Unit: Molecular Genetics of Development - CNRS URA 2578
Director: BUCKINGHAM Margaret
Our research focusses on the study of myogenesis, with the aim of understanding how muscle cells are specified in the embryo and how this leads to the formation of different skeletal muscles. These studies extend to muscle progenitor cells in the adult and the potential contribution of stem cells to skeletal muscle regeneration. We are also interested in cardiogenesis, with the analysis of the origin and behaviour of different populations of myocardial precursor cells and how they contribute to heart morphogenesis. The animal model is the mouse, using experimental approaches on gene manipulation.
Myf5 and Mrf4 genes, as well as Pax3 and its paralogue Pax7, act upstream in the genetic hierarchy that determines the entry of a cell into the myogenic differentiation programme. Myf5 and Mrf4 are members of the family of b-HLH myogenic regulatory factors (Myf5, Mrf4, MyoD and Myogenin), while Pax3 belongs to the Pax family of homeo-domain transcription factors, members of which play a key role in organogenesis and in the appearance of specialized cell types during development.
Pax genes and myogenic progenitor cells in the embryo
[F. Relaix, D. Rocancourt, in collaboration with A. Mansouri (Göttingen)]
The nlacZ and GFP reporters that we have introduced into Pax3 label the cells that express this gene during development. In the case of Pax7, the existence of reliable antibodies as well as a Pax7LacZ/+ mouse line provide tools for following its expression. We have observed the presence of Pax3+, Pax7+ cells, which do not express myogenic markers and which divide, within muscle masses. Such cells are initially present in the central region of the dermomyotome, the epithelial structure that gives rise to muscle cells, and then in all late embryonic and ftal muscles. The stability of the GFP reporter, which acts as a chase, shows that cells that have expressed Pax3, become myoblasts with activation of the myogenic determination genes, Myf5 and MyoD, and then differentiate into muscle fibres. In Pax3-/-/Pax7-/- double mutants, these cells do not enter the myogenic programme but die or become incorporated into other tissues. In the absence of these muscle progenitor cells there is a major deficit in skeletal muscle ; the only muscles present are those formed during the first wave of myogenesis which can occur in the absence of Pax3/7. This depends on Myf5 and also Mrf4, but all the subsequent formation and growth of skeletal muscles depends on the Pax3+, Pax7+ progenitor cell population.
The role of Pax3 and Pax7 in the satellite cells of postnatal muscle
[D. Montarras, F. Relaix, S. Zaffran, en collaboration avec A.Mansouri (Göttingen), S. Tajbakhsh et B. Gayraud-Morel (IP) et A. Cumano (IP)]
Postnatal muscle growth and regeneration depend on satellite cells. Like pre-natal muscle progenitor cells, which assume a satellite cell position under the basal lamina of the fibre just prior to birth, they express Pax7. In Pax7 mutant mice, satellite cells are initially present and therefore specified, but are progressively lost. Pax3, which is also expressed in the satellite cells of most muscles, as visualised by targeted reporter genes, does not save the situation. We show that satelllite cells die in the absence of Pax7. The anti-apoptotic action of Pax7 is illustrated by the death of cultured satellite cells infected with an adenovirus expressing Pax7-En, a dominant-negative form of Pax7. Pax3-En does not show the same effect. We conclude that Pax7 plays an anti-apoptotic role and that Pax3 does not perform this function in these adult muscle progenitor cells. In contrast, the manipulation of dominant negative forms of both Pax proteins shows that Pax3 and Pax7 play a role in the activation of MyoD, thus regulating the entry of a satellite cell into the myogenic programme. Myf5 also plays this role, independently of Pax3/7 ; and it is only in satellite cells from Myf5-/- mutant mice, in the presence of Pax3-En/Pax7-En, that myogenesis does not take place.
Isolation of satellite cells from Pax3GFP/+ mice
[D. Montarras, F. Relaix, in collaboration with A. Cumano (IP), J. Morgan and T. Partridge (London)]
In Pax3GFP/+ mice many satellite cells are GFP positive, permitting for the first time their purification. We have isolated these cells by flow cytometry and analysed their properties in vitro and in vivo. In culture, a clonal analysis shows that they form muscle cells (>95%). In vivo grafting experiments into mdx mice demonstrate the efficiency of their contribution to regenerating fibres, marked by the expression of dystrophin, absent in mdx mice. The reserve of satellite cells is also reconstituted by these GFP+ cells, which self renew. A comparison of different preparations of GFP+ cells shows a loss of efficiency with crude preparations and/or after cell culture. The isolation of fluorescent satellite cells from Pax3GFP/+ mice has led to the definition of parameters of cell size and granulosity which have been used to isolate satellite cells from wild type mice. These cells behave like satellite cells both in vitro and in vivo. We conclude that purified satellite cells have the properties of adult muscle progenitor cells, with considerable regenerative potential.
The regulation of Myf5 and Pax3 targets
[L. Bajard, C. Crist, P. Daubas, M. Lagha, F. Relaix, D. Rocancourt, S. Vincent]
The analysis of Pax3Pax3-En-IRESnlacZ/+ embryos, in which it is possible to examine the role of Pax3 in the hypaxial somite and its derivatives in the absence of pronounced cell death, as well as our analysis of Myf5 regulatory sequences, suggest that Pax3 acts directly upstream of Myf5 in this domain of myogenesis.
We are also pursuing the regulation of Myf5 in adult skeletal muscle and its unexpected transcription in cells of the central nervous system.
We are characterising Pax3 targets in muscle progenitor cells which migrate into the limb bud, as well as the co-factors which interact with Pax3.
Blimp1 and myogenesis
Blimp1 plays a definitive role in the emergence of the slow muscle cell lineage in the zebrafish embryo. Blimp1 is also expressed in the mouse myotome where we are styding its function.
The smooth muscle cells of the dorsal aorta derive from the same progenitor cells as the skeletal muscle of the myotome
[M. Esner, F. Relaix, in collaboration with J.-F. Nicolas (IP) and G. Cossu (Milan)]
A retrospective clonal analysis led to the demonstration that cells in the dorsal aorta and cells in the skeletal muscle of the myotome derive from the same progenitor cell. Different classes of clone reflect the developmental time when this event took place. In particular, clones present in the hypaxial dermomyotome of a single somite and in the dorsal aorta at the same axial level suggest a recent somitic origin (Figure 1A). This suggestion is reinforced by observations on Pax3GFP/+ embryos. Pax3 protein is only present in the dermomyotome and in myogenic progenitor cells that have left this structure. GFP, on the other hand, is also detected in cells located between the somite and the aorta as well as in all the smooth muscle cells of the aorta (Figure 1B,C). These are not neural crest cells and we conclude that the smooth muscle of the dorsal aorta is derived from a progenitor cell that expressed Pax3, consistent with a somitic/paraxial mesoderm origin. These results have implications for the source of mesoangioblasts, mesodermal stem cells present in the wall of the aorta with the potential to form smooth or skeletal muscle.
[F. Bajolle, D. Galli, E. Pecnard (S. Zaffran), Y. Watanabe, in collaboration with N. Brown (London), R. Kelly (Marseilles) and S. Meilhac (Cambridge)]
The second heart field
We have shown by retrospective clonal analysis that two cell lineages participate in the formation of the myocardium of the mouse embryo. Their contributions can be distinguished by the absence of cells from the second lineage in the left ventricle, while the outflow tract of the heart is formed exclusively from the second lineage. This correlates with the presence of two heart fields, identified by explant experiments and Di-I labelling. The second heart field is characterised by the expression of a network of genes, which when mutated give phenotypes expected of interference with the second lineage.
We are currently investigating the posterior extent of the second heart field and its contribution to the venous pole of the heart. This also involves looking at the acquisition of right/left identity of cells which contribute to the atria. Where the anterior part of the second heart field is concerned, we are examining Fgf signalling in this sub-region which is characterised by the expression of Fgf8 and Fgf10. Fgf10-/- mutant embryos do not have an early cardiac phenotype, probably due to the presence of Fgf8. Mutants for the receptor of Fgf10, Fgfr2-IIIb, do have a number of cardiac malformations including some hypoplasia of the right ventricle which may reflect a second heart field effect. We are constructing Fgf8-/-/Fgf10-/- double mutants, with a conditional Fgf8 allele targeted to the second heart field, with the expectation that we should see more pronounced effects on the outflow tract and right ventricle.
Morphogenesis of the outflow tract
[F. Bajolle, E. Pecnard, S. Zaffran, in collaboration with D. Bonnet (Necker Hospital) and R. Kelly (Marseilles)]
Two transgenic mouse lines which mark sub-regions of outflow tract myocardium have led us to examine the morphogenesis of this region of the heart. They mark myocardial cells that will contribute to the base of the aorta or pulmonary trunk respectively. These marked cells undergo a counterclock-wise rotation, confirmed by Di-I labelling. In embryos with mutations that affect the neural crest cell contribution (Pax3-/-) or left/right signalling (Pitx2-/-), we observe that rotation is arrested. In these mutants the outflow tract does not undergo correct morphogenesis and there are malformations of the outflow region. This observation suggests that myocardial rotation is influenced by both contributions and that it is probably important for the correct morphogenesis of this region of the heart.
Figure 1 : Cells in the dorsal aorta originate from somites. A. A section showing clonally related β-galactosidase positive cells in the skeletal muscle of the hypaxial myotome (HM) and the smooth muscle of the dorsal aorta (DA). Muscle is marked by an actin antibody (red). B. Pax3-GFP positive cells in the dermomyotome (DM) of the somite and in the intervening region between the somite and dorsal aorta (DA). C. A section from a Pax3GFP/GFP embryo showing that the smooth muscle cells of the dorsal aorta (DA) are GFP positive (green).
Keywords: Myogenesis, Pax3, Pax7, Myf5, Dordal Aorta, Cargiogenesis, Fgfs