ectopic expression of mef2 in the epidermis induces ... · pdf fileto understand further the...

DEVELOPMENTAL BIOLOGY 182, 240–255 (1997)ARTICLE NO. DB968484

Ectopic Expression of MEF2 in the EpidermisInduces Epidermal Expression of Muscle Genesand Abnormal Muscle Development in Drosophila

Meei-Hua Lin,* Barbara A. Bour,† Susan M. Abmayr,†and Robert V. Storti*,1

*Department of Biochemistry M/C536, University of Illinois College of Medicine, 1853 W. PolkStreet, Chicago, Illinois 60612; and †Department of Biochemistry and Molecular Biology,Pennsylvania State University, University Park, Pennsylvania 16802

Myocyte-specific enhancer-binding factor 2 (MEF2) is a myogenic regulatory factor in vertebrates and Drosophila. Whereasthe role of MEF2 in regulating vertebrate myogenesis and muscle genes has been extensively studied, little is known ofthe role of MEF2 in regulating Drosophila myogenesis. We have shown in a recent analysis of the regulation of theDrosophila Tropomyosin I (TmI) gene in transgenic flies that MEF2 is a positive regulator of TmI expression in the somaticbody-wall muscles of embryos, larvae, and adults. To understand further the role of MEF2 in myogenesis and test the roleof MEF2 in regulating TmI expression, we have used the yeast GAL4/UAS system to generate embryos in which MEF2 isectopically expressed in tissues where it is not normally expressed or embryos in which MEF2 is overexpressed in themesoderm and muscles. We observe that ectopic expression of MEF2 in the epidermis and the ventral midline cells inembryos activates the expression of TmI and other muscle genes in these tissues and that this activation is stage-dependentsuggesting a requirement for additional factors. Furthermore, ectopic expression of MEF2 in the epidermis results in adecrease in the expression of signaling molecules in the epidermis and a failure of the embryo to properly form body-wallmuscles. These results indicate that MEF2 can function out of context in the epidermis to induce the expression of musclegenes and interfere with a requirement for the epidermis in muscle development. We also find that the level of MEF2 inthe mesoderm and/or muscles in embryos is critical to body-wall muscle formation; however, no effect is observed on thedevelopment of the visceral muscle or dorsal vessel. q 1997 Academic Press

INTRODUCTION closest known Drosophila homologue of the vertebratemyogenic factor family (Michelson et al., 1990; Paterson etal., 1991). However, mutation analyses of the nau gene haveMuch of our knowledge about skeletal muscle develop-revealed that the function of nau is probably limited to onlyment has come from the cloning of the vertebrate myogenica subset of body-wall muscle precursors (Abmayr et al.,family of basic helix–loop–helix (bHLH) proteins. This1995).family contains four muscle-specific transcription factors,

A second factor important in regulating vertebrate myo-MyoD, Myf5, MRF4, and Myogenin. Cell culture studiesgenesis is myocyte-specific enhancer-binding factor 2have shown that each of these myogenic bHLH proteins(MEF2). MEF2 was initially identified as a DNA-bindingis able to activate skeletal muscle-gene transcription andactivity that was induced when skeletal myoblasts differen-skeletal muscle differentiation when expressed in a varietytiated into myotubes (Gossett et al., 1989). It recognizes anof non-muscle-cell types (reviewed in Olson and Klein,AT-rich sequence shown to be important for the activation1994). The roles of these factors in muscle developmentof many muscle genes in differentiating myotubes. MEF2have been confirmed by gene-knockout experiments inbelongs to the MADS (MCM1, Agamous, Deficiens, andmice (Braun et al., 1992; Hasty et al., 1993; Nabeshima etserum response factor) box-containing family of transcrip-al., 1993; Rudnicki et al., 1993). nautilus (nau) encodes thetion factors (reviewed in Shore and Sharrocks, 1995). Fourmef2 genes have been identified in vertebrates (Pollock andTreisman, 1991; Yu et al., 1992; Breitbart et al., 1993; Mar-1 To whom correspondence and reprint requests should be ad-

dressed. Fax: (312) 413-0364. tin et al., 1993, 1994; McDermott et al., 1993; Chambers

240

0012-1606/97 $25.00Copyright q 1997 by Academic Press

All rights of reproduction in any form reserved.

AID DB 8484 / 6x19$$$201 01-18-97 03:24:43 dbal

241Ectopic Expression of MEF2 in Drosophila Epidermis

et al., 1994; reviewed in Olson et al., 1995). Their products sufficient by itself to direct TmI enhancer-directed expres-sion in muscles. Thus the function of MEF2 in regulatingshare more than 85% amino-acid homology within the

MADS box and the adjacent conserved region referred to as Drosophila muscle genes and differentiation is likely to becomplex and involve multiple interacting factors. One waythe MEF2 domain. MEF2 recognizes a consensus (C/T)T(A/

T)2AAATA(A/G) sequence through the MADS box and the to gain insight into the possible function of MEF2 is togenerate gain-of-function mutations by expressing MEF2 inMEF2 domain (Gossett et al., 1989; Pollock and Treisman,

1991; Yu et al., 1992; Breitbart et al., 1993; Martin et al., tissues where it is not normally expressed. Another is tomodulate the level of expression of MEF2 in the tissue in1993; McDermott et al., 1993). Several studies have shown

that MEF2 plays an important role in the transcriptional which it is normally expressed and thus alter its interactionwith other regulatory molecules. We therefore testedregulation of the myogenic bHLH genes (Edmondson et al.,

1992; Yee and Rigby, 1993; Buchbager et al., 1994; Leibham whether ectopic expression of MEF2 in other tissues couldactivate ectopic expression of TmI and other muscle geneset al., 1994; Black et al., 1995; Naidu et al., 1995) and other

muscle genes (Bassel-Duby et al., 1992; Morisaki and and whether elevated levels of MEF2 in the mesoderm andmuscles have any effect on normal muscle differentiation.Holmes, 1993; Chambers et al., 1994; Kaushal et al., 1994;

Li and Capetanaki, 1994; Liu et al., 1994; Parmacek et al., In this report, we used the yeast GAL4/UAS system to ex-press MEF2 in different tissues (Brand and Perrimon, 1993).1994; Andres et al., 1995; Molkentin et al., 1995). mef2

genes are expressed in skeletal, cardiac, and smooth muscles We find that ectopically expressing MEF2 in the epidermisand the ventral midline cells of embryos results in the tran-during mouse development, suggesting that they may play

important roles in activating transcription within each scriptional activation of TmI and other muscle genes. Theseresults are discussed in terms of the possible requirementmyogenic lineage (Edmondson et al., 1994).

The analysis of MEF2 functions has been facilitated by for additional factors present in the epidermis to activatemuscle genes. In addition, ectopic expression of MEF2 inthe isolation and characterization of the mef2 gene in Dro-

sophila (Lilly et al., 1994; Nguyen et al., 1994; Taylor et the epidermis results in a failure of the embryo to formbody-wall muscles and a decrease in the expression of sig-al., 1995). Drosophila contains a single mef2 gene that en-

codes a protein having greater than 85% amino-acid homol- naling molecules in the epidermis, suggesting a require-ment for the epidermis in muscle development. We alsoogy within the MADS box and the MEF2 domain to the

corresponding regions of vertebrate MEF2 proteins. It recog- find that the level of MEF2 in the mesoderm and/or musclesis critical to body-wall muscle formation. An increase innizes the same target sequence as its vertebrate homologues

and can transactivate reporter constructs in transient trans- the level of MEF2 in the mesoderm leads to embryoniclethality and defective body-wall muscle formation.fection assays in a MEF2-binding site-dependent manner.

Drosophila mef2 is expressed in the mesoderm of early em-bryos shortly after gastrulation, and continues to be ex-pressed in somatic, visceral, and cardiac muscle lineages MATERIALS AND METHODSthroughout embryogenesis. This pattern of expression sug-gests that mef2 may be important in regulating the earlier Drosophila Stocks and Crosses to Ectopically

Express MEF2stages of myogenesis that establish mesoderm and musclelineages as well as the later stages of myogenesis that regu-

Fly stocks were maintained at 22 or 257C on standard medialate myoblast differentiation. Mutations of the mef2 gene, (Carolina Biological Supply Company). The GAL4-expressing lineshowever, suggest that the role(s) of mef2 in regulating Dro- 24B, 69B, hsGAL42-1 (Brand and Perrimon, 1993), and GAL4-1407sophila myogenesis may be limited to later aspects of myo- (Luo et al., 1994), and the UAS-mef2 line (Bour et al., 1995) havegenesis because mef2 null mutant embryos show normal been described. To direct expression of MEF2 in a tissue-specific

way, the 24B, 69B, or GAL4-1407 enhancer lines expressing GAL4specification and patterning of muscle precursors (Bour etwere crossed to the UAS-mef2 line and embryos from these crossesal., 1995; Lilly et al., 1995; Ranganayakulu et al., 1995).were collected for 16 hr at 257C or 13 hr at 297C on grape-agarThese muscle precursors, however, fail to undergo fusionplates. To heat-induce expression of MEF2 in embryos, theand differentiation to form muscle fibers.hsGAL42-1 line was crossed to the UAS-mef2 line and heat-shockWe have recently demonstrated a role for MEF2 in thetreatments were carried out as follows. Embryos from this crosstranscriptional regulation of the Drosophila Tropomyosinwere collected for 2 hr at 257C, aged for 2 hr at 257C, and then

(TmI) gene during muscle fiber differentiation. The TmI placed for 30 min in a 377C water bath, followed by a 30-mingene contains two partially redundant muscle enhancers incubation at room temperature. The heat-shock treatment waswithin the first intron. Both enhancers can drive high-level repeated once more and embryos were aged at 257C for 0, 2, 4, 6,expression of reporter genes in muscles of transgenic em- 8, 10, or 12 more hours before fixation for immunohistochemical

staining and in situ hybridization.bryos, larvae, and adults (Schultz et al., 1991; Gremke etal., 1993; Lin et al., 1996). Furthermore, both enhancerscontain a MEF2-binding site and mutation of either MEF2-

Immunohistochemical Staining of Whole-Mountbinding site leads to a dramatic decrease in the enhancer-Embryosdirected reporter gene expression in muscles of transgenic

flies (Lin et al., 1996). Additional experiments indicate, Ectopically expressed MEF2 protein was determined by immuno-histochemical staining as described (Hoshizaki, 1994). Polyclonalhowever, that the MEF2-binding site while necessary is not

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID DB 8484 / 6x19$$$202 01-18-97 03:24:43 dbal

242 Lin et al.

MEF2 antibody (Bour et al., 1995) was used at a 1:1000 dilution. null mutant embryos do not make body-wall muscles (BourGoat anti-rabbit IgG conjugated with alkaline phosphatase (Sigma) et al., 1995; Lilly et al., 1995; Ranganayakulu et al., 1995)was used at a 1:500 dilution. In each GAL4 enhancer 1 UAS-mef2 and do not express TmI (Lin et al., 1995). Accordingly, wecross high levels of MEF2 were detected at the stages and tissues considered the possibility that gain-of-function mutationsdictated by the GAL4 enhancer construct (Brand and Perrimon, might be generated by ectopic expression of MEF2 through1993; Luo et al., 1994; see below).

activation of part or all of the myogenic program and TmIexpression in tissues where it is not normally expressed.We tested this possibility by ectopic expression of MEF2Whole-Mount in Situ Hybridization of Embryosin the epidermis and nervous system using the yeast

Whole-mount in situ hybridization using antisense RNA as aGAL4/UAS system. In this system transgenic flies con-probe was done as described (Tautz and Pfeifle, 1989; Jiang et al.,taining a mef2 cDNA clone under the control of yeast up-1991). Antisense RNA probes were synthesized using a digoxi-stream activator sequence (UAS) are crossed to transgenicgenin-labeled uracil analogue as described by the manufacturerflies containing a GAL4 cDNA clone driven by a tissue-(Boehringer-Mannheim). TmI antisense RNA was prepared from aspecific genomic enhancer or a heat-inducible enhancer2.3-kb TmI genomic DNA inserted into pGem-1 which contains

sequence from exon 2 to exon 5 (Basi and Storti, 1986). The plasmid (Brand and Perrimon, 1993). The 69B epidermal enhancerwas linearized by digestion with HindIII and transcribed using T7 line expresses GAL4 in epidermal precursors beginning atRNA polymerase. Antisense b3-Tubulin (b3-Tub) RNA was synthe- late stage 9 and then in the epidermis (Brand and Perrimon,sized from a 0.6-kb b3-Tub cDNA cloned into pGem-1 at the PstI– 1993). It also expresses GAL4 in a set of ventral midlineSacI site (Gasch et al., 1988). The plasmid was linearized by diges- cells underneath the epidermis starting at stage 10. Whiletion with PstI and transcribed using T7 RNA polymerase. Anti- the identities of these ventral midline cells are unknown,sense S59 RNA was synthesized from a 2.6-kb S59 cDNA (Dohr-

their position in the mid-ventral region of the embryo sug-mann et al., 1990). The plasmid was linearized by digestion withgests that they may be ectodermally derived neuroblast pre-PstI and transcribed using T3 RNA polymerase. Antisense naucursors that give rise to midline glial cells. Embryos ob-RNA was synthesized from a 1.3-kb nau cDNA cloned into pBlue-tained from the cross of flies carrying the 69B epidermalscript KS at the EcoRI site (Michelson et al., 1990). The plasmid

was linearized by digestion with BamHI and transcribed using T3 enhancer to flies carrying the UAS-mef2 transgene ex-RNA polymerase. Antisense DMLP1 RNA was synthesized from a pressed high levels of MEF2 in the epidermal precursors and0.5-kb DMLP1 cDNA cloned into pBluescript SK at the EcoRI site the ventral midline cells from stage 10 onward (not shown)(Arber et al., 1994). The plasmid was linearized by digestion with as determined by MEF2 antibody staining. These embryosXhoI and transcribed using T3 RNA polymerase. Antisense aPS1 did not hatch and died as late embryos and showed no tra-integrin RNA (Leptin et al., 1989) was synthesized from a Ç5-kb

chea formation and no body-wall movement. These em-aPS1 cDNA clone obtained from Dr. T. Bunch. The plasmid was

bryos also showed a failure to undergo dorsal closure, re-linearized by digestion with HindIII and transcribed using T7 RNAsulting in a large opening on the dorsal side, through whichpolymerase. Antisense aPS2 integrin RNA (Bogaert et al., 1987) wasthe gut was extruded (Fig. 1E). In embryos in which MEF2synthesized from a 4.4-kb aPS2 cDNA clone obtained from Dr. R.was ectopically expressed in the epidermis and the ventralSchulz. The plasmid was linearized by digestion with XhoI and

transcribed using T3 RNA polymerase. Antisense crumbs (crb) midline cells, TmI expression was detected in these twoRNA was synthesized from an Ç6-kb crb cDNA subclone of crb- tissues as well as in muscles. The ectopic expression of TmIpUAST (Tepass and Knust, 1993) inserted into pGEM-3Zf at the in the epidermis was observed beginning at stage 13 andEcoRI–XbaI site. The plasmid was linearized by digestion with persisted at later stages. The ectopic expression was de-EcoRI and transcribed using SP6 RNA polymerase. Antisense deli- tected uniformly throughout the epidermis and most clearlylah (dei) RNA was synthesized from a 1.3-kb dei cDNA cloned into

seen on the ventral side due to the absence of underlyingpBluescript KS at the EcoRI site (Armand et al., 1994). The plasmidmuscles expressing TmI (compare Figs. 1B and 1D with 1Awas linearized by digestion with BamHI and transcribed using T3and 1C, respectively) and the anterior and posterior regionsRNA polymerase. Antisense decapentaplegic (dpp) RNA was syn-(Fig. 1E) of the embryo where there was less underlying TmIthesized from an Ç3-kb dpp cDNA clone obtained from Dr. E. L.transcript in muscles. The ectopic expression of TmI in theFerguson (Ferguson and Anderson, 1992). The plasmid was linear-

ized by digestion with BamHI and transcribed using T7 RNA poly- ventral midline cells was also observed starting at stage 13merase. The 2.9-kb wingless (wg) cDNA in pSP65 (Rijsewijk et al., and was strongest at stage 15 (Fig. 1F). These results indicate1987) was subcloned into pGEM-3Zf at the BamHI site. The plas- that MEF2 can function to induce TmI expression in themid was linearized by digestion with EcoRI and transcribed using epidermis and the ventral midline cells.SP6 RNA polymerase to make antisense wg RNA. Embryos that ectopically expressed MEF2 in the epider-

mis and the ventral midline cells also showed ectopic ex-pression of other muscle genes such as b3-Tub, nau, andRESULTSaPS2 genes in these two tissues and the DMLP1 gene in theventral midline cells. b3-Tub encodes a cytoskeletal protein

Ectopic Expression of MEF2 in Epidermis and found in myoblasts, muscle cells, and some nonmuscle cellsVentral Midline Cells Results in Ectopic (Gasch et al., 1988; Leiss et al., 1988). It is expressed inExpression of TmI and Other Muscle Genes normal embryos in the somatic mesoderm beginning at

stage 11 when the germ band is fully extended. Its expres-MEF2 functions as a positive regulator of TmI gene tran-scription in developing muscle (Lin et al., 1995) and mef2 sion persists during germ-band retraction and can be fol-


AID DB 8484 / 6x19$$$202 01-18-97 03:24:43 dbal


FIG. 1. Ectopic expression of TmI in embryos that ectopically express MEF2 in the epidermis and the ventral midline cells. Embryosobtained from the cross of flies carrying the 69B epidermal enhancer to flies carrying the UAS-mef2 transgene were assayed for TmIexpression by whole-mount in situ hybridization. All embryos are stage 15 and oriented with the anterior to the left. (A) Wild-type UAS-mef2 embryo, ventral view. No ectopic TmI expression is detected in the epidermis of the embryo. TmI expression in the ventral body-wall muscles lateral to the ventral midsection of the embryo and beneath the epidermis is out of focus. (B) Ectopic MEF2 embryo, ventralview. Ectopic expression of TmI is seen clearly in the epidermis of the ventral side of the embryo. The ectopic expression of TmI in theanterior and lateral epidermis and ventral midline cells beneath the epidermis (arrow) is out of focus. C and D are higher magnificationsof A and B, respectively. (E) Ectopic MEF2 embryo, dorsal view. Ectopic expression of TmI is seen clearly in the epidermis in the anteriorand posterior regions on the dorsal side of the embryo. A failure in dorsal closure, which leads to extrusion of the gut, is also observed.(F) Ectopic MEF2 embryo, ventral view. Ectopic expression of TmI is observed in the ventral midline cells (arrow). TmI expression in thethoracic and anterior abdominal epidermis is out of focus.

lowed into somatic muscle fibers at later stages (Gasch et sion continued to later stages and was most obvious in theventral epidermal and midline cells (compare Figs. 2B andal., 1988; Leiss et al., 1988). In embryos that ectopically

expressed MEF2 in the epidermis and the ventral midline 2D with 2A and 2C, respectively) and the anterior and poste-rior regions of the epidermis (not shown). nau is expressedcells, the ectopic expression of b3-Tub in these tissues was

first observed at stage 12 following the stage when b3-Tub in a subset of body-wall muscle precursor cells at late germ-band extension (Michelson et al., 1990; Paterson et al.,is normally activated in the mesoderm. The ectopic expres-


AID DB 8484 / 6x19$$$203 01-18-97 03:24:43 dbal

244 Lin et al.

1991) and is required for the transition of these cells into Embryos derived from the cross of flies carrying the 69Bepidermal enhancer to flies carrying the UAS-mef2syncytial muscle fibers (Abmayr et al., 1995). In embryos

that ectopically expressed MEF2 in the epidermis and ven- transgene expressed MEF2 ectopically in the epidermal pre-cursors and the ventral midline cells from stage 10 onwardtral midline cells, nau was also expressed in these tissues;

however, the ectopic expression of nau was not detected as examined by MEF2 antibody staining (not shown). How-ever, the ectopic expression of TmI and the other muscleuntil stage 14, later than when nau is first detected in the

mesoderm in normal embryos. In addition, the ectopic ex- genes tested in the epidermis and the ventral midline cellswas not detected until stage 12 or later after stages whenpression appeared mostly in the epidermis of the head re-

gion (compare Fig. 2F with 2E), the ventral midline cells they are normally activated in the mesoderm. One explana-tion for this delayed activation is that the accumulated levellocated in the anterior region (not shown), and at a lower

level in the epidermis of the body wall (compare Fig. 2H of ectopically expressed MEF2 in the epidermis and the ven-tral midline cells at earlier stages is not sufficient to activatewith 2G). aPS2 encodes the a subunit of muscle integrin and

its transcripts are detected in the mesoderm in early-stage ectopic expression of muscle genes in these tissues. An al-ternative possibility, however, is that some other factor(s)embryos and then in the visceral and somatic muscles in

late-stage embryos (Bogaert et al., 1987; Leptin et al., 1989). not present in the epidermis and the ventral midline cellsuntil stage 12 or later is needed to act in conjunction withIn embryos that ectopically expressed MEF2 in the epider-

mis and the ventral midline cells, the ectopic expression of MEF2 to induce ectopic expression of muscle genes in thesetissues. We tested these possibilities by crossing flies car-aPS2 was detected in the ventral midline cells starting at

stage 13 and reached maximum by stage 15 (compare Fig. 2L rying the hsGAL42-1 enhancer to flies carrying the UAS-mef2 transgene and heat-shocking embryos obtained at gas-with 2K) and detectable at about stage 16 in the epidermis of

the head region (compare Fig. 2J with 2I). DMLP1 encodes trulation to induce ectopic expression of MEF2 at earlierstages. These heat-shocked embryos had a phenotype simi-a Drosophila homologue of the vertebrate muscle LIM pro-

tein which is a positive regulator of myogenic differentia- lar to flies carrying the 69B epidermal enhancer and theUAS-mef2 transgene in that they died late in embryogene-tion (Arber et al., 1994). DMLP1 transcripts are detected in

the visceral and somatic mesoderms in normal embryos sis, and did not show tracheal formation, or body-wallmovement. Furthermore, they also showed a defect in dor-starting at stage 13 (Arber et al., 1994). In contrast to the

above, ectopic expression of DMLP1 was observed only in sal closure, resulting in a large opening on the dorsal side.Non-heat-shocked embryos from the same cross or wild-the ventral midline cells but not in the epidermis of em-

bryos that ectopically expressed MEF2 in the epidermis and type embryos heat-shocked at gastrulation were onlyslightly affected since more than 80% of the embryos grewventral midline cells. The ectopic expression of DMLP1 in

the ventral midline cells was detected starting at stage 13 to adults.The heat-shocked embryos from the cross of flies carryingwhen DMLP1 is normally activated in the mesoderm and

peaked at about stage 15 (compare Fig. 2N with 2M). An- the hsGAL42-1 enhancer to flies carrying the UAS-mef2transgene showed strong ectopic expression of MEF2 in theother muscle gene examined, S59, a homeobox gene that is

expressed at late germ-band extension in a small subset presumptive epidermis and amnioserosa cells by stage 7which reached maximum between stages 8 and 12 (Fig. 3A).of myoblasts that give rise to three muscles in abdominal

hemisegments 2 to 8 (Dohrmann et al., 1990), did not show Ectopic expression of TmI in the epidermis, however, wasnot observed until stage 12 (Figs. 3B–3D). We also examinedany ectopic expression in the epidermis or the ventral mid-

line cells of these embryos. these heat-shock embryos for MEF2 induction of nau and

FIG. 2. Ectopic expression of b3-Tub, nau, aPS2, and DMLP1 in embryos that ectopically express MEF2 in the epidermis and the ventralmidline cells. Embryos derived from the cross of flies carrying the 69B epidermal enhancer to flies carrying the UAS-mef2 transgene wereassayed for expression of b3-Tub (A–D), nau (E–H), aPS2 (I–L), and DMLP1 (M and N) by whole-mount in situ hybridization. All embryosare stage 15 and oriented with the anterior to the left. (A) Wild-type UAS-mef2 embryo, ventral view. No ectopic b3-Tub expression isdetected in the epidermis. (B) Ectopic MEF2 embryo, ventral view. Ectopic expression of b3-Tub is observed in the epidermis and theventral midline cells (arrow; out of focus) beneath the epidermis. C and D are higher magnifications of A and B, respectively. (E) A highermagnification of a dorsal view of a wild-type UAS-mef2 embryo. nau expression is seen in the body-wall muscle of the first thoracicsegment. No ectopic expression of nau is detected in the epidermis of the head. (F) A higher magnification of a dorsal view of an ectopicMEF2 embryo. Ectopic expression of nau is seen in the epidermis of the head. (G) A higher magnification of a lateral view of a wild-typeUAS-mef2 embryo. No ectopic expression of nau is detected in the epidermis on the lateral side of the body wall. (H) A higher magnificationof a lateral view of an ectopic MEF2 embryo. Ectopic expression of nau is detected in the epidermis on the lateral side of the body wall.(I) A higher magnification of a dorsal view of a wild-type UAS-mef2 embryo. No ectopic aPS2 expression is detected in the epidermis ofthe head region. (J) A higher magnification of a dorsal view of an ectopic MEF2 embryo. Ectopic expression of aPS2 is observed in theepidermis of the head region. (K) Wild-type UAS-mef2 embryo, ventral view. No ectopic aPS2 expression is detected in the ventral midlinecells. (L) Ectopic MEF2 embryo, ventral view. Ectopic expression of aPS2 is observed in the ventral midline cells (arrow). (M) Wild-typeUAS-mef2 embryo, ventral view. No ectopic DMLP1 expression is detected in the ventral midline cells. (N) Ectopic MEF2 embryo, ventralview. Ectopic expression of DMLP1 is observed in the ventral midline cells (arrow).


AID DB 8484 / 6x19$$$203 01-18-97 03:24:43 dbal


01-18-97 03:24:43 dbal

246 Lin et al.

shown). These embryos showed no gross defects in formationof the central and peripheral nervous systems using themonoclonal antibody 22C10 (not shown) which recognizesa cell-surface neuronal antigen (Fujita et al., 1982). ThusMEF2 in the nervous system affects dorsal closure and sur-vival and may affect innervation leading to slow body-wallmovement. MEF2 expression in the nervous system, how-ever, is not sufficient to activate muscle gene expression.

Ectopic Expression of MEF2 in Epidermis andVentral Midline Cells Results in AbnormalFIG. 3. Ectopic expression of MEF2 and TmI in the epidermis ofBody-Wall Muscle Formationembryos by heat shock. Embryos derived from the cross of flies

carrying the hsGAL42-1 enhancer to flies carrying the UAS-mef2 In addition to the ectopic expression of TmI in the epider-transgene were heat-shocked at gastrulation and assayed for the

mis and the ventral midline cells, the pattern of TmI expres-expression of MEF2 (A) by antibody staining using a serum con-sion in the body-wall muscle of embryos ectopically ex-taining MEF2 antibody and for the expression of TmI (B–D) bypressing MEF2 in the epidermis and the ventral midlinewhole-mount in situ hybridization. All embryos are oriented withcells was abnormal. In a typical embryo, only muscle VA2the anterior to the left. (A) Stage 8, lateral view. Ectopic expression(nomenclature according to Bate, 1993) showed a nearly nor-of MEF2 is observed in the nuclei of the presumptive epidermis

and amnioserosa cells which are located dorsally and less packed mal level of TmI expression among the ventral group ofthan the presumptive epidermis. (B) Stage 10 heat-shocked embryo, body-wall muscles, although muscle VA2 was much shorterlateral view. Ectopic expression of TmI is not detected in the epider- than the equivalent muscle of wild-type embryos (comparemis. (C) Early stage 12 heat-shocked embryo, lateral view. Ectopic Fig. 4B with 4A). The other ventral group muscles showedexpression of TmI is detected in the epidermis. (D) A higher magni- a substantially decreased level of TmI expression (comparefication of C. TmI expression in the underlying developing muscles

Fig. 4D with 4C). The few muscles located at the mostis not detected until slightly later at late stage 12.ventral region, VO4 to VO6, were also much shorter thanthe equivalent muscles of wild-type embryos and they wereoften fused to one another. The four lateral group body-wall muscles, LT1, LT2, LT3, and LT4 (or possibly DT1),b3-Tub. b3-Tub and nau showed epidermal expression be-

ginning at stages 12 and 16, respectively (not shown), which expressed TmI but were disorganized (compare Fig. 4F with4E). The other lateral group muscles, on the other hand,is the same as the results obtained from embryos carrying

the 69B epidermal enhancer and the UAS-mef2 transgene. were not visualized by detection of TmI expression (com-pare Fig. 4H with 4G). It is possible that these musclesThese results suggest that even though MEF2 can be acti-

vated and accumulated to high levels in the epidermis of were missing or unable to express TmI. Similarly, severalmuscles of the dorsal group body-wall muscles in these em-earlier stage embryos, MEF2 is not sufficient to activate

TmI and that another factor(s) is required which is not pres- bryos were not visualized by detection for TmI expression(not shown). The defective muscle formation would accountent in the epidermis until stage 12.

As an additional test of the above we examined TmI ex- for the fact that these embryos showed no movement. Theformation of the viscera and dorsal vessel, on the otherpression in flies in which MEF2 is ectopically expressed in

the nervous system. The GAL4-1407 neuronal enhancer line hand, appeared normal and thus was not severely affectedby ectopic expression of MEF2 in the epidermis and theexpresses GAL4 in neuroblasts beginning at stage 10 and

subsequently in most neurons of the central nervous system ventral midline cells.and all neurons in the peripheral nervous systems (Luo etal., 1994). In embryos obtained from the cross of the GAL4- Effect of Ectopic Expression of MEF2 in Epidermis1407 neuronal enhancer to flies carrying the UAS-mef2

and Ventral Midline Cells on Myoblast Patterningtransgene we observed high levels of MEF2 in the nervoussystem comparable to that observed in the epidermis of em- Ectopic expression of MEF2 in the epidermis and the ven-

tral midline cells of embryos results in abnormal body-wallbryos carrying the 69B epidermal enhancer and the UAS-mef2 transgene. However, in contrast to experiments in muscle formation as examined by TmI expression in mus-

cles. To test whether these embryos have a disrupted my-which MEF2 is ectopically expressed in the epidermis andthe ventral midline cells, embryos which expressed MEF2 oblast population, we analyzed by whole-mount in situ hy-

bridization the expression of b3-Tub which serves as aectopically in the nervous system never showed ectopic ex-pression of TmI or any other muscle markers in neurons, marker of myoblasts. In embryos that ectopically expressed

MEF2 in the epidermis and the ventral midline cells, theand showed normal muscle morphology and TmI expressionlevels in muscles (not shown). These embryos, however, died expression of b3-Tub in the mesoderm and muscle in late

germ-band retraction appeared to be normal (Figs. 5A andat late stages. They had normal looking trachea and slowbody-wall movement. They also showed incomplete dorsal 5B). However, its expression at later stages during body-wall

muscle formation was abnormal (not shown) and similar toclosure, resulting in a small opening on the dorsal side (not


AID DB 8484 / 6x19$$$203 01-18-97 03:24:43 dbal


To further assess how body-wall muscles were abnor-mally formed in these embryos, we examined muscle pre-cursor cells for their expression of nau and S59. In embryosin which MEF2 was ectopically expressed in the epidermisand the ventral midline cells, the expression of nau in themesoderm was moderately reduced compared to wild-typeembryos beginning at stage 11 when nau is normally acti-vated in the mesoderm (compare Fig. 5D with 5C). Thereduction became more distinct, however, in the body-wallmuscle at the end of germ-band shortening (compare Fig.5F with 5E). The reduced level of nau expression in themesoderm could in part account for the defects in the forma-tion of some nau-expressing muscle fibers in these embryos;e.g., VO4 to VO6 were shorter and often fused to one an-other and showed a reduced level of TmI expression (Fig.4D). The expression of S59 in these embryos, on the otherhand, appeared to be normal (not shown), consistent withthe results described earlier that muscles VA2 and possiblyDT1, two S59-expressing muscle fibers, were formed inthese embryos. These results suggest that the formation ofthe general myoblast population in embryos that ectopi-

FIG. 4. Abnormal body-wall muscle formation in embryos thatectopically express MEF2 in the epidermis and the ventral midlinecells. Embryos obtained from the cross of flies carrying the 69Bepidermal enhancer to flies carrying the UAS-mef2 transgene wereassayed for TmI expression by whole-mount in situ hybridization.All embryos are stage 17 with abdominal muscles viewed at highmagnifications and oriented with the anterior to the left. (A) Wild-type UAS-mef2 embryo, ventrolateral view. Muscles VA1 and VA2of the ventral group body-wall muscles (bracket) are visualized byTmI expression. (B) Ectopic MEF2 embryo, ventrolateral view. Mus-cle VA1 of the ventral group body-wall muscles (bracket) is missing.Muscle VA2 (arrow) is much shorter than that found in wild-typeembryos and shows a nearly normal level of TmI expression. (C)A deeper focal plane of A. The other underlying ventral group body-wall muscles are now visualized by TmI expression. The threemuscles located most ventrally, VO4 to VO6, are indicated. (D) A

FIG. 5. Expression of b3-Tub and nau in the mesoderm and mus-deeper focal plane of B. The other ventral group body-wall musclescles in embryos that ectopically express MEF2 in the epidermisshow a substantially decreased level of TmI expression, e.g., mus-and the ventral midline cells. Embryos derived from the cross ofcles VO4 to VO6 (bracket). Muscles VO4 to VO6 are also muchflies carrying the 69B epidermal enhancer to flies carrying the UAS-shorter than those found in wild-type embryos and often fused tomef2 transgene were assayed for expression of b3-Tub (A and B)one another. (E) Wild-type UAS-mef2 embryo, lateral view. Musclesand nau (C–F) by whole-mount in situ hybridization. All embryosLT1 to LT4 of the lateral group body-wall muscles (bracket) andare oriented with the anterior to the left. (A) Wild-type UAS-mef2muscle DT1 of the dorsal group body-wall muscles are visualizedembryo, stage 12, lateral view. b3-Tub is detected in the body-wallby TmI expression. (F) Ectopic MEF2 embryo, lateral view. TmI ismuscle (arrowhead) and the visceral muscle (arrow). (B) Ectopicexpressed in muscles LT1 to LT3 (small bracket) of the lateral groupMEF2 embryo, stage 12, lateral view. b3-Tub expression in thebody-wall muscles (large bracket); however, these muscles are dis-body-wall muscle and the visceral muscle appears normal. Someorganized. Another TmI-expressing muscle observed is possiblyof the out of focus staining is due to the ectopic expression of b3-LT4 or DT1 (arrow). (G) A deeper focal plane of E. The other under-Tub in the epidermis. (C) Wild-type UAS-mef2 embryo, stage 11,lying lateral group body-wall muscles are now visualized by TmIdorsal view. nau expression is detected in clusters of cells in theexpression. (H) A deeper focal plane of F, showing an absence ofmesoderm. (D) Ectopic MEF2 embryos, stage 11, dorsal view. nauthe other lateral group body-wall muscles.expression in the mesoderm is moderately reduced. (E) Wild-typeUAS-mef2 embryo, stage 13, lateral view. nau is expressed in clus-

the pattern of TmI expression. Thus the initial cell-fate ters of cells aligned in three rows in each segment of the body-determination of the general myoblast population was not wall muscle. (F) Ectopic MEF2 embryo, stage 13, lateral view. nau

expression level in the body-wall muscle is dramatically decreased.markedly affected in these embryos.


AID DB 8484 / 6x19$$$203 01-18-97 03:24:43 dbal

248 Lin et al.

cally expressed MEF2 in the epidermis and the ventral mid-line cells is not affected. However, the development of nau-expressing founder cells is defective.

We also examined the expression of MEF2 in these em-bryos by antibody staining to assess whether the abnormalbody-wall muscle formation resulted from aberrant expres-sion of MEF2 in the mesoderm and muscles. The expressionof MEF2 by stage 12 in the mesoderm of these embryos wasnormal (not shown). In addition, we could not detect anyincrease in programmed cell death using acridine orange toselectively stain apoptotic cells (Abrams et al., 1993). Thusit appears that ectopic expression of MEF2 in the epidermisand the ventral midline cells results in the failure of mus-cles to fuse and differentiate properly to form muscle fibers.

Ectopic Expression of MEF2 in Epidermis AffectsEpidermal Differentiation and Signalingof dpp and wg

Several studies have reported that the ectoderm may play FIG. 6. Expression of aPS1 and dei in embryos that ectopicallyexpress MEF2 in the epidermis and the ventral midline cells. Em-an inductive role in muscle development (Beer et al., 1988;bryos derived from the cross of flies carrying the 69B epidermalBate et al., 1993; Baker and Schubiger, 1995). Thus the ab-enhancer to flies carrying the UAS-mef2 transgene were assayednormal body-wall muscle formation in the embryos thatfor expression of aPS1 (A and B) and dei (C–F) by whole-mount inectopically expressed MEF2 in the epidermis and the ventralsitu hybridization. All embryos are oriented with the anterior tomidline cells is likely due to a disrupted development ofthe left. (A) Wild-type UAS-mef2 embryo, stage 12, lateral view.the epidermis. As mentioned previously, the epidermis inaPS1 expression is detected in the segmental boundaries (arrow). (B)

these embryos failed to undergo dorsal closure, suggesting Ectopic MEF2 embryo, stage 12, lateral view. aPS1 expression in thethat the epidermal cells of these embryos lacked the ability segmental boundaries appears normal. The staining of the midgutto undergo dorsal migration. To assess whether the epider- is out of focus. (C–F) Stage 16, viewed at high magnifications. (Cmis in these embryos was capable of differentiation, we and E) The lateral and ventrolateral views of a wild-type UAS-mef2

embryo, respectively. dei is expressed in the muscle attachmentexamined the expression of crumbs (crb) and the secretionsites in the segmental boundaries (arrows) and intrasegmental re-of cuticle. crb encodes an EGF-like protein that is localizedgions (brackets). (D and F) The lateral and ventrolateral views ofon the apical membranes of epithelial cells and is requiredan ectopic MEF2 embryo, respectively. The pattern of dei expres-for the organization of epithelial cells (Tepass et al., 1990;sion in the segmental boundaries (arrows) and intrasegmental re-Grawe et al., 1996). The expression of crb in these embryosgions (brackets) is diffused and distorted.appeared to be normal (not shown). These embryos also

secreted cuticle as examined by the formation of ventraldenticle belts (not shown) (Struhl, 1989); however, becausethese embryos had distorted body walls due to a failure to at stage 8 and becomes concentrated in the segmental

boundaries at later stages (Leptin et al., 1989). The expres-undergo dorsal closure, it was not possible to determinewhether the pattern of ventral denticles was normal. Thus sion pattern of aPS1 in embryos that ectopically expressed

MEF2 in the epidermis appeared to be normal (Figs. 6A andthe epidermis of these embryos was capable of some differ-entiation in that it could express crb and secreted cuticle 6B). The expression of dei in normal embryos is observed

in a small cluster of cells in the middle of the body-walldespite the fact that MEF2 and other muscle genes wereectopically expressed in this tissue. segments, which may correspond to epidermal cells that

form intrasegmental clusters of muscle-attachment sitesEctopic expression of MEF2 in the epidermis, however,may affect the ability of the epidermis to form proper mus- (Armand et al., 1994). Subsequently, dei is expressed in the

muscle-attachment sites in the segmental boundaries. Em-cle-attachment sites, signal the mesoderm to differentiateto form muscle fibers, or serve as a matrix for mesoderm bryos ectopically expressing MEF2 in the epidermis showed

what appeared to be normal levels of dei expression; how-migration and myoblast fusion. To test whether embryosthat ectopically expressed MEF2 in the epidermis formed ever, the pattern of expression was slightly diffused and

distorted compared to normal embryos (compare Figs. 6Dproper muscle-attachment sites in the epidermis, we exam-ined these embryos by whole-mount in situ hybridization and 6F with 6C and 6E, respectively). This might be due to

the irregular shape of epidermal cells and the distortion offor the expression of aPS1 which encodes the a subunit ofepidermal integrin and dei which encodes a bHLH protein. the epidermis resulting from the defect in epidermal migra-

tion and dorsal closure. Thus it appears that muscle-attach-Both of these proteins are expressed in the muscle-attach-ment sites of the epidermis. In normal embryos, aPS1 tran- ment sites in the epidermis form in these embryos. How-

ever, it is possible that they are not completely normal.script is detected in the presumptive epidermis beginning


AID DB 8484 / 6x19$$$203 01-18-97 03:24:43 dbal


7B with 7A). It has been reported that the GAL4/UAS sys-tem is temperature-sensitive and more extreme phenotypesare observed when embryos are collected at a higher temper-ature (297C) (Staehling-Hampton et al., 1994). The level ofexpression of dpp in the visceral mesoderm, however, wasunaffected (compare Fig. 7B with 7A).

Several reports have demonstrated that wg plays a role inintercellular communication (Morata and Lawrence, 1977;Wieschaus and Riggleman, 1987; Baker, 1988). Indeed, re-cent studies have shown that WG expressed in the ectodermcan act as an inductive signal for the initiation of the devel-opment of subsets of somatic muscles (Baylies et al., 1995;Ranganayakulu et al., 1996). In normal embryos, wg is ex-pressed in 14 stripes in the ectoderm at blastoderm andFIG. 7. dpp and wg expression in embryos that ectopically express

MEF2 in the epidermis and the ventral midline cells. Embryos ob- gastrulation and during germ-band extension (Baker, 1988;tained from the cross of flies carrying the 69B epidermal enhancer van den Heuvel et al., 1989). Before germ-band retraction,to flies carrying the UAS-mef2 transgene were assayed for expres- the striped pattern of wg expression becomes discontinuoussion of dpp (A and B) and wg (C and D) by whole-mount in situ and separated into dorsal (Fig. 7C) and ventral (Fig. 7C, nothybridization. All embryos are stage 13, lateral view, and oriented in focus) sectors. wg expression is also detected in paraseg-with the anterior to the left. (A) Wild-type UAS-mef2 embryo. dpp ment 8 in the visceral mesoderm after germ-band retractionis expressed in two longitudinal stripes in the epidermis and in

is complete (Fig. 7C). In embryos that ectopically expressedparasegments 4 and 7 of the visceral mesoderm (arrows; out ofMEF2 in the epidermis, wg expression in the dorsal sectorfocus). (B) Ectopic MEF2 embryo. dpp expression in both stripes inof the epidermis was dramatically reduced in stage 13 em-the epidermis is dramatically decreased, whereas that in the vis-bryos, whereas that in the ventral sector appeared to be onlyceral mesoderm appears unaffected (arrows; out of focus). (C) Wild-

type UAS-mef2 embryo. wg is expressed in two sectors in the epi- slightly affected (not shown). However, when embryos weredermis. The ventral sector is not in focus. wg is also expressed in collected at a higher temperature (297C), the reduced expres-parasegment 8 of the visceral mesoderm (arrow; out of focus). (D) sion of wg in the dorsal sector became more distinct (com-Ectopic MEF2 embryo. wg expression is dramatically reduced in pare Fig. 7D with 7C), and the expression of wg in the ven-the dorsal sector and substantially reduced in the ventral sector (not tral sector was substantially decreased (Fig. 7D, out of fo-in focus) in the epidermis, whereas that in the visceral mesoderm

cus). wg expression in the visceral mesoderm, however, wasappears unaffected (arrow; out of focus).unaffected (compare Fig. 7D with 7C). These results suggestthat disruption of the epidermal signaling of wg is one ofthe factors resulting in the abnormal body-wall muscle for-mation observed in embryos in which MEF2 is ectopicallyRecent reports have revealed that the epidermal signaling

of dpp, a member of the TGF-b family of secreted growth expressed in the epidermis.factors (Rijsewijk et al., 1987), and wg, a segment polaritygene which encodes a secreted protein (St. Johnston and

Misexpression of MEF2 in Mesoderm and MusclesGelbart, 1987), are important for the formation of visceralResults in a Failure of Myoblast Fusionand body-wall muscles, respectively (Bate and Rushton,

1993; Staehling-Hampton et al., 1994; Frasch, 1995; Baylies Since MEF2 is a transcriptional regulator of TmI expres-sion and body-wall muscle differentiation, we suspectedet al., 1995; Ranganayakulu et al., 1996). To assess whether

embryos that ectopically expressed MEF2 in the epidermis that its ability to function might be sensitive to its level ofexpression in the mesoderm and muscles. The 24B mesoder-are capable of sending signals from the epidermis to the

mesoderm, we examined the expression of dpp and wg in mal enhancer line expresses GAL4 in most, if not all, cellsin the mesoderm starting at stage 10 and then in the somaticthese embryos. dpp is expressed in normal embryos in the

dorsal ectoderm at blastoderm and gastrulation and and visceral muscles and the muscles of the dorsal vessel(Brand and Perrimon, 1993; Bour et al., 1995). Embryos de-throughout germ-band extension (St. Johnston and Gelbart,

1987). Prior to the start of germ-band shortening, the expres- rived from the cross of flies carrying the 24B mesodermalenhancer to flies carrying the UAS-mef2 transgene died assion pattern is transformed into two thin horizontal stripes

on each side of the embryo, one stripe of dorsal cells that late embryos. These embryos showed some body-wallmovement; however, analysis of TmI expression in theseborder the amnioserosa, and one stripe of ventrolateral cells

that border the neurogenic ectoderm (Fig. 7A). dpp is also embryos showed an abnormal pattern of body-wall musclesand several unfused myoblasts expressing TmI were ob-expressed in parasegments 4 and 7 of the visceral mesoderm

starting at this stage (Fig. 7A). In embryos that ectopically served in the body-wall muscles at late stages (compare Fig.8B with 8A). This phenotype was considerably enhanced inexpressed MEF2 in the epidermis, the expression of dpp in

the two stripes of the epidermis was decreased beginning embryos collected from this cross and raised at 297C. Atthe higher temperature embryos had no body-wall move-at stage 12. The decrease was more dramatic when embryos

were collected at a higher temperature (297C) (compare Fig. ment and showed a more profound derangement of body-


AID DB 8484 / 6x19$$$203 01-18-97 03:24:43 dbal

250 Lin et al.

FIG. 8. Overexpression of MEF2 in the mesoderm and muscles leads to abnormal body-wall muscle formation, abnormal expression ofnau, and ectopic expression of DMLP1 in embryos. Embryos derived from the cross of flies carrying the 24B mesodermal enhancer to fliescarrying the UAS-mef2 transgene were collected at 257C (A, B, and E–J) or 297C (C and D) and were assayed for TmI (A–D), nau (E–H),and DMLP1 (I and J) expression by whole-mount in situ hybridization. All embryos are oriented with the anterior to the left. (A–D) Stage16, viewed at high magnifications. (A) Wild-type UAS-mef2 embryo, ventrolateral view, showing TmI expression in the ventral groupmuscles (bracket) of the body wall. (B) MEF2 overexpression embryo, ventrolateral view, showing some unfused myoblasts expressing


AID DB 8484 / 6x19$$8484 01-18-97 03:24:43 dbal


wall muscle pattern and many more unfused myoblasts, DISCUSSIONespecially in the dorsal group muscles (compare Fig. 8Dwith 8C). Therefore, the level of MEF2 in the mesoderm Ectopic Expression of MEF2 in the Epidermis Leadsand/or muscles is critical to myoblast fusion and/or differ- to Activated Transcription of Muscle Genesentiation and body-wall muscle formation. However, over-expression of MEF2 in the mesoderm and/or muscles did The experiments presented here demonstrate that ectopicnot have a noticeable effect on the formation of the viscera expression of MEF2 in the epidermis and the ventral mid-and the dorsal vessel or TmI expression in these two mus- line cells, tissues where MEF2 is not normally expressed,cles (not shown), suggesting that MEF2 is not a major deter- can activate transcription of TmI and a number of otherminant of the formation of these two muscle types, which muscle genes tested. There are at least three possible mech-is consistent with previous experiments in analyzing TmI anisms to account for this activation. First, TmI and theexpression in mef2 mutant embryos (Lin et al., 1996). Alter- other muscle genes assayed contain cis-acting regulatorynatively, the level of MEF2 required for the formation of elements that are direct targets for MEF2 and binding ofthe visceral muscle and the dorsal vessel is not as critical MEF2 to these sites alone is able to activate transcription.as it is for the formation of the body-wall muscle. Our previous analysis of TmI enhancer function in

To assess how the body-wall muscle was abnormally transgenic flies has shown that both intron enhancers thatformed in embryos that overexpressed MEF2 in the meso- regulate TmI expression contain MEF2-binding sites andderm and muscles, we examined the expression of nau and that each of these sites is necessary for enhancer functionS59, which serve as markers for founder cells that give rise in body-wall muscles (Lin et al., 1996). These experimentsto specific muscle fibers. In these embryos, the expression also showed, however, that a MEF2-binding site by itself isof nau in the mesoderm at stage 11 when nau is normally not sufficient for transcription in the context of the en-activated appeared unaffected (compare Fig. 8F with 8E); hancer constructs tested. These studies would suggest thathowever, the expression of nau in the developing body-wall MEF2 alone cannot activate TmI, and possibly the othermuscles was substantially decreased by the end of germ- muscle genes tested as well, in the epidermis by bindingband retraction (compare Fig. 8H with 8G). Embryos col- and activation through cognate binding sites. This notionlected at a higher temperature (297C) from this cross showed is further supported by the fact that not all cells ectopicallya more significant decrease in nau expression by the end of expressing MEF2 activate TmI. For instance, expressinggerm-band retraction (not shown). On the other hand, the MEF2 in the nervous system does not lead to TmI expres-expression of S59 was normal (not shown). Thus the early sion. We have also noted that the 24B mesodermal enhancerpatterning of nau and S59 muscle precursors was not af- ectopically expresses MEF2 in small groups of cells abovefected by elevated levels of MEF2; however, subsequent dif- the ventral nerve cord and beneath the epidermis on theferentiation of nau-expressing cells was affected suggesting ventral side of the embryo and that TmI is not ectopicallythat some muscle precursors may be more sensitive to lev- expressed in these cells. Furthermore, ectopically express-els of MEF2 than others. ing MEF2 in the ectoderm and mesoderm earlier in em-

An interesting observation was made when the expres- bryogenesis does not lead to transcriptional activation ofsion pattern of DMLP1 was analyzed. In embryos that over- TmI until much later than would be expected if TmI expres-expressed MEF2 in the mesoderm and muscles, the expression were dependent solely upon the presence of MEF2. Wesion of DMLP1 in the somatic muscle was abnormal (not cannot, however, rule out the possibility that ectopicallyshown), similar to the phenotypes shown by TmI expres- expressed MEF2 in the nervous system or in the early ecto-sion. Interestingly, DMLP1 which is not normally expressed derm and mesoderm is somehow present in an inactive formin the dorsal vessel was ectopically expressed in the dorsal and thus incapable of transcriptional activation.vessel in embryos overexpressing MEF2 in the mesoderm A second hypothesis is that MEF2 activates a pathwayand muscles (compare Fig. 8J with 8I). That overexpression that leads to transcriptional activation of TmI and otherof MEF2 in the dorsal vessel is able to induce the expression muscle genes. MEF2 mutant embryos do not form body-of DMLP1 suggests that DMLP1 may be a target of MEF2- wall muscles suggesting that MEF2 is a key regulator of

muscle differentiation (Bour et al., 1995; Lilly et al., 1995;regulated gene expression in muscles.

TmI in the body-wall muscle (arrows). (C) Wild-type UAS-mef2 embryo, dorsolateral view. TmI expression is detected in the dorsal groupmuscles (bracket) of the body wall. (D) MEF2 overexpression embryo, dorsolateral view. Many more unfused myoblasts expressing TmIand a deranged muscle pattern are observed in the body-wall muscle of embryos raised at the higher temperature. (E) Stage 11 wild-typeUAS-mef2 embryo, dorsal view, showing nau expression in clusters of cells in the mesoderm. (F) Stage 11 MEF2 overexpression embryo,dorsal view, showing nearly normal expression of nau in the mesoderm. (G) Stage 13 wild-type UAS-mef2 embryo, lateral view. nau isexpressed in clusters of cells aligned in three rows in each segment of the body-wall muscle. (H) Stage 13 MEF2 overexpression embryo,lateral view, nau expression level in the body-wall muscle is substantially decreased. (I) Stage 17 wild-type UAS-mef2 embryo, dorsalview. DMLP1 is not detected in the dorsal vessel. (J) Stage 17 MEF2 overexpression embryo, dorsal view. Ectopic expression of DMLP1is observed in the dorsal vessel (arrow).


AID DB 8484 / 6x19$$$204 01-18-97 03:24:43 dbal

252 Lin et al.

Ranganayakulu et al., 1995) and thus ectopic expression of pressed in both the mesoderm and epidermis of later stagewild-type embryos and is not expressed in the nervous sys-MEF2 might lead to activation of a pathway(s) involved in

regulating myogenesis. There are conflicting reports on tem. Thus this factor has characteristics of a common factorthat might facilitate ectopic expression of muscle genes inwhether forced expression of vertebrate MEF2A in fibro-

blasts can activate expression of myogenic bHLH and other the epidermis of late-stage embryos.That the genes,b3-Tub, nau, aPS2 , and DMLP1, are ectopi-muscle genes and lead to the formation of differentiated

myotubes (Kaushal et al., 1994; Molkentin et al., 1995). cally expressed in the epidermis and/or ventral midline cellsof embryos ectopically expressing MEF2 in these tissuesWe have not observed any morphological changes in the

epidermal cells expressing MEF2 that would suggest that indicates that these genes may also be direct targets of MEF2or targets of downstream genes that are regulated by MEF2.they are taking on a muscle-like phenotype. In addition, we

have not detected di- or trinucleate clusters of cells about There are as yet no reports of regulation of these genes byMEF2; however, our results showing that aPS2 might be ato fuse as has been reported for cells ectopically expressing

Twist (Baylies and Bate, 1996). Furthermore, both early em- target of MEF2 are consistent with a report showing thatthe aPS2 gene contains a binding site for MEF2 and thatbryos ectopically expressing MEF2 by heat-shock treatment

in the ectoderm prior to epidermal differentiation and later the aPS2 transcript is not detected in mef2 mutant embryos(Ranganayakulu et al., 1995). Although ectopic expressionstage embryos expressing MEF2 under the control of the 69B

enhancer show some functional epidermal differentiation in of MEF2 in the ventral midline cells is able to activate theexpression of DMLP1 in this tissue, ectopic expression ofthat the epidermis expresses crumbs, an EGF-like molecule

required for epithelial organization, secretes cuticle, and MEF2 in the epidermis is unable to activate the epidermalexpression of DMLP1. This suggests that the ectopicallyexpresses markers indicative of muscle-attachment sites.

Therefore, the cell fate of the epidermis has not been dra- expressed MEF2 in these two ectodermal derivatives is act-ing differently to activate DMLP1 than it is to activate TmI,matically changed by ectopic expression of MEF2 in this

tissue. The epidermis, however, is affected in that the epi- b3-Tub, nau, and aPS2 . In addition, not all epidermal cellsrespond the same to ectopically expressed MEF2 becausedermal cells fail to migrate dorsally to undergo dorsal clo-

sure and fail to form some epidermal-derived organs, such nau and aPS2 are ectopically expressed primarily on the ante-rior dorsal epidermis, whereas TmI and b3-Tub are ex-as the trachea and the filzkorper. It is possible that by the

time ectopic MEF2 is expressed and accumulated to high pressed more uniformly throughout the epidermis. Thesedifferences may reflect differences in the abilities of epider-enough levels to have an effect on the epidermal precursors,

their fates to undergo epithelial organization, secrete cuti- mal cells in different regions of the embryo to respond toectopically expressed MEF2 or that muscle genes are sensi-cle, and form muscle-attachment sites have been deter-

mined while their fates to undergo dorsal closure and form tive to different levels of ectopically expressed MEF2 and/orother factors that might be necessary in different epidermalthe trachea and filzkorper are determined later. Some of

these phenotypes may be in part related to the decreased cells.amounts of DPP made in the epidermis. For instance, theincomplete dorsal closure in these embryos is similar to

Requirement for the Epidermis in Myogenesisthose caused by mutations in genes involved in the dppsignaling pathway (Arora et al., 1995). Our results showing that ectopic expression of MEF2 in

the epidermis of embryos leads to abnormal muscle forma-A third possibility is that an additional factor(s) in theepidermis is required to act with MEF2 to induce ectopic tion suggest a requirement for the epidermis in myogenesis.

Several studies have indicated that the ectoderm may playexpression of TmI and the other muscle genes tested. Forinstance, it has been shown that vertebrate MEF2 acts coop- a role in muscle development. For instance, transplantation

experiments have shown that muscle formation might beeratively with myogenic bHLH proteins to activate skeletalmuscle genes (Molkentin et al., 1995). This notion suggests induced by ectoderm (Beer et al., 1988). Analyses of neuro-

genic mutant embryos have also suggested a role for thethat the epidermis contains a factor(s) in common with themesoderm that can act in conjunction with ectopically ex- ectoderm in muscle development because myoblast fusion

in these embryos is restricted to regions of the mesodermpressed MEF2 to induce TmI expression. Indirect supportfor this notion comes from the fact that we do not observe underlying the residual epidermis caused by the hyperplasia

of the nervous system (Bate et al., 1993). In addition, a re-ectopic expression of muscle genes in the epidermis priorto stage 12 or in the nervous system. Furthermore, in our cent study by arresting gastrulation in Drosophila embryos

has also revealed that muscle determination depends onstudies on TmI regulation we have shown that transcriptionactivation by MEF2 requires a second regulator region called signals provided by the ectoderm (Baker and Schubiger,

1995).the muscle activator region (Gremke et al., 1993; Lin et al.,1996). Moreover, recent studies in one of our laboratories WG secreted by the ectoderm has been shown to act as

an inductive signal for the formation of subsets of body-have identified a novel transcription factor that binds to acis-element within the muscle activator region and regu- wall muscles (Baylies et al., 1995; Ranganayakulu et al.,

1996). Thus the decreased level of wg expression in embryoslates at least in part the function of the muscle activatorregion possibly by interacting with MEF2 (S.-C. Lin and that ectopically express MEF2 in the epidermis is likely to

be responsible in part for the reduced number of body-wallStorti, unpublished data). Interestingly, this factor is ex-


AID DB 8484 / 6x19$$$204 01-18-97 03:24:43 dbal


muscle fibers and the abnormal patterning of the muscle tion (Ranganayakulu et al., 1995). It is likely that too high ortoo low a level of MEF2 will affect concentration-dependentfibers that are formed in these embryos. Although WG se-

creted by the ectoderm has been shown to act as an induc- interactions among factors that regulate enhancer functionand myogenesis.tive signal for the formation of S59-expressing founder cells

(Baylies et al., 1995), embryos that ectopically express MEF2in the epidermis show a normal expression pattern of S59in the mesoderm and muscles. It is possible that the fates ACKNOWLEDGMENTSof S59-expressing founder cells have been determined bythe stage at which epidermal expression of wg is decreased We are grateful to Drs. S. Arber, A. Bejsovec, T. Bunch, M. Cole,or that wg is only slightly affected in those epidermal cells E. L. Ferguson, M. Frasch, E. Knust, R. Renkawitz-Pohl, and R.

Schulz for the cloned DNAs provided. We also thank Drs. A. H.that regulate S59 founder cells.Brand and N. Perrimon for the 24B, 69B, and hsGAL42-1 fly linesIt has been shown that dpp expressed in the ectoderm isand Dr. Y. N. Jan for the GAL4-1407 fly line. This research wasimportant for the formation of the amnioserosa, the dorsalsupported by grants from the National Institute of Health to R.V.S.epidermis-derived organs (Irish and Gelbart, 1987) and theand National Science Foundation to S.M.A.visceral muscle (Staehling-Hampton et al., 1994; Frasch,

1995). In embryos that ectopically express MEF2 in the epi-dermis, the amnioserosa and the visceral muscle are formed

REFERENCESbut the filzkorper, a dorsal epidermis-derived organ, is miss-ing. It is possible that the fate of the epidermal precursors

Abmayr, S. M., Erickson, M. S., and Bour, B. A. (1995). Embryonicto form the filzkorper is determined after the stage at whichdevelopment of the larval body wall musculature of Drosophilaepidermal expression of dpp is decreased. DPP, however,melanogaster. Trends Genet. 11, 153–159.has not been shown to act as an inductive signal to induce

Abrams, J. M., White, K., Fessler, L. I., and Steller, H. (1993). Pro-body-wall muscle formation, and it is unknown whether thegrammed cell death during Drosophila embryogenesis. Develop-

reduced level of dpp expression in embryos that ectopically ment 117, 29 –43.express MEF2 in the epidermis is also involved in the abnor- Andres, V., Fisher, S., Wearsch, P., and Walsh, K. (1995). Regulationmal body-wall muscle formation in these embryos. It has of Gax homeobox gene transcription by a combination of positivealso been shown that dpp expressed in the ectoderm sup- factors including myocyte-specific enhancer factor 2. Mol. Cell.

Biol. 15, 4272–4281.presses the mesodermal expression of pox meso, a putativeArber, S., Halder, G., and Caroni, P. (1994). Muscle LIM protein,myogenic regulatory factor for body-wall muscle formation

a novel essential regulator of myogenesis, promotes myogenic(Staehling-Hampton et al., 1994). Thus altered levels of dpp,differentiation. Cell 79, 221 –231.wg, and probably other signaling molecules as well in em-

Armand, P., Knapp, A. C., Hirsch, A. J., Wieschaus, E. F., and Cole,bryos expressing MEF2 in the epidermis have the potentialM. D. (1994). A novel basic helix–loop–helix protein is expressedto affect a number of myogenic processes leading to thein muscle attachment sites of Drosophila epidermis. Mol. Cell.

phenotypes observed here. Biol. 14, 4145–4154.Our results also indicate that the defects in body-wall Arora, K., Dai, H., Kazuko, S. G., Jamal, J., O’Connor, M. B., Letsou,

muscle formation in the embryos that ectopically express A., and Warrior, R. (1995). The Drosophila schnurri gene acts inMEF2 in the epidermis are not due to an increase in cell the Dpp/TGFb signaling pathway and encodes a transcription

factor homologous to the human BMP family. Cell 81, 781–790.death but due to a failure of the myoblast to differentiate,Baker, N. E. (1988). Localization of transcripts from the winglessfor example, the differentiation of nau-expressing founder

gene in whole Drosophila embryos. Development 103, 289 –298.cells. However, it remains to be determined if and how theBaker, R., and Schubiger, G. (1995). Ectoderm induces muscle-spe-epidermal signaling of wg and dpp affects the expression of

cific gene expression in Drosophila embryos. Development 121,nau in the mesoderm.1387–1398.

Basi, G. B., and Storti, R. V. (1986). Structure and DNA sequenceof the tropomyosin I gene from Drosophila melanogaster. J. Biol.The Amount of MEF2 in the Mesoderm IsChem. 261, 817–827.Important for Normal Myogenesis

Bassel-Duby, R., Hernandez, M. D., Gonzalez, M. A., Krueger, J. K.,The results presented indicate that the level of MEF2 in and Williams, R. S. (1992). A 40-kilodalton protein binds specifi-

cally to an upstream sequence element essential for muscle-spe-the mesoderm and muscle is critical to body-wall musclecific transcription of the human myoglobin prompter. Mol. Cell.formation since overexpression of MEF2 in the mesodermBiol. 12, 5024–5032.and muscles of embryos results in the failure of myoblast

Bate, M. (1993). The mesoderm and its derivatives. In ‘‘The Devel-fusion to form body-wall muscles. These defects may be inopment of Drosophila melanogaster’’ (Bate, M., and Martinezpart due to the failure of muscle precursor founder cellsArias, A., Eds.), pp. 1013–1090. Cold Spring Harbor Press, Coldto differentiate properly since these embryos show normalSpring Harbor, NY.

apoptosis (not shown) and a decrease in nau expression in Bate, M., and Rushton, E. (1993). Myogenesis and muscle patterningthose founder cells that initially express nau. It remains to in Drosophila. C.R. Seances Acad. Sci. 316, 1055–1061.be determined, however, whether this decrease is a direct Bate, M., Rushton, E., and Frasch, M. (1993). A dual requirementconsequence of overexpression of MEF2. Similarly, too low for neurogenic genes in Drosophila myogenesis. Development

119(Suppl.), 149–161.a level of MEF2 can also affect body-wall muscle differentia-


AID DB 8484 / 6x19$$$204 01-18-97 03:24:43 dbal

254 Lin et al.

Baylies, M. K., and Bate, M. (1996). twist: A myogenic switch in expression of b1 and b3 tubulin genes of Drosophila melanogas-ter is spatially regulated during embryogenesis. Mol. Gen. Genet.Drosophila. Science 272, 1481–1484.211, 8–16.Baylies, M. K., Martinez Arias, A., and Bate, M. (1995). wingless is

required for the formation of a subset of muscle founder cells Gossett, L. A., Kelvin, D. J., Sternberg, E. A., and Olson, E. N.(1989). A new myocyte-specific enhancer-binding factor that rec-during Drosophila embryogenesis. Development 121, 3829–

3837. ognizes a conserved element associated with multiple muscle-specific genes. Mol. Cell. Biol. 9, 5022–5033.Beer, J., Technau, G., and Campos-Ortega, J. A. (1988). Lineage anal-

ysis of transplanted individual cells in embryos of Drosophila Grawe, F., Wodarz, A., Lee, B., Knust, E., and Skaer, H. (1996).The Drosophila genes crumbs and stardust are involved in themelanogaster. IV. Commitment and proliferative capabilities of

mesodermal cell. Roux’s Arch. Dev. Biol. 196, 222–230. biogenesis of adherens junctions. Development 122, 951–959.Gremke, L., Lord, P. C. W., Sabacan, L., Lin, S.-C., Wohlwill, A.,Black, B. L., Martin, J. F., and Olson, E. N. (1995). The mouse MRF4

promoter is trans-activated directly and indirectly by muscle- and Storti, R. V. (1993). Coordinate regulation of Drosophila tro-pomyosin gene expression is controlled by multiple muscle-type-specific transcription factors. J. Biol. Chem. 270, 2889–2892.specific positive and negative enhancer elements. Dev. Biol. 159,Bogaert, T., Brown, N., and Wilcox, M. (1987). The Drosophila513–527.PS2 antigen is an invertebrate integrin that, like the fibronectin

receptor, becomes localized to muscle attachments. Cell 51, 929– Hasty, P., Bradley, A., Morris, J. H., Edmondson, D. G., Venuti,940. J., Olson, E. N., and Klein, W. H. (1993). Muscle deficiency and

neonatal death in mice with a targeted mutation in the myogeninBour, B. A., O’Brien, M. A., Lockwood, W. L., Goldstein, E. S.,Bodmer, R., Taghert, P. H., Abmayr, S. M., and Nguyen, H. T. gene. Nature 364, 501–506.(1995). Drosophila MEF2, a transcription factor that is essential Hoshizaki, D. K., Blackburn, T., Price, C., Ghosh, M., Miles, K.,for myogenesis. Genes Dev. 9, 730–741. Ragucci, M., and Sweis, R. (1994). Embryonic fat-cell lineage in

Drosophila melanogaster. Development 120, 2489–2499.Brand, A. H., and Perrimon, N. (1993). Targeted gene expression as ameans of altering cell fates and generating dominant phenotypes. Irish, V. F., and Gelbart, W. M. (1987). The decapentaplegic gene isDevelopment 118, 401–415. required for dorsal–ventral patterning of the Drosophila embryo.

Genes Dev. 1, 868–879.Braun, T., Rudnicki, M. A., Arnold, H. H., and Jaenisch, R. (1992).Targeted inactivation of the muscle regulatory gene Myf-5 results Jiang, J., Kosman, D., Ip, Y. T., and Levine, M. (1991). The morpho-in abnormal rib development and perinatal death. Cell 71, 369– gen gradient regulates the mesoderm determinant twist in early382. Drosophila embryos. Genes Dev. 5, 1881–1891.

Breitbart, R. E., Liang, C.-S., Smoot, L. B., Laheru, D. A., Mahdavi, Kaushal, S., Schneider, J. W., Nadal-Ginard, B., and Mahdav, V.V., and Nadal-Ginard, B. (1993). A fourth human MEF2 transcrip- (1994). Activation of the myogenic lineage by MEF2A, a factortion factor, hMEF2D, is an early marker of the myogenic lineage. that induces and cooperates with MyoD. Science 26, 1236–1240.Development 118, 1095–1106. Leibham, D., Wong, M.-W., Cheng, T.-C., Schroeder, S., Weil, P. A.,

Buchberger, A., Ragge, K., and Arnold, H.-H. (1994). The myogenin Olson, E. N., and Perry, M. (1994). Binding of TFIID and MEF2gene is activated during myocyte differentiation by pre-existing, to the TATA element activates transcription of Xenopus MyoDanot newly synthesized transcription factor MEF-2. J. Biol. Chem. promoter. Mol. Cell. Biol. 14, 686–699.269, 17289–17296. Leiss, D., Hinz, U., Gasch, A., Mertz, R., and Renkawitz-Phol, R.

Chambers, A. E., Logan, M., Kotecha, S., Towers, N., Sparrow, D., (1988). b3 tubulin expression characterizes the differentiatingand Mohun, T. J. (1994). The RSRF/MEF2 protein SL1 regulates mesodermal germ layer during Drosophila embryogenesis. De-cardiac muscle-specific transcription of a myosin light-chain velopment 104, 525–531.gene in Xenopus embryos. Genes Dev. 8, 1324–1334. Leptin, M., Bogaert, T., Lehmann, R., and Wilcox, M. (1989). The

Dohrmann, C., Azpiazu, N., and Frasch, M. (1990). A new Drosoph- function of PS integrins during Drosophila embryogenesis. Cellila homeobox gene is expressed in mesodermal precursor cells 56, 401–408.of distinct muscles during embryogenesis. Genes Dev. 4, 2098– Li, H., and Capetanaki, Y. (1994). An E box in the desmin promoter2111. cooperates with the E box and MEF-2 sites of a distal enhancer

Edmondson, D. G., Cheng, T.-C., Cserjesi, P., Chakraborty, T., and to direct muscle-specific transcription. EMBO J. 13, 3580–3589.Olson, E. N. (1992). Analysis of the myogenin promoter reveals Lilly, B., Galewsky, S., Firulli, A. B., Schulz, R. A., and Olson, E. N.an indirect pathway for positive autoregulation mediated by the (1994). D-MEF2: A MADS box transcription factor expressed inmuscle-specific enhancer factor MEF-2. Mol. Cell. Biol. 12, 3665– differentiating mesoderm and muscle cell lineages during Dro-3677. sophila embryogenesis. Proc. Natl. Acad. Sci. USA 91, 5662–

Edmondson, D. G., Lyons, G. E., Martin, J. F., and Olson, E. N. 5666.(1994). Mef2 gene expression marks the cardiac and skeletal mus- Lilly, B., Zhao, B., Ranganayakulu, G., Paterson, B. M., Schulz,cle lineages during mouse embryogenesis. Development 120, R. A., and Olson, E. N. (1995). Requirement of MADS domain1251–1263. transcription factor D-MEF2 for muscle formation in Drosophila.

Ferguson, E. L., and Anderson, K. V. (1992). Decapentaplegic acts Science 267, 688–693.as a morphogen to organize dorsal–ventral pattern in the Dro- Lin, M.-H., Nguyen, H. T., Dybala, C., and Storti, R. V. (1996). Myo-sophila embryo. Cell 71, 451–461. cyte-specific enhancer factor 2 acts cooperatively with a muscle

Frasch, M. (1995). Induction of visceral and cardiac mesoderm by activator region to regulate Drosophila tropomyosin gene muscleectodermal Dpp in the early Drosophila embryo. Nature 374, expression. Proc. Natl. Acad. Sci. USA 93, 4623–4628.464–467. Liu, M. L., Olson, A. L., Edgington, N. P., Moye-Rowley, W. S., and

Fujita, S., Zipursky, S. L., Benzer, S., Ferrus, A., and Shotwell, S. L. Pessin, J. E. (1994). Myocyte enhancer factor (MEF2) binding site(1982). Monoclonal antibodies against the Drosophila nervous is essential for C2C12 myotube-specific expression of the ratsystem. Proc. Natl. Acad. Sci. USA 79, 7929–7933. GLUT4/muscle-adipose facilitative glucose transporter gene. J.

Biol. Chem. 269, 28514–28521.Gasch, A., Hinz, U., Leiss, D., and Renkawitz-Pohl, R. (1988). The


AID DB 8484 / 6x19$$$205 01-18-97 03:24:43 dbal


Luo, L., Liao, Y. J., Jan, L. Y., and Jan, Y. N. (1994). Distinct morpho- Ranganayakulu, G., Zhao, B., Dokidis, A., Molkentin, J. D., Olson,E. N., and Schulz, R. A. (1995). A series of mutations in the D-genetic functions of similar small GTPases: Drosophila Drac1 is

involved in axonal outgrowth and myoblast fusion. Genes Dev. MEF2 transcription factor reveal multiple functions in larval andadult myogenesis in Drosophila. Dev. Biol. 171, 169–181.8, 1787–1802.

Martin, J. F., Miano, J. M., Hustad, C. M., Copeland, N. G., Jenkins, Ranganayakulu, G., Schulz, R. A., and Olson, E. N. (1996). Winglesssignaling induces nautilus expression in the ventral mesodermN. A., and Olson, E. N. (1994). A Mef2 gene that generates a

muscle-specific isoform via alternative mRNA splicing. Mol. of the Drosophila embryo. Dev. Biol. 176, 143–148.Rijsewijk, F., Schuermann, M., Wagenaar, E., Parren, P., Weigel,Cell. Biol. 14, 1647–1656.

Martin, J. F., Schwarz, J. J., and Olson, E. N. (1993). Myocyte en- D., and Nusse, R. (1987). The Drosophila homolog of the mousemammary oncogene int-1 is identical to the segment polarityhancer factor (MEF) 2C: A tissue-restricted member of the MEF-

2 family of transcription factors. Proc. Natl. Acad. Sci. USA 90, gene wingless. Cell 50, 649–657.Rudnicki, M. A., Braun, T., Hinuma, S., and Jaenisch, R. (1992).5282–5286.

McDermott, J. C., Cardoso, M. C., Yu, Y.-T., Andres, V., Leifer, D., Inactivation of MyoD in mice leads to up-regulation of the myo-genic HLH gene Myf-5 and results in apparently normal muscleKrainc, D., Lipton, S. A., and Nadal-Ginard, B. (1993). hMEF2C

gene encodes skeletal muscle- and brain-specific transcription development. Cell 71, 383–390.Schultz, J. R., Tansey, T., Gremke, L., and Storti, R. V. (1991). Afactors. Mol. Cell. Biol. 13, 2564–2577.

Michelson, A. M., Abmayr, S. M., Bate, M., Arias, A. M., and Ma- muscle specific intron enhancer required for rescue of indirectflight muscle and jump muscle function regulates Drosophilaniatis, T. (1990). Expression of a MyoD family member prefigures

muscle pattern in Drosophila embryos. Genes Dev. 4, 2086– tropomyosin I gene expression. Mol. Cell. Biol. 11, 1901–1911.Shore, P., and Sharrocks, A. D. (1995). The MADX-box family of2097.

Molkentin, J. D., Black, B. L., Martin, J. F., and Olson, E. N. (1995). transcription factors. Eur. J. Biochem. 229, 1–13.Staehling-Hampton, K., Hoffmann, F. M., Baylies, M. K., Rushton,Cooperative activation of muscle gene expression by MEF2 and

myogenic bHLH proteins. Cell 83, 1125–1136. E., and Bate, M. (1994). dpp induces mesodermal gene expressionin Drosophila. Nature 372, 783–786.Morata, G., and Lawrence, P. A. (1977). The development of wing-

less, a homeotic mutation of Drosophila. Dev. Biol. 56, 227–240. St. Johnston, R. D., and Gelbart, W. M. (1987). Decapentaplegictranscripts are localized along the dorsal–ventral axis of the Dro-Morisaki, T., and Holmes, E. W. (1993). Functionally distinct ele-

ments are required for expression of the AMPD1 genes in myo- sophila embryo. EMBO J. 6, 2785–2791.Struhl, G. (1989). Differing strategies for organizing anterior andcytes. Mol. Cell. Biol. 13, 5854–5860.

Nabeshima, Y., Hanaoka, K., Hayasaka, M., Esumi, E., Li, S., No- posterior body pattern in Drosophila embryos. Nature 338, 741–744.naka, I., and Nabeshima, Y. (1993). Myogenin gene disruption

results in perinatal lethality because of severe muscle defect. Tautz, D., and Pfeifle, C. (1989). A non-radioactive in situ hybrid-ization method for the localization of specific RNAs in Drosoph-Nature 364, 532 –535.

Naidu, P. S., Ludolph, D. C., To, R. Q., Hinterberger, T. J., and Ko- ila embryos reveals translational control of the segmentationgene hunchback. Chromosoma 98, 81 –85.nieczny, S. F. (1995). Myogenin and MEF2 function synergisti-

cally to activate the MRF4 promoter during myogenesis. Mol. Taylor, M. V., Beatty, K. E., Hunter, H. K., and Baylies, M. K. (1995).Drosophila MEF2 is regulated by twist and is expressed in both ofCell. Biol. 15, 2707–2718.

Nguyen, H. T., Bodmer, R., Abmayr, S. M., McDermott, J. C., and the primordia and differentiated cells of the embryonic somatic,visceral and heart musculature. Mech. Dev. 50, 29 –41.Spoerel, N. A. (1994). D-mef2: A Drosophila mesoderm-specific

MADS box-containing gene with a biphasic expression profile Tepass, U., Theres, C., and Knust, E. (1990). crumbs encodes anEGF-like protein expressed on apical membranes of Drosophiladuring embryogenesis. Proc. Natl. Acad. Sci. USA 91, 7520–

7524. epithelial cells and required for organization of epithelia. Cell61, 787–799.Olson, E. N., and Klein, W. H. (1994). bHLH factors in muscle de-

velopment: Dead lines and commitments, what to leave in and van den Heuvel, M., Nusse, R., Johnston, P., and Lawerence, P. A.(1989). Distribution of wingless gene product in Drosophila em-what to leave out. Genes Dev. 8, 1–8.

Olson, E. N., Perry, M., and Schulz, R. A. (1995). Regulation of bryos: A protein involved in cell–cell communication. Cell 59,739–749.muscle differentiation by the MEF2 family of MADS box tran-

scription factors. Dev. Biol. 172, 2–14. Wieschaus, E., and Riggleman, R. (1987). Autonomous require-ments for the segment polarity gene armadillo during DrosophilaParmacek, M. S., Ip, H. S., Jung, F., Shen, T., Martin, J. F., Vora,

A. J., Olson, E. N., and Leiden, J. M. (1994). A novel myogenic embryogenesis. Cell 49, 177–184.regulatory circuit controls slow/cardiac troponin C gene tran- Yee, S. P., and Rigby, P. W. (1993). The regulation of myogenin genescription in skeletal muscle. Mol. Cell. Biol. 14, 1870–1885. expression during embryonic development of the mouse. Genes

Paterson, B. M., Walldorf, U., Eldridge, J., Dubendorfer, A., Frasch, Dev. 7, 1277–1289.M., and Gehring, W. (1991). The Drosophila homologue of verte- Yu, Y.-T., Breitbart, R. E., Smoot, L. B., Lee, Y., Mahdavi, V., andbrate myogenic determination genes encodes a transiently ex- Nadal-Ginard, B. (1992). Human myocyte-specific enhancer fac-pressed nuclear protein marking primary myogenic cells. Proc. tor 2 comprises a group of tissue-restricted MADS box transcrip-Natl. Acad. Sci. USA 88, 3782–3786. tion factors. Genes Dev. 6, 1783–1798.

Pollock, R., and Treisman, R. (1991). Human SRF-related proteins:Received for publication October 22, 1996DNA-binding properties and potential regulatory targets. Genes

Dev. 5, 2327–2341. Accepted November 25, 1996


AID DB 8484 / 6x19$$$205 01-18-97 03:24:43 dbal

ectopic expression of mef2 in the epidermis induces ... · pdf fileto understand further the...

Documents