The hernandez and fernandez genes of Drosophila specify eye and antenna

Available online at www.sciencedirect.com R

Developmental Biology 260 (2003) 465– 483

www.elsevier.com/locate/ydbio

The hernandez and fernandez genes of Drosophila specify eye and antenna Magali Suzanne, Carlos Estella, Manuel Calleja, and Ernesto Sa´nchez-Herrero* Centro de Biologı´a Molecular Severo Ochoa, Universidad Auto´noma de Madrid, Cantoblanco, 28049 Madrid, Spain Received for publication 17 December 2002, revised 8 April 2003, accepted 16 April 2003

Abstract The formation of different structures in Drosophila depends on the combined activities of selector genes and signaling pathways. For instance, the antenna requires the selector gene homothorax, which distinguishes between the leg and the antenna and can specify distal antenna if expressed ectopically. Similarly, the eye is formed by a group of “eye-specifying” genes, among them eyeless, which can direct eye development ectopically. We report here the characterization of the hernandez and fernandez genes, expressed in the antennal and eye primordia of the eye–antenna imaginal disc. The predicted proteins encoded by these two genes have 27% common amino acids and include a Pipsqueak domain. Reduced expression of either hernandez or fernandez mildly affects antenna and eye development, while the inactivation of both genes partially transforms distal antenna into leg. Ectopic expression of either of the two genes results in two different phenotypes: it can form distal antenna, activating genes like homothorax, spineless, and spalt, and it can promote eye development and activates eyeless. Reciprocally, eyeless can induce hernandez and fernandez expression, and homothorax and spineless can activate both hernandez and fernandez when ectopically expressed. The formation of eye by these genes seems to require Notch signaling, since the induction of ectopic eyes and the activation of eyeless by the hernandez gene are suppressed when the Notch function is compromised. Our results show that the hernandez and fernandez genes are required for antennal and eye development and are also able to specify eye or antenna ectopically. © 2003 Elsevier Inc. All rights reserved. Keywords: Drosophila; Antenna; Eye; eyeless; homothorax; spineless; Psq motif

Introduction One important issue in developmental biology is to understand how specific structures are determined. In Drosophila melanogaster, this is accomplished in part through the activity of a group of genes called selector or selectorlike genes, which confer identity to structures as diverse as legs, eyes, or wings (Mann and Morata, 2000). Hox (homeotic) genes, which specify structures along the anterior–posterior axis of most animals (Gellon and McGinnis, 1998), are also commonly referred to as selector genes (Garcı´a-Bellido, 1975). The functional distinction between Hox and non-Hox selector genes is not straightforward. In general, Hox genes can differentiate between struc* Corresponding author. Fax: ⫹34-913974799. E-mail address: [email protected] (E. Sa´nchez-Herrero). 0012-1606/03/$ – see front matter © 2003 Elsevier Inc. All rights reserved. doi:10.1016/S0012-1606(03)00249-5

tures considered homologous (like wings and halteres), whereas non-Hox selector genes can specify structures in regions of the fly in which the homology with the new, ectopic organ, is not evident. Thus, Distal-less (Dll), a non-Hox selector gene, induces leg tissue when expressed in the wing (Gorfinkiel et al., 1997; Campbell and Tomlinson, 1998), and the eyeless (ey) gene makes ectopic eye tissue if present in antennae, legs, or proximal wing (Halder et al., 1995). There are structures in Drosophila which do not require Hox information and that can be formed ectopically by different non-Hox selector genes. These structures include two head organs, the eye and the antenna, derived from the eye–antenna imaginal disc. The antennae are ventral structures homologous to legs (Postlethwait and Schneiderman, 1971). Both types of appendages share common positional information determined by the activity of the hedgehog, decapentaplegic, and wing-

466

M. Suzanne et al. / Developmental Biology 260 (2003) 465– 483

less (wg) products (Campbell and Tomlinson, 1995; Brook et al., 1996). These molecules determine the distribution of the homothorax (hth), Dll, and dachsund (dac) genes, which genetically subdivide both appendages along the proximal– distal axis (Dı´az-Benjumea et al., 1994; Lecuit and Cohen, 1997; Gorfinkiel et al., 1997; Gonza´ lez-Crespo et al., 1998; Abu-Shaar and Mann, 1998; Wu and Cohen, 1999; Dong et al., 2001; reviewed in Morata, 2001). The difference between antennae and legs relies in the distinct spatial distribution of the Hox gene Antennapedia (Struhl, 1981) and of hth, the selector gene for the antenna. hth and Dll form an antenna in its wild-type location or in other positions if coexpressed (Casares and Mann, 1998; Dong et al., 2000). The spineless (ss) gene is also able to specify distal antennal development (Duncan et al., 1998), but it seems to mediate Dll and hth function (Duncan et al., 1998; Dong et al., 2002). In the leg disc, where Dll is also expressed, the Antennapedia (Antp) Hox gene prevents hth expression, thus directing this primordium toward leg development (Casares and Mann, 1998). Although several genes are differently expressed in the antennal primordium and the leg disc (Dong et al., 2001), only one gene, spalt (sal), is transcribed in the antenna and not in the leg (Wagner-Bernholz et al., 1991). However, sal is also expressed in other discs (de Celis et al., 1996; Barrio et al., 1999) and no gene expressed exclusively in the antennal primordium has yet been described. The Drosophila eye is a complex organ derived from the eye primordium of the eye–antenna imaginal disc. The ey gene was first identified as a “master” or selector gene for eye development, since it is required for the formation of the eye and can induce eye development when ectopically expressed (Halder et al., 1995). Subsequent studies, however, identified other genes, such as twin of eyeless, eyegone, eyes absent, dachsund, and sine oculis, with similar characteristics. Most of them are involved in complex regulatory networks and protein–protein interactions that determine eye development (reviewed in Treisman, 1999; Kumar, 2001). Interestingly, ectopic activation of the Notch (N) gene, encoding a receptor involved in many developmental decisions (Artavanis-Tsakonas et al., 1999), induces ectopic eyes in a wild-type head and transforms the eye into antenna in an ey mutant background (Kurata et al., 2000). Moreover, recent work suggests that the N and Epidermal growth factor Receptor (Egfr) signaling pathways are involved in the distinction between eye and antenna early in development (Kumar and Moses, 2001). These results suggest that there is not a single selector gene for the determination of the eye and that signaling pathways are somehow involved in the regulatory loops that specify this structure. To identify genes involved in antennal development, we have used the UAS-yellow method (Calleja et al., 1996) to select GAL4 lines showing restricted expression patterns either in the leg or the antenna. We have characterized four of these lines, which are inserted in two related genes, that we called hernandez (hern) and fernandez (fer) (Spanish

names for the twin brothers in Tintin comic-books). A recent report (Emerald et al., 2003) has characterized these same genes and given the names of distal antenna (dan) to the fernandez gene and distal antenna related (danr) to the hernandez one. These genes are expressed in the eye–antennal discs and not in the leg or other imaginal discs. Reduced hern or fer expression leads to small defects in antennal and eye development, and when the activity of both genes is compromised, there is a partial transformation of the distal antenna into leg. When either hern or fer is expressed ectopically, they activate hth and ss expression and transform leg tissue into distal antenna. These genes can also make ectopic eye tissue and activate ey. Our results suggest that the hern and fer genes are involved in antenna and eye specification.

Materials and methods Isolation of GAL4 lines and genetic strains The method used to isolate the GAL4 lines has been previously described (Calleja et al., 1996). The UAS-lacZ (Brand and Perrimon, 1993), UAS-y (Calleja et al., 1996), UAS-GFP (Ito et al., 1997), UAS-Necd (Klein et al., 1997), Antp73b (Hazelrigg and Kaufman, 1983), UAS-hth (Pai et al., 1998), Dll3 (Cohen and Ju¨ rgens, 1989), UAS-ENHTH1-430 (Inbal et al., 2001), ssD115.7, UAS-ss (Duncan et al., 1998), and hs-hth (Pai et al., 1998) stocks have been previously described. Ectopic expression was obtained by using the GAL4/UAS system (Brand and Perrimon, 1993) with the following GAL4 lines: dpp-GAL4 (StaehlingHampton et al., 1994), Dll-GAL4 (MD23 and EM212 lines; Calleja et al., 1996), GMR-GAL4 (Yamaguchi et al., 1999), and ptc-GAL4 (Hinz et al., 1994). Df slo8 is a deficiency removing the genomic region where the hern and fer genes are located (Atkinson et al., 1991). Isolation of derivatives from the P-element insertions The MD634 and CES115 P-GAL4 lines were mobilized to obtain imprecise excisions of the P-element. yw; MD634 (or CES115)/TM3, ⌬ 2-3 males were individually crossed to yw; TM3/TM6B females, and individual w⫺ males were isolated and used to establish yw; MD634* (or CES115*)/ TM6B stocks (the asterisk marks a possible imprecise excision). The phenotype of MD634* or CES115* homozygotes from the stocks was analyzed. Clonal analysis Ectopic expression clones (Pignoni and Zipurski, 1997) were induced by giving a 10-min heat shock at 34°C to y hs-flp; act ⬎y⫹⬎ GAL4 UAS-GFP/ UAS-fer or UAS-hern second instar larvae. The act ⬎y⫹⬎ GAL4 UAS-GFP (AyGAL4.25 UAS-GFP S65T) stock has been previously de-

M. Suzanne et al. / Developmental Biology 260 (2003) 465– 483

467

Fig. 1. Expression of the hern and fer genes and of the P-GAL4 elements inserted in them. (A) Antenna of a y fly. II and III stand for second and third antennal segments, respectively; ar, arista. (B, C) Antennae of y; MD634/UAS-y⫹ (B) and y; AC116/UAS-y⫹ (C) flies. The MD634 insertion is located close to the hern gene and the AC116 insertion close to the fer gene. Note that the third antennal segment, the arista, and a few bristles of the second antennal segment are y⫹ in both cases. Symbols as in (A). (D, E) lacZ expression in third instar eye–antennal imaginal discs of MD634/UAS-lacZ (D) and AC116/UAS-lacZ (E) larvae, showing stronger signal in a ring at the position of the third antennal segment and in the central region of the antennal primordium (a), with weaker lacZ levels in between. There is also expression in the eye primordium, mostly in its posterior region. The morphogenetic furrow is indicated by an arrow. In this and following figures, the dorsal part of the antennal disc is up. (F–H) RNA expression of the hern (F) and fer (G) genes in the eye–antennal disc, resembling the expression driven by the GAL4 lines in the antennal primordium (compare with D and E, respectively); in the eye primordium, strong signal is observed anterior to the morphogenetic furrow (indicated by an arrow), while the posterior, differentiated eye (posterior to the morphogenetic furrow) only shows weak expression. Detail of the signal in the eye is shown in (H). A and P stand for regions anterior and posterior to the morphogenetic furrow (arrow), respectively.

M. Suzanne et al. / Developmental Biology 260 (2003) 465– 483

scribed (Ito et al., 1997). The clones were identified by the GFP expression. Molecular techniques, pUAS constructs The P-Gal4 lines were localized by inverse PCR (http:// www.fruitfly.org/about/methods/inverse.pcr.html). The remaining molecular techniques were done according to standard protocols (Sambrook et al., 1989). The molecular structure of the genes and the reconstruction of the open reading frame (ORF) were done according to the annotation of the Drosophila genome (Adams et al., 2000). The pUAST-fer construct was done by cloning first a partially digested EcoRI/XhoI 3.1-kb fragment from the LD40883 EST into the Bluescript vector; this new vector was then digested with XhoI and NotI and inserted into the pUAST vector. To make the pUAST-hern construct, we first amplified a 1.4 kb genomic fragment by PCR (Unit Expand Long Template PCR system; Roche) with the 5⬘-ttcgaggagtgcacaccggtggtc-3⬘ upper primer and the 5⬘-ctactgttcggcgtcgctggacccg-3⬘ lower primer, and cloned it in a modified Bluescript vector, digested with EcoRV and then treated with Taq Polymerase (Unit Expand Long Template PCR System; Roche) to add dTTP in its 3⬘ end; then, we cloned a XbaI– XhoI fragment from this construct in the pUAST vector. PCR analysis of the derivatives To analyze the ferI49-1 deletion, we amplified by PCR a genomic fragment in the region of the P-element insertion with 5⬘-ggcggctggcaaggagctgcag-3⬘ and 5⬘-gccttcttccagcgaaggcggg-3⬘ primers, which hybridize to sequences located at both sides of the insertion. In wild-type flies, the fragment amplified was about 3.5 kb long, and in ferI49-1 homozygous flies, about 1 kb long. RNA interference The GAL4-inducible constructs for RNA interference were made as follows: for the hern gene, a 524-bp fragment from the hern exon was amplified by PCR with the 5⬘tgaggatccggatgagcacgcgcggc-3⬘ upper primer and the 5⬘gcaggtaccaggacatcggcagtctgtggg-3⬘ lower primer; for the

469

fer gene, a 605-bp fragment from the second exon of the fer transcript was amplified by PCR with the 5⬘-acaggatccatgccatccagaggatccacg-3⬘ upper primer and the 5⬘ cgcggtaccgggattcagctgggccagc-3⬘ lower primer. Each PCR product was independently cloned as a BamHI–KpnI fragment in the pHIBS vector (Nagel et al., 2002), to make the pHIBS-hern or pHIBS-fer constructs. The BamHI–SacI fragments from pHIBS-hern or pHIBS-fer were subcloned in the Bluescript vector, generating the BS-INT-hern and BS-INT-fer constructs, respectively. Finally, SalI–KpnI fragments from pHIBS-hern and BS-INT-hern (or pHIBS-fer and BS-INTfer) constructs were cloned together in the pUAS vector at its KpnI site, thus forming the final RNAi constructs, pUASihern (ihern) and pUAS-ifer (ifer). The RNAi constructs (and the UAS-constructs) were injected into y w1118 embryos by standard procedures. In situ hybridization In situ hybridization was performed according to Azpiazu and Frasch (1993). The DNA probes of the hern and fer genes were synthesized by random priming with the Boehringer Mannheim kit, using as templates a 1361-bp fragment from the coding region of the hern gene and a 1192-bp fragment from the second exon of the fer gene, both PCR-amplified with the following primers: 5⬘-ctgttcggcgtcgctggaccc-3⬘ and 5⬘-gtgcacaccggtggtcgcatccgc-3⬘ for the hern gene, and 5⬘-cgccaagctgttcgacaacggcct-3⬘ and 5⬘gcccgacggtgtggacgagggtgt-3⬘ for the fer gene. The hern RNA probe, used to detect inactivation of hern by RNAi, was synthesized from a hern cDNA cloned in the Bluescript vector. This construct was digested with BamHI so that the probe was specific for the 3⬘ end of the gene (not included in ihern). The ss probe was synthesized from a ss cDNA cloned in the Bluescript vector and provided by Ian Duncan. Immunostaining and histochemical staining of Drosophila imaginal discs The antibody staining was done according to Azpiazu and Morata (2000). The antibodies used are: guinea pig anti-Hth (Azpiazu and Morata, 2002), mouse anti-Dac (mAbdac2-3; Hybridoma Bank, The University of Iowa),

Fig. 2. Location of the P-GAL4 elements, molecular characterization of the ferI49-1 mutation, molecular structure and sequence of the hern and fer genes, and comparison of the Psq domains of the proteins encoded by them. (A) DNA map showing the DNA region containing the hern/danr and fer/dan transcription units, their transcript structure according to the annotation of the Drosophila genomic sequence (Adams et al., 2000), and the location of the P-GAL4 transposable elements. Solid red boxes represent exons, the lines joining them indicate the introns, and the arrows indicate the direction of transcription. There are two predicted transcription units between the hern and fer transcription units (CG13652 and CG13661), shown as small black squares which represent predicted exons. The distance between the insertion points of two GAL4 lines and the predicted transcription start site of the two genes is also indicated. The extension of the ferI49-1 deletion is shown at the top of the figure, and the ihern and ifer constructs used for RNA interference at the bottom. (B) Aligned protein sequence of the two genes obtained from the conceptual translation of the hern and fer transcription units. Residues in blue indicate similar amino acids, and those in red are identical ones. The open box encloses a highly conserved 69-amino-acid domain, and the 48-amino-acid Pipsqueak (Psq) motif is underlined. (C) Alignment of the Psq motifs of the hern and fer genes, as well as that of a third predicted gene (CG13496 transcription unit), which share a high similarity in this domain with the other two (see Siegmund and Lehmann, 2002). Color of the amino acids indicates similarity and identity as in (B).

470

M. Suzanne et al. / Developmental Biology 260 (2003) 465– 483

M. Suzanne et al. / Developmental Biology 260 (2003) 465– 483

mouse anti-Dll (Duncan et al., 1998), rabbit anti-Sal (Ku¨ hnlein et al., 1994), mouse anti-Wg (Brook and Cohen, 1996), rat anti-Elav (Elav-9F8A9; Hybridoma Bank, The University of Iowa), mouse anti-Dan (Emerald et al., 2003), rabbit anti-Ey (U. Walldorf, unpublished), and rabbit anti-␤-galactosidase (Cappel). Secondary antibodies are coupled to Red-X, FITC, and Cy5 fluorochromes (Jackson ImmunoResearch) The staining was analyzed in a Zeiss confocal microscope. Histochemical staining to detect ␤-galactosidase expression was done according to Ashburner (1989). Heat-shock experiments. hs-hth; MD634/UAS-GFP or hs-hth; CES115/UAS-GFP third instar larvae were given a 15-min heat shock at 37°C in a water bath. Discs were dissected out of the larvae and fixed 1 h after the heat-shock.

Results The hern and fer genes are similarly expressed in the antenna and eye primordia From a collection of P-GAL4 lines that drive expression in restricted areas of the adult cuticle (Calleja et al., 1996), we selected four lines which specifically direct a similar expression of the yellow (y) gene: the third antennal segment, the arista, and a few bristles in the second antennal segment express the y gene (Fig. 1A–C). These lines, when crossed to a line carrying an UAS-lacZ reporter construct, show ␤-galactosidase expression in three regions of the eye–antennal disc: in the third antennal segment, in the arista, and within the eye, in the differentiated eye and in a region just anterior to the morphogenetic furrow (Fig. 1D and E). We located by inverse PCR the insertion site of these four lines (Fig. 2A): two of them (CES75 and MD634) are in the 5⬘ upstream region of the predicted CG13651 transcription unit and the other two (AC116 and CES115) are in the 5⬘ upstream region of the CG11849 transcription unit, according to the annotated Drosophila genome (Adams et

471

al., 2000). The two transcription units are located in the chromosomal position 96C2-4, in the same orientation and near adjacent, separated by about 45 kb. The CG13651 transcription unit is 1.259 kb long, bears no introns, and produces a 1259-bp mRNA, whereas the CG11849 transcription unit is 7.175 kb long, has 3 exons, and makes a mRNA of 2232 bp (Fig. 2A). There is an EST for the CG11849 unit, indicating that this predicted gene is transcribed. The CG13651 and CG11849 transcription units encode predicted proteins of 419 and 743 amino acids, respectively, that have 27% identity and 37% similarity between them (Fig. 2B). Both proteins contain similar 69-amino-acid regions, located in their N-terminal part, with 78% identity and 88% similarity between them (Fig. 2B). This region includes a previously characterized 48-amino-acid Pipsqueak (Psq) motif, present in several proteins in Drosophila and other species, including human (Horowitz and Berg, 1996; Lehmann et al., 1998; Siegmund and Lehmann, 2002). The Psq motif is a helix–turn– helix DNA-binding domain similar to the DNA-binding domain of prokaryotic recombinases (Lehmann et al., 1998), which, in turn, also show similarities with the homeodomain (Affolter et al., 1991). The CG13496 predicted Drosophila gene codes for a protein with a Psq domain very similar to those of CG13651 and CG11849 genes (Siegmund and Lehmann, 2002) (Fig. 2C). The CG13496 transcription unit is not expressed in the eye–antennal imaginal disc like the CG13651 and CG11849 genes (data not shown). In situ hybridization experiments show that, among all the imaginal discs, the CG13651 and CG11849 transcription units are only and equally expressed in the eye–antennal disc. Their expression is very similar to that observed with the P-GAL4 lines (Fig. 1F and G, compare with Fig. 1D and E), except that, in the eye primordium, the RNA signal in the differentiated eye is very weak (Fig. 1F–H). The same antennal and eye expression persists in early pupal stages (not shown). Therefore, these genes present a similar sequence, expression in the eye–antennal disc, regulation, and function (see below). We have called these genes hernandez (hern, CG13651) and fernandez (fer, CG11849), and the two of them, the Tintin genes.

Fig. 3. Comparison of MD634-GAL4 expression with those of the hth, Dll, sal, and dac genes in the antennal primordium, and regulation of MD634-GAL4 expression by Antp. (A) GFP (green) expression in the antennal primordium of a MD634 UAS-GFP larva. The strong MD634 expression overlaps with that of hth (B, red) in the proximal part of the third antennal segment. Merged image in (C). (D) Expression of Dll (D, blue) in the same antennal primordium. (E) Expression of Dll (blue) and MD634 (green), showing overlap in the presumptive third antennal segment and the arista. (F) Merged image of GFP, hth and Dll expression. (G) GFP expression (green) in an antennal disc of a MD634 UAS-GFP larva. This disc is from an older larva than that shown in (A). (H) sal (red) and (J) dac (blue) expression in the same disc. (I, K, L) Merged images of MD634 UAS-GFP and sal (I), MD634 UAS-GFP and dac (K), and MD634 UAS-GFP, sal, and dac (L). Note that, in the third antennal segment, the distal limit of strong GFP expression in the third antennal segment coincides with that of sal (I) and the proximal one with that of dac (K). (M) Scheme of the expression of the hern and fer transcription units in relation to Dll, hth, dac, and sal, after Dong et al. (2001). a1, a2, and a3 refer to the first, second, and third antennal segments, respectively. a4 and a5 represent the base of the arista, and ar stands for arista. The light pink color indicates weak hern and fer expression. It is possible that hern and fer expression extends into a few cells of a2. (N) Antennae of an Antp73b/⫹ adult, transformed into legs except in the most proximal and more distal regions (arrows, aristae). (O) GFP (green) expression in an Antp73b/MD634 UAS-GFP larva. Note the patchy green expression, which is almost absent in the primordium of the third antennal segment (arrow; compare with A). The expression in the center of the primordium (arrowhead) is almost unchanged, which corresponds with the normal aristae in (N).

472

M. Suzanne et al. / Developmental Biology 260 (2003) 465– 483

Fig. 4. Phenotype of the AC116 and ferI49-1 mutations and of the GAL4-driven ihern and ifer constructs in the antenna. (A) Wild-type antenna. (B) Antenna of a AC116/Dfslo8 fly. Note the presence of ectopic bristles in the third antennal segment (III). (C) Antenna of a ferI49-1/Dfslo8 fly, showing a similar phenotype to that shown in (B). (D–F). In situ hybridization showing RNA transcription of the fer gene in the third instar eye–antennal discs of wild-type (D), AC116/AC116 (E), and ferI49-1/ferI49-1 (F) larvae. Note the reduced RNA expression in the antennal primordium of the mutants (arrow), whereas the levels of RNA signal in the eye are normal. (G–I) Antennal phenotype of Dll-GAL4/⫹ (G), Dll-Gal4/⫹; ifer/⫹ (H), and Dll-GAL4/⫹; ihern/⫹ (I) flies. (G⬘–I⬘) Insets showing the base of the arista in detail. Note the presence of ectopic bristles in the third antennal segment of (H) and (I) (arrows) and the enlarged base of the arista (H⬘, I⬘) in the antennae carrying RNAi constructs. (J) Phenotype of a Dll-GAL4/⫹; ihern/ifer fly. There is a clear transformation of the distal antenna into leg structures, including many bracted bristles (arrowheads and inset). (K) Wild-type expression pattern of the Dan (Fer) protein in the eye–antenna disc detect with an anti-Dan antibody (Emerald et al., 2003). (L) Expression of Dan (Fer) protein in a Dll-Gal4/⫹; ifer/⫹ eye-antenna disc. Note the absence of Fer protein in the antennal primordium (a).

M. Suzanne et al. / Developmental Biology 260 (2003) 465– 483

hern and fer expression patterns in the antennal primordium To delimit more precisely the expression of the hern and fer genes in the antennal primordium, we carried out double staining of the MD634 and CES115 GAL4 lines (revealing hern and fer transcription, respectively) with genes expressed in restricted areas of mature eye–antennal discs, like Dll, hth, sal, and dac (Wagner-Bernholz et al., 1991; Dı´azBenjumea et al., 1994; Mardon et al., 1994; Gorfinkiel et al., 1997; Rieckhof et al., 1997; Pai et al., 1998; Barrio et al., 1999; Dong et al., 2001). Both lines gave the same results. Within the third antennal segment, the hern and fer expression is included within the Dll and hth domains (Fig. 3A–F). The hern and fer proximal limit of expression seems to coincide with that of dac, and their distal limit of strong expression in the third antennal segment with that of sal (Fig. 3G–L). We cannot exclude that the GAL4 lines drive expression in a few cells from the second antennal segment. A scheme of the results obtained is shown in Fig. 3M. Since hern and fer are not expressed in the leg discs, they may be part of the mechanism that distinguishes legs from antennae. The Antennapedia (Antp) Hox gene, which is expressed in the leg but not in the antennal discs (Wirz et al., 1986; Condie et al., 1991), prevents hth expression and antenna formation in the leg primordia (Struhl, 1981; Casares and Mann, 1998). As expected, in Antp73b/⫹ flies, which show a strong transformation of antennae into legs (Hazelrigg and Kaufman, 1983; Fig. 3N), the expression driven by the MD634 GAL4 line is clearly reduced (Fig. 3O, compare with Fig. 3A or G). Inactivation of the hern and fer genes partially transforms distal antenna into leg To characterize the function of hern and fer in normal development, we studied the phenotype of flies without hern or fer activity. One of the P-GAL4 lines, AC116 (see Fig. 2), is mutant for the antennal function of the fer gene. Homozygous AC116 third instar larvae show no fer transcription in the antennal primordium but present normal fer expression in the eye primordium (Fig. 4E, compare with the wild-type expression in Fig. 4D). AC116/Df adults show one or more bristles in the third antennal segment, normally devoid of them (Fig. 4B, compare with the wild-type antenna shown in Fig. 4A). To obtain more mutations in the hern and fer genes, we mobilized the MD634 and CES115 P-GAL4 lines (the two closer to the hern and fer transcription units) to isolate imprecise excisions of the transposons. We isolated one w⫺ derivative of the CES115 insertion (102 analyzed), ferI49-1, which, in hemizygosis, shows a phenotype very similar to that described for the AC116/Df adults (Fig. 4C). PCR analysis of the mutation revealed that it is a small deletion of about 2.5 kb in the 5⬘ upstream region of the fer transcription unit (Fig. 2A). Larvae homozygous for the ferI49-1 mutation present reduced expression of the fer

473

RNA in the antennal primordium of mature eye–antennal discs, whereas the expression in the eye primordium is normal (Fig. 4F). We conclude that the AC116 insertion and the ferI49-1 deletion may have affected an antennal regulatory region. This regulatory region would control only the fer gene, since in AC116 and ferI49-1 homozygous larvae, we cannot detect any change in hern expression. We have not obtained any mutation for the hern gene (279 w⫺ derivatives analyzed), although the MD634 line used for this purpose is closer to the origin of transcription of the hern gene than the CES115 line is to the transcription unit of the fer gene (see Fig. 2). The weak phenotype of the AC116 and ferI49-1 mutations suggested that there may be a partial functional redundancy between hern and fer, and that the inactivation of both genes may result in a stronger phenotype. To check this, to ascertain the phenotype of hern inactivation, and to study the effect of the absence of Tintin products in the eye, we used double-stranded (ds)-mediated RNA interference (RNAi; Fire et al., 1998; Kennerdell and Carthew, 1998; Sharp, 2001) to inactivate hern and fer functions. The inactivation in the antennal primordium of either the hern or the fer genes with a Dll-GAL4 (MD23) driver causes a phenotype similar to that described for the AC116/Df and ferI49-1/Df adults: there is one or more bristles in the third antennal segment and the base of the arista is slightly enlarged (Fig. 4H and I, compare with the wild-type in Fig. 4G). However, when the ds-hern and ds-fer RNAs are induced together by the Dll-GAL4 driver, a clear transformation of part of the distal antenna into leg is observed: the third antennal segment and the proximal arista are substantially enlarged and covered with bristles; those in the base of the arista bear bracts, indicating a transformation into leg (Fig. 4J). We checked that the RNAi technique is working by doing in situ hybridization with a hern probe not included in ihern or by staining eye–antennal imaginal discs with a specific antibody against the Fer product (Dan protein; Emerald and Cohen, 2003). hern RNA expression is strongly reduced or eliminated in dpp-GAL4/⫹; ihern/⫹ antennal discs (not shown) and no Fer protein is detected in Dll-Gal4/⫹; ifer/⫹ antennal primordia (Fig. 4K and L). These results indicate that both hern and fer are required to develop part of the antenna as opposed to leg and that these two genes are partially redundant in this function. The hern and fer genes produce antennal tissue and activate hth, ss, and sal when ectopically expressed We have shown that hern and fer are specifically expressed in the eye–antennal disc and that both genes are needed for normal antennal development. To test whether those genes are sufficient to induce eye or antennal development, we expressed them ectopically using the GAL4/ UAS system (Brand and Perrimon, 1993). When either the hern or the fer genes are misexpressed in the leg discs with dpp-GAL4 or Dll-GAL4 (EM212) drivers, distal legs are

474

M. Suzanne et al. / Developmental Biology 260 (2003) 465– 483

M. Suzanne et al. / Developmental Biology 260 (2003) 465– 483

475

Fig. 6. Hth and ss activate the hern and fer genes ectopically. (A, B) Leg disc of a MD634 UAS-GFP/ hs-hth larva, after heat shock. There are scattered cells showing ectopic hth expression (in red, A) and also ectopic activation of GFP expression in small patches (in green, B). (C) dpp-GAL4/UAS-ss leg disc, showing ectopic expression of the fer transcript.

transformed to aristae (Fig. 5A–C). These transformations are accompanied by the ectopic expression of hth, sal, and ss, three genes expressed in the antennal primordium but not in the distal region of mature wild-type leg disc (Wagner-Bernholz et al., 1991; Rieckhof et al., 1997; Pai et al., 1998; Casares and Mann, 1998; Duncan et al., 1998; Dong et al., 2002). Clones expressing either the hern or the fer genes in the leg or wing disc have smooth borders and frequently activate the sal and hth genes cell-autonomously (Fig. 5D–I), but do not affect Dll expression (not shown). In dpp-GAL4/UAS-fer or ptc-GAL4/UAS-hern leg (or wing) discs, the expression of ss is also activated (Fig. 5K, compare with the wildtype ss expression in Fig. 5J). Curiously, although ss is downstream of hth in the antenna and leg (Dong et al., 2002), ectopic ss in the leg disc can also activate hth in a few cells (Fig. 5L). Dll, hth, and ss are required for hern and fer expression in the antenna The hth or ss genes, together with Dll, are sufficient to develop ectopic distal antennae when expressed in different regions of the adult (Casares and Mann, 1998; Duncan et al., 1998; Dong et al., 2000). We have shown above that the hern or fer genes are also able to elicit this transformation in the leg and that they activate hth and ss. Conversely, when we induce high levels of the Hth or Ss

products in the leg discs, we find ectopic expression of the hern and fer genes (Fig. 6A–C; and data not shown). To study the interactions between these genes in normal development, we analyzed the relationship between Dll, hth, ss, and hern/fer in the antennal primordium. A reduction of Hth activity using a dominant negative form of hth (UAS-EN-HTH1-430; Inbal et al., 2001) results in a decreased activity of the MD634 and AC116 GAL4 lines, which reveal hern and fer expression, respectively (Fig. 7B; and data not shown, compare with the wild-type expression in Fig. 7A). Similarly, in antennal discs of a Dll strong hypomorph or a ss null mutation, the expression of hern and fer disappears (Fig. 7C and D; and data not shown, compare with the wild-type expression in Fig. 1G or 4D). These results suggests that hth, Dll, and ss are required to maintain hern and fer expression in the antenna. By contrast, high levels of hern or fer may reduce hth expression. In dpp-GAL4/UAS-fer or dpp-GAL4/ UAS-hern larvae, the expression of hth (and sal) in the third antennal segment is eliminated or strongly reduced dorsally (where levels of hern and fer are high) and does not change or is ectopically activated ventrally (where levels of hern and fer are low; Fig. 7E and F; and data not shown). Similarly, fer-expressing clones are able to downregulate hth expression in the antennal primordium (Fig. 7G and H). These results suggest that levels of hern and fer expression may be important for a normal antennal development.

Fig. 5. The ectopic expression of either the hern or fer genes transforms leg into antenna and activates the ss, hth, and sal genes. (A) Distal legs of a Dll-GAL4 (EM212)/UAS-hern fly, showing transformation to aristae (arrows); t, tibia. (B, C) Transformation of distal legs to aristae (arrows in B) in a dpp-GAL4/ UAS-fer fly. A detail of an ectopic arista (ar) is shown in (C), close to a claw (c). (D–F) Clone expressing the hern gene in the leg disc, marked by the GFP marker (D, in green), which activates ectopically hth (red in E). Merged image in (F). (G–I) A similar clone, marked with GFP (G), also activates cell-autonomously the gene sal (red in H). Merged image in (I). (J) Wild-type third instar leg disc hybridized with a ss probe, showing almost no expression in the central region (Duncan et al., 1998). (K) dpp-GAL4/UAS-fer leg disc, showing abnormal development and ectopic ss transcription (arrow). (L) dpp-GAL4/UAS-ss leg disc. A few cells of the dorsal part of the disc activate hth ectopically (arrow).

476

M. Suzanne et al. / Developmental Biology 260 (2003) 465– 483

Fig. 7. Regulatory activations between Dll, ss, hth, and hern/fer genes in the antennal disc. (A) Antennal primordium of a MD634 UAS-GFP larva, showing GFP expression (in green). (B) Similar antennal disc of a MD634 UAS-GFP/UAS-EN-HTH1-430 larva. The GFP signal is now much reduced compared with that in (A). (C) Dll3 antennal disc hybridized with a fer probe, showing there is no fer transcription in the antennal primordium (a), but normal expression in the eye. (D) Antennal primordium in a ssD115.7 mutant larva. The expression of fer in the antenna (a) also disappears. (E, F) Antennal primordium of a dpp-GAL4/UAS-fer third instar larva, showing ectopic activation of hth (red in E) and sal (green in F) in the ventral distal part of the third antennal segment (arrowheads) and repression of both genes in the dorsal part (arrows). (G, H) Antennal primordium of a third instar imaginal disc carrying clones marked with GFP (in green in H, arrows) which ectopically activate the fer gene. These clones cell-autonomously repress hth expression (arrows; hth marked in red in G and H).

The hern and fer genes are required for normal eye development and form eye tissue and activate ey when ectopically expressed To study the possible role of the hern and fer genes in eye development, we looked to the eye phenotype when

either the hern or fer genes are inactivated by RNAi and also express them ectopically. Expression of ds-hern or ds-fer RNA in the eye primordium with a GMR-GAL4 driver causes a slightly rough eye, with some bristles irregularly positioned (Fig. 8A–D). Curiously, the phenotype is not increased if the ds-hern and ds-fer RNAs are induced in the same fly. Misexpression experiments also suggest that both hern and fer are involved in eye development. Thus, the expression of either hern or fer with different GAL4 drivers causes the appearance of ectopic eye tissue in the third antennal segment or rostral membrane (Fig. 8E–G). These transformations are accompanied by the ectopic expression of ey (Fig. 8H and I), although this effect may also indicate the maintenance of a previous ey expression. Conversely, the misexpression of ey activates the hern and fer genes ectopically (Fig. 8J; and data not shown). Both hern and fer also activate embryonic lethal abnormal vision (elav), a marker of neuronal differentiation (Robinow et al., 1988), when ectopically expressed (Fig. 8I; and data not shown). The analysis of clones expressing the fer gene in the leg, eye–antennal, or wing discs shows that elav activation is strictly nonautonomous, and only occurs in some cells adjacent to some of these clones (Fig. 8K–N). The formation of the morphogenetic furrow in the eye is limited laterally by wg signaling (Ma and Moses, 1995; Treisman and Rubin, 1995). We have observed that hern and fer expression within the eye primordium includes the more lateral wg-expressing regions (Fig. 9A). Interestingly, we find that both hern and fer activate wg transcription when ectopically expressed. In ptc-GAL4/UAS-hern or dpp-GAL4/UAS-fer flies, the wings show several alterations, including the appearance of marginal bristles in the middle of the wing blade (Fig. 9B and C). This phenotype is characteristic of ectopic wg signaling (Dı´az-Benjumea and Cohen, 1995), and in fact, wg is ectopically expressed in the wing discs of these larvae (Fig. 9D; and data not shown). Clones expressing the fer genes in the eye–antenna, leg, or wing discs also show induction of wg, mostly within but also outside the clone (Fig. 9E, F, and H). We have previously described that the elav gene is also induced nonautonomously by these clones. We note, however, that cells ectopically expressing elav do not coincide with those expressing wg (Fig. 9G and H) and that this reproduces the wild-type situation in the eye. Interaction of hern and fer genes with N signaling Signaling pathways can modify the activity of selector genes and are needed for proper organ formation. N signaling, for instance, is needed for eye formation (Go et al., 1998; Kurata et al., 2000; Kumar and Moses, 2001) and can activate ey when ectopically activated (Kurata et al., 2000). Moreover, N has been recently implicated in the decision of making eye or antenna, directing eye development, and suppressing antenna formation (Kurata et al., 2000; Kumar and Moses, 2001). Therefore, we investigated whether N

M. Suzanne et al. / Developmental Biology 260 (2003) 465– 483

signaling could alter the ey and elav expression induced by the Tintin genes. As shown in Fig. 9I–K, the coexpression of the hern gene and a dominant negative form of the Notch receptor (Klein et al., 1997) substantially reduces ey and eliminates elav ectopic signals (compare with Fig. 8H and I). Accordingly, no ectopic eyes are formed in this genetic combination. This indicates that the effect of hern on ey expression and eye formation requires N signaling.

Discussion We have isolated and characterized two adjacent genes, hern and fer, which share several characteristics: first, they encode similar protein sequences, including a Psq motif; second, they are expressed in the eye–antennal imaginal discs, and not in other discs, in a very similar or identical pattern; third, they are needed for normal antenna and eye development; and fourth, both genes can make eyes and antennae when ectopically expressed. Our genetic analysis indicates that the specific role of hern and fer is to contribute to antenna and eye development and that hern and fer genes are partially redundant, at least as to their function in the antenna. The role of these same genes in the antenna has been recently characterized (Emerald et al., 2003). These authors call these genes distal antenna (dan; the fer gene) and distal antenna-related (danr; the hern gene; see Fig. 2) and they have reached similar conclusions. Particularly, a deficiency for both genes results in an antennal transformation similar to what we see by inactivating both hern and fer genes by RNAi (Emerald et al., 2003). The Psq motif is present in at least 14 Drosophila proteins, including those encoded by the pipsqueak, Bric a` brac I and II, Piefke, E93, tyrosine-kinase related, and ribbon genes (Baehrecke and Thummel, 1995; Weber et al., 1995; Horowitz and Berg, 1996; Lehmann et al., 1998; Haller et al., 1987; Bradley and Andrew, 2001; Shim et al., 2001; Siegmund and Lehmann, 2002; Couderc et al., 2002). Based on sequence similarity and other characteristics, proteins including Psq domains in distinct species can be classified into different groups (Siegmund and Lehmann, 2002). The Psq motif of the hern and fer genes (as well as that of the CG13496 predicted gene) shows higher similarity to those included in the products of the human centromere protein B and the human predicted protein CAB66474 (Siegmund and Lehmann, 2002). This motif is a helix–turn– helix DNAbinding domain (Lehmann et al., 1998), structurally similar to the DNA-binding domain of the Hin and other recombinases (Lehmann et al., 1998; Siegmund and Lehmann, 2002). This suggests that hern and fer function as transcription regulators. hern and fer specify distal antenna The concept of selector genes in Drosophila has evolved from a precise and restricted definition (Garcı´a-Bellido,

477

1975) to a more loose interpretation (Mann and Morata, 2000). Selector, or selector-like genes, are now considered as those required to make a particular structure and able to form it in different positions when the gene is expressed ectopically. hern and fer fit this definition as selector genes for the distal antenna. They also can make ectopic eyes, although their requirement for eye development is not so evident as that for antenna formation. The differentiation of legs or antennae depends on the activity of the hth and Antp genes (Struhl, 1981; Casares and Mann, 1998). The ss gene, however, is also able to transform distal leg (and also maxillary palp and rostral membrane) into distal antenna (Duncan et al., 1998), and the absence of ss, like that of hth, transforms antenna into leg (Struhl, 1982; Burgess and Duncan, 1990; Casares and Mann, 1998; Pai et al., 1998). Although ss seems to be downstream of Dll and hth in antenna specification (Duncan et al., 1998; Dong et al., 2001; I. Duncan, cited in Dong et al., 2002), we have shown that ectopic ss can activate hth in some cells of the leg disc. Similarly, misexpression of ss in the rostral membrane induces Dll expression (Duncan et al., 1998). It seems, therefore, that ss can trigger an antennal genetic program when misexpressed in certain places. The hern and fer genes are transcribed, at the late third instar stage, in the third antennal segment and the arista. When the activity of these two genes is impaired, the distal antenna is partially transformed into a leg. This transformation is barely detected when only one of the genes is not active, implying that they perform a partially redundant role. In fact, both the expression and protein sequence of Hern and Fer are very similar. When either the hern or fer genes are ectopically expressed in the leg disc, they direct distal antennal development and activate antennal genes like hth, ss, and sal. Therefore, the fer and hern genes are both required and sufficient to make part of the distal antenna. Interactions of Dll, hth, and ss with hern and fer Four different genes, hth, ss, hern, and fer, are able to form distal antenna, together with Dll, when ectopically expressed (Casares and Mann, 1998; Duncan et al., 1998; Dong et al., 2000; Emerald et al., 2003; this report). Their mutual regulation seems to differ when misexpressed in the leg disc or when normally expressed in the antennal primordium. In the leg disc, hern or fer activates hth and ss and, reciprocally, hth and ss induce hern and fer expression. Moreover, even ss can promote hth transcription, although just in a few cells. Taken together, our results suggest that the four genes can form distal antenna by activating each other’s transcription when ectopically expressed (Fig. 10). In the third antennal segment, Dll, hth, and ss are required to activate hern/fer expression. Since ss is downstream of Dll and hth in the antenna (Duncan et al., 1998; Dong et al., 2002), the activation of hern/fer by Dll and hth could be mediated by ss. We note, however, that the levels of hern and fer may modulate hth expression. Moderately

478

M. Suzanne et al. / Developmental Biology 260 (2003) 465– 483

M. Suzanne et al. / Developmental Biology 260 (2003) 465– 483

increased levels of fer can activate hth in dpp-GAL4/UASfer discs but, when the levels of hern or fer in the antenna are highly increased, the transcription of hth is prevented. These results suggest that the total amount of hern and fer expression may be regulated in the antennal primordium. Accordingly, in clones mutant for danr (hern), the expression of dan (fer) is upregulated (Emerald et al., 2003). Also supporting the conclusion that levels of hern and fer have to be regulated, we have found that, in ey-GAL4/UAS-hern or ey-GAL4/UAS-fer flies, where levels of either hern or fer are highly increased in the eye–antennal disc, both the eye and the antenna disappear (not shown). hern and fer specify eye development Several eye-specifying genes have been identified, and they fulfill two conditions: they are required to make the eye and they can form ectopic eyes when expressed in different parts of the body (Treisman, 1999; Kumar, 2001). The hern and fer genes probably form part of this network of “eyespecification” genes: first, they are expressed in the eye primordium, with higher levels of expression anterior to the morphogenetic furrow; second, they activate ey and elav and make ectopic eyes when expressed ectopically; finally, ey also activates the hern and fer genes when ectopically expressed. A recent report has also identified the hern and fer genes as downstream of ey in eye ectopic formation (Michaut et al., 2003). We observe, however, that the inactivation of both hern and fer genes by RNAi with the GAL4 driver does not grossly affect eye development, like mutants in the eye-specification genes do (Treisman, 1999; Kumar, 2001). The nonautonomous induction of elav when hern or fer are ectopically expressed reproduces the wild-type situation, in which high levels of hern and fer are observed adjacent to the differentiating, elav-expressing, photoreceptor cells. Another similarity of hern and fer with some of the “eye-specification” genes is that we have obtained ectopic eye tissue in the antennae. The eye-specification genes eya and dac also form eyes predominantly, when ectopically expressed, in this same position (Bonini et al., 1997; Pignoni et al., 1997; Shen and Mardon, 1997). This is perhaps due to ey being expressed in the antennal primordium in late embryos (Quiring et al., 1994; Kumar and Moses, 2001), thus providing a favorable genetic context for eye formation. In accordance, when we express either the hern or the

479

fer gene ectopically, we only detect ectopic ey expression in the antennal disc. We also obtain eyes in the rostral membrane when ectopically expressing the fer gene. This may be due to the absence of hth, since high levels of either hern or fer repress hth and removal of this gene in the rostral membrane forms ectopic eyes (Pai et al., 1998; Pichaud and Casares, 2000). hern and fer genes and the decision to make eye or antenna The hern and fer genes can form ectopic aristae and eye tissue, but only in a limited number of regions of the adult cuticle. This is similar to what happens with other genes making ectopic antennae (hth, ss) or eye (eye-specification genes) (Halder et al., 1995; Bonini et al., 1997; Chen et al., 1997; Pignoni et al., 1997; Shen and Mardon, 1997; Casares and Mann, 1998; Duncan et al., 1998; Czerny et al., 1999; Dong et al., 2000; Seimiya and Gehring, 2000). This is due to the particular developmental context of the region where the genes are ectopically activated (Chen et al., 1999; Dong et al., 2000). We have observed transformations of third antennal segment, where hern and fer are normally transcribed, to eye tissue, in Dll-GAL4/UAS-hern or dpp-Gal4/UAS-fer flies. This suggests that the levels of Hern and Fer products may be important in inducing or maintaining ey expression and distinguishing eye from antenna. Accordingly, when Hern or Fer products are increased in the antennal primordium, the expression of hth, an inhibitor of eye development (Pai et al., 1998), is eliminated. We also note that, in the wildtype eye–antennal discs, hern and fer show higher levels of expression in the eye primordium than in the antennal one, where these genes are coexpressed with hth. However, the amount of Tintin products is not the only factor in this distinction, since, for instance, in Dll-GAL4/UAS-hern eyeantennal discs, the area of ectopic ey transcription in the antenna is smaller than the area of hern overexpression. The activity of other genes will probably contribute to the formation of either eye or antenna. Thus, the ectopic expression of either hern or fer induces wg, an inhibitor of morphogenetic furrow formation (Ma and Moses, 1995; Treisman and Rubin, 1995), and this probably limits the places where the eye can develop. Two recent models have been proposed to explain the

Fig. 8. The inactivation of hern and fer genes alters eye development and the ectopic expression of the hern or fer genes transforms parts of the head into eye tissue and activates ey and elav. (A) Wild-type eye. (B) Slightly rough eye in a GMR-GAL4; ifer/⫹ fly. (C) Detail of eye bristles in a wild-type eye. (D) Similar detail in a GMR-GAL4; ifer/⫹ fly. Note that the bristles are irregularly positioned (arrows). (E, F) Antennae of a Dll-GAL4 (MD23)/UAS-hern fly: ectopic eye tissue appears in the third antennal segment (arrows in E). A detail of the transformation is shown in (F). (G) A similar ectopic eye (arrow) is obtained when the fer gene is expressed under dpp-GAL4 control. Symbols in (E–G) are as in Fig. 1. (H, I) Eye–antennal disc of a Dll-GAL4/UAS-hern larva, showing ectopic expression of elav (red in I, arrows) and ey (H, I, green; arrow in H). Note that, as in wild-type discs, ey and elav expression do not coincide. The asterisks in (H) and (I) mark the normal ey and elav expression. a, antennal primordium. (J) Leg disc of a dpp-GAL4/UAS-ey larva, showing ectopic activation of the fer gene. (K) Eye–antennal imaginal disc with clones ectopically expressing the fer gene, marked with GFP (green), and activating elav (in red). (L–N) Detail of the clone boxed in (K), showing the nonautonomous induction of elav (red, M) in one side of the clone expressing fer (green in L). (N) Merged image.

480

M. Suzanne et al. / Developmental Biology 260 (2003) 465– 483

Fig. 9. Ectopic expression of the hern or fer genes activates wg, and N signaling is required for eye formation by the hern gene. (A) Eye primordium of a wg-lacZ larva hybridized with a probe of the fer gene, showing coexpression of wg and fer in the lateral regions (squares); detail of this overlap in the inset. (B) Wing of a dpp-GAL4/UAS-fer fly showing the ectopic triple row of bristles in the middle of the wing (arrow), a partial duplication of the wing blade (arrowhead), and a greatly enlarged hinge. (C) Detail of the ectopic triple row appearing in the middle of the wing blade. (D) Wing disc of the same genotype stained with an anti-Wg antibody (red). There is ectopic wg expression (arrow) in the dpp domain in addition to the normal wg signal (arrowhead). (E–H) Clone ectopically expressing the fer gene in the wing disc (in green, marked with GFP in E). This expression causes induction of wg expression (red in F) mostly within the clone, whereas elav induction is strictly cell-nonautonomous (marked in blue in G). (H) Merged image. (I–K) Eye–antennal disc of a Dll-GAL4/UAS-Necd UAS-hern larva. The ectopic expressions of elav (red in I) and ey (green in J, arrow) are substantially reduced when compared with those observed in Dll-GAL4/UAS-hern larvae (Fig. 8H and I). The asterisks in (I) and (J) indicates normal elav and ey expression. Merged image in (K).

M. Suzanne et al. / Developmental Biology 260 (2003) 465– 483

481

Fig. 10. Model of activity of the hern and fer genes. When expressed ectopically in the leg disc, hth, ss, hern, and fer activate one another (Dong et al., 2002; and this report), although activation of hth by ss is limited to a few cells. In the eye–antennal disc, the N and Egfr signaling pathways (Kumar and Moses, 2001) determine whether eye or antennal development follows. In the antennal primordium, hth, together with Dll, activates hern/fer through ss. In the eye primordium, hern and fer expression probably depends on the eye-specification gene group.

specification of eye and antenna within the eye–antennal disc. Both models suggest that the activation of the N signaling pathway is a key element in this process. Kurata et al. (2000) suggest that N signaling activates both ey and Dll in the eye and antennal primordia; subsequently, ey represses Dll in the eye and perhaps the hth and extradenticle genes repress ey in the antenna. In this way, the exclusive expression of ey (in the eye) and Dll and hth (in the antenna) determine eye and antenna identity, respectively. Kumar and Moses (2001) propose that the N and Egfr signaling pathways (together with the hedgehog and wg genes) are instrumental in the decisions to make eye or antenna (Kumar and Moses, 2001). According to these authors, N signaling promotes eye development and prevents formation of the antenna, whereas Egfr signaling does the opposite. We have found that ectopic expression of either hern or fer in the antenna induces ectopic eyes and activates ey and elav, but that the coexpression of hern and an N dominantnegative protein does not result in ectopic eyes and almost eliminates ey and elav activation. This suggests that N function impinges on hern activity to form ectopic eyes (Fig. 10). As in other cases, the combined activity of signaling pathways and selector genes determine the specification of different structures.

Acknowledgments We thank G. Morata for his help and encouragement throughout the work and for the GAL4 lines; S. Cohen for antibodies, stocks, and for communicating results before publication; J.F. de Celis and G. Morata for comments on the manuscript; N. Azpiazu for the gift of the ss probe; A. Cantarero for providing the AC116 line; R. Gonza´ lez for

preparation of the adult cuticles; and N. Azpiazu, R. Barrio, S. Cohen, D. Cribbs, I. Duncan, Gine´ s Morata, A. Preiss, A. Salzberg, R. Schuh, H. Sun, and U. Walldorf for the antibodies, probes, and stocks. M.S. is recipient of an European Marie Curie fellowship and M.C. is recipient of a Ramon y Cajal contract from the Spanish Ministerio de Ciencia y Tecnologı´a. This work has been supported by grants from the Direccio´ n General de Investigacio´ n Cientı´fica y Te´ cnica (N° PB98-0510) and the Comunidad Auto´ noma de Madrid (N° 08.9/0003/98 and 08.1/0031/2001.1) (to E.S.), and an Institutional Grant from the Fundacio´ n Ramo´ n Areces to the Centro de Biologı´a Molecular Severo Ochoa. References Abu-Shaar, M., Mann, R., 1998. Generation of multiple antagonistic domains along the proximodistal axis during Drosophila leg development. Development 125, 3821–3830. Adams, M.D., Celniker, S.E., Holt, R.A., Evans, C.A., Gocayne, J.D., Amanatides, P.G., Scherer, S.E., Li, P.W., et al., 2000. The genome sequence of Drosophila melanogaster. Science 287, 2185–2195. Affolter, M., Percival-Smith, A., Mu¨ ller, M., Billeter, M., Qian, Y.Q., Otting, G., Wu¨ trich, K., Gehring, W.J., 1991. Similarities between the homeodomain and the Hin recombinase DNA-binding domain. Cell 64, 879 – 880. Artavanis-Tsakonas, S., Rand, M.D., Lake, R.J., 1999. Notch signaling: cell fate control and signal integration in development. Science 284, 770 –776. Ashburner, M., 1989. Drosophila: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Atkinson, N.S., Robertson, G.A., Ganetzky, B., 1991. A component of calcium-activated potassium channels encoded by the Drosophila slo locus. Science 253, 551–555. Azpiazu, N., Frasch, M., 1993. tinman and bagpipe: two homeo box genes that determine cell fates in the dorsal mesoderm of Drosophila. Genes Dev. 7, 1325–1340.

482

M. Suzanne et al. / Developmental Biology 260 (2003) 465– 483

Azpiazu, N., Morata, G., 2000. Function and regulation of homothorax in the wing imaginal disc of Drosohila. Development 127, 2685–2693. Azpiazu, N., Morata, G., 2002. Distinct functions of homothorax in leg development in Drosophila. Mech. Dev. 119, 55– 67. Baehrecke, E.H., Thummel, C.S., 1995. The Drosophila E93 gene from the 93F early puff displays stage- and tissue-specific regulation by 20hydroxyecdysone. Dev. Biol. 171, 85–97. Barrio, R., de Celis, J.F., Bolshakov, S., Kafatos, F.C., 1999. Identification of regulatory regions driving the expression of the Drosophila spalt complex at different developmental stages. Dev. Biol. 215, 33– 47. Bonini, N.M., Bui, Q.T., Gray-Board, G.L., Warrick, J.M., 1997. The Drosophila eyes absent gene directs ectopic eye formation in a pathway conserved between fly and vertebrates. Development 124, 4819 – 4826. Bradley, P.L., Andrew, D.J., 2001. ribbon encodes a novel BTB/POZ protein required for directed cell migration in Drosophila melanogaster. Development 128, 3001–3015. Brand, A., Perrimon, N., 1993. Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118, 401– 415. Brook, W.J., Cohen, S.M., 1996. Antagonistic interactions between wingless and decapentaplegic responsible for dorsal–ventral pattern in the Drosophila leg. Science 273, 1373–1377. Brook, W.J., Dı´az-Benjumea, F.J., Cohen, S.M., 1996. Organizing spatial pattern in limb development. Annu. Rev. Cell Biol. 12, 161–180. Burgess, E.A., Duncan, I., 1990. Direct control of antennal identity by the spineless-aristapedia gene of Drosophila. Mol. Gen. Genet. 221, 347– 352. Calleja, M., Moreno, E., Pelaz, S., Morata, G., 1996. Visualization of gene expression in living adult Drosophila. Science 274, 252–255. Campbell, G., Tomlinson, A., 1995. Initiation of the proximodistal axis in insect legs. Development 121, 619 – 628. Campbell, G., Tomlinson, A., 1998. The roles of the homeobox genes aristaless and Distal-less in patterning the legs and wings of Drosophila. Development 125, 4483– 4493. Casares, F., Mann, R.S., 1998. Control of antennal versus leg development in Drosophila. Nature 392, 723–726. Chen, R., Amoui, M., Zhang, Z., Mardon, G., 1997. Dachsund and eyes absent proteins form a complex and function synergistically to induce ectopic eye development in Drosophila. Cell 91, 893–904. Chen, R., Halder, G., Zhang, Z., Mardon, G., 1999. Signaling by the TGF-␤ homolog decapentaplegic functions reiteratively within the network of genes controlling retinal cell fate determination in Drosophila. Development 126, 935–943. Cohen, S.M., Ju¨ rgens, G., 1989. Proximal– distal pattern formation in Drosophila: graded requirements for Distal-less gene activity during limb development. Roux’s Arch. Dev. Biol. 198, 157–169. Condie, L.M., Mustard, J.A., Brown, D.L., 1991. Generation of antiAntennapedia monoclonal antibodies and Antennapedia protein expression in imaginal discs. Dros. Inf. Serv. 70, 52–54. Couderc, J.-L., Godt, D., Zollman, S., Chen, J., Li, M., Tiong, S., Cramton, S.E., Sahut-Barnola, I., Laski, F., 2002. The bric a` brac locus consists of two paralogous genes encoding BTB/POZ domain proteins and acts as a homeotic and morphogenetic regulator of imaginal development in Drosophila. Development 129, 2419 –2433. Czerny, T., Halder, G., Kloter, U., Souabni, A., Gehring, W.J., Busslinger, M., 1999. twin of eyeless, a second Pax-6 gene of Drosophila, acts upstream of eyeless in the control of eye development. Mol. Cell. 3, 297–307. de Celis, J.F., Barrio, R., Kafatos, F.C., 1996. A gene complex acting downstream of dpp in Drosophila wing morphogenesis. Nature 381, 421– 424. Dı´az-Benjumea, F., Cohen, B., Cohen, S.M., 1994. Cell interaction between compartments establishes the proximal– distal axis of Drosophila legs. Nature 372, 175–178. Dı´az-Benjumea, F.J., Cohen, S.M., 1995. Serrate signals through Notch to establish a Wingless-dependent organizer at the dorsal/ventral compartment boundary of the Drosophila wing. Development 121, 4215– 4225.

Dong, P.D.S., Chu, J., Panganiban, G., 2000. Co-expression of the homeobox genes Distal-less and homothorax determines Drosophila antennal identity. Development 127, 209 –216. Dong, P.D.S., Chu, J., Panganiban, G., 2001. Proximodistal domain specification and interactions in developing Drosophila appendages. Development 128, 2365–2372. Dong, P.D.S., Scholz Dicks, J., Panganiban, G., 2002. Distal-less and homothorax regulate multiple targets to pattern the Drosophila antenna. Development 129, 1967–1974. Duncan, D.M., Burgess, E., Duncan, I., 1998. Control of distal antennal identity and tarsal development in Drosophila by spineless-aristapedia, a homolog of the mammalian dioxin receptor. Genes Dev. 12, 1290 – 1303. Emerald, B.S., Curtiss, J., Mlodzik, M., Cohen, S.M., 2003. distal antenna and distal antenna related encode nuclear proteins containing pipsqueak motifs involved in antenna development in Drosophila. Development 130, 1171–1180. Fire, A., Xu, S., Montgomery, M.K., Kostas, S.A., Driver, S.E., Mello, C.C., 1998. Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 391, 806 – 811. Garcı´a-Bellido, A., 1975. Genetic control of wing disc development in Drosophila. Ciba Found. Symp. 29, 161–182. Gellon, G., McGinnis, W., 1998. Shaping animal body plans in development and evolution by modulation of Hox expression patterns. Bioessays 20, 116 –125. Go, M.J., Eastman, S., Artavanis-Tsakonas, S., 1998. Cell proliferation control by Notch signaling in Drosophila development. Development 125, 2031. Gonza´ lez-Crespo, S., Abu-Shaar, M., Torres, M., Martinez-A, C., Mann, R.S., Morata, G., 1998. Antagonism between extradenticle function and Hedgehog signalling in the developing limb. Nature 394, 196 –200. Gorfinkiel, N., Morata, G., Guerrero, I., 1997. The homeobox gene Distalless induces ventral appendage development in Drosophila. Genes Dev 11, 2259 –2271. Halder, G., Callaerts, P., Gehring, W.J., 1995. Induction of ectopic eyes by targeted expression of the eyeless gene in Drosophila. Science 267, 1788 –1792. Haller, J., Cote, S., Bronner, G., Jackle, H., 1987. Dorsal and neural expression of a tyrosine kinase-related Drosophila gene during embryonic development. Genes Dev. 1, 862– 867. Hazelrigg, T., Kaufman, T.C., 1983. Revertants of dominant mutations associated with the Antennapedia gene complex of Drosophila melanogaster, in: Malacinski, G., Klein, W. (Eds.), Molecular Aspects of Early Development, Plenum Press, New York, pp. 189 –218. Hinz, U., Giebel, B., Campos-Ortega, J.A., 1994. The basic helix–loop– helix domain of Drosophila lethal of scute protein is sufficient for proneural function and activates neurogenic genes. Cell 14, 77– 87. Horowitz, H., Berg, C.A., 1996. The Drosophila pipsqueak gene encodes a nuclear BTB-domain-containing protein required early in oogenesis. Development 122, 1859 –1871. Inbal, A., Halachmi, N., Dibner, C., Frank, D., Salzberg, A., 2001. Genetic evidence for the transcriptional-activating function of Homothorax during adult fly development. Development 128, 3405–3413. Ito, K., Awano, W., Suzuki, K., Hiromi, Y., Yamamoto, D., 1997. The Drosophila mushroom body is a quadruple structure of clonal units each of which contains a virtually identical set of neurons and glial cells. Development 124, 761–771. Kennerdell, J.R., Carthew, R.W., 1998. Use of ds-RNA-mediated genetic interference to demonstrate that frizzled and frizzled 2 act in the Wingless pathway. Cell 95, 1017–1026. Klein, T., Brennan, K., Martı´nez-Arias, A., 1997. An intrinsic dominant negative activity of Serrate that is modulated during wing development in Drosophila. Dev. Biol. 186, 123–134. Ku¨ hnlein, R.P., Frommer, G., Friedrich, M., Gonzalez-Gaita´ n, M., Weber, A., Wagner-Bernholz, J.F., Gehring, W.J., Ja¨ ckle, H., Schuh, R., 1994. spalt encodes an evolutionarily conserved zinc finger protein of novel

M. Suzanne et al. / Developmental Biology 260 (2003) 465– 483 structure which provides homeotic gene function in the head and tail region of the Drosophila embryo. EMBO J. 13, 168 –179. Kumar, J.P., 2001. Signaling pathways in Drosophila and vertebrate retinal development. Nat. Rev. 2, 846 – 857. Kumar, J.P., Moses, K., 2001. EGF receptor and Notch signaling act upstream of eyeless/Pax6 to control eye specification. Cell 104, 687– 697. Kurata, S., Go, M.J., Artavanis-Tsakonas, S., Gehring, W.J., 2000. Notch signaling and the determination of appendage identity. Proc. Natl. Acad. Sci. USA 97, 2117–2122. Lecuit, T., Cohen, S.M., 1997. Proximal– distal axis formation in the Drosophila leg. Nature 388, 139 –145. Lehmann, M., Siegmund, T., Lintermann, K.-G., Korger, G., 1998. The Pipsqueak protein of Drosophila melanogaster binds to GAGA sequences through a novel DNA-binding domain. J. Biol. Chem. 273, 28504 –28509. Ma, C., Moses, K., 1995. wingless and patched are negative regulators of the morphogenetic furrow and can affect tissue polarity in the developing Drosophila compound eye. Development 121, 2279 –2289. Mann, R.S., Morata, G., 2000. The developmental and molecular biology of genes that subdivide the body of Drosophila. Annu. Rev. Coll Dev. Biol. 16, 243–271. Mardon, G., Solomon, N.M., Rubin, G.M., 1994. dachsund encodes a nuclear protein required for normal eye and leg development in Drosophila. Development 120, 3473–3486. Michaut, L., Flister, S., Neeb, M., White, K.P., Certa, U., Gehring, W.J., 2003. Analysis of the eye developmental pathway in Drosophila using DNA microarrays. Proc. Natl. Acad. Sci. USA 100, 4024 – 4029. Morata, G., 2001. How Drosophila appendages develop. Nat. Rev. Mol. Cell Biol. 2, 89 –97. Nagel, A.C., Maier, D., Preiss, A., 2002. Green fluorescent protein as a convenient and versatile marker for studies on functional genomics in Drosophila. Dev. Genes Evol. 212, 93–98. Pai, C.-Y., Kuo, T., Jaw, T., Kurant, E., Chen, C., Bessarab, D., Salzberg, A., Sun, Y., 1998. The homothorax homeoprotein activates the nuclear localization of another homeoprotein, Extradenticle, and suppresses eye development in Drosophila.. Genes Dev. 12, 435– 436. Pichaud, F., Casares, F., 2000. Homothorax and iroquois-C genes are required for the establishment of territories within the developing eye disc. Mech. Dev. 96, 15–25. Pignoni, F., Hu, B., Zavitz, K.H., Xiao, J., Garrity, P.A., Zipurski, S.L., 1997. The eye specification proteins so and eya form a complex and regulate multiple steps in Drosophila eye development. Cell 91, 881– 892. Pignoni, F., Zipurski, S., 1997. Induction of Drosophila eye development by Decapentaplegic. Development 124, 271–278. Postlethwait, J.H., Schneiderman, H.A., 1971. Pattern formation and determination in the antenna of the homeotic mutant Antennapedia of Drosophila melanogaster. Dev. Biol. 25, 606 – 640. Quiring, R., Walldorf, U., Kloter, U., Gehring, W.J., 1994. Homology of the eyeless gene of Drosophila to the Small eye gene in mice and Aniridia in humans. Science 265, 785–789. Rieckhof, G., Casares, F., Ryoo, H.D., Abu-Shaar, M., Mann, R.S., 1997. Nuclear translocation of Extradenticle requires homothorax, which

483

encodes an Extradenticle-related homeodomain protein. Cell 91, 171– 183. Robinow, S., Campos, A.R., Yao, K.M., White, K., 1988. The elav gene product of Drosophila, required in neurons, has three RNP consensus motif. Science 242, 1570 –1572. Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Seimiya, M., Gehring, W.J., 2000. The Drosophila homeobox gene optix is capable of inducing ectopic eyes by an eyeless-independent mechanism. Development 127, 1879 –1886. Sharp, P.A., 2001. RNA interference 2001. Genes Dev. 15, 485– 490. Shen, W., Mardon, G., 1997. Ectopic eye development in Drosophila induced by directed dachsund expression. Development 124, 45–52. Shim, K., Blake, K.J., Jack, J., Krasnow, M.A., 2001. The Drosophila ribbon gene encodes a nuclear BTB domain protein that promotes epithelial migration and morphogenesis. Development 128, 4923– 4933. Siegmund, T., Lehmann, M., 2002. The Drosophila Pipsqueak protein defines a new family of helix–turn– helix DNA-binding proteins. Dev. Genes Evol. 212, 152–157. Staehling-Hampton, K., Jackson, P.D., Clark, M.J., Brand, A.H., Hoffmann, F.M., 1994. Specificity of bone morphogenetic protein-related factors: cell fate and gene expression changes in Drosophila embryos induced by decapentaplegic but not 60A. Cell Growth Differ. 5, 585– 593. Struhl, G., 1981. A homeotic mutation transforming leg to antenna in Drosophila. Nature 292, 635– 638. Struhl, G., 1982. Spineless-aristapedia: a homeotic gene that does not control the development of specific compartments in Drosophila. Genetics 102, 737–749. Treisman, J.E., 1999. A conserved blueprint for the eye? Bioessays 21, 843– 850. Treisman, J.E., Rubin, G.M., 1995. wingless inhibits morphogenetic furrow movement in the Drosophila eye disc. Development 121, 3519 – 1527. Wagner-Bernholz, J.T., Wilson, C., Gibson, G., Schuh, R., Gehring, W., 1991. Identification of target genes of the homeotic gene Antennapedia by enhancer detection. Genes Dev. 5, 2467–2480. Weber, U., Siegel, V., Mlodzik, M., 1995. pipsqueak encodes a novel nuclear protein required downstream of seven-up for the development of photoreceptors R3 and R4. EMBO J. 14, 6247– 6257. Wirz, J., Fessler, L.I., Gehring, W.J., 1986. Localization of the Antennapedia protein in Drosophila embryos and imaginal discs. EMBO J. 5, 3327–3334. Wu, J., Cohen, S.M., 1999. Proximal distal axis formation in the Drosophila leg: primary subdivision into proximal and distal domains by Homothorax, Teashirt and Distal-less expression. Development 126, 109 – 117. Yamaguchi, M., Hirose, F., Inoue, Y.H., Shiraki, M., Hayashi, Y., Nishi, Y., Matsukage, A., 1999. Ectopic expression of human p53 inhibits entry into S phase and induces apoptosis in the Drosophila eye imaginal disc. Oncogene 18, 6767– 6775.