Project Topics
Get Complete Project Material File(s) Now! »
Molecular regulation of muscle stem cell emergence
During early development, muscle stem/progenitor cells migrate underneath the dorsal part of the somites called the dermomyotome (DM) and differentiate into mononucleated myocytes to form the myotome. In response to key transcription factors, committed myocytes align and fuse to generate small multinucleated myofibres during primary myogenesis in the embryo (from E11-E14.5), then myofibres containing a few hundred myonuclei during secondary myogenesis (from E14.5-to birth). During the early and late perinatal period that lasts about 4 weeks, continued myoblast fusion, or hyperplasia, is followed by muscle hypertrophy (Sambasivan and Tajbakhsh, 2007; Tajbakhsh, 2009; White et al., 2010) (Figure 5).
The developmental origin of satellite cells was first shown in a chick-quail chimera study: satellite cells of quail origin were found after replacement of chick somitic mesoderm by one from quail. In addition, electroporation of the central dermomyotome (the dorsal somite) in the trunk with a molecular marker showed that marked cells gave rise to Pax7+ satellite cells after hatching, thereby establishing the dermomyotome origin of satellite cells, in chick (Armand et al., 1983; Gros et al., 2005). Further evidences that satellite cells also originate from Pax3/7+ cells coming from the somites have been reported in the mouse (Kassar-Duchossoy et al., 2005; Relaix et al., 2005). Emerging satellite cells are found underneath a basement membrane from about 2 days before birth in mice and they further proliferate until the mid-perinatal stage (Kassar-Duchossoy et al., 2005). The majority of quiescent MuSCs are established from about 2-4 weeks after birth (Tajbakhsh, 2009; White et al., 2010). During prenatal and postnatal myogenesis, stem cell self-renewal and commitment are governed by a gene regulatory network that includes the paired⁄homeodomain transcription factors Pax3 and Pax7, and basic helix-loop-helix (bHLH) myogenic regulatory factors (MRFs), Myf5, Mrf4, Myod and Myogenin (Figure 5). Pax3 plays a critical role in establishing MuSCs during embryonic development (except in cranial-derived muscles) and Pax7 during late foetal and perinatal growth. Indeed, Pax3:Pax7 double mutant mice exhibit severe hypoplasia due to a loss of stem and progenitor cells from mid embryonic stages, and these Pax genes appear to regulate apoptosis (Relaix et al., 2006; Relaix et al., 2005; Sambasivan et al., 2009). During perinatal growth, Pax7 null mice are deficient in the number of MuSCs and fail to regenerate muscle after injury in adult mice (Lepper et al., 2009; Oustanina et al., 2004; Seale et al., 2000; von Maltzahn et al., 2013).
Experiments using simple or double knockout mice have shown the temporal and functional roles of these different factors during myogenesis. Myf5, Mrf4 and Myod assign myogenic cell fate of muscle progenitor cells to give rise to myoblasts (Kassar-Duchossoy et al., 2004; Rudnicki et al., 1993; Tajbakhsh et al., 1996) whereas Myogenin plays a crucial role in myoblast differentiation prenatally (Hasty et al., 1993; Nabeshima et al., 1993) but not postnatally as the conditional mutation of Myogenin in the adult has a relatively mild phenotype (Knapp et al., 2006; Meadows et al., 2008; Venuti et al., 1995). In the adult, Myod deficient mice that survive have increased precursor cell numbers accompanied by a delay in regeneration (Megeney et al., 1996; White et al., 2000); whereas Myf5 null mice display a slight delay in repair (Gayraud-Morel et al., 2007). These studies suggested that Myf5, Mrf4 and Myod could in some cases compensate for each other’s function. Whereas Mrf4 plays a role in embryonic progenitors, Myf5 and Myod continue to regulate muscle progenitor cell fate throughout foetal and postnatal life. Interestingly, additional transcription factors have been shown to interact with MYOD to regulate myogenesis. For instance, ChiP-seq data demonstrated that KLF5 (Kruppel-like factor, member of a subfamily of zinc-finger transcription factors) (Hayashi et al., 2016) as well as RUNX1 (Umansky et al., 2015) binding to Myod-regulated enhancers is necessary to activate a set of myogenic differentiation genes.
Heterogeneity in the muscle stem cell population
Compelling evidence from several studies has demonstrated that the satellite cell population is heterogeneous regarding their gene set of expression, proliferation rate, differentiation potential, stemness and even survival.
One remarkable example is demonstrated by the heterogeneity in satellite cells derived from skeletal muscle arising from different developmental origins: head (non-segmented paraxial mesoderm) versus limb (somites) that showed distinct molecular signatures. Cranial mesoderm derived muscles (except extraoculars) are Tbx1-dependent, whereas somite-derived muscles are Pax3-dependent (Sambasivan et al., 2011a). Furthermore, Alx4, Pitx1/2 are specifically expressed in the cranial mesoderm-derived extraoccular muscles (EOM) (Sambasivan et al., 2009). In addition, EOM-derived satellite cells showed greater ex vivo growth, self-renewal capacities and in vivo transplantation efficiency (Stuelsatz et al., 2015).
Similarly, single fibre transplantation experiments suggested that heterogeneity exists in muscles with the same developmental origin, but different anatomical location: MuSCs isolated from EDL (Extensor digitorium longus) or soleus muscles have superior engraftment potential compared to MuSCs from TA (Tibialis anterior) (Collins et al., 2005). Given that the MuSCs were grafted with their adjacent fibre in those experiments, this result could also be explained by the heterogeneity in the stem cell niche rather than cell-autonomous properties of the satellite cells.
Strikingly, even within a single muscle cell population, heterogeneity has been reported. Continuous in vivo labelling with the thymidine analogue BrdU (5′-bromo-2′-deoxyuridine) in 4weeks-old rats revealed two populations: about ≈80% of satellite cells readily marked over the first 5 days and a slow cycling minority of cells not fully saturated upon 2 weeks of treatment. This second population named “reserve cells” was proposed to maintain quiescence during muscle growth/homeostasis and enter cell-cycle only upon trauma (Schultz, 1996). Furthermore, freshly isolated single myofibres from Myf5nlacZ and Myf5Cre;R26RYFP mice showed ≈13% of MuSCs that never express Myf5 (Pax7+/β-gal—; Pax7+/YFP—, respectively), suggesting a more stem-like fate (Kuang et al., 2007). This Myf5— population is capable of asymmetric cell division and replenish the stem cell pool upon engraftment, whereas the Myf5+ undergo differentiation. These results suggest a hierarchical organisation of quiescent MuSCs: with a more stem population that will give rise to the more committed cells upon activation while self-renew to repopulate the quiescent niche. However, this phenotype is less pronounced with another Myf5Cre allele, and eventually all satellite cells experience Myf5 expression, therefore it is unclear how the genetically modified mice reflect stem-like behaviour over time (Sambasivan et al., 2013). Indeed, the presence/absence of labelling relies on the efficiency of the Cre-recombinase that has been shown to not faithfully represent Myf5 expression in every condition, a phenomenon that has been reported also for other tissues (Comai et al., 2014).
Satellite cell activation and differentiation
Immediately following muscle injury, Myod expression is rapidly upregulated and MYOD protein is already detectable within satellite cells as early as 12 h after injury, before the first cell division that takes place from about 20h (Rocheteau et al., 2012; Smith et al., 1994). This early expression of Myod is proposed to be associated with a subpopulation of committed satellite cells, which are poised to differentiate without proliferation (Rantanen et al., 1995). In contrast, the majority of satellite cells express either Myod or Myf5 by 24h following injury and subsequently co-express both factors (Cornelison and Wold, 1997; Gayraud-Morel et al., 2012; Zammit et al., 2002) (Figure 6). Interestingly, ectopic expression of Myod in NIH-3T3 and C3H10T1/2 fibroblasts is sufficient to activate the complete myogenic program in these cells (Hollenberg et al., 1993); thus expression of Myod is an important determinant of myogenic commitment and differentiation, and its absence promotes proliferation and delayed differentiation (Myod—/—)(Sabourin et al., 1999). During satellite cell activation, Pax7 and Pax3 target genes to promote proliferation and commitment to the myogenic lineage, while repressing genes that induce terminal myogenic differentiation (Soleimani et al., 2012). For example, PAX7 and PAX3 induce the expression of Myf5 by direct binding to distal enhancer elements and Myod by binding to the proximal promoter (Bajard et al., 2006; Hu et al., 2008). Moreover, p38 kinase (p38γ) also negatively regulates the transcriptional potential of Myod by phosphorylation, which leads to a repressive Myod complex occupying the Myogenin promoter (Gillespie et al., 2009). This observation is supported by the premature expression of Myogenin and reduced proliferation of myoblasts in p38-decificent muscle (Gillespie et al., 2009).
Terminal differentiation is initiated by the expression of Myogenin and later Mrf4 (Smith et al., 1994; Yablonka-Reuveni and Rivera, 1994) (Figure 6). ChIP-on-chip experiments (Bergstrom et al., 2002; Cao et al., 2006) and ChIP-Seq analysis (Cao et al., 2010) revealed MYOD and MYOGENIN specific target genes. These studies suggested a hierarchical organization involved in satellite cell activation and differentiation with regard to MRFs. MYOD directly activates Myogenin and Mef2 transcription factors, a large portion of downstream targets are muscle-specific structural and contractile genes, such as those encoding actins, myosins, and troponins, essential for proper myofibres function.
p38α/β kinase stimulates the binding of MYOD and MEF2s to the promoters of muscle-specific genes, leading to the recruitment of chromatin remodelling complexes promoting myogenesis (Cox et al., 2003; Wu et al., 2000). Besides MRFs and their regulators, other post-transcriptional factors have been shown to be involved in myogenic differentiation such as micro-RNAs (see Chapter 3).
Satellite cell self-renewal
The self-renewing capability of MuSCs has been demonstrated by series of transplantation experiments and clearly showed their remarkable ability to sustain the capacity for muscle repair. For example, transplantation of a single myofibre and its resident MuSCs (7-22/fibre) into irradiated muscles of immunodeficient dystrophic mice (nude; mdx) showed that MuSCs can give rise to over 100 new myofibres, expand and support further rounds of muscle regeneration (Collins et al., 2005). Similarly, purification of MuSCs followed by transplantation showed that they both contribute to muscle repair of nude; mdx mice and colonize the stem cell niche (Montarras et al., 2005). The self-renewing capability of satellite cells was further shown by serial transplantations of isolated Pax7-nGFP cells in pre-injured immunocompromised mice (Rocheteau et al., 2012); GFP+ cells were collected up to seven rounds of transplantations. Finally, single cell transplant experiments demonstrated that a single freshly isolated MuSC is capable to give rise to progeny cells and to self-renew upon injury (Sacco et al., 2008).
To study self-renewal ex vivo, two models are generally used: 1) floating isolated single myofibres where MuSCs will proliferate in clusters formed by activated, differentiated and self-renewed cells within 72h in the absence of cell fusion (Figure 7); 2) reserve cell model, where cells plated at high density will form myotubes and this is accompanied by the emergence of non-proliferative single cells (Pax7+) adjacent to the myotubes (Figure 7).
Extracellular matrix: powerful modulator of cell behaviour
ECM was initially considered to be an inert supportive scaffold, however, it is now clear that by either direct or indirect action, ECM regulates cell behaviour and it plays essential roles during development (Hynes, 2002). Indeed, the dynamism of ECM is provided by its capacities to adapt the production, degradation, and remodelling of its components. First, the ECM possesses both direct and indirect signalling properties, since it can act directly by binding cell surface receptors or by growth factor presentation (Hynes, 2002). Second, ECM components confer biomechanical properties to the ECM such as rigidity, porosity, topography and insolubility that can influence various anchorage-related biological functions, like cell division, tissue polarity and cell migration (Lu et al., 2011b). Indeed, ECM stiffness is an essential property by which cells sense the external forces and respond to the environment in an appropriate manner, a process known as mechanotransduction (DuFort et al., 2011; Mammoto and Ingber, 2010). Experiments performed with decellularized tissues, in which the ECM is preserved, showed capacity to guide stem cell differentiation into the cell types residing in the tissue from which the ECM was derived (Nakayama et al., 2010) (Webster et al., 2016). Despite the well-investigated cellular stem cell niche, details are lacking regarding the specific roles of ECM components (Figure 11).
Collagens constitute a major component of the ECM
One key component of the ECM is collagen, the most abundant protein in animals. In light of what has been described above, collagens provide essential structural support for connective tissues but they can also directly interact with cells through cell surface receptors or via intermediary molecules. Collagens have a triple helical structure composed of three genetically distinct polypeptide chains termed α-chains (α1, α2, α3), characterized by repeating glycine-X-X’ sequence with X and X’ being any amino acid. Looping of the three α-chains requires every third amino acid to be a glycine whereas 4-hydroxyproline-proline confers stability. In vertebrates 46 distinct collagen α -chains assemble to form 29 homodimer or heterodimer collagen types. Most triple helices assemble collagen into macromolecules to form fibrils and fibres that are essential components of tissues and bones. Collagen families include fibrillar collagen (eg. type I, III, V), network-forming collagen (COLIV, major component of basement membranes), fibril-associated collagens with interruptions in their helice (FACIT; eg IX, XII) and filamentous (COLVI; beaded microfibrils) (Mouw et al., 2014).
Upon synthesis, collagens α-chains are targeted to the ER where they assemble and undergo post-transcriptional modifications to form a precursor procollagen molecule. Note that the α1-chain is necessarily present in every collagen form. Procollagens are then secreted by cells into the extracellular space and converted into mature collagen by the removal of the N- and C-propeptides via collagen type-specific metalloproteinase enzymes (Mouw et al., 2014).
Insights from Collagen V
For the purpose of this thesis, we will focus on one specific type of collagen: type V Collagen. Collagen V is a fibrillar collagen involved in the regulation of fibril assembly and it can be classified as a regulatory fibril-forming collagen. The major isoform of Collagen V, [α1(V)]2α2(V) (two α1 chains and one α2), co -assembles with Collagen I to form heterotypic fibrils (Birk et al., 1988). The constitutive deletion of Collagen V in mouse (Col5a1—/—) is lethal at embryonic day E8.5. Interestingly, in the embryonic mesenchyme, even if the number of COLI fibrils is altered, the amount of Collagen I remains normal, suggesting that Collagen V is critical for fibril assembly (Wenstrup et al., 2004). Moreover, Col5a1 heterozygous mice are haploinsufficient and present a phenotype mimicking the human Ehlers-Danlos syndrome (EDS) that is characterized by a connective tissue disorder with broad tissue involvement typified by fragile, hyperextensible skin, widened atrophic scars, joint laxity, a high prevalence of aortic root dilation, and other manifestations of connective tissue (Malfait et al., 2010; Wenstrup et al., 2006). This mouse model of EDS of heterozygous Col5a1 ablation ultimately provides an explanation for the haploinsufficiency observed in Col5a1 mice (Wenstrup et al., 2006).
Native collagen triple helix can interact directly with cells via cell transmembrane receptors triggering diverse functions such as stable adhesion or migration. To date, four classes of vertebrate receptors have been described: collagen-binding integrins (α1β1, α2β1, α11β1, α10β1), discoidin domain receptors (DDRs), glycoprotein VI (GPVI), and leukocyte-associated immunoglobulin-like receptor-1 (LAIR-1). Although collagen-binding integrins and DDRs have different structures, they both bind to specific amino acid motifs within the collagen triple helix, and have overlapping cellular functions (Leitinger, 2011). In contrast, the structurally related receptors GPVI and LAIR-1 have similar collagen-binding motifs but mediate opposing functions: GPVI is an activating receptor on platelets, and LAIR-1 is an inhibitory receptor on immune cells (Leitinger, 2011).
Intriguingly, G-protein coupled receptors (GPCRs) have been shown to bind to collagen as well; to date, only two examples have been described in vitro: 1) Collagen III (COLIIIa1) interacts with GPR56 and induces RhoA downstream pathway to inhibit neural migration (Luo et al., 2011); and 2) Collagen IV binds to GPR126 and activates the cAMP downstream pathway (Paavola et al., 2014) in HEK293T cells.
Table of contents :
Abstract
Résumé
INTRODUCTION
Chapter 1. Skeletal muscle and its resident stem cells
1. Skeletal muscle structure and function
1.1. Skeletal muscle as a contractile unit
1.2. Muscle regeneration
2. Satellite cells as adult skeletal muscle stem cells
2.1. A brief history
2.2. Molecular regulation of muscle stem cell emergence
2.3. Heterogeneity in the muscle stem cell population
3. Functions of muscle stem cells
3.1. Adult myogenesis
3.1.1. Satellite cell activation and differentiation
3.1.2. Satellite cell self-renewal
Chapter 2. Stem cell niche is essential for quiescence
1. Stem cell quiescence
1.1. Identification of quiescent stem cells
1.2. Ex vivo induction of quiescence
1.3. Molecular signature of quiescence
1.3.1. Epigenetic control
1.3.2. Cell cycle regulators
2. Molecular signature of MuSCs
2.1.1. Calcitonin receptor
2.1.2. Teneurin-4 or Odz4
2. The stem cell niche
2.1. Extracellular matrix: powerful modulator of cell behaviour
2.2. ECM-cell interaction
2.3. Biophysical properties of ECM
2.4. Collagens constitute a major component of the ECM
2.4.1. Insights from Collagen V
3. The MuSCs niche
3.1. Extracellular matrix and associated factors
Chapter 3. Post-transcriptional regulation of myogenesis: a role for microRNAs
1. The discovery of microRNAs
2. MicroRNAs: Genomics, biogenesis, mechanism and function
2.1. Biogenesis of microRNAs
2.2. MicroRNAs arise from distinct genomic loci
2.3. MicroRNA prediction tools
3. MicroRNAs in cell and tissue regulation
4. Regulation of myogenesis by microRNAs
5. Inhibition of microRNAs using “Antagomirs”
Chapter 4. Notch signalling is a pleiotropic regulator of stem cells
1. An introduction to the world of Notch
2. Notch receptors, ligands and the cascade
3. Notch targets genes and their regulation
4. Notch signalling in the regulation of stem cell fate
5. Notch signalling in skeletal muscle and satellite cells
RESULTS
Part I: Notch-induced Collagen V maintains muscle stem cells by reciprocal activation of the Calcitonin Receptor
Part II: The Notch-induced microRNA-708 maintains quiescence and regulates migratory behavior of adult muscle stem cells .
CONCLUSIONS AND PERSPECTIVES
1. Context of this thesis project
2. Notch signalling regulates ECM niche components
3. Notch signalling positions MuSCs in their niche
4. Potential regulation of Notch signalling by microRNAs
ANNEX 1: Review Regulation and phylogeny of muscle regeneration
ANNEX 2: Resource paper Comparison of multiple transcriptomes using a new analytical pipeline Sherpa exposes unified and divergent features of quiescent and activated skeletal muscle stem cells
ANNEX 3: Small-RNA sequencing identifies dynamic microRNA deregulation during muscle lineage progression
REFERENCES
