and the N-terminal portion of ATBF 1-A were expressed in bacteria as fusion proteins with a 6xHis tag as describe above. 6xHi.s-SRF was a generous gf3from
Keith Weat on and Karl Riabowol (University of Calgary). 35S-~et-labelledMyoD and MyoD-E47 were generated by coupled in vitro transcription/translation reactions as described above. Niz+-resinwas initially activated by an incubation in 5 volumes of 1X charge buffer (5 mM irnidozole, 0.5 NaCl, 20 rnM Tris, pH 7.9).
Five micrograms of the 6xHis proteins were added to 10 pl of resin and incubated for 1 hour at 4°C. Following this incubation, 5 pl of 35S-Met-labelledMyoD or
MyoD --E47, diluted in 500 pl RIPA buffer, was added to the resin and incubated for 2 hours at 4OC. The resin was washed 3 times in 60 mM imidazole, 0.5 M
NaCl, and 20 mM Tris. pH 7.9. Proteins were eluted from the resin with elution buffer (1 M imidazole, 0.5 M NaC1, 20 rnM Tris, pH 7.9) and the supernatants were collected and analysed by SDS-PAGE as described above. The gels were dried and subjected to autoradiography.
2.1 3 Tibialis anterior regeneration.
Male mice aged 10-12 weeks were anaesthetized with methoxyfluorate.
Four transverse crush lesions along the length of the right tibialis anterior were made with a fine-nosed pair of hemostats. The left limb was uninjured and served as a control. After either a 4 or 10 day recovery period, the mice were sacrificed and RNA was extracted from both the injured and untreated muscle as described above. At least 4 animals were used for each time point and two independent regeneration experiments were performed. 3. Results
3.1 Expression of ATBFl-A during myogenic differentiation.
We were interested in determining whether ATBF1-A participated in
myogenic differentiation since previous work in our laboratory suggested a role for
ATBF 1-A as a potential regulator of neurogenic differentiation (Ido et al., 1994;
Miura et al., 1995; Watanabe et al., 1996). Myogenic and neurogenic
differentiation employ some common regulatory strategies. As described
previously, both events are regulated by a hierarchy of bHLH transcription factors
that establish and maintain the differentiated neural or muscle cell type. Other regulatory transcription factors such as the MEF2 family and Pax3 have also been implicated in the development of both lineages. The C2C12 myoblast cell line provides an excellent model system to study the molecular events governing muscle differentiation. These cells represent a population of proliferating cells committed to the rnyogenic lineage (Yafffe and Saxel, 1977). When cultured in GM containing high concentrations of serum, these cells divide and proliferate normally. However, upon transfer to low mitogenic media (DM) they will initiate the differentiation process. The cells will cease to proliferate, align themselves with one another, and fuse to form multinucleated myotubes expressing structural proteins specific to differentiated muscle. The regulatory events that drive muscle differentiation in this in vitro system accurately represent the events occurring in vivo. However, as C2C12 cells do not exhibit cellular senescence, they may differ from normal muscle cells in this important property (Florini, et al.. 199 1).
Analysis of ATBFl-A mRNA expression by RNase protection assay showed that ATBFl-A mFWA was expressed in undifferentiated C2C 12 cells in GM Figure
2A). However upon transfer of cells to DM to induce myogenesis. ATBFl-A mRNA expression declined markedly within 24 horn and remained low for at least
72 hours (Figure 2A). Expression of ATBFI-B rnRNA was not detected in cells treated with either GM or DM (data not shown). When differentiated cells. treated with DM for 4 days were switched back to GM. ATBF1-A expression was elevated within 1 hour and maximal expression detected after 24 hours (Figure
2B). The expression pattern of ATBFl-A mRNA in C2C 12 cells suggests that it is not expressed or expressed at low levels in differentiated muscle. To examine this possibility. we analysed ATBF l -A expression by RNase protection assays in differentiated muscle isolated from a postnatal day 7 mouse. As indicated in Figure
2C. ATBFl-A mRNA was detected at very low levels in hindlirnb, forelimb, back and rib muscles and diaphragm. The levels detected in these cells are comparable to the levels detected in differentiated C2C12 cells and are much lower than those found in undifferentiated C2C12 myoblasts. Muscle from the hindlirnb. forelimb and back are derived from distinct mesoderm populations compared to the muscles of the ribs and diaphragm. Therefore the absence of ATBFl-A expression in differentiated muscle is not limited to individual muscle subtypes. High levels of
ATBF 1-A mRNA were detected in 10T1/2 and MH3T3 fibroblasts which are believed to represent a less committed mesodermal state (Figure 2C). Together, Figure 2. Expression of ATBFl-A mRNA in differentiated and undifferentiated muscle tissues.
(A). Subconfluent C2C12 myoblasts grown in GM were transferred to DM, total RNA was isolated from the cells after 0, 24, 48, and 72 hours and analysed for ATBF la mRNA by RNase protection assay as described under "Materials and
Methods." A parallel set of RNA samples were analysed for GAPDH mRNA expression by Northern blotting.
(B). stimulation of ATBFI-A expression by GM. DM was removed from terminally differentiated C2C 12 cells and replaced with fresh GM. ATBF 1-A expression was analysed by RNase protection assay after 1, 6 and 24 hours of treatment.
(C). Expression levels of ATBF1-A mRNA were compared by RNase protection assay between C2C12 myoblasts and myotubes with muscle isolated from a postnatal day 7 mouse. ATBF1-A expression was also analysed in mouse embryonic fibroblast cell Lines 10T1/2and NM3T3. 0 24 48 72 Hours inDM
0 1 6 24 Hours in GM
GADPH these data suggest that ATBFI-A may play a role in the maintenance of the undifferentiated state of myoblasts and its down-regulation in response to serum depletion is either required for terminal muscle differentiation or is a result of this process.
3.2 ATBFl-A inhibits myogenic differentiation of CZC 12 cells.
To examine the possibility that ATBFl-A zcts as a regulator of myogenesis, we transfected C2C 12 cells with a vector expressing human ATBFl -A under the control of the mouse pgk promoter. Three stably-transfected clones, A2, A6 and
A 19, were selected from twenty-four colonies for Merstudies.
Southern blot analysis confirmed the incorporation of the transfected ATBF1-A cDNA into the genome (Figure 3B). In clones A2, A6 and A19 bands corresponding to the transfected human ATBF 1-A cDNA and mouse ATBF la gene were detected (Figure 3B). In A2 cells, a band with a higher molecular weight than the transfected ATBFl-A band was detected. The exact origin of this band is not known, but it may result from an incomplete digestion of the genomic DNA, or a rearrangement of the transfected ATBFl -A cDNA upon integration into the genome. We next verified expression of the human ATBF1-A transgene using
RNase protection assays. There is a sufficient degree of dissimilarity between the human and mouse ATBFl-A sequence at the location of our riboprobe. Therefore mouse ATBF1-A mRNA will not protect the human derived riboprobe from RNase digestion, and we can specifically distinguish the transfected human ATBF 1 A Figure 3, Conformation of ATBF1-A stable transfected C2C12 cell lines,
(A) The full length ATBF1-A cDNA subcloned into pPOP expression vector
is indicated by the open bar. The initiation codon (ATG), termination codon
(TAA) , and alternative splicing site are indicated. Closed bar. ATBF 1 I-probe (see
Materials and Methods). Digestion with Hindm and EcoRI will produce a
predicted 7178 bp fragment that is recognized by the I-probe.
(B) Southern blot analysis of C2C12 cells transfected with a human ATBF1-
A expression vector. Genornic DNA digested with HindIII and EcoRI was blotted
to positively charged nylon membrane and hybridized with a 32~-d~~~-labelledI-
probe corresponding to the coding region of the human ATBF1-A cDNA. Mouse
and human hybridization complexes are indicated by arrows. A2, A6. A1 9
correspond to colonies stably transfected with ATBFl-A expression vector; Vector
lane corresponds to cells transfected with an empty pPOP expression vector:
C2C 12, parental, untransfected cell line; pPOPME, ATBF 1-A expression vector (1
pg) digested with HindIII and EcoR.1.
(C). Expression of human ATBFI-A mRNA in transfected C2C12 . Total
mRNA was isolated from cells treated with growth medium (G) or with
differentiation medium @) for 4 days. Expression of exogenous ATBF 1-A mRNA was monitored by RNase protection assays using the MD 14 riboprobe specific to
human ATBFl (Miura et al., 1995). The MD 14 riboprobe and the protected
ATBF 1-A and B products are shown below the figure. Probe lane represents
undigested riboprobe. A~G I ' -Probe Alternative splicing site TAA !
a2 C2C12 Vector A2 A6 A19 nt E GDGDGDGDGD
MD 14 (Probe)
256bp mF1-A 216bp ATBF1-B 60
mRNA from the endogenous mouse message. As indicated in Figure 3C, clone A2,
expressed high levels of human ATBFl-A mFNA when cells were cultured in both
GM or DM. Clones A6 and A19 expressed relatively lower levels of human
ATBF 1-A mRNA compared to A2 clones.
Once ectopic ATBF1-A expression in C2C12 cells had been established, the
ability of these cells to differentiate was monitored and the extent of myogenesis
was assessed by the formation of multinucleated myotubes and MHC
imrnunocytochemistry. Nontransfected C2C 12 cells and cells transfected with the
empty expression vector showed extensive formation of MHC-positive myotubes
after 4 days treatment with DM (Fig 4A). In contrast, A2 and A6 cultures generated
no MHC-positive cells, and A19 cells contained only a small number of MHC-
positive cells, mostly mononucleated (Figure 4A). Even treatments with DM up to
2 weeks did not produce multinucleated myotubes in the ATBF1-A-transfected cells
(data not shown). These results indicate that ATBFl-A overexpression inhibits the differentiation of C2C12 cells. To quant@ these effects of forced ATBFl-A
expression, the percentage of nuclei present in multinucleated myotubes per field of view was determined and expressed as the fusion index (Yoshida et al., 1996). As shown in Figure 4B, fusion indices were reduced 99. 90. and 80% in A2. A6 and
A19 cells, respectively, as compared to nontransfected or vector-msfected cells.
The progression of myogenic differentiation has been shown to be marked by the activation of the bHLH MRFs. including MyoD and myogenin, and the Figure 4. Constitutive expression of ATBFl-A inhibits myogenic differentiation of
C2C12 cells.
(A). ATBF 1 -A transfectants (A2. A6. A19) were grown either in growth medium (GM) or in differentiation medium for 4 days (DM) and analysed for
MHC expression by immunocytochemistry. C 12, untransfected C2C 12 cells;
Vector, C2C12 cells transfected with an empty expression vector.
(B) . Fusion indices of ATBF 1A transfected cells (A2. A6 and A 19). untransfected C2C12 and vector transfected cells. The mean percentage of nuclei present in multinucleated myotubes after 4 days growth in DM from 5 random fields was determined. This graph is a representation of a single experiment repeated in triplicate. Error bars correspond to the standard error of the mean
(SEW -
63 down-regulation of the inhibitory HLH proteins, such as Id3 (Lassar et al., 1994;
Olson, 1992; Walsh, 1997; Yun and Wold, 1996). Northern blot analysis confirmed that induction of differentiation of nontransfected or vector-transfected
C2C 12 cells was accompanied by an elevation in expression of myogenin and
MyoD rnRNA and a reduction in Id3 rnRNA levels (Figure 5, lanes 1-4). The expression patterns for MyoD mRNA during C2C 12 differentiation is variable from report to report. For example, in some investigators demonstrate an induction of MyoD mRNA expression during differentiation Bennett and Tonks,
1997; Cenciarelli et al., 1999; and Grendinger et al., 1998), while others demonstrate no change in MyoD rnRNA levels (Epstein et al., 1995; Park et al.,
1992; Puri et al., 2000). and others even demonstrate a decrease in MyoD expression upon differentiation (Lassar et al.. 1989; Miller, 1990). These reasons for these discrepancies is unknown and may result from differences in growth conditions of the cells. In A6 cells, the expression of MyoD and myogenin appeared extinguished in GM and the transfer of the cells to DM was not accompanied by increases in their expression (Figure 5, lanes 5 and 6).
Transfection of MyoD into A6 cells was unable to rescue the impaired myogenic properties of these cells. suggesting that the effect of ATBF1-A is downstream from
MyoD expression (data not shown). Expression of MyoD is subject to positive feedback regulation by myogenin (Yun and Wold, 1996). Therefore, the lack of
MyoD induction in A6 cells may be a consequence of decreased levels of rnyogenin mRNA and the failure of myogenin to activate MyoD, rather than a direct Figure 5. Alteration in the expression patterns of myogenic and cell cycle regulatory factors in ATBF 1-A transfected cells.
Total RNA (10 pg) was isolated from C2C12. Vector-transfected (Vector) and C 12A6 myoblasts in growth medium (G) and from cells treated with differentiation medium for 4 days 0).Membranes were hybridized to 32P-labelled cDNA probes corresponding to myogenin, MyoD, Id3, p2 1CIPINYM1 . cyclin D 1, and GAPDH mRNAs. Myogenin MyoD Id3 regulation by ATBF1-A. Expression of Id3 mRNA in undifferentiated A6 cells was found to be much higher than that in control cells. The level of Id3 expression in A6 cells was reduced after transfer to DM but remained higher than those in control cells (Figure 5, lanes 5 and 6). Comparable expression patterns for MyoD, myogenin and Id3 were obtained in the analysis of A2 and A19 clones. These data indicate that ATBF 1 -A may be inhibiting myogenesis of C2C 12 cells by inactivating expression of a positive regulator, myogenin, as well as enhancing expression of negative regulators of differentiation.
It has been shown that in addition to the activation of MRFs and the inhibition of Id3, induction of C2C12 differentiation is accompanied by changes in expression of proteins regulating the cell cycle, including induction of p2 lCIP1"NAF1 and a reduction of cyclin DL mRNA (Lassar et al., 1994; Missero et al., 1995;
Walsh, 1997). We observed that these changes took place in nontransfected or vector-transfected cells after transfer from GM to DM (Figure 5, lanes 1-4).
However, A6 cells displayed neither an induction of p2 lC'PINAflmRNA nor a reduction of cyclin D 1 rnRNA after transfer to DM (Figure 5, lanes 5 and 6). We also observed that cyclin D 1 mRNA Levels in A6 in GM were much higher than those in control cells. After the cells were transferred to DM, cyclin D 1 expression was reduced, but its levels were still comparable to that in undifferentiated, control cells (Figure 5). These results suggest that ATBF 1-A may block exit from the cell cycle or other early events required for terminal differentiation of C2C12 myo blasts. 67
Finally, the expression of ATBFl-A protein in C2C12 and A6 cells treated with either GM or DM was determined by Western Blotting. The large size of the predicted ATBF 1-A protein presented several technical challenges that needed to be resolved in order to analyse ATBFl-A protein expression. For example, this large protein may be subjected to increased proteolyhc degradation during cell harvesting as the probability of protease target sites would increase with the size of the protein. An extensive combination of protease inhibitors were added to the cell lysis buffer to minimize protein degradation. In addition the electrophoretic separation and transfer to nitrocellulose membrane may occur inefficiently with large proteins. Therefore these challenges may pose limits in the quantitative analysis and interpretation of ATBF 1A protein levels since the entire ATBFl-A protein population may not be represented on an immunoblot. Taking these limitations into consideration, we observed that ATBFl-A protein was expressed at much higher leveIs in undifferentiated C2C12 cells in GM compared to differentiated cells in DM Figure 6A, lanes 1 and 2). ATBF1-A protein levels were also elevated in A6 cells treated with GM compared to cells treated with DM
(Figure 6A, lanes 3 and 4). However the extent of this decrease was less than that observed in control C2C12 cells. The D 1 antibody recognizes both human and mouse ATBF1-A. Therefore the ATBF1-A protein detected in A6 cells in GM likely represents translation products from both the endogenous mouse gene and transfected human cDNA. Since the levels of endogenous mouse ATBFl-A mRNA in A6 cells decrease in response to DM treatment (see Figure 26), the ATBFl-A 68
Figure 6. Western blot analysis of ATBFI-A protein in differentiating C2C12 cells.
(A) Protein samples (30 pg) from C2C 12 and C12A6 cell homogenates treated with growth medium (G) or with differentiation medium @) for 4 days were separated on a 2-12% SDS-polyacrylarnide gradient gel and transferred to a nitrocellulose membrane. ATBF 1-A protein was detected by irnrnunoblotting with anti-ATBF 1 polyclonal antibody (D 1) and visualized through enhanced chemiluminescence. Parallel blots were analysed for desmin, cdk-4, p27 and a- tub& protein expression.
@) Total cellular lysates (30 pg) from A6 cells and 10T1/2 cells transiently transfected with an HA-tagged-ATBF1-A expression vector were separated on a 2-
12% SDS-polyacrylamide gradient gels and transferred to nitrocellulose membranes as described in "Materials and Methods." Lane 5, lysates from A6 cells transferred to nitrocellulose and immunoblotted with anti-ATBF 1 antibodies @ 1). Lane 6 and
7, lysates from A6 cells and HA-transfected cells, respectively, immunoblotted with anti-HA antibodies (1 2CA5) .
(C) COS cells were transiently transfected with 18 pg of either pc3-HA-
ATBFl -A expression vector (lane 8) or empty pcDNA3 expression vector (lane 9).
Cellular lysates were harvested and electrophoretically separated on a 4.5% polyacrylamide gel. Membranes were irnmuno blotted with anti-ATBF 1 antibodies
(a-D1) as described above. Desmin CDK-4
Western Blot aHD4- a-HA 70
protein detected in A6 cells in DM is likely derived from the transfected ATBF 1-A
cDNA. The down-regulation of the endogenous mouse ATBF1-A protein may
account for the decrease in protein levels observed between A6 cells cultured in
GM versus DM. In addition, this decrease in ATBFl-A protein may be attributed
to regulation of the pgk promoter in response to DM treatment, as human ATBFl-
A mRNA levels decreased modestly in A6 cells (Figure 3B).
To venfy that the transfected expression vectors can indeed produce a full
length ATBF 1-A protein, we monitored the expression of an epitope-tagged
ATBF 1-A expression construct in transfected cells. 10T1/2 cells were transfected with an HA-tagged ATBF1-A expression vector and lysates were irnrnunoblotted with anti-HA antibodies. As indicated in Figure 6B, the anti-HA antibodies
recognize a band in cells transfected with the HA-ATBF1-A vector that migrates with approximately the same molecular weight as ATBF 1-A (compare lane 7 &th
5). These antibodies did not react against proteins that did not receive the HA-
ATBF1 -A vector (Figure 6B, lane 6). To further venfy the production of ATBF 1-A protein from expression vectors, we transfected COS cells with an ATBF1-A expression vector and monitored protein production by irnmunoblotting with anti-
ATBFl antibodies (a-D1). Again, a high molecular weight was detected in cells transfected with ATBF 1-A expression vector. but not in cells that received the empty expression vector (Figure 6C, lanes 8 and 9). Together, these transfection experiments demonstrate that a fd-length ATBF 1-A protein can be produced. The expression of muscle structural proteins and cell cycle regulatory proteins were also examined in control C2C12 and A6 cells. The intermediate fdarnent protein, desmin, is an early marker of muscle dBerentiation (Bischoff,
1994; Kablar et al.. 1997). In C2C12 cells this protein is detected in cells treated with DM but not GM, whereas in A6 cells desmin is not expressed (Figure 6A).
Again, these results indicated that ectopic ATBF1 -A expression in C2C 12 cells inhibited the activation of muscle-specific genes. Expression of cdk4 and the Cdk- inhibitor, p27, were also examined in CZC12 and A6 cells. Over the course of myogenic differentiation, cdk4 and p27 protein levels remain constant (Guo and
Walsh. 1997; Wanget al.. 1997). As shown inFigure 6A, there is no change in the cdk4 levels between undifferentiated and differentiated C2C12 cells. In A6 cells, cdk4 protein levels are abundant in GM, but are prominently decreased in
DM (Figure 6A). Since cdk4 is normally not responsive to induction of differentiation, ectopic ATBF 1-A expression had disregulated the expressior. pattern of this protein. On the other hand, p27 protein levels remain relatively unchanged in C2C12 cells cultured in GM and DM, whereas this protein levels in
A6 cells are slightly elevated in comparison to control cells and increased slightly when cells were switched from GM to DM (Figure 6A). Thus in A6 cells treated with DM p27 levels are elevated, while p21~'~"~'levels are reduced in comparison to control C2C12 cells. 72
3.3 ATBFl -A expressing CZCl2 cells undergo growth arrest in response to DM.
To cordi~rmwhether ATBF1-A functions to prevent the normal cell cycle arrest required for rnyogenic differentiation, we examined BrdU incorporation over the course of differentiation as an indicator of entry into S phase. In these experiments, cells were grown to 80% confluency in GM and medium was replaced with DM. Differentiation medium was removed at the indicated time points and replaced with DM containing BrdU for 4 hours. As expected, the percentage of BrdU incorporation decreased in control cells over the four day differentiation period. indicating that the proportion of cells entering S-phase had declined (Figure 7 , shaded bars). A6 cells also displayed a decrease, although to a lesser extent (Figure 7, open bars). At 4 days DM, there was a three fold higher percentage of A6 cells incorporating BrdU than the control cells. When DM was replaced with fresh GM on day 4, the proportion of A6 cells incorporating BrdU retuned to levels observed in GM, whereas the percentage of control cells incorporating BrdU only reached levels 60% of the levels in GM. These data suggest A6 cells, like control cells, entered cell cycle arrest, but are much more capable of reentering into the S-phase (or growth phase) than their control counterparts.
3.4 The ATBF1-B isoform enhances myogenic differentiation of C2ClZ cells.
Although we could not detect expression of the ATBF1-B isoform in C2C12 cells either before or after differentiation, we were interested in knowing whether Figure 7. BrdU incorporation profiles of differentiating and sew-stimulated
C2C12 cells.
Entry into S-phase of the cell cycle in C2C12 (shaded bars) and A6 cells
(open bars) following the induction of differentiation was measured as function of
BrdU incorporation. Cells were grown on glass coverslips and treated with DM or
GM for the indicated time period followed by a 4 hour pulse of BrdU labelling.
Cells were fxed and analysed for BrdU incorporation by indirect imrnuno- fluorescence and counterstained with DAPI. % cells incorporating BrdU was expressed as the percentage of BrdU-positive nuclei present per field of view. The data presented is an average of 3 fields of view consisting of at least 300 nuclei per field. Error bars correspond to the SEM. I r-l 75
this isoforrn has any function in myogenic differentiation of C2C12 cells. For this
purpose, we stably transfected C2C 12 cells with a vector expressing the human
ATBFl-B cDNA. Stable integration of the ATBF 1-B expression vector into the
genome was verified by Southern blotting as described for the ATBFl-A
transfectants (Figure 8A). As with A2 cells, clones B4. B5 and B6 produced larger
fragment than the predicted 7.1 kb band. Again the nature of this band in
unknown but may represent incomplete digestion of the genomic ATBFl DNA.
The colonies containing the transfected ATBF1-B vector were also examined for
expression of human ATBF1-B mRNA by RNase protection assay. All colonies
that contained the ATBF1-B vector (B3. B4, B5, B6, and B8) expressed the
exogenous ATBFL-B mRNA at relatively uniform levels, although clone B3
exhibited the highest levels and B5, the lowest (Figure 8B). Three clonal cell
lines, B3. B5 and B6, were selected and andysed for the ability to differentiate in
DM. We found that the fusion indices of these cells after 4 days in DM were 25 to
40% higher than those of nontransfected or vector-transfected cells (Figure 8C).
Time-course analysis of MHC expression was compared between B6 and vector-transfected cells. The results showed that B6 cells generated MHC-positive
multinucleated cells within 24 hours after the transfer to DM (Figure 9, middle right panel) and formed well-advanced rnyotubes by 36 hours (Figure 9, bottom right panel). In contrast, vector transfected cells were devoid of MHC-positive cells at 24 hours (Figure 9, middle left panel) and exhibited early stages of myotube formation only after 36 hours of incubation in DM (Figure 9 bottom left panel). Figure 8. Enhanced myogenic differentiation in ATBFI-B transfected CZCIZ cells.
(A). The stable integration of the ATBFl-B expression vector was addressed by Southern blot analysis.
(B) . ATBF 1B expression in stably transfected C2C 12 cells was analysed
RNase protection assay. B2 ,B3,B4, B5, B6 and B8 correspond to individual ATBF 1-
B transfected colonies. P, undigested probe. V, cells transfected with empty expression vector.
(C) . Untransfected C2C 12 cells, cells transfected with the empty expression
(Vector) or ATBF 1-B @3, B5, B6) were allowed to differentiate for 4 days in DM and irnrnunostained for MHC with MF20, and the fusion index was calculated by determining the percentage of nuclei found in multinucleated myotubes in a field of view. Error bars correspond to the SEM. I ; Probe TAA ATG I 7178 bp Figure 9. Early onset of MHC expression detected in ATBF1 -B-transfected cells.
Vector-transfected and B6 C2C 12 cells were induced to differentiate by transfer to DM. Cells were fuced at 18, 24 and 36 hours after treatment with DM and MHC expression was detected by MF20 immunocytochemistry. Arrows indicate MHC-positive myo tubes.
80 These results show that the ATBFl-B isofom promotes myogenic differentiation of
C2C12 cells.
Since ATBF1-B expressing cells exhibited an enhanced degree of differentiation we wished to know whether these cells displayed an altered expression pattern of molecular markers indicative of terminal differentiation. The expression of myogenin and ~2 lCmlMAF1rnRNAs were examined in ATBF1-B- transfected cells by Northern Blotting. As demonstrated in Figure 10, both vector- and ATBF1-B-transfected cells treated with DM for 4 days expressed myogenin and p2 1CIP1'vAF' rnRNA. Expression of myogenin and p2 lCIPIM'AF1mRNA levels were comparable between vector- and ATBF1-B-transfected cells. However, clone B4 did exhibit reduced myogenin expression in DM compared to controls. None of the ATBF1-B colonies expressed myogenin rnRNA in GM. These data suggest that the enhanced myogenic differentiation observed in these cells was not due to elevated myogenin expression. It should be noted however, that myogenin expression in ATBF 1-B-transfected cells may be activated at a time point earlier than what we had analysed.
3.5 MyoD induced transdifferentiation of 10Tlf2 fibroblasts to myotubes, is inhibited by ATBFI-A and enhanced by ATBFl-B.
10T 112 fibroblasts have been shown to transdifferentiate to the myogenic lineage when transfected with MyoD (Davis et al., 1987). We examined the effect of forced expression of ATBF1-A or -B on the MyoD-dependent myogenic Figure 10. Normal expression patterns of differentiation markers in ATBFI-B expressing cells.
Myogenin and p2 1 cplmMLmRNA levels were compared between C2C 12
cells and ATBF1-B (B4. B5. B6 and B8) -transfected cells by Northern blotting.
RNA was isolated form cells treated with growth medium (G)or with
differentiation medium (D) for 4 days. Blots were hybridized with myogenin.
p2lCPwAFLcDNA probes and with a GAPDH probe to monitor RNA loading.
83 conversion of these cells. In this assay, 10T112 cells were transfected with a MyoD expression vector with or without the ATBFl-A or -B expression vector. The cells were then transferred to DM, incubated for 4 days and stained for MHC expression (Figure 11 A). We found that co transfection of ATBF 1-A and MyoD in an equal amount resulted in a 60% decrease in the number of MHC-positive cells as compared to transfection of MyoD alone (Figure 11B). When cells were transfected with ATBF 1-A and MyoD at a ratio of 5: 1, the number of MHC- positive cells was reduced by 80% (Figure 11B). In contrast, cotransfection of
ATBF 1-B and MyoD at a ratio of 1 :1 resulted in a 40% increase in the number of
MHC-positive cells (Figure 11B). Co-transfection of vectors antisense to ATBF 1-A with MyoD was also found to increase the number of MHC-positive cells by 35%
(Figure 11B). These results show that ATBFl-A inhibits MyoD-dependent myogenic conversion of 10T 1/2 fibroblasts, whereas ATBF 1-B promotes this conversion. Antisense ATBFl may reduce levels of endogenous ATBFl-A present in these cells (Figure 2C) and relieve any myogenic repression exerted by ATBF 1-A.
The similarity of the effects of ATBF 1-B to antisense ATBFl suggests that ATBFI-B exerts the positive effect by counteracting the negative effect of ATBF1-A.
3.6 Activation of the MRF4 promoter by MyoD is inhibited by ATBF1-A.
In C2C 12 cells transfected with ATBF 1-A, we observed a decrease in muscle-specific gene expression including MyoD and myogenin (Figures 3 and 4).
To determine whether ATBFl-A affected the expression of MyoD-MRFs directly, Figure 11. ATBFl regulates the MyoD-mediated myogenic conversion of 10T112 cells.
A. Photomicrographs of MHC staining in 10T1/2 mouse fibroblast cells transfected with 1 pg of pgkMyoD and 1 pg of pPOP (Control). pPOP-ME(F)
(ATBF 1-A), pPOP-E (ATBFl-B) or pPOP-ME(R) (Antisense) DNAs. Following transfection, the cells were switched to DM for 48 hours to induce differentiation and then irnmunostained for MHC expression.
B. Graphical representation of ATBF 1 regulation of 10T1/2 myogenic differentiation. 10T1/2 were transfected as with 1 pg of pgkMyoD. and either 1 pg
ATBF 1-A, 5 pg ATBF1-A, 1 pg ATBF 1-B, 1 pg antisense ATBF 1. In each case, the total amount of DNA transfected was adjusted to 6 pg with pPOP vector. Five random fields of view from each group were scored for the presence of MHC positive cells. Values obtained in the Control group were set at 100 percent. Error bars indicate the standard error of the mean.
86
we examined the effect of transfected ATBFl-A on the activation of the luciferase gene linked to the MRF4 promoter. This promoter consists of a 390 bp upstream fragment of the mouse MRF4 gene containing one MEF2 and two E-box cis-acting elements which have been demonstrated to direct muscle-specific transcription
(Figure 12A: Black et al.. 1995). The expression of luciferase from this plasmid was low in 10T1/2 fibroblast cells but was increased 35-fold in response to MyoD transfection (Figure 12B). Co-transfection of ATBF 1-A and MyoD at a 1: 1 ratio resulted in an 80% decrease in luciferase activity (Figure 12B). ATBF1-B, on the other hand, had little effect on the MyoD-induced expression of luciferase (Figure
I2B). These data indicate that ATBF 1-A inhibits MyoD-induced activation of the
MRF4-promoter.
3.7 An AT-rich motif is necessary for ATBF1 -A-mediated repression.
To determine the cis-acting elements mediating ATBF 1-A inhibition, we analysed the effect of ATBF1-A on MyoD-induced activation of the 4Rt.k-luc reporter, which is composed of four tandem repeats of the muscle creatine kinase
(MCK) E-box elements upstream of the tk minimal promoter. 10T1/2 cells were cotransfected with this reporter and MyoD with or without ATBFl-A or B. As indicated in Figure 13, the 4Rtk-luc reporter exhibited very little activity when transfected alone in 10T1/2 cells. However, the addition of MyoD induced this reporter about 350 fold. In contrast to MRF4-luc, activation of 4Rtk-luc by MyoD was inhibited by ATBF 1-A by only 15% (Figure 13). Like MRF4-Iuc, ATBF 1-B Figure 12. ATBFl-A inhibits MyoD-dependent transcriptional activation of the
MRF4 promoter.
(A) Schematic representation of cis-acting element present in the mouse
MW4 promoter.
(B) 10T112 cells were transiently transfected with 0.5 pg MMR-luc reporter, either 0.5 pg MyoD expression vector (pgkMyoD) or empty expression vector
@POP), along with 0.5 pg of ATBF1-A (pPOP-A), ATBFI-B @POP-B) or an additional 0.5 pg of pPOP. A B-galactosidase expression vector. pCH110 (0.5 pg) was included in each transfection. Following transfection, cells were allowed to differentiate 2 days in DM and protein harvested. Luciferase activity was normalized to P-galactosidase activity and expressed relative to control (MRF-luc) .
The data presented are an average from 4 independent experiments utilizing two independent plasmid prepartions. Error bars correspond to the standard error of the mean. - - I A MRF'4-Luc -- - LUC E2 MEF2/ AT El TATA Rich
pPOP MyoD MyoD MyoD + + ATIBFI-A ATBFI-B Figure 13. The E-box element is resistant to ATBFl-A repression.
10T1/2 cells were transfected with 4Rtk-luc reporter (0 -5 pg) , along with either 0 -5 p.g pggkMyD or no activator at all (0 -5 pg empty pPOP expression vector.
In addition, these cells were co-transfected with 5 pg of empty pPOP expression vector (open bars), ATBF 1-A (solid bars) or ATBF I-B (shaded ban). All cells were transfected with the pCHl 10 P-galactosidase vector (0.5 pg). Cellular lysate were harvested 48 hours after transfection. Luciferase activity was normalized to P-gal and expressed relative to cells ~ansfectedwith 4Rtk-luc alone. Emor bars indicate the standard error of the mean. Luc MCK E-Box
Vector MyoD Activator 91 had Little effect on MyoD-mediated activation of 4Rtk-luc (Figure 13). Since this reporter is composed solely of four E-box sites and lacks any identifiable AT-rich sequences (Weintraub et al., 1990), these data suggest that ATBF 1-A does not act directly on the E-box elements. The increased activity observed for the 4Rtk-luc reporter in comparison to the MRF4-luc reporter may be due to an increased affinity of MyoD for the 4Rtk promoter. Alternatively, endogenous ATBFl-A may bind to the AT-rich element in the MRF4- promoter and exert negative effects on this promoter. The removal of the AT-rich element in the 4Rtk-luc reporter may relieve this inhibitory effect.
3.8 ATBFI -A homeodomain 4 binds to the AT-rich element of the MRF4 promoter.
ATBFl has been demonstrated to bind AT- rich enhancer sequences through its fourth homeodomain motif (Morinaga et al., 1991; Yasuda et al., 1994). There is an AT rich sequence at position +1 1 to + 19, near the El E-box in the mouse
MRF4 gene that bears similarity to the ATBF 1 binding site in the AFP enhancer
(E3Iack et al., 1995; Yasuda et al., 1994). To examine whether ATBF1-A binds to the MRF4 promoter, we attempted gel mobility shift assays using ATBF1-A produced by in vitro transcription and translation in reticulocyte lysates or expression in bacteria. Due to difficulties in producing a full-length ATBF 1-A in this manner, we produced in vibo the C-terminal half of ATBFl-A from amino acid 2 162 to 3703, termed DM2 (see Figure 2 I), which contains the second, third 92 and fourth homeodomain and six zinc-fmger motifs. Incubation of DM2 with a
3Z~-labelledoligonucleotide probe corresponding to the AT-rich/E-box element of the MRF4 promoter (A/E probe) resulted in the formation of three binding complexes (Figure 14A, lane 2). The lower two bands were also observed when unprogrammed in vih-o translations reactions were examined and therefore represent non-specific complexes (data not shown). The formation of the top band was effectively inhibited by an excess of unlabelled probe (Figure 14A, lanes 3-5) but not by AP-2 probe (Figure 14A, lane 6), indicating that this band represents a specific complex formed between DM2 and the AT-rich/E-box element.
There are several potential DNA binding motifs in DM2 that may interact with the AT-rich/E-box region. Potential DNA binding sites may also exist outside of DMZ. To further delineate the portion of DM2 that binds to the A/E probe, portions of the ATBF1 -A molecule were expressed in bacteria (see below).
Mobility shift assays using these products demonstrated that a peptide containing the fourth homeodornain of ATBF I -A (amino acids 2792 to 3378, HD4) formed a complex when incubated with labelled A/E probe (Figure 14B, lane 8). The addition of antibodies raised against HD4 (a-D4)of ATBF 1-A resulted in a supenhifted complex, confirming the presence of HD4 in this A/E DNA:protein complex (Figure 14B, lane 10). Other regions of ATBF 1-A including the unique
N-terminus, several zinc-fmger clusters as well as homeodomains 1,2 and 3 did not bind to the AEprobe (K. Mihara, unpublished observations). Together these data indicate that HD4 is directly interacting with the MFS4 promoter. It is possible Figure 14. The fourth homeodomain of ATBFl-A binds to the AT rich/E-box
element of the MRF4 promoter.
(A). Gel mobility shift assays examining the DNA binding properties of the
ATBFZ-A C-terminus in vitro translated protein DMZ. "P-labelIed A/E
oligonucleotide probe was incubated alone (Lane 1) or with 5 pL of DM2
translation lysate (Lane 2-6). For competition experiments, either unlabelled A/E probe was added at a 10, 50 or 100 fold molar excess (lanes 3-5). or a 100 fold molar excess of unlabelled AP-2 was added to lysate (Lane 6). A specific complex formed by DM2 (Complex 1) and the non-specific ones formed by the reticulocyte lysate(Comp1ex 2) are noted.
) Bacterial-expressed fusion proteins (5 ng) corresponding to the fourth homeodomain (HD4) of ATBF1-A were incubated with labelled A/E probe (lane 8).
To each binding reaction, either an excess of cold competitor (lane 9). an antibody directed against HD4 (lane 10) or an anti-IgG antibody (lane 11) was added. The binding complex formed by HD4 as well as the supershifted complex are noted as
Complexes 3 and 4, respectively. Free probe corresponds to uncomplexed, labelled A/E probe. AE AP-2 Competitor - - A +
Antibody aD4 aIgG Competitor - - + - - HD4 -+++ +
Free Probe that the full-length ATBF 1-A protein is also interacting with MRF4 promoter through it fourth homeodomain. However as HD4 represents a small portion of the intact ATBF la protein (60 kDa vs 404 ma). the DNA-binding activity of this smaller protein may not accurately represent the DNA-binding activity of the fidl length ATBF 1A molecule.
To confirm the appropriate binding of the 60 kDa HD4 protein to the A/E probe, we performed W-cross-linking experiments. When HD4- 3ZP-labelledA/E probe binding reactions were exposed to W-light and separated on a 10% polyacrylamide gel. an approximately 70 kDa DNA-protein complex was detected
(Figure 15). This molecular weight is in agreement with the predicted molecular weight of the HD4 (60 kDa) and the 25 bp A/E probe (16 kDa ) complex. The fact that HD4 was cross-linked to the A/E probe suggests that the binding is specific since X-linking is an inefficient reaction. A lower molecular weight complex was also detected in the cross linking reaction (Figure 15). This band may represent a truncated fom of HD4 since multiple bands were also observed in the A/E gel mobility shift and supershift reactions (Figure 14).
The E 1 E-box in the A/E probe is a potential binding site for MyoD-MRFs and is separated by the AT-rich element by only 2 bp. Therefore HD4 may binding to this sequence and displace MyoD-MRFs and this displacement may inhibit the activity of the MRF4 promoter. Initially we wanted to identify whether MyoD could indeed bind to this promoter. Gel mobility shift assays were performed using nuclear extracts isolated from undifferentiated or differentiated C2C12 cells Figure 15. W-crossllnkingof HD4 binding complexes to the MRF4 NE probe.
HD4 protein (20ng) was incubated with the 32~-labelledA/E probe and incubated under a W-transilluminator and exposed to a wavelength of 254 nm for
1 hour. As a control, a parallel binding reaction was not exposed to W-radiation.
The samples were the electrophoretically separated in a 10% polyacrylarnide gel and exposed to X-ray film. Two X-linked complexes, one corresponding to HD4-
A/E pro be, were observed following crosslinking, while no complexes were observed in the absence of UV-crosshking. probe incubated with labelled A/E probe. Undifferentiated C2C12 extracts produced only a single fast migrating complex (Figure 16, Band B) , while differentiated extracts produced a larger. slower migrating complex (Figure 16. Band A) in addition to the lower molecular weight band observed in cells treated with GM
(Figure 16, compare lanes 2 and 3). However, when HD4 was added to C2C12 nuclear extracts the formation of the differentiation-specific complex did not occur
(Figure 16, lane 9). In an attempt to ascertain the identity of this differentiation- specific band. antibodies directed against MyoD were added to the reaction mixtures. No supershift of the band was observed indicating that MyoD is not present in this differentiation-specific complex (Figure 16 lanes 4 and 5). These results indicate that MyoD is activating the MRF4 promoter in an indirect manner or its interacting with sites are outside of the El E-box.
The displacement of regulatory factors from the MRF4 promoter by HD4 was examined in further detail. As indicated in Figure 17, the incubation of nuclear extracts from differentiated CZC12 cells with the AIE probe produced two shifted complexes (Figure 17, lane 2). The fast-migrating band represents nonmuscle-specific complexes. since nuclear extracts derived from 10T1/2 fibroblasts yielded this band (Figure 17, lanes 1 and 2). The slow-migrating band likely contains myogenin-El complexes since its formation was disrupted by the addition of myogenin antibodies to the reaction mixture (Figure 17. lane 6). It is possible that the addition of a-myogenin antibody prevented the formation of a myogenin-E/A box complex, and thus no supenhift was observed. The myogenin Figure 16. A muscle ditrerentiation-specific complex binds the MRF4 A/E probe.
Nuclear extracts prepared from C2C12 cells treated with GM (G) or induced to differentiate by culturing DM for 4 days (D) were used for mobility shift assays. A differentiation-specific (A) and a nonmuscle-specific (B) protein-
DNA complex were noted with arrows. Supershift assays were performed by adding MyoD (lanes 4 and 5) or c-myc (lanes 6 and 7) antibodies to iden* proteins present in the shifted bands. The muscle-specific bands were competed away by the addition of 5 ng of HD4 to the C2C12 NX (lanes 8 and 9).
Figure 17. HD4 binding to the A/E probe disrupts myogenin-E-box interactions.
Nuclear extracts from 10T112 cells (he 1) or terminally differentiated
C2C12 (C2) myotubes (lane 2) were incubated with labelled A/E probe. C2
myotube nuclear extracts were supplemented with HD4 where indicated (lanes 3
and 4). HD4 (5ng) was added to labelled ALE probe in lane 5. Antibodies raised
against myogenin (a-myogenin) or HD4 (a-D4) were added to C2 myotube extracts
(lanes 6 and 7, respectively) and to C2 myotube extracts supplemented with HD4
(lanes 8 and 9, respectively) DNA-protein complexes formed by myogenin and HD
for are noted. HD4* corresponds to the altered HD4 complex formed in the presence of C2 myotube nuclear extracts. a-D4 a-myogenin 4 (ng) C2C12 DM 1OT1/2
Myogenin- HD4-
Complex 1 103 band disappeared when HD4 was added to the reaction mixture (Figure 17, lanes
2-4). Instead, two new bands were formed (Figure 17A. lanes 3 and 4). The faster of the two bands represents the HD4 complex described above. The slow migration of this band, marked HD4* in Figure 17, may be due to the formation of complexes containing HD4 post-translationally modified or dimerized or bound to proteins in C2C 12 rnyotube extracts. Although the addition of a-HD4antibodies produced a supershifted band Figure 17, lane 9), the intensity of the HD4-A/E probe complexes did not reduce significantly. It is possible that the affhity of the a-HD4 is low and can not recognize all of the HD4 bound to the A,E probe.
These datz suggest that ATBF1-A may bind to the MRF4 promoter through the fourth homeodomain, preventing myogenin from associating with its E 1 E-box site.
The binding to and the displacement of myogenin from the A/E element of the MRF4 promoter is specifc to the HD4 region of ATBF1-A. As indicated in
Figure 18, NH, an ATBFl -A N-terminal fusion protein containing four zinc-fmger motifs (amino acids 1 to 550), did not bind to the AIE probe as did HD4 (compare lanes 3 and 4). Furthermore, the addition 5 ng of NH to C2C12 myotube nuclear extracts did not displace myogenin from this E box element Figure 18, compare lanes 5 with 6 and 7 with 8). These data suggest the observed displacement of myogenin from A/E probe was due to specific interactions between AE probe and
HD4. However. this binding may not entirely explain the ATBF1-A inhibitory mechanism since the fourth homeodomain is also present in the ATBF1-B isoform which does not inhibit MRF4 promoter activation. In addition, preventing Figure 18. Displacement of myogenin-E-box interactions is speac to HD4 of
ATBF 1-A.
Differentiated C2C 12 nuclear extracts (C2C 12) were incubated with 32P- labelled A/E probe and with HD4 or an ATBF1-A N-terminal region (NH) bacterial expressed fusion protein. Supershift assays were performed by the addition of myogenin (Lanes 7 and 8) or D4 (Lanes 8 and 10) antibodies to binding reactions.
The supershifted HD4 complex is designated. The DNA-protein complex corresponding to the myogenin:E-box interaction is indicated, as is the nonmuscle- specific band (complex 1) and HDCspecific complexes. HD4* corresponds to the slower migrating band observed when HD4 is added to C2C12 NX. HD4 supershift Myogenin HD4+ Complex2 HD4 Complex 1
Probe 106
myogenin from binding the El site may not disrupt promoter activity since there is
a second E-box site located further upstream and only one functional E-box
element is required to support muscle-specific activation of the MRF4 promoter
(Black et al., 1995).
Although ATBF1-A HD4 binds to the A/Eprobe, it is unclear whether it
interacts specificaIly with the AT-rich sequence, the E-box sequence or both. To
address this issue, we generated oligonucleotide probes corresponding to the AT-
rich or E-box portions of the MRF4 promoter to use in gel mobility shift assays.
When HD4 was incubated with 32~-labelledAT-rich probe, a DNA-protein
complex was formed that was efficiently competed with unlabelled AT-rich probe
(Figure 19A, lanes 1 and 2). The formation of this complex was slightly affected by
an excess of unlabelled E-box probe (Figure 19A, lane 4). No bands were formed when HD4 was incubated with 32P-labelledE-box probe (data not shown). An
excess of a consensus MEF2 probe, on the other hand, reduced the intensity of this
band suggesting that ATBF 1-A binds to the MEF2 site (Figure 19A. lane 3). The
specificity of myogenin binding to the El probe was also examined. As
demonstrated in Figure 19B,the formation of the slower migrating myogenin
complex was competed by an excess of unlabelled A/E probe but not with
unrelated probes such as MEFZ or AP2. The faster migrating non-specific band was only slightly reduced in the presence of an excess of cold probe (Figure 19B). Figure 19. HD4 binds to the AT-rich portion of the MRF4 promoter.
(A) HD4 was incubated with labelled A/E probe and binding was analysed by EMSA. To determine whether HD4 binds to the AT-rich region or the E-box element of the A/E probe, unlabelled oligonucleotides corresponding to the AT-rich region or the E-box, as well as the MEF2 sites were added as competitors to the binding reaction. The HD4 complex was formed when no competitor DNA was added (-) and this band was competed away by an excess of AT probe but not by the E-box probe.
) The specifcity of myogenin binding to the A/E was analysed in a similar manner described in (A). EMSA were performed by incubating C2C12 nuclear extracts with labelled A/E probe (-) along with unlabelled AE,MEFZ, or
AP-2 probes. Probe
Competitor
f-- myogenin
Probe 109 3.9 The AT-rich MEFZ element represents an additional ATBFl-A binding site.
In addition to the E-box element, a second a major regulatory element
found in genes expressed in a muscle specific fashion is the AT-rich MEF2 site.
This site is found in almost all myogenic genes and is often positioned in close
association to E-box elements (Gossett et al., 1989; Wright et al., 1991). MEFZ
binds to an its element adjacent to the TATA box in the MRF4 promoter. and
along with myogenin binding to the E 1 E-box, synergistically activates transcription
of the MRF4 gene Black et al., 1995; Naidu et al., 1995). However, MEF2 does
not bind to the AT rich element of the El E-box (Naidu et al., 1995). Wethe
AT-rich binding site of ATBF1-A is similar to the MEFZ site, it does not match the
consensus binding sequence, CTA(A/T) ,TA(G/A) . However the exact binding consensus for ATBF1-A has not been established, therefore we can not rule out the possibility of ATBFl-A binding to a subset of MEFZ elements. To this end, we isolated a fragment of the MRF4 promoter that contains the MEF2 site and utilized it for a probe in gel mobility shift assays. When this region was incubated with
HD4 protein, several bands displaying different mobilities were observed (Figure
20A). Increasing the HD4 protein concentration in the reaction led to an increase in the abundance of the higher molecular weight DNA-protein complexes formed, suggesting that HD4 binds as multimers to this MEF2-containing fragment. The ability of homeodomain proteins to homodimerize and bind DNA has been documented for MEFZ, OW2 and Pit-1 proteins (Black and Olson, 1998; Briata et d., 1999; Jacobson et al.. 1997). Alternatively, the formation of the slow Figure 20. The AT-rich MEF2 element may represents a novel target for HD4 binding.
(A). HD4 binds to a region of the MRF4 promoter outside of the AT-rich/
E-box region. A 290 bp fragment of the MRF4 promoter which contains the
MEF2 element was isolated and used as a probe for EMSA. Increasing amounts
HD4 protein was added to the binding reaction (Lanes 2-5). In the absence of protein, no binding activity was observed (Lane 1).
(B). Binding of HD4 to the MEF2 element. An oligonucleotide corresponding to the MEF2 consensus binding site was used for a probe in EMSA.
MEF2C and luciferase protein were produced by in vitro transcription/translation reactions and HD4 was expressed in bacteria. The luciferase reaction was included to evaluate non-specific complexes generated by the Lysate alone. The proteins were added to the labelled MEF2 probe and analysed on 5% polyacrylarnide gel.
Lane 1, contains no protein; Lane 2, luciferase control reaction, Lane 3, MEF2C protein: Lane 4, HD4; Lanes. MEFZC and HD4. The protein-DNA complexes for
MEF2C and HD4 are indicated, by brackets and arrows, respectively. HD4 Protein 1
MEF HD4
Free 112 migrating band may have arisen from proteins in the bacterial lysate binding to the probe, or from multiple HD4 binding to more than one site on this large DNA probe.
We next determined whether ATBF 1-A interact. specifically to MEF2 sites and prevents MEFZ proteins from binding to this site. An oligonucleotide corresponding to the MEF2 consensus site was used for mobility shift assays.
MEF2C protein, produced through in vitro transcription and translation reactions, was incubated with the 32~-labelledconsensus MEF2 probe. As indicated in Figure
20B, this reaction formed a protein-DNA complex (Lane 8). A band was also observed when HD4 was incubated with MEFZ probe, indicating that ATBF1-A binds to the MEF2 site (Lane 10). However, the addition of HD4 to the MEF2C binding reaction did not disrupt the formation of MEFZC-DNA complexes. HD4 displays a weak affimity for the MEF2 element, which may explain why HD4 can not disrupt the MEF2C-MEF2 complex. These results indicate that although HD4 is capable of binding to the MEF2 element, it does not inhibit MEFZC from binding.
3.10 The ATBFl-A amino-terminus is necessary but not suffZaent to inhibit
MyoD activity.
To further idenw important functional domains required for inhibition. we constructed a series of ATBFl-A deletions and examined whether they would affect
MyoD-mediated activation of the MRF4-luc reporter. As shown previously, 113
ATBF 1-A inhibited MRF4-luc expression. while ATBF 1-B had no effect (Figure
12). indicating that the N-terrninal920- amino acid region of ATBFl-A is involved
in inhibition of the MRF4 promoter. However, expression of the ATBF1-A N-
terminal 1311 amino acid region, M3, alone did not inhibit but rather enhanced luciferase activiv of the MRF4-luc reporter (Figure 2 1). Similarly, the internal region carrying a cluster of 13 zinc fmgers (amino acids 92 0-2 16 2), DM 12, enhanced luciferase expression mediated by MyoD (Figure 2 1). The C-terminal half of ATBFl-A including homeodomains 2, 3 and 4 and zinc fmgen 18-23, DM2 had no effect on luciferase expression. These results suggest that the ATBF1-A specific region is necessary but not suffcient to mediate the inhibitory activity observed with the full-length molecule. These regions likely need the ATBF1-A
DNA-binding domain to mediate their inhibitory activity. To determine ATBF1-A- specific region involved, we tested the effect of removing short segments of the N- terminus- A deletion of the first 113 amino acids of the ATBFI-A, ANN, eliminated inhibitory activity. Similarly a deletion of the C-terminal segment of the ATBF1-A-specific region (amino acids 550-894), ASB, almost completely eliminated inhibitory activity. In contrast, deletion of amino acids 113 to 550,
ANS, produced a protein capable of repressing the MRF4 promoter (Figure 2 1).
These results suggest that to inhibit MRF4 promoter activity, two ATBF1-A- specific regions (amino acids 1- 113 and 550-894) in addition to its DNA-binding domain are required. Figure 21 The ATBF1-A N-terminal and DNA binding domain is necessary inhibition of MyoD- induced activation of MRF4 promoter.
IOT 112 cells were cotransfected with various deletions of ATBF 1-A (0.5 pg) along with 0.5 pg MyoD expression vector and 0.5 pg MRF4-Luc reporter. Vector cells were transfected with 0.5 pg MRF4-Luc reporter plus 1 pg of empty expression vector. Following transfection cells were induced to differentiate by replacing GM with DM. After 48 hours of differentiation, cells were harvested and luciferase activity was determined. Transfection efficiency was normalized to P- galactosidase activity and luciferase activity was expressed relative to cells transfected with MyoD and MRF-luc alone. Values expressed are the average of at least three independent experiments. Emor bars correspond to the standard error of the mean. MRF4-luc ATBF 1 Deletions Relative Luciferase Activity 0 50 100 150 200
Vector
MyoD
A
B
M3
DM12
DM2
ANN
m - ASB 3.1 1 ATBF1-A is a transcriptional repressor.
The experiments described above suggest that the ATBF 1-A-specific region is associated with transcriptional repressor activity- To examine this possibility, we created plasmid constructs, ATBF 1-A-Gal4,, and ATBF 1-B-Gal4,,, which express chimeric proteins of the Gal4 DNA binding domain and ATBF1-A and B, respectively. These constructs were transfected into C2C12 cells along with a luciferase reporter, G4tk-luc, consisting of five Gal4 binding sites upstream of the tk promoter. This reporter is constitutively active in C2C12 cells and is designed to detect transcriptional repression domains (Hsieh and Hayward, 1995). As indicated in Figure 22, this reporter is active in C2C12 cells and transfection of a construct expressing Gal4,, had Little effect on reporter activity. However, the expression of ATBF 1-A-Gal4,, resulted in a 70% decrease in the activiv of the tk promoter. This inhibition required the Gal4,, since expression of ATBF1-A alone displayed only a 20% repression of the G4tk-luc reporter (Figure 22). The requirement of the Gal4,, for ATBF 1-A to elicit its transcriptional inhibitory effects, suggests ATBF1-A must bind to the promoter for its full repressive activity.
The expression of ATBF 1-B-Gal4,,, on the other hand, decreased the activity of the reporter only by 20%. The expression of a fusion protein consisting Gai4,, and the ATBF1-A-specific N-terminal region also resulted in inhibition of G4tk-Iuc expression, although to a lesser extent than that attained with the full-length protein (Figure 22). These results show that ATBF la acts as a transcriptional repressor through the N-terminal ATBF1-A-specific region, although an additional Figure 22. ATBF1-A contains a transcription repressor domain.
The G4tk-luc reporter, containing five Gal4 binding sites upstream of the constitutively active thymidine kinase promoter, pPOP, pPOP-ATBF 1-A or ATBF 1-
Ga14,, fusion proteins were cotransfected into C2C12 cells. CelIs were harvested
48 hours after transfection. Luciferase activity obtained for cells transfected with the Gal4 DNA binding domain alone was set at 100. Values expressed are the average of four independent experiments. Error bars correspond to the standard error of the mean.
2 19 region common to ATBFl-A and B may also be involved to exhibit the fd repressor activity. In addition, the reduced transcriptional repression activity displayed by the N-term-Gal$,, may arise from a reduced stability or from impaired nuclear localization of this truncated ATBF 1-A. Furthermore, the N- term-Gal4,, protein may adopt a three-dimensional conformation that is structurally different from that of the N-terminaI region of full length ATBF 1-A molecule. This difference in protein folding may partially mask any transcriptional repression domains located in the ATBF 1-A-specific N-terminal region.
3.12 A tethered MyoD-E47 protein is resistant to ATBFI-A inhibition of MRF4 promoter.
One mechanism utilized by inhibitory myogenic regulatory factors such as the Id family, is competitive binding for bHLH dimerization partners which prevents the formation of transcriptionally active MyoD-E protein complexes.
To investigate whether such an inhibitory mechanism was responsible for negative effect of ATBF1-A on the activation of the MRF4 promoter by MyoD, we utilized a MyoD protein tethered to an E47 protein by a flexible 22 amino acid linker
(Neuhold and Wold, 1993). This forced dirner of MyoD and E47 is insulated from inhibition by proteins such as Id, that sequester E-proteins from MRFs (Neuhold and Wold. 1993). MyoD-E47 strongly activated luciferase expression from the
MRF4-luc reporter (Figure 23). Since MyoD was covalently attached to its Figure 23. Forced MyoD-E47 dimers can overcome ATBF1-A-mediated inhibition of the MRF4 promoter.
10T1/2 cells were transiently transfected with the MRF4-luc reporter and either MyoD or MyoD-E47 activator expression vectors. The ability oPATBF1-A to inhibit activation of the MFW4 promoter by MyoD or a tethered MyoD-E47 complex was addressed by co-transfecting ATBF la along with the activator vectors. The luciferase activity obtained from MyoD-MRF4-luc transfections was set at 100%. MyoD MyoD+A MyoD-E47 MyoD-E47+A dimerization partner and did not have to form dirnen with an independent E- protein, this activation was about 3 times higher than that observed when MyoD was transfected alone. Transfection of ATBF1-A resulted in only a 25% decrease in transcriptional activation of MRF4-luc by MyoD -E47, where as ATBF 1-A inhibited MyoD-induced activation of MRF4-luc by about 80% Figure 23).
These results suggest that disrupting MyoD-E protein dimer formation may contribute to the inhibitory mechanism of ATBF1-A. This disruption may occur through protein-protein interaction between ATBFl-A and bHLH proteins.
We examined whether ATBF1-A may interact with bHLH proteins such as
MyoD by mammalian two-hybrid assays. These experiments were based on the transcriptional repression activity displayed by ATBF 1-A-Gal4,, fusion proteins
(Figure 2 2). Vectors expressing ATBF 1-A-Gal4,, along with the G4t.k-luc were cotransfected into C2C12 cells with or without plasrnids expressing MyoD fused with the VP16 activation domain (MyoD-VP16). If these proteins interact, the
VP 16-activation domain may overcome the repression exerted by ATBF 1-A and induce luciferase expression from this promoter. As indicated in Figure 24, cells transfected with ATBF 1-A-Gal4,, express about 60% lower luciferase levels from the G4tk-luc reporter than cells transfected with the Ga14,, alone. The addition of MyoD-VP16 to the ATBFl-A-Gal4,, transfected ceIls relieved this transcriptional repression as luciferase expression reached levels observed in control cells expressing Gal4,. Cells co-transfected with ATBF 1-A-Gal4,, and MyoD lacking the VP 16 activation domain still retain the transcriptional inhibition Figure 24. Analysis of MyoD-ATBFI A interactions by mammalian two-hybrid assays.
Potential protein-protein interactions between ATBF1-A and MyoD were addressed through the mammalian two-hybrid assay. In this procedure, ATBF 1-
Ga14,, fusion proteins were co-transfected into C2C12 cells with the constitutively- active G4tk-luc reporter along with either MyoD-VP 16 fusion protein, wild-type b1yoD or an empty expression vector. As a positive control, the Id-Gal4,, fusion protein was co-transfected with MyoD-VP 16 vector. Values obtained for Ga14, alone were set at 100% and the results were expressed relative to this control.
125 observed when ATBFI-A-Gal4,, was transfected done. There was a 20% reduction in luciferase activity when ATBF1-B-Ga14,, was transfected, and the addition of
MyoD-VP16 did not Iead to any significant change in G4tk-luc activation. indicating that these proteins do not interact. The ATBF1-A-specific N-terminus fused to Gd4,, displayed about a 25% decrease in activation of the G4tk-luc reporter. This inhibition was not relieved by cotransfection of MyoD-VP 16. As a control we used Id-GaL4,, which expresses the Gd4,, fused to Id which is known to interact with MyoD. As expected, transfection of both MyoD-VP16 and Id-
Ga14,, led to an enhanced activation of the G4t.k-luc promoter. Alleviation of the transcriptional repression of ATBF1 -A-Gal4,, by the addition of MyoD-VP 16 suggests that MyoD-may interact with ATBF 1-A. This potential interaction domain may overlap the ATBF1-A-specific N-terminal region and the N-terminus of the
ATBFI-B since MyoD-VP 16 activate transcription from the G4tk-luc reporter when co-transfected with ATBF 1-A-Gal4,, ,but not when transfected with ATBF 1-
B-Ga14,, or N-tern-Gal4,,. However, since we did not analyse whether the VP 16 activation domain alone was able to relieve ATBF1-A mediated transcriptional repression it is possible that the VP16 transactivation domain may exert its effects on ATBFl-A independent of its fusion to MyoD.
A possible interaction between ATBF1-A and MyoD was examined biochemically through irnmunoprecipitation and His-tag pull-down assays. Total lysates from C2C12 cells and A6 cells cultured in GM were fxst immune- precipitated with antibodies raised against ATBF1-A, MyoD or a-actin and then Figure 25. ATBF1-A does not interact with MyoD.
(A) The possibility of MyoD-ATBF 1-A interactions was addressed by immunoprecipitation experiments. Lysates obtained from C2ClZ and A6 cells
cultured in GM were immunoprecipitated with antibodies against ATBF 1 (D 1.
D4), MyoD, actin, or mocked immunoprecipitated with no antibody, resolved on a
10% polyacrylamide gel and immunoblotted with a MyoD anabody. As a control for MyoD expression, total C2C12 cell lysate was included in the Western blot
(- lane) .
(B) In vitro protein interactions were assessed by Ni2+-affinity chromatography. Bacterial-expressed. histidine-tagged fusion proteins corresponding to the N-terminal regions of ATBF1-A, and SRF were incubated with 3%-Met-labelled MyoD or MyoD-E47 in vitro translated products. The reaction mixtures were passed over a charged Nickel purificaMon resin and complexes bound to the resin were eluted and separated on a SDS -polyacrylarnide gel. MyoD and MyoD-E47 complexes were only retained wiith SRF reactions.
The N-tenninal regions of ATBF1-A did not associated with eiither MyoD or
MyoD-E47. C2C12 C12A6
Antibodies agains 128 irnmunoblotted with anti-MyoD antibodies. As indicated in Figure 2 5A, MyoD protein was detected only in C2C12 lysates immunoprecipitated with MyoD antibodies and not in lysates ~eatedwith ATBFI-A antibodies. A6 cells produced no MyoD immunoprecipitated complexes indicating that this protein is not expressed in these cells. Although no MyoD-ATBFl-A complexes were detected in these analyses, alterations in salts concentrations or detergents types in the reaction buffers may be required to detect such complexes.
As a more sensitive analysis, we examined whether regions of the ATBF1 -A molecule containing a 6xHis-tag could interact with in vitro translated, 35~- methionine (Met)-labelled, MyoD or MyoD -E47 proteins. The ATBF 1-A-specific regions were immobilized to a Nizr-resin column and 35S-Met-labeUed MyoD or
MyoD-E47 were passed over the proteins bound to the resin. As controls. 35S-
Met-labelled MyoD and MyoD-E47 were added to tubes containing the resin alone or His-tagged-serum response factor (SRF), a documented MyoD-binding protein (Groisman et al., 1996). Figure 25B demonstrates that only SRF was able to bind MyoD and MyoD-E47. Neither of the ATBFl-A N-terminal fusion proteins (NH or NB) interacted with labelled proteins. Together, these results indicate that ATBFl-A and MyoD do not interact directly to disrupt MyoD-E protein heterodirnerization. Therefore, a portion of the inhibitory mechanism of
ATBF1-A does not include a direct protein-protein interaction with MyoD. Rather it may disrupt dirnerization indirectly through elevated expression of Id3 as observed in A6 cells. It is tempting to speculate that Id3 is a target of ATBFl-A NOTE TO USERS
Page(s) not included in the original manuscript and are unavailable from the author or university. The manuscript was microfilmed as received.
This reproduction is the best copy available. 130 regulation as both ID3 and ATBFl-A are expressed in C2C12 myoblasts but not in differentiated C2C12 cells (Christy et al., 1991). However, we did not observe any activation of the Id3 promoter by ATBFl-A (data not shown), indicating that Id3 is indirectly regulated by ATBF1-A in A6 cells.
3.13 The myogenic differentiation stimuli promotes the down-regulation of
ATBF1-A expression in C2C12.
The studies mentioned above indicate that ATBF1-A is an inhibitor of myogenic differentiation. We were interested in whether the down-regulation of
ATBFl-A expression was a consequence of the differentiation process or the removal of serum factors present in GM. To address this issue, we examined the expression of the endogenous, mouse ATBF 1-A rnRNA in A6 cells aeated with GM or DM. As indicated in Figure 26, mouse ATBF1-A mRNA was expressed in untransfected and vector-transfected C2C12, as well as A6 cells treated with GM.
Replacing GM with DM resulted in a down-regulation of mouse ATBF1-A mRNA in untransfected and vector-transfected C2C 12 cells and also in A6 cells even though A6 cells did not undergo myogenic differentiation (Figures 4-6). These cells fail to differentiate due the elevated levels of exogenous ATBFl -A (Figure 3B).
These data indicate that ATBFl -A is responding to the stimulus to differentiate rather than the differentiation mechanism itself, Furthermore, these results, together with the observation of ATBF 1-A mRNA re-expression in differentiated myotubes by GM (Figure 2B), suggests that the expression of ATBFl-A may be Figure 26. The down-regulation of ATBFl -A expression in C2C 12 cells occurs independent of muscle ditrerentiation.
The expression of the endogenous. mouse ATBF1-A mRNA was analysed in
C2C 12 cells as well as in vector-transfected and ATBF1 -A-tramfected C2C12 cells.
Total RNA was isolated from cells in GM and from cells treated with DM for 4 days @) and ATBF 1-A expression was monitored by RNase protection assay using human (top panel) and mouse-specific (bottom panel) riboprobes. Although A6 cells are unable to differentiate when grown in DM for 4 days (Figure 3-5). the endogenous ATBFl-A mRNA is still down-regulated in its normal manner. The lane designated P, corresponds to the undigested. human ATBFl riboprobe. P GDGDGD 133 regulated by serum factors. Myogenic differentiation is inhibited by mitogenic
stimuli such as the expression of oncoproteins including SV40 large T antigen, or c-
fos, or the addition of serum growth factors such as bFGF or TGF-P (reviewed in
Lassar et al.. 1994; and Olson. 1992). We were interested in whether agents that
block C2C 12 rnyogenesis prevented the down-regulation of ATBF 1-A expression,
and if ATBFl-A contributes to these inhibitory mechanisms. As indicated in Figure
27A, okadaic acid, the p42 MAP Kinase pathway inhibitor, PD98059. bone
morphogenetic protein (BMP)-2, bFGF, and lysophosphatidic acid (LPA) inhibited
MHC expression to various degrees. Okadaic acid and PD98059 inhibited differentiation the strongest. with very few MHC-positive cells. Cells treated with
BMP-2. bFGF, and LPA did contain some MHC expressing cells, but the extent of expression was reduced compared to the control cells. The different responses of
C2C12 cells towards these inhibitors may be explained by the nature in which these agents act. For example, okadaic acid and PD98059 display the strongest degree of myogenic inhibition. These agent permeate the cells and inhibit intracellular signal transduction pathways and may exert a more potent effect. On the other hand LPA and bFGF are extracellular compounds and require the expression of corresponding receptors in order to transduce their effects.
Expression of these receptors may be regulated during the differentiation process and the smali percentage of cells that were able to differentiate may have down- regulated expression of these receptors prior to the exposure of the inhibitory signals or may not have been responsive to these extracellular signals. We then Figure 27. Serum growth factors inhibit C2C12 myogenic differentiation and modulate ATBF1-A expression patterns.
(A). Pharmacological inhibition of C2C 12 muscle differentiation. C2C12 cells were induced to differentiate by replacing GM with DM (control). To inhibit myogenic differentiation, DM was supplemented with okadaic acid (10 pM),
PD98059 (50 pM). BMP-2 (300 ng/ml ), bFGF (10 ng/d ) or LPA (10 pg/d ).
After 48 hours of treatment the cells were fixed and MHC expression was determined by MF-20 irnmunocytochemistry.
(B) . ATBF 1-A mRNA expression was analysed after 24 hours from parallel samples described in (A) by RNase protection assay. The levels of ATBF 1-A expression in cells treated with myogenic inhibitors was compared with C2C 12 cells treated with GM alone or with DM for 24 hours- In addition, ATBFl-A mRNA expression was monitored from cells cultured for 24 hours in DM containing TGFP (10ngIml). GAPDH expression was included to control for variation in RNA quantities analysed. ATBF1-A GAPDH 1 2 3 4 5 6 7 8 136
examined whether ATBF la mRNA was expressed in the treated cells. ATBF 1-A
mRNA was detected in control C2C12 cells cultured in GM and expression levels
were reduced in cells after 24 how in DM (Figure 27B, Lanes 1 and 2). Cells
treated with DM containing bFGF, LPA or TGF-P for 24 hours contained ATBFl-
A mRNA levels comparable to those observed in GM (Figure 27B, Lanes 6-8).
although ATBF1-A mRNA levels were lower in bFGF treated ceh. In contrast,
the treatment with DM containing okadaic acid. PD098059 or BMP-2, did not
prevent ATBF1-A mRNA down-regulation (Figure 27B. Lanes 3-5). These data
indicate that these inhibitors may affect myogenesis in different manners, which
may or may not affect ATBFl-A expression. Only the inhibition of myogenic
differentiation by serum growth factors, such as bFGF, TGF-B and LPA, can
stimulate the expression of ATBF 1-A mRNA. Therefore. it is possible that ATBFl-
A may participate in the myogenic inhibitory mechanisms utilized by bFGF, TGF-P
or LPA.
We were interested in the significance of the ATBF1-A responsiveness to
serum factors (Figure 27B) and the re-expression of ATBF 1-A in differentiated
C2C12 cells treated with GM (Figure 2B). Differentiated muscle cells are unable
to reenter the cell cycle upon mitogenic stimuli (Schneider et al., 1994; Tiainen et al.. 1996). However, there are two populations of muscle cells, reserve and satellite cells, that can reenter the cell cycle when properly stimulated. Reserve cells are a quiescent, subpopulation of C2C 12 cells that remain refractory to the differentiation program but can re-enter the cell cycle when treated with GM 137
(Kitzmann et al., 1998; Yoshida et al.. 1998). Satellite cells are adult, myogenic
stem cek that are normally mitotically quiescent but are stimulated to enter the
cell cycle in response to muscle trauma (Bischoff. 1994). These cell types share
two common attributes with A6 cells: the entry into a reversible quiescence and
suppression of MyoD expression (Cooper et al., 1999: Kitzrnann et al.. 1998;
Yoshida et al.. 1998). Since C2C 12 cells are likely derived from satellite cells
(Yaffe and Saxel, 1977). the reactivation of ATBF 1-A mRNA observed in differentiated C2C 12 cells stimulated with GM (Figure 2B) suggests that a similar
induction of ATBF1-A mRNA expression may occur in satellite cells in response to muscle injury.
We examined whether ATBFl-A mRNA was expressed in regenerating tibialis anterior muscle following a crush injury. In the uninjured control limb of an adult mouse, ATBF 1-A mRNA was detected at low levels (Figure 28B. Lane 1).
However, expression of ATBFl-A was induced after 4 days (Figure 28B. Lane 2) of recovery when satellite cells are at their replicative peak prior to the onset of differentiation (McGeachie and Grounds, 1987). However, care must be taken in this interpretation since we have not examined whether ATBF 1-A mRNA is expressed in the satellite cells or in the surrounding muscle tissue. By 10 days of recovery, ATBF1-A expression declined to the residual levels observed in the control limb (Figure 28. Lane 3). These data, together with the observation that
ATBF 1-A mRNA expression can be stimulated by GM, suggest that ATBF 1-A may be involved in the activation of quiescent satellite to proliferating myoblasts. 138
ATBF la may play a role in the maintenance of the undifferentiated state of myoblasts and its down-regulation in response to serum depletion is required for terminal differentiation. Figure 28. Induction of ATBFl-A mRNA expression following a crush injury to the tibialis anterior muscle.
(A). Schematic representation of the regeneration events following muscle injury.
(B) . Up-regulation of ATBF 1A mRNA in regenerating muscle. The tibialis anterior muscle was subjected to a injury/regeneration protocol as outlined in the
"Materials and Methods." RNA was isolated from the uninjured limb (U) as well as from injured tissue after a 4 day (4d) or 10 day (10d) recovery period. ATBF 1-A expression was analyzed by RNase protection assays. Expression of GAPDH by
RNase protection assay was used to venfy that equivalent amounts of RNA were analysed. Muscle Injury Activation of Initiation of satellite ceU myogenic proliferation differentiation
-1 -1 :*-:: .;:'&, ;. , *,{& : V? 'I. . -2-..- . ._.. -.+. . .. ATBF1-A 4. Discussion
4.1 Ectopic ATBFl-A expression inhibits myogenic differentiation.
Muscle differentiation is regulated by a network of transcription factors that control the correct spatial and temporal sequence of differentiation events.
Members of the MyoD family of bHLH proteins have gained prominence for their roles as positive regulators of the determination and differentiation of the skeletal muscle lineage. The abundance of ATBF1-A in undifferentiated myoblasts, but not in differentiated muscle, suggests a function for this transcription factor in myogenesis. Indeed, transfection of CZC 1 2 myo blasts with ATBF 1-A blocks the myogenic differentiation program of these cells. This program includes changes in morphology, cell cycle arrest and expression of previously identified positive and negative myogenic regulatory factors and muscle specific genes. The effects of
ATBFl-A expression suggest that this protein is acting at multiple levels or at a very early nodal point of the differentiation process and that down-regulation of
ATBF 1-A is an essential step required for terminal muscle differentiation.
4.1.1 ATBF1-A perturbs expression of ~ositiveand negative rewlators of muscle differentiation.
The expression of the MyoD-MRF4 factors is vital for muscle differentiation. and disruption of their expression can inhibit myogenesis in vivo
(Hasty et al., 1993; Nabeshima et al., 1993; Rudnicki et al., 1993; Venuti et al., 142
1995). In C2C12 cells. expression of these factors are stimulated during myogenic differentiation by DM treatment. In cells expressing ATBF 1-A, these marker genes were completely suppressed in both GM and DM, which may explain the inability of these cells to differentiate into myotubes. It is uncertain whether ATBF 1-A inhibits MyoD expression directly, since both factors are expressed in undifferentiated myoblasts. Although MyoD is epistatic to myogenin. MyoD expression can be activated by myogenin in a positive-feedback mechanism, which is thought to reinforce the differentiation process (Yun and Wold. 1996).
Therefore. by blocking myogenin expression, ATBF1-A may indirectly prevent activation of MyoD.
The reduced levels of MyoD and myogenin expression in A6 may result from the inhibition of MR.4 promoter activity by ATBF1-A. Of all the MyoD- family members, MRF4 is expressed the latest during the differentiation program and inhibition of its expression would not be expected to have profound effects on the expression of upstream factors such as MyoD and my~genin. However, mice homozygous for an inactive MRF4 allele display reduced MyoD and myogenin mRNA expression at the developmental stages when MRF4 is normally expressed
(Patapoutian et al., 1995). Therefore the disruption of MRF4 promoter activity by
ATBFI-A may partially account for the inhibition of MyoD and myogenin in A6 cells.
The Id family of HLH proteins are negative regulators of differentiation processes (Benezra et al.. 1990; Christy et al.. 1991; Jen et al., 1992; Melnikova 143 and Christy, 1996). They interact with bHLH proteins such as MyoD and E47. preventing the formation of a transcriptionally competent MyoD-E-protein complex. Id3 is expressed in undifferentiated cells and is down-regulated during muscle differentiation. Forced expression of Id3 can inhibit myogenic differentiation in vitro (Atherton et al.. 1996 ; Melnikova and Christy. 1996). Thus the elevated levels of Id3 that we observe in A6 cells may participate in the impaired myogenic phenotype exhibited in these cells. The contribution of Id proteins is fbrther exemplified in experiments using the tethered MyoD-E47 protein which is insulated from the negative effects of Id proteins (Neuhold and
Wold, 1993). ATBFl-A was unable to prevent activation of the MRF4 promoter when the MyoD-E47 construct was transfected, suggesting that a portion of the
ATBF1-A inhibitory mechanism requires a disruption of MyoD-E protein complexes. We observed elevated Id3 mRNA levels in ATBF 1-A-transfected cells under growth condition in both GM and DM. However we were unable to establish whether ATBF 1-A directly activated an Id3-luciferase reporter. It is possible that the elevated expression observed in A6 cells is an indirect consequence of constitutive ATBF1-A expression, or that the upstream region of the Id gene used in the reporter did not contain the appropriate cis-acting element.
4.1.2 Mechanisms of MRF4 promoter inhibition.
The disrupted expression patterns of MyoD-MRFs in A6 cells suggest that these genes may be targets for ATBF1-A regulation. The mouse MRF4 promoter 144 was used as a representative MyoD-MRF member to determine whether it was a target of ATBF 1-A regulation. Activation of this promoter by MyoD in 10T 1/2 cells was inhibited by ATBF1-A. We utilized a bacterially expressed protein containing HD4 of ATBF1-A. This protein was capable of binding to an AT-rich sequence located between the El-box and the transcription start site in the mouse
MRF-4 promoter and this interaction prevented myogenin from binding the E 1 E- box. Since HD4 itself does not bind to E-box elements, this interference is likely through HD4 binding to the AT-rich site and physically blocking myogenin from accessing the adjacent E-box element. Alternatively HD4 may bind to the MRF4 promoter as a higher order protein complex, and this complex prevents myogenin from contacting it binding site. HD4 is common to ATBF1-B which is unable to inhibit the MRF4 promoter. Thus for its inhibitory activity, ATBF la requires both the C- terminus DNA binding domain plus its unique N-terminus.
Although HD4 was able to bind to the AT-rich sequence of the MW4 promoter, this protein may not accurately represent the DNA-binding properties of the full-length ATBFl-A molecule. Therefore we cannot completely conclude that
ATBF1-A is binding to the MRF4 promoter through this domain. The full-length
ATBFL-A protein may adopt a protein conformation different from that of the smaller HD4 protein. This conformational difference may alter the DNA-binding amity or specificity of the intact ATBF1-A molecule. However, DM-2, the 200 kDa C-terminal half of ATBF1-A, displayed DNA-binding properties similar to the smaller HD4 protein. The results from HD4 analyses can only be used as a 145 suggestive guideline for possible protein-DNA interactions between ATBF 1-A and the MRF4 promoter,
A proposed model for the differential regulation of the MRF4 promoter by
ATBFl-A and ATBFl-B is as follows. Both ATBFI-A and B bind to the AT-rich element through HD4 which prevents myogenin from binding the El E-box.
However, as HD4 does not interact with E-box sequences, myogenin can still bind to the upstream E2 E-box @lack et al., 1995; Naidu et al., 1995). Since ATBF1-A can block basal transcription, its binding to the MRF4 promoter can inhibit activation of this promoter regardless of whether or not myogenin binds to the distal E2 E-box. On the other hand, ATBFl-B does not posses this transcriptional inhibitory activity in muscle cells, and can not interfere with myogenin activation of the MRF4 promoter through this secondary E-box site.
4.1-3 DNA binding properties of ATBF 1-A.
ATBF1-A is characterized by the large number of potential DNA binding domains, We have demonstrated that HD4 binds to an AT-rich motif located in the MRF4 promoter. The binding of HD4 was specific to the AT-rich region of the A/E probe and HD4 did not interact with E-box elements. In addition, only the HD4 domain of ATBFI-A was able to bind to this DNA element. HD 1, 2 and
3, and the zinc finger clusters of ATBFl-A displayed no affinity towards the MRF4 promoter. However. this does not preclude these regions of ATBFl-A from binding DNA. just not to the MRF4 promoter. Because the DNA-binding properties of the zinc fingers and homeodomains of ATBFl-A have not been extensively analysed, it is possible that these domains may bind to yet unidentified sequences. possibly outside of the 390 bp MRF4 promoter. The DNA-binding properties of these regions outside of HD4 could be determined though gel- mobility shift experiments utilizing oligonucleotides probes containing random sequences. The preferred binding sites can then be compared with promoter sequence databases to determine candidate target genes. It is possible that ATBFl-
A may interact with multiple cis-elements though its numerous potential DNA- binding motifs. This interaction may result in DNA looping and bring together other positive and negative transcriptional regulators that may regulate the basal transcriptional machinery. In this situation, ATBF 1-A would act as a scaffold protein, however, this possibility remains speculative until the DNA-binding properties of the remaining regions of ATBFl-A are determined.
Our laboratory has previously reported that HD4 binds to the enhancer element of the human AFP gene (Morinaga et al.. 1991 ; Yasuda et al.. 1994). The interaction between HD4 and the mouse MRF4 promoter represents a second identified target for ATBF la and the fmt in the mouse genome. The AT-rich motif present in the MRF4 promoter displayed a striking degree of similarity to the ATBFl binding site in the AFP promoter (Table 2). There is a common 6 nucleotide sequence (5'-TAATTA-3') found in both elements that contains the core homeodomain ATTA recognition site (Gehring et al., 1994) and may represent an
ATBF 1-A binding motif. This site is also found in the muscle-specific enhancer Table 2. Comparison of ATBFI-A binding element with AT-rich motifs.
Enhancer Element Seauence Reference AFP enhancer ATAATTACA (Morinaga et al., 1991) MRF4 AT-rich CTAATTAAA (Black et d.,1995) MCK AT-rich TTAATTATA (Cserjesi et al., 1992) Drosophila Ddc GTAATTAAG uohnson et al., 1989) MCK MEF2 CTAAAJUATA (Gossett et al., 1989) tTtT
The ATBF 1 binding site from the human a-fetoprotein (AFP) enhancer was compared to AT-rich elements in the MRF4 promoter, muscle creatine kinase (MCK) enhancer, and the Drosophila DOPA-decarboxylase @dc) enhancer. A conserved TAATTA sequence was detected in these elements (bold). in addition. the ATBFl binding site was compared to the MEFZ consensus of the MCK enhancer. A to T substitutions that do not affect MEF2 binding are listed as T, while substitutions that weaken MEF2 binding are listed as t (Cserjesi et al., 1992). 148 element of the MCK gene. It is also a binding site for the homeodomain protein
MHox, and is an essential site for optimal MCK expression, as mutation of this element grossly inhibits the activation of the MCK promoter (Cserjesi et al., 1992).
The AT-rich motif is not detected in the promoter sequence of other MRFs including MyoD and myogenin. This absence may simply reflect an incomplete promoter-sequence database or that the entire transcriptional regulatory regions of these factors have not been fully characterized. However, ATBF1-A may not bind directly to these elements. The presence of this motif in the distal enhancer of the
Drosophila Ddc gene is very interesting since Ddc expression is believed to be regulated by the Drosophila-ATBF1 ortholog, ZFH-2 (LundeU and Hirsh, 1992).
While HD4 of ATBF 1-A binds to the AT-rich element of the MRF4 promoter, it is not known whether this represents the only site of ATBF1-A activity. ATBF1-A may inhibit MRF4 promoter activation through the MEF2 element as we found binding of HD4 to a MEFZ consensus sequence. MEF2 is an important transcription factor involved in emcient muscle-specific gene expression and the interference of MEFZ function may result in transcriptional repression and inhibition of myogenesis (Gossett et al., 1989; Ornatsky et al., 1997).
The similarity between the MRF4-ATBF1 binding site and the MEF2 consensus sequence uable 2) suggests that it may represent a binding target for
ATBF1-A. Although HD4 was able to bind to the MEF2 consensus sequence, it was unable to compete with in vitro translated MEF2C protein for binding sites.
This suggests that MEF2 elements are low-afFiiity binding sites for HD4. It is also 149 possible that the DNA-binding affinity of the full-length ATBF1-A protein is greater than HD4 alone. However, given that ATBF1-A is expressed primarily in undifferentiated cells and MEF2 is expressed only upon induction of differentiation, these two proteins are likely never present in the same cell, and these differential binding affinities may not come in to play. In addition, interference of MEF2 binding may not be sufficient to inhibit transcription, since
MEFZ can enhance transcriptional activation of muscle-specific promoter in the absence of an MEFZ binding site, through protein-protein interacts with the
MyoD-MWs (Black et al., 1995; Mollsentin et al., 1995; Naidu et al., 1995).
ATBF 1-A may bind to the MEF2 sites of muscle-regulated promoters in undifferentiated cells, placing these genes under a transcriptional repressive environment and preventing inappropriate activation. As differentiation proceeds, expression of ATBF 1-A declines and MEFZ increases, relieving these promoters from this inhibitory constraint. In A6 cells. ATBF1-A expression does not decline and will continue to occupy MEF2 sites and repress transcriptional activation. This constraint of elevated ATBF1-A expression must be exerted early in the differentiation program, prior to the onset of MEF2 expression, since MEF2 proteins can out-compete HD4 for MEF2 binding sites. Alternatively, MEFZ expression may be impaired in A6 cells due to the reduced levels of MyoD detected, and would not affect ATBF1 -A binding. Thus ATBFl-A binding to
MEFZ sites in other muscle regulated promoters may explain the impaired expression of MyoD and myogenin observed in A6 cells. At present, cis-acting 150 elements controlling expression of any of the MEF2 genes have not been identified. and therefore it is unknown whether the MyoD family directly regulates MEF2
expression. In addition. the levels of MEF2 expression in A6 cells treated with GM or DM are not known. However, since MEF2 is unable to activate the myogenic differentiation program on its own (Molkentin et al., 1995). its presence in A6
cells, in the absence of MyoD or myogenin expression. would not be sufTicient to
activate muscle-specific gene expression.
A search of the TRANSFAC database revealed the AT-rich element located in the MRF4 promoter shares similarity to the binding sites for a number of non-
myogenic transcription factors (Wingender et al., 2000) . These include the
Drosophila homeodomain proteins Antennapedia (Hirsch et al.. 199 1). Deformed
(ReguIski et al., 199 I), Engrailed (Kassis et al., 19891, as well as mammalian transcription factors such as HoxA5 (Hirsch et al., 199 1) , Isl-1 (Karlsson et al,.
1990). HNF-1 (Feuerman et al.. 1989) and ATF/CREB (Mathey-Prevot et al.,
1990). Therefore it is possible that ATBF1-A may compete for binding sites with other transcription factors that also recognize this AT-rich sequence. Furthermore
ATBF1-A may employ two distinct mechanisms of transcriptional repression: active r~pressionand competition for binding site occupancy. The Cut/CDP/Cux homeodomain protein has been demonstrated to repress transcription through these two modes of operation (Mailly et al.. 1996). It can bind to CCAAT and
Spl sites in the tk promoter which down-regulates expression fiorn this promoter by preventing CBP and Spl from binding their respective elements. In addition, a 151 separable trans-repression domain has been mapped to the C-terminal region of the
Cut protein (Mailly et al.. 1996). However ATBF1-A may not employ both mechanisms to inhibit the MRF4 promoter since expression of the HD4 DNA- binding domain alone was not sufficient to inhibit MRF4 activation. We can not rule out that ATBF1-A utilizes these dual modes of transcriptional repression in other targets. For example. ATBF1-B is thought to repress transcription of the human AFP promoter by competition with HNFl for common. AT-rich binding sites (Morinaga et al., 199 1). In addition. ATBFl-A greatly reduces activation of the Aminopeptidase N promoter which also contains an HNF-1 site (Kataoka et al..
2000). Whether ATBF 1-A binds to and prevents access of HNF-1 to this promoter remains to be determined.
The DNA-binding properties of the homeodomain to their targets has been extensively studied in Drosophila. However, little is known about vertebrate homeodomain-DNA interactions since few target genes have been identified. The data generated from Drosophila have provided many insights into how homeodomain proteins function. Many homeodomain proteins recognize a short sequence that often contains an ATTA core. These proteins are promiscuous in their target specificity and many homeodomain proteins can recognize several target sites. The amino acid at position 50 of homeodomain can make direct contact to the two base pain 5' of the ATTA sequence and can determine the specificity for a particular DNA sequence (Hanes and Brent. 1991 ; Schier and
Gehring, 1992). Proteins that contain a glutamine residue at position 50 are 152 known as 450 homeodomain proteins. 450 proteins are suggested to control the expression of a large number of genes (Biggin and McGinnis. 1997; Liang and
Biggin, 199 8; Mannervik, 1 999). Experiments in Drosophila characterizing the expression of randomly selected genes demonstrate that the expression of large number of genes are directly regulated by the Q50 proteins EVE and F'TZ (Liang and Biggin, 1998). A wide spread binding model has been proposed to explain how these proteins can regulate the expression of a large set of target genes (Biggin and McGinnis. 1997; Mannervik, 1999). This model is based upon the observation that 450 proteins bind to a broad range of DNA targets with similar specificity
piggin and McGinnis. 1997; Walter et al.. 1994). The low DNA-binding specificity of Q50 proteins directs these proteins to a large number of binding sites.
The effect of the homeodomain protein once bound to its target site is dictated by interactions with co-factors. Since HD4 contains a glutamine at residue 50. this wide-spread binding model may apply to ATBF 1-A and explain wholesale changes in gene expression observed in A6 cells. HD2 and HD3 of ATBFl-A are also 450 horneodomain proteins. However, these proteins do not bind to the MRF4 A./E probe and therefore display differential DNA-binding properties. Residues apart from Q50 are important in mediating homeodomain-DNA interactions (Geh~get al., 1994: Gehring et al., 1994). It is not unusual that the different HDs of ZFH proteins may interact with specific DNA elements, for example. HD2 from
Drosophila ZFH-2 interacts with the Ddc enhancer, while HD3 binds the opsin promoter (Fortini et al.. 199 1; Lundell and Hirsh. 1992). Moreover, HD2 has a 153 binding preference for the Ddc element over the opsin element gundell and Hinh,
1992). Therefore, ATBF 1-A may directly regulate the expression of a large number of genes including MyoD, myogenin, MRF4 and cyclin D 1, by specifically binding to AT-rich motifs in the transcriptional regulatory regions of these genes through HD4.
4.1.4 ATBF1-A-s~ecificfunctional domains,
In addition to the C-terminal DNA-binding domain, HD4, two regions of the ATBFl-A-specific N-terminal domain were required for myogenic inhibition.
One region from amino acid 550 to 894 contains 4 zinc-fmger motifs unique to
ATBFl-A which may provide a surface for protein-protein interactions as described for other zinc-fmgers proteins &ee et al., 1998; Tsai and Reed, 1998).
This ATBF1-A-specific region may bind DNA, but this region did not exhibit any affinity towards MRF4 promoter sequences. The other region from amino acid 1 to 1 13 contains a short stretch of residues rich in proline, a characteristic of some transcriptional repression domains (Han and Manley, 1993; Madden et al., 1993;
Mailly et al., 1996; Venot et al., 1998). Fusion of the ATBF 1-A N-terminal region to yeast Ga14,, domain produced a protein that displayed transcriptional repression activity 60% of the inhibition achieved by the full-length ATBF1-A . Thus to achieve maximal activity as a transcriptional repressor, ATBF1-A requires its unique N-tenninal domain as well as regions common to ATBF1-B isofom. These regions on their own are insufficient to mediate the inhibitory activiv of ATBFl-A. 154
The thyroid hormone receptor repression domain is also composed of several regions that contribute to its repression activity and alI are required for its maximal activity (Baniahmad et al., 1992). It is not surprising that an intact ATBF 1A molecule is required for its fX activity, since the structural organization of ATBF 1 is evolutionarily conserved between human. mouse and DrosopMa orthologs
(Fortini et al., 1991; Hashirnoto et al., 1992; Kostich and Sanes, 1995).
4.1.5 Possible mechanisms of ATBF 1-A-mediated transcriptional reoression,
Transcriptional repressor proteins are thought to inhibit transcription by two main mechanisms: 1) they may sterically hinder the binding of another transcription factor to the promoter by binding to the same DNA element or an overlapping one or 2 ) they may actively repress transcriptional activation independently of interference of protein-DNA interactions. Active transcriptional repressor proteins are characterized by a modular repression domain that is independent from the DNA-binding domain (Hanna-Rose and Hansen. 1996; Licht et al., 1990). There is little consensus in the amino acid sequence found in these domains, but regions rich in alanine, proline or hydrophobic residues have been attributed to transcriptional repression activity (Hanna-Rose and Hansen, 1996).
ATBF1-A is suspected to act as an active transcriptional repressor for several reasons. First, expression of the DNA-binding domain alone is not sufficient to impede activation of the MRF4 promoter. Second, ATBF1-A does not inhibit
MRF4 transcriptional activation solely through steric hindrance dictated by its 155 large size since the ATBFl-B isoforrn is unable to inhibit MRF4 activation even though it is a large protein that shares the same DNA binding element as ATBF1-A.
Third. the transcriptional repression activity of ATBF1-A is transferable to a Gal4
DNA-binding domain fusion protein. Fourth, a segment required for transcriptional repression activity contains a region rich in proline, characteristic of some transcriptional repression domains. (Miura et al., 19 9 5).
How might ATBF1-A affect the activation of the MRF4 promoter? It is possible that ATBF 1-A strictly blocks the basal transcription of this promoter. For example, ATBF1-A may bind to the MRF4 promoter and prevent the formation of the preinitiation compIex by possibly interacting with TBP or other basal transcription factors. Alternatively, ATBFI-A may bind in close proximity to activator proteins, such as rnyogenin and antagonize their action which prevents
MRF4 induction without effecting the basal transcriptionat machinery. The experiments employing ATBF 1-A-Gal4,, fusion proteins indicate that ATBF 1-A is capable of inhibiting basal transcription of the tk promoter. Therefore these results suggest that ATBFl -A may inhibit the basal transcription of the MRF4 promoter as well, and does not directly interfere with activation by MyoD or myogenin. However, ATBFl-A was unable to prevent MRF4 activation by the tethered MyoD-E47 protein, suggesting that ATBF1-A may not be entirely capable of repressing basal transcription of this promoter or that the MyoD-E47 protein binds to the MRF promoter with a higher affinity than ATBF1-A and displaces it from the AT-rich region. One could determine whether ATBF1-A inhibits basal or 156 activated expression by increasing the distance between the binding sites for
ATBF 1-A and the activator proteins. If ATBF 1-A-mediated repression is independent of its distance to the basal promoter and the activator protein, then it is acting on the basal transcription and not on the activator.
Transcriptional repression may be mediated by interactions between inhibitory transcription factors and co-repressors proteins. These factors have garnered much attention over the last few years since many of these co-repressors such as mSin3A or N-CoR are associated with complexes possessing histone deacetylase activity (AUand et d., 1997; Hassig et al.. 1997; Hassig et al., 1998 ;
Heinzel et al., 1997; Lahem et ai., 1997; Nagy et al., 1997). For example, the bHLH transcriptional inhibitor Mad interacts with the co-repressors mSinJA/B which ultimately forms a complex that contains the histone deacetylase proteins
HDAC1 and HDAC2 (Hassig et al., 1997; Laherty et al., 1997). The transcriptional state of a gene is influenced by the degree of acetylation of core histones in the nucleosome (reviewed in Kuo and Aliis, 1998; Wolffe, 1996) .
Co-repressors do not interact with DNA, but rather interact with transcription factors bound to DNA. This interaction recruits deacetylase complexes to the promoter and prevents positive-acting transcription factors from accessing the promoter. Because co-repressors do not bind DNA directly they can inhibit transcription of a wide range of genes by interacting with different transcription factors that recognize specific promoter elements.
An attractive possibility is an interaction between ATBF 1-A and such 157 transcriptional co-repressors. It is not known whether ATBFl-A associates with these co-repressor proteins; however, interactions between ATBF 1-A and other proteins may be key in establishing a functional complex. The repressive activity of ATBFl-A may be mediated through interactions between the ATBF1-A-specific
N-terminal domain and regions common to both isoforms with other transcription factors. Recently, it has been reported that an interaction between the extreme C- terminal portion of both ATBF 1-A and ATBF 1-B and the proto-oncoprotein, c-myb represses activation-of myb-responsive promoters (Kaspar et al., 1999). In this example, ATBF 1 is acting as a co-repressor. It is possible that ATBF 1-A activity is modulated by myb. Furthermore, we have observed the potential for ATBF1-A protein-protein interactions in gel mobility shift experiments. The addition of
HD4 protein to differentiated C2C12 nuclear extracts produced a complex that was not observed with either HD4 or C2C12 extracts alone. This complex may represent an interaction between HD4 and proteins present in the nuclear extract.
Alternatively. this larger complex may reflect a post-translational modification of bacterially-expressed HD proteins. Protein-protein interactions have shown to be established via the homeodomain (Bendall et al., 1998; Di Rocco et al., 1997;
Ohneda et al., 2000; Um et al., 1995). In particular, Ohneda et al., (2000) demonstrate that the homeodomain of PDX-1 mediates interactions between bHLH proteins including E47 and BETAZNeuroD 1 in pancreatic P cells. We were unable to detect a direct interaction between ATBFl-A and MyoD. However,
ATBF1-A -interacting proteins remain to be identifed and the activity of ATBFl-A in a given cellular environment may be influenced by these interactions.
4.1.6 A comparison with mvo~enicinhibitorv transcription factors.
Several transcriptional regulatory proteins have been identified that inhibit myogenic differentiation, with each protein possessing specific inhibitory mechanisms. The active transcriptional repression mechanism displayed by
ATBF la is reminiscent of that of the zinc-finger/homeodomain protein family member ZEB (Genetta et al., 1994; Yasuda et al., 1994). Like ATBF 1-A, an intact
ZEB protein consisting of the DNA-binding domain and repression domain is required to inhibit myogenesis (Postigo and Dean, 1997; Postigo et al.. 1999).
However, these proteins differ in their DNA-binding domains: ATBF 1-A binds through HD4, while ZEB binds DNA primarily with its zinc-finger motifs (Ikeda and Kawakarni, 1995). They also differ in their recognition sequences: ATBF 1-A recognizes AT-rich motifs whereas ZEB binds to a subset of E-box elements
(Genetta et al., 1994; Yasuda et al., 1994).
Helix-loop-helix proteins are also involved in myogenic inhibitory processes. Although not bona fide transcription factors since they lack DNA- binding domains, Id proteins disrupts myogenic differentiation through competitive protein-protein interaction (Benezra et al., 1990; Christy et al., 1991; Jen et al.,
1992; Neuhold and Wold, 1993). These proteins bind and sequester E proteins away from the MyoD-bHLH proteins, preventing the formation of transcriptionally active complexes. The inhibitory mechanisms of MyoR and 259 Mistl, on the other hand, are composed of a combination of E-protein squelching and active transcriptional repression. Like Id, these bHLH proteins can form heterodimers with E proteins. but unlike Id, they can bind to E-box elements. It is believed that these proteins compete with MyoD for E-box binding sites in muscle- specfic genes and actively repress transcription through these elements (Lemercier et al., 1998; Lu et al., 1999). ATBFl-A represents a class of inhibitory proteins that displays an alternative inhibitory mechanism from the above described factors.
By targeting AT-rich sequences, including MEF2 sites. ATBF 1-A may regulate the expression of a number of genes expressed in the myogenic lineage, while other proteins such as ZEB and MyoR target the E-box binding site specifically (Lu et al..
1999; Postigo and Dean, 1997).
4.1 -7 Differential activity of ATBF 1 isoforms.
While ATBF1-A has been demonstrated to be a negative regulator of
C2C 12 myogenic differentiation, the ATBF 1-B isofom does not possess these inhibitory properties. ATBF 1-B may act in a negative manrler towards the A isoform since it contains the same HD4 DNA binding domain but lacks the repressor domain associated with N-terminus of ATBF1 -A. The similar results obtained from ATBF1-B and antisense ATBFl in the MyoD-conversion of 10T1/2 cells supports this notion of ATBFl-B inhibiting the function of the -A isoform. A number of genes are known to generate more than one rnRNAs through alternative splicing and/or alternative promoter usage, often yielding functionally different protein isoforms. In the case of transcription factors, isoforms with opposing
regulatory activities can be produced due to modulation of DNA-binding specificity
or affinity. transactivation or protein-protein interaction (Foulkes and Sassone-
Coei. 1992; Laoide et al., 1993; Lbpez, 1995; Tanaka et al.. 1995). One
interesting example relevant to muscle differentiation is illustrated by the
alternative splice variants of the ITF-2 gene (Skejanc et al., 1996). ITF-2 is a
member of the ubiquitous bHLH E-proteins that forms functional heterodimers
with MyoD (Lassar et al., 199 1). An alternatively spliced form of this protein,
termed MITF-2B. containing a unique N-terminal region was found to inhibit
MyoD trans-activation by forming a transcriptionally inactive heterodirner
(Skerjanc et al., 1996). In the case of ATBF1-B, this isoform may compete with its
ATBFl A isoform for AT-rich sites in muscle-specific enhancer elements preventing the inhibitory ATBF 1-A protein from gaining access to DNA. Such a mechanism may explain why transfection of ATBF1-B into C2C 12 and MyoD- converted 10T 112 cells display enhanced myogenic differentiation. At the moment, we do not know whether the ATBF1-B isoform is involved in activating myogenic differentiation of C2C 12 cells. Accurate quanMication of ATBF 1-B expression in C2C12 cells has been difficult due to very low levels of ATBFI-B mRNA and protein in these cells. 161 4.2 ATBFl-A alters cell cycle progression during C2C12 myogenic differentiation and promotes the reserve cell fate.
Myogenic differentiation of C2C12 cells is tightly coupled with cell cycle withdrawal into a GJG, arrest (Clegg et al., 1987; Lassar et al., 1994; Nadal-
Ginard. 1978; Olson, 1992; Walsh and Perlman. 1997). MyoD can inhibit cell proliferation independent of myogenic differentiation (Crescenzi et al., 1990;
Sorrentino et al., 1990). In addition, the Cdk inhibitor p2 lC'P'NAF'is up-regulated during myogenesis, while in parallel, Cdk activities and the expression of cyclin D 1 decline (Guo et al.. 1995; Halevy et al.. 1995; Parker et al.. 1995; Skapek et al.,
1995). In C2C 12 cells constitutively expressing ATBF1-A, we did not observe an induction of p2 lclPINM'mRNA when the medium was switched from GM to DM.
MyoD can activate and enhance the expression of p21CIPl/wAFl , although the exact mechanisms have yet to be established (Halevy et al.. 1995; Otten et al., 1997).
The reduced MyoD levels detected in A6 cells may contribute to the impaired expression of p21Cm1Nm1in response to DM aeatrnent. We also demonstrated that cyclin D 1 expression is elevated in ATBF1-A expressing cells. Since forced expression of cyclin D 1 has been shown to prevent activation of muscle-specific promoters by myogenic bHLH (Rao et al., 1994; Rao and Kohtz. 1995). the elevated cyclin D 1 expression observed in A6 cells may contribute to the inhibition of myogenic differentiation.
ATBFl -A may inhibit myogenesis in C2C 12 cells by altering the cell cycle progression and arrest required for differentiation. Since ATBF 1-A is normally 162 expressed in proliferating myoblasts it may function to maintain cells in their undifferentiated state by promoting increased cell proliferation, and thereby inhibiting myogenic differentiation. Expression of DNA turno r virus oncogenes including E1A and SV40 large T antigen, or the addition of mitogens such as bFGF can inhibit muscle cell differentiation and promote cell proliferation (reviewed in
Lassar et al., 1994; Olson, 1992). The elevated levels of cyclin D 1 and the reduced levels p21C1P"M1 in A6 cells cultured in DM would suggest that these cells are capable of cell proliferation. However, analysis of DNA synthesis revealed that these cells exit from the cell cycle in DM, albeit with delayed kinetics, in the absence of p2 lCIP1mml.In the case of A6, cell cycle arrest appears to occur at a point different from normal cells; a point where it is "easier" to re-enter the cell cycle. It is not know how A6 cells exit from the cell cycle in the absence of p2 lCIP1mAF1expression, possibly other Cdk-inhibitors such as p27 or p57 may compensate (Zhang et al., 1999).
The altered expression patterns of regulatory proteins such as MyoD. Id3, p2 lCTPINmlor cyclin Dl may affect how A6 cells respond to changes in growth conditions. Several investigators have described a subpopulation of C2C 12 cells that fail to form myotubes when induced to differentiate by DM (Kitzmann et al.,
1998; Miller, 1990; Yoshida et al., 1998). These non-differentiated reserve cells are in the quiescent Go phase but can be stimulated to re-enter the cell cycle when treated with GM. MyoD expression is absent in these quiescent cells, but is re- expressed during the progression from Go to G,. The down-regulation of MyoD is 163 a causal effect as forced MyoD expression severely reduces the number reserve cek formed (Yoshida et al., 1998). ATBF 1-A may direct C2C 12 myoblasts to this
MyoD-deficient, reserve cell subpopulation. However, these non-differentiated cells differ from ATBF1 -A expressing cells in that they retain their differentiation capacity after they re-enter the cell cycle, whereas A6 lack MyoD expression and are unable to differentiate into a mature myotubes following GM stimulation- The level of p2 lCIP1"NAF1expression may also influence the decision of whether myoblasts enter an irreversible differentiation program or reversible quiescence.
Cells expressing p2 lCIPL'WAF1may undergo growth arrest associated with differentiation, while those cells not expressing p2 lC1pl"NAF1may undergo quiescence. There was Little p2 lCIP1/WAF'expression detected in A6 cells treated with DM, or quiescent C2ClZ reserve cells, yet these cells can reenter S-phase when stimulated with GM. This Cdk inhibitor may act to permanently withdraw cells from their growth cycle at the onset of differentiation. Again the induction of p2 lCIP'maLexpression may be regulated by MyoD, and the reduced levels of p21CpL/WAF1observed in A6 cells may be related to the low MyoD expression levels.
Therefore, ATBF1-A is not inhibiting myogenesis simply by preventing cells from withdrawing from the cell cycle, but rather may promote a reversible entry into a quiescent Go phase.
This transition from Go into the cell growth cycle is similar to the activation of a population of muscle stem cells known as satellite cells in response to muscle injury. Our observed expression of ATBFl-A during muscle regeneration in vivo is 164 consistent with the activation of satellite cells. However, since we did not define the cell types expressing ATBF 1-A, these changes in expression may occur in all cell types in the injured muscle rather than in satellite cells alone. These quiescent satellite cells are located between the muscle fibre and the basement membrane, and are activated to enter the cell cycle in response to muscle trauma. Stimulation to proliferate coincides with activation of MyoD expression followed by an induction to differentiate as marked by the expression of myogenin and fmally a permanent withdrawal from the cell cycle. The differentiated myotubes will fuse with the existing muscle fibre, but a portion of these cells wiU remain as undifferentiated satellite cells. Expression of MyoD is believed to regulate the transition from the quiescent cell to the proliferative state, and fmally to the differentiated myotube, as mice deficient for MyoD display an impaired regenerative response (Megeney et al., 1996; Yablonka-Reuveni et al., 1999).
In the injured muscle, ATBF 1-A mRNA was transiently expressed during the satellite cell-activation period of muscle regeneration, but was not expressed after recovery from injury. It is thought that C2C 12 cells are composed primarily of activated satellite cells, since C2C12 cells were isolated from cultures of adult muscle in which satellite cells are believed to be the only source of proliferating myoblasts (Yaffe and Saxel, 1977). Based upon the expression of ATBF 1-A in undifferentiated C2C12 myoblasts, its down-regulation in response to induction to differentiate in low serum and the re-expression of ATBF 1-A following GM stimulation, it is possible that ATBFl-A may participate in the activation of satellite cells and their transition to myoblasts.
4.3 Regulation of ATBFl-A expression by growth promoting stimuli.
If ATBFl-A functions in the activation of quiescent satellite cells to proliferating myoblasts, it would be expected that ATBF1-A expression is regulated by growth promoting stimuli. Expression of endogenous mouse ATBF 1-A mRNA was down-regulated in A6 cells treated with DM even though these cells did not differentiate into mature muscle. Therefore the down-regulation of endogenous
ATBF la expression observed is in response to the stimulus to differentiate (the removal of growth factors), rather than in response to the myogenic differentiation process itself (the induction of MRFs). Moreover, the addition of serum components LPA, bFGF or TGF-P to DM inhibits myogenic differentiation C2C 12 cells (this study and Olson, 1992; Rao and Kohtz. 1995; Salminen et al., 199 1;
Vaidya et al.. 1989; Yoshida et al., 1996) and prevents the down-regulation of
ATBF1-A expression. These data suggest that ATBFl-A expression is downstream of these factors and the inhibitory activity of ATBF1-A may contribute to the anti- rnyogenic actions of bFGF and TGF-P (discussed below). ATBF1-A may partially contribute to the myogenic inhibitory mechanism for LPA. since C2C 12 cells treated with DM containing LPA share characteristics common with A6 such as an impaired myogenic differentiation program and the suppression of myogenin expression in DM. However, unlike ATBFl-A-expressing cells, MyoD expression is not affected in LPA-treated cells (Yoshida et al.. 1996). Therefore other inhibitory components mediate the effects of LPA. Yet the addition of okadaic acid, PD98059 or BMP-2 to DM disrupted C2C 12 myogenic differentiation did not prevent the reduction in ATBFl-A rnRNA levels in response to DM. This further demonstrates that the removal of serum factors, rather than the differentiation process, promotes the down-regulation ATBF 1-A expression. The protein phosphatase 1 and 2A inhibitor, okadaic acid, is thought to inhibit myogenesis by preventing a dephosphorylation event required to initiate MyoD
DNA-binding activity (Km et al.. 1992; Park et al., 1992). The treatment of
C2C 12 cells with BMP2 inhibits myogenic differentiation by diverting the cells from the myogenic to the osteogenic Lineage (Katagiri et al.. 1997; Katagiri et al.,
1994). Finally inactivation of myogenic differentiation by the MEKl inhibitor
PD98059 did not inhibit ;2TBF1-A down-regulation, suggesting that the MAP kinase pathway is not involved in the inactivation of ATBF1-A expression at the onset of myogenesis. The fact that ATBF1-A expression was not affected by these agents suggests that their inhibitory effects are dowmstrearn or independent of
ATBF 1-A action.
The myogenic inhibitory activity of ATBF1-A may comprise a component of the bFGF and the TGF-P inhibitory mechanisms for several reasons. The disruption of ATBF1-A down-regulation following transfer to DM occurred in response to these growth factors. In addition there are many similarities between
A6 cells cultured in DM and C2C 12 cells treated with DM containing bFGF or TGF-P. These include a reduced fusion index. a suppression of MyoD and
myogenin mRNA levels and induction of cych Dl expression (Rao and Kohtz.
1995; Salminen et al.. 1991 ; Vaidya et al., 1989; Yoshida et al., 1996).
Furthermore these growth factors can inhibit myogenic differentiation by inhibiting
DNA-binding and/or the transcriptional activities of MyoD-MRFs (Brennan et al..
1991 ; Hardy et al.. 1993; Kong et al., 1995). One striking similarity observed
between A6 cells and C2C 12 cells treated with TGF-P is the withdrawal from the cell cycle in response to DM which occurs in the absence of p2 lC'P1~AF1expression
(Allen and Boxhorn, 1989; Frith-Terhune et al., 1998 ; Johnson and Allen, 1990;
Massague et al.. 1986). These similarities between TGF-P-treated C2C 12 cells and
A6 cells suggest that ATBF 1-A may mediate the inhibitory effects exerted by TGF- p. The inhibition by bFGF, however, is characterized by a high percentage of cells entering S-phase in response to DM treatment (Yoshida et al., 1996), whereas in A6 cells this percentage decreases with DM incubation. Therefore not all of the characteristics described above for bFGF-mediated inhibition can be attributed to
ATBF 1-A function and other factors may be involved that prevent the cell cycle arrest observed in FGF-treated cells. Since FGF receptors can stimulate activation of the MAP kinase pathway, components regulated by this signalling cascade may constitute the portion of the bFGF inhibitory mechanism that leads to cell proliferation. Although ATBF1-A expression was not affected by inhibition of
MAP kinase signalling and may be independent from this pathway, the signal 168 mediated by FGF receptors may bifurcate and utilize other signalling swategies such as heterotrimeric G proteins (Fedorov et al.. 1998) that may activate ATBF 1-A expression in response to bFGF.
The role for the MAP kinase pathway in myogenic differentiation is somewhat controversial. In some studies there is little activation of this pathway during C2C12 muscle differentiation (Cuenda and Cohen, 1999). While in others, this signalling pathway appears to be important in maintaining both the undifferentiated and differentiated state. Inhibiting MAP kinase signalling when
C2C12 cells are transferred from GM to DM will block myogenic differentiation
(Bennett and Tonks. 1997; Gredinger et al., 1998). However, inactivating this pathway in C2C12 cells cultured in GM mrill induce muscle differentiation in the presence of rnitogens (Bennett and Tonks, 1997). It is unknown how this lvIAP kinase signalling pathway can achieve such a degree of specificity in the undifferentiated versus the differentiated cell. Potentially, the MAP kinase pathway may represent a single signal transduction cascade that exists in a multi-component signalling network.
The transformation of C2C 12 cells through constitutive expression of the cellular proto-oncogenes, including activated Ras, c-myc, c-jun or c-fos, can also inhibit n~yogenicdifferentiation (Endo, 1992; Konieczny et al.. 1989; Lassar et al.,
1989; Miner and Wold, 1991). Ras signalling can lead to the activation of the
MAP kinase pathway, which then can ultimately lead to the phosphorylation and activation transcription factors including c-fos . The inhibitory action of transformation is independent of cell proliferation. rather it may target MyoD locus and the differentiation apparatus directly. The impaired myogenic differentiation phenotype elicited by ATBF 1-A expression displayed characteristics common to those in cells transformed by activated Ras and c-fos: these cells fail to differentiate, and MyoD and rnyogenin expression are extinguished (Konieczny et d.,1989; Lassar et al., 1989). There are notable differences, however between transformation-mediated and ATBF 1-A-mediated inhibition. For example, inhibition of muscle differentiation by activated Ras is reversed by exogenous sources of MyoD kassar et al., 1989), whereas transfection of MyoD did not rescue myogenesis in A6 cells. In addition, the inhibitory effect of activated Ras is likely modulated through the MAP kinase pathway. However, ATBFl-A expression was not affected by inhibition of the MAP kinase activity. The inhibitory action of ATBF1-A may belong to a pathway independent from that of
Ras and c-fos.
Although bFGF. TGF-P and LPA have been implicated as negative regulators of muscle differentiation they may play important roles in the initiation of muscle repair and the specification of muscle identity during embryonic development. Combinatorial signalling of bFGF and TFGP from the neural tube can promote myogenesis in chicken paraxial mesoderm (Stern et al.. 1997). In addition. TGF-P signalling may promote myogenesis in the presence of growth factors (Zentella and Massague, 1992). The release of LPA has been documented in response to cellular trauma (Fukarni and Takenawa, 1992; Moolenaar, 1994) , and 170
the paracrine secretion of other factors such as bFGF may represent a satellite cell
activation signal. In fact, the addition of bFGF has been demonstrated to increase
the number proliferating satellite cells in culture and is thought to recruit quiescent
satellite cek into the activated state (Yablonka-Reuveni et d., 1999). In this
hypothesis, bFGF or LPA may be released from injured and surrounding tissues to
activate satellite cell proliferation concomitant with the induction of ATBF 1-A and
MyoD expression. MyoD is unable to initiate the myogenic program in the
presence of ATBF1-A, and this inhibition may ensure the generation of a sufficient
number of precursor cells. ATBF1-A expression subsequently decreases as
myogenic differentiation proceeds. It is tempting to speculate that the levels of
bFGF, LPA or other growth factors present at the injury site decline over time.
which leads to a down-regulation of ATBFl-A expression and initiates
differentiation, mimicking the effect of growth factor removal in vih-o.
4.4 Future Directions
4.4.1 Expression of ATBF 1-A during mvogenic differentiation in vivo.
In this study we have identified and presented biological roles for ATBF1-A in muscle differentiation in vitro. In addition. potential targets for ATBF 1-A transcriptional regulation have been elucidated. A number of questions have been
raised in this study that remain to be answered which can be addressed experimentally. First, it would be of great interest to determine the expression patterns for ATBFl -A during somitogenesis and inyogenesis in vivo. ATBF 1-A 171 mRNA has been detected in whole mouse embryos extracts as early as 8.5 dpc at the time of somite formation (Y. Miura, F. Berry and T. Tamoaki, unpublished observations). In situ hybridization analysis of ATBF 1-A mRNA localization in the mouse embryo has been performed, but the earliest time point examined was 15 dpc. well after the onset of myogenesis and ATBF1-A mRNA was not detected in any muscle tissue (Watanabe et al.. 1996). In addition, the expression of ATBF1-A in muscle cells following an injury suggests a function for ATBF1-A in muscle regeneration in vivo (Figure 28). The expression patterns of ATBF 1-A could be determined at earlier time points (8 to 12 dpc) in conjunction with the expression of differentiation markers such as MyoD, myf5, rnyogenin and MRF4. This analysis would determine whether ATBFl-A is expressed in the developing somite and whether its expression changes with the onset of differentiation. In addition,
ATBF1 knockout mice have been generated and mice homozygous for a mutated
ATBFl allele die prior to birth due to an unidentified defect. The analysis of muscle defects in these mice may reveal a function for ATBF la in myogenesis in vivo. The data generated from these experiments would extend the role ATBF 1-A as a negative regulator muscle differentiation in vih-o to such a role in vim.
The formation of somites and the induction of MyoD and myf5 expression are thought to be initiated though combinatorial signalling by SHH originating from the notochord and by Wnt members. bFGF and TGF-/3 signals emerging from the neural tube (Miinsterberg et al.. 1995; Miinsterberg and Lassar. 1995; Stem et al.. 1997). Since these inductive signals are not specific to the muscle lineage and 172 the expression of MyoD and Myf5 can induce ectopic muscle differentiation, there
is a need for negative regulators such as ATBF 1-A to prevent non-somitic tissues
exposed to these signals from differentiating into muscle cells. For example, the
Pax-3 transcription factor con&ols the activation of MyoD expression in vivo
(Maroto et al., 1997; Rawls and Olson, 1997; Tajbakhshet al., 1997). However,
Pax 3 expression is not limited to somites (Goulding et al.. 199 1; Mansouri and
Gruss, 1998; Tremblay et al., 1995) and ectoptic Pax3 expression in the neural
tube can induce MyoD expression (Tajbakhsh et al., 1997). Moreover, Pax3
expression is induced in the developing spinal cord by SHH signals originating
from the notochord (Goulding et al., 1993). Therefore, Pax3 must act in a
combinatorial manner to commit cells to the myogenic program in somites and not
in other tissues such as the neural tube. Since ATBFl-A is expressed in the
developing nervous system (Ido et d., 1996; Watanabe et al., 1996). it may act to
prevent Pax3 from initiating myogenesis in the developing neural tube. To address
this issue, the expression of MyoD and myogenin can be examined in ATBFl
knockout mice. If ATBF1 -A acts to suppress ectopic myogenesis in tissues such as
the neural tube. the expression patterns for rnyogenin and possibly MyoD would be
expanded in mice lacking a functional ATBF 1-A protein.
A second role for the inhibitory action of ATBF1-A may be in the formation
of the limb musculature. The limb muscles are forrned from migratory muscle precursor cells (mpc) derived from the demarnyotome of the somite. As these cells
begin their migration from the somite, expression of MyoD and other MRFs is 173 extinguished. The molecular mechanisms regulating these expression patterns in the mpc are relatively unknown, but Pax3 is required since Splotch mutant mice, which lack a functional Pax3 protein, display limb defects (I3ober et al., 1994;
Chalepakis et al., 1994: Goulding et al.. 1993 ). Once again, a possible role for
ATBF 1-A in this process may be identified by examining ATBF 1-A expression patterns in mpcs and by comparison between wild type and ATBF1-deficient mice of the expression profiles of MRFs in these cells . As well, the limb musculature may be examined for defects in muscle formation that may be attributed to the loss of ATBF 1-A activity.
4.4.2 Identification of ATBF 1-A targets of transcriptional redation.
A second set of experiments would be to determine the preferred binding sequences for ATBFl-A and whether ATBF 1-A directly regulates the expression of
MyoD, myogenin and other muscle regulated promoters in addition to MRF4.
Interestingly, expression of MyoD in the mouse embryo is directed by a distal enhancer element containing an AT-rich segment (Asakura et al., 1995). In addition, it is necessary to determine whether ATBF 1-A regulates expression through the AT-rich MEF2 element. The MEFZ element is vital for activating myogenin expression in vitro and in vim (Cheng et al.. 1995; Cheng et al.. 1993;
Edmondson et al., 1992). Transcriptional inhibitory actions of ATBF la through this element may explain the repressed myogenin expression observed in ATBF1- A-expressing C2C 12 cells.
Finally as a means of identifying the ATBFl-A transcriptional inhibitory mechanisms, it would be interesting to determine if any protein-protein interactions occur between ATBFl-A and other nuclear factors. A yeast two-hybrid screen may identify a wide range of proteins potentially interacting with ATBF1-A. including existing proteins as well as novel clones. In addition, the interaction between ATBFl-A and known transcriptional co-repressors or activators such as N-
CoR, p300 or PCAF, can be addressed directly through in vitro assays such as irnunoprecipitations or GST-pull downs. Identlfylng proteins that interact with
ATBF1-A will be of great interest as it is becoming clear that transcriptional regulatory factors are not acting alone, but rather they are components of a higher- order regulatory complex.
4.5 Summary
The role of the multiple-zinc fmger homeodomain transcription factor,
ATBF 1-A was investigated during skeletal muscle differentiation in vih-o. This study has demonstrated that ATBF la is expressed in proliferative, undifferentiated
C2C 12 myoblasts and its expression subsequently declines as these cells initiate muscle differentiation. This down-regulation of ATBF 1-A may have a causal role in muscle differentiation as cells constitutively expressing ATBF 1-A fail to differentiate as characterized by aberrant expression patterns of a wide range of 175 genes including myogenic regulatory transcription factors, cell cycle regulatory
proteins, and muscle structural proteins. This inhibitory effect was limited to the
ATBF la isoform, as ATBFI-B expression enhanced rather than inhibited muscle
differentiation. Together these results suggest that the two ATBFl isofoms may
act antagonistically in C2C12 cell differentiation and that regulatory regions
required for inhibition lie in the ATBFl -A specific N-tenninal portion of the
protein.
In addition. this study has identified the mouse MFV4 promoter as a target
of ATBF 1-A transcriptional regulation. The fourth homeodomain of ATBF 1-A
binds to an AT-rich element in this promoter and actively represses transcriptional
activation of MRF4 by MyoD . Transcriptional repression by ATBF 1-A requires
both the DNA-binding domain and the N-terminal region specific to ATBF1-A.
The possibility exists that the MEF2 promoter is also regulated by ATBF1-A.
Whether ATBFl-A binds and regulates other MyoD-MRF promoters remains to be
determined.
Ectopic ATBF la expression in C2C12 cells alters the cell cycle response in
low serum medium. Instead of entering a permanent G, mest required to initiate muscle differentiation, these cells enter a reversible, Goquiescence. Therefore,
ATBFl-A may divert myoblasts from entering terminal differentiation and may promote these cells into adopting a reserve cell identity. These reserve cells share similar characteristics with satellite cells, which are the only ceh capable of 176 proliferation in adult muscle. We propose that ATBFI-A functions in regulating the temporal order of muscle differentiation by maintaining myoblasts in their undifferentiated state until cells receive differentiation signals. ATBFl-A expression is down-regulated in response to this stimuli and cells are able to enter the differentiation program. Negative regulation by factors such as ATBF 1-A may be necessary to ensure the formation of sufficient numbers of muscle precursor cells prior to the onset of muscle differentiation. 5. References.
Alland. L.. Muhle. R.. Hou. H.. Jr., Potes, J., Chin. L., Schreiber-Agus, N., and DePinho. R. A. (1 997). Role for N-CoR and histone deacetylase in Sin3-mediated transcriptional repression. Nature 38 7, 49-55.
Men. R. E.. and Boxhorn. L. K. (1989). Regulation of skeletal muscle satellite cell proliferation and differentiation by transforming growth factor-beta, insulin-like growth factor I. and fibroblast growth factor. J Cell Physiol 138, 3 1 1-5.
Andrks. V.. and Walsh. K. (1996). Myogenin expression. cell cycle withdrawal, and phenotypic differentiation are temporally separable events that precede cell fusion upon myogenesis. Journal Of Cell Biology 132. 657-66.
Arnold. H.H., and Winter. B. (1998) Muscle differentiation: more complexity to the network of myogenic regulators. Current Opinion in Genetics and Development 8. 539-544.
Asakura. A.. Lyons. G. E.. and Tapscott. S. J. (1995). The regulation of MyoD gene expression: conserved elements mediate expression in embryonic axial muscle. Dev Biol 171. 386-98.
Atherton. G. T.. Travers. H., Deed. R.. and Norton, J. D. (1996). Regulation of cell differentiation in C2C 12 myoblasts by the Id3 helix-loop-helix protein. Cell Growth And Differentiation 7. 1059-66.
Bader, D.. Masaki, T., and Fischman. D. A. (1982). Immunochemical analysis of myosin heavy chain during avian myogenesis in vivo and in vitro. Journal Of Cell Biology 95. 763-70.
Bailey, P.. Downes. M.. Lau. P.. Harris. J., Chen, S. L.. Harnarnori, Y., Sartorelli, V.. and Muscat, G. E. (1999). The nuclear receptor corepressor N-CoR regulates differentiation: N-CoR directly interacts with MyoD. Mol Endocrinol 13, 1155-68.
Baniahmad, A., Kohne, A. C.. and Renkawitz. R. (1992). A transferable silencing domain is present in the thyroid hormone receptor. in the v-erbA oncogene product and in the retinoic acid receptor. Embo J I I, 10 15-23.
Barthelemy. I., Carrarnolino. L.. Gutierrez, J., Barbero. J. L.. Marquez. G., and Zaballos. A. (1996). zhx-1: a novel mouse homeodomain protein containing two zinc-fingers and five homeodomains. Biochem Biophys Res Commun 224. 870-6. Bendall, A. J., Rincdn-Limas, D. E., Botas, J., and Abate-Shen, C. (1998). Protein complex formation between Msxl and Lhx2 homeoproteins is incompatible with DNA binding activity- Differentiation 63. 151-7.
Benezra, R., Davis, R. L.. Lassar, A., Tapscott, S., Thayer, M., Lockshon, D., and Weintraub, H. (1990). Id: a negative regulator of helix-loop-helix DNA binding proteins. Control of terminal myogenic differentiation. Annals Of The New York Academy Of Sciences 599. 1-1 1.
Benezra, R., Davis, R. L., Lockshon, D ., Turner, D . L., and Weintraub, H. (1990). The protein Id: a negative regulator of helix-loop-helix DNA binding proteins. Cell 61, 49-59.
Bennett, A. M., and Tonks. N. K. (1 997) . Regulation of distinct stages of skeletal muscle differentiation by mitogen-activated protein kinases. Science 278, 1288-9 1.
Berg, J. M. (1990). Zinc fmger domains: hypotheses and current knowledge. Annu Rev Biophys Biophys Chem 19, 405-2 1.
Berg. J. M. (1990). Zinc fmgers and other metal-binding domains. Elements for interactions between macromolecules. J Biol Chem 265, 6 5 13-6.
Biggin, M. D., and McGinnis, W. (1997). Regulation of segmentation and segmental identity by Drosophila homeoproteins: the role of DNA binding in functional activity and specificity. Development 124, 4425-33.
Biggin, M. D., and Tjian. R. (1989). Transcription factors and the control of Drosophila development. Trends Genet 5, 377-83.
Billeter, M., Qian, Y., Otting, G.. Muller, M., Gehring, W. J., and Wuthrich. K. (1990). Determination of the three-dimensional structure of the Antemapedia homeodomain from Drosophila in solution by 1H nuclear magnetic resonance spectroscopy. J Mol Biol214, 183-97.
Billeter, M., Qian, Y. Q.. Otting, G., Muller, M., Gehring, W., and Wuthrich, K. (1993). Determination of the nuclear magnetic resonance solution structure of an Antemapedia homeodomain-DNA complex. J Mol Biol234, 1084-93.
Bischoff, R. (1994). The satellite cell and muscle regeneration. In Myology, E. A. G and F.-A. C, eds. (New York NY: McGraw-Hill), pp. 97-133.
Black. B. L., Ligon, K. L., Zhang, Y.. and Olson, E. N. (1996). Cooperative transcriptional activation by the neurogenic basic helix-loop-helix protein MASHl and members of the rnyocyte enhancer factor-2 (MEF2) family. Journal Of Biological Chemistry 271, 26659-63.
Black, B. L., Martin. J. F., and Olson. E. N. (1995). The mouse MW4 promoter is tram-activated directly and indirectly by muscle-specific transcription factors. Journal Of Biological Chemistry 270. 2889-92.
Black. B. L.. Molkentin, J. D., and Olson, E. N. (1998). Multiple roles for the MyoD basic region in transmission of transcriptional activation signals and interaction with MEF2. Molecular And Cellular Biology 18, 69-77.
Black, B. L.. and Olson, E. N. (1998). Transcriptional control of muscle development by myocyte enhancer factor-2 (MEF2) proteins. Annu Rev Cell Dev Bi01 14, 167-96.
Bober, E.. Franz, T.. Arnold. H. H.. Gms, P., and Tremblay, P. (1994). Pax-3 is required for the developrnent of limb muscles: a possible role for the migration of dermomyotomal muscle progenitor cells. Development 120. 603- 12.
Borycki, A. G., Brunk, B., Tajbakhsh, S., Buckingham, M., Chiang. C., and Emerson, C. P., Jr. (1 999). Sonic hedgehog controls epaxial muscle determination through MyfS activation. Development 126. 4053-63.
Braun, T., Buschhausen-Denker, G.. Bober, E., Tannich. E., and Arnold, H. H. (1989). A novel human muscle factor related to but distinct from MyoD 1 induces myogenic conversion in 10T1/2 fibroblasts. Embo Journal 8. 70 1-9.
Braun, T., Rudnicki, M. A., Arnold, H. H., and Jaenisch. R. (1992). Targeted inactivation of the muscle regdatory gene Myf-5 results in abnorrnal rib development and perinatal death. Cell 71. 369-82.
Brennan. T. J., Edmondson, D. G., Li. L., and Olson, E. N. (1991). Transforming growth factor beta represses the actions of myogenin through a mechanism independent of DNA binding. Proc Natl Acad Sci U S A 88, 3822-6.
Briata, P., Ilengo. C.,Bobola. N., and Corte, G. (1999). Binding properties of the hurnan homeodomain protein 0TX2 to a DNA target sequence. FEBS Let. 445, 1160-4.
Cdaerts, P.. Halder. G.. and Gehring. W. J. (1997). PAX-6 in development and evolution. Annu Rev Neurosci 20, 483-532. Cenciarelli, C., De Santa, F-, Puri, P.L., Mattei, E., Ricci, L., Bucci, F., Felsani, A., and Caruso, M. (1999). Critical role played by cyclin D3 in the MyoD-mediated arrest of cell cycle during myoblast differentiation. Mol Cell Biol. 19, 5203-52 17.
Chalepakis, G., Goulding, M., Read, A., Strachan, T., and Gruss, P. (1994). Molecular basis of splotch and Waardenburg Pax-3 mutations. Proc Natl Acad Sci U S A 91, 3685-9.
Chen, C., and Okayama, H. (1987). High-efficiency transformation of mammalian cells by plasmid DNA. Molecular And Cellular Biology 7, 2745-52.
Cheng, T. C., Tseng, B. S., Merlie, J. P., Klein, W. H., and Olson, E. N. (1995). Activation of the myogenin promoter during mouse embryogenesis in the absence of positive autoregulation. Proceedings Of The National Academy Of Sciences Of The United States Of America 92. 56 1-5.
Cheng, T. C., Wallace. M. C.. Merlie, J. P., and Olson, E. N. (1993). Separable regulatory elements governing myogenin transcription in mouse embryogenesis. Science 261, 2 15-8.
Chomczynski, P., and Sacchi, N. (1987). Single-step method of RNA isolation by acid guanidinium thiocyanate- phenol-chloroform extraction. Anal Biochem 162, 156-9.
Christy. B. A., Sanders, L. K.,Lau, L. F., Copeland, N. G., Jenkins, N. A., and Nathans, D. (199 1). An Id-related helix-loop-helix protein encoded by a growth factor-inducible gene. Proceedings Of The National Academy Of Sciences Of The United States Of America 88, 1815-9.
Clegg, C. H., Linkhart, T. A., Olwin, B. B.. and Hauschka, S. D. (1987). Growth factor control of skeletal muscle differentiation: commitment to terminal differentiation occurs in G1 phase and is repressed by fibroblast growth factor. J Cell Biol 105, 949-56.
Cooper, R. N., Tajbakhsh, S., Mouly, V., Cossu, G., Buckingharn, M.. and Butler- Browne, G. S. (1999). In vivo satellite cell activation via myf5 and MyoD in regenerating mouse skeletal muscle. J Cell Sci 112, 2895-901.
Cossu, G., Kelly, R., Tajbakhsh, S., Di Donna, S., Vivarelli, E., and Buckingham, M. (19 96). Activation of different myogenic pathways: myf-5 is induced by the neural tube and MyoD by the dorsal ectoderm in mouse paraxial mesoderm. Development 122, 429-37. Crescenzi, M., Fleming, T. P., Lassar, A. B., Weintraub. H., and Aaronson, S. A. (1990). MyoD induces growth arrest independent of differentiation in normal and transformed cells. Proceedings Of The National Academy Of Sciences Of The United States Of America 87, 8442-6.
Csejesi. P.. Lilly, B., Bryson. L., Wang, Y., Sassoon, D. A., and Olson, E. N. (1992). MHox: a mesodermally restricted homeodomain protein that binds an essential site in the muscle creatine kinase enhancer. Development 115, 1087-10 1.
Cuenda, A.. and Cohen, P. (1999). Stress-activated protein kinase-2/p38 and a raparnycin-sensitive pathway are required for C2C 12 myogenesis. J Biol Chem 2 74, 4341-6.
Davis, R. L., and Weintraub, H. (1992). Acquisition of myogenic specificity by replacement of three amino acid residues from MyoD into E12. Science 256, 1027- 30.
Davis, R. L., Weintraub. H., and Lassar, A. B. (1987). Expression of a single transfected cDNA converts fibroblasts to myoblasts. Cell 51, 987-1000.
Dezan, C., Meierhans, D., Kunne, A. G. E., and Allernann. R K. (1999). Acquisition of myogenic specificity through replacement of one amino acid of MASH-1 and introduction of an additional a-helical turn. Biological Chemistry 380, 705-7 10.
Di Rocco. G., Mavilio, F.. and Zappavigna, V. (1 997). Functional dissection of a transcriptionally active, target-specific Hox-Pbx complex. Embo Journal 16, 3644- 54,
Edrnondson, D. G., Cheng, T. C., Cserjesi, P., Chakraborty, T.. and Olson. E. N. (1992). Analysis of the myogenin promoter reveals an indirect pathway for positive autoregulation mediated by the muscle-specific enhancer factor MEF-2. Mol Cell Biol 12, 3665-77.
Endo. T. (1992). SV40 large T inhibits myogenic dserentiation partially through inducing c-jun. J Biochem (Tokyo) 112, 32 1-9.
Epstein, J.A.. Lam, P.. Jepael, L., Maas. R.L., and Shapiro. D.N. (1995). Pax3 inhibits myogenic differentiation of cultured myoblast cells. J. Biol. Chem. 270, 11719-22.
Fedorov, Y. V.. Jones, N. C., and Olwin, B. B. (1998). Regulation of myogenesis by fibroblast growth factors requires beta- gamma subunits of pertussis toxin-sensitive G proteins. Mol Cell Biol 18, 5780-7.
Feuerman, M. H., Godbout, R., Ingram, R S. and TiIghman, S. M. (1989).Tissue-Specific Transcription of the Mouse alpha-Fetoprotein Gene Promoter Is Dependent of HNF-1. Mol. Cell. Biol. 9,4204-42 12.
Fortini. M. E., Lai, 2. C.. and Rubin, G. M. (1991). The Drosophila zfh-1 and zfh- 2 genes encode novel proteins containing both zinc-fmger and homeodomain motifs. Mechanisms Of Development 34, 113-22.
Foulkes, N. S., and Sassone-Corsi, P. (1992). More is better: activators and repressors from the same gene. Cell 68, 41 1-4.
Frith-Terhune, A.. Koh, K. N., Jin, W. J., Chung, K. B., Park, S. K., and Koh, G. Y. (1998). Programmed changes of cell cycle regulators by serum deprivation regardless of skeletal myocyte differentiation. Mol Cells 8, 637-46.
Fukarni, K., and Takenawa, T. (1992). Phosphatidic acid that accumulates in platelet-derived growth factor- stimulated Balb/c 3T3 cells is a potential mitogenic signal. J Biol Chem 267, 10988-93.
Funahashi, J., Sekido, R., Murai, K., Kamachi, Y., and Kondoh, H. (1993). Delta- crystallin enhancer binding protein delta EF1 is a zinc finger- homeodornain protein implicated in postgastrulation embryogenesis. Development 119, 433-46.
Gehring, W. J., Affolter, M., and Burglin, T. (1994). Homeodomain proteins. Annu Rev Biochem 63, 487-526.
Gehring. W. J., Qian, Y. Q., BiUeter, M., Furukubo-Tokunaga, K., Schier, A. F., Resendez-Perez, D., Affolter, M., Otting, G., and Wuthrich, K. (1994). Homeodomain-DNA recognition. Cell 78, 2 11-23.
Genetta, T., Ruezinsky, D., and Kadesch, T. (1994). Displacement of an E-box- binding repressor by basic helix-loo p-helix proteins : implications for B-cell specificity of the immunoglobulin heavy-chain enhancer. Molecular And Cellular Biology 14, 6 153-63.
Gerber, A. N., Klesert, T. R., Bergstrom, D. A., and Tapscott, S. J. (1997). Two domains of MyoD mediate transcriptional activation of genes in repressive chromatin: a mechanism for lineage determination in myogenesis. Genes Dev 1 I, 436-50. Godbout, R., Ingram, R. S., and Tilghman, S. M. (1988). Fine-structure mapping of the three mouse alpha-fetoprotein gene enhancers. Mol Cell Biol8, 1169-78.
Godbout, R., and Tilghman, S. M. (1988). Configuration of the alpha-fetoprotein regulatory domain during development. Genes Dev 2, 949-56.
Goldhamer, D. J., Brunk, B. P., Faeman, A., King, A., Shani, M., and Emerson, C. P., Jr. (19 9 5). Embryonic activation of the myoD gene is regulated by a highly conserved distal control element. Development 121, 637-49.
Gossett, L. A., Kelvin, D. J., Stemberg, E. A., and Olson, E. N. (1989). A new myocyte-specific enhancer-binding factor that recognizes a conserved element associated with multiple muscle-specific genes. Mol Cell Biol 9, 5022-33.
Goulding, M., Sterrer, S., Fleming, J., Balling, R., Nadeau, 5.. Moore, K. J., Brown, S. D., Steel, K. P., and Gruss, P. (1993). Analysis of the Pax-3 gene in the mouse mutant splotch. Genomia 17, 355-63.
Goulding, M. D., Chalepakis, G., Deutsch, U., Enelius, J. R., and Gruss, P. (1391). Pax-3,a novel murine DNA binding protein expressed during early neurogenesis. Embo J 10, 1135-47.
Goulding, M. D., Lumsden, A., and Gruss, P. (1993). Signals from the notochord and floor plate regulate the region-specific expression of two Pax genes in the developing spinal cord. Development 11 7, 100 1-16.
Gredinger, E., Gerber, A. N., Tarnir, Y., Tapscott, S. J., and Bengal, E. (1998). Mitogen-activated protein kinase pathway is involved in the differentiation of muscle cells. Journal Of Biological Chemistry 2 73, 10436-44.
Groisman, R-,Masutani, H., Leibovitch, M. P., Robin, P., Soudant, I., Trouche, D., and Harel-Bellan, A. (1996). Physical interaction between the mitogen- responsive serum response factor and myogenic basic-helix-loop-helix proteins. J Biol Chem 271, 5258-64.
Gu, W., Schneider, J- W., Condorelli, G., Kaushal, S., Iviahdavi, V., and Nadal- Ginard, B. (1993). Interaction of myogenic factors and the retinoblastoma protein mediates muscle cell commitment and differentiation. Cell 72. 309-24.
Guo, K., and Walsh, K. (1997). Inhibition of myogenesis by multiple cyclin-Cdk complexes. Coordinate regulation of myogenesis and cell cycle activity at the level of E2F. J Biol Chem 272, 79 1-7. Guo, K.,Wang, J., Andra, V., Smith, R. C., and Walsh, K. (1995). MyoD-induced expression of p2 1 inhibits cyclin-dependent kinase activity upon myocyte terminal differentiation. Molecular And Cellular Biology 15, 3823-9.
Halevy, 0..Novitch, B. G., Spicer, D. B., Skapek, S. X., Rhee, J., Hannon, G. J., Beach, D., and Lassar, A. B. (1995). Correlation of terminal cell cycle arrest of skeletal muscle with induction of p2 1 by MyoD. Science 26 7, 1018-2 1.
Han, K.,and Manley, J. L. (1993). Transcriptional repression by the Drosophila even-skipped protein: defmition of a minimal repression domain. Genes And Development 7, 49 1-503.
Hanes, S. D.,and Brent, R. (199 1). A genetic model for interaction of the homeodornain recognition helix with DNA. Science 251, 426-30.
Hanna-Rose, W., and Hansen, U. (1996). Active repression mechanisms of eukaryotic transcription repressors. Trends In Genetics 12, 229-34.
Hardy, S., Kong, Y., and Konieczny, S. F. (1993). Fibroblast growth factor inhibits MRF4 activity independently of the phosphorylation status of a conserved threonine residue within the DNA- binding domain. Mol Cell Biol 13. 5943-56.
Hashirnoto, T., Nakano, Y., Morinaga, T., and Tamaoki, T. (1992). A new family of homeobox genes encoding multiple homeodomain and zinc finger motifs. Mechanisms Of Development 39, 125-6.
Hassig, C. A., Fleischer, T. C.. Billin, A. N., Schreiber, S. L., and Ayer, D. E. (1997). Histone deaceqlase activity is required for full transcriptional repression by mSin3A. Cell 89, 34 1-7.
Hassig, C. A., Tong, J. K., Fleischer, T. C., Owa, T., Grable, P. G., Ayer, D. E., and Schreiber, S. L. (1998). A role for histone deacetylase activity in HDAC1- mediated transcriptional repression. Proc Natl Acad Sci U S A 95, 35 19-24.
Hasty, P., Bradley, A., Moms, J. H., Edmondson, D. G., Venuti, J. M., Olson, E. N., and Klein, W. H. (1993). Muscle deficiency and neonatal death in mice with a targeted mutation in the myogenin gene. Nature 364, 50 1-6.
Hebbes, T. R., Thome, A. W., and Crane-Robinson, C. (1988). A direct link between core histone acetylation and transcriptionally active chromatin. Embo J 7, 1395-402. Heinzel, T., Lavinsky, R. M., Mullen, T. M., Sodeirstrom, M., Laherty, C. D., Torchia, J., Yang, W. M., Brard, G., Ngo, S. D., Davie, J. R., Seto, E., Eisenrnan, R. N.. Rose, D. W., Glass, C. K., and Rosenfeld, M. G. (1997). A complex containing N-CoR, mSin3 and histone deacetylase mediates transcriptional repression. Nature 38 7, 43-8.
Higashi, Y., Moribe, H., Takagi, T., Sekido, R., Kawakami, K., Kikutani, H., and Kondoh, H. (1997). Impairment of T cell development in deltaEFl mutant mice. J Exp Med 185, 1467-79.
Hirsch, M-R., Valarche, I., Deagostini-Bazin. H., PerneIle, C., Joliot, A., Goridis, C. (1991). An upstream regulatory element of the NCAM promoter contains a binding site for homeodomains. FEBS Lett. 287, L97-202.
Hogan. B., Beddington, R., Costantini, F., and Lacy, E. (1994). Manipulating the mouse embryo. A laboratory manual.: Cold Spring Harbour Laboratory Press), pp. 74-8 1.
Hollemann, T., Bellefroid, E., Stick, R., and Pieler, T. (1996). Zinc fmger proteins in early Xenopus development. Int J Dev Biol 40, 291-5.
Hsieh, JJ-D., and Hayward, S.D. (1995). Masking of the CBF l/RBPJ, transcriptional repression domain by Epstein-Barr Mrus EBNAZ. Science 268, 560- 563.
Hunt, P., and KNmlauf, R. (1992). Hox codes and positional specification in vertebrate embryonic axes. Annu Rev Cell Biol8, 227-56.
Ido, A., Miura, Y., and Tarnaoki, T. (1994). Activation of ATBF1, a multiple- homeodomain zinc-fmger gene, during neuronal differentiation of murine embryonal carcinoma cells. Developmental Biologqr 163, 184-7.
Ido, A., Miura, Y., Watanabe, M., Sakai, M., Inoue, Y., Miki, T., Hashirnoto, T., Morinaga, T., Nishi, S., and Tamaoki, T. (1996). Cloning of the cDNA encoding the mouse ATBFl transcription factor. Gene 168, 227-3 1.
Ikeda, K., Halle, J. P., Stelzer, G., Meisterernst, M.. and Kawakami, K. (1998). Involvernent of negative cofactor NC2 in active repression by zinc fmger- homeodomain transcription factor AREB6. Mol Cell Biol 18, 10-8.
Ikeda, K., and Kawakami. K. (1995). DNA binding through distinct domains of zinc-fmger-homeodomain protein AREB6 has different effects on gene transcription. Eur J Biochem 233, 73-82.
Jacobson, E.M., Li, P.. Leon-del-Rio, A., Rosenfield, M.G., and Aggmal, A.K. (1997). Structure of Pit-1 POU domain bound to DNA as a dirner: unexpected arrangement and flexibility- Genes and Dev. 1I, 198-2 12.
Jen, Y., Weintraub, H., and Benezra. R. (1992). Overexpression of Id protein inhibits the muscle differentiation program: in vivo association of Id with E2A proteins. Genes And Development 6, 1466-79.
Johnson, S. E., and Allen, R. E. (1990). The effects of bFGF, IGF-I, and TGF-beta on RMo skeletal muscle cell proliferation and differentiation. Exp CeIl Res 187, 250-4.
Johnson, W. A., McCormick, C. A., Bray, S. J., and Hinh, J. (1989). A neuron- specific enhancer of the Drosophila dopa decarboxylase gene. Genes And Development 3, 676-86.
Kablar, B., Krastel, K., Ying, C., Asakura, A., Tapscott, S. J., and Rudnicki, M. A. (1997). MyoD and Myf-5 differentially regulate the development of Limb versus trunk skeletal muscle. Development 124, 4729-38.
Kablar, B., Krastel, K., Ying, C., Tapscott, S. J., Goldharner, D. J., and Rudnicki, M. A. (1999). Myogenic determination occurs independently in somites and limb buds. Dev Biol206, 2 19-31.
Kageyama, R., and Nakanishi, S. (1997). Helix-loop-helix factors in growth and differentiation of the vertebrate nervous system. Curr Opin Genet Dev 7, 659-65.
Karnps, M. P., Taylor, S. S., and Sefton, B. M. (1984). Direct evidence that oncogenic tyrosine kinases and cyclic AMP- dependent protein kinase have homologous ATP-binding sites. Nature 31 0,589-92.
Karlsson, 0..Thor, S., Norberg, T., Ohlsson, H. and Edlund, T. (1990). Insulin gene enhancer binding protein Isl-1 is a member of a novel class of proteins containing both a homeo- and a Cys-His domain. Nature 344,879-882.
Kaspar, P., Dvorakova, M., Kralova, J., Pajer, P., Kozrnik, 2.. and Dvorak, M. (1999). Myb-interacting protein, ATBF 1, represses transcriptional activity of Myb oncoprotein. J Biol Chem 274, 14422-8.
Katagiri, T., Akiyarna, S., Narniki, M., Komaki, M., Yarnaguchi, A., Rosen, V., Wozney, J. M., Fujisawa-Sehara, A., and Suda, T. (1997). Bone morphogenetic protein-2 inhibits terminal differentiation of myogenic cells by suppressing the transcriptional activity of MyoD and myogenin. Exp Cell Res 230,342-5 1.
Katagiri, T., Yamaguchi, A.. Komaki, M., Abe, E., Takahashi, N., Ikeda, T., Rosen, V., Wozney, J. M., Fujisawa-Sehara. A., and Suda, T. (1994). Bone morphogenetic protein-2 converts the differentiation pathway of C2C 12 myoblasts into the osteoblast lineage. J Cell Biol 127, 1755-66.
Kataoka, H,, joh, T., Miura, Y., Tamaoki, T., Senoo, K.,Ohara, H., Nornura, T., Tada, T., Asai, K., Kato, T., and Itoh, M. (2000). AT motif binding factor 1-A (ATBF1-A) negatively regulates transcription of the aminopeptidase N gene in the crypt-villus axis of small intestine. Biochem Biophys Res Commun 26 7, 9 1-5.
Kassis, J. A.. Desplan, C., Wright, D. K. and O'Farrell, P. H. (1989) Evolutionary Conservation of Homeodomain-Binding Sites and Other Sequences Upstream and within the Major Transcription Unit of the Drosophila Segmentation Gene engrailed. Mol. Cell. Biol. 9, 4304-431 1.
Kim, S. J., Kim, K. Y., Tapscott, S. J., Winokur, T. S., Park, K., Fujiki, H.. Weintraub, H., and Roberts, A. B. (1 992). Inhibition of protein phosphatases blocks myogenesis by fmt altering MyoD binding activity. Journal Of Biological Chemistry 267, 15140-5.
Kingston. R. E., and Narlikar, G. J. (1999). ATP-dependent remodeling and acetylation as regulators of chromatin fluidity. Genes Dev 13, 2339-52.
Kitrmann, M., Camac, G., Vandromme, M., Primig. M.. Lamb, N. J., and Fernandez, A. (1998). The muscle regulatory factors MyoD and myf-5 undergo distinct cell cycle-specific expression in muscle cells. Journal Of Cell Biology 142, 1447-59.
Kitzrnann, M., Vandromme, M., Schaeffer, V., Carnac, G., Labbe, J. C., Lamb, N., and Fernandez, A. (1999). cdkl- and cdk2-mediated phosphorylation of MyoD Ser200 in growing C2 myoblasts : role in modulating MyoD half-life and rnyogenic activity. Molecular And Cellular Biology 19, 3 167-76.
Klemm, J. D., Rould, M. A., Aurora. R.. Herr, W., and Pabo, C. 0. (1994). Crystal structure of the Oct-1 POU domain bound to an octamer site: DNA recognition with tethered DNA-binding modules. CeIl 77, 2 1-32.
Koenig, H. (199 1). Cell-type specific multiprotein complex formation over the c-fos serum response element in vivo: ternary complex formation is not required for the induction of c-fos. Nuc. Acids Res. 19:3607-3611 (1991).
Kong. Y., Johnson, S. E.. Taparowsky. E. J.. and Konieczny. S. F. (1995). Ras p2 lVal inhibits myogenesis without altering the DNA binding or transcriptional activities of the myogenic basic helix-loop-helix factors. Mol Cell Biol 15, 5205-13.
Koniecmy, S. F.. Drobes, B. L.. Menke. S. L., and Taparowsky. E. J. (1989). Inhibition of myogenic differentiation by the H-ras oncogene is associated with the down regulation of the MyoD l gene. Oncogene 4.473-8 1.
Kostich. W. A., and Sanes. J. R. (1995). Expression of zfh-4.a new member of the zinc fmger-homeodomain family, in developing brain and muscle. Developmental Dynamics 202, 145-52.
Krempler. A., and Brenig, B. (1999). Zinc fmger proteins: watchdogs in muscle development. Molecular And General Genetics 261,209- 15.
Kucharczuk. K. L.. Love. C. M.. Dougherty. N. M.. and Goldharner, D. J. (1999). Fine-scale transgenic mapping of the MyoD core enhancer: MyoD is regulated by distinct but overlapping mechanisms in myotomal and non- myotomal muscle lineages. Development 126. 195 7-65.
Kuo, M. H.. and AUis. C. D. (1998). Roles of histone acetyltransferases and deacetylases in gene regulation. Bioessays 20. 6 15-26.
Laernmli. U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227. 680-5.
Laherty. C. D.. Yang, W. M., Sun. J. M.. Davie, J. R.. Seto. E., and Eisenman. R. N. (1997). Histone deacetylases associated with the mSin3 corepressor mediate mad transcriptional repression. Cell 89, 349-56.
Lai. 2. C.. Fortini, M. E.. and Rubin. G. M. (1991). The embryonic expression patterns of zfh-1 and zfh-2. two Drosophila genes encoding novel zinc-finger homeodomain proteins. Mechanisms Of Development 34. 123-34.
Lai, 2. C.. Rushton. E.. Bate. M.. and Rubin, G. M. (1993). Loss of hnction of the Drosophila zfh- l gene results in abnormal development of mesodermally derived tissues. Proc Natl Acad Sci U S A 90. 4 122-6.
Laird. P. W.. Zijderveld. A.. Linders, K.. Rudnicki, M. A., Jaenisch, R.. and Berns, A. (199 1). Simplified rnarnmallan DNA isolation procedure. Nucleic Acids Res 19, 4293.
Laoide, B- M., Foulkes, N. S., Schlotter, F., and Sassone-Corsi, P. (1993). The functional versatility of CREM is determined by its modular structure. Embo Journal 12, 1179-9 1-
Lassar, A. B., Davis, R. L., Wright, W. E., Kadesch, T., Murre. C., Voronova, A., Baltimore. D., and Weintraub, H. (199 1). Functional activity of myogenic HLH proteins requires hetero-oligomerization with E 12/E47-like proteins in vivo. Cell 66, 305-15.
Lassar, A. B.. Skapek. S. X., and Novitch, B. (1994). Regulatory mechanisms that coordinate skeletal muscle differentiation and celI cycle withdrawal. Current Opinion In Cell Biology 6, 788-94.
Lassar, A. B., Thayer. M. J., Overell, R. W., and Weintraub, H. (1989). Transformation by activated ras or fos prevents myogenesis by inhibiting expression of MyoD 1. Cell 58, 659-67.
Lee, Y-,Shioi, T-,Kasahara, H., Jobe, S. M., Wiese, R. J., Markham, B. E., and Izumo, S. (1998). The cardiac tissue-restricted homeobox protein CsxNkx2.5 physically associates with the zinc fmger protein GATA4 and cooperatively activates atrial natriuretic factor gene expression. Molecular And Cellular Biology 18, 3120-9.
Lemercier, C., To, R. Q., Carrasco, R. A., and Konieczny. S. F. (1998). The basic helix-loop-helix transcription factor Mist 1 functions as a transcriptional repressor of myoD. Embo J 17, 1412-22.
Li, T.. Stark, M. R., Johnson, A. D., and Wolberger, C. (1995). Crystal structure of the MATal/MAT alpha 2 homeodomain heterodimer bound to DNA. Science 270, 262-9.
Liang, 2.. and Biggin, M. D. (1998). Eve and ftz regulate a wide array of genes in blastoderm embryos: the selector homeoproteins directly or indirectly regulate most genes in Drosophila. Development 125, 447 1-82.
Licht, J. D., Grossel, M. J., Figge, J., and Hansen, U. M. (1990). Drosophila Kriippel protein is a transcriptional repressor. Nature 346, 76-9. Lopez, A. J. (1995). Developmental role of transcription factor isoforms generated by alternative splicing. Developmental Biology 172, 396-4 11.
Lu, J-,Webb, R., Richardson, J. A,, and Olson, E. N. (1 999). MyoR: a muscle- restricted basic helix-loop-helix transcription factor that antagonizes the actions of MyoD. Proceedings Of The National Academy Of Sciences Of The United States Of America 96. 55 2-7.
Lundell. M. J., and Hirsh, J. (1992). The zfh-2 gene product is a potential regulator of neuron-specific dopa decarboxylase gene expression in Drosophila. Developmental Biology 154, 84-94.
Lyons, G. E., and Buckingharn, M. E. (1992) Developmental regulation of myogenesis in the mouse. Semin. Dev. Biol. 3, 243-243.
Lyons. G. E.. Micales, B. K., Schwarz. J.. Martin. J. F.. and Olson. E. N. (1995). Expression of mef2 genes in the mouse central nervous system suggests a role in neuronal maturation- J Neurosci 15, 5727-38.
Ma. D. (1997). Structure and function of mouse ATBF 1 gene. M.Sc. Thesis. University of Calgary.
Ma. P. C., Rould, M. A., Weintraub, H., and Pabo, C. 0. (1994). Crystal structure of MyoD bHLH domain-DNA complex: perspectives on DNA recognition and implications for transcriptional activation. Cell 77, 45 1-9.
Madden, S. L., Cook. I3. M., and Rauscher. F. J. d. (1993).A structure-function analysis of transcriptional repression mediated by the WT 1, Wilms' tumor suppressor protein. Oncogene 8, 1713-20.
Madly, F.. Berube, G., Harada, R.. Mao. P. L., Phillips, S.. and Nepveu, A. (1996). The human cut homeodomain protein can repress gene expression by two distinct mechanisms: active repression and competition for binding site occupancy. Molecular And Cellular Biology 16, 5346-57.
Mannervik. M. (1999). Target genes of homeodomain proteins. Bioessays 21. 267- 70.
Mansouri, A., and Gruss, P. (1998). Pax3 and Pax7 are expressed in cornmissural neurons and restrict ventral neuronal identity in the spinal cord. Mech Dev 78, 171-8. Maroto, M., Reshef, R., Munsterberg, A. E.. Koester. S.. Goulding, M., and Lassar, A. B. (1997). Ectopic Pax-3 activates MyoD and Myf-5 expression in embryonic mesoderm and neural tissue- Cell 89, 139-48.
Massague, J., Cheifetz, S., Endo, T., and Nadal-Ginard, B. (1986). Type beta transforming growth factor is an inhibitor of rnyogenic differentiation. Proc Natl Acad Sci U S A 83, 8206-10.
Mathey-Prevot, B.. Andrews, N. C., Murphy, H. S., Kreissman, S. G. and Nathan, D . G. (1990). Positive and negative elements regulate human interleukin 3 expression. Proc. Natl. Acad. Sci. USA 87,5046-5050 .
McBurney, M. W. (1 99 3). P 19 embryonal carcinoma cells. International Journal Of Developmental Biology 37. 135-40.
McBurney. M. W., Jones-Villeneuve, E. M., Edwards. M. K.,and Anderson, P. J. (1982). Control of muscle and neuronal differentiation in a cultured embryonal carcinoma cell line- Nature 299, 165-7-
McGeachie, J. K., and Grounds, M. D. (1987). Initiation and duration of muscle precursor replication after mild and severe injury to skeletal muscle of mice. An autoradiographic study. Cell Tissue Res 248, 125-30.
McGinnis, W., and Krurnlauf, R. (1992). Homeobox genes and axial patterning. Cell 68, 283-302.
Megeney, L. A., Kablar, B.. Garrett, K., Anderson, J. E., and Rudnicki, M. A. (1996). MyoD is required for myogenic stem cell function in adult skeletal muscle. Genes Dev 10, 1173-83.
Melnikova, I. N.. and Christy, B. A. (1996). Muscle cell differentiation is inhibited by the helix-loop-helix protein Id3. Cell Growth Differ 7, 1067-79.
Miller. J. B. (1990). Myogenic programs of mouse muscle cell lines : expression of myosin heavy chain isoforms. MyoD1, and myogenin. J Cell Biol 111, 1149-59.
Miner. J. H., and Wold, B. (1990). Herculin, a fourth member of the MyoD family of myogenic regulatory genes. Proceedings Of The National Academy Of Sciences Of The United States Of America 87, 1089-93.
Miner, J. H., and Wold, B. J. (199 1).c-myc inhibition of MyoD and myogenin- initiated myogenic differentiation. Molecular And Cellular Biology 1 1. 28 42-5 1. Missero, C., Calautti, E., Eckner, R.. Chin, J., Tsai, L. H., Livingston, D. M., and Dotto. G. P. (1995). Involvement of the cell-cycle inhibitor Cipl/WAFl and the E1A-associated p300 protein in terminal differentiation. Proceedings Of The National Academy Of Sciences Of The United States Of America 92. 545 1-5.
Mitchell, P. J.. and Tjian, R. (1989). Transcriptional regulation in mammalian cells by sequence-specific DNA binding proteins. Science 245. 371-8.
Miura, Y., Tam, T., Ido, A*. Morinaga. T., Miki, T., Hashimoto, T., and Tamaoki, T. (1995). Cloning and characterization of an ATBFl isofom that expresses in a neuronal differentiation-dependent manner. Journal Of Biological Chemistry 270, 26840-8.
Molkentin, J. D., Black. B. L., Martin, J. F., and Olson. E. N. (1995). Cooperative activation of muscle gene expression by MEF2 and myogenic bHLH proteins. Cell 83, 1125-36.
Molkentin, J. D., and Olson, E. N. (1996). Combinatorial control of muscle development by basic helix-loop-helix and MADS-box transcription factors. Proceedings Of The National Academy Of Sciences Of The United States Of America 93, 9366-73.
Moolenaar, W. H. (1994). LPA: a novel lipid mediator with diverse biological actions. Trends in Cell Biology 4, 2 13-2 19.
Morinaga, T., Yasuda, H., Hashimoto, T., Higashio, K., and Tarnaoki, T. (199 1). A human alpha-fetoprotein enhancer-binding protein. ATBF 1, contains four homeodomains and seventeen zinc fmgers. Molecular And Cellular Biology 11, 6041-9.
Miinsterberg. A. E., Kitajewski, J., Burncrot. D. A.. McMahon. A. P., and Lassar, A. B. (1995). Combinatorial signaling by Sonic hedgehog and Wnt family members induces myogenic bHLH gene expression in the somite. Genes And Development 9, 2911-22,
Miinsterberg, A. E., and Lassar, A. B. (1995). Combinatorial signals from the neural tube, floor plate and notochord induce myogenic bHLH gene expression in the somite. Development 121, 65 1-60.
Nabeshima, Y.. Hanaoka, K., Hayasaka, M.. Esumi, E., Li, S., and Nonaka, I. (199 3). Myogenin gene disruption results in perinatal lethality because of severe muscle defect. Nature 364. 532-5. Nadal-Ginard, B . (1 978). Commitment, fusion and biochemical differentiation of a myogenic cell line in the absence of DNA synthesis. Cell 15. 855-64.
Nagy. L.. Kao. H. Y., Chakravarti, D., Lin, R. J.. Hassig. C. A.. Ayer. D. E.. Schreiber. S. L.. and Evans. R. M. (1997). Nuclear receptor repression mediated by a complex containing SMRT. mSin3A. and histone deacetylase. Cell 89. 373-80.
Naidu, P. S.. Ludolph. D. C.. To. R. Q.. Hinterberger, T. J., and Konieczny. S. F. (1995). Myogenin and MEF2 function synergistically to activate the MRF4 promoter during myogenesis. Mol Cell Biol 15, 2707-18.
Neuhold. L. A.. and Wold. B. (1993). HLH forced dimers: tethering MyoD to E47 generates a dominant positive myogenic factor insulated from negative regulation by Id. Cell 74, 1033-42.
Novitch, B. G.. Mulligan. G. J., Jacks. T., and Lassar, A. B. (1996). Skeletal muscle cells lacking the retinoblastoma protein display defect. in muscle gene expression and accumulate in S and G2 phases of the cell cycle. J Cell Biol 135, 441-56.
Ohneda. K.. Mirmira, R. G.. Wang, J., Johnson. J. D.. and German. M. S. (2000). The homeodomain of PDX-1 mediates multiple protein-protein interactions in the formation of a transcriptional activation complex on the insulin promoter. Mol Cell Biol20. 900- 11.
Olson, E. N. (1992). Interplay between proliferation and differentiation within the myogenic lineage. Developmental Biology 154, 26 1-72.
Olson. E. N.. Arnold. H. H.. Rigby. P. W.. and Wold. B. J. (1996). Know your neighbors: three phenotypes in null mutants of the myogenic bHLH gene MW4. Cell 85, 1-4.
Olson, E. N.. Perry, M.. and Schulz. R. A. (1995). Regulation of muscle differentiation by the MEFZ family of MADS box transcription factors. Developmental Biology 1 72, 2-14.
Ordahl, C. P.. and Williams, B. A. (1998). Knowing chops from chuck: roasting myoD redundancy. Bioessays 20, 357-62.
Ornatsky. 0.I., Andreucci, J. J.. and McDermott, J. C. (1997). A dominant- negative form of transcription factor MEFZ inhibits myogenesis. .J Biol Chem 272, 33271-8. Ott, M. O., Bober. E.,Lyons, G., Arnold, H., and Buckingham, M. (1991). Early expression of the myogenic regulatory gene, myf-5, in precursor cells of skeletal muscle in the mouse embryo. Development 11 1, 1097-1 107.
Otten, A. D., Firpo, E. J., Gerber. A. N., Brody, L. L., Roberts. J. M., and Tapscott, S. J. (1997). Inactivation of MyoD-mediated expression of p21 in tumor cell lines. Cell Growth Differ 8, 1 151-60.
Otting, G., Qian, Y. Q., Billeter, M., Muller, M., Affolter, M., Gehring, W. J., and Wuthrich, K. (1990). Protein-DNA contacts in the structure of a homeodomain-- DNA complex determined by nuclear magnetic resonance spectroscopy in solution. Embo J 9, 3085-92.
Park, K.. Chung, M., and Kim, S. J. (1 992). Inhibition of myogenesis by okadaic acid, an inhibitor of protein phosphatases, 1 and 2A, correlates with the induction of AP1. J Biol Chem 267, 108 10-5.
Parker, S. B., Eichele, G.. Zhang, P., Rawls, A.. Sands, A. T., Bradley. A.. Olson, E. N.. Harper, J. W., and Elledge, S. J. (1995). p53-independent expression of p2l Cipl in muscle and other terminally differentiating cells. Science 267. 1024-7.
Passner, J. M., Ryoo, H. D., Shen, L., Mann, R. S., and Aggarwal, A. K. (1999). Structure of a DNA-bound Ultrabithorax-Extradenticle homeodomain complex. Nature 39 7, 7 14-9.
Patapoutian, A., Miner, J. H.. Lyons, G. E., and Wold, B. (1993). Isolated sequences from the linked Myf-5 and MRF4 genes drive distinct patterns of muscle-specific expression in transgenic mice. Development 118, 6 1-9.
Patapoutian, A., Yoon, J. K., Miner, J. H.. Wang, S.. Stark. K.. and Wold, B. (1995). Disruption of the mouse MRF4 gene identifies multiple waves of myogenesis in the myotome. Development 121, 3347-58.
Postigo, A. A,, and Dean, D. C. (1999). Independent repressor domains in ZEB regulate muscle and T-cell differentiation. Mol Cell Biol 19, 796 1-7 1.
Postigo, A. A., and Dean, D. C. (1997). ZEB, a vertebrate homolog of Drosophila Zfh-1, is a negative regulator of muscle differentiation. Embo Jourral 16, 3935-43.
Postigo, A. A., Ward, E., Skeath. J. B., and Dean, D. C. (1999). zfh-1, the Drosophila Homologue of ZEB, Is a Transcriptional Repressor That Regulates Somatic Myogenesis. Mol Cell Biol 19, 7255-7263. Puri, P. L., Avantaggiati, M. L., Balsano. C., Sang, N.. Graessmann, A., Giordano. A., and Levrero, M. (1997). p300 is required for MyoD-dependent cell cycle arrest and muscle- specific gene transcription. Embo J 16, 369-83.
Puri, P. L., SartoreIli, V., Yang, X. J.. Hamamori. Y., Ogryzko. V. V., Howard, B. H.. Kedes, L., Wang, J. Y., Graessmann. A., Nakatani, Y.. and Levrero, M. (1997). Differential roles of p300 and PCAF acetyltransferases in muscle differentiation. Mol Cell 1, 35-45.
Puri, P.L., Wu, Z., Zhang, P., Wood, L.D., Bhakta, K.S., Han, J.. Fermisco. J.R.. Karin, M., and Wang, J.Y. J. (2000). Induction of terminal differentiation by constitutive activation of p38 MAP base in human rhabdomyosarcoma cells. Genes and Dev. 14, 574-584.
Qian. Y. Q., Furukubo-Tokunaga, K.. Resendez-Perez. D.. Muller, M., Gehring, W. J., and Wuthrich. K. (1994). Nuclear magnetic resonance solution structure of the fushi tarazu horneodomain from Drosophila and comparison with the Antennapedia homeodomain. J Mol Biol238, 333-45.
Qian. Y. Q.. Otting, G., Furukubo-Tokunaga, K.. Affolter. M., Gehring, W. J., and Wuthrich, K. (1992). NMR structure determination reveals that the homeodomain is connected through a flexible linker to the main body in the Drosophila Antennapedia protein. Proc NatI Acad Sci U S A 89,10738-42.
Rao, S. S.. Chu, C., and Kohtz, D. S. (1994). Ectopic expression of cyclin D 1 prevents activation of gene transcription by myogenic basic helix-loop-helix regulators. Molecular And Cellular Biology 14, 5259-67.
Rao. S. S., and Kohtz, D. S. (1995). Positive and negative regulation of D-type cyclin expression in skeletal myoblasts by basic fibroblast growth factor and transforming growth factor beta. A role for cyclin D 1 in control of myoblast differentiation. Joumal Of Biological Chemistry 2 70, 4093- 100.
Rawls, A., and Olson, E. N. (1997). MyoD meets its maker. Cell 89, 5-8.
Regulski, M., Dessain, S., McGinnis, N.. McGinnis. W. (19 9 1). High-affinity binding sites for the deformed protein are required for the function of an autoregulatory enhancer of the Deformed gene. Genes Dev. 5, 278-86.
Reynaud. E. G., Pelpel, K., Guillier, M., Leibovitch. M. P.,and Leibovitch, S. A. (1999). p57(Kip2) stabilizes the MyoD protein by inhibiting cyclin E-Cdk2 kinase activity in growing myoblasts. Mol Cell Biol 19. 762 1-9. Rhodes, S. J., and Konieany, S. F. (1989). IdentiCication of MRF4: a new member of the muscle regulatory factor gene family. Genes And Development 3, 2050-6 1.
Rudnicki, M. A., Braun, T., Hinuma, S., and Jaenisch, R. (1992). Inactivation of MyoD in mice leads to up-regulation of the myogenic HLH gene Myf-5 arid results in apparently normal muscle development. Cell 71,383-90.
Rudnicki. M. A., and McBurney, M. W. (1987). Cell culture methods and induction of differentiation of embryonal carcinoma cell lines. In Teratocarcinomas and Embryonic Stem Cells: A Practical Approach., E. J. Robertson. ed. (Oxford: IRL Press) , pp. 19-49.
Rudnicki, M. A., Schnegelsberg, P. N.. Stead, R. H., Braun. T., Arnold, H. H., and Jaenisch, R. (1 993) . MyoD or Myf-5 is required for the formation of skeletal muscle. Cell 75, 1351-9.
Sabourin. L. A*, Girgis-Gabardo, A*,Seale, P., Asakura, A.. and Rudnicki, M. A. (1999). Reduced differentiation potential of primary MyoD-/- myogenic cells derived from adult skeletal muscle. Journal Of Cell Biology 144. 631-43.
Salminen, A., Braun, T., Buchberger, A.. Jiirs, S., Winter, B., and Arnold, H. H. (199 1).Transcription of the muscle regulatory gene Myf4 is regulated by serum components, peptide growth factors and signaling pathways involving G proteins. Journal Of Cell Biology 115, 905-17.
Sartorelli, V., Humg, J., Harnarnori, Y., and Kedes. L. (1997). Molecular mechanisms of myogenic coactivation by p300: direct interaction with the activation domain of MyoD and with the MADS box of MEFPC. Mol Cell Biol 17, 1010-26.
Sassoon, D., Lyans, G., Wright, W. E., Lin, V., Lassar, A., Weintraub, H., and Buckingham, M. (1989). Expression of two myogenic regulatory factors myogenin and MyoD 1 during mouse embryogenesis. Nature 341, 303-307.
Schier, A. F., and Gehring, W. J. (1992). Direct homeodomain-DNA interaction in the autoregulation of the fushi tarazu gene. Nature 356, 804-7.
Schneider, J. W., Gu, W., Zhu, L., Mahdavi, V., and Nadal-Ginard, B. (1994). Reversal of terminal differentiation mediated by p 107 in Rb-/- muscle cells. Science 264, 1467-7 1.
Sekido. R., Murai, K., Funahashi, J., Karnachi, Y., Fujisawa-Sehara, A., Nabeshirna, Y., and Kondoh, H. (1994). The delta-crystallin enhancer-binding protein delta EFI is a repressor of E2-box-mediated gene activation. Molecular And Cellular Biology 14, 5692-700.
Skapek, S. X.,Rhee, J., Kim, P. S., Novitch, B. G., and Lassar, A. B. (1996). Cyclin-mediated inhibition of muscle gene expression via a mechanism that is independent of pRB hyperphosphorylation. Mol Cell Biol 16, 7043-53.
Skapek. S. X.,Rhee, J., Spicer, D. B., and Lassar, A. B. (1995). Inhibition of myogenic differentiation in proliferating myo blasts by cyclin D 1-dependent kinase. Science 267, 1022-4.
Skerjanc, I. S., Truong, J., Filion, P., and McBumey, M. W. (1996). A splice variant of the ITF-2 transcript encodes a transcription factor that inhibits MyoD activity. J Biol Chem 271, 3 555-6 1.
Smith, T. H., Kachinsky, A. M., and Miller, J. B. (1994). Sornite subdomains, muscle cell origins, and the four muscle regulatory factor proteins. J Cell Biol 127, 95-105.
Song, A., Wang, Q., Goebl, M. G., and Harrington, M. A. (1998). Phosphorylation of nuclear MyoD is required for its rapid degradation. Mol Cell Biol 18, 4994-9.
Sorrentino, V., Pepperkok, R., Davis, R. L., Ansorge, W., and Philipson, L. (1990). Cell proliferation inhibited by MyoD 1 independently of myogenic differentiation. Nature 345, 8 13-5.
Spence, M. S., Yip, J., and Erickson, C. A. (1996). The dorsal neural tube organizes the dermamyotome and induces axial myocytes in the avian embryo. Development 122, 231-41.
Stern, H. M., Brown, A. M., and Hauschka, S. D. (1995). Myogenesis in paraxial mesoderm: preferential induction by dorsal neural tube and by cells expressing Wnt-1. Development 121, 3675-86.
Stern, H. M., Lin-Jones, J., and Hauschka, S. D. (1997). Synergistic interactions between bFGF and a TGF-beta family member may mediate myogenic signals from the neural tube. Development 124, 35 11-23.
Tajbakhsh, S., Borello, U., Vivarelli, E., Kelly, R.. Papkoff, J., Duprez, D., Buckingham, M., and Cossu, G. (1998). Differential activation of Myf5 and MyoD by different Wnts in explants of mouse paraxial mesoderm and the later activation of myogenesis in the absence of Myf5. Development 125, 4 155-62.
Tajbakhsh, S., Rocancourt, D., Cossu, G., and Buckingham, M. (1997). Redefining the genetic hierarchies controlling skeletal myogenesis: Pax- 3 and Myf-5 act upstream of MyoD. Cell 89, 127-38.
Takagi, T., Moribe, H., Kondoh, H., and Higashi, Y. (1998). (delta)EF 1, a zinc fmger and homeodomain transcription factor, is required for skeleton patterning in multiple lineages. Development 125, 2 1-3 1.
Tanaka. T., Tanaka, K.. Ogawa, S., Kurokawa, M., Mitani, K., Nishida, J., Shibata, Y., Yazaki, Y., and Hirai, H. (1995). An acute myeloid leukemia gene, AML 1, regulates hemopoietic myeloid cell differentiation and transcriptional activation antagonistically by two alternative spliced forms. Embo Journal 14, 34 1-50.
Tiainen. M., Pajalunga, D., Femntelli, F., Soddu, S., Salvatori, G., Sacchi, A., and Crescenzi, M. (1996). Terminally differentiated skeletal myotubes are not confined to GO but can enter GI upon growth factor stimulation. Cell Growth Differ 7, 1039-50.
Tremblay, P., Kessel, M., and Gruss, P. (1995). A transgenic neuroanatomical marker identifies cranial neural crest deficiencies associated with the Pax3 mutant Splotch. Dev Biol 171, 3 17-29.
Tsai. R. Y., and Reed, R. R. (1998). Identification of DNA recognition sequences and protein interaction domains of the multiple-Zn-finger protein Roaz. Molecular And CeUuIar Biology 18, 6447-56.
Urn. M., Li, C., and Manley, J. L. (1995). The transcriptional repressor even- skipped interacts directly with TATA-binding protein. Molecular And Cellular Biology 15, 5007-16.
Vaidya, T. B., Rhodes. S. J., Taparowsky, E. J., and Konieczny, S. F. (1989). Fibroblast growth factor and transforming growth factor beta repress transcription of the myogenic regulatory gene MyoDl. Mol Cell Biol9, 3576-9.
Venot, C., Maratrat, M., Dureuil, C., Conseiller, E., Bracco, L., and Debussche, L. (1998). The requirement for the p53 proline-rich functional domain for mediation of apoptosis is correlated with specific PIG3 gene transactivation and with transcriptional repression. Embo Journal I 7, 4668-79.
Venuti, J. M., Moms, J. H., Vivian, J. L., Olson, E. N., and Klein, W. H. (1995). Myogenin is required for late but not early aspects of rnyogenesis during mouse development. Journal Of Cell Biology 128, 563-76.
Walsh, K. (1997). Coordinate regulation of cell cycle and apoptosis during myogenesis. Prog Cell Cycle Res 3, 53-8.
Walsh, K., and Perlman, H. (1997). Cell cycle exit upon myogenic differentiation. Curr Opin Genet Dev 7, 597-602.
Walter, J., Dever, C. A., and Biggin, M. D. (1994). Two homeo domain proteins bind with similar specificity to a wide range of DNA sites in Drosophila embryos. Genes Dev 8, 1678-92.
Wang, J., Guo, K., Wills, K. N., and Walsh, K. (1997). Rb functions to inhibit apoptosis during myocyte differentiation. Cancer Res 57, 35 1-4.
Watanabe, M., Miura, Y., Ido, A., Sakai, M., Nishi, S., Inoue, Y., Hashirnoto, T., and Tamaoki, T. (1996) . Developmental changes in expression of the ATBF 1 transcription factor gene. Brain Research. Molecular Brain Research 42, 344-9.
Watanabe, Y., Kawakarni, K., Hirayama, Y., and Nagano, K. (1993). Transcription factors positively and negatively regulating the Na,K- ATPase alpha 1 subunit gene. J Biochem (Tokyo) 114, 849-55.
Weintraub, H., Davis, R., Lockshon, D., and Lassar, A. (1990). MyoD binds cooperatively to two sites in a target enhancer sequence: occupancy of two sites is required for activation. Proc Natl Acad Sci U S A 87, 5623-7.
Weintraub, H., Tapscott, S. J., Davis, R. L., Thayer, M. J., Adam, M. A., Lassar, A. B., and Miller, A. D. (1989). Activation of muscle-specific genes in pigment, nerve, fat, liver, and fibroblast cell lines by forced expression of MyoD. Proceedings Of The National Academy Of Sciences Of The United States Of America 86, 5434-8.
Williams, T. M., Moolten, D., Burlein, J., Romano, J., Bhaerrnan, R., Godillot, A., Mellon, M., Rauscher, F. J. d., and Kant, J. A. (1991). Identification of a zinc fmger protein that inhibits IL-2 gene expression. Science 254, 1791-4.
Wingender, E., Chen, X., Hehl, R., Karas, H., Liebech, I., Matys, V., Meinhardt, T., Prub, M.. Reuter, L, and Schacherer, F. (2000). TRANSFEC: an integrated system for gene expression regulation. Nuc. Acids Res. 28, 3 16-3 19.
WoHe, A. P. (1996). Histone deacetylase: a regulator of transcription. Science Wright, W. E.. Binder, M.. and Funk. W. (1991). Cyclic arnplfication and selection of targets (CASTing) for the myogenin consensus binding site. Mol Cell Biol 11, 4104-10.
Wright. W. E., Sassoon, D. A.. and Lin, V. K. (1989). Myogenin. a factor regulating myogenesis. has a domain homologous to MyoD. Cell 56, 607-17.
Yablonka-Reuveni, Z., Rudnicki, M. A., Rivera. A. J.. Primig. M., Anderson, J. E.. and Natanson, P. (1999). The transition from proliferation to differentiation is delayed in satellite cells from mice lacking MyoD. Developmental Biology 210. 440-55.
Yablonka-Reuveni. 2.. Seger, R., and Rivera, A. J. (1999). Fibroblast growth factor promotes recruitment of skeletal muscle satellite cells in young and old rats. Journal Of Histochemistry And Cytochernistry 47, 23-42.
Yaffe. D.. and Saxel, 0. (1977). Serial passaging and differentiation of myogenic cells isolated from dystrophic mouse muscle. Nature 270, 725-7.
Yasuda, H., Mizuno, A., Tamaoki, T., and Morinaga, T. (1994). ATBFI, a mu1 tiple-homeodomain zinc finger protein, selectively down-regulates AT-rich elements of the human alpha-fetoprotein gene. Molecular And Cellular Biology I 4. 1395-40 1.
Yoshida. N., Yoshida, S., Koishi, K.. Masuda. K., and Nabeshima, Y. (1998). Cell heterogeneity upon myogenic differentiation: down-regulation of MyoD and Myf-5 generates 'reserve cells'. Journal Of Cell Science 11 1 (Pt 6), 769-79.
Yoshida, S., Fujisawa-Sehara, A., Taki, T.. Arai, K.. and Nabeshima. Y. (1996). Lysophosphatidic acid and bFGF control different modes in proliferating myoblasts. Journal Of Cell Biology 132, 18 1-93.
Yun, K., and Wold, B. (1996). Skeletal muscle determination and differentiation: story of a core regulatory network and its context. Current Opinion In Cell Biology 8, 877-89.
Zentella, A., and Massague, J. (1992). Transforming growth factor beta induces myoblast differentiation in the presence of mitogens. Proc Natl Acad Sci U S A 89, 5 176-80. Zhang, J. M.. Wei. Q.. Zhao. X.. and Paterson. B. M. (1999a). Coupling of the cell cycle and myogenesis through the cyclin D 1- dependent interaction of MyoD with cdk4. Embo J 18, 926-33,
Zhang, J. M.. Zhao. X., Wei, Q., and Paterson, B. M. (1999b). Direct inhibition of G(l) cdk kinase activiv by MyoD promotes myoblast cell cycle withdrawal and terminal differentiation. Embo J 18. 6983-93.
Zhang. P., Wong, C., Liu, D.. Finegold, M., Harper. J. W.. and EUedge. S. J. (1999). p2 1(CIP I) and p57(Kll?2) control muscle differentiation at the myogenin step. Genes Dev 13, 2 13-24.
Zhang, W., Behringer, R. R., and Olson. E. N. (1995). Inactivation of the myogenic bHLH gene MRF4 results in up-regulation of myogenin and rib anomalies. Genes And Development 9, 1388-99.
Zweigerdt, R., Braun, T.. and hold,H. H. (1997). Faithful expression of the Myf-5 gene during mouse myogenesis requires distant control regions: a transgene approach using yeast dicial chromosomes. Developmental Biology 192, 172-80.