![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Previous Article | Next Article 
Molecular and Cellular Biology, February 2006, p. 1414-1423, Vol. 26, No. 4
0270-7306/06/$08.00+0 doi:10.1128/MCB.26.4.1414-1423.2006
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
Department of Medicine (Division of Cardiology),1 Department of Developmental and Molecular Biology, Albert Einstein College of Medicine, Bronx, New York 104612
Received 3 August 2005/ Returned for modification 18 September 2005/ Accepted 25 November 2005
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
Fusion-competent myoblasts also express specific markers, such as Stick-and-stones (6) and Hibris (1, 12), and are capable of limited differentiation in the absence of fusion to muscle founders, thereby suggesting that distinct regulatory pathways exist to control their development. Of note, recent studies have identified lame duck (lmd) (also known as gleeful [glc] and myoblast incompetent [minc]) as a gene essential for the specification and differentiation of fusion-competent myoblasts (11, 14, 24). In embryos lacking lmd function, muscle founders are not affected, as inferred from normal expression of identity genes such as Kruppel (Kr) and nautilus (nau). However, there is a specific loss of myocyte enhancer factor 2 (mef2) and sticks-and-stones (sns) expression in fusion-competent myoblasts, as well as an absence of multinucleate muscle fibers, thus suggesting that lmd is required for proper specification of fusion-competent myoblasts. The lame duck gene codes for a new and distinct member of the Gli superfamily of transcription factors, and molecular data showed that lmd is a direct upstream regulator of mef2 expression in the fusion-competent myoblasts (11). Moreover, activation of Mef2 in fusion-competent myoblasts is found to be associated with increased nuclear-localized Lmd protein expression. In the aggregate, our previous findings supported the notion that the development of fusion-competent myoblasts occurs via a two-step process. The initial step involves the transcriptional activation of lmd in nonfounder myoblasts, which provides mesodermal cells with the potential to become fusion-competent myoblasts. The subsequent step involves the nuclear translocation of the Lmd protein, which results in the activation of genes critical for fusion and terminal differentiation.
Here, we show that spatial restriction of Lmd expression and appropriate levels of Lmd protein activity in fusion-competent myoblasts must be maintained; otherwise, inappropriate activation of downstream target genes would ensue, leading to muscle defects. We provide evidence that the generation of active Lmd protein is regulated via distinct posttranscriptional mechanisms. First, our detailed structure-function analysis of the Lmd protein has identified two separate regions, one at the most-N-terminal portion and the other within the carboxyl-terminal portion, which are critical for modulating the balance between translocation of the Lmd protein into the nucleus and its retention within the cytoplasm. Second, mutations affecting a putative binding site for the Src homology 3 (SH3) domain protein or potential protein kinase A (PKA) phosphorylation sites altered Lmd activity without any apparent changes in the subcellular localization of the protein, thereby underscoring the importance of additional mechanisms which involve posttranslational modifications.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
Generation of UAS-lmd derivative constructs and transgenic flies. Deletion derivatives of lmd were first obtained by direct PCR amplification of the regions of interest with appropriate primers and a high-fidelity PCR system (Roche Molecular Biochemicals). Full-length lmd with specific mutations was first generated by using an ExSite PCR-based site-directed mutagenesis kit (Stratagene, CA). These mutated full-length constructs included the following: (i) the PXXPmt construct, in which the region between amino acid residues 163 to 179, which encompassed several copies of the PYTP motif, was deleted; (ii) the NLSmt construct, in which the region spanning a putative nuclear localization sequence (aa residues 477 to 483) was deleted; and (iii) the Pka1mt and Pka2mt constructs, in which residues Arg596 and Arg597 and residues Arg617 and Arg618, respectively, were changed to Ala. During generation of the constructs, one copy of the hemagglutinin (HA) protein epitope tag was added in frame to the amino-terminal end of the full-length protein (aa 1 to 866) and all mutant Lmd protein derivatives. All tagged constructs were verified by sequencing prior to being recloned into the pUAST plasmid (8). Transgenic flies harboring UAS full-length or mutant lmd constructs were generated by using standard P-element-mediated transformation. For each construct, multiple independent insertions were generated and tested.
Misexpression and rescue experiments. The UAS/GAL4 system (8) was used to achieve misexpression of full-length or specific mutant Lmd protein derivatives in somatic mesodermal cells (twi-Gal4 driver), muscle founders (rP298-Gal4 driver), or the ectoderm (en-Gal4 driver). Staged embryo collections were done at 27°C. For phenotypic analysis, embryos were stained with an antitropomyosin antibody to visualize the mature muscle pattern. For a subset of constructs, embryos were also stained with an anti-Mef2 antibody to monitor specific gene activation. In protein localization studies, embryos were double-stained with antibodies against HA and lamin to identify the HA-tagged Lmd protein derivatives and nuclear membrane, respectively.
To examine the rescue potential of particular constructs, twi-Gal4 and UAS-lmd lines (corresponding to full-length or specific mutant derivatives) were crossed into the lmd1 mutant background. Stocks which were homozygous for the Gal4 driver or UAS-lmd construct (on the second chromosome) and heterozygous for the lmd mutation, which was balanced over TM3Sb ftz-LacZ, were then generated. For assessing the degree of rescue conferred by the Lmd protein derivative of interest, embryos were double-stained with antibodies against tropomyosin and ß-galactosidase (ß-gal), with the latter for identifying mutant embryos that showed an absence of LacZ staining. To examine the subcellular localization of the exogenously provided Lmd protein derivatives in the rescued embryos, embryos were triple-stained with antibodies against the carboxy terminus of Lmd (this antibody does not recognize the truncated Lmd protein in lmd1 mutant embryos), ß-galactosidase, and lamin.
Antibody immunocytochemistry of whole-mount embryos. Immunocytochemistry of whole-mount embryos was performed, essentially as described previously (20). In brief, embryos were dechorionated and fixed in 4% formaldehyde-phosphate-buffered saline-heptane. After being blocked in phosphate-buffered saline-10% bovine serum albumin, the embryos were incubated with primary antibodies of interest, washed, and then incubated with biotinylated or fluorescent secondary antibodies. For detecting the biotinylated antibody, a Vectastain ABC Elite peroxidase kit (Vector Labs) was used, followed by color visualization with diaminobenzidine as the substrate. Fluorescent-labeled samples were directly examined, or a TSA fluorescence system (NEN Life Science/Perkin Elmer) was used for signal amplification when necessary. The following primary antibodies were used: mouse monoclonal anti-HA (1:400; Roche Molecular Biochemicals), rat polyclonal antitropomyosin (1:400; Babraham Tetronix), mouse monoclonal anti-ß-gal (1:10; Developmental Studies Hybridoma Bank), rabbit polyclonal anti-ß-gal (1:1,000; ICN), guinea pig polyclonal anti-amino-terminal portion of Lmd (1:250) (11), guinea pig polyclonal anti-carboxy-terminal portion of Lmd (1:250) (this study), rabbit polyclonal anti-Mef2 (1:750) (7), rabbit polyclonal antilamin (1:1,000; gift from P. Fisher), rabbit polyclonal anti-Eve (1:1,000; gift from M. Frasch), and guinea pig polyclonal anti-Eve (1:400; gift from J. Reinitz). Biotinylated (Vector Labs) and fluorescent (Jackson ImmunoResearch) secondary antibodies were used, as recommended by the suppliers. Embryos were photographed with Nomarski differential interference contrast optics on an Olympus AX70 microscope with a 20x UPlan objective or analyzed with a Leica TCS4D confocal microscope with a 100x objective.
Yeast-based assay for mapping activation domain. Truncated lmd cDNA fragments were amplified by PCR and cloned into the pGBKT7 vector (Clontech), which contains the sequences encoding the GAL4 DNA binding domain and a tryptophan selection marker. These plasmids were transformed into PJ69-4A Saccharomyces cerevisiae cells (provided by G. Prelich) that harbor an integrated copy of Gal4 binding sites cloned upstream of the histidine reporter gene. Transformants were plated on selective medium lacking histidine. The presence of an activation domain within any of the testing fragments was indicated by colony growth on the minimal medium.
Western blot analysis. Staged embryos were homogenized and lysed in RIPA buffer (150 mM NaCl-50 mM Tris-HCl [pH 8.0]-1% Nonidet P-40-0.5% deoxycholate-0.1% sodium dodecyl sulfate-1 mM dithiothreitol) plus 1 mM phenylmethylsulfonyl fluoride and complete protease inhibitor cocktail (Roche). Fifty micrograms of total protein per lane was loaded on standard 8% polyacrylamide gels and transferred to Immobilon-P membranes, using the manufacturer's recommendations (Millipore). After being blocked with 5% nonfat milk in TBST (Tris-buffered saline-0.1% Tween 20), the membranes were incubated with a rabbit polyclonal antibody against the amino-terminal portion of Lmd (1:1,000 dilution) (11) or a mouse monoclonal anti-Flag antibody (1:5,000 dilution; Sigma) overnight at 4°C. Following washes in TBST, the membranes were incubated with peroxidase-conjugated anti-rabbit or anti-mouse immunoglobulin G (Bio-Rad) for 1 h at room temperature. The protein complexes were detected by using an enhanced chemiluminescence system (Pierce).
|
RESULTS AND DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
aa residue 691 are critical for the cytoplasmic retention of the Lmd protein.
|
|
|
Taken together, our data establish the existence of two distinct regions, one located at the amino-terminal end of Lmd (region I, aa residues 1 to 141) and the other immediately carboxyl to the Zn finger/DNA binding domain (region II, aa residues 691 to 866), which are required for the cytoplasmic localization of the Lmd protein (summarized in Table 1). The data further indicate that these regions are necessary but not sufficient, within particular contexts, for mediating the cytoplasmic retention of Lmd. Notably, the inclusion of region I in the aa1-650 protein is not sufficient for retaining the protein in the cytoplasm. Similarly, the inclusion of region II in the aa141-866 protein is also not sufficient. Furthermore, the full-length Lmd protein is both nuclear and cytoplasmic, indicating the action of additional regulatory inputs. At present, our data cannot distinguish between the possibility that region I or region II (or both) interacts with cytoplasmic factors which could provide a link between Lmd and the microtubule network/cytoskeleton and the possibility that interaction of specific factors with either (or both) of these regions leads to the masking of the NLS that is present within the DNA binding/Zn finger domain. These two possibilities are of course not mutually exclusive. Interestingly, it has been proposed that subcellular localization of the related Drosophila Cubitus interruptus (Ci) protein is regulated via microtubule-dependent and -independent mechanisms (27).
Abnormally high levels of nuclear Lmd protein in mesodermal cells result in severe defects in muscle formation. To assess the functional consequences arising from altered subcellular localization of the Lmd protein and the importance of the elements that were uncovered in the above analysis, we examined the effects of overexpressing full-length Lmd or particular Lmd protein derivatives specifically in somatic mesodermal cells, by using the twist-Gal4 driver, on muscle formation (summarized in Table 1). For analysis, embryos were collected and then stained with an antibody against tropomyosin to visualize the mature muscle pattern. We first examined embryos that overexpress full-length Lmd protein with a mutation of the predicted NLS within the DNA binding domain (NLSmt). These embryos exhibit a normal muscle pattern, thereby providing functional evidence that the NLS is indeed necessary for nuclear localization of the Lmd protein (Fig. 2A and B). Interestingly, compared to wild-type embryos, embryos overexpressing the full-length protein (aa 1 to 866) show some muscle malformations that include a loss of a limited number of muscles and the presence of a small number of rounded cells that correspond to unfused cells (Fig. 2C; compare with panel A). These observations suggest that levels of endogenous Lmd protein might already be in excess and that, hence, its overexpression is mostly inconsequential. Alternatively, the subcellular distribution of the exogenous full-length Lmd protein might also be tightly regulated to ensure that appropriate levels of active Lmd are maintained in the nucleus. This second possibility is supported by the observation that embryos expressing the aa1-780 protein, which includes both cytoplasmic retention elements (region I and most if not all of region II), also show relatively minor muscle defects (data not shown). By contrast, embryos that express the aa1-650, aa141-866, or aa183-866 protein, all of which were shown to be exclusively nuclear, exhibit a severe mutant phenotype that is characterized by a significant loss of muscles and the presence of a large number of rounded, unfused cells (Fig. 2D and E; also data not shown). Combined with the observation that the nuclear-localized aa382-650 and aa382-691 proteins do not cause any muscle defects and hence are presumed to be inactive, our data further indicate that aa residues 183 to 382 encompass a functionally important domain, likely to be an activation domain, that effects the activity of Lmd during transcriptional regulation of its target genes. Indeed, when constructs that consist of different portions of Lmd fused to the GAL4 binding domain were tested in a yeast-based assay system, two regions within the amino-terminal portion of the Lmd protein, mapping within aa 1 to 139 and aa 183 to 349, that confer transcriptional activation capability were identified (Fig. 3).
|
|
In the aggregate, our in vivo analysis demonstrates that deregulation of the normal nuclear-cytoplasmic ratio of Lmd protein has deleterious effects on myogenesis. Our data strongly suggest that forced expression of the abovementioned nuclear Lmd protein derivatives in fusion-competent myoblasts and muscle founders yields abnormally high levels of Lmd activity that lead to increased activation of downstream target genes. In turn, inappropriate levels of some of these gene products and their inappropriate presence in muscle founders are likely to compromise the normal process of myoblast fusion and differentiation. Our data further suggest that the mechanisms for regulating Lmd activity in fusion-competent myoblasts are active throughout the somatic mesoderm, since the activity of exogenous full-length Lmd is also regulated when driven into muscle founders with the rP298-Gal4 driver.
Increased nuclear Lmd activity results in inappropriate gene activation. Previous studies had identified mef2 and sns, two genes that are critical for differentiation and fusion, as downstream target genes of lmd (11, 14, 24). It has also been established that appropriate levels of mef2 expression were critical for proper muscle formation and that overexpression of mef2 resulted in muscle defects (15, 18). Hence, our observations that overexpression of the nuclear-localized Lmd protein derivatives result in muscle defects suggest the possibility that this is due in part to increased levels of mef2 expression, arising from increased levels of Lmd activity. To test this possibility more directly, we took advantage of the observation that ectopic expression of full-length Lmd in the central nervous system activates both mef2 and sns expression while ectopic expression of full-length Lmd in the ectoderm activates only sns (14, 24; H. Duan and H. T. Nguyen, unpublished data). For this purpose, we used the en-Gal4 driver to direct the full-length Lmd protein (aa 1 to 866) or the nuclear Lmd protein derivatives (aa 1 to 650, aa 141 to 866, and aa 183 to 866) in the ectoderm and assayed for sns and mef2 expression by RNA in situ hybridization and antibody staining, respectively. As anticipated, the full-length Lmd protein and the nuclear protein versions activate sns RNA expression, with higher levels being observed with the nuclear Lmd protein derivatives (data not shown). Of particular note, prominent Mef2 protein expression is observed with all three nuclear Lmd protein derivatives, while mef2 activation is not observed with the full-length Lmd protein, as documented in earlier reports (Fig. 4A to C). Consequently, these results support the proposal that the muscle defects that are associated with overexpression of nuclear Lmd protein derivatives are partly due to increased levels of mef2 expression and possibly additional genes that are important for muscle differentiation and fusion. These data argue against dominant effects of the nuclear protein versions.
|
To assess the functional importance of the PXXP motif, we generated a UAS-lmd construct in which the region spanning the PXXP motifs (aa residues 163 to 179) was deleted and determined the consequences of overexpressing it with the twist-Gal4 and rP298-Gal4 drivers. Embryos were collected and stained with an antibody against tropomyosin to examine the mature somatic musculature. Compared to wild-type embryos, embryos in which the PXXP-deleted mutant protein (PXXPmt) was overexpressed in somatic mesodermal cells by the twist-Gal4 driver exhibit a significant loss of muscles and a detectable number of unfused cells (Fig. 5A; compare with Fig. 2A). Overexpression of the PXXPmt protein in muscle founders with the rP298-Gal4 driver also results in muscle defects, but these are slightly less severe than those observed with the twi-Gal4 driver (Fig. 5C). To explore the possibility that the above mutant phenotypes were due to increased levels of mef2 expression, as documented previously with the nuclear-localized Lmd protein derivatives, we used the en-Gal4 driver to direct the PXXPmt construct into ectodermal cells. Indeed, the PXXPmt protein is capable of activating mef2 expression in ectodermal cells, albeit less prominently than the nuclear-localized mutant proteins (Fig. 5E; compare with Fig. 4B and C).
|
We also examined the importance of the putative PKA phosphorylation sites. As described above, UAS-lmd constructs in which either or both PKA phosphorylation motifs were mutated were generated. When the Pka1mt or Pka2mt protein was overexpressed with the twist-Gal4 driver, a significant number of muscles were missing and unfused cells were detectable (Fig. 5B; also data not shown). Not surprisingly, overexpression of the protein with both mutations results in more-severe muscle defects (data not shown). As observed with the PXXPmt protein, ectopic expression of Pka1mt (or Pka2mt) in muscle founders with the rP298-Gal4 driver yields less severe muscle defects (Fig. 5D; also data not shown). Collectively, these observations suggest that these mutant proteins are hyperactive, and therefore we tested for their ability to active mef2 expression in ectodermal cells by using the en-Gal4 driver. As predicted, the Pka1mt protein activates mef2 to appreciable levels (compare Fig. 5F with Fig. 4B and C). Notably, the hyperactivity of the Pka1mt protein is reminiscent of the documented observation that the Ci protein with mutated PKA sites was also much more active than the wild-type Ci protein (9, 22). For further insight, we assessed the effects of reduced or increased PKA activity on muscle formation by using the twi-Gal4 driver to direct the UAS-R* or UAS-mC* transgene, which encodes a mutant PKA regulatory subunit or a constitutively active PKA catalytic subunit, respectively, into mesodermal cells (17). This analysis showed that perturbations in these PKA activities do not result in any obvious muscle malformations, thereby suggesting that a novel PKA or another type of kinase is involved (data not shown).
In order to determine whether the hyperactivity of the PKA mutant proteins could reflect a change in subcellular localization, we also monitored the exogenous Pka1mt protein in the lmd mutant background, as described above. Interestingly, the Pka1mt protein is also detectable in both the nucleus and the cytoplasm at a ratio that appears to be comparable to that observed with the full-length protein (compare Fig. 6A and C). We also tested whether the Lmd protein undergoes cleavage and whether the putative phosphorylation sites influence this process, as has been documented for the Ci protein (reviewed in references 2 and 16). For this analysis, we examined extracts derived from wild-type embryo and transgenic embryos that overexpress the full-length Lmd or Pka1mt protein. As shown in Fig. 7, Western blot analysis with the Lmd antibody against the amino-terminal portion identifies one prominent specific band, which corresponds in size to the full-length protein, with extracts from 6- to 9-h embryos and similarly staged embryos that overexpress the full-length (twi>1-866) or Pka1mt (twi>Pka1mt) protein. Additional detectable bands are nonspecific since they are observed with all extracts, including those derived from control 0- to 3-h embryos that have been shown not to express lmd RNA or protein, as assayed by in situ RNA hybridization and immunocytochemistry, respectively. A similar analysis with a Flag antibody, which recognizes the epitope tag at the carboxy-terminal end of the Lmd protein, also reveals one prominent specific band of a size similar to the full-length protein with extracts from transgenic embryos that overexpress the full-length (twi>1-866) or Pka1mt (twi>Pka1mt) protein. Nonspecific bands are observed with extracts from both control wild-type embryos and transgenic embryos. These data argue against proteolytic cleavage as a mechanism for regulating Lmd protein activity.
|
Differential rescue potential of the various types of hyperactive Lmd protein derivatives. Based upon the data obtained from the above-described overexpression and misexpression studies, it appears that the Lmd protein derivatives that exhibit a change in their subcellular distribution from nucleus and cytoplasm to exclusively nuclear (i.e., aa 1 to 650 and aa 141 to 866) are more potent activators than the proteins that are activated by blocking modifications (i.e., PXXPmt and Pka1mt). To gain further insight, we assessed the capability of these various hyperactive Lmd protein derivatives to rescue the lmd mutant phenotype (summarized in Table 2). For comparison, we also tested the rescue potential of the full-length Lmd protein (aa 1 to 866) and the full-length protein with a mutation of the NLS (NLSmt). The analysis entailed double-staining the transgenic embryos, which also carry the twi-Gal4 driver, with antibodies against tropomyosin and ß-galactosidase, with the latter for identifying mutant embryos. Based upon the observed high degree to which the mature muscle pattern is restored, the exogenously provided full-length protein is able to rescue the lmd mutant phenotype to nearly wild type (Fig. 8A). The resulting phenotype is similar to the one obtained with forced expression of full-length Lmd in wild-type embryos (Fig. 2C). As predicted, the NLSmt protein could not impart any rescue potential, which is consistent with the observation that the NLSmt protein is predominantly cytoplasmic (Fig. 8B; also data not shown). Interestingly, both the aa1-650 and aa141-866 protein versions have poor rescue potential, as shown by the exhibited deranged musculatures (Fig. 8C and D). By contrast, the PXXPmt and Pka1mt proteins are capable of providing partial rescue, albeit to differing degrees (Fig. 8E and F).
|
|
Concluding remarks.
In this study, we have provided evidence that appropriate levels of active Lmd protein are regulated by at least two distinct mechanisms. First, Lmd activity is modulated by changes in nuclear translocation of the protein versus its retention in the cytoplasm. Indeed, protein localization studies showed that two separate regions (I and II) are critical for the cytoplasmic retention of the Lmd protein. The amino-terminal region I (aa 1 to 141) is necessary but not sufficient, in any context, for effecting cytoplasmic retention, whereas region II (aa 691 to 866) is necessary and sufficient when present in truncated Lmd proteins. It is noteworthy that region II is not sufficient within the context of the full-length protein (e.g., aa141-866 or aa183-866 protein) and that the inclusion of both regions in the wild-type protein does not make it exclusively cytoplasmic, thus suggesting the existence of a yet-to-be-defined regulatory component(s) that counteracts the action of the two regulatory regions. Based upon these data, we propose that some unknown factor binds in a signal-dependent manner to the region that spans aa 183 to 382 and abrogates the capability of region I and/or region II to mediate interactions (direct or indirect) with the microtubule network/cytoskeleton. Alternatively, interactions (direct or indirect) between regions I and II could be important for masking the NLS and the binding of this unknown factor may disrupt this masking (Fig. 9). In general terms, our findings regarding the regulation of the subcellular distribution of Lmd are reminiscent of those associated with the Ci protein. For example, a recent study showed that the kinesin-like protein Costal2 (Cos2) interacts with two separate domains of Ci to inhibit its nuclear translocation (27). It was also reported that Cos2 relies on microtubule-dependent and microtubule-independent mechanisms to retain the Ci protein in the cytoplasm. Furthermore, it has been suggested that the formation of distinct protein complexes, consisting of Cos2, Fu, Su(fu), and Ci, is a mechanism by which Ci can be sequestered in the cytoplasm (26, 27; also reviewed in reference 16). Consequently, it will be important to characterize the components that recognize the different regulatory regions within the Lmd protein. By analogy to the Ci protein, it will also be interesting to determine whether there could exist in vivo distinct Lmd protein-containing complexes.
|
|
ACKNOWLEDGMENTS |
|---|
This work was supported by Public Health Service grant RO1-AR4628 from the National Institutes of Health (to H.T.N.).
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
2. Aza-Blanc, P., and T. B. Kornberg. 1999. Ci: a complex transducer of the Hedgehog signal. Trends Genet. 15:458-462.[CrossRef][Medline]
3. Bate, M. 1990. The embryonic development of larval muscles in Drosophila. Development 110:791-804.
4. Baylies, M. K., and M. Bate. 1996. twist: a myogenic switch in Drosophila. Science 272:1481-1484.[Abstract]
5. Baylies, M. K., and A. M. Michelson. 2001. Invertebrate myogenesis: looking back to the future of muscle development. Curr. Opin. Genet. Dev. 11:431-439.[CrossRef][Medline]
6. Bour, B. A., M. Chakravarti, J. M. West, and S. M. Abmayr. 2000. Drosophila SNS, a member of the immunoglobulin superfamily that is essential for myoblast fusion. Genes Dev. 14:1498-1511.
7. Bour, B. A., M. A. O'Brien, W. L. Lockwood, E. S. Goldstein, R. Bodmer, P. H. Taghert, S. M. Abmayr, and H. T. Nguyen. 1995. Drosophila MEF2, a transcription factor that is essential for myogenesis. Genes Dev. 9:730-741.
8. Brand, A. H., and N. Perrimon. 1993. Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118:401-415.[Abstract]
9. Chen, Y., J.-R. Cardinaux, R. H. Goodman, and S. M. Smolik. 1999. Mutants of cubitus interruptus that are independent of PKA regulation are independent of hedgehog signaling. Development 126:3607-3616.[Abstract]
10. Dohrmann, C., N. Azpiazu, and M. Frasch. 1990. A new Drosophila homeo box gene is expressed in mesodermal precursor cells of distinct muscles during embryogenesis. Genes Dev. 4:2098-2111.
11. Duan, H., J. B. Skeath, and H. T. Nguyen. 2001. Drosophila Lame duck, a novel member of the Gli superfamily, acts as a key regulator of myogenesis by controlling fusion-competent myoblast development. Development 128:4489-4500.
12. Dworak, H. A., M. A. Charles, L. B. Pellerano, and H. Sink. 2001. Characterization of Drosophila hibris, a gene related to human nephrin. Development 128:4265-4276.
13. Frasch, M. 1999. Controls in patterning and diversification of somatic muscles during Drosophila embryogenesis. Curr. Opin. Genet. Dev. 9:522-529.[CrossRef][Medline]
14. Furlong, E. E., E. C. Andersen, B. Null, K. P. White, and M. P. Scott. 2001. Patterns of gene expression during Drosophila mesoderm development. Science 293:1629-1633.
15. Gunthorpe, D., K. E. Beatty, and M. V. Taylor. 1999. Different levels, but not different isoforms, of the Drosophila transcription factor DMEF2 affect distinct aspects of muscle differentiation. Dev. Biol. 215:130-145.[CrossRef][Medline]
16. Ingham, P. W., and A. P. McMahon. 2001. Hedgehog signaling in animal development: paradigms and principles. Genes Dev. 15:3059-3087.
17. Li, W., J. T. Ohlmeyer, M. E. Lane, and D. Kalderon. 1995. Function of protein kinase A in hedgehog signal transduction and Drosophila imaginal disc development. Cell 80:553-562.[CrossRef][Medline]
18. Lin, M. H., B. A. Bour, S. M. Abmayr, and R. V. Storti. 1997. Ectopic expression of MEF2 in the epidermis induces epidermal expression of muscle genes and abnormal muscle development in Drosophila. Dev. Biol. 182:240-255.[CrossRef][Medline]
19. Menon, S. D., and W. Chia. 2001. Drosophila Rolling pebbles: a multidomain protein required for myoblast fusion that recruits D-Titin in response to the myoblast attractant Dumbfounded. Dev. Cell 1:691-703.[CrossRef][Medline]
20. Nguyen, H. T., and X. Xu. 1998. Drosophila mef2 expression during mesoderm development is controlled by a complex array of cis-acting regulatory modules. Dev. Biol. 204:550-566.[CrossRef][Medline]
21. Pawson, T. 1995. Protein modules and signalling networks. Nature 373:573-580.[CrossRef][Medline]
22. Price, M. A., and D. Kalderon. 1999. Proteolysis of cubitus interruptus in Drosophila requires phosphorylation by protein kinase A. Development 126:4331-4339.[Abstract]
23. Ruiz-Gomez, M., N. Coutts, A. Price, M. V. Taylor, and M. Bate. 2000. Drosophila dumbfounded: a myoblast attractant essential for fusion. Cell 102:189-198.[CrossRef][Medline]
24. Ruiz-Gomez, M., N. Coutts, M. L. Suster, M. Landgraf, and M. Bate. 2002. myoblasts incompetent encodes a zinc finger transcription factor required to specify fusion-competent myoblasts in Drosophila. Development 129:133-141.
25. Rushton, E., R. Drysdale, S. M. Abmayr, A. M. Michelson, and M. Bate. 1995. Mutations in a novel gene, myoblast city, provide evidence in support of the founder cell hypothesis for Drosophila muscle development. Development 121:1979-1988.[Abstract]
26. Stegman, M. A., J. E. Vallance, G. Elangovan, J. Sosinski, Y. Cheng, and D. J. Robbins. 2000. Identification of a tetrameric hedgehog signaling complex. J. Biol. Chem. 275:21809-21812.
27. Wang, G., and J. Jiang. 2004. Multiple Cos2/Ci interactions regulate Ci subcellular localization through microtubule dependent and independent mechanisms. Dev. Biol. 268:493-505.[CrossRef][Medline]
28. Wientjes, F. B., and A. W. Segal. 2003. PX domain takes shape. Curr. Opin. Hematol. 10:2-7.[CrossRef][Medline]
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| J. Bacteriol. | J. Virol. | Eukaryot. Cell |
|---|
| Microbiol. Mol. Biol. Rev. | Clin. Vaccine Immunol. | All ASM Journals |
|---|
