Developmental Biology 245, 213–223 (2002) doi:10.1006/dbio.2002.0638, available online at http://www.idealibrary.com on
Essential and Redundant Functions of the MyoD Distal Regulatory Region Revealed by Targeted Mutagenesis Jennifer C. J. Chen, Rageshree Ramachandran, and David J. Goldhamer 1 Department of Cell and Developmental Biology, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19104
Transgenic analyses have defined two MyoD enhancers in mammals, the core enhancer and distal regulatory region (DRR); these enhancers exhibit complementary activities and together are sufficient to recapitulate MyoD expression in developing and mature skeletal muscle. DRR activity is restricted to differentiated muscle and persists postnatally, suggesting an important role in maintaining MyoD expression in myocytes and muscle fibers. Here, we use targeted mutagenesis in the mouse to define essential functions of the DRR in its normal chromosomal context. Surprisingly, deletion of the DRR resulted in reduced MyoD expression in all myogenic lineages at E10.5, at least 1 day prior to detection of DRR activity in limb buds and branchial arches of transgenic mice. At later embryonic and fetal stages, however, no defect in MyoD expression was observed, indicating that the DRR is dispensable for regulating MyoD during muscle differentiation. Expression analyses in wild-type and Myf-5 mutant embryos also indicate that the DRR is not an obligate target for Myf-5and Pax-3-dependent regulation. In contrast to embryonic and fetal stages, deletion of the DRR resulted in a pronounced reduction in MyoD mRNA levels in adults, showing a functional requirement for DRR activity in mature muscle. These data reveal essential and redundant functions of the DRR and underscore the importance of loss-of-function enhancer analyses for understanding cis transcriptional circuitry. © 2002 Elsevier Science (USA) Key Words: MyoD; enhancer; myogenesis; gene targeting; Cre recombination; somite; transcription; redundancy.
INTRODUCTION In vertebrate embryos, commitment of cells to the skeletal muscle lineage requires either MyoD or Myf-5, two members of the muscle-specific basic helix–loop– helix transcription factor family. In the absence of these muscle regulatory factors (MRFs), myogenesis fails to occur (Rudnicki et al., 1993), and presumptive myogenic cells are diverted to nonmuscle lineages (Tajbakhsh et al., 1996; Kablar et al., 1999). During somite development, MyoD and Myf-5 are activated in a precise temporal and spatial pattern in response to signaling molecules produced by the notochord, neural tube, surface ectoderm, and lateral plate mesoderm (reviewed by Borycki and Emerson, 2000). Although considerable progress has been made in identifying these signaling proteins, which include Sonic hedgehog (Shh), Wnt proteins, and BMPs and their antagonists, little 1
To whom correspondence should be addressed. Fax: (215) 8989871. E-mail:
[email protected]. 0012-1606/02 $35.00 © 2002 Elsevier Science (USA) All rights reserved.
is known of how MyoD and Myf-5 transcription is initiated and maintained in response to these signals. Previous gain- and loss-of-function experiments have begun to define the hierarchical relationship between MyoD and other transcriptional regulatory factors expressed in somites. For example, MyoD expression is abrogated in the body muscles of mouse embryos mutant for both Myf-5 and the paired box transcription factor, Pax-3, indicating that these factors function upstream of MyoD (Tajbakhsh et al., 1997). In addition, a number of transcription factors and cofactors, including Pax-3, Six1, Dach2, and Eya2, have been shown to activate MyoD when overexpressed in somite explant cultures (Maroto et al., 1997; Heanue et al., 1999). It remains to be determined, however, whether any of these factors regulate MyoD through direct transcriptional interactions with MyoD regulatory sequences. Mutations in Myf-5 (Tajbakhsh et al., 1996; Grass et al., 1996; Tajbakhsh et al., 1997), Pax-3 (Franz et al., 1993; Tajbakhsh et al., 1997; Borycki et al., 1999b), and Paraxis (Burgess et al., 1996; Wilson-Rawls et al., 1999), for
213
214
Chen, Ramachandran, and Goldhamer
example, disrupt somite patterning, raising the possibility that effects on MyoD expression are secondary to their roles in somite morphogenesis. Indeed, in Myf-5 and Pax-3 mutant embryos, muscle precursor cells exhibit migratory defects and may not be positioned properly to receive myogenic signals (Tajbakhsh et al., 1996, 1997). As an essential step toward defining the regulatory pathways that govern myogenic determination and MyoD activation, we have used transgenic mouse analysis to characterize mammalian enhancer sequences that control MyoD expression. These studies showed that 24 kb of human MyoD 5⬘ flanking sequence is sufficient to recapitulate endogenous MyoD expression in mouse embryos (Chen et al., 2001), as well as in normal and injured adult skeletal muscles (unpublished observations). Two muscle-specific enhancers with distinct but overlapping specificity have been characterized within MyoD flanking sequence in humans and mice. A highly conserved core enhancer sequence approximately 20 kb upstream of MyoD is sufficient for early MyoD activation in somites, limb buds, and branchial arches (Goldhamer et al., 1995; Faerman et al., 1995; Kablar et al., 1998). Linker-scanner mutagenesis of this 258-bp sequence has identified multiple cis elements required for its activity, including an enhancer region specifically required in myotomal lineages that may represent a target for Myf-5-dependent regulation of MyoD (Kucharczuk et al., 1999; unpublished observations). The core enhancer also exhibits MRF-independent transcriptional activity (Kablar et al., 1999), likely reflecting a role in initiating MyoD expression in a subpopulation of muscle progenitor cells early in the myogenic program. However, the core enhancer is not sufficient to maintain MyoD expression in differentiated skeletal muscle, being down-regulated in fetal and neonatal muscle and essentially inactive in adult muscle (Faerman et al., 1995; unpublished observations). Five kilobases upstream of MyoD is a second enhancer, the distal regulatory region (DRR), which is unrelated in sequence to the core enhancer and exhibits largely complementary activity in transgenic mice (Tapscott et al., 1992; Goldhamer et al., 1995; Asakura et al., 1995; Kablar et al., 1998; Chen et al., 2001). DRR activity is entirely MRFdependent (Kablar et al., 1999) and is restricted to differentiated skeletal muscle in vivo (Kablar et al., 1997), which is reflected as a significant delay in DRR-driven transgene expression in several sites of myogenesis relative to the endogenous MyoD gene (Asakura et al., 1995). Unlike the core enhancer, the DRR remains active in adult muscle, showing a similar expression profile as the endogenous MyoD gene (Hughes et al., 1993). Collectively, these data indicate that the core enhancer and DRR have distinct activation and maintenance functions, respectively, that collaborate to establish the dynamic pattern of MyoD expression in embryonic and adult skeletal muscle (Chen et al., 2001). Although transgenic studies are a powerful means to characterize activities of transcriptional regulatory sequences, transgenesis is inherently limited in its ability to
assess required functions of such elements, as restricted sequences are used that are inserted at heterologous genomic loci (reviewed by Fiering et al., 1999). To circumvent these limitations, we have used targeted mutagenesis in mice to determine the essential functions and regulatory relationships of the MyoD enhancers in their normal chromosomal context. Here, we report the consequences of targeted deletion of the DRR. Replacement of the DRR with the neo-positive selection marker (DRR neo allele) resulted in a dramatic reduction in MyoD expression in all muscle lineages and at all pre- and postnatal stages tested. Removal of the neo cassette by Cre-mediated recombination (DRR loxP allele), however, produced stage-specific defects in MyoD expression. Analysis of E9.75 embryos showed that the DRR is dispensable for the initiation of MyoD expression in branchial arches and somites. Surprisingly, however, MyoD expression was reduced in E10.5 DRR loxP/loxP embryos, revealing an essential DRR function earlier in the myogenic program than expected from previous transgenic work. The DRR is also required to maintain MyoD expression in adult skeletal muscle, as MyoD levels were reduced by approximately 65% in DRR loxP/loxP leg muscles. In contrast, MyoD expression was not affected in differentiating muscle fibers of embryos and fetuses. In addition, in Myf-5 mutant embryos, Pax-3-dependent recovery of MyoD expression (Tajbakhsh et al., 1997) occurred with normal kinetics in the absence of the DRR, showing that the DRR is not an obligate target for Pax-3-dependent MyoD expression. These data reveal essential and redundant DRR functions in developing and mature skeletal muscle.
MATERIALS AND METHODS Targeted Mutagenesis and Generation of Chimeric Mice A 1.9-kb ApaI/XhoI genomic fragment (⫺5870 to ⫺4000 relative to the MyoD transcriptional start site) containing the 714-bp DRR (Tapscott et al., 1992) was targeted for deletion by homologous recombination. Approximately 11 kb of 129/SvJ mouse genomic DNA containing the DRR was isolated by screening a gt11 library (Clontech), cloned into pBluescript KS⫹ (Stratagene), and mapped by restriction enzyme digests using standard methods. The 1.9-kb fragment was replaced with a PGKneo-positive selection cassette flanked by loxP sites (from ploxPneo-1; kindly provided by Dr. Marisa Bartolomei), cloned in a 3⬘ to 5⬘ orientation to avoid possible read-through transcription into the MyoD gene. The targeting construct (Fig. 1) has 5⬘ and 3⬘ homology arms of approximately 5 and 3 kb, respectively, as well as an HSVtk-negative selection cassette adjacent to the 5⬘ homology arm. Approximately 4 ⫻ 10 7 GS1 129/SvJ ES cells (Incyte) were electroporated with 40 g of NotI-linearized targeting vector (800 V, 3 F) and plated on a monolayer of mitomycin C (Sigma)-treated, G418-resistant primary embryonic fibroblasts on 0.1% gelatin (Sigma)-coated tissue culture dishes in ES media [15% FBS (Hyclone), 100 units/ml Penicillin (Gibco), 100 g/ml Streptomycin (Gibco), 2 mM L-glutamine (Gibco), 100 M nonessential amino acids (Gibco), 10 mM Hepes (Gibco), 100 M -mercaptoethanol (Sigma), and 10 3 units/ml ESGRO (Calbiochem) in high glucose
© 2002 Elsevier Science (USA). All rights reserved.
215
Essential and Redundant Functions of the MyoD DRR
DMEM (Gibco)]. After 24 h, selection media was added [ES media containing 200 g/ml Geneticin (LTI) and 1 M ganciclovir] and cells were cultured for 8 –12 days. Resistant ES colonies were expanded and targeted clones identified by Southern analysis of XbaI-digested DNA with a radioactive probe to a 1.1-kb XbaI/ EcoRI fragment that lies outside of the 5⬘ homology arm (Fig. 1). Targeting was confirmed by Southern analysis of XhoI-digested DNA with a HindIII/StuI 3⬘ probe (data not shown). Targeting frequency was approximately 1 in 150 G418/ganciclovir-resistant clones. To remove the neo cassette, targeted ES clones were electroporated with 25 g of the Cre expression plasmid pBS185 (Sauer and Henderson, 1990; kindly provided by Dr. Brian Sauer) and plated at clonal density. Individual colonies were expanded and screened by Southern blotting with flanking probes to test for the loss of the PGKneo gene and to verify the fidelity of the recombined locus. PCR was used to verify lack of integration of the excised neo cassette and the Cre plasmid. Targeted ES clones, both with (DRR neo) and without (DRR loxP) the neo cassette, were injected into C57 blastocysts by the University of Pennsylvania Transgenic and Chimeric Mouse Facility. Chimeric mice were crossed to C57Bl/6J mice, and germline transmission was assessed by Southern analysis of tail DNA.
electrophoresed on formaldehyde agarose gels, transferred to GeneScreen nylon membrane (NEN), and hybridized with a 32P-labeled full-length mouse MyoD cDNA probe (Radprime kit; Gibco) in PerfectHyb Plus (Sigma) by using standard methods. After exposing blots to phosphor screens and scanning with a Storm Scanner 860 (Molecular Dynamics), blots were reprobed with an 800-bp EcoRI fragment from the rat GAPDH cDNA and rescanned. MyoD hybridization signals were quantified by using ImageQuant software (Molecular Dynamics) and normalized to GAPDH. Analyses were performed at least three times at each time point.
Mouse Breeding and Embryo Genotyping
Generation of DRR Mutant ES Cells and Chimeric Mice
For in situ and Northern analyses, heterozygous DRR mutant mice were mated and litters harvested at various stages with noon of the day of plug considered E0.5. Somite number was used for more accurate staging of embryos from E9.75 to E11.5, according to Rugh (1991). To obtain DRR loxP;Myf-5 double mutant embryos, DRR loxP/loxP mice were interbred with Myf-5 ⫹/⫺ mice (Braun et al., 1992) to generate DRR ⫹/loxP;Myf-5 ⫹/⫺ mice, which were subsequently crossed to DRR loxP/loxP mice to generate DRR loxP/loxP;Myf-5 ⫹/⫺ mice. Embryos used for analysis were generated by intercrossing DRR ⫹/loxP;Myf-5 ⫹/⫺ mice, or by crossing DRR ⫹/loxP;Myf-5 ⫹/⫺ and DRR loxP/loxP;Myf-5 ⫹/⫺ mice. DNA was isolated from yolk sacs and genotyping performed by Southern hybridization as described above.
Whole-Mount in Situ Hybridization E9.75 to E12.5 embryos were fixed in freshly prepared 4% paraformaldehyde in PBS, and processed for in situ hybridization with a digoxigenin-labeled MyoD RNA probe as previously described (Chen et al., 2001). Embryos within a comparison group were processed simultaneously and developed for identical times to allow direct comparisons of staining intensities. In order to detect small differences in staining intensities, embryos were inspected frequently during the staining process, and occasionally stained in a fourfold dilution of NBT/BCIP and X-phosphate (Roche).
RNA Isolation and Northern Analysis Tissues were quick-frozen in liquid nitrogen, and total RNA was isolated by either the LiCl-urea method [Auffray and Rougeon, 1980; E11.5 embryos, decapitated (to increase the representation of muscle-specific mRNAs) E15.5 embryos, and 7 day neonatal pooled forelimb and hindlimb muscles] or the guanidine isothiocyanateCsCl method (Chirgwin et al., 1979; adult tibialis anterior, gastrocnemius, or pooled hindlimb muscles) using a Polytron homogenizer for tissue dissociations. Total RNA (5–10 g) was
Photomicroscopy Embryo whole-mount images were photographed by using a Hamamatsu C5810 color video camera and Nikon SMZU stereomicroscope under dark-field illumination. Identical light and camera conditions were used to capture images of littermate embryos processed side by side for in situ hybridization. Images were captured and assembled by using Adobe Photoshop and Illustrator, respectively.
RESULTS
To define essential functions of the mouse DRR, the minimal genomic element shown to direct muscle-specific expression in transgenic mice was deleted (Asakura et al., 1995). This 1.9-kb ApaI/XhoI fragment contains the originally defined 714-bp DRR, which directs muscle-specific expression in cell culture (Tapscott et al., 1992) but has not been assayed in vivo. A standard replacement strategy was used whereby a PGKneo cassette was inserted in place of the 1.9-kb DRR fragment (Fig. 1A). As cases of long-range transcriptional inhibition by the neo cassette have been well-documented (reviewed by Olson et al., 1996; Fiering et al., 1999), the neo cassette was flanked by loxP sites to allow its excision by Cre-mediated recombination (Sauer and Henderson, 1988). A single transcriptionally inert loxP site (Arango et al., 1999) was present at the DRR locus following Cre recombination. Chimeric mice were generated from DRR-targeted ES cells, and germ line transmission was achieved for both DRR neo and DRR loxP mutations. Mutant lines were maintained by interbreeding to C57Bl/6J mice, and are in a mixed B6, 129 background. Crosses of heterozygous mice generated viable and fertile DRR neo and DRR loxP homozygous null adult offspring at expected Mendelian ratios (data not shown). In contrast, MyoD null mice produced by a heterozygous cross exhibit reduced survival at weaning (Rudnicki et al., 1992).
DRR neo Mice Exhibit a Pronounced Decrease in MyoD Expression in All Myogenic Populations Whole-mount in situ hybridization and Northern analysis revealed a dramatic decrease in MyoD expression in
© 2002 Elsevier Science (USA). All rights reserved.
216
Chen, Ramachandran, and Goldhamer
FIG. 1. (A) Targeting strategy for the mouse DRR. The 1.9-kb ApaI/XhoI fragment, which contains the DRR, was replaced by a floxed PGKneo cassette. The targeting vector (pDRRneotk) contains a 5-kb 5⬘ homology arm delineated by the EcoRI and ApaI restriction sites, and a 3-kb 3⬘ homology arm delineated by the XhoI and HindIII restriction sites (ApaI and XhoI sites were destroyed during subcloning). The structure of the locus following homologous recombination (DRR neo) and Cre-mediated excision of the neo cassette (DRR loxP) are shown. Diagnostic bands generated using the 5⬘ probe on XbaI-digested DNA are indicated above the locus diagrams. A 3⬘ probe was also used to verify the integrity of the locus following homologous recombination. (B) Representative Southern blot using the 5⬘ probe on XbaI-digested DNA. DNA was isolated from offspring of a cross of heterozygous DRR loxP (lanes 1–7) and DRR neo (lanes 8 –14) mice. All genotypic classes were generated at expected Mendelian ratios.
DRR neo embryos and adult muscle. Embryos exhibited a generalized reduction in MyoD expression in all muscle populations and at all developmental stages examined (E10.5–E15.5), although the timing of MyoD expression was not affected (data not shown). Quantitative Northern blots of RNA isolated from E15.5 embryos and neonatal and adult muscle revealed an 80% reduction in MyoD mRNA levels. Induction of MyoD expression in crush-injured muscle also was severely curtailed (data not shown).
Cre-Mediated Excision of the Neo Cassette Restores Near-Normal Levels of MyoD Expression in Embryos The dramatic effect on MyoD expression of the DRR neo mutation could result from the deletion of the DRR, inhibitory effects of the neo cassette, or both. To distinguish between these possibilities, MyoD expression was analyzed in DRR loxP embryos by northern hybridization (Fig. 2).
© 2002 Elsevier Science (USA). All rights reserved.
217
Essential and Redundant Functions of the MyoD DRR
FIG. 2. DRR loxP/loxP embryos exhibit wild-type MyoD mRNA levels at E11.5 and E15.5. Total RNA was isolated from wild-type and DRR mutant E11.5 embryos and decapitated E15.5 fetuses, and consecutively hybridized with MyoD- and GAPDH-specific, 32P-labeled probes. MyoD hybridization signals were quantified and normalized to GAPDH levels by using a Storm Scanner 860 and ImageQuant software.
Interestingly, MyoD mRNA levels were similar in wildtype, heterozygous, and homozygous null embryos at E11.5 and E15.5, showing that the severe, constitutive expression defect observed in DRR neo mutants is the result of inhibition by the neo cassette, and not deletion of the DRR. Thus, the DRR is sufficient (Hughes et al., 1993; Asakura et al., 1995; Kablar et al., 1997) but not required to maintain MyoD expression in differentiating embryonic and fetal muscles. Whole-mount in situ hybridization was performed on DRR loxP embryos to determine whether the DRR is required in a specific myogenic cell population that would not be discernible by Northern analysis. MyoD expression was observed in epaxial and hypaxial myotome subdomains, in limb buds, and in branchial arches (Fig. 3), showing no lineage-specific defects in MyoD expression. Interestingly, however, a relatively small but consistent decrease in staining intensity was observed in all myogenic populations at E10.5. The decrease in MyoD expression in the limb buds and branchial arches (Figs. 3B and 3F) reveals that the DRR is necessary but not sufficient (Asakura et al., 1995; Chen et al., 2001) for normal MyoD expression in these myogenic populations at this stage. In the mouse, MyoD mRNA is first detected by whole-mount in situ hybridization in the mandibular arch and interlimb somites at E9.75 (Tajbakhsh et al., 1997; Chen et al., 2001). We compared MyoD expression in wild-type and DRR loxP/loxP embryos at E9.75 to determine whether the initiation of MyoD expression was impaired by deletion of the DRR. No consistent difference in MyoD expression was observed between wild-type and DRR null embryos based on both staining intensity and the craniocaudal extent of somite expression (Figs. 3A and 3E), indicating that the earliest induction of MyoD expression is not dependent on the DRR. MyoD expression in the myotome is thought to be directed by the complementary activities of the core enhancer and DRR; core enhancer transgenes are active in the
medial and lateral aspects of the myotome (Faerman et al., 1995; Goldhamer et al., 1995; Kablar et al., 1998), whereas DRR transgenes are strongly expressed only in the central myotome at E10.5 and E11.5 (Asakura et al., 1995; Kablar et al., 1998; Chen et al., 2001). Surprisingly, however, the DRR is not required for MyoD expression in the central myotome at these stages (Figs. 3B, 3C, 3F, and 3G), suggesting that additional cis elements direct expression to this region normally, or can provide this function in the absence of the DRR.
The DRR Is Not Required for Pax-3-Dependent MyoD Expression Embryos mutant for both Myf-5 and Pax-3 fail to activate MyoD in the body proper, and are essentially devoid of trunk and limb musculature (Tajbakhsh et al., 1997). In embryos mutant only for Myf-5, MyoD expression in myotomes is delayed by approximately 2 days, eventually recovering through a Pax-3-dependent mechanism (Tajbakhsh et al., 1997). To determine whether the DRR is essential for Pax-3-dependent MyoD expression, embryos mutant for both Myf-5 and the DRR were analyzed by in situ hybridization. Deletion of the DRR did not affect the timing or the extent of Pax-3-dependent MyoD expression (Fig. 4). As expected, MyoD expression in myotomes was delayed in Myf-5 ⫺/⫺ (data not shown) and DRR ⫹/loxP;Myf-5 ⫺/⫺ embryos, with recovery beginning in the lateral myotome by E11.5 (Fig. 4B; Tajbakhsh et al., 1997). Similarly, MyoD expression was observed in the lateral myotomal domains of DRR loxP/loxP;Myf-5 ⫺/⫺ embryos at E11.5 (Fig. 4C). Staining was slightly reduced in DRR loxP/loxP;Myf-5 ⫺/⫺ embryos relative to DRR ⫹/loxP;Myf-5 ⫺/⫺ embryos (Figs. 4B and 4C), reflecting the minor quantitative effect of the DRR mutation that was more apparent at E10.5 (Fig. 3). By E12.5, both heterozygous and null DRR loxP embryos exhibited robust MyoD expression comparable to that of Myf-5 ⫹/⫹ embryos (Figs. 4D–
© 2002 Elsevier Science (USA). All rights reserved.
218
Chen, Ramachandran, and Goldhamer
FIG. 3. Whole-mount in situ hybridization for MyoD mRNA in wild-type (A–D) and DRR loxP/loxP (E–H) littermates from E9.75 to E12.5. MyoD expression was indistinguishable in wild-type and mutant embryos at E9.75 (A, E). Weak staining was observed in the branchial arches (red arrows) and to a similar craniocaudal extent in interlimb somites. Black arrows indicate the position of the forelimb buds. At E10.5 (B, F), a decrease in staining intensity was consistently observed in DRR loxP/loxP embryos [compare forelimb buds (black arrows) and branchial arches (red arrows) in B and F], although the spatiotemporal patterning of MyoD expression was preserved. Prolonged development revealed staining in the branchial arches of DRR loxP/loxP embryos (data not shown). By E11.5 (C, G), MyoD was expressed at near wild-type levels in DRR loxP/loxP embryos and staining was observed in all major myogenic populations. At E12.5 (D, H), MyoD staining intensity and pattern were indistinguishable in wild-type and DRR loxP/loxP embryos.
4F), although developing trunk musculature was disorganized, as reported previously (Figs. 4E and 4F; Braun et al., 1992). We conclude that the DRR is not required for Pax-3-dependent MyoD expression in Myf-5 mutant embryos.
The DRR Is Essential for Normal MyoD Expression in Adults We tested the requirement for DRR function in postnatal muscle by measuring MyoD mRNA levels in neonatal and adult leg muscles. Deletion of the DRR had only a slight effect (average decrease of 15%) on MyoD expression in pooled 7-day neonatal limb muscles (Figs. 5A and 5B). In contrast, MyoD mRNA levels were reduced by 65% in the adult tibialis anterior muscle (Figs. 5A and 5B). A similar reduction was observed in the gastrocnemius and pooled
hindlimb muscles of 24-week adults (data not shown). The stage-specificity of the DRR loxP phenotype strongly suggests that the pronounced effect on MyoD expression is not a consequence of disrupted chromatin structure or the presence of a loxP sequence at the MyoD locus. These data demonstrate that the DRR is required to maintain normal levels of MyoD expression in adult muscle and indicate that distinct regulatory mechanisms control MyoD expression in embryonic and mature musculature.
DISCUSSION An inherent limitation of transgenic studies is their inability to elucidate required functions of transcriptional elements in a native chromosomal context, due to the
© 2002 Elsevier Science (USA). All rights reserved.
219
Essential and Redundant Functions of the MyoD DRR
FIG. 4. Deletion of the DRR did not affect Pax-3-dependent rescue of MyoD expression in Myf-5 mutant embryos. (A) Embryos heterozygous for both DRR loxP and Myf-5 exhibit MyoD expression throughout the mediolateral extent of the myotome at E11.5 (black arrows), comparable to wild-type embryos. (B, C) At E11.5, DRR ⫹/loxP; Myf-5 ⫺/⫺ and DRR loxP/loxP; Myf-5 ⫺/⫺ embryos show weak MyoD expression in the ventrolateral myotomes of interlimb somites (black arrows), representing an early phase of Pax-3-dependent compensation. A slight decrease in MyoD staining was observed in DRR loxP/loxP embryos (most evident by comparing regions posterior to the black arrows in B and C), which is unrelated to the Myf-5 mutant genotype (see Fig. 3). MyoD expression in limb buds (red arrows) and branchial arch-derived myogenic cells (arrowheads) of E11.5 embryos was essentially unaffected by the Myf-5 mutation, consistent with previous observations (Tajbakhsh et al., 1997; Kablar et al., 1997). (D–F) By E12.5, MyoD expression levels were normal in myotomal muscles of Myf-5 mutant embryos both in the presence (E) and absence (F) of a functional DRR allele. Disorganization of trunk musculature in Myf-5 mutant embryos (arrows) is the result of rib defects associated with the Myf-5 m1 allele (Braun et al., 1992).
necessarily limited sequences used as transgenes and their integration at heterologous sites in the genome. While reporter gene assays in transgenic mice have shown that the combined activities of the core enhancer and DRR are sufficient to regulate MyoD expression in embryonic and adult skeletal muscle, whether these enhancers serve unique and necessary functions and possible redundant networks governing MyoD transcription had not been explored previously. In the present study, targeted mutagenesis revealed both essential and redundant DRR functions.
Long-Range Disruption of MyoD Expression by the Neo Cassette Replacement of the DRR with the neo cassette resulted in a dramatic reduction in MyoD expression to approximately 20% of wild-type levels at all embryonic and postnatal stages tested. Comparable inhibition of MyoD expression also was observed when the neo cassette replaced the core enhancer, approximately 20 kb upstream of MyoD (unpublished observations). Because long-range interference of gene expression by positive selection cassettes such as
© 2002 Elsevier Science (USA). All rights reserved.
220
Chen, Ramachandran, and Goldhamer
neo have been observed in a number of enhancer- and gene-targeting studies (reviewed by Olson et al., 1996; Fiering et al., 1999), neo was flanked by loxP sequences to allow its removal by Cre-mediated recombination. At most embryonic stages, removal of the neo cassette restored MyoD expression essentially to wild-type levels (see below), demonstrating that insertion of neo, and not deletion of the DRR, was responsible for the constitutive reduction in MyoD mRNA levels. Genetic studies have shown that MyoD heterozygotes, which express 50% of normal MyoD mRNA levels, can support normal myogenesis in the absence of Myf-5 (Rudnicki et al., 1992, 1993). In contrast, one wild-type Myf-5 allele is not sufficient to support normal myogenesis in a MyoD mutant background, indicating different threshold requirements for these key regulators of myogenesis (Rudnicki et al., 1993). Preliminary analysis has shown that late gestational DRR neo/neo;Myf-5 ⫺/⫺ fetuses exhibit histologically normal skeletal muscle (unpublished data), suggesting that 20% of normal MyoD mRNA levels are sufficient to support embryonic myogenesis. Additional analysis, including further titration of MyoD levels by combining MyoD knockout and DRR neo alleles, will provide important insights into MyoD dosage requirements for embryonic myogenesis. Kaul et al. (2000) recently developed a viable Myf-5 loxP knockout strain that lacks rib defects characteristic of the original mutation. Thus, MyoD knockout, DRR neo, and DRR loxP alleles will provide an allelic series with which to study genetic interactions between MyoD and Myf-5 in regulating postnatal muscle growth and regeneration (see Seale and Rudnicki, 2000).
The DRR Is Not Required in Differentiating Embryonic Skeletal Muscle Based on transgenic studies, a simple prediction was that DRR loxP embryos would exhibit reduced MyoD expression in differentiating muscle, since expression of DRRlacZ transgenes are restricted to differentiating muscle and core enhancer activity is down-regulated as skeletal muscle differentiates and matures (reviewed in Chen et al., 2001). In the somite, for example, core enhancer activity predominates in the younger myocytes of the medial and lateral myotomal domains, but is down-regulated in older myocytes of the central myotome (Faerman et al., 1995; Kablar et al., 1998; see Ordahl et al., 2000 and Cinnamon et al., 2001 for discussions of myotome formation). In contrast, DRR activity is largely complementary, and is the only known MyoD regulatory element that exhibits robust activity in the central myotome (Asakura et al., 1995; Kablar et al., 1998; Chen et al., 2001). Deletion of the DRR, however, did not reduce MyoD expression in the central myotome at E10.5 and E11.5 relative to other myogenic populations, nor was expression significantly diminished in differentiating muscle at any embryonic stage tested. These results demonstrate that the DRR is dispensable for MyoD expression in differentiating skeletal muscle, and suggest
that redundant transcriptional networks regulate MyoD in embryos. Surprisingly, although DRR transgene activity is not detected in limb buds and branchial arches at E10.5 (Asakura et al., 1995), DRR loxP/loxP embryos exhibited a consistent decrease in MyoD expression in somites, limb buds, and branchial arches at this stage. Thus, the DRR plays an essential role early in the myogenic program, likely cooperating with the core enhancer and perhaps other elements to confer wild-type expression levels. The DRR does not regulate the initiation of MyoD expression, however, as deletion of the DRR did not affect MyoD expression at E9.75, the stage at which MyoD expression is first detected by in situ hybridization.
The DRR Is Dispensable for Myf-5- and Pax-3Dependent MyoD Regulation Myf-5 and Pax-3 function in parallel genetic pathways to regulate MyoD (Tajbakhsh et al., 1997; Kablar et al., 1997; Borycki et al., 1999a), although their specific roles have not been elucidated. Both Myf-5 (Tajbakhsh et al., 1996) and Pax-3 (Bober et al., 1994; Williams and Ordahl, 1994; Daston et al., 1996; Tremblay et al., 1998) regulate cell migration of myogenic precursors, raising the possibility that their effects on MyoD expression are an indirect consequence of improper positioning of muscle precursor cells relative to myogenic signals (Daston et al., 1996; Tajbakhsh et al., 1996, 1997). Alternatively, these transcription factors may play a more direct role in MyoD transcriptional control, a notion consistent with Pax-3 overexpression studies (Maroto et al., 1997), and the fact that MyoD induction by Shh in somite explants is dependent on Myf-5 (Borycki et al., 1999a). In addition, both the core enhancer and the DRR contain multiple E-boxes (Tapscott et al., 1992; Goldhamer et al., 1995; Chen et al., 2001) and a potential binding site for the paired DNA binding domain of Pax factors (Phelan and Loeken, 1998). That deletion of the DRR did not phenocopy the MyoD expression defect in Myf-5 mutant embryos indicates that the DRR is not an obligate target for Myf-5 regulation of MyoD. Interestingly, deletion of a conserved E-box in the core enhancer results in a pattern of transgene expression that closely resembles endogenous MyoD expression in Myf-5 mutant embryos (Kucharczuk et al., 1999; unpublished observations) suggesting that Myf-5-dependent regulation of MyoD may function through the core enhancer, a possibility being addressed by targeted mutagenesis. In Myf-5 mutant embryos, MyoD expression in the trunk is entirely dependent on Pax-3 (Tajbakhsh et al., 1997). The DRR apparently can mediate Pax-3-dependent expression, as a DRR transgene was expressed in the myotomes of Myf-5 mutant embryos (Kablar et al., 1997). Targeted deletion of the DRR, however, did not affect the timing or extent of Pax-3-dependent MyoD expression, showing that the DRR is not an essential Pax-3 target and that other sequences can also mediate Pax-3-dependent regulation. A likely candidate is the core enhancer, as a core enhancer
© 2002 Elsevier Science (USA). All rights reserved.
221
Essential and Redundant Functions of the MyoD DRR
FIG. 5. Deletion of the DRR significantly reduces MyoD expression in adult muscle. (A) Representative northern analysis of MyoD expression in skeletal muscle of postnatal DRR loxP/loxP mice. Total RNA was isolated from pooled forelimb and hindlimb muscles of neonates and pooled tibialis anterior muscles from 12-week-old mice, hybridized to MyoD and GAPDH-specific probes, and quantified as in Fig. 2. Northern analysis of 24-week adult gastrocnemius and pooled hindlimb muscles gave similar results (data not shown). (B) Quantification of MyoD expression in neonatal and adult muscle. MyoD northern hybridization signals in DRR loxP muscle were normalized to wild-type values, and each bar represents the average of four independent experiments. Error bars, ⫾ SEM.
transgene was also expressed in Myf-5 mutants (Kablar et al., 1998). In this regard, it will be important to determine how two enhancers of unrelated sequence and distinct activities can function in the Pax-3-dependent transcriptional pathway, and whether Pax-3 directly binds these enhancers or functions by regulating genes involved in cell migration (reviewed by Birchmeier and Brohmann, 2000) or other processes upstream of MyoD expression.
The DRR Is Required to Maintain Normal MyoD Expression Levels in Adult Skeletal Muscle Deletion of the DRR resulted in a 65% reduction in MyoD expression in the adult tibialis anterior, gastrocnemius, and pooled hindlimb muscles. It is likely that this
mutation affects MyoD transcription in muscle fibers themselves, as MyoD is not expressed in quiescent satellite cells (Maley et al., 1994; Smith et al., 1994; YablonkaReuveni and Rivera, 1994; Cornelison and Wold, 1997), and few myoblasts are present in mature skeletal muscle (Snow, 1977). The present study did not address whether all muscle fiber types were affected similarly by the DRR mutation. In this regard, the endogenous MyoD gene as well as DRR transgenes are expressed at higher levels in fast glycolytic fibers than in slow fibers (Hughes et al., 1993); a requirement for the DRR in a subset of fast fiber types could potentially explain a reduction in MyoD mRNA levels of the magnitude observed. Nevertheless, these data demonstrate that maintenance of normal MyoD levels in mature skeletal muscle depends on DRR function, and point to differences in transcriptional pathways or compensatory mechanisms operative in immature and mature muscle. One similarity with regulation of MyoD in embryos, however, is a reliance on multiple regulatory pathways, as deletion of the DRR did not abrogate MyoD expression in adult muscle. A 1.9-kb DRR-containing fragment was targeted, as this is the minimum sequence that has been tested for DRR activity in transgenic mice (Asakura et al., 1995). Although the precise DRR sequences required for MyoD expression in adult muscle and early embryos remain to be delineated, we favor the view that the required regulatory information is contained within the originally defined 714-bp DRR (Tapscott et al., 1992). First, the 714-bp DRR exhibits muscle specificity in cell culture (Tapscott et al., 1992) and contains several motifs for known muscle regulatory factors (Tapscott et al., 1992; Chen et al., 2001). Second, it shows extended blocks of sequence conservation between humans and mice, whereas other sequences within the 1.9-kb fragment exhibit minimal conservation (Chen et al., 2001). Third, the 714-bp DRR sequence is subject to chromatin modification by MyoD in cell culture (Gerber et al., 1997), providing a possible means for maintaining MyoD gene expression by autoregulation (see Thayer et al., 1989). In this latter regard, it will be interesting to determine whether chromatin modification of the DRR occurs in vivo and whether this modification is MyoD-dependent.
Enhancer Redundancy Taken together with previous transgenic data, the present results indicate that redundancy is an integral aspect of the transcriptional circuitry regulating MyoD expression. We are currently investigating whether other enhancers exist with DRR-like specificity, or whether the core enhancer, despite its apparently distinct activities, can compensate in the absence of the DRR. Using similar targeting strategies, gene regulation by redundant cis elements has also been documented for -globin (Fiering et al., 1995), HoxD (Zakany et al., 1997; Beckers and Duboule, 1998), and the T cell receptor variable region genes (Monroe et al., 1999), raising the possibility that gene regulation by functionally
© 2002 Elsevier Science (USA). All rights reserved.
222
Chen, Ramachandran, and Goldhamer
interchangeable cis motifs is a common theme in development, as suggested by transgenic studies in a variety of species (reviewed by Arnone and Davidson, 1997; Dover, 2000). Redundancy of regulatory regions such as the DRR could result from the presence of multiple enhancers that can each act as targets for transcription factor mediators of a particular developmental pathway. Alternatively, functionally equivalent but distinct signaling and transcriptional pathways may each activate transcription through a unique DNA target. In this regard, enhancer-targeted mice will be valuable reagents to directly link Shh, Wnt, and other signaling pathways with transcriptional responses that regulate MyoD expression.
ACKNOWLEDGMENTS We thank Okechukwu Nwoko for excellent technical assistance, Dr. Jean Richa of the University of Pennsylvania Medical School Transgenic Facility for ES cell injections, and Christopher Watt for helpful comments on the manuscript. This work was supported by grants from the NIH (AR44878) and Muscular Dystrophy Association, and was done during the tenure of an Established Investigatorship from the American Heart Association.
REFERENCES Arango, N. A., Lovell-Badge, R., and Behringer, R. R. (1999). Targeted mutagenesis of the endogenous mouse Mis gene promoter: In vivo definition of genetic pathways of vertebrate sexual development. Cell 99, 409 – 419. Arnone, M. I., and Davidson, E. H. (1997). The hardwiring of development: Organization and function of genomic regulatory systems. Development 124, 1851–1864. Asakura, A., Lyons, G. E., and Tapscott, S. J. (1995). The regulation of MyoD gene expression: Conserved elements mediate expression in embryonic axial muscle. Dev. Biol. 171, 386 –398. Auffray, C., and Rougeon, F. (1980). Purification of mouse immunoglobulin heavy-chain messenger RNAs from total myeloma tumor RNA. Eur. J. Biochem. 107, 303–314. Beckers, J., and Duboule, D. (1998). Genetic analysis of a conserved sequence in the HoxD complex: Regulatory redundancy or limitations of the transgenic approach? Dev. Dyn. 213, 1–11. Birchmeier, C., and Brohmann, H. (2000). Genes that control the development of migrating muscle precursor cells. Curr. Opin. Cell Biol. 12, 725–730. Bober, E., Franz, T., Arnold, H. H., Gruss, P., and Tremblay, P. (1994). Pax-3 is required for the development of limb muscles: A possible role for the migration of dermomyotomal muscle progenitor cells. Development 120, 603– 612. Borycki, A. G., Brunk, B., Tajbakhsh, S., Buckingham, M., Chiang, C., and Emerson, C. P., Jr. (1999a). Sonic hedgehog controls epaxial muscle determination through Myf5 activation. Development 126, 4053– 4063. Borycki, A. G., and Emerson, C. P., Jr. (2000). Multiple tissue interactions and signal transduction pathways control somite myogenesis. Curr. Top. Dev. Biol. 48, 165–224. Borycki, A. G., Li, J., Jin, F., Emerson, C. P., and Epstein, J. A. (1999b). Pax3 functions in cell survival and in pax7 regulation. Development 126, 1665–1674.
Braun, T., Rudnicki, M. A., Arnold, H. H., and Jaenisch, R. (1992). Targeted inactivation of the muscle regulatory gene Myf-5 results in abnormal rib development and perinatal death. Cell 71, 369 –382. Burgess, R., Rawls, A., Brown, D., Bradley, A., and Olson, E. N. (1996). Requirement of the paraxis gene for somite formation and musculoskeletal patterning. Nature 384, 570 –573. Chen, J. C. J., Love, C. M., and Goldhamer, D. J. (2001). Two upstream enhancers collaborate to regulate the spatial patterning and timing of MyoD transcription during mouse development. Dev. Dyn. 221, 274 –288. Chirgwin, J. M., Przbyla, A. E., MacDonald, R. J., and Rutter, W. J. (1979). Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 18, 5294 –5299. Cinnamon, Y., Kahane, N., Bachelet, I., and Kalcheim, C. (2001). The sub-lip domain—a distinct pathway for myotome precursors that demonstrate rostral-caudal migration. Development 128, 341–351. Cornelison, D. D., and Wold, B. J. (1997). Single-cell analysis of regulatory gene expression in quiescent and activated mouse skeletal muscle satellite cells. Dev. Biol. 191, 270 –283. Daston, G., Lamar, E., Olivier, M., and Goulding, M. (1996). Pax-3 is necessary for migration but not differentiation of limb muscle precursors in the mouse. Development 122, 1017–1027. Dover, G. (2000). How genomic and developmental dynamics affect evolutionary processes. BioEssays 22, 1153–1159. Faerman, A., Goldhamer, D. J., Puzis, R., Emerson, C. P., Jr., and Shani, M. (1995). The distal human myoD enhancer sequences direct unique muscle-specific patterns of lacZ expression during mouse development. Dev. Biol. 171, 27–38. Fiering, S., Bender, M. A., and Groudine, M. (1999). Analysis of mammalian cis-regulatory DNA elements by homologous recombination. Methods Enzymol. 306, 42– 66. Fiering, S., Epner, E., Robinson, K., Zhuang, Y., Telling, A., Hu, M., Martin, D. I., Enver, T., Ley, T. J., and Groudine, M. (1995). Targeted deletion of 5⬘HS2 of the murine beta-globin LCR reveals that it is not essential for proper regulation of the beta-globin locus. Genes Dev. 9, 2203–2213. Franz, T., Kothary, R., Surani, M. A., Halata, Z., and Grim, M. (1993). The Splotch mutation interferes with muscle development in the limbs. Anat. Embryol. 187, 153–160. Gerber, A. N., Klesert, T. R., Bergstrom, D. A., and Tapscott, S. J. (1997). Two domains of MyoD mediate transcriptional activation of genes in repressive chromatin: A mechanism for lineage determination in myogenesis. Genes Dev. 11, 436 – 450. Goldhamer, D. J., Brunk, B. P., Faerman, A., King, A., Shani, M., and Emerson, C. P., Jr. (1995). Embryonic activation of the myoD gene is regulated by a highly conserved distal control element. Development 121, 637– 649. Grass, S., Arnold, H. H., and Braun, T. (1996). Alterations in somite patterning of Myf-5-deficient mice: A possible role for FGF-4 and FGF-6. Development 122, 141–150. Heanue, T. A., Reshef, R., Davis, R. J., Mardon, G., Oliver, G., Tomarev, S., Lassar, A. B., and Tabin, C. J. (1999). Synergistic regulation of vertebrate muscle development by Dach2, Eya2, and Six1, homologs of genes required for Drosophila eye formation. Genes Dev. 13, 3231–3243. Hughes, S. M., Taylor, J. M., Tapscott, S. J., Gurley, C. M., Carter, W. J., and Peterson, C. A. (1993). Selective accumulation of MyoD and myogenin mRNAs in fast and slow adult skeletal muscle is controlled by innervation and hormones. Development 118, 1137–1147.
© 2002 Elsevier Science (USA). All rights reserved.
223
Essential and Redundant Functions of the MyoD DRR
Kablar, B., Asakura, A., Krastel, K., Ying, C., May, L. L., Goldhamer, D. J., and Rudnicki, M. A. (1998). MyoD and Myf-5 define the specification of musculature of distinct embryonic origin. Biochem. Cell Biol. 76, 1079 –1091. Kablar, B., Krastel, K., Ying, C., Asakura, A., Tapscott, S. J., and Rudnicki, M. A. (1997). MyoD and Myf-5 differentially regulate the development of limb versus trunk skeletal muscle. Development 124, 4729 – 4738. Kablar, B., Krastel, K., Ying, C., Tapscott, S. J., Goldhamer, D. J., and Rudnicki, M. A. (1999). Myogenic determination occurs independently in somites and limb buds. Dev. Biol. 206, 219 – 231. Kaul, A., Koster, M., Neuhaus, H., and Braun, T. (2000). Myf-5 revisited: Loss of early myotome formation does not lead to a rib phenotype in homozygous Myf-5 mutant mice. Cell 102, 17–19. Kucharczuk, K. L., Love, C. M., Dougherty, N. M., and Goldhamer, D. J. (1999). Fine-scale transgenic mapping of the MyoD core enhancer: MyoD is regulated by distinct but overlapping mechanisms in myotomal and non- myotomal muscle lineages. Development 126, 1957–1965. Maley, M. A., Fan, Y., Beilharz, M. W., and Grounds, M. D. (1994). Intrinsic differences in MyoD and myogenin expression between primary cultures of SJL/J and BALB/C skeletal muscle. Exp. Cell Res. 211, 99 –107. Maroto, M., Reshef, R., Munsterberg, A. E., Koester, S., Goulding, M., and Lassar, A. B. (1997). Ectopic Pax-3 activates MyoD and Myf-5 expression in embryonic mesoderm and neural tissue. Cell 89, 139 –148. Monroe, R. J., Sleckman, B. P., Monroe, B. C., Knor, B., Claypool, S., Ferrini, R., Davidson, L., and Alt, F. W. (1999). Developmental regulation of TCR␦ locus accessibility and expression by the TCR␦ enhancer. Immunity 10, 503–513. Olson, E. N., Arnold, H. H., Rigby, P. W., and Wold, B. J. (1996). Know your neighbors: three phenotypes in null mutants of the myogenic bHLH gene MRF4. Cell 85, 1– 4. Ordahl, C. P., Williams, B. A., and Denetclaw, W. (2000). Determination and morphogenesis in myogenic progenitor cells: An experimental embryological approach. Curr. Top. Dev. Biol. 48, 319 –367. Phelan, S. A., and Loeken, M. R. (1998). Identification of a new binding motif for the paired domain of Pax-3 and unusual characteristics of spacing of bipartite recognition elements on binding and transcription activation. J. Biol. Chem. 273, 19153– 19159. Rudnicki, M. A., Braun, T., Hinuma, S., and 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. Rudnicki, M. A., Schnegelsberg, P. N., Stead, R. H., Braun, T., Arnold, H. H., and Jaenisch, R. (1993). MyoD or Myf-5 is required for the formation of skeletal muscle. Cell 75, 1351–1359.
Rugh, R. (1991). “The Mouse.” Oxford University Press, New York. Sauer, B., and Henderson, N. (1988). Site-specific DNA recombination in mammalian cells by the Cre recombinase of bacteriophage P1. Proc. Natl. Acad. Sci. USA 85, 5166 –5170. Sauer, B., and Henderson, N. (1990). Targeted insertion of exogenous DNA into the eukaryotic genome by the Cre recombinase. New Biol. 2, 441– 449. Seale, P., and Rudnicki, M. A. (2000). A new look at the origin, function, and “stem-cell” status of muscle satellite cells. Dev. Biol. 218 115–124. Smith, C. K. n., Janney, M. J., and Allen, R. E. (1994). Temporal expression of myogenic regulatory genes during activation, proliferation, and differentiation of rat skeletal muscle satellite cells. J. Cell. Physiol. 159, 379 –385. Snow, M. H. (1977). The effects of aging on satellite cells in skeletal muscles of mice and rats. Cell Tissue Res. 185, 399 – 408. Tajbakhsh, S., Rocancourt, D., and Buckingham, M. (1996). Muscle progenitor cells failing to respond to positional cues adopt non-myogenic fates in myf-5 null mice. Nature 384, 266 –270. Tajbakhsh, S., Rocancourt, D., Cossu, G., and Buckingham, M. (1997). Redefining the genetic hierarchies controlling skeletal myogenesis: Pax-3 and Myf-5 act upstream of MyoD. Cell 89, 127–138. Tapscott, S. J., Lassar, A. B., and Weintraub, H. (1992). A novel myoblast enhancer element mediates MyoD transcription. Mol. Cell. Biol. 12, 4994 –5003. Thayer, M. J., Tapscott, S. J., Davis, R. L., Wright, W. E., Lassar, A. B., and Weintraub, H. (1989). Positive autoregulation of the myogenic determination gene MyoD1. Cell 58, 241–248. Tremblay, P., Dietrich, S., Mericskay, M., Schubert, F. R., Li, Z., and Paulin, D. (1998). A crucial role for Pax3 in the development of the hypaxial musculature and the long-range migration of muscle precursors. Dev. Biol. 203, 49 – 61. Williams, B. A., and Ordahl, C. P. (1994). Pax-3 expression in segmental mesoderm marks early stages in myogenic cell specification. Development 120, 785–796. Wilson-Rawls, J., Hurt, C. R., Parsons, S. M., and Rawls, A. (1999). Differential regulation of epaxial and hypaxial muscle development by paraxis. Development 126, 5217–5229. Yablonka-Reuveni, Z., and Rivera, A. J. (1994). Temporal expression of regulatory and structural muscle proteins during myogenesis of satellite cells on isolated adult rat fibers. Dev. Biol. 164, 588 – 603. Zakany, J., Gerard, M., Favier, B., and Duboule, D. (1997). Deletion of a HoxD enhancer induces transcriptional heterochrony leading to transposition of the sacrum. EMBO J. 16, 4393– 4402.
© 2002 Elsevier Science (USA). All rights reserved.
Received for publication October Revised February Accepted February Published online April
29, 25, 25, 14,
2001 2002 2002 2002
