- Email: [email protected]
Developmental program expression of myosin alkali light chain and skeletal actin genes in the rainbow trout Oncorhynchus mykiss1
Biochimica et Biophysica Acta 1519 (2001) 139^142
www.bba-direct.com
Short sequence-paper
Developmental program expression of myosin alkali light chain and skeletal actin genes in the rainbow trout Oncorhynchus mykiss1 Pierre Thie¨baud a
a;
*, Pierre-Yves Rescan b , Wilfrid Barillot a , Ce¨cile Rallie©re b , Nadine The¨ze¨ a
Unite¨ INSERM 441, Avenue du Haut-Le¨veªque, Universite¨ Bordeaux 2, 33600 Pessac, France b SCRIBE-INRA, Campus de Beaulieu, 35042 Rennes, France Received 8 January 2001; received in revised form 28 March 2001; accepted 3 April 2001
Abstract We have isolated MLC1F (tMLC1F ), MLC3F (tMLC3F ) and skeletal actin cDNAs from the teleost Oncorhynchus mykiss. Sequence analysis indicates that tMLC1F and tMLC3F are not produced from differentially spliced mRNAs as reported in avians and rodents but are encoded by different genes. Results from RNase protection analysis showed that the corresponding transcripts are expressed in fast skeletal muscles. Whole-mount in situ hybridisation revealed distinct expression patterns of the myosin alkali light chains and skeletal actin genes during skeletal muscle development in the embryo. ß 2001 Elsevier Science B.V. All rights reserved. Keywords : Myosin alkali light chain; Skeletal actin; cDNA ; Skeletal muscle; Embryo ; Rainbow trout
The contractile proteins myosin and actin exist in multiple, tissue-speci¢c and developmentally regulated isoforms encoded by families of genes [1]. In vertebrate adult fast skeletal muscle, myosin consists of two heavy chains and four associated light chains, two regulatory light chains (MLC2) and two alkali light chains (MLC1F , MLC3F ). In rodents and avian, MLC1F and MLC3F , are encoded by a single gene and produced by alternative transcription and two modes of splicing [2^5]. cDNA clone analysis indicates that there is a similar situation in the amphibian Xenopus [6] but this is in sharp contrast with ¢sh where MLC1F and MLC3F are produced from distinct genes [7,8]. This suggests a divergent evolution of the process that has lead to MLC1F and MLC3F proteins production in vertebrates. Although mouse and Xenopus show a conserved MLC1F /3 F and skeletal actin gene structure, they present a distinct development regulation of these genes. In mouse, MLC1F , MLC3F and skeletal actin transcripts show independent pattern of expression during embryonic muscle development and are not detected before somites formation [9^11]. In Xenopus there is a synchronous ex* Corresponding author. Fax: +33-5-5636-8979; E-mail : [email protected] 1 The sequence data in this paper have been submitted to the EMBL/ GenBank data libraries under the accession numbers AF330140, AF330141, AF330142 and AF360980.
pression of the three mRNAs whose expression commences before segmentation of the dorsal mesoderm into somites [6,12]. In this respect, the timetable of myogenesis is di¡erent between mammals and amphibian. Little is known about the expression of myosin light chain and skeletal actin genes during early stages of skeletal muscle development in ¢sh. We report here the cloning of MLC1F , MLC3F and skeletal actin cDNAs from the trout Oncorhynchus mykiss and the expression of the corresponding genes in the embryo and adult tissues. A rainbow trout embryo Vgt10 cDNA library was screened with Xenopus laevis MLC1F /3F and skeletal actin cDNA probes. The cDNA clones tMLC1F and tMLC3F isolated are full length and encode proteins of respectively 193 and 158 amino acids (Fig. 1). The two proteins show 74^77% identity with the corresponding proteins from carp and mullet species and the homology rises to 85% after allowing for conservative changes in the sequences. There is only 63^70% identity with the Xenopus and mouse homologous proteins (Fig. 1A,B). When the trout tMLC1F and tMLC3F proteins are aligned, they show a 144-amino-acid overlapping region (amino acid 40^193 of tMLC1F ) that presents 37 amino acid substitutions and a 3-amino-acid deletion in the tMLC3F protein (Fig. 1C). This suggests that the two proteins are, like other known ¢sh but on the contrary to mammals and amphibian sequences, produced by distinct genes [7,8]. The trout skel-
0167-4781 / 01 / $ ^ see front matter ß 2001 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 7 8 1 ( 0 1 ) 0 0 2 2 1 - 4
BBAEXP 91549 29-5-01
Cyaan Magenta Geel Zwart
140
P. Thie¨baud et al. / Biochimica et Biophysica Acta 1519 (2001) 139^142
Fig. 1. (A) Comparison of the amino acid sequences of MLC1F from trout (tMLC1F ) and those from other species. Identical amino acid residues are represented by dots. Hyphen represent gaps introduced into the amino acid sequence in order to obtain optimal sequence homology. (B) Comparison of the amino acid sequences of MLC3F (tMLC3F ) from trout and those from other species. (C) Sequence alignment between tMLC1F and tMLC3F proteins.
etal actin cDNA clone isolated, although uncomplete, codes for the last 337 amino acids of the protein (amino acid region 25^370) and has only ¢ve amino acid di¡erences with the skeletal actin clones from other ¢sh species (data not shown). We analysed by RNase protection assay the expression of tMLC1F , tMLC3F and skeletal actin mRNAs in embryo and adult tissues. The three mRNAs are detected in total RNA from eyed stage (stage 21 according to Ballard
staging [13]) and in hatching embryos (stage 23) (Fig. 2A,B). In 280 and 500 g immature ¢sh, the tMLC1F , tMLC3F and skeletal actin mRNAs are expressed in fast white muscle (Fig. 2A,B) but not in slow red muscle. In adult tissues, tMLC1F , tMLC3F and skeletal actin mRNAs are not detected in cardiac, smooth muscle or
C
Fig. 2. Expression of tMLC1F , tMLC3F and skeletal actin mRNAs by RNase protection analysis. Two Wg of total RNA were hybridised with anti-sense probes and the RNA duplexes were digested with a mix of RNase A and T1 before analysis on a 5% denaturing gel. (A) tMLC1F and tMLC3F mRNA analysis. The two anti-sense probes are mixed in the assay allowing co-detection of mRNAs. (B) Skeletal actin (sk actin) mRNA analysis. (C) GAPDH mRNA analysis control to ensure the quality and quantity of RNA samples. The size discrepancies between the probes and the protected fragments are due to the polylinker sequences. RNA was extracted from stage 21 embryo (St. 21), stage 23 embryo (St. 23), fast or slow muscle from 280 g (280) and 500 g (500) juveniles, intestine, heart, and liver. Control tRNA (^) and undigested probes (P).
BBAEXP 91549 29-5-01
Cyaan Magenta Geel Zwart
P. Thie¨baud et al. / Biochimica et Biophysica Acta 1519 (2001) 139^142
141
Fig. 3. Expression pattern of tMLC1F (A^C), tMLC3F (D^F) and skeletal actin (G^I) mRNAs analysed by whole-mount in situ hybridisation. The anti-sense RNA probes used in RNAse protection assays have been labeled with UTP-digoxygenin. (A,D,G) Stage 12 embryo (approximately 20 somites) : tMLC1F transcripts are present in the most rostral somites while tMLCF3 and skeletal actin transcripts are still undetectable. (B,E,H) Stage 17 embryo (approximately 45 somites) : tMLC1F labels more caudal newly formed somites compared to tMLC3F and skeletal actin. (C,F,I) Stage 21 embryo (segmentation is complete): all the chevron-shaped myotomes are completely stained for the three genes.
non muscle tissues (Fig. 2A,B). Thus, the restricted expression pattern of the tMLC1F , tMLC3F , and skeletal actin genes to fast skeletal muscle con¢rm their identity. The spatio-temporal expression of the three genes in the developing embryo was analysed by whole-mount in situ hybridisation. None of the transcripts were detected before stage 12 (20 somites) (data not shown). Therefore, in agreement with the situation found in mammals and amphibian, the onset of expression of these structural genes during ¢sh development is later than that of the myogenic regulatory bHLH genes like TMyoD or Tmyogenin whose expression begins before somitogenesis (TMyoD) or in forming somites (TMyoD2 and Tmyogenin) [14]. In stage 12 embryo, only tMLC1F mRNA is detected and its expression is restricted to the most rostral somites (Fig. 3A). The expression of tMLC1F progresses caudally as the somites formed in a rostral to caudal wave (Fig. 3A,B). tMLC3F expression is ¢rst detected in stage 16 embryo (data not shown) and it labels the most rostral somites but not the newly formed posterior somites in stage 17 embryo (Fig. 3E). The expression of skeletal actin
BBAEXP 91549 29-5-01
mRNA is similar to that of tMLC3F mRNA following a rostral^caudal gradient as the somites are forming (Fig. 3H). The delayed expression of tMLC3 F and skeletal actin genes compared with that of tMLC1F gene is apparent especially in stage 17 embryo, where only the rostral somites express tMLC3F and skeletal actin transcripts while the expression of tMLC1F transcripts extends more caudally in more newly formed somites (Fig. 3B,E,H). When the segmentation is complete (stage 21) the three genes are expressed in all chevron-shaped myotomes but none of them is expressed at a detectable level in embryonic heart (Fig. 3C,F,I). In the mouse, MLC1F transcripts can ¢rst be detected between 8.5 and 9.5 days post coitum (p.c.) when about 25 somites are formed [10]. This marks a striking similitude with the temporal expression of the MLC1F gene in trout embryo. This suggests that the MLC1F regulatory regions, responsible for the temporal expression of the gene, have been kept under evolutionary constraints between mammals and ¢sh. In contrast, the expression of the MLC3F gene is markedly delayed in mouse muscle development
Cyaan Magenta Geel Zwart
142
P. Thie¨baud et al. / Biochimica et Biophysica Acta 1519 (2001) 139^142
when compared to trout development as MLC3F transcripts are not detected before the end of somitogenesis (i.e., 15 days p.c. in mouse embryo) when there are about 60 somites [10]. Moreover, MLC3F transcripts are expressed in mouse embryonic heart [15] but this is not the case in trout embryo or in Xenopus as we have shown before [6]. Our analysis reveals also a striking di¡erence between teleost and mammal skeletal muscle development with respect to the skeletal actin gene expression. In mouse, skeletal actin gene expression, although weak, precedes MLC1F gene expression while in trout there is an opposite situation ([10]; this study). This makes the MLC1F gene a very early and sensitive marker for fast skeletal muscle di¡erentiation in trout. Our studies revealed that the asynchronous temporal expression of muscle genes in the developing vertebrate embryo has evolved di¡erently between species and probably re£ects intrinsic species speci¢c events such as the gradual buildup of di¡erent levels of regulatory factors within the muscle cell. We thank David Lepetit for help with cDNA library screening and Michelle Olive for critical reading of the manuscript. This work was supported by the Centre National de La Recherche Scienti¢que, the Institut National de la Sante¨ et de la Recherche Me¨dicale and a grant from The Association Franc°aise contre les Myopathies.
BBAEXP 91549 29-5-01
References [1] E. Bandman, Dev. Biol. 1 (1992) 273^283. [2] P.J.R. Barton, M. Buckingham, Biochem. J. 231 (1985) 249^261. [3] B. Robert, P. Daubas, M.-A. Akimenko, A. Cohen, I. Garner, J.-L. Guenet, M. Buckingham, Cell 39 (1984) 129^140. [4] M. Periasamy, E. Strehler, L.I. Gar¢nkel, R.M. Gubits, N. Ruiz-Opazo, B. Nadal-Ginard, J. Biol. Chem. 259 (1984) 13595^ 13604. [5] Y.-I. Nabeshima, Y. Fujii-Kuriyama, M. Muramatsu, K. Ogata, Nature 308 (1984) 333^338. [6] N. The¨ze¨, S. Hardy, R. Wilson, M.-R. Allo, T. Mohun, P. Thie¨baud, Dev. Biol. 171 (1995) 352^362. [7] L. Dalla Libera, E. Carpene, J. Theibert, J. Collins, J. Muscle Res. Cell Motil. 12 (1991) 366^371. [8] Y. Hirayama, S. Kanoh, M. Nakaya, S. Watabe, J. Exp. Biol. 200 (1997) 693^701. [9] D. Sassoon, I. Garner, M. Buckingham, Development 104 (1988) 155^164. [10] G.E. Lyons, M. Ontell, R. Cox, D. Sassoon, M. Buckingham, J. Cell Biol. 111 (1990) 1465^1476. [11] M. Ontell, M.P. Ontell, M.M. Sopper, R. Mallonga, G. Lyons, M. Buckingham, Development 117 (1993) 1435^1444. [12] T.J. Mohun, S. Brennan, N. Dathan, S. Fairman, J.B. Gurdon, Nature 311 (1983) 716^721. [13] W.W. Ballard, J. Exp. Zool. 184 (1973) 7^26. [14] J.-M. Delalande, P.-Y. Rescan, Dev. Genes Evol. 209 (1999) 432^ 437. [15] R.G. Kelly, P.S. Zammit, V. Mouly, G. Butler-Browne, M.E. Buckingham, J. Mol. Cell. Cardiol. 30 (1998) 1067^1081.
Cyaan Magenta Geel Zwart
www.bba-direct.com
Short sequence-paper
Developmental program expression of myosin alkali light chain and skeletal actin genes in the rainbow trout Oncorhynchus mykiss1 Pierre Thie¨baud a
a;
*, Pierre-Yves Rescan b , Wilfrid Barillot a , Ce¨cile Rallie©re b , Nadine The¨ze¨ a
Unite¨ INSERM 441, Avenue du Haut-Le¨veªque, Universite¨ Bordeaux 2, 33600 Pessac, France b SCRIBE-INRA, Campus de Beaulieu, 35042 Rennes, France Received 8 January 2001; received in revised form 28 March 2001; accepted 3 April 2001
Abstract We have isolated MLC1F (tMLC1F ), MLC3F (tMLC3F ) and skeletal actin cDNAs from the teleost Oncorhynchus mykiss. Sequence analysis indicates that tMLC1F and tMLC3F are not produced from differentially spliced mRNAs as reported in avians and rodents but are encoded by different genes. Results from RNase protection analysis showed that the corresponding transcripts are expressed in fast skeletal muscles. Whole-mount in situ hybridisation revealed distinct expression patterns of the myosin alkali light chains and skeletal actin genes during skeletal muscle development in the embryo. ß 2001 Elsevier Science B.V. All rights reserved. Keywords : Myosin alkali light chain; Skeletal actin; cDNA ; Skeletal muscle; Embryo ; Rainbow trout
The contractile proteins myosin and actin exist in multiple, tissue-speci¢c and developmentally regulated isoforms encoded by families of genes [1]. In vertebrate adult fast skeletal muscle, myosin consists of two heavy chains and four associated light chains, two regulatory light chains (MLC2) and two alkali light chains (MLC1F , MLC3F ). In rodents and avian, MLC1F and MLC3F , are encoded by a single gene and produced by alternative transcription and two modes of splicing [2^5]. cDNA clone analysis indicates that there is a similar situation in the amphibian Xenopus [6] but this is in sharp contrast with ¢sh where MLC1F and MLC3F are produced from distinct genes [7,8]. This suggests a divergent evolution of the process that has lead to MLC1F and MLC3F proteins production in vertebrates. Although mouse and Xenopus show a conserved MLC1F /3 F and skeletal actin gene structure, they present a distinct development regulation of these genes. In mouse, MLC1F , MLC3F and skeletal actin transcripts show independent pattern of expression during embryonic muscle development and are not detected before somites formation [9^11]. In Xenopus there is a synchronous ex* Corresponding author. Fax: +33-5-5636-8979; E-mail : [email protected] 1 The sequence data in this paper have been submitted to the EMBL/ GenBank data libraries under the accession numbers AF330140, AF330141, AF330142 and AF360980.
pression of the three mRNAs whose expression commences before segmentation of the dorsal mesoderm into somites [6,12]. In this respect, the timetable of myogenesis is di¡erent between mammals and amphibian. Little is known about the expression of myosin light chain and skeletal actin genes during early stages of skeletal muscle development in ¢sh. We report here the cloning of MLC1F , MLC3F and skeletal actin cDNAs from the trout Oncorhynchus mykiss and the expression of the corresponding genes in the embryo and adult tissues. A rainbow trout embryo Vgt10 cDNA library was screened with Xenopus laevis MLC1F /3F and skeletal actin cDNA probes. The cDNA clones tMLC1F and tMLC3F isolated are full length and encode proteins of respectively 193 and 158 amino acids (Fig. 1). The two proteins show 74^77% identity with the corresponding proteins from carp and mullet species and the homology rises to 85% after allowing for conservative changes in the sequences. There is only 63^70% identity with the Xenopus and mouse homologous proteins (Fig. 1A,B). When the trout tMLC1F and tMLC3F proteins are aligned, they show a 144-amino-acid overlapping region (amino acid 40^193 of tMLC1F ) that presents 37 amino acid substitutions and a 3-amino-acid deletion in the tMLC3F protein (Fig. 1C). This suggests that the two proteins are, like other known ¢sh but on the contrary to mammals and amphibian sequences, produced by distinct genes [7,8]. The trout skel-
0167-4781 / 01 / $ ^ see front matter ß 2001 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 7 8 1 ( 0 1 ) 0 0 2 2 1 - 4
BBAEXP 91549 29-5-01
Cyaan Magenta Geel Zwart
140
P. Thie¨baud et al. / Biochimica et Biophysica Acta 1519 (2001) 139^142
Fig. 1. (A) Comparison of the amino acid sequences of MLC1F from trout (tMLC1F ) and those from other species. Identical amino acid residues are represented by dots. Hyphen represent gaps introduced into the amino acid sequence in order to obtain optimal sequence homology. (B) Comparison of the amino acid sequences of MLC3F (tMLC3F ) from trout and those from other species. (C) Sequence alignment between tMLC1F and tMLC3F proteins.
etal actin cDNA clone isolated, although uncomplete, codes for the last 337 amino acids of the protein (amino acid region 25^370) and has only ¢ve amino acid di¡erences with the skeletal actin clones from other ¢sh species (data not shown). We analysed by RNase protection assay the expression of tMLC1F , tMLC3F and skeletal actin mRNAs in embryo and adult tissues. The three mRNAs are detected in total RNA from eyed stage (stage 21 according to Ballard
staging [13]) and in hatching embryos (stage 23) (Fig. 2A,B). In 280 and 500 g immature ¢sh, the tMLC1F , tMLC3F and skeletal actin mRNAs are expressed in fast white muscle (Fig. 2A,B) but not in slow red muscle. In adult tissues, tMLC1F , tMLC3F and skeletal actin mRNAs are not detected in cardiac, smooth muscle or
C
Fig. 2. Expression of tMLC1F , tMLC3F and skeletal actin mRNAs by RNase protection analysis. Two Wg of total RNA were hybridised with anti-sense probes and the RNA duplexes were digested with a mix of RNase A and T1 before analysis on a 5% denaturing gel. (A) tMLC1F and tMLC3F mRNA analysis. The two anti-sense probes are mixed in the assay allowing co-detection of mRNAs. (B) Skeletal actin (sk actin) mRNA analysis. (C) GAPDH mRNA analysis control to ensure the quality and quantity of RNA samples. The size discrepancies between the probes and the protected fragments are due to the polylinker sequences. RNA was extracted from stage 21 embryo (St. 21), stage 23 embryo (St. 23), fast or slow muscle from 280 g (280) and 500 g (500) juveniles, intestine, heart, and liver. Control tRNA (^) and undigested probes (P).
BBAEXP 91549 29-5-01
Cyaan Magenta Geel Zwart
P. Thie¨baud et al. / Biochimica et Biophysica Acta 1519 (2001) 139^142
141
Fig. 3. Expression pattern of tMLC1F (A^C), tMLC3F (D^F) and skeletal actin (G^I) mRNAs analysed by whole-mount in situ hybridisation. The anti-sense RNA probes used in RNAse protection assays have been labeled with UTP-digoxygenin. (A,D,G) Stage 12 embryo (approximately 20 somites) : tMLC1F transcripts are present in the most rostral somites while tMLCF3 and skeletal actin transcripts are still undetectable. (B,E,H) Stage 17 embryo (approximately 45 somites) : tMLC1F labels more caudal newly formed somites compared to tMLC3F and skeletal actin. (C,F,I) Stage 21 embryo (segmentation is complete): all the chevron-shaped myotomes are completely stained for the three genes.
non muscle tissues (Fig. 2A,B). Thus, the restricted expression pattern of the tMLC1F , tMLC3F , and skeletal actin genes to fast skeletal muscle con¢rm their identity. The spatio-temporal expression of the three genes in the developing embryo was analysed by whole-mount in situ hybridisation. None of the transcripts were detected before stage 12 (20 somites) (data not shown). Therefore, in agreement with the situation found in mammals and amphibian, the onset of expression of these structural genes during ¢sh development is later than that of the myogenic regulatory bHLH genes like TMyoD or Tmyogenin whose expression begins before somitogenesis (TMyoD) or in forming somites (TMyoD2 and Tmyogenin) [14]. In stage 12 embryo, only tMLC1F mRNA is detected and its expression is restricted to the most rostral somites (Fig. 3A). The expression of tMLC1F progresses caudally as the somites formed in a rostral to caudal wave (Fig. 3A,B). tMLC3F expression is ¢rst detected in stage 16 embryo (data not shown) and it labels the most rostral somites but not the newly formed posterior somites in stage 17 embryo (Fig. 3E). The expression of skeletal actin
BBAEXP 91549 29-5-01
mRNA is similar to that of tMLC3F mRNA following a rostral^caudal gradient as the somites are forming (Fig. 3H). The delayed expression of tMLC3 F and skeletal actin genes compared with that of tMLC1F gene is apparent especially in stage 17 embryo, where only the rostral somites express tMLC3F and skeletal actin transcripts while the expression of tMLC1F transcripts extends more caudally in more newly formed somites (Fig. 3B,E,H). When the segmentation is complete (stage 21) the three genes are expressed in all chevron-shaped myotomes but none of them is expressed at a detectable level in embryonic heart (Fig. 3C,F,I). In the mouse, MLC1F transcripts can ¢rst be detected between 8.5 and 9.5 days post coitum (p.c.) when about 25 somites are formed [10]. This marks a striking similitude with the temporal expression of the MLC1F gene in trout embryo. This suggests that the MLC1F regulatory regions, responsible for the temporal expression of the gene, have been kept under evolutionary constraints between mammals and ¢sh. In contrast, the expression of the MLC3F gene is markedly delayed in mouse muscle development
Cyaan Magenta Geel Zwart
142
P. Thie¨baud et al. / Biochimica et Biophysica Acta 1519 (2001) 139^142
when compared to trout development as MLC3F transcripts are not detected before the end of somitogenesis (i.e., 15 days p.c. in mouse embryo) when there are about 60 somites [10]. Moreover, MLC3F transcripts are expressed in mouse embryonic heart [15] but this is not the case in trout embryo or in Xenopus as we have shown before [6]. Our analysis reveals also a striking di¡erence between teleost and mammal skeletal muscle development with respect to the skeletal actin gene expression. In mouse, skeletal actin gene expression, although weak, precedes MLC1F gene expression while in trout there is an opposite situation ([10]; this study). This makes the MLC1F gene a very early and sensitive marker for fast skeletal muscle di¡erentiation in trout. Our studies revealed that the asynchronous temporal expression of muscle genes in the developing vertebrate embryo has evolved di¡erently between species and probably re£ects intrinsic species speci¢c events such as the gradual buildup of di¡erent levels of regulatory factors within the muscle cell. We thank David Lepetit for help with cDNA library screening and Michelle Olive for critical reading of the manuscript. This work was supported by the Centre National de La Recherche Scienti¢que, the Institut National de la Sante¨ et de la Recherche Me¨dicale and a grant from The Association Franc°aise contre les Myopathies.
BBAEXP 91549 29-5-01
References [1] E. Bandman, Dev. Biol. 1 (1992) 273^283. [2] P.J.R. Barton, M. Buckingham, Biochem. J. 231 (1985) 249^261. [3] B. Robert, P. Daubas, M.-A. Akimenko, A. Cohen, I. Garner, J.-L. Guenet, M. Buckingham, Cell 39 (1984) 129^140. [4] M. Periasamy, E. Strehler, L.I. Gar¢nkel, R.M. Gubits, N. Ruiz-Opazo, B. Nadal-Ginard, J. Biol. Chem. 259 (1984) 13595^ 13604. [5] Y.-I. Nabeshima, Y. Fujii-Kuriyama, M. Muramatsu, K. Ogata, Nature 308 (1984) 333^338. [6] N. The¨ze¨, S. Hardy, R. Wilson, M.-R. Allo, T. Mohun, P. Thie¨baud, Dev. Biol. 171 (1995) 352^362. [7] L. Dalla Libera, E. Carpene, J. Theibert, J. Collins, J. Muscle Res. Cell Motil. 12 (1991) 366^371. [8] Y. Hirayama, S. Kanoh, M. Nakaya, S. Watabe, J. Exp. Biol. 200 (1997) 693^701. [9] D. Sassoon, I. Garner, M. Buckingham, Development 104 (1988) 155^164. [10] G.E. Lyons, M. Ontell, R. Cox, D. Sassoon, M. Buckingham, J. Cell Biol. 111 (1990) 1465^1476. [11] M. Ontell, M.P. Ontell, M.M. Sopper, R. Mallonga, G. Lyons, M. Buckingham, Development 117 (1993) 1435^1444. [12] T.J. Mohun, S. Brennan, N. Dathan, S. Fairman, J.B. Gurdon, Nature 311 (1983) 716^721. [13] W.W. Ballard, J. Exp. Zool. 184 (1973) 7^26. [14] J.-M. Delalande, P.-Y. Rescan, Dev. Genes Evol. 209 (1999) 432^ 437. [15] R.G. Kelly, P.S. Zammit, V. Mouly, G. Butler-Browne, M.E. Buckingham, J. Mol. Cell. Cardiol. 30 (1998) 1067^1081.
Cyaan Magenta Geel Zwart