This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Boettner, B.
Right arrow Articles by Van Aelst, L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Boettner, B.
Right arrow Articles by Van Aelst, L.

 Previous Article  |  Next Article 

Molecular and Cellular Biology, November 2007, p. 7966-7980, Vol. 27, No. 22
0270-7306/07/$08.00+0     doi:10.1128/MCB.01275-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

The Rap GTPase Activator Drosophila PDZ-GEF Regulates Cell Shape in Epithelial Migration and Morphogenesis{triangledown}

Benjamin Boettner and Linda Van Aelst*

Cold Spring Harbor Laboratory, Watson School of Biological Sciences, 1 Bungtown Road, Cold Spring Harbor, New York 11724

Received 16 July 2007/ Returned for modification 20 August 2007/ Accepted 30 August 2007


arrow
ABSTRACT
 
Epithelial morphogenesis is characterized by an exquisite control of cell shape and position. Progression through dorsal closure in Drosophila gastrulation depends on the ability of Rap1 GTPase to signal through the adherens junctional multidomain protein Canoe. Here, we provide genetic evidence that epithelial Rap activation and Canoe effector usage are conferred by the Drosophila PDZ-GEF (dPDZ-GEF) exchange factor. We demonstrate that dPDZ-GEF/Rap/Canoe signaling modulates cell shape and apicolateral cell constriction in embryonic and wing disc epithelia. In dPDZ-GEF mutant embryos with strong dorsal closure defects, cells in the lateral ectoderm fail to properly elongate. Postembryonic dPDZ-GEF mutant cells generated in mosaic tissue display a striking extension of lateral cell perimeters in the proximity of junctional complexes, suggesting a loss of normal cell contractility. Furthermore, our data indicate that dPDZ-GEF signaling is linked to myosin II function. Both dPDZ-GEF and cno show strong genetic interactions with the myosin II-encoding gene, and myosin II distribution is severely perturbed in epithelia of both mutants. These findings provide the first insight into the molecular machinery targeted by Rap signaling to modulate epithelial plasticity. We propose that dPDZ-GEF-dependent signaling functions as a rheostat linking Rap activity to the regulation of cell shape in epithelial morphogenesis at different developmental stages.


arrow
INTRODUCTION
 
Epithelial morphogenetic processes constitute genetically programmed cell movement or tissue growth that shapes the developing embryo or organ. Such processes entail alterations in tissue shape that are brought about by a variety of different cellular mechanisms, including cell adhesion, cell contractility, cell intercalation, and polarized cell division (3, 27, 56). Concurrently, the integrity of the epithelium has to be maintained at all times. The dorsal closure (DC) process late in Drosophila gastrulation has proven to be a superb system with which to study the signaling pathways and mechanical alterations involved in cell shape changes. DC is initiated after the germ band has retracted and the transient amnioserosa is dorsally exposed. Following initiation, the bilateral ectoderm sweeps over the amnioserosa in the dorsal direction to finally engage in zippering at the dorsal midline. In this manner the embryo is tightly sealed. This process, both in terms of genetic contributions and its physiological aspects, is highly reminiscent of the closure of an epithelial wound after a mechanical insult (reviewed in reference 39). The sequence of events that comprises DC is accompanied by dramatic cell shape changes in the contributing tissues. Ectodermal cells elongate along the dorsal-ventral axis, and while gradually spreading over the space occupied by the shrinking amnioserosa, they morph substantially from a thick- to a thin-layered epithelium (29, 39). In contrast, the differentiation of imaginal disc epithelia is driven largely by the orchestration of the growth and orientation of cell division. Newly born cells have to adopt and maintain a well-defined position in the epithelium and adjust their shape to the requirements of their environment. The shape of individual cells critically depends on the polarity, adhesion, and tension parameters that govern their neighbors (8, 16, 23).

Cell shape changes in DC and other epithelial migration events are driven largely by nonmuscle myosin, also referred to as myosin II (MyoII), which generates mechanical force at the leading edge, the lateral ectoderm, and the amnioserosa by modulating contractile events proximal to cell-cell adhesion complexes (13). In proliferating and differentiating epithelia, MyoII provides general cell tension and is involved in the establishment of compartment boundaries (38). MyoII assembly and activity are tightly controlled by signal transduction pathways, among which the Rho GTPase/Rho-kinase pathway is the best studied (11, 43, 55). However, recent studies have documented the existence of Rho-independent pathways that regulate MyoII assembly and contractility. For instance, in the slime mold Dictyostelium, multiple MyoII heavy-chain kinases have been identified that affect MyoII assembly but function without a regulatory input from the Rho pathway (49, 62). In Xenopus gastrulation, a pathway involving the Shroom protein and the small GTPase Rap1 likely engages MyoII in the closure of the neural tube (17, 20, 21). The precise operational mechanisms of these pathways largely remain to be deciphered, but they underscore that epithelial cell shape can be regulated by both Rho-dependent and Rho-independent means.

The Rap GTPases have emerged as key regulators not only of integrin-mediated adhesion to extracellular substrates but also of intercellular adhesion and cell motility (reviewed in references 5, 6, and 32). In Drosophila, Rap1 mutations are associated with failures in several morphogenetic processes. Rap1 mutant embryos are defective in ventral and dorsal closure, and mesodermal precursors as well as primordial germ cells are impaired in their stereotypical migration patterns (2, 4). Thus, Rap1 is vital for various morphogenetic processes that require cell migration and associated cell shape changes. We previously demonstrated that Rap1 relies to a large extent on its effector protein AF-6/Canoe (Cno) to mediate its action in the DC process (4). Here, we report that the exchange factor Drosophila PDZ-GEF (dPDZ-GEF) serves as a Rap1 activator to confer cell shape changes and to modulate cell plasticity in epithelial cells. In analyzing DC in the embryo and wing disc morphogenesis at later developmental stages, we find that dPDZ-GEF links via Rap1 to the effector protein Cno to regulate MyoII function and to generate and adjust lateral cell contractility. Thus, our data reveal a novel Rap1-dependent pathway that is critical for the maintenance and modulation of cell shape and plasticity in epithelial tissues.


arrow
MATERIALS AND METHODS
 
Drosophila strains and genetics. Canton S flies were used for germ line transformation and control experiments. dPDZ-GEF alleles were the following: Gef261, also called l(2)k13720, dizzyP, and pdz-gef1 (35); Gef262, also called EP(2)0388, dizzyEP, and pdz-gef2 (35) (Szeged Stock Center); Gef263 (57); and Gef26C1 and Gef26C2, which were generated in this study. For the isolation of the latter mutant alleles, the gmrGAL4 driver line was crossed to EP(2)0388 males that had been mutagenized with 25 mM ethyl methanesulfonate. G1 males subsequently were scored for a loss of a gmrGAL4>EP(2)0388-specific rough-eye phenotype and were mated to Gef261/CyO females for complementation testing. Positive alleles then were subjected to a second complementation cross with females carrying the Df(2)BSC5 deficiency, which uncovers the region 26B1-2;26D1-2 (FlyBase). Other alleles used in this study were Df(2L)BSC5 (Bloomington 6299); cno2 and cnomis1; Rap1B3 and Rap1CD3 (both provided by I. Harihanan); zip1 and zip2 (Bloomington 4199 and 8739); and zipebr (obtained from D. Kiehart).

P-element reversion was performed with Sp/CyO; Sb {Delta}2-3/TM6 flies as a transposase source. The following driver lines were used: gmrGAL4 (Bloomington 321), engrailed-GAL4 (enGAL4) (a gift from H. D. Ryoo), and dPDZ-GEFGAL4 (this study; see below). Transgenes used include UAS-2xEGFP (Bloomington 6658); UAS-DRap1V12, UAS-DRap1WT, UAS-DRap1N17, and UAS-cnoWT (4); and UAS-dPDZ-GEF2, UAS-myrdPDZ-GEF2, UAS-dPDZ-GEFEGFP, UAS-dPDZ-GEF-IR, and the genomic dPDZ-GEFEGFP rescue construct (see below). Transgenic lines were generated by P-element-mediated transformation (48). Embryos in genetic interaction and rescue experiments were genotyped using green fluorescent protein (GFP)-expressing balancers.

Standard recombination techniques were used to generate the FRT40A, Gef261 chromosome, and genetic mosaics were generated by FLP/FRT-mediated recombination. Clones in adult wings were induced with the Minute technique and were marked with forked using w f36A hsFLP1.22; FRT40A M(2)Z f+ (14). Heat shocks were applied for 1 h at 37°C 4 days after egg deposition (AED). For wing disc analysis, y w hsFLP1.22; FRT40A, armLacZ, or y w hsFLP1.22; FRT40A M(2)24 armLacZ/CyO females were crossed to y w; FRT40A Gef261/CyO males, and 24-h egg collections were heat shocked for 30 min 3 or 4 days AED. For the generation of germ line clones (GLCs) with the dominant female sterile/FLP technique (10), y w hsFLP1.22; FRT40, Gef263/CyO females were mated with w; FRT40 ovoD1 males. Recombination was induced by successive heat shocks at days 4, 5, and 6 AED. Female progeny were mated to Df(2L)BSC5/CyO-GFP males, and GFP-negative embryos were selected for cuticle analysis. Fly culture, crosses, and recombination techniques were performed at 25°C according to standard methods.

YTH screen. Yeast two-hybrid (YTH) screening and assaying were performed as previously described (4). A Drosophila embryonic cDNA library in pGAD (4) was screened in the L40 yeast reporter strain (54) using wild-type Rap1 cloned in pBTM116 as bait.

Molecular biology. A partial cDNA clone starting at nucleotide 667 of the dPDZ-GEF coding sequence was isolated in a YTH screen using Rap1 as bait. The sequence encoded an alternative exon 6, contributing the alternative C-terminal peptide sequence GPPNFMDSTKCTICPMPPMMQ. For this variation, we termed the cDNA dPDZ-GEF2. The full-length cDNA of dPDZ-GEF2 was produced by PCR from an embryonic cDNA library using 5' AAAGGAAAAAAGCGGCCGCCACCATGGATCCGTATCACCATATCAG and 3' CTAGTCTAGATTATTGATTCATGGGTGGCATGGG primers and was subcloned in pUAST to generate pUAST-WTdPDZ-GEF2. pUAST-myrdPDZ-GEF2 was constructed by inserting the sequence encoding the first 69 amino acids of Dsrc41 upstream of the ATG start codon of dPDZ-GEF2. The genomic rescue construct dPDZ-GEFEGFP was obtained from BAC R05J04 (RPCI-98 05.J.4; Children's Hospital Oakland Research Institute) by PCR amplification and insertion into pCaSpeR4. A novel NotI site was introduced upstream of the native ATG start codon, which subsequently served to introduce an EGFP moiety in frame with the 5'-protein-coding region of the ATG start codon. For dPDZ-GEFGAL4, a dPDZ-GEF promoter fragment starting 2,975 bp upstream of the ATG start codon was inserted into pCaSpeR4 and C-terminally fused to the GAL4/VP16 transcriptional activator sequence (kindly gifted by J. Rodriguez). pUAST-dPDZ-GEFEGFP was made by transferring the genomic sequence downstream of the ATG start codon from dPDZ-GEFEGFP to pUAST and subsequent addition of an enhanced GFP (EGFP) tag upstream of the ATG. The YTH constructs pLexA-Rap1V12, pLexA-Rap1N17, pLexA-Rap1WT, and pGAD-cnoN were previously described (4). pLexA-DRas1V12, pLexA-Ras1N17, and pLex-Ras1WT were generated by insertion of the respective cDNAs into pBTM116 (54). For RNA interference (RNAi) purposes, the dPDZ-GEF coding sequence from bp 277 to 1031 was amplified with AvrII- or NheI-linked complementary primer sequence 5' AGTCTTACCGTGCGGGATGCC and XbaI-linked primer sequence 5' TCCTGCGTCTTGTGCGCATCC and inserted stepwise as inverted repeats in a tail-to-tail orientation into pWIZ (36).

Immunohistochemistry. Embryos were fixed with heat/methanol (53) or 8% paraformaldehyde (25), stained with the indicated antibodies, and mounted in Vectashield medium (Vector Laboratories). Third-instar larval imaginal discs were fixed in 4% paraformaldehyde-phosphate-buffered saline. Antibodies used were the following: rat anti-Cno (1:50) (4); mouse anti-Arm (N2 7A1; 1:25) (Developmental Studies Hybridoma Bank [DSHB]); mouse anti-Dlg (4F3; 1:50) (DSHB); mouse anti-Fas3 (7G10; 1:20) (DSHB); rabbit anti-Zip/MyoII (1:800) (a gift of R. Karess); mouse anti-ß-galactosidase (1:2,000) (Sigma); rabbit anti-ß-galactosidase (1:5,000) (Cappell); rabbit anti-GFP (1:500) (MBL). Alexa 594-conjugated phalloidin (Molecular Probes) was used at 2 U ml–1. Alexa fluorophore-coupled secondary antibodies were from Molecular Probes and were used at 1:250. Confocal images were captured on an LSM 510 Zeiss laser-scanning microscope.

For DC survival analysis, embryos not carrying GFP-expressing balancers were selected, transferred to fresh grape plates, and aged for 36 h at 25°C. Cuticle preparations were as described in Takahashi et al. (50), and differential interference contrast images of cuticles were collected with an Axioplan 2 Zeiss microscope. Adult wings were dehydrated in isopropanol, mounted in Canada Balsam (Sigma), and digitally photographed with bright-field illumination using a Nikon SMZ microscope. For quantification, tricomb regions from five independent wings were photographed at x40 magnification with bright-field illumination. Files were imported into the MatLab 7.1 program (Mathworks), and tricomb numbers per fixed area were counted and compiled.


arrow
RESULTS
 
dPDZ-GEF is expressed in the gastrulating embryonic ectoderm and colocalizes with Cno. In a previous study, we demonstrated an essential role for the Rap1/Cno GTPase/effector complex in the process of epithelial migration driving DC (4). It remained, however, unknown how Rap1 becomes activated in DC and which cellular processes Rap1/Cno signaling impinges upon. To further elucidate the signaling networks to which Rap1 is integral, we performed a YTH screen using Rap1 as bait. This resulted in the isolation of a cDNA encoding a splice variant of the Rap exchange factor dPDZ-GEF or dizzy, which we termed dPDZ-GEF2. Computational sequence analysis revealed six coding exons, of which the last exon deviates from the previously described dPDZ-GEF/dizzy sequences (Fig. 1A; also see Materials and Methods) (22, 57). In common with the previously described protein sequences, dPDZ-GEF2 harbors a cyclic nucleotide monophosphate-binding domain, a GefN domain, a PDZ domain, and a Ras-associating (RA) domain, as well as a consensus catalytic guanine nucleotide exchange factor (GEF) domain (Fig. 1B), all of which are encoded by the third coding exon. We observed that dPDZ-GEF2 binds to the wild-type form of Rap1 as well as to both a constitutively active (CA) and a dominant-negative (DN) mutant form of Rap1 (Fig. 1C). The interaction with CA Rap1 is mediated by the RA domain, whereas binding to DN Rap1 occurs through the GEF domain, which carries out the nucleotide exchange reaction on Rap1 (data not shown). No interaction was detected with analogous mutants of the Drosophila Ras1 protein (Fig. 1C). The domain composition and Rap-binding and activation specificity are conserved among the mammalian and Caenorhabditis elegans homologues of the protein (12, 46). Of note, the mammalian homologue was found to reside in cell-cell junctional complexes of epithelial cells (26, 40). So far, dPDZ-GEF isoforms have been implicated in macrophage and male stem cell biology as well as in eye differentiation (22, 35, 57); however, a potential role in epithelial migration and homeostasis has not been reported.


Figure 1
View larger version (73K):
[in this window]
[in a new window]

  		FIG. 1. dPDZ-GEF gene and protein structure and expression in the embryonic ectoderm. (A) Genomic organization of the dPDZ-GEF locus. dPDZ-GEF2 contains an alternatively spliced last exon. Coding sequences are in green, and untranslated regions (UTR) are in red. Depicted underneath are the genomic regions covered by the dPDZ-GEFEGFP, UAS-dPDZ-GEFEGFP, and dPDZ-GEFGAL4 transgenes. (B) The dPDZ-GEF2 protein sequence encompasses 1,569 amino acids (aa) and encodes cNMP (cyclic nucleotide monophosphate; yellow)-binding, GefN and GEF (blue), PDZ (Psd95/Dlg/ZO-1; red), and RA (green) motifs. (C) YTH interaction between dPDZ-GEF2 and Rap1 and Ras1 mutants. The L40 yeast strain was transformed with plasmids expressing dPDZ-GEF2 or an N-terminal fragment of Cno (CnoN) fused to the GAL4 activation domain (GAD) and the indicated Rap1 and Ras1 mutants, or lamin control, fused to LexA and tested for ß-galactosidase activity. (D to D''') Visualization of dPDZ-GEFEGFP in a Gef261 mutant background during early (D) and late (D') DC stages and zippering at the dorsal midline (D'''). Scale bar, 50 µm. (E to E''') Colocalization (E''') of dPDZ-GEF (E') with Arm (E) during ectodermal elongation. The dorsal side of the embryo is up. (F to F''') Colocalization (F''') of dPDZ-GEF (F') with Fas3 (F) in a more basal section of the ectoderm. (G to G''') dPDZ-GEF (G') codistributes with Cno (G) in the lateral ectoderm and the amnioserosa at the onset of DC. An overlay is shown in panel G'''. (H to H''') Dorsal view of an embryo late in DC in which dPDZ-GEF (H') apically colocalizes with Cno (H). An overlay is shown in panel H'''. (I to I''') More basal sections in the confocal stack still exhibit significant dPDZ-GEF levels (I') at the membrane, while Cno (I) is absent. Sections in panels G, H, and I are 0.7 µm thick, and the sections in panels H and I are 1.4 µm apart. Scale bars for panels E to I''', 10 µm.

To explore whether dPDZ-GEF plays a role in epithelial processes, we began by investigating its embryonic localization pattern. We developed a fully functional fluorescent genomic transgene, which we termed dPDZ-GEFEGFP. Importantly, dPDZ-GEFEGFP reverted all of the defects associated with a dPDZ-GEF loss-of-function (LOF) allele to wild type (Fig. 1A; also see below). When monitoring dPDZ-GEFEGFP expression in a dPDZ-GEF mutant background during embryogenesis, we found that the fluorescent protein is strongly expressed in the ectoderm during cellularization and subsequently in the cellular blastoderm (data not shown). The strong presence of dPDZ-GEFEGFP protein prior to the onset of zygotic transcription indicates that a large maternal component contributes to gene expression in the early embryo and may partially persist into later stages. At subsequent stages, we detected high protein levels in primordial germ cells, in the ectoderm during ventral invagination, in internalized cells that will differentiate into mesodermal and neuroectodermal structures, and in the ectodermal tissue performing the germ band elongation and retraction movements (data not shown). Of particular interest, at DC stages, during initiation, sweeping, and zippering, all epithelial tissues involved in the closure process displayed large amounts of dPDZ-GEFEGFP protein (Fig. 1D to D'').

Both Rap1 and Cno, the putative downstream effectors of dPDZ-GEF, are expressed at near-ubiquitous levels throughout gastrulation. Both proteins overlap in their subcellular localization patterns at the adherens junction (AJ) of cells that undergo DC (4). To trace the subcellular localization of dPDZ-GEF, we again employed dPDZ-GEFEGFP in a mutant background. In embryos at DC stages, we observed dPDZ-GEF protein to be present at AJs of leading-edge and lateral ectodermal cells, where it colocalizes with the ß-catenin homologue Armadillo (Arm) (Fig. 1E). Moreover, we found that dPDZ-GEF codistributes with Cno at the AJ at early (stage 13) and late (stage 14) DC stages during ectodermal sweeping and suturing (Fig. 1G to I). Notably, the dPDZ-GEF protein also is detectable in more basal sections of the membrane, as demonstrated by costaining with anti-fasciclin 3 (anti-Fas3) antibodies, which decorate the lateral membrane (Fig. 1F). Overall, the distribution of dPDZ-GEF is highly reminiscent of Rap1's localization in the embryonic ectoderm (4). Thus, dPDZ-GEF, Rap1, and Cno colocalize at the AJ in the epithelial cells involved in DC, supporting the notion that they participate in a common signaling pathway.

dPDZ-GEF signals through Rap1 and Cno during DC. Our binding and protein expression data are suggestive of a scenario in which the dPDZ-GEF gene product functions as a relevant Rap1 activator during DC. We next sought genetic evidence of a functional pathway that links dPDZ-GEF with Rap1 signaling during DC. We made use of a P element (Gef261) that is inserted in the first untranslated exon of the dPDZ-GEF gene. Gef261, also termed l(2)k13270, represents a recessive lethal allele, and it has been associated with undetectable dPDZ-GEF transcript levels (22, 35, 57). To determine a role for dPDZ-GEF in DC, we examined the Gef261 allele for phenotypic effects in the embryonic ectoderm. Thorough analysis of a Gef261 mutant population (n = 277) revealed a significant percentage of embryos that display late gastrulation defects, ranging from a complete failure in DC (31%) to milder closure phenotypes exemplified by smaller anterior holes (29%) or the inability to invaginate the head cuticle with the adhering head skeleton (18 out of 23% of embryos with a closed cuticle) (Fig. 2C and D; Table 1). A small portion of anterior-open and -closed cuticles (5% of the total population) displayed a tail-up phenotype (Fig. 2C), indicative of a not-entirely-completed germ band retraction at the onset of DC. Seventeen percent of Gef261 mutants completed embryogenesis and survived into larval stages. We found this phenotypic distribution to be closely mirrored by the Gef261 allele in conjunction with the Df(2L)BSC5 deficiency that covers the locus (Fig. 2B and Table 1). Of note, the observed phenotypic variability also has been found in mutant situations for other genes impacting DC (4, 33). One likely reason for this variability is the presence of a maternal component that varies in strength. Our analyses of dPDZ-GEF GLCs (see Materials and Methods) that lack maternal and zygotic expression support this notion. We found that the majority of embryos derived from these GLCs exhibit a dorsal-open phenotype (Fig. 2E and Table 1). To ascertain that the observed defects are due to the Gef261 insertion and a specific obliteration of dPDZ-GEF gene function, we performed two additional experiments. First, we precisely excised the transposon, which resulted in a complete reversion of the DC phenotypes described above, and conferred viability to the chromosome (Table 1). Second, when a genomic rescue construct encompassing the complete dPDZ-GEF gene, dPDZ-GEFEGFP, was introduced into the Gef261 background, the ectodermal defects were abolished (Table 1). Together, these data indicate that dPDZ-GEF plays an important role in the DC process.


Figure 2
View larger version (77K):
[in this window]
[in a new window]

  		FIG. 2. dPDZ-GEF functions upstream of Rap1 and cno. (A) Representative wild-type cuticle; (B) transheterozygous Gef261/Df(2l)BSC5 cuticle; (C and D) Gef261 cuticles that are (C) anterior open (the white arrowhead indicates the anterior hole) and (D) severely dorsal open; and (E) Gef263 germ line mutant embryo carrying a paternal Df(2L)BSC5 deficiency. (F to I) Representative cuticles from genetic interaction experiments between dPDZ-GEF and Rap1 or cno. Double transheterozygotes of the genotypes Gef26C2/Gef261; Rap1CD3/Rap1B3 (F); Gef26C1/Gef261; Rap1CD3/Rap1B3 (G); Gef26C2/Gef261; cnomis1/cno2 (H); and Gef26C1/Gef261; cnomis1/cno2 (I) show late gastrulation defects. (J to L) Representative cuticle defects investigated in epistasis tests for dPDZ-GEF, Rap1, and cno. (J) Dorsal-open cno2 mutant embryo. (K) Dorsal-open dPDZ-GEFGAL4>UAS-Rap1N17 embryo. (L) Anterior-open dPDZ-GEFGAL4>UAS-Rap1V12 embryo. (M and N) Ectodermal dPDZ-GEFGAL4>UAS-EGFP reporter activities in early (M) and late (N) DC embryos.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Quantitative analysis of cuticular phenotypes of genetic interaction and rescue experiments

In subsequent experiments, we examined whether the dPDZ-GEF gene is indeed linked to Rap1/Cno signaling in DC. Our previous study, in which we removed zygotic Rap1 from a heteroallelic cno background, demonstrated a strong genetic interaction between Rap1 and cno (4). Both genes provided sufficient dose sensitivity to shift the phenotypic spectrum to stronger DC defects. To assess a genetic interaction between dPDZ-GEF and Rap1 and between cno and dPDZ-GEF, we generated two new hypomorphic alleles (Gef26C1 and Gef26C2, respectively) of the dPDZ-GEF gene (see Materials and Methods). Both hypomorphs, when homozygous, survived into late larval stages, and embryogenesis was completed without noticeable defects. When combined with the strong Gef261 allele, neither of the two hypomorphic mutations exhibited gastrulation defects, and only a small percentage of the embryos displayed embryonic lethality: 9 (n = 278) and 13% (n = 264) for Gef26C1 and Gef26C2, respectively (see Table 1). Due to a strong maternal contribution, Rap1 zygotic mutants are not compromised during embryogenesis and survive into larval stages (Table 1) (2, 4). However, eliminating zygotic Rap1 expression by combining the recessive null alleles Rap1CD3 and Rap1B3 in the Gef26C1/Gef261 background significantly exacerbated the severity of defects. Eighteen percent of Gef26C1/Gef261; Rap1CD3/Rap1B3 embryos (n = 212) showed severe dorsal-open phenotypes (Table 1). Sixty-seven percent of the mutant embryos are mildly closure defective, with smaller anterior holes (Fig. 2G), and 14% have a closed cuticle. Within the latter group, 4% of the embryos were unable to involute their head skeletons. The high survival rates of Gef26C1/Gef261 and Rap1CD3/Rap1B3 embryos (91 and 100%, respectively) were reduced to 1% in the double transheterozygotic embryos. Similar results were obtained with Gef26C2 (Fig. 2F and Table 1). We next analyzed double transheterozygous combinations of dPDZ-GEF and cno alleles and found an even more severe distribution of late gastrulation phenotypes (Fig. 2H and I and Table 1). Sixty-nine percent and 75% of double mutant embryos (Gef26C1/Gef261; cnomis1/cno2 and Gef26C2/Gef261; cnomis1/cno2, respectively) remained open, with strong ectodermal migration defects, and about a quarter of the mutant embryos exhibited weaker anterior-open defects (Table 1). Thus, allelic combinations between dPDZ-Gef and Rap1 or cno almost entirely blocked progression into larval stages. These data indicate a remarkable degree of cooperation between all three genes in the final stages of embryonic gastrulation and support the notion that dPDZ-Gef acts in a common pathway with Rap1 and cno.

To explore the epistatic relationships of dPDZ-GEF to Rap1 and cno, we made use of the GAL4/UAS system (9). In order to perform a dPDZ-GEF tissue-specific rescue analysis, we developed a driver line by fusing a GAL4VP16 transcriptional activator sequence to a functional dPDZ-GEF promoter fragment (henceforth referred to as dPDZ-GEFGAL4; see Materials and Methods) and generated transgenic lines. We validated the functionality of dPDZ-GEFGAL4. First, GAL4 activity was strong in the lateral ectoderm throughout DC, as determined by EGFP expression (Fig. 2M and N). Second, dPDZ-GEFGAL4 did not noticeably alter the phenotypic spectrum of Gef261 mutant embryos (Table 1). Third, driving expression of a dPDZ-GEF2 cDNA with dPDZ-GEFGAL4 significantly alleviated the ectodermal phenotypes of dPDZ-GEF mutant animals (Table 1). Lastly, forced expression of DN Rap1N17 and CA Rap1V12 cDNAs with dPDZ-GEFGAL4 gave rise to a consistent inability to proceed through DC (Fig. 2K and L; Table 1). The fact that both DN and CA versions of Rap1 are able to halt the DC process may reflect the requirement for active cycling between the active and inactive GTPase states. Such oscillation would allow for a more flexible modulation of effector pathway activities during the different stages of ectodermal migration. Interferences by both types of dominant mutants of small GTPases have been observed in other migratory processes as well (37, 44). In sum, these observations prove the gene-specific dPDZ-GEFGAL4 driver to be an efficient tool for the analysis of the epistatic relationships between our genes of interest. We then examined whether ectopic expression of Rap1 and Cno can alleviate the dPDZ-GEF LOF phenotype. Upon overexpression of Rap1, 80% of Gef261 mutant embryos have a closed cuticle, with 44% proceeding into the first-instar larval stage (n = 194), as opposed to only 32% closed cuticles and 14% larval survival in the control Gef261 mutant population (n = 335) analyzed in parallel (Table 1). Ectopic expression of Cno in dPDZ-GEF LOF mutants yielded 74% of embryos with closed cuticles and 31% surviving into larval stages (n = 291). In both experiments, the number of open cuticles was significantly reduced from 68% in dPDZ-GEF LOF embryos to 20 and 26% in Rap1- and Cno-overexpressing mutant embryos, respectively. Thus, both Rap1 and cno transgenes were able to rescue the dPDZ-GEF LOF phenotype to a large extent. In the converse experiments, we found that ectopic expression of wild-type dPDZ-GEF2 or a myristoylated form of dPDZ-GEF2 had no effect on the mutant phenotype of Rap1N17 or cno (Fig. 2J and K; Table 1), indicating that dPDZ-GEF acts upstream of Rap1 and cno. Together with data from our previous study demonstrating that cno is epistatic to Rap1 (4), these data establish a link between the dPDZ-GEF, Rap1, and cno genes, with dPDZ-GEF acting as the most upstream element in this signaling module that promotes DC.

dPDZ-GEF promotes ectodermal cell shape changes during DC. In light of the ectodermal defects associated with the dPDZ-GEF LOF phenotypes at the cuticle level, we looked for cellular abnormalities in the lateral ectoderm and leading edge of mutant embryos undergoing DC. We began by analyzing the shape of cells in the lateral ectoderm of dPDZ-GEF mutant embryos by staining for Fas3, a lateral membrane marker. In severely compromised embryos, the lateral ectodermal cell sheets of dPDZ-GEF mutants displayed an inability to migrate towards the dorsal midline. This was due to a failure of ectodermal cells to elongate dorsally (compare Fig. 3A to A '). In addition, about 20% of dorsal-open animals displayed a local bunching of leading-edge cells (Fig. 3A'). We observed that local segments of the leading-edge collapse into foci and that cells in adjacent areas are abnormally extended in the anterior-posterior (A/P) direction toward the bunching regions (Fig. 3A'). Immunostaining for Arm and Cno as an AJ marker revealed similar ectodermal stretching deficiencies and local bunching defects at the leading edge (Fig. 3B' and C'), suggesting that the entire basolateral membrane compartment facing the amnioserosa is involved in this phenomenon. Notably, this phenotype also is present in embryos overexpressing a DN mutant version of Rap1 and in embryos deficient for cno (4), and thus it likely represents a common regulatory defect in mutants of all three genes. Since Cno and Arm as AJ components were properly localized in dPDZ-GEF mutant embryos that showed impaired stretching towards the dorsal midline (Fig. 3B' and C'), we infer that AJ integrity is unperturbed in a dPDZ-GEF LOF situation. This is in agreement with earlier findings with Rap1 germ line mutant embryos, in which Cno and Arm localization also remained unaffected (4).


Figure 3
View larger version (43K):
[in this window]
[in a new window]

  		FIG. 3. dPDZ-GEF mutant ectoderm displays intact AJs but is characterized by cell shape abnormalities. (A and A') Wild-type (A) and dPDZ-GEF mutant (A') embryos immunostained for Fas3 as a lateral membrane marker. (B and B') Wild-type (B) and dPDZ-GEF mutant (B') embryos immunostained for Arm. (C and C') Wild-type (C) and dPDZ-GEF mutant (C') embryos immunostained for Cno. Note that the mutant ectoderm in panels A', B', and C' fails to elongate and that the leading-edge organization is dyscoordinated and locally contracted (see the arrowheads in panels A', B', and C'). Scale bar, 10 µm. AS, amnioserosa; LE, leading edge; E, ectoderm.

Together, these findings indicate that dPDZ-GEF function is critical for bringing about the epithelial cell shape changes in the lateral ectoderm that are required to organize a coherently moving leading edge, and that the underlying mechanism does not seem to involve regulation of the levels or distribution of AJ components. Moreover, they underscore that dPDZ-GEF, Rap1, and Cno compose a module that regulates the proper dorsoventral elongation of ectodermal cells involved in DC.

Tissue integrity of the larval epithelium depends on dPDZ-GEF function. Our finding that dPDZ-GEF-deficient embryos display cell shape abnormalities raises the questions of whether these also are manifest at later stages in development and whether they can be further defined. We noted that rare homozygous dPDZ-GEF mutant adult escapers exhibit a marked wing phenotype, in which the wing blades are not extended in a straight fashion but bend downwards, suggesting that cell shape abnormalities might interfere with proper wing morphogenesis. Therefore, we decided to address dPDZ-GEF mutant situations in the wing disc epithelium at late larval stages. Analysis of the wing epithelium also provides the advantage of being able to differentiate between cell-autonomous and nonautonomous effects. We first examined dPDZ-GEF expression throughout disc morphogenesis. Visualizing dPDZ-GEFGAL4 expression with a UAS-EGFP reporter revealed high levels of expression in the entire cephalic complex, which represents the developing epithelium in larval life (data not shown). Analysis of the wing imaginal disc showed that dPDZ-GEFEGFP is present in all developing wing compartments, including the pouch, margin, and hinge regions (Fig. 4A). Furthermore, confocal cross-sectioning showed a marked dPDZ-GEF accumulation at the lateral membrane that strongly resembles the distribution of Rap1 (Fig. 4B) (31) and a colocalization with Arm at the apical border of the lateral membrane (Fig. 4C to C''). The conservation of dPDZ-GEF's subcellular localization between the embryonic ectoderm and the wing disc epithelium prompted us to perform a mosaic analysis employing the FLP/FRT system, which enables the generation of mitotic mutant clones in an otherwise wild-type epithelium (60). The clones we obtained were, however, relatively small, which impeded a detailed analysis of larger contiguous areas of mutant cells. To generate larger clones, we exploited the Minute technique (42). Minute mutations impose a growth disadvantage on wild-type cells, which often allows for clonal expansion and survival of mitotic clones into adult stages. This technique previously has been employed to define the roles of adhesion molecules and cytoskeletal regulators in wing disc morphogenesis (15, 58). We analyzed the distribution of cell-cell junctional markers and the shape of cells in clonal and surrounding wild-type tissues. Consistent with our findings for the embryo, we found that the AJ components, Arm and DE-cadherin, were evenly distributed around the circumference of dPDZ-GEF mutant cells (Fig. 4D to D''' and data not shown), indicating that AJ integrity is not perturbed by the loss of dPDZ-GEF function. However, we noticed a frequent widening of mutant cell perimeters in zones in which mutant cells contacted surrounding wild-type cells (Fig. 4D'''). In contrast, wild-type cells that border mutant clones often displayed diminished circumferences (Fig. 4D'''). When examining the more basally located septate junction (SJ), using Discs large (Dlg) as a marker, we also found that the Dlg protein is present evenly around the entire circumference of mutant cells. However, here as well we observed that mutant cell perimeters were widened to an even larger extent than that at the AJ of dPDZ-GEF cells bordering on wild-type tissue (Fig. 4E' and E'''). Cell circumferences were vastly extended not only at the borders of mutant clones but also in interiorly residing mutant cells (Fig. 4E'''). Wild-type cells close to the border of mutant clones, in contrast, were even more shrunken than those at the more apical AJ (Fig. 4E'''). Thus, a diminished circumference in wild-type cells contrasts with a circumferential expansion in mutant cells, which suggests a loss of contractile force around the apicolateral membrane domain in dPDZ-GEF mutant cells. Importantly, these effects are not due to the Minute background, since wild-type control clones within the same Minute background did not exhibit cell shape abnormalities. Neither Arm nor Dlg staining revealed any alterations in cell constriction at the clonal boundaries of wild-type control clones (Fig. 4F to G'). These findings suggest that dPDZ-GEF regulates lateral contractility and that its function is important for proper wing morphogenesis.


Figure 4
View larger version (72K):
[in this window]
[in a new window]

  		FIG. 4. Loss of dPDZ-GEF is associated with a reduction of apicolateral cell contractility. (A) x-y section through a third-instar larval Gef261; dPDZ-GEFEGFP wing disc. (B) x-z section through a Gef261; dPDZ-GEFEGFP wing disc. dPDZ-GEF is ubiquitously expressed throughout the epithelium. (C to C'') x-z sections though a Gef261; dPDZ-GEFEGFP wing disc immunostained for Arm (C'). dPDZ-GEF (C) localizes to the lateral membrane and displays overlap with Arm staining at the AJ (C"). (D to D''') Gef261 Minute+ mutant clones (marked by a lack of ß-galactosidase expression, which labels wild-type tissue in green) immunostained for Arm (red in panels D and D" or white in panels D' and D'''). (D" and D''') High magnification of the region of Gef261 mosaic tissue. Mutant cells bordering on wild-type areas (encircled with a dashed yellow line) often are stretched (white arrowheads in panel D'''), whereas more distal wild-type cells are slightly contracted (yellow arrowhead). (E to E''') Gef261 Minute+ mutant clones lacking ß-galactosidase expression (green) and immunostained with anti-Dlg as an SJ marker (red in panels E and E" or white in panels E' and E'''). (E and E') Clonal areas are less tightly packed (white arrowheads) than adjacent wild-type areas (yellow arrowheads). (E" and E''') High magnification of Gef261 mosaic tissue stained for Dlg. Clonal areas (white arrowheads) display markedly widened cell perimeters compared to those of wild-type cells (yellow arrowhead). (F and F') High magnification of a Minute control mosaic region labeled with an anti-Arm antibody (red in panel F and white in panel F') and ß-galactosidase expression (green) present in Minute mutant cells. (G and G') High magnification of the Minute control region immunostained with anti-Dlg (red in panel G and white in panel G'). Yellow lines in panels F' and G' demarcate clone boundaries. (H to M) Representative pictures of wild-type and Gef261 Minute+ wings. (H) Wild-type control wing. (I) The Gef261 Minute+ wing is characterized by a rounded geometry. (J to M) Magnified view of the anterior margin in wild-type (J and L) and Gef261 Minute+ (K and M) wings. Images in panels L and M represent enlarged views of the boxed areas in panels J and K, respectively. Margin bristles in mutant areas display wider, uneven spacing or are not formed. (N and O) Tricomb fields in wild-type (N) and Gef261 Minute+ mutant backgrounds. Mutant tricomb patterns exhibit widened spacing and lack the parallel organization visible in a wild-type wing. Scale bars in panels A, B, D, and E are 50 µm; those in panels C, D", E", F, and G are 10 µm.

Due to the growth disadvantage of Minute tissue, features of mutant clones originating in the larval wing disc epithelium often survive through pupal development into the adult (42). Hence, we next surveyed adult wings for the phenotypic expression of dPDZ-GEF mutant clones. We found that wings harboring multiple dPDZ-GEF mutant clones are more rounded and broadened than wild-type wings or wings harboring wild-type control clones (Fig. 4H and I). Each cell in the adult wing epithelium has an apically protruding hair-like structure, which in cells at the margin is a large bristle and in interiorly located cells is a smaller-sized tricomb. In wild-type wing blades, bristles and tricombs are regularly organized and spaced, the former lining the wing margin and the latter running in parallel rows through intervein regions (Fig. 4J, L, and N). dPDZ-GEF mutant clones display vastly disorganized bristle and tricomb patterns, with an uneven and markedly increased spacing between them (Fig. 4K, M, and O). Since the spacing between bristles and tricombs is a direct reflection of the circumferential sizes of the cells that produce them, this phenotype is consistent with the extension of lateral cell perimeters at earlier stages during the morphogenesis of dPDZ-GEF mutant clones in imaginal discs and suggests difficulties in maintaining proper tissue architecture.

We reasoned that if the loss of dPDZ-GEF function affects cell shape and constriction, then excess amounts of dPDZ-GEF should impact the same parameters. To address this, we ectopically expressed a UAS-dPDZ-GEFEGFP transgene (Fig. 1A) in a defined wing territory. In particular, enGAL4 was used as a driver, which results in high levels of dPDZ-GEF expression in the posterior compartment of the wing (Fig. 5D and E). Interestingly, when staining wing imaginal discs with an antibody against Arm, we observed that dPDZ-GEF-overexpressing posterior cells that directly face the A/P compartment boundary are aberrantly stretched in the A/P direction (Fig. 5E and F). Importantly, overexpression of EGFP in control discs did not yield any appreciable stretching effects in posterior cells lining the A/P boundary (compare Fig. 5C to F). The dPDZ-GEF overexpression effect also is represented in emerging adult enGAL4>UAS-dPDZ-GEFEGFP wings (Fig. 6 F). Adult wings posteriorly overexpressing dPDZ-GEF assumed a more rounded geometry, with the anterior cells pulling at the posterior margin (Fig. 6F). The longitudinal veins L2 and L3 were shifted towards the anterior direction, describing a curve instead of the relatively straight line typically observed in a wild-type wing. Furthermore, posterior tricombs close to the A/P compartment boundary displayed increased distances between each other (Fig. 5G and H). Conversely, tricombs at the opposite side of the compartment boundary were positioned much closer to each other than in a wild-type control wing (Fig. 5G and H; for quantifications, see Fig. 5I). Of note, these phenotypes differ from the ones observed for flies overexpressing growth regulators (for example, Rheb) in posterior cells. Growth-regulatory genes primarily alter the size of targeted wing territories without drastically affecting cell shape in adjoining territories (45). Interestingly, our data reveal that both the loss and gain of dPDZ-GEF function produce similar phenotypes, supporting the notion that tightly controlled cycling between active and inactive forms of Rap GTPase is important to establish tensional strength around lateral cellular circumferences.


Figure 5
View larger version (131K):
[in this window]
[in a new window]

  		FIG. 5. Overexpression of dPDZ-GEF triggers cell shape abnormalities. (A to C) Wing pouches from third-instar larval wing discs overexpressing EGFP with enGAL4. (A) enGAL4>UAS-EGFP (green) discs were immunostained for Arm (red). (B) Arm immunostaining shown alone. (C) Enlarged view of the boxed area in panel B. Cells normally are organized along the A/P boundary (between the yellow arrowheads). (D to F) enGAL4>UAS-dPDZGefEGFP disc stained for Arm (red in panel D, white in panel E). (F) Enlarged view of the boxed area in panel E. Ectopic dPDZGefEGFP (green) provokes a loss of normal cell contractility along the A/P boundary (between the yellow arrows) within the posterior compartment. Anterior is left, and posterior is right. The scale bars in panels A, B, D, and E are 50 µm, and those of panels C and F are 10 µm. (G and H) Hair densities in enGAL4>UAS-EGFP (G) and enGAL4>UAS-dPDZGefEGFP (H) adult wings. Anterior is up, and posterior is down. Posterior cells of enGAL4>UAS-dPDZGefEGFP wings in the proximity of the A/P boundary (between yellow arrowheads) are dilated (white arrowhead in panel H) compared to those in an equivalent area of a wild-type wing (white arrowhead in panel G). As a consequence, wild-type cells facing the boundary anteriorly display a more contracted surface (black arrowheads in panels G and H). (I) Quantification of tricomb densities in wings overexpressing dPDZ-GEFEGFP or EGFP with enGAL4. Tricomb densities were determined in the intervein region between the L3 and L4 veins anterior (A) and posterior (P) of the A/P boundary. Overexpression of dPDZ-GEFEGFP with enGAL4 decreases tricomb densities posteriorly and increases densities anteriorly; P < 10–4 for comparisons of enGAL4>UAS-dPDZ-GEFEGFP anterior to enGAL4>UAS-EGFP anterior, enGAL4>UAS-dPDZ-GEFEGFP posterior to enGAL4>UAS-EGFP posterior, and enGAL4>UAS-dPDZ-GEFEGFP anterior to enGAL4>UAS-dPDZ-GEFEGFP posterior, whereas P = 0.44 for the comparison of enGAL4>UAS-EGFP anterior to enGAL4>UAS-EGFP posterior (Student's t tests).


Figure 6
View larger version (111K):
[in this window]
[in a new window]

  		FIG. 6. dPDZ-GEF gain of function and LOF alter wing geometry and are modified by Rap and cno. (A) ptcGAL4>UAS-EGFP expression (green) along the A/P compartment boundary in a wing disc stained with an anti-Arm antibody (red). The scale bar is 50 µm. (B and B') Expression of UAS-dPDZ-GEF-IR with ptcGAL4 in a wing disc immunostained for Arm (red in panel B) eliminates the dPDZ-GEFEGFP signal (green in panels B and B'). (C to K) Effects of modulated dPDZ-GEF levels in the posterior compartment of the wing. (C) Wild-type control wing. A and P indicate the anterior and posterior sides of the wing, respectively. (D and E) enGAL4>UAS-dPDZ-GEF-IR-expressing wings that are curled downwards with the margin folding back (D) or dorsally ruptured due to enhanced tension (E). (F) enGAL4>UAS-dPDZ-GEFEGFP-expressing wing that shows a rounded geometry, particularly at the anterior margin and the longitudinal L2 and L3 veins. The posterior cross vein is forked at its anterior attachment site. (G) enGAL4>UAS-dPDZGEF-IR/UAS-dPDZ-GEFEGFP wing. RNAi and overexpression effects ameliorate each other. (H) enGAL4>UAS-Rap1WT wing with venation defects but intact margins and overall geometry. (I) enGAL4>UAS-dPDZ-GEF-IR/UAS-Rap1WT wing. Ectopic levels of Rap largely rescue the dPDZ-GEF RNAi-specific tension defects. (J) enGAL4>UAS-dPDZ-GEFEGFP/cno2 wing. Reduction of the cno gene dosage partially suppresses the dPDZ-GEF overexpression phenotypes. (K) Heterozygous cno2 wing with a wild-type morphology. (L to N) Posterior margins of a control enGAL4>UAS-EGFP wing (L), a defective enGAL4>UAS-dPDZ-GEF-IR wing (M), and an enGAL4>UAS-dPDZ-GEF-IR/UAS-Rap1WT wing in which the RNAi-induced defects are rescued (N).

To assess whether dPDZ-GEF regulates tissue organization through the Rap/Cno complex, we established a dPDZ-GEF-directed RNAi transgene, UAS-dPDZ-GEF-IR (see Materials and Methods). Before examining the effect of the RNAi transgene on wing architecture, we confirmed that it (i) was able to significantly decrease the fluorescence signal produced by dPDZ-GEFEGFP (Fig. 6B and 6B') and (ii) could counteract the overexpression phenotype (compare Fig. 6F to G). Thus, dPDZ-GEF-IR was effective in knocking down dPDZ-GEF expression levels. We noted that expressing dPDZ-GEF-IR posteriorly using enGAL4 forced the wing to round up and curl downwards (Fig. 6D), most likely as a result of a loss of lateral contractility. Two to 3 days after eclosion, wings frequently (8/32 wings) ruptured at the posterior margin (Fig. 6E). This differs somewhat from the aberrations observed in wings overexpressing enGAL4>UAS-dPDZ-GEFEGFP, in which rupture never occurred (Fig. 6F). Bristles of enGAL4>UAS-dPDZ-GEF-IR wings were wider and unevenly spaced at the posterior wing margin (Fig. 6M), and in rare cases the anterior cross vein (9/32 wings) and posterior cross vein (1/32 wings) failed to form. To determine whether dPDZ-GEF functions in a Rap1- and Cno-dependent manner in wing development, we conducted two types of experiments. First, we tested whether concomitant overexpression of wild-type Rap1 can rescue the enGAL4>UAS-dPDZ-GEF-IR-associated posterior wing shape defect. While ectopic levels of Rap1 interfered with the correct establishment of posterior cross veins and occasionally interfered with that of longitudinal veins, it had no influence on the overall wing geometry, and the wing blades were extended in a straight orientation (Fig. 6H). We found, however, that Rap1 overexpression could rescue dPDZ-GEF-IR-induced wing bending substantially. Downward curling of the wing blades and the irregular spacing of margin bristles triggered by dPDZ-GEF-IR expression were largely suppressed, suggesting that these phenotypes are caused by insufficient Rap activation (Fig. 6I and N). Overexpression of Cno with enGAL4 had lethal consequences, and no viable adults emerged. Hence, we performed an alternative interaction study in which we reduced the endogenous cno gene dose in an enGAL4>UAS-dPDZ-GEFEGFP overexpression background. Removing one copy of cno, which by itself has no effect on wing morphogenesis (Fig. 6K), weakened the effects of dPDZ-GEF overexpression (compare Fig. 6F to J). Posterior cell stretching along the A/P boundary was ameliorated, and the positions of longitudinal veins L2 and L3 shifted toward a wild-type pattern (Fig. 6J). These modifications indicate that Rap and Cno participate in the regulation of apicolateral cell constriction downstream of dPDZ-GEF. Together, our data imply that a dPDZ-GEF/Rap/Cno pathway is reiteratively used in epithelial morphogenesis to adjust apicolateral cell circumferences and cell shape in morphogenetic processes.

MyoII regulation by dPDZ-GEF signaling. Our observation that the levels and distribution of junctional components appear to be unperturbed in dPDZ-GEF LOF situations during DC and wing disc morphogenesis raises the question of which other cellular elements might be dysregulated and causative for the observed cell shape phenotypes. Increasing evidence supports the idea that contractile events in the epithelium help drive morphogenesis and that myosin proteins, particularly the nonmuscle myosin/MyoII, generate force to modulate cell shape in morphogenetic movements and in developing epithelia (13, 23). The Drosophila MyoII heavy chain is encoded by the zipper gene (zip) (28). It forms a dimer and, together with two essential and two regulatory light chains, constitutes a functional hexamer. In light of our observation that lateral contractility is weakened in dPDZ-GEF LOF clones, we decided to assess whether this effect is dependent on MyoII. We began by investigating the localization of MyoII in dPDZ-GEF LOF clones in wing imaginal discs by immunostaining with a MyoII-specific antibody. Strikingly, we observed that mutant cells display an abnormal, less compact organization of the MyoII network, whereas wild-type tissues surrounding mutant clones frequently exhibit foci with excessively high MyoII accumulation (Fig. 7A to B '). These differences in MyoII distribution are paralleled by the loss and gain of apicolateral constriction that we previously described (Fig. 4D''' and E'''). Thus, a failure of dPDZ-GEF mutant cells to sufficiently assemble MyoII in order to establish homeostatic contraction levels provides a likely explanation for the expansion of mutant cell perimeters and constriction in the surrounding wild-type tissue.


Figure 7
View larger version (121K):
[in this window]
[in a new window]

  		FIG. 7. dPDZ-GEF signaling is linked to MyoII function. (A and A') Wing pouch displaying Gef261 Minute+ mutant clones (marked by a lack of ß-galactosidase expression, which labels wild-type tissue in green) and immunostained with anti-MyoII (red in panel A, white in panel A'). The scale bar is 50 µm. (B and B') High magnification of a clonal area. MyoII (red in panel B, white in panel B') is less compact in mutant tissue, whereas it is overassembled in juxtaposing wild-type tissue (green in panel B, encircled with a dashed yellow line in panel B'). The scale bar is 10 µm. (C and C') Wild-type (C) embryo immunostained for MyoII (red) and Fas3 (green). (C') MyoII staining in white indicates the purse string formed at the leading edge (LE). (D and D') dPDZ-GEF mutant embryo immunostained for MyoII (red) and Fas3 (green in panel B, encircled with a dashed yellow line in panel B') (D) with aberrant A/P stretching of LE cells and markedly reduced accumulation of MyoII (D') at the LE. (E and E') dPDZ-GEF mutant embryo immunostained for MyoII (red) and Fas3 (green) (E). MyoII (E') does not form the wild-type bars-on-a-string pattern of MyoII at the LE and displays a distinct membrane-associated localization. (F and F') cno mutant embryo immunostained for MyoII (red) and Fas3 (green) (F). Ectodermal cells that initiated the elongation process still show MyoII staining (F') at the membrane but do not assemble a continuous MyoII cable at the LE. Scale bars in panels C to F are 20 µm. (G to I) Genetic interaction between dPDZ-GEF and cno with zip during DC. (G) Wild-type cuticle. (H) Gef26C2, zipebr/Gef261, and zip2 anterior-open defects. (I) zipebr/zip2; cnomis1/cno2 dorsal-open phenotype. (J to L) Genetic interaction between a hypomorphic allelic cno combination and zipebr in the adult wing: (J) zipebr wing, (K) cnomis1/cno2 wing, and (L) a strongly malformed zipebr; cnomis1/cno2 wing.

We then asked whether a dysregulation of MyoII also could be one of the reasons why dPDZ-GEF pathway deficiencies display DC defects. MyoII in wild-type embryos appears in a typical bars-on-a-string-type distribution on the leading edge (Fig. 7C and C'). Immunostaining of dPDZ-GEF mutant embryos, however, revealed a striking failure to assemble MyoII at the leading edge (Fig. 7D and D'). This is particularly apparent in ectodermal segments for which the leading-edge cells are stretched along the dorsoventral axis (Fig. 7E and E'). Analysis of cno mutant embryos showed a similar mislocalization of MyoII in leading-edge cells. cno mutant embryos that failed to proceed through DC often show MyoII localizing discontinuously in their leading edges (Fig. 7F and F'). These data are in agreement with earlier studies showing that compromising MyoII function prevents the proper generation of contractile forces in the leading edge and DC (13).

To obtain genetic evidence for a role of dPDZ-GEF signaling in MyoII regulation, we asked whether a concomitant reduction of gene dosages at the zip and dPDZ-GEF or cno loci would reveal genetic interactions. To this end, we made use of a transheteroallelic zip combination (zipebr/zip2) that does not provoke significant embryonic lethality (Table 2) (38). Reduced dosages of the zip and dPDZ-GEF (Gef26C2/Gef261) genes together yielded a significant number of dead embryos (52%; n = 248). Six percent of the double transheterozygotic animals remained dorsal open and 25% retained anterior-open defects, while the remaining 21% proceeded through DC but did not involute their head cuticles (Fig. 7H and Table 2). Viability into the first-instar larval stage was reduced to 48%, whereas it was 93 and 87% for zipebr/zip2 and Gef26C2/Gef261 embryos, respectively (Table 2). Thus, these findings indicate that the dPDZ-GEF and zip genes cooperate in DC. We next lowered gene dosages concomitantly at the zip (zipebr/zip2) and cno (cnomis1/cno2) loci and found an even more profound effect. Embryonic lethality increased to 95%. Among them, 29% of embryos remained dorsal open and 57% remained anterior open, while 9% of embryos completed DC but did not hatch into larvae (n = 93) (Fig. 6I and Table 2). To analyze the interaction we observed between the cno and zip genes during DC later in wing morphogenesis, we modified the zip gene dose by introducing zipebr into the viable hypomorphic cnomis/cno2 background. Flies carrying these genetic modifications developed malformed wings at a significant frequency (7/25 wings), while heterozygous zipebr and hypomorphic cno wings were not affected (Fig. 7J to L). Together, these data demonstrate a genetic interaction of the zip gene with dPDZ-GEF and cno and corroborate our immunocytochemical observations. This finding, in conjunction with the decline in lateral constriction we observed in mitotic clones and mutant embryos, compels us to propose a model in which dPDZ-GEF signaling via Rap GTPases regulates MyoII to modulate cell contractility in epithelial migration and homeostasis.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Quantitative analysis of cuticular phenotypes for genetic interactions of zip and dPDZ-GEF or cno


arrow
DISCUSSION
 
In developing tissues, Rap has been found to promote various morphogenetic processes, ranging from epithelial migration and invagination in embryogenesis to the maintenance of epithelial integrity in proliferating tissues at later stages (2, 4, 17, 31, 46). However, the mechanisms by which Rap is regulated and mediates its effects in morphogenetic episodes remain poorly understood. In this report, we delineate a pathway in which the Drosophila GEF dPDZ-GEF links Rap activity to MyoII and the regulation of lateral contractility and cell shape in different epithelial morphogenetic episodes.

We identified dPDZ-GEF as a putative activator of Rap GTPases in a YTH screen and subsequently demonstrated that it specifically associates with Rap, but not Ras, GTPases. PDZ-GEF is highly conserved among metazoans (12, 26, 40, 46), suggesting that it might serve common physiological roles. We found that dPDZ-GEF is highly expressed in epithelial tissues involved in embryonic DC, and, importantly, our data revealed that it functions as an activator of Rap1/Cno signaling in this process. First, as in the case of Rap1 and cno, loss of zygotic dPDZ-GEF function is associated with an ectodermal failure, which is manifested by dorsal-open phenotypes. Eliminating both zygotic and maternal dPDZ-GEF elevates the frequency of late gastrulation defects. Second, our genetic analysis places dPDZ-GEF upstream of the Rap/Cno GTPase/effector complex, as both Rap1 and Cno were able to rescue the dPDZ-GEF LOF phenotype to a large extent. Third, all three proteins show an overlapping localization at AJs in ectodermal cells involved in DC. Thus, our findings demonstrate that dPDZ-GEF serves as a Rap1/Cno activator to promote late epithelial gastrulation movements. In support of a conserved role of dPDZ-GEF in epithelial morphogenesis, studies with C. elegans demonstrated that pxf-1, the dPDZ-GEF homolog, is vital for epithelial integrity (46). pxf-1 mutant animals often are confronted with hypodermal malfunctions; the underlying cellular basis of these defects, however, remains to be elucidated.

Epithelial migration processes often entail striking alterations in cell shape, and much effort has been devoted to unraveling the underlying cellular and molecular mechanisms. Our study highlights that dPDZ-GEF as a Rap activator adjusts cell shape to the demands of morphogenetic movements and imaginal disc morphogenesis. We observed that dPDZ-GEF mutant embryos involved in DC often exhibit bunched regions in their leading edge and an incompetence of ectodermal cells to elongate dorsally. These phenotypes also characterize embryos that either overexpress DN Rap1 or are mutant for cno (4). Thus, signaling through dPDZ-GEF, Rap, and Cno (i) is vital for the organization of a coherently moving leading edge and (ii) enables the typical dorsoventral stretching of lateral ectodermal cell sheets. Our studies also unveiled a requirement for dPDZ-GEF for the adjustment of epithelial cell shape in the differentiation program of the wing imaginal disc. We found that a dPDZ-GEF LOF situation generated in a clonal analysis of mosaic wing discs is associated with a decline in apicolateral contractility in the vicinity of junctional complexes. Loss of contractility, as visualized by a widening of apicolateral circumferences, is coupled to a partially compensating gain of contractility in adjacent wild-type tissue. Wild-type cells in close proximity to mutant clones display smaller apicolateral circumferences. Interestingly, overexpression of dPDZ-GEF in restricted areas of the wing disc causes contractile aberrations. When ectopically expressed in the posterior compartment, dPDZ-GEF leads to a loss of apicolateral contractility in cells lining the A/P boundary. Thus, both gain and loss of dPDZ-GEF function compromise normal contractile strength and result in aberrant adult tissue formation. These observations suggest that a finely tuned level of Rap activation is crucial for normal cellular and organismal development to occur. Tight requirements for activation of small GTPases in vivo have been documented previously, e.g., for Rho GTPases and their function in axon guidance (37). Importantly, we found in genetic modification experiments that reduced or enhanced dPDZ-GEF activity in the developing wing can be rescued by ectopic Rap1 or lowered cno doses, respectively, suggesting that signaling through the dPDZ-GEF/Rap/Cno module at least partially controls disc morphogenesis. This, together with the vital cooperative roles of all three genes in embryonic cell sheet migration, corroborates the reiterative function of dPDZ-GEF/Rap/Cno signaling during epithelial development.

What are the mechanisms that translate Rap signaling downstream of dPDZ-GEF into the modulation of cell shape? Our analysis of dPDZ-GEF LOF situations during gastrulation and wing disc morphogenesis showed that junctional integrity is not corrupted. Both AJ and SJ belts around the apicolateral circumference are seamlessly maintained in dPDZ-GEF LOF tissue. However, our data support a role for the MyoII heavy chain, the product of the zip gene, as an effector. MyoII assembly and disassembly in migrating cells and tissue homeostasis are tightly balanced processes. In epithelial cells, MyoII localizes to cell-cell junctional complexes and is essential for establishing and maintaining intercellular adhesion and tension (16). We found that the decline in apicolateral constriction associated with dPDZ-GEF LOF in mitotic clones in the wing disc epithelium is accompanied by a less compact MyoII localization and that adjacent constricted wild-type cells display overassembled MyoII, which concentrates in ectopic focal structures. Also, in the DC paradigm, dPDZ-GEF and cno mutant embryos that are involved in DC exhibited failures of leading-edge cells to properly assemble MyoII. The abundant MyoII localization at the leading edge that characterizes wild-type embryos during DC is significantly diminished, and the bars-on-a-string-like MyoII distribution is lost in these mutants. In particular, regions of the leading edge adjacent to the bunched segments retain only minimal amounts of assembled MyoII. These observations strongly suggest that loss of MyoII control at the leading edge is contributing to the bunching phenotype observed in dPDZ-GEF and cno mutants. Consistent with a spatiotemporal regulation of MyoII in distinct regions of the leading edge are elegant life-imaging studies undertaken with embryos undergoing DC (13). The latter study visualized dynamic cycles of MyoII-dependent contraction and relaxation that are limited to smaller regions within the leading edge during the migration process.

In further support of the notion that dPDZ-GEF signaling acts on MyoII, we obtained evidence that dPDZ-GEF and cno genetically interact with zip in late gastrulation and, moreover, that cno is genetically linked to zip in wing morphogenesis. In particular, our data show a strong enhancement of dorsal-open frequencies in embryos that are double transheterozygous for hypomorphic combinations of zip and either dPDZ-GEF or cno. We also found that combined mutations at the zip and cno loci cause a malformation of wings. Together, our findings imply that signaling through the dPDZ-GEF/Rap/Cno module is required for MyoII function at different stages of epithelial development. Future experimentation will be required to determine the precise biochemical link between this module and MyoII regulation. Of note, a recent study demonstrated that the mammalian Cno homolog, AF-6/Afadin, in a two-dimensional tissue culture system moves together with MyoII at the edge of wounds induced by laser ablation (51). At the onset of wound closure, a subpopulation of MyoII resides apically in the lateral membranes of cells lining the wound. However, when closure progresses into advanced stages, MyoII, together with AF-6/Afadin, migrates basalward to constrict both the wound perimeter and the apicobasal membranes facing the wound.

Our data do not rule out additional Rap effector pathways acting in synergy with Cno to bring about appropriate lateral cell contractility and MyoII regulation. The most extensively investigated signaling pathway that regulates MyoII activity in diverse systems is the Rho/Rho kinase pathway. Several studies have demonstrated that the zip gene exhibits strong interactions with genes encoding Rho signaling components (18, 43, 55, 59). Abundant evidence from mammalian cells indicates that Rho-stimulated Rho kinase activity leads to the phosphorylation of the regulatory myosin light chain (MLC) as well as an inhibition of MLC phosphatase, both of which in concert are stimulating MyoII assembly and contractility (1, 30). In light of recently uncovered molecular links between Rap and Rho involving Rap-associating RhoGAPs, it is an intriguing possibility that Rap is connected to Rho signaling in certain aspects of epithelial morphogenesis (34, 61). Alternatively, however, Rap could signal directly to an enzyme modifying the MyoII complex, such as kinases and phosphatases targeting MLC. Interestingly, the heavy chains of MyoII also have been reported to be subject to phosphorylation (62). Studies with chemotactic Dictyostelium have identified MyoII heavy-chain kinases that regulate the subcellular distribution of assembled MyoII complexes (7, 62). More recently, the Ser/Thr kinase Phg2, which acts as a potential Rap1 effector in Dictyostelium, also has been reported to induce MyoII heavy-chain phosphorylation (24). However, we failed to identify an obvious Drosophila homolog of Phg2 by extensively searching numerous databases. Future studies will be important to delineate how far Rap and Rho signaling activities are intertwined in epithelial morphogenetic events.

Our data support a model in which dPDZ-GEF, through Rap activation and MyoII regulation, contributes to the adjustment of lateral cell contractility in epithelial cells of the embryo and the developing wing. In a previous study, the analysis of Rap1 mutant clones in the wing imaginal disc revealed a direct effect of Rap1 on the reorganization of AJs at the end of cytokinesis, where resealing of their belts has to occur between daughter cells (31). Since our data showed that AJ integrity is unperturbed in clones comprised of dPDZ-GEF LOF cells, we surmise that dPDZ-GEF either is not relevant for Rap1 activation in the reconstitution of a seamless AJ belt during cytokinesis or is compensated for by a still-unknown factor conferring the necessary exchange activity. On the other hand, the apicolateral constriction defects we detected as a consequence of clonal loss of dPDZ-GEF function so far have not been described for Rap1 mutant clones in the same scenario. We presume either that they have escaped scrutiny or, more likely, that Rap1 acts redundantly with its close homolog Rap2l in adjusting apicolateral constriction, while the reorganization of AJs during cytokinesis relies solely on Rap1. In fact, Rap1 and Rap2l have been shown to compensate for each other in the male stem cell niche (57). In this context, both Rap proteins cooperate downstream of dPDZ-GEF to anchor germ line stem cells to their niche. In future experiments, we plan to generate Rap1 and Rap2l mitotic clones in parallel and to examine and compare their effects on cell shape and contractility.

A picture is emerging in which specialized GEFs activate Rap GTPases and selective effectors in different morphogenetic scenarios and cellular processes. For example, Rap1 signaling has been implicated in cell/extracellular matrix-dependent force transduction at focal adhesion sites of cultured cells (52). In this scenario, Rap1 is regulated by an Src/p130Cas/C3G-triggered mechanism (47). Also, apical constriction during neural tube closure in the Xenopus blastula has been demonstrated to depend on Rap1 function downstream of the Shroom protein (17); however, the relevant GEF in this scenario remains to be identified. The notion that Rap activation in distinct developmental processes is specified by dedicated GEFs also suggests that Rap effectors are selected in order to fulfill pathway requirements. In light of this, dPDZ-GEF and Rap1 have been implicated in the regulation of mitogen-activated protein kinase activity during differentiation of the Drosophila compound eye (19, 35), and a more recent study reported that D-Raf relays a signal from Rap to mitogen-activated protein kinase in Torso-receptor-dependent terminal differentiation of the early Drosophila embryo (41). Together, these findings suggest the possibility that dPDZ-GEF could trigger the activation of the Rap/D-Raf pathway to regulate certain differentiation processes. Our data reveal a novel function for dPDZ-GEF as an activator of Rap in the implementation of epithelial cell shape changes required for sheet migration and homeostatic cell shape maintenance in the genesis of the wing imaginal disc epithelium. We further provide evidence that Cno functions as a relevant effector of Rap downstream of dPDZ-GEF in these events and that the dPDZ-GEF/Rap/Cno module is connected to the regulation of MyoII and the generation and modulation of appropriate lateral cell contractility. Thus, our findings have unveiled a pathway linking the Rap activator dPDZ-GEF to MyoII and the regulation of lateral contractility and cell shape in epithelium migration and homeostasis. Further elucidation of dPDZ-GEF-interacting proteins and the molecular underpinnings of MyoII regulation downstream of this module in epithelial cells will be key to our understanding of these aspects of tissue morphogenesis.


arrow
ACKNOWLEDGMENTS
 
We thank the Bloomington and Szeged stock centers for providing Drosophila stocks and the DSHB for antibodies. We are indebted to D. Kiehart, R. Karess, I. Harihanan, A. Garcia-Bellido, T. Xing, H. D. Ryoo, and J. Rodriguez for kindly sharing reagents and M. Betson for critical reading of the manuscript.

This work was supported by NIH grant R01-CA096882 to L.V.A., and B.B. is a recipient of an NIH postdoctoral training grant (S T32 CA009176).


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Cold Spring Harbor Laboratory, 1 Bungtown Road, Cold Spring Harbor, NY 11724. Phone: (516) 367-6829. Fax: (516) 367-8815. E-mail: vanaelst{at}cshl.edu Back

{triangledown} Published ahead of print on 10 September 2007. Back


arrow
REFERENCES
 
    1
  1. Amano, M., M. Ito, K. Kimura, Y. Fukata, K. Chihara, T. Nakano, Y. Matsuura, and K. Kaibuchi. 1996. Phosphorylation and activation of myosin by Rho-associated kinase (Rho-kinase). J. Biol. Chem. 271:20246-20249.[Abstract/Free Full Text]
    2. 2
  2. Asha, H., N. D. de Ruiter, M. G. Wang, and I. K. Hariharan. 1999. The Rap1 GTPase functions as a regulator of morphogenesis in vivo. EMBO J. 18:605-615.[CrossRef][Medline]
    3. 3
  3. Baena-López, L. A., A. Baonza, and A. Garcia-Bellido. 2005. The orientation of cell divisions determines the shape of Drosophila organs. Curr. Biol. 15:1640-1644.[CrossRef][Medline]
    4. 4
  4. Boettner, B., P. Harjes, S. Ishimaru, M. Heke, H. Q. Fan, Y. Qin, L. Van Aelst, and U. Gaul. 2003. The AF-6 homolog canoe acts as a Rap1 effector during dorsal closure of the Drosophila embryo. Genetics 165:159-169.[Abstract/Free Full Text]
    5. 5
  5. Bos, J. L. 2005. Linking Rap to cell adhesion. Curr. Opin. Cell Biol. 17:123-128.[CrossRef][Medline]
    6. 6
  6. Bos, J. L., K. de Bruyn, J. Enserink, B. Kuiperij, S. Rangarajan, H. Rehmann, J. Riedl, J. de Rooij, F. van Mansfeld, and F. Zwartkruis. 2003. The role of Rap1 in integrin-mediated cell adhesion. Biochem. Soc. Trans. 31:83-86.[Medline]
    7. 7
  7. Bosgraaf, L., and P. J. van Haastert. 2006. The regulation of myosin II in Dictyostelium. Eur. J. Cell Biol. 85:969-979.[CrossRef][Medline]
    8. 8
  8. Braga, V. M., and A. S. Yap. 2005. The challenges of abundance: epithelial junctions and small GTPase signalling. Curr. Opin. Cell Biol. 17:466-474.[CrossRef][Medline]
    9. 9
  9. 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]
    10. 10
  10. Chou, T. B., and N. Perrimon. 1996. The autosomal FLP-DFS technique for generating germline mosaics in Drosophila melanogaster. Genetics 144:1673-1679.[Abstract]
    11. 11
  11. Dawes-Hoang, R. E., K. M. Parmar, A. E. Christiansen, C. B. Phelps, A. H. Brand, and E. F. Wieschaus. 2005. Folded gastrulation, cell shape change and the control of myosin localization. Development 132:4165-4178.[Abstract/Free Full Text]
    12. 12
  12. de Rooij, J., N. M. Boenink, M. van Triest, R. H. Cool, A. Wittinghofer, and J. L. Bos. 1999. PDZ-GEF1, a guanine nucleotide exchange factor specific for Rap1 and Rap2. J. Biol. Chem. 274:38125-38130.[Abstract/Free Full Text]
    13. 13
  13. Franke, J. D., R. A. Montague, and D. P. Kiehart. 2005. Nonmuscle myosin II generates forces that transmit tension and drive contraction in multiple tissues during dorsal closure. Curr. Biol. 15:2208-2221.[CrossRef][Medline]
    14. 14
  14. Garcia-Bellido, A., and E. B. Lewis. 1976. Autonomous cellular differentiation of homoeotic bithorax mutants of Drosophila melanogaster. Dev. Biol. 48:400-410.[CrossRef][Medline]
    15. 15
  15. Garcia-Bellido, A., P. Ripoll, and G. Morata. 1976. Developmental compartmentalization in the dorsal mesothoracic disc of Drosophila. Dev. Biol. 48:132-147.[CrossRef][Medline]
    16. 16
  16. Gibson, M. C., and N. Perrimon. 2003. Apicobasal polarization: epithelial form and function. Curr. Opin. Cell Biol. 15:747-752.[CrossRef][Medline]
    17. 17
  17. Haigo, S. L., J. D. Hildebrand, R. M. Harland, and J. B. Wallingford. 2003. Shroom induces apical constriction and is required for hingepoint formation during neural tube closure. Curr. Biol. 13:2125-2137.[CrossRef][Medline]
    18. 18
  18. Halsell, S. R., B. I. Chu, and D. P. Kiehart. 2000. Genetic analysis demonstrates a direct link between rho signaling and nonmuscle myosin function during Drosophila morphogenesis. Genetics 155:1253-1265.[Abstract/Free Full Text]
    19. 19
  19. Hariharan, I. K., R. W. Carthew, and G. M. Rubin. 1991. The Drosophila roughened mutation: activation of a rap homolog disrupts eye development and interferes with cell determination. Cell 67:717-722.[CrossRef][Medline]
    20. 20
  20. Hildebrand, J. D. 2005. Shroom regulates epithelial cell shape via the apical positioning of an actomyosin network. J. Cell Sci. 118:5191-5203.[Abstract/Free Full Text]
    21. 21
  21. Hildebrand, J. D., and P. Soriano. 1999. Shroom, a PDZ domain-containing actin-binding protein, is required for neural tube morphogenesis in mice. Cell 99:485-497.[CrossRef][Medline]
    22. 22
  22. Huelsmann, S., C. Hepper, D. Marchese, C. Knoll, and R. Reuter. 2006. The PDZ-GEF dizzy regulates cell shape of migrating macrophages via Rap1 and integrins in the Drosophila embryo. Development 133:2915-2924.[Abstract/Free Full Text]
    23. 23
  23. Jamora, C., and E. Fuchs. 2002. Intercellular adhesion, signalling and the cytoskeleton. Nat. Cell Biol. 4:E101-E108.[CrossRef][Medline]
    24. 24
  24. Jeon, T. J., D. J. Lee, S. Merlot, G. Weeks, and R. A. Firtel. 2007. Rap1 controls cell adhesion and cell motility through the regulation of myosin II. J. Cell Biol. 176:1021-1033.[Abstract/Free Full Text]
    25. 25
  25. Kaltschmidt, J. A., N. Lawrence, V. Morel, T. Balayo, B. G. Fernandez, A. Pelissier, A. Jacinto, and A. Martinez Arias. 2002. Planar polarity and actin dynamics in the epidermis of Drosophila. Nat. Cell Biol. 4:937-944.[CrossRef][Medline]
    26. 26
  26. Kawajiri, A., N. Itoh, M. Fukata, M. Nakagawa, M. Yamaga, A. Iwamatsu, and K. Kaibuchi. 2000. Identification of a novel beta-catenin-interacting protein. Biochem. Biophys. Res. Commun. 273:712-717.[CrossRef][Medline]
    27. 27
  27. Keller, R., L. A. Davidson, and D. R. Shook. 2003. How we are shaped: the biomechanics of gastrulation. Differentiation 71:171-205.[CrossRef][Medline]
    28. 28
  28. Kiehart, D. P. 1990. Molecular genetic dissection of myosin heavy chain function. Cell 60:347-350.[CrossRef][Medline]
    29. 29
  29. Kiehart, D. P., C. G. Galbraith, K. A. Edwards, W. L. Rickoll, and R. A. Montague. 2000. Multiple forces contribute to cell sheet morphogenesis for dorsal closure in Drosophila. J. Cell Biol. 149:471-490.[Abstract/Free Full Text]
    30. 30
  30. Kimura, K., M. Ito, M. Amano, K. Chihara, Y. Fukata, M. Nakafuku, B. Yamamori, J. Feng, T. Nakano, K. Okawa, A. Iwamatsu, and K. Kaibuchi. 1996. Regulation of myosin phosphatase by Rho and Rho-associated kinase (Rho-kinase). Science 273:245-248.[Abstract]
    31. 31
  31. Knox, A. L., and N. H. Brown. 2002. Rap1 GTPase regulation of adherens junction positioning and cell adhesion. Science 295:1285-1288.[Abstract/Free Full Text]
    32. 32
  32. Kooistra, M. R., N. Dube, and J. L. Bos. 2007. Rap1: a key regulator in cell-cell junction formation. J. Cell Sci. 120:17-22.[Abstract/Free Full Text]
    33. 33
  33. Köppen, M., B. G. Fernandez, L. Carvalho, A. Jacinto, and C. P. Heisenberg. 2006. Coordinated cell-shape changes control epithelial movement in zebrafish and Drosophila. Development 133:2671-2681.[Abstract/Free Full Text]
    34. 34
  34. Krugmann, S., R. Williams, L. Stephens, and P. T. Hawkins. 2004. ARAP3 is a PI3K- and rap-regulated GAP for RhoA. Curr. Biol. 14:1380-1384.[CrossRef][Medline]
    35. 35
  35. Lee, J. H., K. S. Cho, J. Lee, D. Kim, S. B. Lee, J. Yoo, G. H. Cha, and J. Chung. 2002. Drosophila PDZ-GEF, a guanine nucleotide exchange factor for Rap1 GTPase, reveals a novel upstream regulatory mechanism in the mitogen-activated protein kinase signaling pathway. Mol. Cell. Biol. 22:7658-7666.[Abstract/Free Full Text]
    36. 36
  36. Lee, Y. S., and R. W. Carthew. 2003. Making a better RNAi vector for Drosophila: use of intron spacers. Methods 30:322-329.[CrossRef][Medline]
    37. 37
  37. Luo, L., Y. J. Liao, L. Y. Jan, and Y. N. Jan. 1994. Distinct morphogenetic functions of similar small GTPases: Drosophila Drac1 is involved in axonal outgrowth and myoblast fusion. Genes Dev. 8:1787-1802.[Abstract/Free Full Text]
    38. 38
  38. Major, R. J., and K. D. Irvine. 2006. Localization and requirement for Myosin II at the dorsal-ventral compartment boundary of the Drosophila wing. Dev. Dyn. 235:3051-3058.[CrossRef][Medline]
    39. 39
  39. Martin, P., and S. M. Parkhurst. 2004. Parallels between tissue repair and embryo morphogenesis. Development 131:3021-3034.[Abstract/Free Full Text]
    40. 40
  40. Mino, A., T. Ohtsuka, E. Inoue, and Y. Takai. 2000. Membrane-associated guanylate kinase with inverted orientation (MAGI)-1/brain angiogenesis inhibitor 1-associated protein (BAP1) as a scaffolding molecule for Rap small G protein GDP/GTP exchange protein at tight junctions. Genes Cells 5:1009-1016.[Abstract]
    41. 41
  41. Mishra, S., S. M. Smolik, M. A. Forte, and P. J. Stork. 2005. Ras-independent activation of ERK signaling via the torso receptor tyrosine kinase is mediated by Rap1. Curr. Biol. 15:366-370.[CrossRef][Medline]
    42. 42
  42. Morata, G., and P. Ripoll. 1975. Minutes: mutants of drosophila autonomously affecting cell division rate. Dev. Biol. 42:211-221.[CrossRef][Medline]
    43. 43
  43. Nikolaidou, K. K., and K. Barrett. 2004. A Rho GTPase signaling pathway is used reiteratively in epithelial folding and potentially selects the outcome of Rho activation. Curr. Biol. 14:1822-1826.[CrossRef][Medline]
    44. 44
  44. Paladi, M., and U. Tepass. 2004. Function of Rho GTPases in embryonic blood cell migration in Drosophila. J. Cell Sci. 117:6313-6326.[Abstract/Free Full Text]
    45. 45
  45. Patel, P. H., N. Thapar, L. Guo, M. Martinez, J. Maris, C. L. Gau, J. A. Lengyel, and F. Tamanoi. 2003. Drosophila Rheb GTPase is required for cell cycle progression and cell growth. J. Cell Sci. 116:3601-3610.[Abstract/Free Full Text]
    46. 46
  46. Pellis-van Berkel, W., M. H. Verheijen, E. Cuppen, M. Asahina, J. de Rooij, G. Jansen, R. H. Plasterk, J. L. Bos, and F. J. Zwartkruis. 2005. Requirement of the Caenorhabditis elegans RapGEF pxf-1 and rap-1 for epithelial integrity. Mol. Biol. Cell 16:106-116.[Abstract/Free Full Text]
    47. 47
  47. Sawada, Y., M. Tamada, B. J. Dubin-Thaler, O. Cherniavskaya, R. Sakai, S. Tanaka, and M. P. Sheetz. 2006. Force sensing by mechanical extension of the Src family kinase substrate p130Cas. Cell 127:1015-1026.[CrossRef][Medline]
    48. 48
  48. Spradling, A. C., and G. M. Rubin. 1982. Transposition of cloned P elements into Drosophila germ line chromosomes. Science 218:341-347.[Abstract/Free Full Text]
    49. 49
  49. Steimle, P. A., S. Yumura, G. P. Cote, Q. G. Medley, M. V. Polyakov, B. Leppert, and T. T. Egelhoff. 2001. Recruitment of a myosin heavy chain kinase to actin-rich protrusions in Dictyostelium. Curr. Biol. 11:708-713.[CrossRef][Medline]
    50. 50
  50. Takahashi, K., T. Matsuo, T. Katsube, R. Ueda, and D. Yamamoto. 1998. Direct binding between two PDZ domain proteins Canoe and ZO-1 and their roles in regulation of the jun N-terminal kinase pathway in Drosophila morphogenesis. Mech. Dev. 78:97-111.[CrossRef][Medline]
    51. 51
  51. Tamada, M., T. D. Perez, W. J. Nelson, and M. P. Sheetz. 2007. Two distinct modes of myosin assembly and dynamics during epithelial wound closure. J. Cell Biol. 176:27-33.[Abstract/Free Full Text]
    52. 52
  52. Tamada, M., M. P. Sheetz, and Y. Sawada. 2004. Activation of a signaling cascade by cytoskeleton stretch. Dev. Cell 7:709-718.[CrossRef][Medline]
    53. 53
  53. Tepass, U. 1996. Crumbs, a component of the apical membrane, is required for zonula adherens formation in primary epithelia of Drosophila. Dev. Biol. 177:217-225.[CrossRef][Medline]
    54. 54
  54. Van Aelst, L. 1998. Two-hybrid analysis of Ras-Raf interactions. Methods Mol. Biol. 84:201-222.[Medline]
    55. 55
  55. Verdier, V., G. C. Chen, and J. Settleman. 2006. Rho-kinase regulates tissue morphogenesis via non-muscle myosin and LIM-kinase during Drosophila development. BMC Dev. Biol. 6:38.[CrossRef][Medline]
    56. 56
  56. Wallingford, J. B., S. E. Fraser, and R. M. Harland. 2002. Convergent extension: the molecular control of polarized cell movement during embryonic development. Dev. Cell 2:695-706.[CrossRef][Medline]
    57. 57
  57. Wang, H., S. R. Singh, Z. Zheng, S. W. Oh, X. Chen, K. Edwards, and S. X. Hou. 2006. Rap-GEF signaling controls stem cell anchoring to their niche through regulating DE-cadherin-mediated cell adhesion in the Drosophila testis. Dev. Cell 10:117-126.[CrossRef][Medline]
    58. 58
  58. Wei, S. Y., L. M. Escudero, F. Yu, L. H. Chang, L. Y. Chen, Y. H. Ho, C. M. Lin, C. S. Chou, W. Chia, J. Modolell, and J. C. Hsu. 2005. Echinoid is a component of adherens junctions that cooperates with DE-cadherin to mediate cell adhesion. Dev. Cell. 8:493-504.[CrossRef][Medline]
    59. 59
  59. Winter, C. G., B. Wang, A. Ballew, A. Royou, R. Karess, J. D. Axelrod, and L. Luo. 2001. Drosophila Rho-associated kinase (Drok) links Frizzled-mediated planar cell polarity signaling to the actin cytoskeleton. Cell 105:81-91.[CrossRef][Medline]
    60. 60
  60. Xu, T., and G. M. Rubin. 1993. Analysis of genetic mosaics in developing and adult Drosophila tissues. Development 117:1223-1237.[Abstract]
    61. 61
  61. Yamada, T., T. Sakisaka, S. Hisata, T. Baba, and Y. Takai. 2005. RA-RhoGAP, Rap-activated Rho GTPase-activating protein implicated in neurite outgrowth through Rho. J. Biol. Chem. 280:33026-33034.[Abstract/Free Full Text]
    62. 62
  62. Yumura, S., M. Yoshida, V. Betapudi, L. S. Licate, Y. Iwadate, A. Nagasaki, T. Q. Uyeda, and T. T. Egelhoff. 2005. Multiple myosin II heavy chain kinases: roles in filament assembly control and proper cytokinesis in Dictyostelium. Mol. Biol. Cell 16:4256-4266.[Abstract/Free Full Text]

Molecular and Cellular Biology, November 2007, p. 7966-7980, Vol. 27, No. 22
0270-7306/07/$08.00+0     doi:10.1128/MCB.01275-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:

  • Jaffe, A. B., Kaji, N., Durgan, J., Hall, A. (2008). Cdc42 controls spindle orientation to position the apical surface during epithelial morphogenesis. JCB 183: 625-633 [Abstract] [Full Text]  

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Boettner, B.
Right arrow Articles by Van Aelst, L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Boettner, B.
Right arrow Articles by Van Aelst, L.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»