- Email: [email protected]
Muscle determination: Another key player in myogenesis?
R620
Dispatch
Muscle determination: Another key player in myogenesis? Anne-Gaelle Borycki and Charles P. Emerson
The steps that commit multipotential somite cells to muscle differentiation are being elucidated. Recent results show that pax3 is an upstream regulator of myoD, one of the key genes in myogenic lineage determination. Address: Department of Cell and Developmental Biology, University of Pennsylvania School of Medicine, 245 Anatomy-Chemistry Building, Philadelphia, Pennsylvania 19104-6058, USA. E-mail: [email protected] Current Biology 1997, 7:R620–R623 http://biomednet.com/elecref/09609822007R0620 © Current Biology Ltd ISSN 0960-9822
During embryonic development, cells are produced with a progressively more restricted lineage potential. The genetic mechanisms that control this process of lineage ‘determination’ are poorly understood, but the discovery of the myoD gene family opened the door to studies of muscle cell determination in vertebrate embryos. The first clue that the myoD genes, which all encode basic helix–loop–helix proteins, are important in myogenic determination came with the demonstration that they can stably convert non-muscle cells to skeletal muscle cells in culture [1]. Subsequent studies have confirmed this conclusion and shown that two members of the myoD family, myoD itself and myf5, have early functions in the determination process that commits multipotential somite cells to the myogenic lineage [2]. The two other myoD genes, myogenin and MRF4, have later functions in the initiation of myoblast differentiation [3].
An initially puzzling observation was that, in mouse embryos, myf5 and myoD are activated in somites with different temporal and spatial patterns (Figure 1). The first to be activated is myf5 — in day 8.0 embryos, signals from the neighboring neural tube–notochord complex induce myf5 expression in the dorsal-medial somites, in cells that are progenitors of the trunk and intercostal (epaxial) muscles [4]. Much later, in day 10.0 embryos, signals from the overlying surface ectoderm induce myoD expression in the dorsal-lateral somites, in cells that are progenitors of body wall and limb (hypaxial) muscles (Figure 1) [4]. After myf5 and myoD have both been activated, however, most, if not all, myotomal cells coexpress both genes [4]. This may account for the apparent redundancy of myf5 and myoD in the determination of the epaxial and hypaxial myogenic lineages [2]. Remarkably, double mutant ‘knockout’ mice, deficient in both myoD and myf5, lack all myogenic progenitor cells [2]. In either single mutant, however, muscle differentiation is not perturbed at a gross level, although in myf5 mutant embryos the initial differentiation of epaxial myotomal muscle is delayed until myoD is activated and compensates for the loss of myf5 [5,6]. Not only are myf5 and myoD essential for muscle formation, but their expression appears to be the key determinative step that commits multipotential somite cells to the myogenic lineage. This conclusion is based on recent lineage-tracing experiments using a lacZ reporter gene cloned into the myf5 locus. In homozygous mutant myf5 embryos, lacZ-expressing somite cells adopt other somite
Figure 1 (a) 8.0
9.0
10.0
11.0
12.0
13.0
14.0
pax7 pax3 myf5 myoD
© 1997 Current Biology
Expression of pax3, pax7, myf5 and myoD during mouse embryogenesis. (a) Colored boxes show the timing of expression of pax7 (purple) [22], pax3 (red) [11], myf5 (yellow) [9,14,23,24] and myoD (green) [25] in the neural tube and in somites. Dashed boxes indicate either that expression level is reducing (pax3) or has not been examined further (myf5).
15.0
16.0
(b) Epaxial muscles
Time (day) Neural tube Somite
DM
Neural tube Somite Neural tube Somite Neural tube Somite
SC NT
Hypaxial muscles
NC Dorsal Medial
Lateral Ventral
Significantly, pax7, pax3 and myf5 are transiently expressed, whereas myoD is expressed throughout development until birth. (b) Expression domains of pax3 (red), myf5 (yellow) and myoD (green) in the neural tube and somites of a day 10.5 mouse embryo. NT, neural tube; NC, notochord; SC, sclerotome; DM, dermomyotome.
Dispatch
lineage fates, expressing sclerotome and dermis genes, whereas in heterozygotes, all lacZ-expressing cells are committed to the myogenic lineage [7]. Taking all the evidence together, it is now clear that myoD and myf5 are redundant genes whose activation — at least in the case of myf5 — is the determinative step that commits multipotential somite cells to the myogenic lineage. If myoD and myf5 are the critical determination genes, then what are the upstream regulatory mechanisms responsible for their activation in somitic mesoderm? Two recent papers by Maroto et al. [8] and Tajbakhsh et al. [9] begin to address this important question by providing compelling molecular and genetic evidence that pax3 has dominant myoD-inducing activity in cultured chick cells, and is an essential upstream gene for myoD expression in the mouse embryo. These findings also raise intriguing questions about the specific functions of pax3 in myogenic determination. A role for pax3 in body muscle determination
The pax3 gene encodes a member of the subfamily of homeodomain transcription factors that is defined by the presence of a ‘paired’ motif [10]. The first sign of pax3 expression is in the neural tube at 8 days of gestation. By day 8.5, pax3 is expressed in presomitic mesoderm and in newly formed somites, and its expression becomes progressively restricted to the dorsal dermomyotome, and finally to the ventral-lateral part of the dermomyotome in the population of cells that will migrate to form limb muscles (Figure 1) [11]. Several spontaneous or X-ray induced pax3 mutant mice, known as splotch mutants, have been identified. splotch embryos die before day 15; they display neural defects — spina bifida, exencephaly and neural crest defects — and lack limb muscles as a consequence of impaired migration of the pax3-expressing somite cells in the lateral dermomyotome [10,12]. splotch mice also display altered muscle expression of myoD, myf5 and myogenin, in both the dorsal-medial and ventral-lateral domains of the dermomyotome, and lack some of the deep back and ventral body muscles, further implicating pax3 in at least some aspects of myogenesis [9]. The myogenic function of pax3 was most dramatically revealed when Tajbakhsh et al. [9] generated splotch/myf5 double mutant embryos. The myf5 deficiency of these mice forces myogenic determination to occur through the myoD lineage [6]. Remarkably, Tajbakhsh et al. [9] discovered that the double mutant embryos fail to express myoD in the somitic mesoderm and completely lack all body muscles. Despite this severe body muscle phenotype, head muscles form normally in splotch/myf5 double mutants, indicating that the head muscles are determined by a pax3-independent mechanism.
R621
The absence of body muscles in the splotch/myf5 double mutants resembles the phenotype of myoD/myf5 double mutants, consistent with the conclusion that pax3 is an upstream regulator of myoD. Indeed, myoD expression is nearly absent in explant co-cultures of somitic mesoderm and surface ectoderm isolated from splotch mice [9]. These findings thus provide unexpected, compelling evidence that pax3 is an essential upstream regulator in the myoDdependent body muscle lineage, but not in the head muscle lineage. pax3 is a dominant myogenic regulatory gene
Further evidence that pax3 is essential for myoD regulation comes from the studies of Maroto et al. [8], who expressed pax3 ectopically in cultured chick embryo tissues using a retrovirus expression vector. They found that, in presomitic mesoderm explants, pax3 induces myoD expression and maintains myf5 expression. This fits nicely with the related finding that myoD is activated in somitic mesoderm explants that are in contact with the surface ectoderm, which is known to provide signals for pax3 expression [13]. The more significant and surprising finding, however, is that 90% of lateral plate mesoderm cells and neural tube cells infected with a pax3-expressing retrovirus also express myoD, whereas only 8% of primary dermal fibroblasts, and no 10T½ fibroblasts activate myoD in response to pax3 expression. Curiously, one common feature of the three tissues affected by pax3 expression — paraxial mesoderm, neural tube and lateral plate — is that they all express myf5 at the time of explantation, indicating that they contain cells with a myogenic potential that is realized when pax3 expression is provided [8,14]. The results of these chick tissue culture experiments thus strikingly complement the genetic studies of pax3/myf5 mutant mice, by showing that pax3 has an essential function in the activation of myoD — and maintenance of myf5 expression — in the somitic mesoderm, leading to myogenic determination and myogenesis. What is the myogenic function of pax3?
Does pax3 act directly on myogenic determination by controlling the transcription of myoD (Figure 2a)? Or does it instead act indirectly, via genes that are essential for somitic mesoderm functions (Figure 2b)? The most compelling evidence that pax3 has a direct role in myogenic determination is the finding that ectopic pax3 expression converts non-muscle neural tube and lateral mesoderm in tissue culture to muscle [8]. Furthermore, a pax3-expressing retrovirus can replace both surface ectoderm and neural tube/notochord signals for myoD activation in somitic mesoderm explants [8], showing that pax3 acts downstream of these myoD-inducing signals. A significant caveat, however, is that this dominant myogenic activity has only been demonstrated in tissue culture, which is a notoriously permissive environment for skeletal myogenesis [15]. It will
R622
Current Biology, Vol 7 No 10
Figure 2 Body muscles (a) Direct model
Signal A (neural tube/ notochord)
(b) Indirect model
Signal B (surface ectoderm)
Signal B (surface ectoderm)
Signal A (neural tube/ notochord) myf5
pax3
pax3
Signal C (surface ectoderm)
Essential dorsal mesoderm function
? myoD
myf5
myoD ?
Epaxial muscles
Hypaxial muscles
Epaxial muscles
Hypaxial muscles
Head muscles (c)
myf5 Signal X pax3-independent pathway
© 1997 Current Biology
?
Head muscles
Models for genetic pathways of muscle determination and pax3 function. (a) The direct pathway model for body muscle formation hypothesizes that Pax3 activates myoD either directly or via an intermediary transcription factor. The notochord/neural tube complex provides signal A responsible for myf5 activation in somites, and the surface ectoderm provides signal B, responsible for pax3 activation [5,8]. myf5 expression in the dorsal medial domain of the dermomyotome defines the progenitor cells of epaxial muscles, and myoD expression in the ventral lateral domain of the dermomyotome defines the progenitor cells of hypaxial muscles. Signal A comprises Sonic hedgehog and a Wnt; signal B is unknown. (b) The indirect model for body muscle determination assumes that Pax3 is an essential transcriptional activator for dorsal mesoderm function, controlling mitotic activity, cell survival, or cell migration [12,15,26]. In this model, pax3 expression confers competence to respond to signals A and C, which induce myoD and myf5 expression. It remains to be determined whether signals B and C are distinct or identical factors. (c) A pax3independent pathway under the control of signal X activates myf5 and myoD, allowing formation of head muscles.
myoD
thus be important to test the myogenic activity of pax3 in vivo, by determining whether ectopic pax3 expression in chick or mouse embryos converts non-myogenic cells to the myogenic lineage. Ectopic pax3 expression has already been shown to block ventral neural tube differentiation [16]; it will be of interest to see whether presumptive neural tube misexpressing pax3 adopts a myogenic fate in these transgenic animals. Another reason to believe that pax3 regulates myogenic determination is that, in presomitic mesoderm explants from splotch mouse embryos, myoD is only poorly activated in the presence of either axial or surface ectoderm signals [9]. Furthermore, both myf5 and myoD expression, and the formation of some myf5-dependent muscles, are impaired in splotch mutant embryos [9]. It is unlikely, however, that Pax3 protein binds directly to myoD transcriptional control elements, as the human myoD core enhancer, which is activated in all epaxial and hypaxial muscles and head muscles [17], lacks Pax3-binding sites and is not transactivated by pax3 expression in 10T½ fibroblast cells (D. Goldhamer, personal communication). So while pax3 may be on the direct pathway for myoD and myf5 regulation, it would have to act on these genes via an intermediary regulator.
Alternatively, pax3 may have an indirect role, regulating genes involved in early mesoderm and neural tube functions (Figure 2b). Supporting this model, pax3 is expressed widely in the dermomyotome and neural tube, which do not adopt the myogenic fate [11], and splotch mice have strong neural tube defects [10]. Furthermore, myoD activation in somitic mesoderm explants from splotch mice is deficient in response to neural tube/notochord signals, which regulate myoD indirectly through myf5, as well as to surface ectoderm signals, which regulate myoD directly [9]. These observations indicate that pax3 is required for more general mesoderm functions. Possible mesoderm functions of pax3 include preserving dermomyotome integrity, as at early stages in splotch embryos the epithelial architecture of the dermomyotome is disrupted, particularly in the lateral region where myoD is normally activated [18]. Furthermore, splotch mice lack mesoderm (our unpublished observation), indicating additional defects in the proliferation and/or survival of somitic mesoderm in the absence of pax3 activity. That pax3 has a proliferative function is also suggested by the tumorigenic activity in rhabdomyosarcomas of fusion proteins encoded by pax3 joined to the forkhead gene [19]. The mesoderm phenotype of splotch embryos may be due to a defect in
Dispatch
cell migration, as lateral dermomyotome cells fail to migrate to the limb bud in splotch mutant embryos [12]. Indeed, Pax3 can transcriptionally activate c-met, the receptor for scatter factor, suggesting that the limb bud, as well as the body muscle deficiency phenotype seen in splotch mice may be attributed to the failure of splotch mesoderm to express c-met [20,21]. If this is so, then a strong prediction is that c-met/myf5 and scatter factor/myf5 double mutant embryos will have similar body muscle defects to those of splotch/myf5 embryos. Defects in cell migration could also explain the finding that somitic mesoderm from splotch mice can differentiate into muscle when transplanted into the limb bud of a chick embryo, showing that pax3 is not essential for myogenesis, but is required for maintaining more general functions of mesoderm in the somite environment where body muscles form [12]. This is consistent with the finding that head muscles form independently of pax3 (Figure 2c). The available evidence thus favors the conclusion that pax3 is not directly on the pathway for myogenic determination, but rather regulates essential, general somatic mesoderm functions. References 1. Weintraub H, Davis R, Tapscott S, Thayer M, Krause M, Benezra R, Blackwell TK, Turner D, Rupp R, Hollenberg S, et al.: The myoD gene family: nodal point during specification of the muscle cell lineage. Science 1991, 251:761-766. 2. Rudnicki MA, Schnegelsberg PN, Stead RH, Braun T, Arnold HH, Jaenisch R: MyoD or Myf-5 is required for the formation of skeletal muscle. Cell 1993, 75:1351-1359. 3. Edmondson DG, Olson EN: Helix-loop-helix proteins as regulators of muscle-specific transcription. J Biol Chem 1993, 268:755-758. 4. Cossu G, Kelly R, Tajbakhsh S, Di Donna S, Vivarelli E, Buckingham M: Activation of different myogenic pathways: myf-5 is induced by the neural tube and MyoD by the dorsal ectoderm in mouse paraxial mesoderm. Development 1996, 122:429-437. 5. Rudnicki MA, Braun T, Hinuma S, Jaenisch R: Inactivation of MyoD in mice leads to up-regulation of the myogenic HLH gene Myf-5 and results in apparently normal muscle development. Cell 1992, 71:383-390. 6. Braun T, Bober E, Rudnicki MA, Jaenisch R, Arnold HH: MyoD expression marks the onset of skeletal myogenesis in Myf-5 mutant mice. Development 1994, 120:3083-3092. 7. Tajbakhsh S, Rocancourt D, Buckingham M:. Muscle progenitor cells failing to respond to positional cues adopt non-myogenic fates in myf5 null mice. Nature 1996, 384:266-270. 8. Maroto M, Reshef R, Munsterberg AE, Koester S, Goulding M, Lassar AB: Ectopic pax-3 activates myoD and myf-5 expression in embryonic mesoderm and neural tissue. Cell 1997, 89:139-148. 9. Tajbakhsh S, Rocancourt D, Cossu G, Buckingham M: Redefining the genetic hierarchies controlling skeletal myogenesis: pax-3 and myf-5 act upstream of myoD. Cell 1997, 89:127-138. 10. Strachan T, Read AP: PAX genes. Curr Opin Gen Dev 1994, 4:427-438. 11. Goulding MD, Chalepakis G, Deutsch U, Erselius JR, Gruss P: Pax-3, a novel murine DNA binding protein expressed during early neurogenesis. EMBO J 1991, 10:1135-1147. 12. Daston G, Lamar E, Olivier M, Goulding M: Pax3 is necessary for migration but not differentiation of limb muscle precursor in the mouse. Development 1996, 122:1017-1027. 13. Fan CM, Tessier-Lavigne M: Patterning of mammalian somites by surface ectoderm and notochord: evidence for sclerotome induction by a hedgehog homolog. Cell 1994, 79:1175-1186. 14. Tajbakhsh S, Vivarelli E, Cusella-De Angelis G, Rocancourt D, Buckingham M, Cossu G: A population of myogenic cells derived from the mouse neural tube. Neuron 1994, 13:813-821.
R623
15. George-Weinstein M, Gerhart J, Reed R, Flynn J, Callihan B, Mattiacci M, Miehle C, Foti G, Lash JW, Weintraub H: Skeletal myogenesis: the preferred pathway of chick embryo epiblast cells in vitro. Dev Biol 1996, 173:279-291. 16. Tremblay P, Pituello F, Gruss P: Inhibition of floor plate differentiation by Pax3: evidence from ectopic expression in transgenic mice. Development 1996, 122:2555-2567. 17. Goldhamer DJ, Brunk BP, Faerman A, King A, Shani M, Emerson CP Jr: Embryonic activation of the myoD gene is regulated by a highly conserved distal control element. Development 1995, 121:637-649. 18. Franz T, Kothary R, Surani MA, Halata Z, Grim M: The Splotch mutation interferes with muscle development in the limbs. Anat Embryol 1993, 187:153-160. 19. Galili N, Davis RJ, Fredericks WJ, Mukhopadhyay S, Rauscher FJD, Emanuel BS, Rovera G, Barr FG: Fusion of a forkhead domain gene to pax3 in the solid tumor alveolar rhabdomyosarcoma. Nature Genet 1993, 5:230-235. 20. Bladt F, Riethmacher D, Isenmann S, Aguzzi A, Birchmeier C: Essential role for the c-met receptor in the migration of myogenic precursor cells into the limb bud. Nature 1995, 376:768-771. 21. Epstein JA, Shapiro DN, Cheng J, Lam PY, Maas RL: Pax3 modulates expression of the c-Met receptor during limb muscle development. Proc Natl Acad Sci USA 1996, 93:4213-4218. 22. Jostes B, Walther C, Gruss P: The murine paired box gene, Pax7, is expressed specifically during the development of the nervous and muscular system. Mech Dev 1991, 33:27-38. 23. Ott MO, Bober E, Lyons G, Arnold H, Buckingham M: Early expression of the myogenic regulatory gene, myf-5, in precursor cells of skeletal muscle in the mouse embryo. Development 1991, 111:1097-1107. 24. Tajbakhsh S, Buckingham ME: Lineage restriction of the myogenic conversion factor myf-5 in the brain. Development 1995, 121:4077-4083. 25. Sassoon D, Lyons G, Wright WE, Lin V, Lassar A, Weintraub H, Buckingham M: Expression of two myogenic regulatory factors myogenin and MyoD1 during mouse embryogenesis. Nature 1989, 341:303-307. 26. Bernasconi M, Remppis A, Fredericks WJ, Rauscher F Jr, Schafer BW: Induction of apoptosis in rhabdomyosarcoma cells through down-regulation of PAX proteins. Proc Natl Acad Sci USA 1996, 93:13164-13169.
Dispatch
Muscle determination: Another key player in myogenesis? Anne-Gaelle Borycki and Charles P. Emerson
The steps that commit multipotential somite cells to muscle differentiation are being elucidated. Recent results show that pax3 is an upstream regulator of myoD, one of the key genes in myogenic lineage determination. Address: Department of Cell and Developmental Biology, University of Pennsylvania School of Medicine, 245 Anatomy-Chemistry Building, Philadelphia, Pennsylvania 19104-6058, USA. E-mail: [email protected] Current Biology 1997, 7:R620–R623 http://biomednet.com/elecref/09609822007R0620 © Current Biology Ltd ISSN 0960-9822
During embryonic development, cells are produced with a progressively more restricted lineage potential. The genetic mechanisms that control this process of lineage ‘determination’ are poorly understood, but the discovery of the myoD gene family opened the door to studies of muscle cell determination in vertebrate embryos. The first clue that the myoD genes, which all encode basic helix–loop–helix proteins, are important in myogenic determination came with the demonstration that they can stably convert non-muscle cells to skeletal muscle cells in culture [1]. Subsequent studies have confirmed this conclusion and shown that two members of the myoD family, myoD itself and myf5, have early functions in the determination process that commits multipotential somite cells to the myogenic lineage [2]. The two other myoD genes, myogenin and MRF4, have later functions in the initiation of myoblast differentiation [3].
An initially puzzling observation was that, in mouse embryos, myf5 and myoD are activated in somites with different temporal and spatial patterns (Figure 1). The first to be activated is myf5 — in day 8.0 embryos, signals from the neighboring neural tube–notochord complex induce myf5 expression in the dorsal-medial somites, in cells that are progenitors of the trunk and intercostal (epaxial) muscles [4]. Much later, in day 10.0 embryos, signals from the overlying surface ectoderm induce myoD expression in the dorsal-lateral somites, in cells that are progenitors of body wall and limb (hypaxial) muscles (Figure 1) [4]. After myf5 and myoD have both been activated, however, most, if not all, myotomal cells coexpress both genes [4]. This may account for the apparent redundancy of myf5 and myoD in the determination of the epaxial and hypaxial myogenic lineages [2]. Remarkably, double mutant ‘knockout’ mice, deficient in both myoD and myf5, lack all myogenic progenitor cells [2]. In either single mutant, however, muscle differentiation is not perturbed at a gross level, although in myf5 mutant embryos the initial differentiation of epaxial myotomal muscle is delayed until myoD is activated and compensates for the loss of myf5 [5,6]. Not only are myf5 and myoD essential for muscle formation, but their expression appears to be the key determinative step that commits multipotential somite cells to the myogenic lineage. This conclusion is based on recent lineage-tracing experiments using a lacZ reporter gene cloned into the myf5 locus. In homozygous mutant myf5 embryos, lacZ-expressing somite cells adopt other somite
Figure 1 (a) 8.0
9.0
10.0
11.0
12.0
13.0
14.0
pax7 pax3 myf5 myoD
© 1997 Current Biology
Expression of pax3, pax7, myf5 and myoD during mouse embryogenesis. (a) Colored boxes show the timing of expression of pax7 (purple) [22], pax3 (red) [11], myf5 (yellow) [9,14,23,24] and myoD (green) [25] in the neural tube and in somites. Dashed boxes indicate either that expression level is reducing (pax3) or has not been examined further (myf5).
15.0
16.0
(b) Epaxial muscles
Time (day) Neural tube Somite
DM
Neural tube Somite Neural tube Somite Neural tube Somite
SC NT
Hypaxial muscles
NC Dorsal Medial
Lateral Ventral
Significantly, pax7, pax3 and myf5 are transiently expressed, whereas myoD is expressed throughout development until birth. (b) Expression domains of pax3 (red), myf5 (yellow) and myoD (green) in the neural tube and somites of a day 10.5 mouse embryo. NT, neural tube; NC, notochord; SC, sclerotome; DM, dermomyotome.
Dispatch
lineage fates, expressing sclerotome and dermis genes, whereas in heterozygotes, all lacZ-expressing cells are committed to the myogenic lineage [7]. Taking all the evidence together, it is now clear that myoD and myf5 are redundant genes whose activation — at least in the case of myf5 — is the determinative step that commits multipotential somite cells to the myogenic lineage. If myoD and myf5 are the critical determination genes, then what are the upstream regulatory mechanisms responsible for their activation in somitic mesoderm? Two recent papers by Maroto et al. [8] and Tajbakhsh et al. [9] begin to address this important question by providing compelling molecular and genetic evidence that pax3 has dominant myoD-inducing activity in cultured chick cells, and is an essential upstream gene for myoD expression in the mouse embryo. These findings also raise intriguing questions about the specific functions of pax3 in myogenic determination. A role for pax3 in body muscle determination
The pax3 gene encodes a member of the subfamily of homeodomain transcription factors that is defined by the presence of a ‘paired’ motif [10]. The first sign of pax3 expression is in the neural tube at 8 days of gestation. By day 8.5, pax3 is expressed in presomitic mesoderm and in newly formed somites, and its expression becomes progressively restricted to the dorsal dermomyotome, and finally to the ventral-lateral part of the dermomyotome in the population of cells that will migrate to form limb muscles (Figure 1) [11]. Several spontaneous or X-ray induced pax3 mutant mice, known as splotch mutants, have been identified. splotch embryos die before day 15; they display neural defects — spina bifida, exencephaly and neural crest defects — and lack limb muscles as a consequence of impaired migration of the pax3-expressing somite cells in the lateral dermomyotome [10,12]. splotch mice also display altered muscle expression of myoD, myf5 and myogenin, in both the dorsal-medial and ventral-lateral domains of the dermomyotome, and lack some of the deep back and ventral body muscles, further implicating pax3 in at least some aspects of myogenesis [9]. The myogenic function of pax3 was most dramatically revealed when Tajbakhsh et al. [9] generated splotch/myf5 double mutant embryos. The myf5 deficiency of these mice forces myogenic determination to occur through the myoD lineage [6]. Remarkably, Tajbakhsh et al. [9] discovered that the double mutant embryos fail to express myoD in the somitic mesoderm and completely lack all body muscles. Despite this severe body muscle phenotype, head muscles form normally in splotch/myf5 double mutants, indicating that the head muscles are determined by a pax3-independent mechanism.
R621
The absence of body muscles in the splotch/myf5 double mutants resembles the phenotype of myoD/myf5 double mutants, consistent with the conclusion that pax3 is an upstream regulator of myoD. Indeed, myoD expression is nearly absent in explant co-cultures of somitic mesoderm and surface ectoderm isolated from splotch mice [9]. These findings thus provide unexpected, compelling evidence that pax3 is an essential upstream regulator in the myoDdependent body muscle lineage, but not in the head muscle lineage. pax3 is a dominant myogenic regulatory gene
Further evidence that pax3 is essential for myoD regulation comes from the studies of Maroto et al. [8], who expressed pax3 ectopically in cultured chick embryo tissues using a retrovirus expression vector. They found that, in presomitic mesoderm explants, pax3 induces myoD expression and maintains myf5 expression. This fits nicely with the related finding that myoD is activated in somitic mesoderm explants that are in contact with the surface ectoderm, which is known to provide signals for pax3 expression [13]. The more significant and surprising finding, however, is that 90% of lateral plate mesoderm cells and neural tube cells infected with a pax3-expressing retrovirus also express myoD, whereas only 8% of primary dermal fibroblasts, and no 10T½ fibroblasts activate myoD in response to pax3 expression. Curiously, one common feature of the three tissues affected by pax3 expression — paraxial mesoderm, neural tube and lateral plate — is that they all express myf5 at the time of explantation, indicating that they contain cells with a myogenic potential that is realized when pax3 expression is provided [8,14]. The results of these chick tissue culture experiments thus strikingly complement the genetic studies of pax3/myf5 mutant mice, by showing that pax3 has an essential function in the activation of myoD — and maintenance of myf5 expression — in the somitic mesoderm, leading to myogenic determination and myogenesis. What is the myogenic function of pax3?
Does pax3 act directly on myogenic determination by controlling the transcription of myoD (Figure 2a)? Or does it instead act indirectly, via genes that are essential for somitic mesoderm functions (Figure 2b)? The most compelling evidence that pax3 has a direct role in myogenic determination is the finding that ectopic pax3 expression converts non-muscle neural tube and lateral mesoderm in tissue culture to muscle [8]. Furthermore, a pax3-expressing retrovirus can replace both surface ectoderm and neural tube/notochord signals for myoD activation in somitic mesoderm explants [8], showing that pax3 acts downstream of these myoD-inducing signals. A significant caveat, however, is that this dominant myogenic activity has only been demonstrated in tissue culture, which is a notoriously permissive environment for skeletal myogenesis [15]. It will
R622
Current Biology, Vol 7 No 10
Figure 2 Body muscles (a) Direct model
Signal A (neural tube/ notochord)
(b) Indirect model
Signal B (surface ectoderm)
Signal B (surface ectoderm)
Signal A (neural tube/ notochord) myf5
pax3
pax3
Signal C (surface ectoderm)
Essential dorsal mesoderm function
? myoD
myf5
myoD ?
Epaxial muscles
Hypaxial muscles
Epaxial muscles
Hypaxial muscles
Head muscles (c)
myf5 Signal X pax3-independent pathway
© 1997 Current Biology
?
Head muscles
Models for genetic pathways of muscle determination and pax3 function. (a) The direct pathway model for body muscle formation hypothesizes that Pax3 activates myoD either directly or via an intermediary transcription factor. The notochord/neural tube complex provides signal A responsible for myf5 activation in somites, and the surface ectoderm provides signal B, responsible for pax3 activation [5,8]. myf5 expression in the dorsal medial domain of the dermomyotome defines the progenitor cells of epaxial muscles, and myoD expression in the ventral lateral domain of the dermomyotome defines the progenitor cells of hypaxial muscles. Signal A comprises Sonic hedgehog and a Wnt; signal B is unknown. (b) The indirect model for body muscle determination assumes that Pax3 is an essential transcriptional activator for dorsal mesoderm function, controlling mitotic activity, cell survival, or cell migration [12,15,26]. In this model, pax3 expression confers competence to respond to signals A and C, which induce myoD and myf5 expression. It remains to be determined whether signals B and C are distinct or identical factors. (c) A pax3independent pathway under the control of signal X activates myf5 and myoD, allowing formation of head muscles.
myoD
thus be important to test the myogenic activity of pax3 in vivo, by determining whether ectopic pax3 expression in chick or mouse embryos converts non-myogenic cells to the myogenic lineage. Ectopic pax3 expression has already been shown to block ventral neural tube differentiation [16]; it will be of interest to see whether presumptive neural tube misexpressing pax3 adopts a myogenic fate in these transgenic animals. Another reason to believe that pax3 regulates myogenic determination is that, in presomitic mesoderm explants from splotch mouse embryos, myoD is only poorly activated in the presence of either axial or surface ectoderm signals [9]. Furthermore, both myf5 and myoD expression, and the formation of some myf5-dependent muscles, are impaired in splotch mutant embryos [9]. It is unlikely, however, that Pax3 protein binds directly to myoD transcriptional control elements, as the human myoD core enhancer, which is activated in all epaxial and hypaxial muscles and head muscles [17], lacks Pax3-binding sites and is not transactivated by pax3 expression in 10T½ fibroblast cells (D. Goldhamer, personal communication). So while pax3 may be on the direct pathway for myoD and myf5 regulation, it would have to act on these genes via an intermediary regulator.
Alternatively, pax3 may have an indirect role, regulating genes involved in early mesoderm and neural tube functions (Figure 2b). Supporting this model, pax3 is expressed widely in the dermomyotome and neural tube, which do not adopt the myogenic fate [11], and splotch mice have strong neural tube defects [10]. Furthermore, myoD activation in somitic mesoderm explants from splotch mice is deficient in response to neural tube/notochord signals, which regulate myoD indirectly through myf5, as well as to surface ectoderm signals, which regulate myoD directly [9]. These observations indicate that pax3 is required for more general mesoderm functions. Possible mesoderm functions of pax3 include preserving dermomyotome integrity, as at early stages in splotch embryos the epithelial architecture of the dermomyotome is disrupted, particularly in the lateral region where myoD is normally activated [18]. Furthermore, splotch mice lack mesoderm (our unpublished observation), indicating additional defects in the proliferation and/or survival of somitic mesoderm in the absence of pax3 activity. That pax3 has a proliferative function is also suggested by the tumorigenic activity in rhabdomyosarcomas of fusion proteins encoded by pax3 joined to the forkhead gene [19]. The mesoderm phenotype of splotch embryos may be due to a defect in
Dispatch
cell migration, as lateral dermomyotome cells fail to migrate to the limb bud in splotch mutant embryos [12]. Indeed, Pax3 can transcriptionally activate c-met, the receptor for scatter factor, suggesting that the limb bud, as well as the body muscle deficiency phenotype seen in splotch mice may be attributed to the failure of splotch mesoderm to express c-met [20,21]. If this is so, then a strong prediction is that c-met/myf5 and scatter factor/myf5 double mutant embryos will have similar body muscle defects to those of splotch/myf5 embryos. Defects in cell migration could also explain the finding that somitic mesoderm from splotch mice can differentiate into muscle when transplanted into the limb bud of a chick embryo, showing that pax3 is not essential for myogenesis, but is required for maintaining more general functions of mesoderm in the somite environment where body muscles form [12]. This is consistent with the finding that head muscles form independently of pax3 (Figure 2c). The available evidence thus favors the conclusion that pax3 is not directly on the pathway for myogenic determination, but rather regulates essential, general somatic mesoderm functions. References 1. Weintraub H, Davis R, Tapscott S, Thayer M, Krause M, Benezra R, Blackwell TK, Turner D, Rupp R, Hollenberg S, et al.: The myoD gene family: nodal point during specification of the muscle cell lineage. Science 1991, 251:761-766. 2. Rudnicki MA, Schnegelsberg PN, Stead RH, Braun T, Arnold HH, Jaenisch R: MyoD or Myf-5 is required for the formation of skeletal muscle. Cell 1993, 75:1351-1359. 3. Edmondson DG, Olson EN: Helix-loop-helix proteins as regulators of muscle-specific transcription. J Biol Chem 1993, 268:755-758. 4. Cossu G, Kelly R, Tajbakhsh S, Di Donna S, Vivarelli E, Buckingham M: Activation of different myogenic pathways: myf-5 is induced by the neural tube and MyoD by the dorsal ectoderm in mouse paraxial mesoderm. Development 1996, 122:429-437. 5. Rudnicki MA, Braun T, Hinuma S, Jaenisch R: Inactivation of MyoD in mice leads to up-regulation of the myogenic HLH gene Myf-5 and results in apparently normal muscle development. Cell 1992, 71:383-390. 6. Braun T, Bober E, Rudnicki MA, Jaenisch R, Arnold HH: MyoD expression marks the onset of skeletal myogenesis in Myf-5 mutant mice. Development 1994, 120:3083-3092. 7. Tajbakhsh S, Rocancourt D, Buckingham M:. Muscle progenitor cells failing to respond to positional cues adopt non-myogenic fates in myf5 null mice. Nature 1996, 384:266-270. 8. Maroto M, Reshef R, Munsterberg AE, Koester S, Goulding M, Lassar AB: Ectopic pax-3 activates myoD and myf-5 expression in embryonic mesoderm and neural tissue. Cell 1997, 89:139-148. 9. Tajbakhsh S, Rocancourt D, Cossu G, Buckingham M: Redefining the genetic hierarchies controlling skeletal myogenesis: pax-3 and myf-5 act upstream of myoD. Cell 1997, 89:127-138. 10. Strachan T, Read AP: PAX genes. Curr Opin Gen Dev 1994, 4:427-438. 11. Goulding MD, Chalepakis G, Deutsch U, Erselius JR, Gruss P: Pax-3, a novel murine DNA binding protein expressed during early neurogenesis. EMBO J 1991, 10:1135-1147. 12. Daston G, Lamar E, Olivier M, Goulding M: Pax3 is necessary for migration but not differentiation of limb muscle precursor in the mouse. Development 1996, 122:1017-1027. 13. Fan CM, Tessier-Lavigne M: Patterning of mammalian somites by surface ectoderm and notochord: evidence for sclerotome induction by a hedgehog homolog. Cell 1994, 79:1175-1186. 14. Tajbakhsh S, Vivarelli E, Cusella-De Angelis G, Rocancourt D, Buckingham M, Cossu G: A population of myogenic cells derived from the mouse neural tube. Neuron 1994, 13:813-821.
R623
15. George-Weinstein M, Gerhart J, Reed R, Flynn J, Callihan B, Mattiacci M, Miehle C, Foti G, Lash JW, Weintraub H: Skeletal myogenesis: the preferred pathway of chick embryo epiblast cells in vitro. Dev Biol 1996, 173:279-291. 16. Tremblay P, Pituello F, Gruss P: Inhibition of floor plate differentiation by Pax3: evidence from ectopic expression in transgenic mice. Development 1996, 122:2555-2567. 17. Goldhamer DJ, Brunk BP, Faerman A, King A, Shani M, Emerson CP Jr: Embryonic activation of the myoD gene is regulated by a highly conserved distal control element. Development 1995, 121:637-649. 18. Franz T, Kothary R, Surani MA, Halata Z, Grim M: The Splotch mutation interferes with muscle development in the limbs. Anat Embryol 1993, 187:153-160. 19. Galili N, Davis RJ, Fredericks WJ, Mukhopadhyay S, Rauscher FJD, Emanuel BS, Rovera G, Barr FG: Fusion of a forkhead domain gene to pax3 in the solid tumor alveolar rhabdomyosarcoma. Nature Genet 1993, 5:230-235. 20. Bladt F, Riethmacher D, Isenmann S, Aguzzi A, Birchmeier C: Essential role for the c-met receptor in the migration of myogenic precursor cells into the limb bud. Nature 1995, 376:768-771. 21. Epstein JA, Shapiro DN, Cheng J, Lam PY, Maas RL: Pax3 modulates expression of the c-Met receptor during limb muscle development. Proc Natl Acad Sci USA 1996, 93:4213-4218. 22. Jostes B, Walther C, Gruss P: The murine paired box gene, Pax7, is expressed specifically during the development of the nervous and muscular system. Mech Dev 1991, 33:27-38. 23. Ott MO, Bober E, Lyons G, Arnold H, Buckingham M: Early expression of the myogenic regulatory gene, myf-5, in precursor cells of skeletal muscle in the mouse embryo. Development 1991, 111:1097-1107. 24. Tajbakhsh S, Buckingham ME: Lineage restriction of the myogenic conversion factor myf-5 in the brain. Development 1995, 121:4077-4083. 25. Sassoon D, Lyons G, Wright WE, Lin V, Lassar A, Weintraub H, Buckingham M: Expression of two myogenic regulatory factors myogenin and MyoD1 during mouse embryogenesis. Nature 1989, 341:303-307. 26. Bernasconi M, Remppis A, Fredericks WJ, Rauscher F Jr, Schafer BW: Induction of apoptosis in rhabdomyosarcoma cells through down-regulation of PAX proteins. Proc Natl Acad Sci USA 1996, 93:13164-13169.