PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Dev Biol. Author manuscript; available in PMC 2010 January 19.
Published in final edited form as:
PMCID: PMC2808026
NIHMSID: NIHMS74429

Cad74A is regulated by BR and is required for robust dorsal appendage formation in Drosophila oogenesis

Abstract

Drosophila egg development is an established model for studying epithelial patterning and morphogenesis, but the connection between signaling pathways and egg morphology is still incompletely understood. We have identified a non-classical cadherin, Cad74A, as a putative adhesion gene that bridges epithelial patterning and morphogenesis in the follicle cells. Starting in mid-oogenesis, Cad74A is expressed in the follicle cells that contact the oocyte, including the border cells and most of the columnar follicle cells. However, Cad74A is repressed in two dorsolateral patches of follicle cells, which participate in the formation of tubular respiratory appendages. We show genetically that Cad74A is downstream of the EGFR and BMP signaling pathways and is repressed by the Zn-finger transcription factor Broad. The correlation of Cad74A repression in the cells that bend out of the plane of the follicular epithelium is preserved across Drosophila species and mutant backgrounds exhibiting a range of eggshell phenotypes. Complete removal of Cad74A from the follicle cells causes defects in dorsal appendage formation. Ectopic expression of Cad74A in the roof cells results in shortened, flattened appendages due to the hindered migration of the roof cells. Based on these results, we propose that Cad74A is part of the adhesive machinery that enables robust dorsal appendage formation, and as such provides a link between the patterning of the follicle cells and eggshell morphogenesis.

Keywords: cadherin, Drosophila, oogenesis, morphogenesis, regulation, patterning, Broad, follicle cell, epithelium, adhesion

Introduction

The transformation of epithelial sheets into complex three-dimensional structures is a key process in metazoan development. Epithelial morphogenesis is often preceded by the establishment of spatially nonuniform expression patterns of genes involved in the control of adhesion and cytoskeleton architecture. These patterns then guide the coordinated cell shape changes, movements, and rearrangements that lead to the formation of three-dimensional organs (Martinez Arias and Stewart, 2002). Drosophila oogenesis is an excellent model for studying epithelial patterning and morphogenesis (Berg, 2005; Horne-Badovinac and Bilder, 2005; Wu et al., 2008). In particular, patterning of the follicle cells (FCs), which form an epithelial monolayer encapsulating the developing egg, leads to the formation of several eggshell structures including: a micropyle for sperm entry, an operculum that provides an escape hatch for emerging larvae, and two dorsal appendages (DAs) that serve as respiratory tubes (Hinton, 1969). The DAs form by evagination, cell shape changes and the rearrangements of two symmetric groups of FCs (Dorman et al., 2004; Horne-Badovinac and Bilder, 2005; Ward and Berg, 2005).

The roof of each DA is derived from a dorsolateral patch of follicle cells (henceforth, roof cells) that strongly express Broad (BR), a Zn-finger transcription factor (Deng and Bownes, 1997; Tzolovsky et al., 1999) Similarly, the floor of each appendage is derived from adjacent cells (floor cells) that express rhomboid (rho), a gene encoding an intracellular protease in the epidermal growth factor receptor (EGFR) pathway (Deng and Bownes, 1997; Ruohola-Baker et al., 1993; Ward et al., 2006). The signaling and transcriptional mechanisms that establish the precise expression patterns of BR and rho have become progressively characterized (Dorman et al., 2004; Ward and Berg, 2005; Ward et al., 2006; Yakoby et al., 2008). Furthermore, the morphogenetic steps leading to DA formation have been carefully dissected (Dorman et al., 2004). In particular, the initial stage of DA formation begins with the apico-basal elongation of the DA primordial and the apical constriction and cell intercalation of the roof cells. As the roof cells bend out of the plane of the follicular epithelium, the floor cells slip underneath the roof cells to seal the bottom of the DAs, thereby forming tubular structures.

Elucidating the mechanisms of epithelial morphogenesis requires the identification of genes directly involved in the control of cell shape, adhesion, and motility. In other models of epithelial morphogenesis, cadherin proteins have been shown to play an important role in regulating cell-cell adhesion, cell rearrangements, and cell structure (Tepass, 1999). For example, previous studies established that DE-cadherin mediates cell sorting and cell migration events during oogenesis (Godt and Tepass, 1998; Gonzalez-Reyes and St Johnston, 1998; Niewiadomska et al., 1999; Oda et al., 1997; Pacquelet and Rorth, 2005), and Cad99C, which regulates the length of microvilli, is essential for proper secretion of the vitelline membrane (D'Alterio et al., 2005; Schlichting et al., 2006). Here, we propose that Cad74A, a non-classical cadherin that differs structurally from classical cadherins, provides an additional link between eggshell patterning and morphogenesis.

Cad74A is one of 17 genes in the Drosophila genome encoding putative cadherin proteins and has 13 cadherin domains, a N-terminal signal peptide, a transmembrane domain, and a short uncharacterized cytoplasmic domain (Hill et al., 2001; Hynes and Zhao, 2000) (Fig. 1A, B). A recent study that investigated the regulation of non-classical cadherins in the posterior spiracle reported that Cad74A mediates Ca-dependent homotypic cell-cell adhesion in cultured cells and is located sub-apically and apically in vivo (Lovegrove et al., 2006). Cad74A is also expressed in the early embryo, in the neurectoderm, the developing Drosophila larval brain, and eye imaginal disc, but the functional characterization of Cad74A has not been reported (Fung et al., 2008; Kearney et al., 2004; Schlichting and Dahmann, 2008; Tomancak et al., 2002).

Fig. 1
Gene and protein architecture and expression of Cad74A during oogenesis

Here, we report the dynamic expression of Cad74A mRNA and protein localization in oogenesis and show that this pattern correlates with the formation of multiple structural features of the eggshell. We demonstrate that the dynamic pattern of Cad74A expression is controlled by the EGFR and bone morphogenetic protein (BMP)/decapentaplegic (DPP) signaling pathways, two of the key regulators of follicle cell patterning (Berg, 2005). Specifically, high levels of BR, which is downstream of EGFR and BMP signaling, repress Cad74A in the roof cells while the Ets transcription factor, Pointed (PNT), activates Cad74A expression in the dorsal midline, likely by repressing BR (Deng and Bownes, 1997; Yakoby et al., 2008). Flies homozygous for an allele containing the complete deletion of Cad74A lay eggs that exhibit reproducible defects in the shape of the DAs, implying a possible role for Cad74A in eggshell morphogenesis. Strongly expressing Cad74A in the roof cells result in flattened, shortened appendages which are thickened at the base, suggesting that Cad74A has to be repressed in the floor cells to ensure proper DA elongation and migration. On the basis of these results, we propose that Cad74A is a component in the FC morphogenetic machinery that translates two-dimensional patterns into three-dimensional egg structures, possibly by modulating the adhesive properties of FCs.

Materials and methods

Fly stocks

The control stock used in this study is Oregon R (Ore R). GAL4 drivers and UAS lines that were used in this study include: CY2>UAS-mae (edl) (a gift from J. Duffy), UAS-Broad-Z1 (Zhou et al., 2004), UAS-pntP1 (Morimoto et al., 1996), GR1-GAL4 (Gupta and Schupbach, 2003), E4-GAL4 and CY2-GAL4, UAS-λtop (Queenan et al., 1997), UAS-dnEGFR (provided by A. Michelson), UAS-Dad (Tsuneizumi et al., 1997), UAS-tkv* (Lecuit et al., 1996), and br-GAL4 drivers were a gift of H. Cui and L. Riddiford. Since CY2>pntP1 and CY2>br-Z1 are lethal, the UAS-pntP1 and UAS-Broad-Z1 constructs were driven by the GAL80-GAL4, CY2-GAL4 line (Maximiliano L. Suster, 2004). To drive expression with the GAL80 flies, the flies were incubated at 27 degrees for 24 hours. The GAL80-GAL4, CY2-GAL4>UAS-Broad-Z1 flies were kept on yeast 12 hours prior to dissection, and the GAL80-GAL4, CY2-GAL4>UAS-pntP1 flies were put on yeast 24 hours before dissection. D. pseudoobscura was provided by V. Orgogozo and D. Stern. We also used D. virilis, rho2.2-lacZ (Ip et al., 1992), fs(1)K10 (Wieschaus et al., 1978), Ras85DE62K and Ras85D05703 (Schnorr and Berg, 1996).

Clonal Analysis

The FLP/FRT recombinant technique (Lee and Luo, 2001; Xu and Rubin, 1993) was used to generate loss of function clones where the standard protocol marks null clones with a loss of a GFP marker, and the MARCM technique marks the generation of clones with GFP expression. Ras null clones were generated using the following background: e22c-Gal4 UAS-FLP; FRT82B RasΔC40b/FRT82B ubi-GFP (Hou et al., 1995; James et al., 2002). br clones, marked using the MARCM technique (Lee and Luo, 2001), were generated with the brnpr-3 FRT19A line crossed to P{tubP-GAL80}LL1 hsFLP FRT19A;;UAS-mCD8GFP tubGAL4/MKRS flies (Ward et al., 2006). cut clones were generated with ctdb7 FRT18D/hGFP FRT18D; MKRS hsflp (Blochlinger et al., 1988; Sun and Deng, 2005). The br and ct flies were heat shocked at 37°C in an air incubator for 1–2 hours, kept at 25°C for 2–3 days (ctdb7 flies) or 5 five days (brnpr-3 flies), and then fed during the final 24 hours before dissection.

Cad74A lines

The entire coding region of Cad74A (coordinates 221107-209329 on AE003524) is deleted in Cad74A38A (coordinates of the deletion are 209199-221108 on AE003524), which was generated by an imprecise P-element excision of GE22619 (GenExel, Inc) (Fig. 1A). GFP-tagged lines of Cad74A (UAS-Cad74A-GFP53A, UAS-Cad74A-GFP99X) were a gift of A. Jacinto (Lovegrove et al., 2006). The UAS-Cad74ARNAi line was obtained from the Vienna Drosophila RNAi Center (Dietzl et al., 2007).

In situ hybridization

A modification of the standard in situ hybridization protocol was used for mRNA localization experiments, which did not include the RNase treatment (Wang et al., 2006; Yakoby et al., 2008). Postfixation in 4% paraformaldehyde was done for 15 minutes and ovaries were incubated in prehybridization solution for 3 hr at 60°C. The Cad74A antisense probe was a gift from M. Halfon. The primary antibody used for the enzyme color reaction was mouse anti-Digoxigenin-AP (1:2000, Roche).

Antibody staining and imaging

Antibodies used include mouse anti-BR core (1:50, DSHB), Oregon Green phalloidin (1:1000, Molecular Probes), Hoechst dye to stain for nuclei (1:10,000), rat anti-DE-cadherin (DCAD2, 1:100; DSHB, Oda et al., 1997), and guinea pig anti-Cad99C (1:3000, gift of D. Godt). To generate the mouse polyclonal Cad74A antibody, a 10xHistidine-Tag recombinant peptide (1398–1750 aa) was used for immunization (PrimmBiotech). Secondary antibodies used were the Alexa Fluor and Oregon Green secondary antibodies (1:1000, Molecular Probes). A standard immunostaining protocol was followed with modifications (Laplante and Nilson, 2006). For double staining of the mouse anti-BR core antibody and the mouse anti-Cad74A antibody, the Zenon® Mouse IgG1 Labeling Kit (Molecular Probes) was used at a 1[ratio]5 ratio of mouse anti-BR antibody to labeling reagent and then blocked with blocking reagent at a 1[ratio]6 ratio. The labeled antibody was then diluted 1:4 with PBS+0.2% Triton X-100. Incubation with the Zenon® complex was performed after the traditional immunostaining protocol for the polyclonal Cad74A antibody, followed by a 15 minute fixation with 4% paraformaldehyde in solution with PBST (0.2% Triton). Confocal images were taken with a PerkinElmer RS3 Spinning Disk Confocal microscope, a Zeiss LSM 510 microscope, or a Nikon Eclipse E800 compound microscope. Images were processed and organized with ImageJ (Rasband, 1997–2006), Photoshop (Adobe Systems, Inc., San Jose, CA) or Picasa2 (Google, Mountain View, CA).

Results

Cad74A is expressed in the follicle cells contacting the oocyte during late oogenesis

Cad74A was one of the transcripts identified in a microarray-based search for new targets of EGFR and BMP pathways in the follicular epithelium (Yakoby et al, unpublished). As revealed by in situ hybridization staining, Cad74A is expressed in a dynamic pattern during late oogenesis (Fig. 1C-F). We did not detect Cad74A expression in earlier stage egg chambers (Fig. 1C). From stage 10 and onwards, Cad74A is expressed in all columnar follicle cells except in two dorsolateral patches. It is also expressed in the border cells, but only after migrating from the anterior of the egg chamber to the oocyte/nurse cell boundary (arrow, Fig 1D). Throughout the later stages of oogenesis, Cad74A expression is expressed in the floor cells of the DAs and in the main body follicle cells (MBFCs) (Fig. 1E, F).

Correlation between Cad74A expression and morphogenesis

Cad74A’s ability to promotes homotypic cell-cell adhesion in transfected culture cells (Lovegrove et al., 2006) suggests a possible role for Cad74A in the morphogenesis of the dorsal appendages. One line of evidence that Cad74A plays a role in dorsal appendage formation is the correlation of the expression pattern in different genetic backgrounds with the final morphology of the dorsal anterior eggshell structures. For example, in fs(1)K10 egg chambers, the mislocalization of Gurken, an EGF ligand secreted by the oocyte and required for FC patterning, gives rise to an eggshell with a broad ventral “veil” of dorsal appendage material (Wieschaus et al., 1978). Based on the proposed correlation between the Cad74A pattern and appendage morphology, we expected that the expression of Cad74A in this mutant would be repressed in the ventral anterior band of the follicle cells, as is indeed the case (Fig. 1G, K). In the CY2>mae background, which expresses the inhibitor of the transcription factor Pointed (Yamada et al., 2003) throughout the FCs using the GAL4/UAS system (Duffy, 2002), there is a single broad dorsal appendage (Fig 1L). With the loss of the dorsal midline, the expression of Cad74A is repressed in a large dorsal patch of the follicular epithelium (Fig 1H). Furthermore, repression of Cad74A in the roof cells of other species with varying number of DAs is conserved. This pattern holds for both D. pseudoobscura, which exhibits a similar Cad74A spatial pattern similar to D. melanogaster and has two appendages, and for D. virilis, which has two patches of Cad74A repression that corresponds to the shape of BR protein domain and has four appendages (James and Berg, 2003; Nakamura and Matsuno, 2003) (Fig. 1I,M,J,N). The conservation of the Cad74A pattern across species suggests a functional role in eggshell morphology. Moreover, Cad74A is repressed in cells that undergo apical constriction and bend out of the FC epithelial plane to form appendage material, suggesting the possibility that local repression of Cad74A is a necessary condition for cells to detach from the oocyte (allowing the floor cells to invaginate between the roof cells and the oocyte), an important step in the morphogenesis of DAs (Dorman et al., 2004). Given that Cad74A can promote homotypic cell-adhesion (Lovegrove et al., 2006), repression of Cad74A in the roof cells may decreases adhesion between these cells, allowing increased junctional remodelling and, thus, arrangement between roof cells, a prerequisite for their morphogenetic movements away from the oocyte.

Cad74A localizes to the apical membranes of FCs

To determine the localization of the Cad74A protein, we raised a polyclonal antibody to Cad74A and found that Cad74A protein distribution largely mirrors the mRNA expression pattern. However, the distribution of Cad74A appears more punctuate in the dorsal midline and floor cells (Fig. 2A, B). At later stages, when the medial row and the anterior row of floor cells appear to meet and fuse, Cad74A is located in a puncate pattern that surrounds the floor cell apices (Fig. 2C, D). In the MBFCs, Cad74A is enriched at the apical membrane, but cytoplasmic Cad74A is also detected, possibly representing internalized molecules (Fig. 2E, F). Cad74A strongly marks FC microvilli at stage 10B when protein is first detected. However, microvilli length decreases in the MBFCs during later stages of oogenesis. While Cad99C and Cad74A share similar staining to the apical membrane of FCs, the expression patterns of Cad74A and Cad99C differ dynamically as they do not appear to strongly co-localize in later stages of oogenesis (Fig 2G-I).

Fig. 2
Localization of Cad74A during oogenesis

Antibody immunoreactivity is specific to Cad74A as no staining is detected in Cad74A38A/38A egg chambers that were incubated in the same tube with eggs expressing the Cad74A-GFP construct, which served as an internal control. Co-staining for GFP allowed us to differentiate between the two genotypes (Fig. 2J-M). Furthermore, no immunoreactivity to Cad74A is found at late stages in the posterior FCs when the E4-GAL4 drives expression of a Cad74A RNAi construct (Dietzl et al., 2007; Queenan et al., 1997) (Fig. 2N, O), demonstrating the efficacy of the UAS-Cad74ARNAi line. When Cad74A-GFP was driven by the early driver, GR1-GAL4 (Gupta and Schupbach, 2003), GFP and Cad74A staining co-localized to the same FCs (Fig. 2P-R), showing that the Cad74A antibody recognizes Cad74A-GFP. We did not detect any staining in the oocyte.

EGFR and BMP signaling jointly regulate Cad74A

The repression of Cad74A in the two dorsolateral patches suggests joint regulation by the EGFR and BMP signaling pathways, which have been shown to act as the key patterning signals of the dorsal anterior region of the follicular epithelium (Atkey et al., 2006; Berg, 2005; Dobens and Raftery, 2000; Shravage et al., 2007; Yakoby et al., 2008). Hence, we used the UAS/GAL4 system (Duffy, 2002) to perturb the EGFR and BMP pathways to test how Cad74A expression changes. When activated EGFR signals at high levels in the follicle cells (CY2>λtop), Cad74A is expressed in all of the main body follicle cells in stage 10 egg chambers except for the anteriormost row of cells (Fig. 3A). At later stages, the region of repression expands and then two dark bands of strong Cad74A expression are seen between a narrow band of repressed Cad74A, corresponding to a single band of high levels of BR (Fig 3B, C, arrowhead). This band likely corresponds to the boundary between the operculum and the main body of the egg, which is shown in the dorsalized eggshell with no dorsal appendages (Fig. 3D). The broad, dark band of Cad74A expression matches the band of MAPK signaling seen in CY2>λtop as a result of the rho spatial pattern (not shown) (Queenan et al., 1997). In the opposite perturbation, downregulation of EGFR signaling (CY2>dnEGFR), results in a ventralized phenotype. In this background, Cad74A is expressed in all FCs, where uniform levels of BR are observed (Fig. 3E-H).

Fig. 3
Cad74A expression is regulated by the EGFR and BMP signaling pathways

With ectopic expression of a constitutively active form of TKV receptor (CY2>tkv*, Lecuit et al., 1996; Nellen et al., 1996), Cad74A is initially repressed throughout the dorsal anterior region (Fig. 3I). Later, the dorsal anterior region is filled with Cad74A, but a thin band in the shape of an arc remains (Fig. 3J). This band of Cad74A repression correlates to the band of high BR and may correspond to the boundary between the enlarged operculum and the main body of the egg (Fig. 3J-L). Ectopic expression of an inhibitor of BMP signaling (CY2>Dad) enlarges the roof cells towards the anterior direction (Yakoby et al., 2008), which is accompanied by repression of Cad74A in those cells (Fig. 3M). At later stages, the BR patches decrease in size and are shifted in the dorsal anterior direction (Fig. 3N-O). Consequently, the two dorsal appendages are also shifted toward the anterior edge of the egg chamber (Fig 3P). In each of these genetic perturbations, we noticed that Cad74A is repressed in cells with high levels of BR, which suggests that high levels of BR may repress Cad74A.

Regulation of Cad74A by BR

To clarify the relationship between BR and Cad74A, we first examined cells homozygous for a null allele of Ras, a component of the EGFR signaling pathway. Previously it was shown that loss of Ras in the roof cells leads to a cell autonomous loss of BR (Atkey et al., 2006; Deng and Bownes, 1997; Yakoby et al., 2008). If BR represses Cad74A, then reduced levels of BR in RasΔC40b clones should result in ectopic Cad74A. In 9/12 (75%) of stage 10B/11 egg chambers, we found this to be the case (Fig. 3Q-S). While BR is not eliminated from the patch, it is at a level comparable to the basal level found in the MBFCs, alleviating the repression of Cad74A by BR. Cad74A is found to be strongly expressed in such clones. The negative examples appeared in younger stage 10B egg chambers, suggesting that the loss of EGFR signaling and reduction of BR in the roof cells does not immediately induce Cad74A (Fig. 3T-U). On the other hand, downregulation of BR in the dorsal anterior requires high levels of EGFR signaling. Hypomorphic alleles of Ras would thus be expected to reduce Cad74A expression in the dorsal midline, which is what is observed (Fig. 3V). In each perturbation examined, we find changes in the shape and location of three domains of Cad74A expression: I. Intermediate/delayed expression in the dorsal midline (High EGFR/BR is absent), II: Strong repression in the roof cells (Moderate levels of EGFR/high BR), III: strong expression in the ventral and posterior FCs (Low levels of EGFR/basal levels of BR) (Fig. 3W).

The appearance of Cad74A in the dorsal midline and the absence of Cad74A in the roof cells suggest the possibility that high levels of BR in the dorsal anterior FCs could be sufficient to repress Cad74A (Fig. 4A-C). The co-localization of basal BR and Cad74A in the posterior MBFCs could imply either that BR is not at high enough levels to repress Cad74A or acts as an activator at lower levels. To directly test the regulation of Cad74A by BR directly, we overexpressed the br-Z1 isoform using the GAL80-GAL4, CY2-GAL4 system, which promotes expression in the MBFCs starting in mid-oogenesis (stages 8–10) (Queenan et al., 1997) when the flies are kept at elevated temperatures (see Materials and Methods for details; Fig. 4D). Overexpression of br reduced levels of the Cad74A transcript in the dorsal anterior almost completely and to a lesser extent in the posterior and ventral FCs (Fig. 4E, F). Because the repression of Cad74A by BR was mostly confined to the dorsal anterior region of the FCs, we examined the BR protein distribution in this background. Interestingly, we found that the highest levels of BR in the CY2>br-Z1 background were limited mostly to the anterior, with patchy upregulation of BR in the posterior main body as shown for egg chambers double stained for BR protein and Cad74A, demonstrating not only that Cad74A is negatively regulated by ectopic BR, but also a possibility that BR may undergo post-transcriptional autoregulation (Fig. 4G-I).

Fig. 4
BR represses Cad74A in the roof cells

To confirm that BR is both necessary and sufficient for repressing Cad74A in the roof cells, we examined Cad74A expression in egg chambers mosaic for the null allele, brnpr-3 (Belyaeva et al., 1980; Kiss et al., 1988) using the MARCM technique (Lee and Luo, 2001). As noticed earlier (Ward et al., 2006), loss of BR was not faithfully marked by GFP expression, although the presence of GFP, which was found in a minority of cases, accurately marked homozygous null brnpr-3 clones. We therefore marked clones with a loss of BR staining. The tilt and keystone shape of the roof cells, as well as the invagination of Cad74A-expressing floor cells made it difficult to determine whether ectopic Cad74A expression in the roof cells was completely cell autonomous. To avoid confusing the invaginating Cad74A-expressing floor cells with ectopic expression in brnpr-3 clones, we focused mainly on scoring clones in the roof cells of stage 10B/11 egg chambers. We found ectopic Cad74A expression in 94% (30/32 distinct clones, 20 egg chambers examined) of clones examined in the roof cells, of which 75% (24/32 clones) appeared to be completely cell-autonomous (Fig. 4J-0). In two clones, we did not see expression of Cad74A (both stage 10B egg chambers, one example is shown (arrowhead) in Fig. 4L, M) and in six additional clones, the ectopic expression in the clone did not appear to be span the clone completely. We thus conclude that high levels of BR repress Cad74A in the roof cells, but that other factor(s), including delayed induction or repression independent of BR, result in cases where Cad74A is still repressed in brnpr-3 homozygous cells during stages 10B/11.

Regulation of Cad74A by additional transcriptional factors

Based on the difference in the levels of Cad74A expression in the midline and the main body, we examined the role of pointed (pnt) in regulating Cad74A. Downregulation of BR in the dorsal anterior requires the Ets transcription factor pnt, which is in turn expressed only at the highest levels of EGFR signaling (Deng and Bownes, 1997; Morimoto et al., 1996; Yamada et al., 2003). To confirm the regulatory connection between pnt and Cad74A, the pntP1 isoform was ectopically expressed starting at stage 8, which resulted in the early (earlier than stage 10) and strong expression of Cad74A throughout the main body FC epithelium and the loss of repression of Cad74A in the roof cells (Fig. 5A-C). In the opposite direction, ectopic expression of Mae, an inhibitor of pntP2 transcriptional activity (Yamada et al., 2003), leads to a loss of Cad74A in the dorsal anterior bridge, demonstrating that PNT, likely through repression of BR in the dorsal anterior, is required for Cad74A expression (Fig. 5D-F).

Fig. 5
Regulation of Cad74A by other transcription factors

The regulation of Cad74A changes mainly in the dorsal anterior domain when perturbations in either EGFR or BMP signaling are made, implying that other inputs may govern Cad74A regulation in the ventral and posterior side of the egg chamber. Cad74A was shown to be downstream of cut in the posterior spiracle (Lovegrove et al., 2006). The transcription factor cut is involved in cell-cycle progression and FC differentiation in early stages of oogenesis and later reappears at stage 10B (Sun and Deng, 2005), when Cad74A is first expressed. We examined egg chambers mosaic for the null allele cut, ctdb7 (Blochlinger et al., 1988) and found that expression of Cad74A is unaffected in 100% of cutdb7 clones examined (10/10 egg chambers) (Fig. 5G-I). This result suggests that different enhancers are responsible for expression in the FCs and the posterior spiracle.

Cad74A is required for robust DA formation

To test the hypothesis that Cad74A is functionally important for DA formation, we generated a null allele, Cad74A38A, using imprecise P-element excision that resulted in the complete deletion of the gene with no Cad74A transcript detected (Fig. 1A; Fig 6A, B).

Fig. 6
Phenotypic analysis of the Cad74A38A allele

Homozygous Cad74A38A flies reproducibly lay a subset of eggs with severe DA defects. We found that 17% (315/1820 eggs examined over multiple independent counts) of Cad74A38A/38A eggs laid at room temperature lacked appendages or exhibited appendages that are significantly deformed or shortened to varying degrees (Fig. 6C-F). As a control, the parental line used to generate the deletion does not show loss of DAs (data not shown). In the fraction of Cad74A38A/38A eggs with severely disrupted DAs, a range of DA length was observed, suggesting that tube elongation, a major step in eggshell morphogenesis (Berg, 2005; Dorman et al., 2004), was affected. We did not see signs of collapsed eggs, confirming the observation based on localization that Cad99C (D’Alterio et al., 2005; Schlichting et al., 2006) and Cad74A have different functions. To verify this conclusion, we investigated the permeability of Cad74A38A/38A eggs after chemically removing the outer chorion to the dye Neutral Red (Schlichting et al., 2006) and found that Cad74A38A/38A eggs were impermeable to Neutral Red (data not shown), as expected if the vitelline membrane was deposited correctly.

Ectopic expression of Cad74A in the roof cells hinders DA elongation

To test our hypothesis that repression of Cad74A in the roof cells is important for DA formation, we expressed Cad74A-GFP (UAS-Cad74A-GFP53A) in roof cells using a late stage br-GAL4 driver developed in the Riddiford lab. Overexpression with br>Cad74A-GFP gave a strong phenotype with the DAs largely replaced by thick, flattened ridges (Fig. 7B), compared to br>GFP eggs with properly formed DAs (Fig. 7A) and eggs laid by sister flies to br>Cad74A-GFP (not shown). While the controls also sometimes exhibited shortened DAs, the phenotype was always less penetrant and less severe (Fig. 7C). The strong expression of Cad74A-GFP in the roof cells results in large aggregates of Cad74A in the apical half of the roof follicle cells (Fig. 7D-F). Apical constriction was sometimes affected in br>Cad74A-GFP egg chambers during stage 12, but not absolutely prevented (Fig. 7G-N). In egg chambers at late stage 13/14, the roof cells marked by BR (Fig. 7R compared to the control, Fig. 7O) failed to complete migration towards the anterior tip of the egg chamber, consistent with the observed eggshell phenotypes (Fig. 7B). Based on these overexpression experiments, we suggest that repression of Cad74A in the roof cells is required for robust DA morphogenesis.

Figure 7
Overexpression of Cad74A-GFP hinders the proper migration of the roof cells

Discussion

High levels of BR repress Cad74A in the roof cells

Starting at stage 10, Cad74A is expressed in the main body FCs, except for cells expressing high levels of BR, corresponding to the roof primordia. As oogenesis proceeds, Cad74A continues to be expressed in the cells surrounding the roof cells, which eventually detach from the oocyte as the floor cells slip between the roof cells and oocyte. The dynamic expression pattern of Cad74A and its ability to promote cell adhesion (Lovegrove et al., 2006) suggests that Cad74A is a link between FC patterning and morphogenesis.

While investigating how the EGFR and BMP signaling pathways jointly regulate Cad74A expression, we found that perturbations in either pathway result in transitions of Cad74A expression that can be predicted from the final DA morphology or the corresponding expression pattern of high levels of BR. Based on the expression pattern of Cad74A and BR, two general models of regulation can be envisioned (Fig. 8A, B):

  1. Cad74A is uniformly activated throughout the follicular epithelium from stage 10 and throughout late oogenesis and is repressed in the roof cells by sufficiently high levels of BR, or
  2. Cad74A is uniformly activated at low levels of EGFR signaling in the ventral and posterior MBFCs and repressed by an unknown factor expressed in the dorsal anterior. A second signal in the dorsal midline and floor cells activates Cad74A, which is slightly delayed from ventral and posterior expression. In this model, BR and Cad74A are regulated in parallel.

Figure 8
From patterns to gene regulation and morphogenesis of effector proteins

BR has long served as a marker for DA morphogenesis (Deng and Bownes, 1997; Dorman et al., 2004; Tzolovsky et al., 1999) and is a conserved regulator of morphogenesis during many other stages such as pupation (Konopova and Jindra, 2008; Suzuki et al., 2008). Through ectopic expression and loss-of-function mosaic analysis, we show that the repression of Cad74A in the roof cells is mediated by BR (consistent with model 1). The basal level of BR in the posterior and ventral MBFCs appears to only slightly repress Cad74A (data not shown), suggesting that another activator is present to stimulate Cad74A expression in the MBFCs or that the basal level of BR is not sufficient to repress Cad74A. Additionally, Cad74A is also activated by the Ets transcription factor PNT, possibly indirectly through repression of BR (Deng and Bownes, 1997). In conclusion, while the first model of regulation (uniform activation − high levels of BR) appears to be involved, the second possible mode of regulating Cad74A (uniform activation − dorsal repression + a second dorsal midline signal) cannot be ruled out because: 1) Cad74A in the posterior and ventral MBFCs is expressed at stronger levels and slightly earlier to the dorsal midline expression, 2) PNT acts as an effective ectopic activator of Cad74A expression even before stage 10, and 3) loss of BR in Ras and br clones does not absolutely ensure Cad74A expression at stages when Cad74A is expressed in the other FCs. Finally, we examined a known factor regulating Cad74A during embryogenesis, cut, which is expressed uniformly in the FCs during midoogenesis (Sun and Deng, 2005), and found no effect in the main body FCs. Future work will require identifying the specific transcription factors that stimulate Cad74A expression in the posterior and ventral MBFCs. Establishing the regulatory connection between BR and Cad74A places Cad74A into the larger regulatory network responsible for patterning the FCs during midoogenesis (Yakoby et al., 2008; shown in Fig. 8C).

The role of Cad74A in epithelial morphogenesis

Previous efforts to genetically probe for the function of non-classical cadherins in the posterior spiracle, including Cad74A, Cad88C, Cad86C, and Cad96C, have been confounded by a lack of a strong phenotype and functional redundancy (Lovegrove et al., 2006). This was also found to be the case for Cad86C, which is specifically upregulated in cells of the morphogenetic furrow of the eye disc, but does not show a phenotype in the eye imaginal disc (Schlichting and Dahmann, 2008). The lack of a phenotype in single knock-downs confirms the robustness of the morphogenetic program to a loss of expression of many effector proteins. The redundant function of the non-classical cadherins in the posterior spiracle suggests that single loss-of-function alleles of Cad74A should not lead to strong phenotypes. To test this expectation for Cad74A in oogenesis, we created and analyzed a complete null of Cad74A (Cad74A38A). Surprisingly, Cad74A does not appear to be fully redundant with other adhesion molecules as complete deletion of Cad74A leads to partially penetrant, but severe DA phenotypes, where the migration and tubular elongation of the DAs is severely disrupted.

The disruption of the formation of the DAs suggests that the morphogenesis and migration of the roof cells is compromised. This may be due to the loss of segregation between the roof and surrounding follicle cells due to differences in adhesion between the follicle cells or a change in the surface tension along the apical membrane. Alternatively, Cad74A may play a role in epithelial polarity or in the secretion of the chorion. Overexpression of Cad74A specifically in the roof cells affects cell migration, consistent with a hypothesis that Cad74A acts as a “glue” to the oocyte. Alternatively, ectopic Cad74A in the roof cells prevents remodeling of the adherens junctions during intercalation (Dorman et al., 2004). Down regulation of Cad74A in ectopic patches is not sufficient to cause the FCs to detach from the oocyte as evident in the overall normal FC morphology found in posterior follicle cells when Cad74A was knocked-down (E4>Cad74ARNai background, Fig. 2N-O). As a preliminary conclusion, we suggest that Cad74A could perform a role in modulating the surface tension along the apical domain or in stabilizing cell-cell connections, but this model will require confirmation through future biochemical studies that identify the interacting partners of Cad74A. Finally, the incomplete penetrance of the Cad74A null allele may be a function either of a partial functional redundancy with other adhesive molecules (a strong possibility due to the intricate patterning of multiple adhesive molecules in this tissue) or due to the inherent robustness of morphogenesis (the machinery that performs the unit operation of cellular changes compensates if one aspect if adversely affected) (Kafri et al., 2006). A comprehensive study of double knock-downs of adhesive molecules will shed light onto this problem in the future.

Syn-expression groups of genes involved in morphogenesis

Producing an atlas of differentially expressed effector genes will be important to identify the roles of adhesion and cytoskeleton genes in the morphogenesis of the DAs. Cad74A shares a similar two-dimensional expression pattern with several other documented effector genes, and thus the model for Cad74A regulation can serve as a backbone to compare and contrast the regulation of these other effector genes. For example, the Cad74A mRNA expression pattern during stage 10B is strikingly similar to the expression of Echinoid (ED) protein, a cell adhesion molecule which is implicated in the formation of an actomyosin contraction cable that surrounds the roof cell primordial boundary, with an important difference: expression in the dorsal midline appears later in Cad74A but earlier for ED (Laplante and Nilson; Fig. 8D). The homophilic adhesion molecule Fasciclin 3 (FAS3) also accumulates in the dorsal anterior region of the egg chamber and likely plays a role in setting the boundary between roof and floor cells (Ward and Berg, 2005). A basal pool of DE-Cadherin, separate from apical DE-Cadherin, is found at elevated levels in the dorsal anterior cells (James et al., 2002). The toll-like receptor, 18-wheeler (18w), which may also play a role in adhesion or cell signaling by regulating the Rho-GTPase-signaling pathway (Kolesnikov and Beckendorf, 2007), is also expressed in a ring that appears to surround the roof cells during mid-oogenesis (Kleve et al., 2006). A subpopulation of α-actinin, which is involved in cross-linking and bundling actin filaments, is also expressed in the somatic FCs but repressed in the dorsal anterior follicle cells during mid-oogenesis and only overlaps BR in the posterior edge of the roof primordia (Wahlstrom et al., 2006). Thus, we group Cad74A with Ed, Fas3, E-Cad, 18w and FC-α-actinin, as effector genes that show absence, reduction, or differential expression in the roof cell primordia compared to the surrounding cells (Fig. 8D). This does not imply that these effector proteins are interacting directly, but that expression of high levels of these effector proteins in overlapping patterns is required to coordinate the morphogenesis of the DAs. The differences in the expression of these effector genes between the dorsal midline (on one side of the roof cell boundary) and the posterior and ventral MBFCs (the other side of the roof cell boundary) may reflect differences in the two possible modes of regulation discussed previously.

The correlation between final dorsal appendage morphology and Cad74A expression suggests that the regulation of genes involved in DA formation is highly modular and interconnected (e.g., a shift in BR expression corresponds to a shift in the domain of Cad74A repression that correlates with final dorsal appendage morphology). Further characterization of the regulation and function of Cad74A and other effector genes during epithelial morphogenesis will be important in synthesizing a comprehensive mechanism of dorsal appendage formation.

Acknowledgements

The authors thank T. Schüpbach, D. Godt and L. Riddiford for reviewing earlier versions of the manuscript and for providing flies and reagents. We are indebted to M. Rossi and Y. Gogotsi for expert imaging assistance, members of the Shvartsman lab for reviewing the manuscript, and D. Gong, A. Rinberg, C. Watson, S. Leffler and D. Weiner for technical assistance. We thank J. Duffy, D. Godt, M. Halfon, W. Deng, V. Orgogozo, D. Stern, S. Simoes and A. Jacinto, GenExel, Inc., the Developmental Studies Hybridoma Bank at the University of Iowa, and the Vienna Drosophila RNAi Center for antibody reagents and flies. J.Z. is supported by the Fannie and John Hertz Foundation and the Princeton Wu fellowship, C.B is supported by the Burroughs-Wellcome graduate training program in Biological Dynamics, and X.Z. by the National Institute of Health (GM60122). This work was supported by the following NIH grants: P50 GM071508 and RO1 GM078079.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • Atkey MR, et al. Capicua regulates follicle cell fate in the Drosophila ovary through repression of mirror. Development. 2006;133:2115–2123. [PubMed]
  • Belyaeva ES, et al. Cytogenetic analysis of the 2B3-4--2B11 region of the X-chromosome of Drosophila melanogaster. I. Cytology of the region and mutant complementation groups. Chromosoma. 1980;81:281–306. [PubMed]
  • Berg CA. The Drosophila shell game: patterning genes and morphological change. Trends in Genetics. 2005;21:346–355. [PubMed]
  • Blochlinger K, et al. Primary structure and expression of a product from cut, a locus involved in specifying sensory organ identity in Drosophila. Nature. 1988;333:629. [PubMed]
  • D'Alterio C, et al. Drosophila melanogaster Cad99C, the orthologue of human Usher cadherin PCDH15, regulates the length of microvilli. J Cell Biol. 2005;171:549–558. [PMC free article] [PubMed]
  • Deng WM, Bownes M. Two signalling pathways specify localised expression of the Broad-Complex in Drosophila eggshell patterning and morphogenesis. Development. 1997;124:4639–4647. [PubMed]
  • Dietzl G, et al. A genome-wide transgenic RNAi library for conditional gene inactivation in Drosophila. Nature. 2007;448:151–156. [PubMed]
  • Dobens LL, Raftery LA. Integration of epithelial patterning and morphogenesis in Drosophila ovarian follicle cells. Dev Dyn. 2000;218:80–93. [PubMed]
  • Dorman JB, et al. bullwinkle is required for epithelial morphogenesis during Drosophila oogenesis. Dev Biol. 2004;267:320–341. [PubMed]
  • Duffy JB. GAL4 system in Drosophila: a fly geneticist's Swiss army knife. Genesis. 2002;34:1–15. [PubMed]
  • Fung S, et al. Expression profile of the cadherin family in the developing Drosophila brain. J Comp Neurol. 2008;506:469–488. [PubMed]
  • Godt D, Tepass U. Drosophila oocyte localization is mediated by differential cadherin-based adhesion. Nature. 1998;395:387–391. [PubMed]
  • Gonzalez-Reyes A, St Johnston D. The Drosophila AP axis is polarised by the cadherin-mediated positioning of the oocyte. Development. 1998;125:3635–3644. [PubMed]
  • Gupta T, Schupbach T. Cct1, a phosphatidylcholine biosynthesis enzyme, is required for Drosophila oogenesis and ovarian morphogenesis. Development. 2003;130:6075–6087. [PubMed]
  • Hill E, et al. Cadherin superfamily proteins in Caenorhabditis elegans and Drosophila melanogaster. J Mol Biol. 2001;305:1011–1024. [PubMed]
  • Hinton HE. Respiratory systems of insect egg shells. Annu Rev Entomol. 1969;14:343–368. [PubMed]
  • Horne-Badovinac S, Bilder D. Mass transit: epithelial morphogenesis in the Drosophila egg chamber. Dev Dyn. 2005;232:559–574. [PubMed]
  • Hou XS, et al. The Torso Receptor Tyrosine Kinase Can Activate Raf in a Ras-Independent Pathway. Cell. 1995;81:63–71. [PubMed]
  • Hynes RO, Zhao Q. The evolution of cell adhesion. J Cell Biol. 2000;150:F89–F96. [PMC free article] [PubMed]
  • Ip YT, et al. The dorsal gradient morphogen regulates stripes of rhomboid expression in the presumptive neuroectoderm of the Drosophila embryo. Genes Dev. 1992;6:1728–1739. [PubMed]
  • James KE, Berg CA. Temporal comparison of Broad-Complex expression during eggshell-appendage patterning and morphogenesis in two Drosophila species with different eggshell-appendage numbers. Gene Expr Patterns. 2003;3:629–634. [PubMed]
  • James KE, et al. Mosaic analyses reveal the function of Drosophila Ras in embryonic dorsoventral patterning and dorsal follicle cell morphogenesis. Development. 2002;129:2209–2222. [PubMed]
  • Kafri R, et al. The regulatory utilization of genetic redundancy through responsive backup circuits. Proceedings of the National Academy of Sciences. 2006;103:11653–11658. [PubMed]
  • Kearney JB, et al. Gene expression profiling of the developing Drosophila CNS midline cells. Dev Biol. 2004;275:473–492. [PMC free article] [PubMed]
  • Kiss I, et al. Interactions and developmental effects of mutations in the Broad-Complex of Drosophila melanogaster. Genetics. 1988;118:247–259. [PubMed]
  • Kleve CD, et al. Expression of 18-wheeler in the follicle cell epithelium affects cell migration and egg morphology in Drosophila. Developmental Dynamics. 2006;235:1953–1961. [PubMed]
  • Kolesnikov T, Beckendorf SK. 18 Wheeler regulates apical constriction of salivary gland cells via the Rho-GTPase-signaling pathway. Developmental Biology. 2007;307:53–61. [PMC free article] [PubMed]
  • Konopova B, Jindra M. Broad-Complex acts downstream of Met in juvenile hormone signaling to coordinate primitive holometabolan metamorphosis. Development. 2008;135:559–568. [PubMed]
  • Laplante C, Nilson LA. Differential expression of the adhesion molecule Echinoid drives epithelial morphogenesis in Drosophila. Development. 2006;133:3255–3264. [PubMed]
  • Lecuit T, et al. Two distinct mechanisms for long-range patterning by Decapentaplegic in the Drosophila wing. Nature. 1996;381:387–393. [PubMed]
  • Lee T, Luo L. Mosaic analysis with a repressible cell marker (MARCM) for Drosophila neural development. Trends in Neurosciences. 2001;24:251–254. [PubMed]
  • Lovegrove B, et al. Coordinated Control of Cell Adhesion, Polarity, and Cytoskeleton Underlies Hox-Induced Organogenesis in Drosophila. Current Biology. 2006;16:2206–2216. [PubMed]
  • Martinez Arias A, Stewart A. Molecular principles of animal development. Oxford; New York: Oxford University Press; 2002.
  • Maximiliano L, Suster LSMBMBS. Refining GAL4-driven transgene expression in Drosophila with a GAL80 enhancer-trap. genesis. 2004;39:240–245. [PubMed]
  • Morimoto AM, et al. Pointed, an ETS domain transcription factor, negatively regulates the EGF receptor pathway in Drosophila oogenesis. Development. 1996;122:3745–3754. [PubMed]
  • Nakamura Y, Matsuno K. Species-specific activation of EGF receptor signaling underlies evolutionary diversity in the dorsal appendage number of the genus Drosophila eggshells. Mech Dev. 2003;120:897–907. [PubMed]
  • Nellen D, et al. Direct and Long-Range Action of a DPP Morphogen Gradient. Cell. 1996;85:357. [PubMed]
  • Niewiadomska P, et al. DE-Cadherin Is Required for Intercellular Motility during Drosophila Oogenesis. J. Cell Biol. 1999;144:533–547. [PMC free article] [PubMed]
  • Oda H, et al. Phenotypic analysis of null mutants for DE-cadherin and Armadillo in Drosophila ovaries reveals distinct aspects of their functions in cell adhesion and cytoskeletal organization. Genes to Cells. 1997;2:29–40. [PubMed]
  • Pacquelet A, Rorth P. Regulatory mechanisms required for DE-cadherin function in cell migration and other types of adhesion. J Cell Biol. 2005;170:803–812. [PMC free article] [PubMed]
  • Queenan AM, et al. Ectopic activation of torpedo/Egfr, a Drosophila receptor tyrosine kinase, dorsalizes both the eggshell and the embryo. Development. 1997;124:3871–3880. [PubMed]
  • Rasband WS. ImageJ. Bethesda, Maryland, USA: U. S. National Institutes of Health; 1997–2006.
  • Ruohola-Baker H, et al. Spatially localized rhomboid is required for establishment of the dorsal-ventral axis in Drosophila oogenesis. Cell. 1993;73:953–965. [PubMed]
  • Schlichting K, Dahmann C. Hedgehog and Dpp signaling induce cadherin Cad86C expression in the morphogenetic furrow during Drosophila eye development. Mech Dev. 2008 [PubMed]
  • Schlichting K, et al. Cadherin Cad99C is required for normal microvilli morphology in Drosophila follicle cells. J Cell Sci. 2006;119:1184–1195. [PubMed]
  • Schnorr JD, Berg CA. Differential Activity of Ras1 during Patterning of the Drosophila Dorsoventral Axis. Genetics. 1996;144:1545–1557. [PubMed]
  • Schultz J, et al. SMART, a simple modular architecture research tool: identification of signaling domains. Proc Natl Acad Sci U S A. 1998;95:5857–5864. [PubMed]
  • Shravage BV, et al. The role of Dpp and its inhibitors during eggshell patterning in Drosophila. Development. 2007;134:2261–2271. [PubMed]
  • Sun J, Deng W-M. Notch-dependent downregulation of the homeodomain gene cut is required for the mitotic cycle/endocycle switch and cell differentiation in Drosophila follicle cells. Development. 2005;132:4229–4308. [PMC free article] [PubMed]
  • Suzuki Y, et al. The role of Broad in the development of Tribolium castaneum: implications for the evolution of the holometabolous insect pupa. Development. 2008;135:569–577. [PubMed]
  • Tepass U. Genetic analysis of cadherin function in animal morphogenesis. Curr Opin Cell Biol. 1999;11:540–548. [PubMed]
  • Tomancak P, et al. Systematic determination of patterns of gene expression during Drosophila embryogenesis. Genome Biol. 2002;3 research0088.1-0088.14. [PMC free article] [PubMed]
  • Tsuneizumi K, et al. Daughters against dpp modulates dpp organizing activity in Drosophila wing development. Nature. 1997;389:627–631. [PubMed]
  • Tzolovsky G, et al. The Function of the Broad-Complex During Drosophila melanogaster Oogenesis. Genetics. 1999;153:1371–1383. [PubMed]
  • Wahlstrom G, et al. Drosophila α-actinin in ovarian follicle cells is regulated by EGFR and Dpp signalling and required for cytoskeletal remodelling. Mech Dev. 2006;123:801–818. [PubMed]
  • Wang X, et al. Analysis of cell migration using whole-genome expression profiling of migratory cells in the Drosophila ovary. Dev Cell. 2006;10:483–495. [PubMed]
  • Ward EJ, Berg CA. Juxtaposition between two cell types is necessary for dorsal appendage tube formation. Mech Dev. 2005;122:241–255. [PubMed]
  • Ward EJ, et al. Border of Notch activity establishes a boundary between the two dorsal appendage tube cell types. Dev Biol. 2006 [PubMed]
  • Wieschaus E, et al. Wilhelm Rouxs Archives of Developmental Biology. Vol. 184. 1978. Fs(1)K10, a Germline-Dependent Female Sterile Mutation Causing Abnormal Chorion Morphology in Drosophila-Melanogaster; pp. 75–82.
  • Wu X, et al. Drosophila follicle cells: Morphogenesis in an eggshell. Seminars in Cell & Developmental Biology. 2008;19:271–282. [PMC free article] [PubMed]
  • Xu T, Rubin GM. Analysis of genetic mosaics in developing and adult Drosophila tissues. Development. 1993;117:1223–1237. [PubMed]
  • Yakoby N, et al. Drosophila eggshell is patterned by sequential action of feedforward and feedback loops. Development. 2008;135:343–351. [PubMed]
  • Yamada T, et al. EDL/MAE regulates EGF-mediated induction by antagonizing Ets transcription factor Pointed. Development. 2003;130:4085–4096. [PubMed]
  • Zhou X, et al. Overexpression of broad: a new insight into its role in the Drosophila prothoracic gland cells. J Exp Biol. 2004;207:1151–1161. [PubMed]