SEARCH

SEARCH BY CITATION

Keywords:

  • myosin heavy chain gene;
  • muscle development;
  • hedgehog signaling pathway;
  • muscle pioneer;
  • horizontal myoseptum;
  • engrailed;
  • adaxial cell;
  • medaka;
  • Oryzias latipes

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

We cloned three full-length cDNAs encoding myosin heavy chains (MYHs) previously found to be expressed in embryos or larvae of medaka Oryzias latipes. Based on cDNA sequence information, the three medaka MYH genes, mMYHemb1, mMYHL1 and mMYHL2, were localized on the chromosomes. In vivo promoter assay using the gene encoding green or red fluorescent protein and linked to the 5′-flanking region of mMYH demonstrated that the transcripts of fast-type mMYHemb1, first expressed in embryos but belonging to the adult type in phylogenetic analysis, were located in the horizontal myoseptum. On the other hand, embryonic fast-type mMYHL1 and mMYHL2 were expressed in the whole myotomes. Interestingly, cells expressing mMYHemb1 were localized together with engrailed, and cyclopamine, which blocks hedgehog signaling, inhibited mMYHemb1 expression as well as the formation of the horizontal myoseptum, suggesting that muscle pioneer cells express mMYHemb1 as a key protein in the formation of the horizontal myoseptum. Developmental Dynamics 239:1807–1817, 2010. © 2010 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

The majority of fish trunk muscle is composed of fast muscle, which shows glycolytic properties, whereas slow muscle is localized on the lateral sides of the trunk having oxidative metabolism (Johnston et al.,1974; Johnston,1977; Korneliussen et al.,1978; Sänger and Stoiber,2001). Fast and slow muscles in most fish are morphologically well discriminated, although certain fish contain intermediate muscle fibers having intermediate properties between those of fast and slow muscle fibers.

The development and growth of fish muscle has been well studied and demonstrated to be distinct from that of amniote. Fast and slow muscles in zebrafish Danio rerio are originated from different cell lineages, and the latter is differentiated at early developmental stages from adaxial cells located on both sides of the notochord (Devoto et al.,1996; Daggett et al.,2007). Adaxial cells migrate radially from either sides of the notochord to the superficial part in the trunk, where slow muscle-specific proteins are expressed, and finally developed to slow muscle in a superficial region beneath the skin or remain as muscle pioneers in the horizontal myoseptum (Felsenfeld et al.,1991; Devoto et al.,1996). Such development of slow muscle and differentiation of pioneer cells are induced by hedgehog signaling originated from the notochord (Blagden et al.,1997; Du et al.,1997; Wolff et al.,2003). In association with this migration of adaxial cells, an undifferentiated somitic mesoderm differentiates into fast muscle by expressing its specific genes forming the majority of trunk musculature (Henry and Amacher,2004). Such muscle formation lineage is specific to fish and not observed in amniotes where myotomes are originated from the medial and lateral borders of dermomyotome (Currie and Ingham,1998; Parker et al.,2003; Gros et al.,2004). However, the molecular mechanisms involved in such unique muscle development of fish have remained not fully explained.

Class II myosin composed of two heavy chains with approximately 200 kDa and four light chains of approximately 20 kDa is the major muscle protein and functions as a molecular motor by splitting ATP while bound to actin, another major muscle protein (reviewed by Sellers,2000). It is well known that vertebrate muscles have different myosin heavy chain (MYH) isoforms encoded by specific genes (MYHs) and their expression is changed under different physiological conditions such as endurance training and denervation (Shrager et al.,2000; Huey et al.,2001; Caiozzo et al.,2003). Fish muscles express different MYHs additionally depending on environmental factors such as water temperatures (Hirayama and Watabe,1997; Watabe,2002; Tao et al.,2004; Liang et al.,2007). Therefore, MYH is a useful marker gene for changes in physiological conditions and environmental factors.

Muscle development of fish has been extensively investigated with zebrafish especially focusing on the expression of muscle-specific genes and those encoding transcriptional factors that regulate the expression of muscle proteins in relation to the fate of muscle formation (Xu et al.,2000; Hinits and Hughes,2007). We also previously reported that various sarcomeric MYHs including fast, slow, and cardiac types were differentially expressed during muscle development in common carp Cyprinus carpio (Nihei et al.,2006) and medaka Oryzias latipes (Ono et al.,2006). However, the mechanisms that orchestrate spatially and temporally specific expressions of MYHs during development is still unclear.

Here, three types of fast medaka MYHs (mMYHs) expressed in embryos and larvae were cloned for their full-length cDNA, investigated for their phylogenetic relationship and examined for their expression lineage in association with muscle formation during development. One of these mMYHs was specifically expressed in muscle pioneer cells and considered to be related to the formation of the horizontal myoseptum.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

cDNA Sequence of mMYHs and Their Location on Chromosomes

Sarcomeric MYHs expressed during development of medaka embryos contained embryonic type 1 (mMYHemb1), larval type 1 (mMYHL1), and type 2 (mMYHL2) as predominant mMYHs, which showed different spatiotemporal expression patterns as described in our previous report (Ono et al.,2006). The present study revealed that the full-length cDNA sequences of mMYHemb1, mMYHL1, and mMYHL2 consisted of 5,997 bp, 5,987 bp, and 5,982 bp encoding 1,938, 1,933, and 1,933 amino acids, respectively. Phylogenetic analysis based on the full-length amino acid sequences in comparison with those of other fish species indicated that these three mMYHs are classified into a fast type (Fig. 1) in agreement with our previous analysis using C-terminal sequences (Ono et al.,2006). mMYHemb1 formed unexpectedly the same clade with those of adult fast type and was monophyletic with MYHM2528-2 and MYHM1034 expressed in both fast and slow muscles of adult from torafugu pufferfish Takifugu rubripes (Ikeda et al.,2007; Akolkar et al.,2010). On the other hand, mMYHL1 and mMYHL2 were branched into an embryonic fast type and formed the same clade with MYHM743-2 from torafugu, myhz2 and myhc4 from zebrafish, and MYHemb1 and MYHemb2 from common carp (embryonic type in Fig. 1).

thumbnail image

Figure 1. Phylogenetic tree based on the full-length deduced amino acid sequences of fish sarcomeric myosin heavy chains (MYHs) constructed by the neighbor-joining method. Medaka MYHs (mMYHs) were compared with the following MYHs: common carp (Cc) for adult fast type 10°C-acclimated (F10), 20°C-acclimated (Fint) and 30°C-acclimated (F30), embryonic fast type 1 (emb1) and type 2 (emb2), and embryonic cardiac/slow type (emb3); torafugu (Tr) for MYHM86-1, MYHM743-2, MYHM1034, and MYHM2528-2; and zebrafish (Dr) for embryonic fast type 2 (myhz2) and type 4 (myhc4), ventricular type (vmhc), embryonic slow type 1 (smyhc1), and smooth muscle type. The mMYHs analyzed in this study are boxed. Smooth muscle type MYH of zebrafish was used as outgroup and the bootstrap values from a 1,000 replicate analysis are given at the node in percentage. See the text for accession numbers in the DDBJ/EMBL/GenBank databases of MYHs used for analysis.

Download figure to PowerPoint

To locate the three mMYHs on chromosomes, their full-length cDNA sequences were subjected to BLAST search on the medaka genome database. As a result, mMYHemb1 was found to be located in scaffold 431 on chromosome 5 (Fig. 2A), whereas mMYHL1 and mMYHL2 were both in scaffold 9 on chromosome 6 having a long gap as well as in scaffold 2438 on an unidentified chromosome (Fig. 2B). Meanwhile, adult fast MYHs (mMYH1mMYH11) have been reported to form a cluster on chromosome 8 (Liang et al.,2007).

thumbnail image

Figure 2. Schematic representation for the location of medaka myosin heavy chain genes (mMYHs) on the genome. A: mMYHemb1 is located at scaffold 431 on chromosome 5. Exons are represented by colored boxes. B: mMYHL1 and mMYHL2 are at scaffold 9 on chromosome 6. The two genes were also found to be located closely to each other at scaffold 2438, although its corresponding chromosome has not been identified. Arrows show the direction of transcription. Asterisks and filled circles indicate the initiation and stop codons, respectively. A region having a long sequence gap in scaffold 9 of B is shown by the dashed line.

Download figure to PowerPoint

Expression of Transgenes Containing the 5′-Flanking Regions of mMYHs

The 5′-flanking regions of mMYHs were identified by in silico cloning on the medaka genome database, yielding the sizes of 4,092, 2,693, and 8,141 bp from the translation start codon for mMYHemb1, mMYHL1, and mMYHL2, respectively. Each of the three 5′-flanking regions contained a 5′ untranslated region (UTR) (5′ untranslated region [UTR]) consisting of three exons. To examine their transcriptional activity, the 5′-flanking region was fused to the gene encoding green fluorescent protein (GFP) or red fluorescent protein (RFP) and introduced into fertilized medaka eggs by microinjection. In preliminary experiments, we used the regions of 4,092 and 1,908 bp from the start codon for mMYHemb1. Because both regions functioned as the promoter sequence with comparable activity, we selected the region containing 1,908 bp. Then, we attempted to use the region of 2,659 and 2,030 bp for mMYHL1 and mMYHL2, respectively. While the region of 2,659 bp from mMYHL1 was successful to regulate the activity, the region of 2,030 bp from mMYHL2 failed to be amplified by polymerase chain reaction (PCR) unlike the region of 4,000 bp. Thus we used this 4,000 bp region for constructing the transgene of mMYHL2.

The humanized recombinant GFP (hrGFP) construct containing the 5′-flanking region of mMYHemb1 (emb1-1.9k-hrGFP) showed a transient expression of GFP along the horizontal myoseptum which divides epaxial and hypaxial myotomes of embryos at 2 days postfertilization (dpf; Fig. 3A,B). The transgene was also expressed in larvae at 12 dpf in both trunk and cranial muscles (Fig. 3C,D). Immunohistochemistry was performed using an antibody against hrGFP and F59 antibody against chicken fast-type MYH (Crow and Stockdale,1986) which, interestingly, recognized MYH in adaxial cells in zebrafish (Devoto et al.,1996). The emb1-1.9k-hrGFP transgene was found to be expressed at 2 dpf in the horizontal myoseptum (Fig. 3E) and at 4 and 7 dpf in the lateral surface of the horizontal myoseptum (Fig. 3F–H). On the other hand, F59 labeled the whole myotome in a marked contrast to zebrafish where the antibody labeled adaxial cells and superficial muscle cells (Devoto et al.,1996).

thumbnail image

Figure 3. Transient expression of the emb1-1.9k-hrGFP transgene in medaka embryos and larvae. A: Lateral view of an embryo at 2 days postfertilization (dpf) expressing the transgene (an arrowhead) in trunk somites. Dashed lines indicate the location of the trunk, with anterior to the left. B: Transverse section of an embryo at 2 dpf expressing the transgene in the horizontal myoseptum (an arrowhead). C: Lateral view of a larva at 12 dpf expressing the transgene in the superficial part of the horizontal myoseptum and the trunk. D: Ventral view of the same larva expressing the transgene in cranial muscles. E–G: Immunohistochemistry localizing the transgene expressed in embryos at 2 dpf (E), 4 dpf (F), and 7 dpf (G) in the transverse sections of middle trunks (arrowheads). Sections were immunolabeled with anti-hrGFP polyclonal antibody (green) and F59 monoclonal antibody (red). H: A higher magnification of the boxed region in G. Cell nuclei were stained with 4′, 6-diamidine-2′-phenylindole dihydrochloride (DAPI, blue). Scale bars = 50 μm. NT, neural tube; NC, notochord; EM, epaxial myotome; HM, hypaxial myotome.

Download figure to PowerPoint

The construct containing the 2,659 bp 5′-flanking region of mMYHL1 (L1-2.6k-DsRed) or 4,000 bp that of mMYHL2 (L2-4.0k-hrGFP) expressed the transgene encoding a modified RFP from Discosoma sp. (DsRed) or GFP in both epaxial and hypaxial myotomes of embryos at 4 dpf (Fig. 4A–D). While the L1-2.6k-DsRed transgene was expressed only in inner myotomes, the L2-4.0k-hrGFP transgene was in both inner myotomes and their superficial part (Fig. 4D, arrowheads). Furthermore, the L1-2.6k-DsRed and L2-4.0k-hrGFP transgenes were co-expressed in a single muscle fiber (merged in Fig. 4E,F; arrowheads in Fig. 4F).

thumbnail image

Figure 4. Transient expression of the L1-2.6k-DsRed and L2-4.0k-hrGFP transgenes in medaka embryos at 4 days postfertilization (dpf). A: Lateral view showing expression of the L1-2.6k-DsRed transgene in the trunk myotome. An inset indicates a higher magnification of the expressed transgene. B: Transverse section of the L1-2.6k-DsRed transgene expressed at the middle trunk (green). Epaxial and hypaxial myotomes were labeled with F59 antibody (red) with the same muscle cells expressing the transgene (merged). C: Lateral view showing expression of the L2-4.0k-hrGFP in the trunk myotome. An inset indicates a higher magnification of the expressed transgene. D: Transverse section of the L2-4.0k-hrGFP transgene expressed at the middle trunk (green) with the same muscle cells expressing the transgene (merged). E: Lateral view showing the expression of the L1-2.6k-DsRed and L2-4.0k-hrGFP transgenes in the trunk myotome. F: Higher magnification of the boxed region in E. At least four muscle fibers co-expressed the two transgenes in the same cells (merged, arrowheads). Scale bars = 50 μm. Refer to the legend of Fig. 3 for abbreviations.

Download figure to PowerPoint

Observations With Stable Transgenic Lines

To learn expression patterns of mMYHs during early muscle development more clearly, we attempted to establish stable transgenic lines. Embryos dosed with the transgene constructs were grown to maturity. One male of three matures harbored the emb1-1.9k-hrGFP transgene in the germline, and 29.7% (175 of 589 individuals) of F1 embryos successively expressed this transgene when the male was mated with a wild-type female. The transgenic embryo at 2 dpf expressed hrGFP in the horizontal myoseptum from all somites (Fig. 5A–E) and its localization was identical to that of mMYHemb1 mRNA as demonstrated by immuohistochemical staining and in situ hybridization (Fig. 5F,G). A muscle fiber expressing GFP was mononuclear, which is specific to slow muscle fibers (Supp. Fig. S1, which is available online). GFP was gradually decayed from the trunk during embryonic development (Fig. 5H,I) and in turn appeared in cranial muscles of hatching larvae at 9 dpf with no GFP in the trunk (Fig. 5J,K).

thumbnail image

Figure 5. The expression of the emb1-1.9k-hrGFP transgene in a stable transgenic line. A: Lateral view of the transgene expressed in somites of an embryo at 2 days postfertilization (dpf). B: Dorsal view of the same embryo. C: Lateral view showing an expression pattern of endogenous mMYHemb1 transcripts similar to that of the transgene. D,E: Lateral view showing the transgene expressed in the horizontal myoseptum (green, D) and transverse section showing the localization in the middle trunk of an embryo at 2 dpf (E). Trunk myotomes were labeled with MF20 antibody raised against sarcomeric myosin (red). F: Colocalization (merged) of the emb1-1.9k-hrGFP transgene (green) and endogenous mMYHemb1 transcripts (red) in the horizontal myoseptum of an embryo at 2 dpf. G: Higher magnification of the boxed region in F. H,I: The transgene is expressed in the lateral region at the horizontal myoseptum in an embryo at 4 dpf (arrowheads; H, lateral; I, transverse section at the middle trunk). Trunk myotomes were labeled with MF20 antibody (red). J,K: The expression of the emb1-1.9k-hrGFP transgene in cranial muscles of larvae at 9 dpf (J, ventral view; K, lateral view). Nonspecific fluorescence is observed in pigment cells and liver. Scale bars = 20 μm. HE, heart; LI, liver. Refer to the legend of Fig. 3 for other abbreviations.

Download figure to PowerPoint

Meanwhile, one female of three matures harbored L1-2.6k-DsRed in the germline and 8.2% (27 of 328 individuals) of F1 embryos expressed the transgene. In contrast to emb1-1.9k-hrGFP, the transgenic fish harboring L1-2.6k-DsRed expressed DsRed in whole myotomes of embryos at 3 dpf and additionally in cranial muscles of hatching larva at 9 dpf (Fig. 6). Of interest, the expression of L1-2.6k-DsRed was observed in inner fast muscle of larvae even after 90 days posthatching (dph), but not in superficial slow muscle (Fig. 6H,I). Both transgenics did not express the transgene in heart muscle (Figs. 5J, 6E). We have not yet been successful in the establishment of a transgenic line for the L2-4.0k-hrGFP construct.

thumbnail image

Figure 6. The expression of the L1-2.6k-DsRed transgene in stable transgenics. A,B,F: The L1-2.6k-DsRed transgene is expressed in the myotome of an embryo at 3 days postfertilization (dpf; A, lateral view; B, dorsal view; F, transverse section). C,G: An embryo at 4 and 5 dpf expressing intensely the L1-2.6k-DsRed transgene in the whole myotome (C, lateral view at 4 dpf; G, transverse section at 5 dpf). An inset in C shows a dorsal view. D,E: A hatching larva at 9 dpf expressing the L1-2.6k-DsRed transgene in both trunk (D, lateral view) and cranial muscles (E, ventral view). H,I: The L1-2.6k-DsRed transgene expressed only in inner fast muscle (magenta), but not in lateral superficial slow muscle outlined by dashed lines in larvae after 90 days posthatching (dph). Superficial side is left. Trunk muscle and cell nuclei were labeled by MF20 (green) and 4′, 6-diamidine-2′-phenylindole dihydrochloride (DAPI; blue), respectively. Scale bars = 50 μm in F,G, and 200 μm in H,I. Refer to the legend of Figure 3 for abbreviations in G.

Download figure to PowerPoint

Expression of mMYHemb1 During Differentiation of Adaxial Cells

The expression patterns of emb1-1.9k-hrGFP revealed in the previous section suggest that mMYHemb1 functions in the differentiation of adaxial cells, especially to muscle pioneer cells. Immunohistochemistry using anti-hrGFP together with 4D9 antibody raised against engrailed, a marker protein of muscle pioneer cells (Devoto et al.,1996), demonstrated that cells expressing emb1-1.9k-hrGFP corresponded with those expressing engrailed in the horizontal myoseptum of embryo at 2 dpf, whereas not all engrailed positive cells necessarily expressed emb1-1.9k-hrGFP (Fig. 7).

thumbnail image

Figure 7. The expression of the emb1-1.9k-hrGFP transgene and engrailed. A: Transverse section at the middle trunk of an embryo at 2 days postfertilization (dpf). The emb1-1.9k-hrGFP transgene is expressed in the horizontal myoseptum (arrowhead, green) and colocalized with engrailed protein labeled by 4D9 antibody (red). A dashed line indicates the location of the horizontal myoseptum. B–D: Magnified view around the area indicated by an arrowhead in A. Two cells expressed both the emb1-1.9k-hrGFP transgene and engrailed (arrowheads). Cell nuclei were stained by 4′, 6-diamidine-2′-phenylindole dihydrochloride (DAPI; blue). Scale bars = 20 μm. Refer to the legend of Figure 3 for abbreviations.

Download figure to PowerPoint

Next, we examined the involvement of hedgehog signaling pathway, a positive regulator of adaxial cell differentiation, in mMYH expression using cyclopamine, an inhibitor of hedgehog signaling (Incardona et al.,1998; Chen et al.,2002). Two stable transgenic embryos harboring emb1-1.9k-hrGFP and L1-2.6k-DsRed were treated with 2.5 μg/ml cyclopamine (Fig. 8) where no change was observed in the survival rate of embryos (Supp. Table S1). As a result, the horizontal myoseptum was not formed (Fig. 8A,B), and concomitantly engrailed was not expressed (Fig. 8C,D). Of interest, the rate of embryos at 2 dpf expressing emb1-1.9k-hrGFP was significantly decreased after cyclopamine treatment and found to be only 4.0% compared with those without treatment (75.8%, Fig. 8E). In contrast, the rate of embryos at 4 dpf expressing L1-2.6k-DsRed was 70.0% and 83.3% with and without cyclopamine treatment, respectively (Fig. 8F), suggesting no effect of cyclopamine treatment on the expression of mMYHL1.

thumbnail image

Figure 8. The effects of cyclopamine treatment on muscle development and the expression of the transgenes. A–D: Localization of sarcomeric myosin detected by MF20 antibody in the middle trunk of embryos at 3 days postfertilization (dpf; A,B) and engrailed detected by 4D9 antibody at 2 dpf (C,D). A and C indicate wild embryos (WT), whereas B and D indicate embryos treated with cyclopamine. An arrowhead in A indicates the location of the horizontal myoseptum. Scale bars = 50 μm. Refer to the legend of Figure 3 for abbreviations. E: The rate of transgenic embryos expressing the emb1-1.9k-hrGFP transgene in embryos at 2 dpf with and without treatment of cyclopamine. Numbers in the histogram indicate total individuals examined. F: The rate of transgenic embryos expressing the L1-2.6k-DsRed transgene in embryos at 4 dpf. Chi-squared test was used for statistical analysis with significance at P < 0.01 (*).

Download figure to PowerPoint

Transgene Expression in Zebrafish

We introduced the various constructs of mMYHs into zebrafish fertilized eggs to investigate whether or not the transcriptional regulation by the 5′-flanking region of mMYHs could function in other fish species. Emb1-1.9k-hrGFP was expressed in muscle fibers located at the horizontal myoseptum where muscle fibers possibly derived from adaxial cells were reacted with F59 antibody (Fig. 9A,B). In contrast, L1-2.6k-DsRed and L2-4.0k-hrGFP were expressed in the whole myotome where muscle fibers were reacted with MF20 antibody raised against sarcomeric myosin (Fig. 9C–F). As in the case of medaka, both L1-2.6k-DsRed and L2-4.0k-hrGFP were co-expressed in a single muscle fiber (Fig. 9G,H), whereas only L2-4.0k-hrGFP was expressed in superficial slow muscle (Supp. Fig. S2, see Fig. 4).

thumbnail image

Figure 9. The expression of the transgenes constructed from mMYHs in zebrafish embryos. A,C,E: Lateral view of embryos at 1 days postfertilization (dpf) expressing the genes of emb1-1.9k-hrGFP (A), L1-2.6k-DsRed (C), and L2-4.0k-hrGFP (E) in the trunk myotomes. An inset in each panel indicates a view in higher magnification showing the expression of the transgene. B,D,F: Transverse sections at the middle trunk of embryos at 1 dpf. The emb1-1.9k-hrGFP transgene (B) is expressed in the horizontal myoseptum around superficial slow muscle derived from adaxial cells (labeled by F59, an arrowhead), whereas the L1-2.6k-DsRed (D) and L2-4.0k-hrGFP (F) transgenes are expressed in epaxial and hypaxial myotomes labeled with MF20 antibody (red). Scale bars = 20 μm. G: Lateral view of an embryo at 1 dpf expressing both the L1-2.6k-DsRed (red) and L2-4.0k-hrGFP (green) transgenes in the trunk myotomes. H: Higher magnification of the middle trunk region in G. Several muscle fibers co-expressed the two transgenes (merged in yellow).

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

In this study, we used transgenes in which the 5′-flanking regions of the three mMYHs, mMYHemb1, mMYHL1. and mMYHL2, were fused to the 5′ end of the gene encoding GFP or RFP. Each of these transgenes was expressed at the same site and stage as that of the corresponding endogenous gene previously reported (Ono et al.,2006), demonstrating that the 5′-flanking regions of mMYHemb1, mMYHL1, and mMYHL2 regulate their expression in muscle fibers during development of medaka.

Based on the primary structure, sarcomeric MYHs of fish are classified into fast and slow/cardiac types (McGuigan et al.,2004; Ikeda et al.,2007), and the fast type is further grouped into adult and embryonic types (see Fig. 1; Ikeda et al.,2010). mMYHL1 and mMYHL2 cloned from medaka in the present study were classified into an embryonic fast type with MYHs expressed predominantly during embryonic development in common carp (Nihei et al.,2006), torafugu (Ikeda et al.,2007), and zebrafish (Peng et al.,2002; Bryson-Richardson et al.,2005; see Fig. 1), and both showed their expression in the whole myotomes of embryos and larvae (see Figs. 4, 6). mMYHL1 and mMYHL2 were expressed in the same muscle fibers together, revealing that the 5′-flanking regions of mMYHL1 and mMYHL2 have comparable functions (see Figs. 4, 9). On the other hand, mMYHL1 was expressed only in inner fast muscle, whereas mMYHL2 was expressed in both inner fast and superficial slow muscles. In addition, muscle fibers expressing L1-2.6k-DsRed tended to be localized in an anterior region of trunk muscle compared with those expressing L2-4.0k-hrGFP (see Figs. 4, 9). Because mMYHL1 and mMYHL2 were tandemly arrayed on the same chromosome (see Fig. 2), it is suggested that these MYHs are paralogous genes aroused by gene duplication and have been diverged to function in the different spatial expression mentioned above. Mammalian fast-type MYHs tandemly arrayed on the same chromosome have been reported for their expression to be regulated by natural antisense RNA (Pandorf et al.,2006). It is uncertain whether or not such expressional regulation would be conserved in the two embryonic fast-type mMYHs.

mMYHemb1 was classified into an adult fast type, but unexpectedly expressed in the horizontal myoseptum during embryonic development (see Fig. 5). It is noted that the putative orthologous genes of mMYHemb1 in torafugu, MYHM2528-1 (a homologue of MYHM2528-2) and MYHM1034, are expressed in both adult fast and slow muscles in the trunk (Ikeda et al.,2007; Akolkar et al.,2010). Such different expression patterns among these orthologous MYHs require the comparison in their 5′-flanking regions.

We previously demonstrated that eight adult fast-type mMYHs and three pseudogenes are tandemly arrayed, forming a gene cluster in a narrow region spanning 219 kbp on a single chromosome (Liang et al.,2007). However, neither mMYHemb1, mMYHL1, nor mMYHL2 was included in this gene cluster. Although mMYHemb1, mMYHL1, and mMYHL2 were observed marginally in adult fast muscle as revealed by reverse transcription PCR (Ono et al.,2006), their cDNAs were not cloned from adult fast muscle (Liang et al.,2007), indicating that mMYHemb1, mMYHL1, and mMYHL2 likely have functions different from those of adult fast-type mMYHs.

Among embryonic mMYHs investigated in this study, mMYHemb1 showed its expression specifically to the horizontal myoseptum of embryos, whereas mMYHL1 and mMYHL2 were expressed in the whole myotome of embryos as described above. These results suggest that mMYHemb1 has its expressional regulation different from those of mMYHL1 and mMYHL2. In zebrafish, adaxial cells migrate radially to the surface of the trunk muscle to form slow muscle as well as through the border of epaxial and hypaxial somites in the horizontal myoseptum during development, whereas muscle pioneer cells are produced from adaxial cells and remain only in the horizontal myoseptum as earliest muscle fibers (Felsenfeld et al.,1991; Devoto et al.,1996; Du et al.,1997). Such differentiation of adaxial cells is induced by hedgehog signaling (Blagden et al.,1997; Du et al.,1997; Wolff et al.,2003).

Cells expressing mMYHemb1 also expressed engrailed, a marker protein of muscle pioneer cells, whereas its expression was decreased by the treatment with cyclopamine, an inhibitor of the hedgehog signaling pathway (see Fig. 8). These results indicate that cells expressing mMYHemb1 correspond to muscle pioneer cells, and its gene expression is involved in the differentiation of adaxial cells. Furthermore, muscle fibers expressing mMYHemb1 were mononucleated and moved to a superficial part of the trunk in the progress of embryonic development (see Figs. 3, 5, and Supp. Fig. S1). These muscle fibers had small diameters compared with those observed in inner myotomes (see Fig. 3), suggesting that fibers expressing mMYHemb1 in a superficial part are localized in a region of slow muscle.

Engrailed-positive and mMYHemb1-negative cells were observed around cells expressing mMYHemb1 (see Fig. 7). It has been reported that engrailed positive fast muscle fibers are located in the neighbor of muscle pioneer cells (Wolff et al.,2003). However, these fibers are multinucleated and thus different from mononucleated muscle pioneer cells expressing mMYHemb1 found in the present study. Zebrafish has been reported to express both fast- and slow-type MYHs in adaxial cells (Bryson-Richardson et al.,2005). We also previously reported that medaka mMYHC1 belonging to slow/cardiac type MYH is expressed in the horizontal myoseptum as in the case of mMYHemb1 (Ono et al.,2006). These results suggest that muscle pioneer cells have different types in terms of the expression of MYH isoforms.

It is likely that muscle fibers expressing mMYHemb1 function in the formation of the horizontal myoseptum. Cyclopamine treatment declined the expression of mMYHemb1 with a failure in the formation of the horizontal myoseptum (see Fig. 8). Hedgehog signaling has been reported to participate in the formation of the horizontal myoseptum. For example, zebrafish you-too (yot) and slow-muscle-omitted (smu) mutants do not form the horizontal myoseptum due to loss of hedgehog signaling pathway (van Eeden et al.,1996; Karlstrom et al.,1999; Barresi et al.,2000). Meanwhile, zebrafish smyhc2 belonging to slow-type MYH is expressed in the horizontal myoseptum of embryos through hedgehog signaling (Elworthy et al.,2008). These data together with our present results suggest that certain fish MYHs including both fast and slow types are expressed transiently to form the horizontal myoseptum. Unfortunately, we have not yet been successful to clone slow-type mMYH such as zebrafish smyhc1 (Bryson-Richardson et al.,2005), common carp MYHemb3 (Nihei et al.,2006) and torafugu MYHM5 (Ikeda et al.,2007), which are contained in adaxial cells migrating to the superficial part to form slow muscle.

The muscle type-specific regulation by the 5′-flanking region of mMYH raised the question about which transcriptional factors interact with them. In this regard, it has been claimed by in vivo experiment that MEF2 binding site is crucial for the expression of mMYHs (Liang et al.,2008) and common carp MYHs (Kobiyama et al.,2006). Furthermore, Prdm1 has been reported to determine the developmental lineage of zebrafish embryonic slow muscle by repressing the expression of fast fiber-specific MYH (Elworthy et al.,2008; von Hofsten et al.,2008).

In conclusion, we revealed that spatiotemporally specific expression patterns of embryonic and larval fast types of mMYHs were regulated through their 5′-flanking region during development of medaka embryos. The transcriptional activity of the transgenes containing the 5′-flanking region of these mMYHs was similarly observed in zebrafish embryos, indicating a conserved function of this region in fish during development of fish embryos. Moreover, the expression of mMYHemb1 depended on hedgehog signaling in association with the formation of the horizontal myoseptum.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

Experimental Fish

The HNI inbred strain of medaka (reviewed by Wittbrodt et al.,2002) was used for cDNA cloning and extraction of genomic DNA. Orange–red colored wild medaka and wild zebrafish were used for in vivo promoter assay. Embryos of both species were incubated in a chamber at 26°C.

Determination of the Full-Length cDNA Sequences of mMYHs

Almost full-length cDNA encoding mMYHemb1 was amplified from first-strand cDNA synthesized from total RNA of embryos at 3 dpf using a set of forward primer GAGYRSDGAYGCWGARATGGA designed in reference to the sequences around the start codon of MYHs previously reported in medaka (Liang et al.,2007) and common carp (Hirayama and Watabe,1997), and reverse primer TAGGAGTTGATAATGAAAAG designed in reference to the 3′ UTR sequence of mMYHemb1. cDNA encoding mMYHL1 and mMYHL2 were amplified from embryos at 9 dpf using a set of forward primer ATGAGYACRGACGCGGAGATG designed in reference to the sequences around the start codon of myhz2 (BC071279 in the DDBJ/EMBL/GenBank databases) and myhc4 (AY921650) from zebrafish as well as MYHemb1 (AB231798) and MYHemb2 (AB231799) of common carp, and reverse primer designed in reference to the 3′ UTR sequences of mMYHL1 (ACTTTATTTAGGTGTACCTC) and mMYHL2 (TATGATTCAACATTATAGGA). Amplified DNA fragments were cloned into pUC118 vector using Mighty Cloning Kit (Takara, Otsu, Japan) and sequenced. The 5′ UTR sequences were amplified with CapFishing Full-length cDNA Premix Kit (Seegene, Seoul, Korea) using a forward primer provided by the kit (GTCTACCAGGCATTCGCTTCAT) and gene-specific reverse primers of AGGTACATCTCGTTGGGTTCAGAC for mMYHemb1, CTGTTCTTTCCTCATTACTATCCAG for mMYHL1, and CTGTTCTTTCCTCATTACTATCCA for mMYHL2. Determined cDNA sequences were registered into the DDBJ/EMBL/GenBank databases with accession numbers AB472673 for mMYHemb1, AB472674 for mMYHL1, and AB472675 for mMYHL2.

Bioinformatics Analysis

The full-length deduced amino acid sequences of fish MYHs were aligned by Clustal X multiple sequence alignment program (Thompson et al.,1997), and a phylogenetic tree was constructed using the neighbor-joining method on the software MEGA4 (Tamura et al.,2007). Accession numbers in the DDBJ/EMBL/GenBank databases of MYHs used for analysis are follows: medaka, mMYH1 (BAF34701), mMYH2 (BAF34702), mMYH3 (BAF34703), mMYH5 (BAF34700), mMYH6 (BAF34699), mMYH7 (BAF34704), mMYH9 (BAF34705), and mMYH11 (BAF34706); common carp, adult fast skeletal 10°C acclimated (BAA22067), 20°C acclimated (BAA22068), 30°C acclimated (BAA22069), MYHemb1 (BAE79361), MYHemb2 (BAE79362), and MYHemb3 (BAE79363); zebrafish, myhz2 (AAH71279), myhc4 (AAY26547), smyhc1 (AAY26546), and smooth muscle type (AAY42972). Deduced amino acid sequences of MYHM1034, MYHM2528-2, and MYHM743-2 from torafugu pufferfish were predicted by genomic sequences and partial coding sequences of cloned cDNA (Ikeda et al.,2007) and that of MYHM86-1 by the full-length cDNA sequence (AB465004). The genomic sequences are provided by National Institute of Genetics (http://dolphin.lab.nig.ac.jp/medaka/) for medaka and International Fugu Genome Consortium (http://www.fugu-sg.org/index.html) for torafugu pufferfish. BLAST search was carried out on the genomic databases using the Ensembl Genome Browser (http://www.ensembl.org/index.html).

Preparation of GFP and DsRed Constructs and Microinjection

Various sequences in the 5′-flanking region of mMYHs were amplified from genomic DNA by PCR and cloned into the promoter-less phrGFP vector (Stratagene, LA Jolla, CA) or pDsRed Express-1 vector (Clontech, Mountain View, CA). For the insertion of DNA fragments into the vector, additional sequences recognized by restriction enzymes (CGCGGATCC for BamHI, CGGAATTC for EcoRI, and CCGCAAGCTT for HindIII) were added at the 5′ end of primers (primer sequences are described in Supp. Table S2), and the insertion of DNA fragments was confirmed by PCR and sequencing. In microinjection, the constructs were diluted at 50 ng/μl with sterile distilled water containing 0.025% phenol red and introduced into fertilized eggs at the two-cell stage. The fluorescence derived from transgenes in embryos was observed with a MVX10 macro-zoom microscope (Olympus, Tokyo, Japan) and a FV1000 confocal laser scanning microscope (Olympus).

Immunohistochemistry and In Situ Hybridization

For immunohistochemistry, embryos were fixed with 4% paraformaldehyde in Tris-buffered saline (25 mM Tris, 137 mM NaCl, 2.7 mM KCl, pH 7.4) with 0.1% Tween 20 (TBSTw) overnight at 4°C. Fixed embryos were washed with TBSTw, and blocking was performed using 1.5% blocking reagent (Roche, Mannheim, Germany) in TBSTw. The first antibodies used in this study were as follows. Vitality full-length hrGFP polyclonal antibody (Stratagene) was used at a dilution of 1:2,500 in blocking solution, Living Colors DsRed polyclonal antibody (Clontech) at 1:1,000, and MF20, F59, and 4D9 supplied by Developmental Studies Hybridoma Bank at 1:20. Immunoreaction with the first antibody was performed overnight at 4°C. After the incubation, embryos were washed with TBSTw and labeled with the second antibodies, anti-mouse IgG Alexa Fluor 555 and anti-rabbit IgG Alexa Fluor 488 (Invitrogen, Carlsbad, CA) at a dilution of 1:250 for 3 hr at room temperature. The localization of cell nuclei was observed after staining for 10 min with 4′, 6-diamidine-2′-phenylindole dihydrochloride (DAPI, Roche) diluted at 1 μg/ml in TBSTw. Transverse sections were prepared at a thickness of 16 μm using a HM550 cryostat (Thermo, Walldorf, Germany) before the first immunoreaction.

In situ hybridization was performed with protocols as previously described (Ono et al.,2006), and digoxigenin-labeled RNA probes hybridized were detected with anti-digoxigenin monoclonal antibody (Roche, used at a 330-fold dilution) as the first antibody and anti-mouse IgG Alexa Fluor 555 as the second antibody in double staining by immuohistochemistry and in situ hybridization.

Cyclopamine Treatment

Fertilized eggs of stable emb1-1.9k-hrGFP and L1-2.6k-DsRed transgenic line were obtained by crossing matured heterozygous F1 transgenic fish. Transgenic embryos at 4- to 32-cell stages were transferred into 2.5, 5.0, and 10 μg/ml cyclopamine solution (Wako, Osaka, Japan) containing 0.25, 0.5, and 1.0% ethanol and incubated at 26°C.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

The authors thank T. Kaneko, The University of Tokyo, for his advice in immunohistochemistry and confocal laser scanning microscopy. MF20, F59, and 4D9 antibodies developed by D.A. Fischman, F.E. Stockdale, and C. Goodman were obtained from the Developmental Studies Hybridoma Bank maintained by The University of Iowa, Department of Biological Science (Iowa City, IA).

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
DVDY_22298_sm_suppfig1.tif3766KSupporting Figure 1. A–C: A mononuclear muscle fiber expressing the emb1-1.9k-hrGFP transgene in the horizontal myoseptum of an embryo at 2 dpf (green). The nucleus in a mononuclear muscle fiber is indicated by arrowheads. A,C: Cell nuclei were stained by 4′, 6-diamidine-2′-phenylindole dihydrochloride (DAPI; magenta). Scale bars = 20 μm.
DVDY_22298_sm_suppfig2.tif5034KSupporting Figure 2. A–D: The expression of L1-2.6k-DsRed (A,B) and L2-4.0k-hrGFP (C,D) transgenes in zebrafish embryos. A and C were scanned in the inner part of trunk myotome and B and C in the superficial part of embryos at 1 days postfertilization (dpf). Both transgenes are expressed in the inner myotome (A,C), whereas only L2-4.0k-hrGFP is expressed in the superficial slow muscle labeled by F59 (D, arrowheads). Scale bars = 50 μm. Anterior is left in all panels.
Table_S1.xls16KSupporting Table 1. The effect of cyclopamine treatment on the survival rate of wild-type embryos at 2dpf
Table_S2.xls15KSupporting Table 2. Nucleotide sequence of primers for amplification of 5'-flanking region of mMYHs

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.