Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Non Conservation of Function for the Evolutionarily Conserved Prdm1 Protein in the Control of the Slow Twitch Myogenic Program in the Mouse Embryo Stéphane D. Vincent,*, ,1 Alicia Mayeuf,1 Claire Niro,2 Mitinori Saitou,à,3 and Margaret Buckingham1 1 Department of Developmental Biology, CNRS URA 2575, Institut Pasteur, Paris, France Department of Genetics and Development, Institut Cochin, Université Paris Descartes, CNRS UMR 8104, INSERM U567, Paris, France 3 Laboratory for Mammalian Germ Cell Biology, RIKEN Center for Developmental Biology, Chuo-ku, Kobe, Japan Present address: Department of Development and Stem Cells, Institut de Génétique et de Biologie Moléculaire et Cellulaire, CNRS UMR 7104, Inserm U964, Université de Strasbourg, Illkirch, France àPresent address: Department of Anatomy and Cell Biology, Graduate School of Medicine, Kyoto University, Yoshida-Konoe-cho, Sakyo-ku, Kyoto, Japan *Corresponding author: E-mail: [email protected]. Associate editor: Claudia Kappen 2 Muscles are composed of multinucleated muscle fibers with different contractile and physiological properties, which result from specific slow or fast gene expression programs in the differentiated muscle cells. In the zebra fish embryo, the slow program is under the control of Hedgehog signaling from the notochord and floor plate. This pathway activates the expression of the conserved transcriptional repressor, Prdm1 (Blimp1), which in turn represses the fast program and promotes the slow program in adaxial cells of the somite and their descendants. In the mouse embryo, myogenesis is also initiated in the myotomal compartment of the somite, but the slow muscle program is not confined to a specific subset of cells. We now show that Prdm1 is expressed in the first differentiated myocytes of the early myotome from embryonic day (E)9.5–E11.5. During this period, muscle formation depends on the myogenic regulatory factors, Myf5 and Mrf4. In their absence, Prdm1 is not activated, in apparent contrast to zebra fish where Prdm1 is expressed in the absence of Myf5 and MyoD that drive myogenesis in adaxial cells. However, as in zebra fish, Prdm1 expression in the mouse myotome does not occur in the absence of Hedgehog signaling. Analysis of the muscle phenotype of Prdm1 mutant embryos shows that myogenesis appears to proceed normally. Notably, there is no requirement for Prdm1 activation of the slow muscle program in the mouse myotome. Furthermore, the gene for the transcriptional repressor, Sox6, which is repressed by Prdm1 to permit slow muscle differentiation in zebra fish, is not expressed in the mouse myotome. We propose that the lack of functional conservation for mouse Prdm1, that can nevertheless partially rescue the adaxial cells of zebra fish Prdm1 mutants, reflects differences in the evolution of the role of key regulators such as Prdm1 or Sox6, in initiating the onset of the slow muscle program, between teleosts and mammals. Key words: mouse, embryonic myogenesis, myotome, Prdm1, slow muscle gene expression, myogenic regulatory factors. Introduction Cellular processes such as tissue differentiation are coordinated by the controlled expression of regulatory proteins that trigger specific gene expression programs. Implementation of a program depends on both transcriptional activators and also on transcriptional repressors. The Prdm1 gene encodes a transcriptional repressor, containing five zinc fingers and a PR/SET domain, that binds to a specific DNA sequence and recruits co-repressors such as class I histone deacetylases, Groucho family members or histone methyl transferases (for a review, see Bikoff et al. 2009). Prdm1 was first identified as a master gene controlling the terminal differentiation of B lymphocytes into plasmocytes (Turner et al. 1994). In this context, Prdm1 regulates the expression of hundreds of genes (Shaffer et al. 2002; Sciammas and Davis 2004), by direct repression of targets, such as c-Myc (Lin et al. 1997), and by indirect activation of secondary target genes via the repression of the gene coding for Pax5, that is, itself, a repressor (Lin et al. 2002). Interestingly, Prdm1 is dynamically expressed during mouse embryonic development (de Souza et al. 1999; Chang et al. 2002; Vincent et al. 2005) and has been shown to play major roles in a number of differentiation programs. It is the earliest marker of primordial germ cells (Ohinata et al. 2005; Vincent et al. 2005). In Prdm1 mutants, these cells are not specified correctly (Vincent et al. 2005) and those that can be detected show failure to repress the somatic program (Ohinata et al. 2005). Prdm1 is also expressed in the endoderm, ectoderm, and mesoderm of the branchial arches and is required for their development (Vincent et al. 2005) and for normal heart morphogenesis, reflecting its expression in the Second Heart Field (Vincent et al. 2005; Robertson et al. 2007). In limb buds, Prdm1 is expressed in the posterior domain (Vincent et al. 2005) © The Author 2012. Published by Oxford University Press on behalf of the Society for Molecular Biology and Evolution. All rights reserved. For permissions, please e-mail: [email protected] Mol. Biol. Evol. 29(10):3181–3191. 2012 doi:10.1093/molbev/mss125 Advance Access publication April 20, 2012 3181 Research article Abstract MBE Vincent et al. · doi:10.1093/molbev/mss125 and is required for the maintenance of the zone of polarizing activity cells that express Sonic hedgehog, an essential signaling molecule during limb development (Robertson et al. 2007). The embryonic expression of Prdm1 is mainly conserved among vertebrates, as illustrated by branchial arch and limb/fin bud patterns (Chang et al. 2002; Baxendale et al. 2004; Vincent et al. 2005; Wilm and Solnica-Krezel 2005). During the early formation of skeletal muscle, Prdm1 expression has been reported in the myotomal compartment of somites of mouse and Xenopus and in the adaxial myogenic cells of zebra fish somites (de Souza et al. 1999; Chang et al. 2002; Baxendale et al. 2004; Vincent et al. 2005; Wilm and Solnica-Krezel 2005). A recent report on the chick embryo has also shown Prdm1 expression in myogenic cells, in keeping with the conservation of its expression in the muscle lineage, with only subtle differences between different species (Beermann et al. 2010). Moreover, the Prdm1 protein is also very well conserved among vertebrates. It has been estimated that the Prdm1 proteins of Xenopus and mouse share 70% of overall identity (de Souza et al. 1999). The conservation of the expression pattern and of the protein are in favor of a conserved role during evolution. Indeed, the mouse Prdm1 protein can rescue the phenotype of a zebra fish Prdm1 mutant (Ubo) (Liew et al. 2008). In zebra fish, Prdm1 is expressed in adaxial cells (Baxendale et al. 2004) that are fated to give rise to slow muscle (Devoto et al. 1996). This pathway is regulated by Hedgehog signaling from the adjacent axial structures (notochord, ventral neural tube) (Blagden et al. 1997). Ubo (Prdm1) mutants are characterized by U-shaped somites (van Eeden et al. 1996) and lack of slow-twitch fibers (Roy et al. 2001). Detailed analysis of the phenotype has shown that adaxial cells still differentiate into muscle but fail to activate the slow program and express fast muscle markers (Roy et al. 2001). Prdm1, acting downstream of Hedgehog signaling (Baxendale et al. 2004), has been shown to directly repress some fast muscle genes (Liew et al. 2008; von Hofsten et al. 2008). It also promotes the onset of the slow program by repressing the expression of the transcriptional repressor Sox6 (von Hofsten et al. 2008) through an indirect mechanism (Wang et al. 2011). Sox6, in zebra fish, directly represses the expression of slow muscle genes. Interestingly, Sox6 mouse mutants are characterized by the failure to repress the expression of the slow myosin heavy chain in fetal fibers, indicating that in the mouse Sox6 also acts as a repressor of the slow program at later stages (Hagiwara et al. 2005, 2007). In the mouse embryo, it has been shown recently that Six1 and Six4 homeodomain transcription factors, that regulate muscle progenitor cell behavior at the level of Pax3 and the myogenic regulatory factors, are required for the correct expression of fast muscle genes in the myotome (Niro et al. 2009). In the absence of Six1 and Six4, the slow program is normally activated, but fast genes fail to be expressed. In the mouse myotome, unlike the zebra fish, distinct fast and slow muscle cells are not initially specified and differentiated muscle cells express both fast and slow muscle markers (Niro et al. 2009; this study). 3182 Contrary to the fast muscle program, it is not clear what controls the onset of the slow program in Amniotes, but Sonic hedgehog (Shh) plays a role. Shh promotes terminal differentiation of slow muscle fibers in chick embryos and Shh mouse mutant embryos, which have defective myotomes (Borycki et al. 1999; Kruger et al. 2001), lack slow myosin heavy chain expression in the remaining differentiated muscle cells (Li et al. 2004). Given its function in slow myogenesis and dependence on Hedgehog signaling in zebra fish, and because of its expression pattern in the mouse myotome, Prdm1 is a potentially good candidate to control the onset of the slow muscle program in the mouse embryo. In this study, we analyzed in detail the expression pattern of Prdm1 and show that it is expressed in the first differentiating myocytes of the early myotome. As in the zebra fish embryo, this expression is dependent on Hh signaling. However, contrary to the zebra fish embryo, mutant analysis shows that Sox6 expression is not controlled by Prdm1 and slow as well as fast muscle markers continue to be expressed in the myotome. In conclusion, our results demonstrate that despite the conserved biochemical properties of Prdm1 between mouse and zebra fish, Prdm1 function in the control of the slow muscle program is not conserved during vertebrate radiation. Furthermore, this lack of functional conservation for Prdm1 is explained, at least in part, by further evolutionary differences in slow program regulation between zebra fish and mouse, with respect to other important regulators, such as Sox6. Materials and Methods Generation of the pCAG Prdm1b Vector and pCAG CAT 1b8 Transgenic Line The coding sequence of the murine truncated form of Prdm1 (Prdm1b) (Gyory et al. 2003) cloned by rapid amplification of cDNA-ends by polymerase chain reaction (RACE PCR) (Vincent SD, unpublished data) linked to an IRES green fluorescent protein (GFP) was cloned into the conditional pCAG CAT vector (kindly provided by Y. Saga, National Institute of Genetics, Japan) (Watanabe et al. 2006). Efficient conditional expression of the Prdm1b protein was checked by co-transfection in C2C12 cells with a plasmid expressing Cre followed by green fluorescent protein (GFP) expression and western blot analysis. Linearized vector was microinjected by the Centre d’Ingénierie Génétique Murine platform (Institut Pasteur, France) to generate transgenic mouse lines. Three independent lines were established, only one of which showed bright GFP expression after Cre recombination. This line, Tg(pCAG CAT 1b8), was employed in this study, using the Myf5Cre/þ line. Mouse Strains and Genotyping The Blimp1-mEGFP, Mrf4 (Myf6tm1Eno), Myf5Cre, Myf5lox, Myf5nlacZ, Myod1, Pax3Cre, Prdm1BEH, Prdm1CA, Shh, Smo alleles have been described elsewhere (Rudnicki et al. 1992; Zhang et al. 1995; Tajbakhsh, Bober, et al. 1996; Tallquist et al. 2000; Zhang et al. 2001; Shapiro-Shelef Prdm1 Function Is Not Conserved During Myogenesis · doi:10.1093/molbev/mss125 et al. 2003; Kassar-Duchossoy et al. 2004; Engleka et al. 2005; Ohinata et al. 2005; Vincent et al. 2005). The Myog (Myogenin) allele was obtained from the Myogflox allele (Knapp et al. 2006) by crossing with the PGKCre transgenic line (Lallemand et al. 1998). Mouse care and procedures were in accordance with institutional and national guidelines. Embryonic day (E) 0.5 was counted from the appearance of a vaginal plug. Whole-Mount In Situ Hybridization and Immunolocalization Whole-mount in situ hybridization was performed according to standard protocols (Nagy et al. 2003). Prdm1 (Vincent et al. 2003) and Myog (Sassoon et al. 1989) probes have been previously described. One of the Sox6 probes was produced from a plasmid template kindly given by N. Hagiwara (University of California, Davis): The template was linearized with NotI and the probe was transcribed using SP6 RNA polymerase (Promega). The second Sox6 probe was transcribed using SP6 RNA polymerase from a template generated by PCR (GCA GCC ACA CGG AGT TGA TG and CAT TTA GGT GAC ACT ATA GTG ATG GTG TGG TCG TTG CC primers). Both probes gave similar results. Embryos were dissected at E10.5 and then fixed for 2 h in 4% paraformaldehyde at 4 °C. After three rinses, embryos were transferred to 15% sucrose/phosphate buffered saline and embedded in 7% gelatin and 15% sucrose. Embryos were sectioned on a Leica cryostat, to give 20-lm thick slices (or 50-lm thick slices for Sox6 and Myog wholemount in situ hybridized embryos). As primary antibodies, a chick polyclonal for GFP (Invitrogen), rabbit polyclonal antibodies; Myogenin (Santa Cruz), and Mrf4 (Invitrogen), mouse monoclonal antibodies; MF20 (all MyHC) (DSHB), Pax3 (DSHB), and Ki67 (BD Biosciences), and a rat monoclonal antibody for Prdm1 (6D3; Santa Cruz) were used. As secondary antibodies, conjugated Alexa 488/546 against the different species (Invitrogen) were used. Immunolocalization was performed as described and analyzed with a Zeiss ApoTome system. Quantitative Reverse Transcription–PCR Analysis Total RNA was prepared from E10.5 and E11.5 embryos. Briefly, individual embryos were staged by counting the somite number and lysed in TRIzol (Invitrogen). After genotyping, total RNA was prepared according to the manufacturer’s protocol (Applied Biosystem). RNA was purified on PureLink RNA mini kit columns (Invitrogen). RNA was then quantified using a nanodrop spectrophotometer. Five hundred nanograms of RNA were retro-transcribed with random hexamers and Superscript II (Invitrogen). Quantitative PCR was performed using the Power SYBR Green PCR Master Mix on a StepOne Plus Real-Time PCR system (Applied Biosystem), with RNA extracted from embryos with the same number of somites. Three mutant and two control embryos with the same somite number were analyzed as triplicates. Primer sequences have been published elsewhere (Niro et al. 2009). Results were standardized to GAPDH and Myog transcripts. Data were analyzed MBE using the Mann–Whitney non parametric test (http:// www.anastats.fr/). Results Prdm1 Is Expressed in Differentiated Myocytes of the Early Myotome We investigated the expression of Prdm1 in somites of the developing mouse embryo. Transcripts are first detectable in more mature anterior somites, from about somite 14, at embryonic day (E) 9.5 (fig. 1A). At E10.5 (fig. 1B), Prdm1 is expressed in all somites except the most posterior, whereas by E11.5, as somite maturation proceeds in an anterior/posterior direction, transcripts are mainly detected in more posterior somites (fig. 1C). Prdm1 therefore has a dynamic expression pattern in maturing somites, in a central band of cells extending from the epaxial to the hypaxial domain at E10.5 and also at E11.5, despite myotomal expansion by this stage. Such expression is characteristic of the early myotome and is very reminiscent of the first wave of expression of Mrf4 (Bober et al. 1991). In order to examine this further, we analyzed Prdm1 expression on sections, using co-immunohistochemistry. Since Prdm1 antibodies are difficult to use (fig. 1I), we employed a transgenic line carrying an engineered BAC that contains the Prdm1 locus, with a membrane bound GFP reporter (Blimp1-mEGFP) (Ohinata et al. 2005). This faithfully recapitulates the expression pattern of Prdm1. Pax3, that marks myogenic progenitor cells in the dermomyotome at E10.5 (Goulding et al. 1991), does not overlap with GFP (fig. 1D), whereas coimmunolocalization is seen with an antibody to striated muscle myosin heavy chains (fig. 1G), that marks differentiated muscle cells in the myotome. No co-localization with Ki67 (fig. 1F), expressed in proliferating cells in the dermomyotome, confirms that Prdm1/GFP is present only in the post-mitotic cells of the myotome. Some overlapping expression is seen with the myogenic differentiation factor, Myogenin, (fig. 1E) and with Mrf4 in the myotome (result not shown). In more immature posterior myotomes at E10.5, Mrf4 is present in centrally located differentiated myocytes which have a characteristic elongated shaped nucleus (fig. 1H), whereas Myogenin is strongly expressed in more peripheral myotomal cells at this stage. Using the Prdm1 antibody to confirm the myotomal location seen with the transgene, we observe nuclear expression of Prdm1 in Mrf4-positive nuclei of myocytes (fig. 1I). Co-immunolocalization in a central subgroup of Mrf4positive cells suggests that Prdm1 is present in fully differentiated cells of the early myotome. Prdm1 Expression Is Restricted to the Early Myotome and Depends on the Myogenic Determination Factors Myf5 and/or Mrf4 Since it has been proposed that Prdm1 expression in the adaxial cells of zebra fish somites does not depend on myogenic regulatory factors (Hinits et al. 2009), we investigated this relationship in the myotome of mouse embryos, on different mutant backgrounds. In embryos that lack 3183 Vincent et al. · doi:10.1093/molbev/mss125 FIG. 1. Prdm1 expression in the early myotome. (A–C) Wholemount in situ hybridization with a Prdm1 probe, showing Prdm1 transcripts in more mature anterior somites, at E9.5 (A), in the central myotome domain of somites, at E10.5 (B) and mainly in this domain of more posterior somites, at E11.5 (C). (D–G) Transverse sections in the interlimb region of Prdm1GFP(Blimp1mEGFP) BAC transgenic embryos, at E10.5, treated for co-immunohistochemistry with a GFP antibody, reflecting Prdm1 expression, and a Pax3 antibody that marks progenitor cells in the dermomyotome (D) or with an antibody directed against all striated muscle myosin heavy chains (MyHC) that marks differentiated skeletal muscle cells (G) shows that membrane bound GFP labeling co-localizes with myosinpositive cells in the myotome. No co-localization is seen with an antibody to Ki67, which marks proliferating Pax3-positive cells (F), whereas there is co-localization with Myogenin (E), in keeping with expression of the Prdm1 transgene in differentiating post-mitotic cells of the myotome. (H and I) Transverse sections of less mature posterior somites, at E10.5, show that Mrf4 is principally expressed in elongated nuclei of myocytes, whereas Myogenin is present in the nuclei of more peripheral myotomal cells (H). Nuclear accumulation of Prdm1 is detected in Mrf4-positive myocytes (I). Myf5 (Myf5lox/lox), the onset of myogenesis is slightly delayed but is initiated by Mrf4, also acting as a myogenic determination factor in this context (Kassar-Duchossoy et al. 2004). In these mutant embryos, Prdm1 is expressed in the myotome, although less strongly than in the control at E10.5 (fig. 2A and B), reflecting the delayed onset of myogenesis. In the absence of the other myogenic determination factor, MyoD, which begins to be expressed at this 3184 MBE FIG. 2. The dependence of Prdm1 expression on the myogenic regulatory factors. (A–I) Whole-mount in situ hybridization with a Prdm1 (A–F, H, and I) or a Myog (Myogenin) probe (G), on control (A and D) or MRF mutant embryos at E10.5 (A–C), E10.75 (D–F), or E11.75 (G–I). In the absence of Myf5 (Myf5lox/lox) (B), Prdm1 transcripts are still detectable (arrowhead in A indicates the central domain of the myotome where Prdm1 is expressed, arrowhead in B indicates the epaxial, and arrow the hypaxial, extent of the myotome); in the absence of MyoD (Myod1/) (C), there is robust Prdm1 expression in the myotome. In the absence of the myogenic differentiation factor, Myogenin (Myog/), Prdm1 transcripts are present (E) and are still detectable, although reduced in the absence of Mrf4 (Mrf4/) (F). In Myf5nlacZ/nlacZ embryos (G–I), in the absence of both Myf5 and Mrf4, when Myog (Myogenin) transcripts have begun to accumulate in the myotome (G), Prdm1 expression is not detectable (H, with interlimb somites enlarged in I). time, Prdm1 expression is unaffected (fig. 2C). Prdm1 is also expressed in the myotome of Myogenin (Myog) (fig. 2E) and Mrf4 (fig. 2F) mutant embryos. In the Myf5nlacZ/nlacZ mutant, both Myf5 and Mrf4 are absent (Kassar-Duchossoy et al. 2004) and cells do not enter the myogenic program until Myod1 is expressed independently, from about E11 (Tajbakhsh, Rocancourt, et al. 1996). Under these conditions, the myotome is initially absent and the myogenic differentiation gene, Myog, is only expressed after Myod1 activation. As shown in figure 2G, at E11.75 myotome formation is rescued. However, Prdm1 transcripts are not detectable (fig. 2H and I), showing that Prdm1 is not expressed in the MyoD-dependent myogenic program, despite the presence of Myogenin. In the double Mrf4/Myf5 mutant (Myf5nlacZ/nlacZ), undifferentiated cells that have activated the Myf5 gene (visualized by nlacZ expression) mislocate or accummulate adjacent to the edges of the dermomyotome (Tajbakhsh, Rocancourt, et al. 1996). Prdm1 Function Is Not Conserved During Myogenesis · doi:10.1093/molbev/mss125 MBE FIG. 3. Prdm1 is expressed in the absence of Shh but not detected in the absence of Hedgehog signaling. (A–D) Whole-mount in situ hybridization for Prdm1 transcripts on wild-type (A) and Shh/ mutant (B–D) embryos at E10 (A and B) and E10.75 (C and D). Prdm1 expression in the myotome (arrowhead) is maintained in the remaining myotomal cells of Shh/ mutant embryos (B and C, with magnification of C in D). (E–H) Whole-mount in situ hybridization for Myf5 (E and F) and Prdm1 (G and H) transcripts on wild-type (E and G) and Smo/ mutant (F and H) embryos at E9.5. In the absence of Hedgehog signaling, Myf5 is still transcribed in the somites with a characteristic segmented pattern (arrowhead), contrary to Prdm1, which does not show any somitic expression. As in Shh/ mutant embryos, the Prdm1 expression pattern is still observed in the limb buds (asterisk), gut endoderm and branchial arch region (bracket) in Smo/ mutant embryos. Prdm1 is not expressed in these cells. Prdm1 expression in the mouse embryo therefore depends on the formation of the early myotome and hence on Mrf4 or Myf5. Consistent with the relationship between Prdm1 expression and Myf5/ Mrf4 regulation of the onset of myogenesis, later expression of Prdm1 at E11.5 is confined to a narrow band of cells in the central domain of the myotome (fig. 1C), which by this stage has expanded due to the on-going contribution of MyoD-positive myogenic cells which do not express Prdm1 (fig. 2H and I). MyoD normally takes over the role of myogenic determination during later embryonic development, when Myf5 is downregulated. Consistent with the dependence of Prdm1 expression on Myf5/Mrf4, it is no longer expressed in skeletal muscle at later developmental stages nor in the adult myofibers (data not shown). As Prdm1 expression is associated only with the formation of the early myotome, and it is not re-expressed later during muscle development, contrary to Mrf4, this suggests that it is also controlled by environmental and temporal cues. Prdm1 Expression in the Myotome Requires Hedgehog Signaling Sonic hedgehog (Shh) coming from the notochord and the floor plate of the neural tube is one of the crucial signals implicated in the onset of myogenesis (Munsterberg et al. 1995). During myogenesis in the zebra fish embryo, abrogation of Hedgehog signaling results in the absence of Prdm1 expression in adaxial cells of the myotome (Baxendale et al. 2004). In order to see whether Shh signaling affects Prdm1 expression in the myotome of the mouse embryo, we analyzed Shh/ mutant embryos. In these mutants, the somites are severely affected and the initiation of myogenesis is impaired. However, some cells still activate the myogenic determination gene, Myf5, independently of Shh, leading to partial myotomal muscle formation (Borycki et al. 1999; Kruger et al. 2001; Zhang et al. 2001). In situ hybridization for Prdm1 transcripts in Shh/ mutant embryos shows that these are still detectable in anterior somites at E10 (fig. 3A and B) and in interlimb somites at E10.75 (fig. 3C and D), despite perturbations due to the lack of Shh signaling. In other domains, such as in the branchial arches or early limb buds (fig. 1A and B and fig. 3A), Prdm1 expression is not affected. However, in Shh/ mutant embryos, the Shh target Ptc1 is still detected, suggesting that Indian hedgehog (Ihh) is able to functionally compensate (Zhang et al. 2001). In the somites of Smoothened mutants (Smo), in which all Hedgehog signaling is absent, Myf5 expression is observed as in Shh/ mutant embryos (Zhang et al. 2001). At E9.5, Myf5 is still expressed in the absence of Hedgehog signaling (fig. 3E and F); however, we failed to detect Prdm1 expression in the somites despite normal expression in the other embryonic domains (fig. 3G and H). These results suggest that, as in the zebra fish, Hedgehog signaling is required for Prdm1 expression, in cells already engaged in the myogenic program as a result of Myf5/Mrf4 expression in the mouse embryo. Normal Onset of Myogenesis and of Slow Myosin Heavy Chain Expression in the Absence of Prdm1 In order to investigate a potential role for Prdm1 at the onset of myogenesis in the mouse myotome, we 3185 Vincent et al. · doi:10.1093/molbev/mss125 MBE FIG. 4. Expression of slow and fast embryonic myosin heavy chains in the myotomes of mouse embryos in the presence and absence of Prdm1. (A–L) Co-immunolocalization with antibodies to myogenic differentiation factors Mrf4 (A and B) and Myogenin (C–E, H–J) and to all striated myosin heavy chains (MyHC) (A, B, C, F, G, H, K, and L), slow MyHC (D, F, I, and K), and fast embryonic MyHC (embMyHC) (E, G, J, and L) on transverse sections of control (A and C–G) and conditional Myf5Cre/þ;Prdm1CA/CA mutant (B) or Pax3Cre/þ;Prdm1CA/CA mutant (H–L) embryos at E10.25 in less mature posterior somites (A and B) or interlimb somites at E10.5 (C–L), showing co-localization of fast and slow myosin isoforms in Mrf4-positive and in some Myogenin-positive cells in the myotome, in the presence (A and C–G) or absence (B and H–L) of Prdm1. investigated the expression of myosin heavy chain isoforms. Prdm1 null mutant embryos (Prdm1BEH/BEH) die around E10.5 because of placental defects (Vincent et al. 2005; Robertson et al. 2007); however, some embryos are viable at this stage. In these embryos, although somite development is retarded and the morphology of the myotome is abnormal, both slow and embryonic fast myosin heavy chains are expressed (supplementary fig. 1, Supplementary Material online). In order to overcome the embryonic lethality of the null mutation, we used a conditional Prdm1 line (Shapiro-Shelef et al. 2003), crossed onto a Myf5Cre/þ (Tallquist et al. 2000) or a Pax3Cre/þ (Engleka et al. 2005) genetic background to delete Prdm1 in myogenic cells or myogenic progenitor cells, respectively. Both Cre lines led to efficient deletion of Prdm1 in the myotome (results not shown) and gave similar results. As shown in figure 4A and B and supplementary figure 2 (Supplementary Material online), the formation of the early myotome is not affected in the absence of Prdm1, with myosin heavy chains expressed in Mrf4-positive cells. Both slow myosin heavy chain and fast embryonic myosin heavy chain are present in most myosin-positive differentiated cells of the myotome, some of which are also Myogenin-positive at E10.5 (fig. 4C–G). In the conditional Prdm1 mutant, differentiation of cells in 3186 the myotome is normal, with expression of both slow (fig. 4I and K) and fast embryonic (fig. 4J and L) myosin heavy chains. Furthermore, conditional deletion of Prdm1 with the Myf5Cre allele on different Myf5, Mrf4 or Myog mutant backgrounds did not affect the expression of slow, or fast embryonic myosin heavy chains in the myotome (results not shown). In supplementary figure 3 (Supplementary Material online), this is shown with a Myf5Cre/lox;Prdm1CA/BEH cross in which Myf5 is absent and Mrf4, produced by the Myf5lox allele, is responsible for early myotome formation. We therefore conclude that, unlike the zebra fish, Prdm1 is not required for the slow skeletal muscle program in the mouse myotome. This is further supported by reverse transcription– quantitative polymerase chain reaction (RT-qPCR) analysis (fig. 5), which shows that, although there are some minor differences with the control (statistically not significant), a range of slow as well as fast skeletal muscle genes are transcribed in the myotome of conditional Prdm1 mutant embryos. Expression of a Dominant Negative Truncated Prdm1 in the Myotome Does Not Affect Myogenesis We cannot exclude that the absence of phenotype in the Prdm1 conditional mutant embryos is the result of a Prdm1 Function Is Not Conserved During Myogenesis · doi:10.1093/molbev/mss125 MBE to the myotome since when this transgene is expressed in heart progenitors, it has a strong phenotype (Vincent SD, Buckingham M, unpublished data). We conclude that the slow and fast programs are not under the control of Prdm1 in the mouse myotome. Sox6, a Repressor of the Slow Muscle Program in Zebra Fish, Is Not Expressed in the Myotome and Is Not Repressed by Prdm1 in the Mouse Myotome FIG. 5. Analysis of transcripts of slow and fast skeletal muscle genes expressed in the myotome in the absence of Prdm1. Quantitative PCR analysis of transcripts, expressed relative to Gapdh and Myog transcripts taken as 1, of Myog (Myogenin), Pax3, and slow muscle genes, slow myosin heavy chain (slow MyHC), slow myosin light chain 4 (Myl4), slow myosin light chain 6b (Myl6B), TroponinT1 (TnnT1), ATPase, Caþþ transporting, cardiac muscle, slow twitch 2 (Atp2a2), or fast muscle genes, myosin light chain one (Myl1), phosphorylatable myosin light chain (Mylpf), or ATPase, Caþþ transporting, cardiac muscle, fast twitch 1 (Atp2a1), shows that these are expressed in both control (Myf5Cre/þ;Prdm1CA/þ, n 5 2, triplicates) and Prdm1 conditional mutant (Myf5Cre/þ; Prdm1CA/CA, n 5 3, triplicates) embryos, at E11.5. There is no statistically significant difference in the expression of these markers between mutant and control embryos (ns, not significant; brackets indicate the standard error of the mean). functional compensation. Whole-mount in situ hybridization of several members of the Prdm family did not reveal any obvious expression in the myotome (data not shown). In order to indirectly assess this question, we constructed a transgenic line Tg(pCAG CAT 1b8) conditionally expressing a truncated form of Prdm1 (Prdm1b) and GFP. This truncated form does not have a PR/SET domain and displays dramatically reduced repressive activity, although still able to bind to its DNA targets (Gyory et al. 2003), therefore, it should interfere with the endogenous full length Prdm1 (Angelin-Duclos et al. 2002). The endogenous transcript encoding the Prdm1b protein, which could be potentially generated from an internal promoter as seen in human myeloma cell lines (Gyory et al. 2003), is not expressed in the embryo at the time of myotome formation (Vincent SD, unpublished data). As shown in figure 6A and B, expression of Prdm1b depends on the expression of Cre. Embryos carrying the transgene and one copy of the Myf5Cre allele display the expected GFP expression profile in the somite, although this is mosaic (fig. 6C). Expression of Pax3 in the dermomyotome is not affected in cells that express the transgene (fig. 6D). Expression of myosin heavy chain (all MyHC, slow MyHC and emb MyHC) in cells of the myotome is also not perturbed by the expression of the transgene (fig. 6E–G), suggesting that interference with the repressive activity of Prdm1 does not affect the expression of the slow or fast programs in the mouse myotome. The absence of a phenotype is specific A key repressor of the slow muscle program in zebra fish is Sox6, which is transcriptionally repressed by Prdm1 (von Hofsten et al. 2008). In the mouse also, Sox6 has been shown to inhibit slow muscle myosin heavy chain expression at fetal stages by directly binding to the promoter region of this gene (Hagiwara et al. 2007). At E17.5, Sox6 accumulates in the myonuclei of secondary fibers (data not shown). In order to assess whether Sox6 has a role in the control of the slow muscle program in the mouse myotome, we performed whole-mount in situ hybridization at E10.75. Sox6 transcripts are not detectable in the myotome marked by Myogenin (Myog) expression (fig. 7A and B, sections in E and F, respectively). In immature caudal somites, where the myotome has not yet formed, Sox6 is transcribed in the ventral sclerotome that will give rise to cartilage and bone (fig. 7A and B, sections in I and J, respectively). This dynamic expression pattern in the somites is not the result of Prdm1 repressor activity since in the conditional Prdm1 mutants, the Sox6 expression pattern is not affected (fig. 7C and D section in G and K and H and L, respectively). These results demonstrate that in the mouse embryo, contrary to the zebra fish (Wang et al. 2011), Sox6 is not detected in the myotome. This suggest first, that, in the mouse myotome, Prdm1 expression does not impact on Sox6 expression, and second, that Sox6 repression of slow muscle gene expression may not be relevant to the initial stages of mouse myogenesis, as suggested by the absence of an early myogenic phenotype in Sox6 null mutant embryos (Hagiwara et al. 2005, 2007). Discussion We have investigated the expression and role of Prdm1 at the onset of skeletal muscle differentiation in the mouse, relative to its important function in promoting slow muscle formation in the zebra fish embryo. First, we show that Prdm1 is expressed in the differentiated cells of the early mouse myotome, where the myogenic factor, Mrf4, is also present. Consistent with its myogenic expression, Prdm1 activation depends on Myf5 and/or Mrf4, which are required for early myotome formation. In the absence of these factors, myogenic progenitor cells do not differentiate and are mislocated (Tajbakhsh, Rocancourt, et al. 1996). Prdm1 is not expressed in these cells. In contrast, in the zebra fish embryo where the onset of myogenesis in adaxial cells depends on Myf5 and/or MyoD, in Myf5 mutants, in which Myod1 expression has been downregulated with morpholinos, Prdm1 is still expressed, in the absence of 3187 Vincent et al. · doi:10.1093/molbev/mss125 MBE FIG. 6. Overexpression of Prdm1b does not affect myogenesis. (A and B) Co-transfection of C2C12 cells with the pCAG CAT 1b8 IRES GFP together with the Cre expression pMC13 plasmids. After Cre expression, GFP (A) and expression of the truncated protein Prdm1b as shown by western blot (B), are detected. (C) The GFP expression pattern of a Myf5Cre/þ;Tg(pCAGCAT1b8/þ) embryo at E11 marks the myotome (C1,2). A longitudinal section indicates that the expression of the transgene is mosaic in the tail somites (C3). (D–G) Co-immunolocalization on transverse sections of Myf5Cre/þ;Tg(pCAGCAT1b8/þ) embryos at E10.5 using antibody to GFP and Pax3 (D), all MyHCs (E), slow MyHC (F), and fast embryonic (emb) MyHC (G) indicating that expression of these markers in the somites is not affected by the expression of the transgene. Arrows point to examples of cells that express the transgene. muscle formation (Hinits et al. 2009). The Myf5/Mrf4 mutant analysis indicates that, in the mouse embryo, these myogenic factors are either directly required or that the location of cells in respect to the signaling environment of the myotome is essential for Prdm1 activation. Later specification of myogenic cells that contribute to the formation of skeletal muscle in the myotome depends on MyoD, in the mouse embryo. Prdm1 activation is MyoD-independent and its downregulation from E11.5 is in accordance with its activation during the earlier wave of myogenesis in the myotome. This is further supported by the analysis of Six1/;Six4/ mouse mutant embryos (Grifone et al. 2005). In the absence of these two important upstream myogenic regulators, Prdm1 is still expressed in the residual early myotome formed as a result of Myf5 activation (Niro et al. 2009). In this case, Mrf4 expression is not detected and myogenic differentiation depends on low levels of Myogenin (Grifone et al. 2005). In these double mutants, Prdm1 expression correlates with the expression of Myf5 during early myotome formation. In the zebra fish embryo, Prdm1 activation depends on Hedgehog signaling to adaxial cells from the adjacent notochord and floor plate of the neural tube (Baxendale et al. 2004). In Shh/ mouse mutants, however, Prdm1 activation 3188 is still observed in the remaining myogenic cells. It is possible that other Hedgehog signaling pathways are important for activation, in particular, Ihh has been shown to be responsible for the Hedgehog signaling in Shh/ mutant embryos (Zhang et al. 2001). However, in the Smoothened mutant, in which all Hedgehog signaling is abrogated, Myf5 expression is still induced as in the Shh/ mutant (Zhang et al. 2001). However, we do not detect somitic expression of Prdm1 in Smo/ mutant embryos. Hedgehog-dependent activation of Prdm1 is, therefore, comparable to zebra fish. If one makes the analogy between adaxial cells and the epaxial cells of the early mouse somite, then Myf5 activation and subsequent early epaxial myogenesis, with accompanying Prdm1 expression in the differentiated epaxial myotome, are mainly Shh dependent in the mouse embryo also, as indicated by Shh and Gli2/3 mutant phenotypes (Borycki et al. 1999; McDermott et al. 2005) and the importance of the Gli binding site for activation of the early epaxial enhancer of Myf5 (Gustafsson et al. 2002; Borello et al. 2006). Early hypaxial myogenesis is also affected by the absence of Shh (Kruger et al. 2001); however, some skeletal muscle formation takes place (Borycki et al. 1999) and in these myotomal cells, Prdm1 continues to be expressed, probably under the control of Ihh. A Gli binding region has been Prdm1 Function Is Not Conserved During Myogenesis · doi:10.1093/molbev/mss125 MBE FIG. 7. Sox6 is not expressed in the myotome of normal and conditional Prdm1 mutant embryos. (A and C) Sox6 whole-mount in situ hybridization of control (A, transverse sections in E, I with plane of section indicated in dotted red line in A) or conditional Prdm1 mutant (Myf5Cre/þ;Prdm1CA/CA) (C, transverse sections in G and K with plane of section indicated in dotted red line in B) embryos at E10.75. (B and D). Whole-mount in situ hybridization for Myogenin (Myog) transcripts of control (B, transverse sections in F and J) or conditional Prdm1 mutant (Myf5Cre/þ;Prdm1CA/CA) embryos (D, transverse sections in H and L). No expression of Prdm1 is detected (with two independent probes) in the myotome (My) as shown on section (F), only rostro-caudal lips of the dermomyotome (Dm) display weak Sox6 expression. Faint expression is detected in the posterior dorsal root ganglia (A and C arrowhead). In the newly formed somites (I), Sox6 expression is detected in the notochord (arrowhead) and in the sclerotome (asterisk), in the absence of the myotome that has not yet begun to form, as shown by the lack of Myogenin expression (J). In the absence of Prdm1, the expression pattern of Sox6 is not affected (C, G, and K). identified in the 5# of the Prdm1 locus by chromatin immunoprecipitation from E11.5 limb buds (Vokes et al. 2008). Prdm1 expression in the limb buds is not affected in Smo/ mutant embryos, suggesting that this region is not active in the limb buds, but could be responsible of its expression in the somites. In mouse, in comparison with zebra fish, the single early myotomal compartment expresses both slow and fast muscle genes within the same cells. In zebra fish, initial expression domains of smyhc1 (slow) and myhc4 (fast) show some overlap in the adaxial cell region at the 10 somite stage (Bryson-Richardson et al. 2005). Expression of myhc4 is much weaker and its domain more restricted compared with smyhc1 expression, suggesting that this co-expression is only transient, due to the subsequent activity of Prdm1. In the mouse embryo, however, divergence of slow and fast muscle phenotypes occurs later, after the onset of innervation during fetal development. In zebra fish, as in Xenopus, embryos innervation occurs much earlier in the myotome (Blackshaw and Warner 1976; Westerfield et al. 1986), in response to the requirement for rapid muscular movement, mainly mediated by myotomal muscle, already, at the larval/tadpole stage. In contrast, during mammalian development, there is no such early deployment of muscular activity (Deries et al. 2008). Perhaps, as a consequence of this premature requirement for efficient fast/slow muscle function, the zebra fish embryo has implemented a different regulatory strategy to that operational in the mammalian myotome where fast and slow muscle genes are co-expressed. Contrary to the situation in zebra fish, Sox6, that encodes a repressor of the slow program, is not expressed in the mouse myotome and is not de-repressed in the absence of the murine Prdm1. Moreover, in the mouse myotome, Prdm1 does not act as a repressor of the fast muscle program. Therefore, regulatory factors, that in the zebra fish promote the slow skeletal muscle program, are not functional in this context in the mouse embryo. Mouse Prdm1 has conserved a functional role, as indicated by its capacity to replace the zebra fish protein during a limited time window in adaxial cells of the 3189 Vincent et al. · doi:10.1093/molbev/mss125 Ubo mutant (Liew et al. 2008). It remains to be seen whether Prdm1, which can therefore potentially function as a repressor in the mouse myotome, actually has a role. This was not revealed by our analysis of conditional Prdm1 myogenic mutant embryos or of transgenic embryos expressing a dominant negative form of Prdm1. It also remains unclear how the slow skeletal muscle genes are activated in the early mouse myotome. Neither Prdm1 nor Sox6 appear to be implicated. Our study clearly indicates divergences in Prdm1 regulation and function during early myogenesis in the zebra fish compared with the mouse embryo. Supplementary Material Supplementary figures 1–3 are available at Molecular Biology and Evolution online (http://www.mbe.oxfordjournals.org/). Acknowledgments We thank S. Tajbakhsh, Y. Lallemand, and Y. Ohinata for reagents and Emmanuel Pecnard, Sabrina Coqueran, and Catherine Bodin for technical help. We are grateful to Kathryn Calame for generously giving us the conditional Prdm1 mouse line. We thank Emmanuel Perret and Pascal Roux (Imagopole, Institut Pasteur, Paris) for advice on the imaging. We thank the Centre d’Ingénierie Génétique Murine (CIGM) plateform (Institut Pasteur, Paris) for micronuclear injections to make transgenic lines. We especially thank Pascal Maire and Phil Ingham for helpful discussion. Work in MB’s laboratory on Prdm1 was funded by the Institut Pasteur and the Centre National de la Recherche Scientifique (CNRS) with grants from the Association Franc xaise contre les Myopathies (AFM), the European Union consortia Cells into Organs (NoE, 6th framework programme) and CardioCell (7th framework programme). S.D.V. is an Institut National de la Santé et de la Recherche Médicale (INSERM) investigator. References Angelin-Duclos C, Johnson K, Liao J, Lin KI, Calame K. 2002. An interfering form of Blimp-1 increases IgM secreting plasma cells and blocks maturation of peripheral B cells. Eur J Immunol. 32:3765–3775. Baxendale S, Davison C, Muxworthy C, Wolff C, Ingham PW, Roy S. 2004. The B-cell maturation factor Blimp-1 specifies vertebrate slow-twitch muscle fiber identity in response to Hedgehog signaling. Nat Genet. 36:88–93. Beermann ML, Ardelt M, Girgenrath M, Miller JB. 2010. Prdm1 (Blimp1) and the expression of fast and slow myosin heavy chain isoforms during avian myogenesis in vitro. PLoS One 5:e9951. Bikoff EK, Morgan MA, Robertson EJ. 2009. An expanding job description for Blimp-1/PRDM1. Curr Opin Genet Dev. 19:379–385. Blackshaw S, Warner A. 1976. Onset of acetylcholine sensitivity and endplate activity in developing myotome muscles of Xenopus. Nature 262:217–218. Blagden CS, Currie PD, Ingham PW, Hughes SM. 1997. Notochord induction of zebrafish slow muscle mediated by Sonic hedgehog. Genes Dev. 11:2163–2175. Bober E, Lyons GE, Braun T, Cossu G, Buckingham M, Arnold HH. 1991. The muscle regulatory gene, Myf-6, has a biphasic pattern of 3190 MBE expression during early mouse development. J Cell Biol. 113:1255–1265. Borello U, Berarducci B, Murphy P, Bajard L, Buffa V, Piccolo S, Buckingham M, Cossu G. 2006. The Wnt/beta-catenin pathway regulates Gli-mediated Myf5 expression during somitogenesis. Development 133:3723–3732. Borycki AG, Brunk B, Tajbakhsh S, Buckingham M, Chiang C, Emerson CP Jr. 1999. Sonic hedgehog controls epaxial muscle determination through Myf5 activation. Development 126:4053–4063. Bryson-Richardson RJ, Daggett DF, Cortes F, Neyt C, Keenan DG, Currie PD. 2005. Myosin heavy chain expression in zebrafish and slow muscle composition. Dev Dyn. 233:1018–1022. Chang DH, Cattoretti G, Calame KL. 2002. The dynamic expression pattern of B lymphocyte induced maturation protein-1 (Blimp-1) during mouse embryonic development. Mech Dev. 117:3 05–309. de Souza FS, Gawantka V, Gomez AP, Delius H, Ang SL, Niehrs C. 1999. The zinc finger gene Xblimp1 controls anterior endomesodermal cell fate in Spemann’s organizer. EMBO J. 18:6062–6072. Deries M, Collins JJ, Duxson MJ. 2008. The mammalian myotome: a muscle with no innervation. Evol Dev. 10:746–755. Devoto SH, Melancon E, Eisen JS, Westerfield M. 1996. Identification of separate slow and fast muscle precursor cells in vivo, prior to somite formation. Development 122:3371–3380. Engleka KA, Gitler AD, Zhang M, Zhou DD, High FA, Epstein JA. 2005. Insertion of Cre into the Pax3 locus creates a new allele of Splotch and identifies unexpected Pax3 derivatives. Dev Biol. 280:396–406. Goulding MD, Chalepakis G, Deutsch U, Erselius JR, Gruss P. 1991. Pax-3, a novel murine DNA binding protein expressed during early neurogenesis. EMBO J. 10:1135–1147. Grifone R, Demignon J, Houbron C, Souil E, Niro C, Seller MJ, Hamard G, Maire P. 2005. Six1 and Six4 homeoproteins are required for Pax3 and Mrf expression during myogenesis in the mouse embryo. Development 132:2235–2249. Gustafsson MK, Pan H, Pinney DF, Liu Y, Lewandowski A, Epstein DJ, Emerson CP Jr. 2002. Myf5 is a direct target of long-range Shh signaling and Gli regulation for muscle specification. Genes Dev. 16:114–126. Gyory I, Fejer G, Ghosh N, Seto E, Wright KL. 2003. Identification of a functionally impaired positive regulatory domain I binding factor 1 transcription repressor in myeloma cell lines. J Immunol. 170:3125–3133. Hagiwara N, Ma B, Ly A. 2005. Slow and fast fiber isoform gene expression is systematically altered in skeletal muscle of the Sox6 mutant, p100H. Dev Dyn. 234:301–311. Hagiwara N, Yeh M, Liu A. 2007. Sox6 is required for normal fiber type differentiation of fetal skeletal muscle in mice. Dev Dyn. 236:2062–2076. Hinits Y, Osborn DP, Hughes SM. 2009. Differential requirements for myogenic regulatory factors distinguish medial and lateral somitic, cranial and fin muscle fibre populations. Development 136:403–414. Kassar-Duchossoy L, Gayraud-Morel B, Gomes D, Rocancourt D, Buckingham M, Shinin V, Tajbakhsh S. 2004. Mrf4 determines skeletal muscle identity in Myf5:Myod double-mutant mice. Nature 431:466–471. Knapp JR, Davie JK, Myer A, Meadows E, Olson EN, Klein WH. 2006. Loss of myogenin in postnatal life leads to normal skeletal muscle but reduced body size. Development 133:601–610. Kruger M, Mennerich D, Fees S, Schafer R, Mundlos S, Braun T. 2001. Sonic hedgehog is a survival factor for hypaxial muscles during mouse development. Development 128:743–752. Lallemand Y, Luria V, Haffner-Krausz R, Lonai P. 1998. Maternally expressed PGK-Cre transgene as a tool for early and uniform Prdm1 Function Is Not Conserved During Myogenesis · doi:10.1093/molbev/mss125 activation of the Cre site-specific recombinase. Transgenic Res. 7:105–112. Li X, Blagden CS, Bildsoe H, Bonnin MA, Duprez D, Hughes SM. 2004. Hedgehog can drive terminal differentiation of amniote slow skeletal muscle. BMC Dev Biol. 4:9. Liew HP, Choksi SP, Wong KN, Roy S. 2008. Specification of vertebrate slow-twitch muscle fiber fate by the transcriptional regulator Blimp1. Dev Biol. 324:226–235. Lin KI, Angelin-Duclos C, Kuo TC, Calame K. 2002. Blimp-1dependent repression of Pax-5 is required for differentiation of B cells to immunoglobulin M-secreting plasma cells. Mol Cell Biol. 22:4771–4780. Lin Y, Wong K, Calame K. 1997. Repression of c-myc transcription by Blimp-1, an inducer of terminal B cell differentiation. Science 276:596–599. McDermott A, Gustafsson M, Elsam T, Hui CC, Emerson CP Jr, Borycki AG. 2005. Gli2 and Gli3 have redundant and contextdependent function in skeletal muscle formation. Development 132:345–357. Munsterberg AE, Kitajewski J, Bumcrot DA, McMahon AP, Lassar AB. 1995. Combinatorial signaling by Sonic hedgehog and Wnt family members induces myogenic bHLH gene expression in the somite. Genes Dev. 9:2911–2922. Nagy A, Gertenstein M, Vintersten K, Behringer RR. 2003. Manipulating the mouse embryo: a laboratory manual. Cold Spring Harbor (NY): Cold Spring Harbor Press. Niro C, Demignon J, Vincent S, Liu Y, Giordani J, Sgarioto N, Favier M, Guillet-Deniau I, Blais A, Maire P. 2009. Six1 and Six4 gene expression is necessary to activate the fast-type muscle gene program in the mouse primary myotome. Dev Biol 338:168–182. Ohinata Y, Payer B, O’Carroll D, et al. 2005. Blimp1 is a critical determinant of the germ cell lineage in mice. Nature 436:207–213. Robertson EJ, Charatsi I, Joyner CJ, et al. 2007. Blimp1 regulates development of the posterior forelimb, caudal pharyngeal arches, heart and sensory vibrissae in mice. Development 134:4335–4345. Roy S, Wolff C, Ingham PW. 2001. The u-boot mutation identifies a Hedgehog-regulated myogenic switch for fiber-type diversification in the zebrafish embryo. Genes Dev. 15:1563–1576. Rudnicki MA, Braun T, Hinuma S, Jaenisch R. 1992. Inactivation of MyoD in mice leads to up-regulation of the myogenic HLH gene Myf-5 and results in apparently normal muscle development. Cell 71:383–390. Sassoon D, Lyons G, Wright WE, Lin V, Lassar A, Weintraub H, Buckingham M. 1989. Expression of two myogenic regulatory factors myogenin and MyoD1 during mouse embryogenesis. Nature 341:303–307. Sciammas R, Davis MM. 2004. Modular nature of Blimp-1 in the regulation of gene expression during B cell maturation. J Immunol. 172:5427–5440. Shaffer AL, Lin KI, Kuo TC, et al. 2002. Blimp-1 orchestrates plasma cell differentiation by extinguishing the mature B cell gene expression program. Immunity 17:51–62. Shapiro-Shelef M, Lin KI, McHeyzer-Williams LJ, Liao J, McHeyzerWilliams MG, Calame K. 2003. Blimp-1 is required for the MBE formation of immunoglobulin secreting plasma cells and preplasma memory B cells. Immunity 19:607–620. Tajbakhsh S, Bober E, Babinet C, Pournin S, Arnold H, Buckingham M. 1996a. Gene targeting the myf-5 locus with nlacZ reveals expression of this myogenic factor in mature skeletal muscle fibres as well as early embryonic muscle. Dev Dyn. 206:291–300. Tajbakhsh S, Rocancourt D, Buckingham M. 1996. Muscle progenitor cells failing to respond to positional cues adopt nonmyogenic fates in myf-5 null mice. Nature 384:266–270. Tallquist MD, Weismann KE, Hellstrom M, Soriano P. 2000. Early myotome specification regulates PDGFA expression and axial skeleton development. Development 127:5059–5070. Turner CA Jr, Mack DH, Davis MM. 1994. Blimp-1, a novel zinc fingercontaining protein that can drive the maturation of B lymphocytes into immunoglobulin-secreting cells. Cell 77:297–306. van Eeden FJ, Granato M, Schach U, et al. 1996. Mutations affecting somite formation and patterning in the zebrafish, Danio rerio. Development 123:153–164. Vincent SD, Dunn NR, Hayashi S, Norris DP, Robertson EJ. 2003. Cell fate decisions within the mouse organizer are governed by graded Nodal signals. Genes Dev. 17:1646–1662. Vincent SD, Dunn NR, Sciammas R, Shapiro-Shalef M, Davis MM, Calame K, Bikoff EK, Robertson EJ. 2005. The zinc finger transcriptional repressor Blimp1/Prdm1 is dispensable for early axis formation but is required for specification of primordial germ cells in the mouse. Development 132:1315–1325. Vokes SA, Ji H, Wong WH, McMahon AP. 2008. A genome-scale analysis of the cis-regulatory circuitry underlying sonic hedgehog-mediated patterning of the mammalian limb. Genes Dev. 22:2651–2663. von Hofsten J, Elworthy S, Gilchrist MJ, Smith JC, Wardle FC, Ingham PW. 2008. Prdm1- and Sox6-mediated transcriptional repression specifies muscle fibre type in the zebrafish embryo. EMBO Rep. 9:683–689. Wang X, Ono Y, Tan SC, Chai RJ, Parkin C, Ingham PW. 2011. Prdm1a and miR-499 act sequentially to restrict Sox6 activity to the fast-twitch muscle lineage in the zebrafish embryo. Development 138:4399–4404. Watanabe Y, Kokubo H, Miyagawa-Tomita S, Endo M, Igarashi K, Aisaki K, Kanno J, Saga Y. 2006. Activation of Notch1 signaling in cardiogenic mesoderm induces abnormal heart morphogenesis in mouse. Development 133:1625–1634. Westerfield M, McMurray JV, Eisen JS. 1986. Identified motoneurons and their innervation of axial muscles in the zebrafish. J Neurosci. 6:2267–2277. Wilm TP, and Solnica-Krezel L. 2005. Essential roles of a zebrafish prdm1/blimp1 homolog in embryo patterning and organogenesis. Development 132:393–404. Zhang W, Behringer RR, Olson EN. 1995. Inactivation of the myogenic bHLH gene MRF4 results in up-regulation of myogenin and rib anomalies. Genes Dev. 9:1388–1399. Zhang XM, Ramalho-Santos M, McMahon AP. 2001. Smoothened mutants reveal redundant roles for Shh and Ihh signaling including regulation of L/R symmetry by the mouse node. Cell 106:781–792. 3191