Abstract
In vertebrates, body skeletal muscles and axial skeleton derive from the paraxial mesoderm which flanks the neural tube and notochord. The paraxial mesoderm forms in the posterior region of the embryo as presomitic mesoderm (PSM), which generates the embryonic segments called somites. Here, we characterized gene signatures identified using microarray series from the mouse PSM and compared the PSM transcriptome dynamics to that of the developing neural tube. In contrast to the PSM where an abrupt transcriptome reorganisation occurs at the level of the determination front, we show that transcriptome changes are progressive during parallel stages of neural tube differentiation. We show that these early differentiation stages of the paraxial mesoderm can be efficiently recapitulated in monolayer culture in vitro using murine Embryonic Stem (ES) cells. We describe a serum-containing protocol which parallels in vivo tissue maturation allowing differentiation of ES cells towards a paraxial mesoderm fate. We show that R-spondin treatment or Wnt activation alone can induce posterior PSM markers in both mouse and human ES/iPS cells but acquisition of a committed posterior PSM fate requires BMP inhibition to prevent induced cells to drift to a lateral plate mesoderm identity. We show that posterior PSM-like cells induced from mouse ES cells can be further differentiated in vitro to acquire an anterior PSM Pax3-positive identity. When grafted into injured adult muscle, these induced PSM-like precursors generated large numbers of immature muscle fibers. We further show that exposing ES-derived PSM-like cells to a brief FGF inhibition step followed by culture in horse serum-containing medium allows efficient recapitulation of the myogenic program. Differentiating ES cells first produce mononucleated embryonic myocytes and subsequently multinucleated myotubes, as well as Pax7-positive cells. The protocol described here results in improved differentiation and maturation of mouse muscle fibers differentiated in vitro over serum-free protocols. It provides an efficient system for the study of myogenic processes otherwise difficult to study in vivo such as fusion or satellite cell differentiation.
Introduction
Skeletal muscles represent a major derivative of the embryonic paraxial mesoderm. Presomitic mesoderm (PSM) cells, which emerge as a result of gastrulation from the primitive streak and later on the tail bud, express specific sets of genes such as Mesogenin1 (Msgn1) and they experience periodic signalling driven by the segmentation clock while located in the posterior PSM (Hubaud and Pourquie, 2014). Once cells become located in the anterior PSM, at the level of the so-called determination front, they acquire their segmental identity and activate expression of genes such as Pax3 that control their subsequent differentiation. Epithelial somites form at the anterior tip of the PSM and soon after their formation, distinct subsets of somitic cells begin to activate the myogenic and chondrogenic differentiation programs (Chal and Pourquie, 2009). At the trunk level, the dorso-lateral portion of epithelial somites forms the dermomyotome which contains the Pax3-expressing myogenic precursors. A subset of these precursors located in the dermomyotome lips first activates Myf5 and then Myogenin and gives rise to mononucleated postmitotic myocytes which form the myotome. Precursors of the limb and girdle muscles delaminate from the ventro-lateral lip of the dermomyotome to migrate to their final locations where they activate the myogenic program. Myocytes subsequently fuse in a highly patterned manner to first generate myotubes which further mature into myofibers. Thereafter muscle fibers continue to be formed during embryogenesis from a pool of proliferating precursors expressing Pax3 and Pax7 (Hutcheson et al., 2009). These cells ultimately form the embryonic, fetal and adult muscle fibers and the satellite cells (Biressi et al., 2007).
We recently identified Rspo3 as a secreted protein expressed in the posterior PSM able to induce the differentiation of ES cells toward a Msgn1-positive posterior PSM fate when combined with BMP inhibitors in chemically-defined conditions (Chal et al., 2015). Using chemically-defined culture conditions, these posterior PSM cells can be further induced to differentiate into anterior PSM fates characterized by Pax3 expression and can be used to subsequently generate large amounts of muscle cells in vitro and in vivo. Here, we describe a robust protocol based on Wnt signaling activation and BMP inhibition to efficiently produce PSM-like cells from mouse ES cells. We show that the transcriptome of in vitro differentiating mouse ES cells exhibits conserved kinetics of gene activation with mouse PSM cells in vivo. Furthermore, these induced PSM cells are able to engraft into adult injured muscles, and to generate large amounts of immature skeletal muscle fibers, supporting true paraxial mesoderm commitment. Finally, the PSM precursors could be further differentiated in vitro into muscle fibers which are more mature than that generated in serum-free conditions and amenable to long term analysis, thus providing an ideal system to study myogenesis in vitro using mouse ES cells.
Results
Comparison of the transcriptional landscape between the PSM and posterior Neural Tube in the mouse embryo
We previously reported the generation of a microarray series of consecutive micro-dissected fragments of the E9.5 mouse PSM (Chal et al., 2015). This data set describes the dynamics of the PSM transcriptome from the tail bud to the forming somite. We used this PSM microarray series to identify lists of approximately 40 to 50 highly specific gene signatures for the posterior and anterior PSM domains, respectively (Chal et al., 2015) (Table S1, S2). In addition to genes belonging to pathways well-known to be involved in PSM specification and differentiation such as Wnt, FGF or Notch signaling, the posterior PSM signature was enriched in genes associated to signalling pathways that have been little studied in the context of paraxial mesoderm development (Table S1, S2). Here, we have performed systematic whole mount in situ hybridization (ISH) using probes for these genes to validate their expression pattern. Genes identified as strongly expressed in the posterior PSM include the carbohydrate (N-acetylglucosamino) sulfotransferase 7 (Chst7), the gene regulated by estrogen in breast cancer product (Greb1), Tropomyosin alpha 1 (Tpm1), the EGF domain-specific O-linked N-acetylglucosamine (GlcNAc) transferase (Eogt), the previously identified cyclic gene Tnfrsf19 (Troy), as well as the Apelin receptor (Aplnr/Apj) (Kalin et al., 2007) and the Sphingosine phosphate receptors (S1pr3 and S1pr5) (Ohuchi et al., 2008) (Figure 1A, data not shown). Genes expressed in the entire PSM but not in somites included Greb1L, the interferon induced transmembrane protein 1 (Ifitm1, Fragilis2) (Tanaka and Matsui, 2002). Genes showing expression in the anterior PSM comprised the adducin gamma (Add3), the lipoma HMGIC fusion partner-like 2 (Lhfpl2) and Fibulin2 (Fbn2) (Figure 1A). A subset of genes was expressed as stripes in the anterior PSM and comprised the transcriptional regulator Myocardin (Myocd), the extracellular matrix protein vitronectin (Vtn), the epithelial remodelling factor Shroom3 (Sousa-Nunes et al., 2003), and a number of genes with unknown function in the PSM including Abca1, Fam101a (Cfm2) (Hirano et al., 2005), Arg1 (Hou et al., 2007), Pgm5 and Ism1 (Tamplin et al., 2008) (Figure 1A, data not shown). Thus our data identifies a new set of molecular players showing tightly restricted spatio-temporal expression in the mouse PSM.
During formation of the posterior body, PSM and neural tube have been shown to share common progenitors in the tail bud, the so-called Neuro-Mesodermal progenitors (NMPs) (Henrique et al., 2015). To evaluate how differentiation programs diverge once cells acquire a neural or paraxial mesoderm fate, we compared the transcriptional program during PSM differentiation with that of the adjacent neural tube. To that end, we generated a series of consecutive microdissected fragments of the posterior neural tube region adjacent to the PSM, from two different E9.5 mouse embryos. Each series comprised six contiguous ~100 μm long neural tube fragments spanning from the tail bud to the level of the newly formed somite (S0) (Figure 1B). Like for the PSM, RNA extracted from each fragment was used to hybridize a single Affymetrix microarray. Analysis of the expression profiles of known genes activated during neural tube differentiation including Pax6, Neurog1, NrCAM, Nkx6.1 or Nkx1.2 showed the expected expression gradients in the neural tube (Figure 1C). Thus, the paraxial mesoderm and neural tube microarray series provide an accurate representation of the transcriptional landscape during early stages of the development of these two tissues. Clustering analysis of the PSM microarray series identified an unbiased molecular subdivision of the PSM corresponding to the well-characterized “determination front” at which the segmental prepattern is first established and where differentiation begins (Chal et al., 2015). Such clear demarcation was not observed when a similar analysis was performed for the neural tube series (Figure 1B), suggesting that progressive transcriptional changes accompany early differentiation of the neural lineage (Table S3, S4). Thus our data identify novel PSM specific genes and argue for a different mode of transcriptome regulation during contemporary stages of paraxial mesoderm and neural tube development.
R-spondin treatment/ Wnt signaling activation in combination with BMP inhibitors promotes posterior PSM differentiation of mouse ES and human iPS cells
In vivo, the first stage of paraxial mesoderm cells differentiation corresponds to their exit from the primitive streak or tailbud to enter the posterior PSM. This stage is characterized by the activation of the gene Mesogenin1 (Msgn1), coding for a basic helix-loop-helix transcription factor specifically expressed in the posterior PSM (Yoon et al., 2000). Examination of gene expression profiles of secreted growth factors in the microarray series of differentiating paraxial mesoderm led to the identification of the secreted Wnt agonist R-spondin3 (Rspo3), which is strongly expressed specifically in the posterior PSM (Chal et al., 2015; Kazanskaya et al., 2004). When we treated monolayers of the Msgn1-repV mES reporter cells with 10 ng/mL Rspo3 in a base culture medium containing 15% Foetal Bovine Serum (FBS), we observed a dramatic increase of the percentage of Msgn1-repV+ cells over control medium to reach up to 70% after 4 days (Figure 2A). Dimethyl sulfoxide (DMSO) has been shown to promote differentiation of several cell types, including mesoderm from the P19 embryonic carcinoma cell line (McBurney et al., 1982) or from ES cells (Chetty et al., 2013). We also found that addition of DMSO synergized with Rspo3 to induce Venus from Msgn1-repV cells in serum-containing medium or in a serum-free medium (Figure S1, data not shown). Thus, 0.5% DMSO was systematically added to the differentiation medium.
We next tested whether the Venus-positive cells induced by Rspo3 exhibited a characteristic PSM identity. Microarrays from sorted Venus-positive cells cultured for 3 and 4 days in Rspo3 containing medium (RD) were generated and compared the expression of known markers of the PSM. At day 3, many characteristic markers of the posterior PSM such as Tbx6 were found to be expressed in differentiated ES cells (Figure 2B). Unexpectedly, the Bmp4 gene, which, in the posterior region of the embryo, is specific for the lateral plate mesoderm, was found to be enriched in the Venus-positive cells differentiated for 4 days in the presence of Rspo3 (Figure 2B). Bmp4 has been shown to promote the lateral plate mesoderm fate at the expense of the paraxial mesoderm and to auto-regulate its own expression (Adelman et al., 2002; Tonegawa et al., 1997). Accordingly, Msgn1-repV+ cells induced in RD media also expressed the lateral plate marker Foxf1a (Figure 2B). The expression of lateral plate markers in the cell population expressing Msgn1 suggests that treatment with Rspo3 alone leads cells to acquire a mixed identity between paraxial mesoderm and lateral plate.
In order to prevent cells to acquire a lateral plate identity, we added the BMP inhibitor Noggin to the culture medium. Microarray and qPCR analysis indicated that Noggin-treated cells down-regulated Bmp4 and Foxf1a and upregulated Tbx6 (Figure 2B). Furthermore, after 4 days in culture, Noggin-treated cells up-regulated the anterior PSM-specific markers Ripply2 and Pax3 (Figure 2B). Stronger inhibition of Bmp4 activity was observed with the chemical BMP inhibitor LDN 193189 (thereafter named Ldn), a more potent derivative of dorsomorphin (Cuny et al, 2008). qPCR analysis on FACS-sorted Msgn1-repV+ cells revealed that Ldn applied together with Rspo3 on mouse ES Msgn1-repV cells acts in a dose-dependent fashion, inhibiting Foxf1a and Bmp4 expression and promoting expression of Tbx6, Hes7, Msgn1 and Pax3 much more efficiently than Noggin (Figure 2C). Addition of Noggin or Ldn to the medium was however found to lead to lower levels of induction of Venus+ cells from the Msgn1-repV cultures (data not shown). Substituting Rspo3 by Rspo2 also led to similar results (Figure S1). Together, our data demonstrates that efficient differentiation of mouse ES cells toward a posterior PSM fate can be achieved by culturing the cells in a medium containing Rspo3, DMSO and Ldn (thereafter named RDL).
We next studied the mechanism of action of Rspo3. R-spondins are secreted molecules that bind Lgr4 and Lgr5 receptors and activate both the canonical and planar cell polarity (PCP) Wnt signalling pathways (de Lau et al., 2012). While Lgr5 expression was not significantly detected in the PSM in our profiling series, expression of Lgr4 was observed both in the PSM and neural tube ((Chal et al., 2015) and data not shown). We next asked whether Rspo3 activates canonical Wnt signaling in differentiating mouse ES cells. Addition of Dkk1 inhibited the induction of Venus+ cells from Msgn1-repV cells by Rspo3 (Figure 2D). Furthermore, substituting Rspo3 by the GSK3-inhibitor and canonical Wnt activator CHIRON 99021 (Chir) led to comparable induction level of the Msgn1-repV+ population (Figure 2E, Table S5 and data not shown). These data demonstrate that Rspo3 efficiently promotes the induction of a Msgn1-positive cell population by activating the canonical Wnt pathway, supporting an important role for this pathway in PSM specification.
We next analyzed induction of the posterior PSM fate from human iPS cells in vitro. To monitor the differentiation of hiPS cells toward paraxial mesoderm, we generated a fluorescent reporter line harboring a MSGN1-Venus fusion transcript by knock-in using the CRISPR/Cas9 system (Figure 3A). hMSGN1 reporter cells were differentiated in serum-free medium containing Chir/Ldn (CL) for 4-5 days to induce paraxial mesoderm differentiation (Chal et al., 2016; Chal et al., 2015). Venus was expressed in induced cells starting at day 3 and the MSGN1 transcript was almost exclusively found in the Venus+ (hM+) fraction (Figure 3B). By 4 days of differentiation, flow cytometry analysis demonstrates that up to 95% of the differentiating hiPS cells were Venus-positive, both in Chir/Ldn (CL) and in Chir only (C) conditions (Figure 3C). hM+ cells induced in the presence of Ldn also expressed the PSM-specific marker TBX6, validating the specificity of the reporter activity and the efficiency of the protocol to induce posterior PSM fate in human cells (Figure 3D). Next, we compared the gene expression profile of FACS-sorted hM+ cells generated in serum-free Chir/Ldn (CL) containing medium to undifferentiated hPSC (Figure 3E, Table S6). Mouse posterior PSM signature genes including DKK1, MSGN1, RSPO3, BRACHURY, HES7 and LFNG were strongly up-regulated in the differentiated hM+ cells (Figure 3E). To evaluate the impact of BMP inhibition on the gene signature of induced hM+ cells, microarrays from FACS-sorted hM+ cells cultured in serum-free Chir/Ldn (CL) or Chir only (C) medium were compared (Figure 3F). Genes of the Wnt (AXIN, DKK1, DKK2, RSPO3, LEF1, WNT3A), Notch (DLL1, DLL3, LFNG, HES7) and FGF (DUSP1, FGF10, FGF8, FGF9) signaling pathways which are important for posterior PSM development were found upregulated in hM+ cells induced in CL conditions (Fold change >2 and False discovery rate <10%). By contrast, hM+ cells induced in the presence of the WNT activator alone up-regulated lateral plate and cardiac markers including FOXF1, BMP4, TBX3, HAND1, HAND2, CXCR7 while posterior PSM markers were expressed at a lower level compared to cells differentiated in CL (Figure 3F, G, Table S7). Strikingly, the anterior paraxial mesoderm marker Pax3 was found to be up-regulated in absence of LDN (Figure 3F). Together these data suggest that these cells exhibit a mixed paraxial/lateral plate mesoderm identity. Thus, like for mouse ES cells, efficient induction of the paraxial mesoderm fate from human iPS cells requires BMP inhibition at the critical stage of posterior PSM differentiation, preventing early paraxial mesoderm cells from drifting to a lateral plate mesoderm fate.
We next compared hM+ progenitors induced in vitro by Wnt activation and BMP inhibition to their mouse counterparts (Chal et al., 2015). As observed for the mouse Msgn1-repV+ differentiated in serum-free CL conditions, hM+ cells upregulated a large number of posterior PSM signature genes, including MSGN1, TBX6, BRACHURY, WNT5A, RSPO3, CDX2, and EVX1 (Figure 3H; Table S5, S6). Interestingly, a subset of genes were found differentially enriched specifically in human versus mouse M+ progenitors. hM+ progenitors expressed genes such as DKK2 and CXCR4 which were detected at significantly lower level in mouse M+ cells, while mouse M+ cells show enriched expression of some FGF pathway components (Dusp2, Dusp4, Fgf3, Fgf15) when compared to their human counterparts. Altogether, our data indicate that as reported for mouse ES cells, in vitro differentiation of human posterior PSM-like cells can be induced in the presence of a WNT activator and a BMP inhibitor. Moreover, our data suggest that the transcriptome of mouse and human posterior PSM-like cells is highly similar.
Generation of Pax3-positive anterior PSM precursors from ES cells in vitro
In vivo, as paraxial mesoderm cells mature, they become located in the anterior PSM, and down-regulate the expression of Msgn1 while activating the expression of Pax3 at the level of the determination front. FACS-sorted mouse M+ cells differentiated in RDL or CDL for 3 days strongly express the posterior PSM markers Tbx6 and Hes7 compared to M- cells, indicating that they share features with posterior PSM cells (Figure 2D, 4A). After 4 days of culture in RDL M+ cells activated anterior PSM and early somite markers including Pax3, Meox1/2, Paraxis/Tcf15, Foxc1/2, Tbx18, and Uncx4.1 (Figure 4A-C). While in vivo, these genes are expressed in a more anterior PSM domain than Msgn1, their co-expression with the Venus protein is due to the stability of the reporter which persists for some time in cells that have ceased expressing Msgn1. M+ cells induced in basal (FBS 15%, F15) medium without Ldn strongly up-regulated BMP4 at day 4-5 (Figure 4A-C). This transition to an anterior PSM fate was highly dependent on the concentration of Ldn, with 100nM leading to the highest induction of anterior PSM markers, and the lowest expression levels of the lateral plate markers (Figure 4B). Chir (CDL medium) was found to be as efficient as R-spondin3 at inducing the anterior PSM fate (Figure 4C) while cells differentiated in base F15 medium failed to activate the anterior PSM program. Therefore, our data suggest that mouse ES cells differentiated in RDL or CDL media are able to efficiently differentiate into anterior PSM and to activate the early somitic differentiation program.
In order to better monitor the differentiation efficiency of mouse ES cells into anterior PSM precursors, we took advantage of the mouse ES Pax3-GFP reporter line (Chal et al., 2015). During development, Pax3 is expressed both in mesodermal and neural derivatives, but fluorescence intensity of the reporter was found to be weaker in the former. When these cells were differentiated in F15 or in R-spondin3-containing medium without Ldn (RD), about 1% of GFP+ cells was detected after 5 days in culture (Figure 5A). In contrast, in RDL medium, up to 20-40% of GFP+ cells were observed at day 5 (Figure 5A). The peak of Pax3-GFP expression followed by approximately one to two days the peak of Msgn1-repV expression. Maximal induction of the Pax3-GFP+ cells was observed when Ldn was added in the differentiation media at day 0 (Figure 5B). We confirmed that at day 5-6, Pax3-GFP+ cells in culture also expressed the Pax3 protein (Figure 5C). Pax3 is expressed both in the paraxial mesoderm and in neural precursors. However, the Pax3-GFP+ cells induced in our RDL/ CDL conditions exhibit a characteristic mesenchymal aspect, distinct from the rosette-forming Pax3+ neural precursors found in neural-inducing (dual Smad inhibition) conditions (Figure 5D) (Chambers et al., 2009). The Pax3-GFP+ cells were also found to be essentially negative for the neural marker Sox2 but expressed the anterior PSM/somitic markers Pax3, Uncx, Meox1 and Foxc2 at levels comparable to those detected in vivo (Figure 5E). Thus, differentiation of ES cells in RDL or CDL media was found to efficiently recapitulate the major stages of PSM differentiation in vitro.
Validation of the paraxial mesoderm identity of mouse ES cells differentiated in vitro
The presence of PDGFRα (CD140a) and the absence of VEGFR2 (also named Flk1, KDR or CD309) have been used to identify the paraxial mesoderm lineage in differentiated ES cells in vitro (Darabi et al., 2008; Nishikawa et al., 1998; Sakurai et al., 2006; Sakurai et al., 2009). In vivo, however, these surface markers are not specific to the paraxial mesoderm and their expression largely overlaps with other mesodermal populations such as the lateral plate (Ding et al., 2013; Ema et al., 2006; Motoike et al., 2003). FACS analysis of the Msgn1-repV cultures differentiated for 3-4 days in RDL medium with or without a pre-differentiation step in N2B27 medium supplemented with 1% Knock-out serum (thereafter, NK1) medium (Chal et al, 2015), revealed that more than 90% of the Msgn1-repV+ (M+) cells are recognized by the PDGFRα antibody (Figure 6A). However, 25% of these M+ cells were also found to express VEGFR2 (Figure 6B). Moreover, PDGFRα was not specific to the Msgn1-repV+ population as it also marked 75% of the Msgn1-repV- (M-) cells (Figure 6B), 45% of which also expressed VEGFR2 (Figure 6A, B). We also analyzed expression of the surface marker CXCR4 (Borchin et al., 2013) in relation to M+ cells and found that while most of the M- were CXCR4+, about 30-50% of the M+ cells were also CXCR4+, suggesting the idea that CXCR4 cannot discriminate for PSM fate in vitro (Figure S2).
To further characterize these cell populations, we analyzed by qPCR the expression profile of various lineage markers in different fractions of differentiated Msgn1-repV+ cells sorted based on Venus and PDGFRα expression (as indicated in Figure 6A). We found that the Msgn1-repV+ PDGFRα+ (M+Pα+) fraction was strongly enriched for paraxial mesoderm markers such as Msgn1, Tbx6, and Pax3 (Figure 6C). The Msgn1-repV- PDGFRα+ (M-Pα+) fraction showed very low levels of Tbx6 suggesting that it does not contain PSM cells (Figure 6C). These M-Pα+ cells expressed Sox10 together with Pax3 but lack Sox2 expression suggesting that this fraction contains neural crest cells (Figure 6C). The Msgn1-repV subpopulations were further analyzed by qPCR following combined PDGFRα and VEGFR2 staining. While the Msgn1-repV+ PDGFRα+ VEGFR2- (M+Pα+V-) population expressed the highest levels of anterior PSM/somitic markers, the triple positive M+Pα+V+ population also expressed a significant level of anterior PSM/somitic markers (Figure 6D). Interestingly, among the Msgn1-repV- subpopulations, the PDGFRα+ VEGFR2- (M-Pα+V-) also expressed comparable level of anterior PSM/somitic markers, suggesting that by day 4 a fraction of cells transited to anterior PSM/ somitic fate and downregulated the Msgn1-repV reporter. Together, our results indicate that the posterior PSM-like cells differentiated in vitro express PDGFRα, but they also argue for the lack of specificity toward the paraxial mesoderm lineage of the anti-PDGFRα antibody, even when combined with the VEGFR2 antibody.
To further establish the specificity or lack thereof of the PDGFRα and VEGFR2 surface markers in identifying somitic mesoderm, the Pax3-GFP mES line was differentiated for 6 days in RDL medium and analyzed for PDGFRα and VEGFR2 surface expression. Pax3-GFP+ cells differentiated with or without pre-differentiation in NK1 medium (Chal et al, 2015) were essentially negative for VEGFR2 while the majority were PDGFRα-positive (Figure 6E). Nevertheless, Pax3-GFP+ PDGFRα- accounted for about 15 to 35% of the total Pax3-GFP+ population. Moreover, staining for CXCR4 showed that Pax3-GFP+ are negative for CXCR4 (Figure S2). Altogether, this suggest that PDGFRα, VEGFR2 and CXCR4 surface expression does not fully capture Paraxial mesoderm identity.
Transcriptomic analysis of the differentiated Msgn1-repV and Pax3-GFP-positive cells
We next compared the identity of the transcriptome of the Msgn1-repV and Pax3-GFP-positive cells differentiated in vitro to their in vivo counterpart. Microarrays were generated for Msgn1-repV+ and Pax3-GFP+ cells sorted by FACS after 3, 4 or 5 days of differentiation in serum-containing RDL and CDL media. Their gene signatures were compared to that of the anterior and posterior PSM transcriptional domains in vivo (Chal et al., 2015) and with the neural tube array series to examine tissue specificity (Figure 7, Table S2, S5). As the differentiating ES cells transited from a Msgn1+ to a Pax3+ stage, they down-regulated a large number of posterior-specific PSM genes including Dusp4, Rspo3, Evx1 and Fgf8 as observed in vivo (Figure 7A, C). Interestingly, while many posterior PSM signature genes where also shared with the posterior-most neural tube, such was not the case for the anterior PSM signature genes. In parallel, progressive activation of a large fraction of the anterior PSM specific signature genes including Ripply2, Mesp2, Nkx3-1, Tbx18, was observed during in vitro differentiation (Figure 7B, D). Thus, in vitro differentiation of the Msgn1-repV and the Pax3-GFP reporter ES cells in the presence of Rspo3 and BMP inhibitors appears to recapitulate the early differentiation stages of the paraxial mesoderm in vivo.
In vivo analysis of the myogenic potential of the ES-derived PSM-like cells
Trunk skeletal muscle tissue is generated exclusively from somitic paraxial mesoderm (Sambasivan et al., 2011). Properly specified PSM-like cells should therefore have the unique potential to generate skeletal muscle. We next analyzed the myogenic potential of the mouse PSM-like cells differentiated in vitro by transplanting them directly into injured adult muscles in vivo. Msgn1-repV+ and Pax3-GFP+ cells were differentiated for 3-4 days and 5-6 days in RDL medium respectively, and subsequently purified by FACS. Sorted cells were permanently labeled with a lentivirus driving ubiquitous expression of GFP in order to follow their fate after they are grafted in vivo. 50,000-100,000 labeled Msgn1-repV+ or Pax3-GFP+ cells were transplanted into cardiotoxin-injured tibialis anterior muscle of adult Rag2-/- γc-/- mice (Figure 8A). Freshly isolated, GFP-labelled adult satellite cells were also used as a positive control. One month posttransplantation, Msgn1-repV+ and Pax3-GFP+ donor cells reconstituted large areas filled with small, poorly organized, striated dystrophin-positive muscle fibers as well as occasionally other derivatives such as fibroblasts, chondrogenic nodules or epithelial cells forming large cysts (Figure 8B, C and data not shown). GFP-labeled muscle fibers expressed embryonic, slow or perinatal/fast isoforms of myosin heavy chain (MyHC) indicating that they span a large array of myogenic differentiation stages (Figure 8D). Thus, when transplanted in vivo into adult injured muscles, PSM-like cells derived in vitro from ES cells are able to continue their differentiation toward the myogenic lineage.
Differentiated PSM-like cells can recapitulate skeletal myogenesis in vitro
We next sought to define conditions in which the anterior PSM-like cells differentiated from ES cells could be reproducibly induced to generate skeletal muscle in vitro. In the embryo, FGF and Wnt signaling are downregulated in the anterior PSM prior to activation of the myogenic program (Aulehla and Pourquie, 2009). We therefore first differentiated ES cells in RDL medium for 4-6 days and then cultured cells for 2 days in a medium (thereafter referred as PDL) lacking the Wnt activator and containing the FGF inhibitor PD173074, while maintaining the BMP inhibition with LDN193189. Subsequently, cells were transferred to a myogenic differentiation medium containing 2% horse serum, in which they were maintained for up to 2 months. To monitor the generation of skeletal muscle cells, we used the Myog-repV ES reporter line described in (Chal et al., 2015). Myog-repV+ myocytes harboring a single centrally located nucleus, strongly resembling early myotome cells produced during primary myogenesis, were visible after 1 week in culture in these conditions (Figure 9A). We found that predifferentiating the mouse ES cell for 2 days in NK1 medium prior to exposure to RDL medium led to a more robust and homogeneous activation of the Myog-repV reporter in culture. Futhermore, substituting R-spondin3 by the GSK3β inhibitor and Wnt activator CHIR 99021 also led to efficient generation of Myog-repV+ myocytes. These myocytes were not observed when cells were maintained in RDL (or CDL) after day 6 or differentiated in base FBS 15% media (data not shown). The number of myocytes steadily increased over time progressively covering the entire surface of the wells (Figure 9D). Large numbers of elongated slow and perinatal/fast Myosin Heavy chain (MyHC) positive fibers progressively appeared from the Myog-repV+ myocytes cells (Figure 9B,C). While immature myogenic cells expressed MyoD and a limited subset also Myf5, myonuclei found in myotubes were positive for MyoD and Myogenin (Figure 9E and data not shown). By 2 weeks of differentiation, myotubes formed by fusion of myocytes which aligned spontaneously in culture, leading by three weeks in vitro to a large number of multinucleated Fast-MyHC+ fibers (Figure 9F-H) containing up to 50-100 myonuclei each, as seen in perinatal fibers in vivo (White et al., 2010) (Figure 9 G; data not shown). Moreover, larger fibers expressed Dystrophin in their sub-sarcolemmal domain (Figure 9I), and in some instance exhibited a single cluster of Acetylcholine receptors in equatorial position, as detected by fluorescent bungarotoxin (Figure 9J). Strikingly, the muscle fibers differentiated in culture for 3 to 5 weeks extended over several millimeters (Figure 9F-H) and reached a density of up to 80 fibers per square millimeter (data not shown). Around 10 to 15,000 such multinucleated muscle fibers could be obtained in wells initially seeded with 20 to 30,000 ES cells. Immunolabeling with anti-perinatal MyHC antibody showed that the generated fibers exhibit highly organized striation as expected from mature muscle fibers (Figure 9H). Many of the differentiated fibers contracted spontaneously indicating that the sarcomeric organization of the fibers is functional (Movie S1, S2). The dimensions of these fibers were between 1-3 millimeters in length and 10-20 micrometers in width, with a sarcomeric length around 2.5 micrometers, which are values similar to those of mouse post-natal fibers (White et al., 2010) (Figure 9L,M). Fibers were individually surrounded by a continuous basal lamina visualized by laminin deposits (Figure 9K). In long term cultures, muscle fibers were frequently found to develop over an epithelial-like sheet of cells (data not shown). Together, these data suggest that our in vitro culture conditions are able to recapitulate a myogenic differentiation sequence resembling that observed during normal embryogenesis in the mouse.
We next used qPCR to analyze the expression level of a set of myogenic markers in 3-week old cultures differentiated in the conditions described above. These cultures were compared with undifferentiated ES cells, with E11.5 trunk muscles (containing primary myofibers) and E17.5 back muscles (in which foetal myogenesis is ongoing) (Figure 9N). While expression of the myogenic factors MyoD and Myogenin was clearly observed in the three muscle-containing samples, Pax3 was only detected in the E11.5 sample. In contrast, Pax7 was robustly expressed in both muscle samples and at a lower level in the differentiated ES myogenic cultures. The marker of fetal muscle fibers Nfix (Messina et al., 2010) and the Myh2 (Fast2A MyHC) were not observed in E11.5 trunk muscles but were significantly enriched both in the E17.5 muscle sample and in the 25-day old cultures (Figure 9N). Thus, both primary and secondary (fetal) skeletal myogenesis can be recapitulated in vitro ultimately resulting in the production of striated contractile muscle fibers exhibiting a phenotype similar to early post-natal fibers.
Production of ES-derived Pax7-positive myogenic precursor cells in vitro
During embryogenesis, muscle fibers are generated by proliferating progenitors expressing Pax7 and/or Pax3 that are located in between the growing fibers and that ultimately become located under the fiber basal lamina during late fetal stage to become satellite cells. We examined the presence of such Pax7+ progenitors in our long term cultures of differentiated ES cells described above. By two to three weeks of differentiation, large streams of Pax7+ cells were observed in the cultures and interspersed with newly formed myocytes suggesting that these correspond to myogenic progenitors (Figure 10A). Between three and four weeks, these populations quickly resolved to generate skeletal myocytes and aligned muscle fibers. By 4 weeks of differentiation and onward, the number of Pax7+ cells in culture decreased drastically. Nevertheless, a small population of Pax7+ was found in close association to large myofibers, a topography strikingly reminiscent of the in vivo situation where adult Pax7+ satellite cells are found in contact with fully differentiated muscle fibers (Yin et al., 2013) (Figure 10B, C). A fraction of Pax7+ progenitors become quiescent as suggested by the loss of Ki67 expression (Figure 10D-F, H). Thus our experiments show that differentiation of Satellite-like Pax7-positive myogenic progenitor cells can be obtained in vitro using our differentiation strategy.
Discussion
Here, we describe a novel protocol recapitulating developmental cues in vitro to promote highly efficient myogenic differentiation of mouse ES cells. This protocol, which contains serum, allows for more effective differentiation and maturation of striated myofibers than the chemically defined protocols recently published (Chal et al., 2015; Shelton et al., 2014). Thus it might provide a useful tool to study aspects of myogenesis otherwise difficult to study in vivo such as myoblast fusion, myofibrillogenesis or satellite cell maturation. Activation of Wnt/β-catenin signaling combined to BMP inhibition leads to efficient differentiation of ES cells toward a posterior PSM identity. Further differentiation of these cells leads to the differentiation of Pax3-positive cells exhibiting characteristics of the anterior PSM fate. We also identified culture conditions allowing these PSM precursors to differentiate into Myogenin-positive mononucleated myotubes after 7 days. In long term culture, these cells produce large numbers of muscle fibers and their associated proliferating Pax7-positive precursors.
When allowed to differentiate in various conditions, embryoid bodies derived from mouse or human embryonic stem cells can yield a wide variety of derivatives from the three germ layers including paraxial mesoderm specific derivatives (Awaya et al., 2012; Braun and Arnold, 1994; Chang et al., 2009; Mizuno et al., 2010; Rohwedel et al., 1994). However, these approaches yield highly mixed populations and differentiation of mature derivatives such as skeletal muscle is poorly efficient and reproducible. Alternatively, differentiation of mouse and human pluripotent cells in monolayers combined with various treatments to modulate the Activin, BMP, and Wnt pathways has been reported to be able to induce a paraxial mesoderm fate (Sakurai et al., 2009; Sakurai et al., 2012; Tanaka et al., 2009). However, the methods reported in these studies lead to limited production of paraxial mesoderm precursors which need to be enriched by FACS-sorting to generate limited amount of paraxial mesoderm derivatives such as muscle or cartilage. Furthermore, these studies relied on the expression of PDGFRα to identify paraxial mesoderm cells. Our data show that whereas the Msgn1+ population of differentiated ES cells is recognized by the antibody against PDGFRα, this antibody also recognizes many other non-paraxial mesoderm cell populations. Hence, even in combination with the anti-VEGFR2 antibody, the anti-PDGFRα antibody is not specific for the paraxial mesoderm lineage and thus cannot be used as the identification criterion for this lineage. This is also supported by lineage tracing experiments in the mouse embryo (Ding et al., 2013; Ema et al., 2006; Motoike et al., 2003). Most studies published so far on the differentiation of paraxial mesoderm cells and their derivatives from ES cells, have relied on this identification method (Chan et al., 2016; Darabi et al., 2008; Filareto et al., 2012; Hwang et al., 2014; Magli et al., 2013; Sakurai et al., 2006; Sakurai et al., 2008; Sakurai et al., 2012; Tanaka et al., 2009) and thus the populations described as paraxial mesoderm likely correspond to a mixture of fates.
Using the Msgn1-repV and Pax3-GFP lines, we demonstrate that activation of the Wnt/β-catenin pathway with Rspo3 or CHIR-99021, can induce up to 70-90% of cells to activate the Msgn1-RepV reporter, which identifies a posterior PSM fate. Such a requirement of Wnt/ β-catenin signaling for paraxial mesoderm differentiation has been well established both in vivo in mouse mutants for the Wnt pathway (Dunty et al., 2008; Galceran et al., 2004; Takada et al., 1994; Yamaguchi et al., 1999), and in vitro in differentiating ES cell cultures (Borchin et al., 2013; Mendjan et al., 2014; Shelton et al., 2014). We further demonstrate that while Wnt activation alone is sufficient to activate expression of Msgn1 in mouse and human pluripotent cells, these cells start to express Bmp4 and progressively drift to a lateral plate fate, expressing markers such as Foxf1 or Hand1 and 2. In vivo, Bmp4 has been shown to be able to divert early paraxial mesoderm progenitors toward a lateral plate fate, consistent with its well described ventralizing activity on the developing mesoderm (Tonegawa et al., 1997). When mouse ES or human iPS cells were treated with Rspo3 or CHIR-99021 alone, we noticed an activation of Bmp4 in the Venus -positive cells together with expression of lateral plate markers such as Foxf1. This raised the possibility that early Msgn1-repV cells exhibit a mixed unresolved paraxial mesoderm/lateral plate identity. Alternatively, some cells might have transiently expressed Msgn1, and retained the stable fluorescent Venus, while subsequently differentiating into lateral plate under the influence of Bmp4, whereas other Msgn1+ cells retained a paraxial identity thus generating an heterogenous population of Msgn1-repV+ cells composed of a mixture of lateral plate and paraxial mesoderm cells. Since Bmp4 has been shown to activate its own expression (Adelman et al., 2002; Blitz et al., 2000; Rojas et al., 2005; Schuler-Metz et al., 2000), we treated the cells with BMP inhibitors such as Noggin to block this endogenous BMP production. This led to an active downregulation of Bmp4 and to the dose-dependent induction of a paraxial mesoderm fate. Interestingly, we observed significant differences towards induction of Pax3-positive cells in vitro between different BMP inhibitors. Whereas Noggin was able to efficiently contribute to induction of the posterior Msgn1-positive fate, only LDN-193189 (which inhibits Bmp signaling more strongly than Noggin) was found to be efficient for generating Pax3-positive cells in vitro. Alternatively, Ldn has been shown to also inhibit other kinases than the BMP receptors, such as FGFR1 (Vogt, 2011). These off-target effects could also account for the improved efficiency of Ldn on Pax3 induction. Together, our data demonstrate that active Wnt signaling together with BMP inhibition is required to differentiate ES cells toward the paraxial mesoderm lineage. BMP modulation as also been shown to be important for intermediate mesoderm and chondrogenic mesoderm induction (Craft et al., 2013; Craft et al., 2015; Morizane et al., 2015; Tanaka et al., 2009; Umeda et al., 2012; Zhao et al., 2014). These requirements are clearly different from those necessary to produce other mesodermal cell types such as hematopoietic and cardiovascular lineages where the BMP and Wnt pathway need to be activated simultaneously, followed by a subsequent Wnt signaling inhibition step required to induce differentiation of the cardiomyocyte lineage (Laflamme et al., 2007; Lian et al., 2012; Naito et al., 2006; Nostro et al., 2008; Ueno et al., 2007; Yang et al., 2008; Zhang et al., 2008).
Finally, we generated a microarray time series from fragments of microdissected embryonic neural tube and compared it to the PSM series (Chal et al., 2015). Clustering analysis of these microarrays revealed that unlike the PSM, the early neural tube differentiation is a progressive process with no abrupt transcriptional fate specification, suggesting that PSM and neural tube differentiation are controlled by distinct mechanisms (Oginuma et al., 2017; Olivera-Martinez et al., 2014; Ozbudak et al., 2010). Interestingly, we found that the gene signature the posterior-most neural tube microdissected fragment was largely overlapping with the Tail bud fragment of the PSM series, supporting the idea that the progenitors of both lineages share a common program/origin (Henrique et al., 2015).
Contributions
J.C. designed and performed experiments, analyzed data and coordinated the project. Z.A.T. designed and performed experiments, generated and characterized the hMSGN1-Venus line with help from B.G. M.O. performed the neural tube microdissection series. B.G., A.M. and G.G. carried out most of the mouse ES cell differentiation and characterization experiments under J.C.′s supervision. B.G. performed the in situ hybridization screen under J.C.′s supervision. O.S. characterized the Myog-repV line. P.M. and O.T. helped with microarray data analysis. A.B. contributed experimentally to the early project. L.K. provided technical support. J.-M.G. generated reporter constructs. M.K. established the hPS culture, provided expertise and coordinated the project. B.G.-M. and S.T. provided expertise and performed transplantation. O.P. supervised the overall project. O.P. and J.C. performed the final data analysis and wrote the manuscript.
Competing interests
The work described in this article is partially covered by patent application no. PCT/EP2012/066793 (publication no. WO2013030243 A1). O.P., J.C. and M.K. are cofounders and shareholders of Anagenesis Biotechnologies, a startup company specializing in the production of muscle cells in vitro for cell therapy and drug screening.
Supplementary Materials
Table S1: List of genes differentially expressed between posterior and anterior PSM transcriptional domains. Expression intensity of each expressed probeset (RMA annotation) were compared between the posterior PSM domain (fragments 2 and 3 of PSM array series) and the anterior PSM domain (fragments 4, 5 and 6 of PSM array series). Comparison was done by Significance Analysis of Microarrays (SAM) analysis using MeV4.5 (TM4 Microarray Software suite). For each probeset, the corresponding gene and the relative expression fold change between the posterior and anterior PSM domains are shown. The False Discovery Rate was fixed at 0.525. Genes validated by in situ hybrization in this study are highlighted (yellow), or validated genes in the literature are also highlighted (orange).
Table S2: List of signature genes for posterior and anterior PSM domains used for expression heatmaps. For each marker gene, Affymetrix probeset identity is shown. Lists were assembled from this study and published literature.
Table S3: Complete lists of signature genes for each Paraxial mesoderm and Neural tube domains (GSL method). Signature gene lists were generated by the GSL method (see Materials and Methods for details) for each paraxial mesoderm domains in the PSM array serie, namely Tail bud, Posterior PSM, Anterior PSM and Somite. The neural tube array series was arbitrarily subdivided into a posterior and anterior domains and signature gene lists were generated for both.
Table S4: Venn diagram comparing the gene signatures list (GSL) of PSM versus Neural tube (NT) array series. (left) Comparison between the core Neural tube GSL genes and the core PSM GSL genes. Core signature genes are genes found significantly upregulated in domain of a serie for a given tissue. (right) Comparison between all the signature Neural tube genes with all the signature PSM genes. For each subgroup, the corresponding list of signature genes is provided.
Table S5: Lists of signature genes for mouse ES cells and differentiated FACS-sorted Msgn1-RepV-positive and Pax3-GFP-positive populations (GSL method). Signature gene lists were generated by the GSL method (see Materials and Methods for details) for undifferentiated mouse ES cells and differentiated populations (Msgn1-repV+, Pax3-GFP+) for 4 to 5 daysof differentiation in Rspo3/DMSO/Ldn (RDL) or Chir/DMSO/Ldn (RDL) medium. The top 400 probe sets are provided for each conditions. For each probe set, the corresponding gene and the relative expression fold change compared to the reference median value are shown.
Table S6: List of genes differentially expressed in in 4 day-old hMSGN1-GFP-positive cells cultured in Chir/Ldn or Chir only media versus undifferentiated hiPSC. (TabA) List of genes upregulated >2 folds in MSGN1-GFP+ cells in Chir/LdnL versus undifferentiated hiPSC. (TabB) List of genes upregulated >2 folds in MSGN1-GFP+ cells in Chir only versus undifferentiated hiPSC
Table S7: List of genes differentially expressed in 4 day-old hMSGN1-GFP-positive cells cultured in Chir/Ldn versus Chir only media. (TabA) List of genes upregulated >2 folds in hMSGN1-GFP+ cells in Chir/LdnL vs Chir only conditions. (TabB) List of genes upregulated >2 folds in MSGN1-GFP+ cells in Chir only vs Chir/Ldn conditions. For each probeset/gene Fold change (red) and False Discovery Rate (FDR(BH), yellow) are highlighted.
Table S8: In situ probes information
For each gene/probe, the ENSEMBL reference sequence number is provided. The primer couples sequences (F: forward, R: reverse) used to amplify the region of interest and the corresponding probe template size are provided.
Movie S1: Spontaneous contractile activity of differentiated mouse ES-derived myofibers. Mouse ES cells were differentiated in to PSM-like cells according to the RDL medium protocol and subsequently cultured in HS2% of additional weeks and imaged with transillumination. Real time speed.
Movie S2: Spontaneous muscle bundle formation and contractile activity in adherent differentiated ES cultures. Mouse ES cells were differentiated in to PSM-like cells according to the CDL medium protocol and subsequently cultured in HS2%. Large fiber bundle formed spontaneously in vitro anchoring only by extremities (not shown). Note synchroneous twitching of the bundle of aligned myofibers. Real time speed.
Acknowledgments:
We thank Christopher Henderson for critical reading of the manuscript. We are grateful to Jennifer Pace and Tania Knauer-Meyer for their help; Laurent Bianchetti for Bioinformatic support. We thank Claudine Ebel from the Cytometry Facility at IGBMC and the IGBMC Microarray Facility for assistance. We thank the Pasteur Institute Cytometry and Animal Facilities and the IGBMC Cell Culture Facility for assistance. This work was supported by an advanced grant from the European Research Council (ERC-2009-AdG 249931 to O.P.), by the FP7 EU grant Plurimes (agreement no. 602423) and by a strategic grant from the French Muscular Dystrophy Association (AFM-Téléthon) to O.P. Microarrays data reported in this paper have been deposited in GEO database under the accession number GSE39615 and pending.