A Misexpression Study Examining Dorsal Thorax Formation in Drosophila melanogaster

Corresponding author: Juan Rafael Riesgo-Escovar, Instituto de Neurobiología, UNAM, Futbol #149, Col. Country Club Churubusco, México, D. F., 04220, México., riesgo{at}mail.cnb.unam.mx (E-mail)

Communicating editor: T. SCHÜPBACH

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS AND DISCUSSION LITERATURE CITED

We studied thorax formation in Drosophila melanogaster using a misexpression screen with EP lines and thoracic Gal4 drivers that provide a genetically sensitized background. We identified 191 interacting lines showing alterations of thoracic bristles (number and/or location), thorax and scutellum malformations, lethality, or suppression of the thoracic phenotype used in the screen. We analyzed these lines and showed that known genes with different functional roles (selector, prepattern, proneural, cell cycle regulation, lineage restriction, signaling pathways, transcriptional control, and chromatin organization) are among the modifier lines. A few lines have previously been identified in thorax formation, but others, such as chromatin-remodeling complex genes, are novel. However, most of the interacting loci are uncharacterized, providing a wealth of new genetic data. We also describe one such novel line, poco pelo (ppo), where both misexpression and loss-of-function phenotypes are similar: loss of bristles and scutellum malformation.

GENETIC screens for loss-of-function mutations have been conventional tools for the identification of genes in model organisms like Drosophila (NUSSLEIN-VOLHARD and WIESCHAUS 1980 ). Unfortunately, early lethality hinders studies of later functions of these genes. Also, redundancy or pleiotropy may prevent identification of many genes. In this regard, MIKLOS and RUBIN 1996 have estimated that two-thirds to three-quarters of all Drosophila genes have no obvious loss-of-function mutant phenotypes. In all these cases, controlled temporal and spatial misexpression is an alternative for identification of these genes and facilitates description of their genetic interactions. RORTH 1996 devised a successful system in Drosophila that allows controlled misexpression of genes randomly tagged by insertion of a transposable element (EP element) containing UAS sites that can bind the yeast transcripton factor Gal4 and a promoter directed outward 3' of the insertion site. Crosses to lines bearing gal4-expressing insertions in defined patterns allow misexpression of genes 3' of the EP elements in the progeny bearing both constructs.

Very successful genetic screens have also been conducted in sensitized genetic backgrounds (SIMON et al. 1991 ; OLIVIER et al. 1993 ). In these cases, the genetic background of the screen is not wild type, but contains a mutation in a gene that "enhances" effects caused by mutations in other genes, especially those with which this gene interacts. To take advantage of a genetically sensitized background while at the same time providing a Gal4 driver line, we have employed mutant alleles of pannier (pnr) and apterous (ap), caused by insertional mutagenesis as thoracic Gal4 driver lines (CALLEJA et al. 1996 ).

pnr is a pleiotropic gene that encodes a zinc-finger protein with homology to vertebrate GATA transcription factors (RAMAIN et al. 1993 ; WINICK et al. 1993 ). In embryos and larvae, pnr is expressed in dorsal tissues (amnioserosa, lateral epidermis, and dorsal-most aspects of wing and eye imaginal discs). In adults, pnr expression runs in a medial region from the head capsule to the posterior end of the abdomen (CALLEJA et al. 1996 ). pnr functions as a prepattern gene required for specification of the dorsocentral aspect of the thorax together with u-shaped (ush; HAENLIN et al. 1997 ; CALLEJA et al. 2000 ). Lateral aspects of the thorax are specified by another set of prepattern genes, the iroquois gene complex (GOMEZ-SKARMETA et al. 1996 ). Null pnr mutants are embryonic lethal with premature loss of the amnioserosa and defects in dorsal closure (HEITZLER et al. 1996 ). Mutant clones in the notum anlagen of the wing imaginal disc produce alterations in the pattern of medial thoracic bristles and a strong dorsal thoracic cleft. The allele used in this study, pnrGal4, has an insertion in the pnr promoter region that results in a hypomorphic allele (CALLEJA et al. 1996 ). Expression of Gal4 in pnrGal4 closely mimics normal pnr expression. In heterozygous condition, adult pnrGal4 flies show a very slight dorsal thoracic cleft and disorganization of bristles, sometimes missing a few inner and postvertical bristles (Fig 1B).

View larger version (188K): In this window In a new window Download PPT slide   Figure 1. Enhancement of the pnrGal4/+ phenotype. Modifications in the number and distribution of macrochaetae and/or microchaetae. All figures are dorsal views of adult thoraxes, except C where the dorsal abdomen is also shown. Anterior is up. (C–H) Misexpression phenotypes of the genes listed using pnrGal4. In the case of G, this leads to antisense expression; all others are overexpression phenotypes. Dashed lines in D show the approximate area where the phenotype is located in A–H, except C, and the arrow in A shows the location of the dorsocentral macrochaetae. Wild type (A), pnrGal4/+ (B), emc (C), sd (D), glec (E), stg (F), antisense stg (G), and EP(X)1353 (H). Note loss of dorsocentral macrochaetae in C–E.

These multiple functions of pnr allow several different phenotypes (and genes required for those phenotypes) to be screened for at the same time. We have exploited this in our screen in order to identify genes required for bristle formation, thoracic closure, thorax and scutellum shape, and viability.

We also used apGal4 as a Gal4 driver counterscreen for some lines to allow us to separate effects due to the combination of particular sensitized backgrounds plus misexpression vs. effects of misexpression alone. apterous is a transcription factor harboring a LIM domain and a homeodomain that plays a key role in wing, haltere, and nervous system formation (COHEN et al. 1992 ; LUNDGREN et al. 1995 ). In larvae, ap is expressed in groups of cells of the dorsal compartment of the wing disc, which also express pnr, and in the region giving rise to the wing hinge and dorsal surface of the wing blade. We used a Gal4 allele of ap that drives Gal4 expression in an ap expression pattern (CALLEJA et al. 1996 ). Heterozygous adult apGal4 flies have wild-type wings and thorax.

The formation of complex structures like the Drosophila thorax putatively requires a large number of genes, including some involved in growth, differentiation, cellular death, and maintenance of the differentiated state. Many genes that control these general processes are known. For example, growth of the wing disc—the imaginal disc that gives rise to the thorax—is controlled by Wingless (Wg), Hedgehog (Hh), mitogen-activated protein kinase (MAPK), and Decapentaplegic (Dpp) signaling pathways (CLIFFORD and SCHUPBACH 1989 ; BLAIR 1995 ). Structures within the thorax, such as bristles, may require different types of genes: selector, prepattern, and proneural genes for regional differentiation, formation, and correct spacing, besides signaling pathways. Dpp and Wg are also important for the formation of bristles in the mesothorax (TOMOYASU et al. 1998 ; PHILLIPS et al. 1999 ; SATO et al. 1999 ). Mutations in the Jun N-terminal kinase (JNK) pathway produce thoracic clefts, indicating a failure in the coordination of imaginal epithelial sheet spreading, a phenomenon similar to dorsal closure during midembryogenesis (RIESGO-ESCOVAR and HAFEN 1997 ; AGNES et al. 1999 ; ZEITLINGER and BOHMANN 1999 ). Limited amounts of apoptosis also occur during thorax formation (MILAN et al. 1997 ). In spite of this, very little is known about the actual overall genetic architecture that governs the formation of this structure.

We screened 2100 EP lines from which 191 (9%), corresponding to 167 loci, produced modifications of the weak pnrGal4 phenotype. These modifier lines identified genes previously known to be important for notum development, but also many novel and hitherto known genes that were previously not characterized as required for thoracic development.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS AND DISCUSSION LITERATURE CITED

Drosophila maintenance:
Fly stocks and crosses were maintained in cotton-stoppered glass vials in a molasses-yeast-gelatin standard medium.

Drosophila stocks:
The collection of EP transgenic lines generated and described by RORTH 1996 was obtained from J. Szabad at the European Drosophila stock center in Szeged, Hungary, and from E. Hafen, University of Zürich (Zurich, Switzerland). pnr ^MD237, the pnrGal4 line (CALLEJA et al. 1996 ), was obtained from G. Morata, Autonomous University of Madrid (Madrid, Spain). apGal4 and UAS-pnr were obtained from J. Modolell, Autonomous University of Madrid. FRT-containing chromosomes and transgenic hs-flp flies were obtained from E. Hafen, University of Zürich. All other stocks used are described in LINDSLEY and ZIMM 1992 or Flybase (http://flybase.bio.indiana.edu/).

EP screen:
pnrGal4 virgin female flies were mated to males from the EP collection. F₁ progeny were scored under a dissecting microscope for modification of the weak pnrGal4/+ phenotype. As most EP lines represent insertions in the promoter region of genes, we expected most of the effects observed to be due to overexpression. A fraction of EP lines are inserted at the 3' end of genes, and so loss-of-function effects due to production of antisense transcripts could also be expected. All lines were tested with pnrGal4 as driver, and some interacting lines were then tested with apGal4. All crosses were carried out at 25°. Lines that modified the original phenotype were retested at least once. In some cases, parents from initial crosses with a strong enhancement of the pnrGal4 phenotype were retested at 25° and progeny were allowed to develop at 18°, 25°, and 29° to allow us to look for effects of different levels of expression of Gal4.

Adult F₁ flies with modification of the pnrGal4/+ phenotype were fixed and stored in the dark in a 3:1 mixture of ethanol:glycerol. Thoraxes of adult flies were viewed in stereo and optical microscopes (Nikon and Olympus) and digitally photographed (Optronics, Pixera). For scanning electron microscopy, adult flies were fixed in 4% glutaraldehyde in phosphate-buffered saline (PBS) for 2 hr at room temperature, after ventral incisions in the thorax, abdomen, and head were made. They were then washed in PBS and dehydrated in a graded series of acetone-PBS (50, 70, and 100%) at room temperature. Afterward the specimens were critical point dried, mounted in stubs dorsal thorax side up, sputter coated with gold, viewed, and photographed in a Zeiss scanning electron microscope.

Clonal analysis:
Mutant lethal alleles of ppo obtained by imprecise P-element excisions were recombined onto FRT chromosomes. These ppo FRT chromosomes were crossed to appropriately marked FRT chromosomes with hs-flp, and heat shocks were applied to progeny larvae for 1 hr at 37° to generate mutant clones in the thorax and wing. The larvae were allowed to develop and examined as adults. We used forked and yellow to mark clones. For the generation of germline clones, we followed the dominant female sterile technique of CHOU and PERRIMON 1996 .

	RESULTS AND DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS AND DISCUSSION LITERATURE CITED

For our screen, we hypothesized that lines where thoracic misexpression leads to alterations of the pnrGal4 phenotype were candidate genes for thoracic development. We found lines that modified the thorax/scutellum structure or that altered significantly the number, size, and/or arrangement of chaetes in the notum. Overall modifications were produced in 9% of the 2100 lines screened. This compares favorably to other misexpression screens (RORTH et al. 1998 ; ABDELILAH-SEYFRIED et al. 2000 ; HUANG and RUBIN 2000 ; KRAUT et al. 2001 ), indicating that our screen was very sensitive. We classified observed modifications in three broad categories: (1) enhancement of the pnrGal4 phenotype in 5.9% (125 of the total lines), (2) lethality (embryonic, larval, or pupal) in <1% (20 lines), and (3) suppression of the pnrGal4 phenotype in 2.1% (45 lines). The lines where the pnrGal4 phenotype was enhanced showed: (a) modifications in the number and distribution of thoracic macrochaetes and/or microchaetes, (b) defects of the scutellum, and (c) formation of a cleft of variable width and depth at the dorsal midline of the thorax, with some lines showing more than one phenotype. Very few lines modified the dorsal aspects of the head and abdomen (0.2%). Table 1 shows a summary of these results. Some modifier lines were retested and progeny allowed to develop both at 18° and 29°. Progeny, except for lines EP(2)0639, EP(2)2148, EP(2)2402, and EP(2)2437, showed similar phenotypes to those observed at 25° (data not shown). These four lines have insertions in the same locus and showed stronger effects at 29°, including pupal lethality.

EP elements inserted in genes with previously known effects in the thorax acted as positive controls of the screen. In the following descriptions, we group genes found in the screen by known function and use the positive controls to validate results.

Enhancement of the pnrGal4/+ phenotype: effects on chaetae:
Selector genes: The scalloped (sd) gene is a selector gene that encodes a transcription factor expressed in several imaginal discs, including the wing imaginal disc and the peripheral and central nervous systems. Sd/Vestigial (Vg) heterodimers bind to cis-regulatory sequences controlling patterns of gene expression for wing development (HALDER et al. 1998 ). In addition, some mutant sd alleles have defects in the peripheral sense organs of the embryo (CAMPBELL et al. 1992 ) and loss of sensory organs in the wing margin (CAMPBELL et al. 1991 ). In our case, misexpression of sd via EP(X)1435, with both pnrGal4 and apGal4, produced loss of several microchaetae and macrochaetae (Table 1, Ia; Fig 1D). In another gain-of-function screen aimed at identifying genes required in external sensory organ development, misexpression of sd using scaGal4 also produced effects on bristles (ABDELILAH-SEYFRIED et al. 2000 ). These findings confirm that misexpression of sd, irrespective of Gal4 driver or of sensitized background, alters numbers of microchaetae. sd mutants have chemosensory defects (ANAND et al. 1990 ). Taken together, these data argue for a function of sd in bristle formation, perhaps in conjunction with other transcription factors like Vg.

Prepattern and proneural genes: Precise positional information is required during development to generate specific two-dimensional patterns. This information is provided by the prepattern (STERN 1954 ), a combination of transcriptional activators and repressors distributed asymmetrically that prefigure the pattern of each region (CAMPUZANO and MODOLELL 1992 ). One such transcriptional repressor in the thoracic prepattern is Hairy. Proneural genes are activated in groups of cells within these competent regions from which sensory organ precursor cells or neuroblasts will form. Extramacrochaetae (Emc) act as antagonists to proneural gene function (MOSCOSO DEL PRADO and GARCIA-BELLIDO 1984 ; ELLIS et al. 1990 ; GARRELL and MODOLELL 1990 ; RAMAIN et al. 1993 ). Prepatterns for bristle formation on the thorax involve the establishment of a zone of competent cells, called proneural territory, where neural precursors arise as sensory organ precursors (SOP). emc overexpression gives rise to loss of SOPs (POSAKONY 1994 ). In EP(3)0415, EP(3)3087, EP(3)3614, and EP(3)3620, the P element is inserted in the emc locus. Misexpression of these lines via pnrGal4 resulted in complete loss of bristles in the pnr region (Table 1, Ic; Fig 1C), a result consistent with emc overexpression experiments. As in the previous case of sd, loss-of-function and misexpression data support evidence of a gene's requirement for a function. This argues that relevant loci can be identified through misexpression screens, notwithstanding the caveat that misexpression can also yield artifactual phenotypes.

Cell cycle regulation genes: Loss of bristles could also arise by alterations during the last stages of their formation, for example, by transformation of the support or external cells (the shaft or trichogen cell or the socket or tormogen cell) into neurons (POSAKONY 1994 ) or loss of sensory mother cells (KIMURA et al. 1997 ), caused by changes in the cell cycle or cell lethality. The EP elements in EP(3)3261 and EP(3)3426 are inserted in string (stg), a gene that encodes a protein tyrosine phosphatase involved in mitotic G2/M transition (EDGAR et al. 1994 ). We observed missing microchaetae in the pnr-expressing region with these lines (Table 1, Ia; Fig 1F). Conversely, a suppression of pnrGal4/+ phenotype was observed with the EP(3)3432 line (Fig 1G). This line also contains an insertion in string, but in an antisense orientation, thus generating a loss-of-function phenotype. Reduction of stg expression is seen in groups of cells where sensory mother cells are emerging (MILAN et al. 1996 ), and stg hypomorphic combinations affect formation of macrochaetae (VERHEYEN et al. 1996 ). A gain-of-function study found that stg overexpression generated transformations of shaft to socket cells (ABDELILAH-SEYFRIED et al. 2000 ).

Misexpression of several other lines such as EP(2)1221, EP(2)2356, EP(3)3072, EP(3)3073, EP(3)3306, and EP(3)3621 also results in loss of bristles. These insertions are in four different novel genes (see Table 1, Ia).

Effects on the scutellum:
Lineage restriction genes: The epidermis of imaginal discs is divided into anterior and posterior compartments. The principal signaling cascades involved in the determination of these regions are the hh and dpp signaling pathways. hh is expressed in posterior compartment cells and signals to adjacent cells in the anterior compartment. Patched (Ptc) is a transmembrane receptor for hh. Genetic analysis indicates that Ptc is a negative regulator limiting Hh diffusion, thus repressing expression of Hh target genes (CHEN and STRUHL 1998 ). In the EP(2)0941 line, the EP element is inserted in ptc. Misexpression in the thorax via pnrGal4 produced flies lacking the scutellum (Table 1, Ib; Fig 2A), thus suggesting that limiting hh signaling disrupts scutellum formation. EP(X)0385 also produced a very similar effect (Table 1, Ib; Fig 2B), but the insertion is in an uncharacterized gene. Misexpression of EP(3)0666, EP(3)0707, EP(2)0853, and other lines (see Table 1) produced other defects on the scutellum, and insertions are in uncharacterized genes as well.

View larger version (190K): In this window In a new window Download PPT slide   Figure 2. Enhancement of the pnrGal4/+ phenotype. Modifications in thorax, scutellum, and other structures. Conventions as in Fig 1, except K, where overexpression using apGal4 is shown. D, E, H, and L are scanning electron microscope micrographs. Arrow in A points to the nearly total absence of scutellum. ptc (A), EP(X)0385 (B), dpp (C), activated version of Tkv (TkvQD construct of NELLEN et al. 1996 ) (D), homozygous weak allele of DFos (E), Dsp1 (F), tara (G), kis (H), faf (I), EP(2)0639 (J), EP(2)0639 with apGal4 (K), and antisense esg (L).

Effects on the thorax and on multiple structures:
Signaling pathway genes: During thorax formation the two hemithoraxes come together and fuse at the dorsal midline in a process known as thoracic closure. Spreading and fusion of the lateral wing discs in the midline are controlled by the JNK and Dpp signaling pathways. Hypomorphic mutants of dpp show lack of filopodia from imaginal leading edge cells (MARTIN-BLANCO et al. 2000 ). dpp also participates in sensory organ patterning in the thorax in conjunction with the Wg signaling pathway (PHILLIPS et al. 1999 ; SATO et al. 1999 ). Misexpression of the EP(2)2232 line, in which the P element is inserted in the dpp gene, produced a cleft in the dorsal midline in the thorax and abnormally placed chaetae, as expected (Table 1, Id; Fig 2C). We also misexpressed an activated version of the Dpp receptor thick veins (tkv) and obtained similar results (Fig 2D). A strong thoracic and scutellar cleft is produced in a weak mutant allele of kayak (kay; kay is the Drosophila DFos homolog, a JNK and Dpp downstream gene; Fig 2E). Thus, both loss- and gain-of-function phenotypes give comparable results, implying that dpp signaling levels are critical for function during thorax formation.

Chromatin organization and transcriptional control genes: In eukaryotes, nucleosome assembly and higher-order packaging produce a general repression of gene expression. Remodeling of chromatin structure is necessary for gene activation. ATP-dependent protein complexes with chromatin-remodeling activity can change nucleosomal patterns and DNA packaging (KINGSTON and NARLIKAR 1999 ). The best-studied Drosophila chromatin-remodeling complexes are the NURF complex (TSUKIYAMA and WU 1995 ), the CHRAC complex (VARGA-WEISZ et al. 1997 ), the ACF complex (ITO et al. 1997 ), and the BRM complex (PAPOULAS et al. 1998 ). The BRM complex includes several genes of the trithorax group. We unexpectedly found in our screen that misexpression of three genes involved in changes in chromatin structure produced strong defects during thoracic closure. One of them is Dsp1, line EP(X)0355, a gene that encodes a member of the high mobility group 1/2 (HMG) family of gene repression proteins (LEHMING et al. 1998 ). We also found two members of the trithorax group: kismet (kis; DAUBRESSE et al. 1999 ), lines EP(2)2240 and EP(2)0563, and taranis (tara; H. N. BOURBON, personal communication to FlyBase), line EP(3)3463.

Dsp1 was first identified as a corepressor of Dorsal (LEHMING et al. 1998 ), but recent evidence has showed that Dsp1 is a regulatory factor involved in several stages during development and is expressed in many tissues (MOSRIN-HUAMAN et al. 1998 ). For example, a gain-of-function screen aimed at identifying genes involved in eye and wing development (RORTH et al. 1998 ) showed that misexpression of Dsp1 produced defects in both the wing and the eye. In our screen, misexpression of Dsp1 produced a strong thoracic and scutellar cleft (Table 1, Id; Fig 2F). It is possible that Dsp1 interacts with other transcription factors during thorax development.

taranis has been reported as an enhancer in a screen aimed at identifying genes that modify ataxin-1-induced neurodegeneration (FERNANDEZ-FUNEZ et al. 2000 ) and as affecting macrochaetae (ABDELILAH-SEYFRIED et al. 2000 ). Our results showed that tara misexpression also alters thoracic closure (Table 1, Id; Fig 2G). Further studies are required to clarify tara's multiple roles during development.

kismet was found in a genetic screen for dominant modifiers of Polycomb mutations (KENNISON and TAMKUN 1988 ). kis has ATPase domains homologous to a subunit of SWI/SNF remodeling complexes (DAUBRESSE et al. 1999 ). We observed a thoracic cleft by misexpression of kis (Table 1, Id; Fig 2H). We found several independent insertions in kis, such as EP(2)2240 and EP(2)0563, which had similar phenotypes. Molecular analysis indicated that these insertions mapped to the start of the kis short transcript, thus implying only the short form of kis in the aforementioned effects. kis is an example of a gene whose misexpression phenotype may be informative: loss-of-function kis alleles are lethal with no obvious phenotype, and clonal analysis has revealed few alterations in mutant cells, perhaps implying redundancy in kis function.

To study this, we crossed several kis lethal alleles to pnrGal4 and found that reducing kis function suppressed the dominant pnrGal4 phenotype (Fig 3A and data not shown), indicating that kis and pnr interact genetically. The effect is the opposite of what was obtained when kis was overexpressed and suggests that dosage of kis is important in modulating pnr function in the dorsal thorax. Moreover, crossing a mutation in Polycomb (Pc) to pnrGal4 enhances dramatically the pnrGal4 phenotype (Fig 3B). Since Pc antagonizes trithorax function, this result is expected. Taken together, the data presented argue for a function of chromatin-remodeling complexes in the formation of the thorax, specifically by regulating pnr activity. pnr is a prepattern gene, and this would point to chromatin-remodeling complexes as regulators of prepattern gene activity. As such, the data imply a novel function for trithorax and Polycomb group genes.

View larger version (214K): In this window In a new window Download PPT slide   Figure 3. kis, Pc, and ppo thoracic phenotypes. Dorsal thoracic views; conventions as in Fig 2. (A) kis suppresses the pnrGal4 phenotype. Genotype shown is kis07816/+; pnrGal4/+. We also did the same cross with kisk13416 and obtained identical results. (B) Pc acts as an enhancer of pnrGal4. The genotype shown is Pc3/pnrGal4. (C and D) Misexpression of ppo alters the scutellum and reduces the number of bristles in the pnr thoracic domain. Genotype in C is EP(2)2478/+; pnrGal4/+; in D it is EP(2)2478/ apGal4. (E) ppo misexpression reduces thoracic bristles and alters scutellum form even in a wild-type background. Genotype shown is UAS-pnr/EP(2)2478; pnrGal4. A control rescue of pnrGal4 is shown in F: UAS-pnr/+; pnrGal4/+. (G) Loss-of-function clones of ppo in the thorax reduce the numbers of bristles. Clone shown is ppo5/ ppo5 and is marked with forked. ppo5 is a lethal allele of ppo obtained by imprecise excision of the EP element in EP(2)2478. The clonal region shows reduced numbers of thoracic bristles. The approximate limits of the clones are marked by the white line. (H and I) ppo interacts genetically with pnr but not with ap. The genotype of H is ppo5/+; pnrGal4/+; note loss of bristles and deformed scutellum, whereas the genotype of I, ppo5/ apGal4, is wild type.

EP(2)0816 also showed a strong thoracic cleft, but the insertion is in an uncharacterized gene with homology to a histone acetyltransferase. It is thus possible that this gene is also involved in chromatin organization and state during thorax formation.

Genes acting in post-translational regulation: Misexpression of fat facets (faf) in EP(3)0381 produced a slight cleft in the thorax and abdomen and a reduced scutellum (Table 1, Id; Fig 2I). faf encodes a Ubp enzyme (ubiquitin-processing protease) required for cellularization of the embryo and for patterning the developing eye (FISHER-VIZE et al. 1992 ; CHEN and FISHER 2000 ). It is now well established that ubiquitin-mediated proteolysis plays an essential role in several key biological processes involving signal transduction (MANIATIS 1999 ). Several lines—(EP(3)0416, EP(3)0562, EP(2)0622, EP(2)0639 (and the other lines in the same locus), EP(X)1501, EP(3)3135, and EP(3)3528—also have multiple effects and lie in genes that are not yet characterized.

We rescreened 20 of the pnrGal4 interacting lines with apGal4. Of these, one-half gave similar effects with both drivers. Examples include line EP(2)0639, where misexpression with both pnrGal4 and apGal4 gave similar results (Table 1, Id; Fig 2F and Fig K), and the previously mentioned sd gene. Six more lines gave different effects, and 4 lines gave no effects with apGal4. This underscores the fact that a large fraction (80%) of the rescreened interacting lines produced effects on the thorax irrespective of genetic background. Lines that did not show effects with apGal4 may represent lines that have specific interactions with pnr.

Effects on viability:
We found effects on viability in 20 lines where EP misexpression via pnrGal4 caused lethality. The EP elements were inserted in known genes such as escargot (esg), inscuteable (insc), and Krüppel-homolog 1 (Kr-h1) in 7 of the lines. esg is expressed in the central nervous system during late embryogenesis and larval stages and in imaginal discs. Loss-of-function mutations of esg produce embryonic or early larval lethality (ASHBURNER et al. 1990 ). esg overexpression causes defects in eye and wing development (RORTH et al. 1998 ). In our screen, overexpression of esg using EP(2)2009 and EP(2)0684 produced lethality during embryonic development. However, one antisense insert in the esg gene in EP(2)0683 produced a strong thoracic cleft and defects in the abdominal dorsal midline but not lethality (Table 1, Id; Fig 2E). This loss-of-function effect could nevertheless be caused also by cell death in these tissues, since esg codes for a zinc-finger protein involved in maintaining diploidy of imaginal cells and probably also in suppressing the endoduplication that occurs in larval cells (HAYASHI et al. 1993 ). Dosage of esg is important and may be required for cell survival.

The Kr-h1 gene [EP(2)2289; Table 1, Id] encodes a transcription factor involved in ecdysone response. This gene has at least two different promoters and produces multiple transcripts during development. Mutants complete normal embryonic and larval development but die during metamorphosis with defects of head eversion (PECASSE et al. 2000 ). We also observed pupal lethality in Kr-h1 overexpression. Kr-h1 transcript levels during metamorphosis may be critical. Besides these lines, we also found that EP(2)0797, EP(2)873, EP(3)1191, EP(X)1595, EP(2)2016, EP(2)2612, EP(3)3085, EP(3)3223, EP(3)3334, and EP(3)3384, all with insertions in unknown genes, produce lethality.

Suppression of the pnrGal4/+ phenotype:
From our screen, only 4 out of 45 EP lines whose misexpression suppresses pnrGal4/+ correspond to characterized genes: EP(X)0382, corkscrew (csw); EP(2)2278, big brain (bib); EP(X)1442, amnesiac (amn); and the already mentioned EP(3)3432, stg (Fig 1G). csw encodes a protein phosphatase identified in an embryonic genetic screen and functions in MAPK signaling pathways. This is consistent with evidence that MAPK pathways are required for normal spacing and number of thoracic bristles (CLIFFORD and SCHUPBACH 1989 ). Due to the weak pnrGal4/+ phenotype, experiments to test suppression in stronger pnr mutant backgrounds should be performed to assess the extent of the rescue. The collection of uncharacterized suppressors could provide a starting point for identification of novel pnr interacting genes, as suppressions are generally thought to evidence more specific genetic interactions.

Novel gene identified in the screen with thoracic defects:
In the foregoing analysis we have concentrated on genes whose functions, in several cases, were already known and whose involvement in thorax formation was sometimes also known (they thus acted as positive controls), but one of the aims of the screen was to identify novel genes required for dorsal thorax formation. A large collection of the lines screened that produced effects fall into uncharacterized loci (142 loci; 85% of the lines screened). From these, we have focused on the analysis of one of them, line EP(2)2478. This line has several effects in the thorax when misexpressed via pnrGal4: it has a reduced number of thoracic bristles and a defective scutellum (Fig 3C), whereas the insertion by itself has no phenotype (data not shown). Misexpression via apGal4 also produced flies with missing bristles (Fig 3D), showing that the effect was not specific to pnrGal4 misexpression. Further confirmation of this, and of the fact that misexpression of EP(2)2478 generates loss of bristles, among other phenotypes, was obtained when we misexpressed both EP(2)2478 and pnr at the same time: the loss of bristles was even more evident (Fig 3E).

We then generated excision alleles of this insertion and identified several lethal lines. We mapped the lethality at the embryonic stage: both zygotic and germline clones show defects at the extended germband stage with a characteristic U-shaped mutant embryo (Fig 4B and Fig C). Since the excisions are lethal, we generated loss-of-function clones in larvae to study their effects in the thorax (Fig 3G). Adult thoracic clones also lack bristles and have a deformed scutellum, implying that both misexpression and loss-of-function phenotypes are similar. For this reason we have named this locus poco pelo (ppo), which means few hairs in Spanish. In view of the fact that both pnr and ppo produce loss of bristles and defects in the thorax, we decided to test whether these two loci genetically interact. We crossed a lethal allele of ppo to pnrGal4. As shown in Fig 3H, ppo genetically interacts with pnrGal4, because it enhances the weak dominant effect of pnrGal4 with respect to bristle loss but also with respect to formation of the thorax. This genetic interaction is specific and not due to expression of Gal4 in the thorax in conjunction with a reduction of ppo function, because crosses of the same lethal allele of ppo to apGal4 produced wild-type progeny (Fig 3I).

View larger version (99K): In this window In a new window Download PPT slide   Figure 4. Lethal alleles of ppo have embryonic phenotypes. Embryonic cuticles are shown; dorsal is up and anterior is to the left. Lateral views. (A) The cuticle of a wild-type embryo; note filzkörper at the dorsal posterior end and ventral denticle belts. (B) A ppo5/ ppo5 zygotic mutant embryo. The embryo has a characteristic U-shape, with part of the ventral cuticle at the posterior end curved upward; note position of the filzkörper and abdominal denticle belts. (C) A germline clone of ppo5 shows the same embryonic lethal phenotype as a zygotic mutant, suggesting that there is no significant maternal contribution.

ppo also alters bristles in the wing: while examining the effects of ppo misexpression using apGal4 we observed loss of bristles and bristle duplications at the dorsal wing margin (Fig 5B). This was then corroborated using clonal analysis: loss-of-function clones of ppo at the wing margin also produced loss of bristles and bristle duplications (Fig 5C and Fig D).

View larger version (181K): In this window In a new window Download PPT slide   Figure 5. ppo function is required for bristle formation at the wing margin. (A) Dorsal view of a wild-type wing margin. Note close spacing of bristles. (B) The effects of misexpression of ppo in the wing margin with apGal4. Note extensive loss of bristles and bristle duplications. (C) A ppo5 mutant clone marked with forked (approximate limits are marked by the white line) showing a bristle duplication. D is a similar clone, marked as in C, showing bristle loss.

ppo maps to the left arm of the second chromosome at 32D. Since both the loss-of-function and misexpression phenotypes of ppo are similar, it is possible that the insertion in EP(2)2478 directs expression of an antisense transcript, but we have been unable to clone the gene and are thus unable to verify this assumption. It is also possible that both loss- and gain-of-function phenotypes yield similar results.

ppo mutant phenotypes at the embryonic and adult stages show that this locus is involved in the formation of dorsal structures, such as pnr, with which it interacts, but also that it is required for the formation of other structures where pnr is not required, such as the bristles of the dorsal wing margin. It thus represents a member of a class of genes that, like pnr, is required for the formation of the thorax both with respect to the number and spacing of bristles but also with respect to the shape and formation of the thorax: pnr mutants also display dorsal thoracic clefts, as do mutant conditions of dpp signaling in the thorax (both loss- and gain-of-function conditions), a class of genes that this screen was aimed at identifying. This also implies that ppo interacts with pnr in some structures, such as the thorax, but that, possibly, it interacts with other genes in some others, such as the wing. It is conceivable that these effects are due to the actions of more than one gene, since we have not cloned ppo, but in any case these genes would have to be in very close proximity and would also correspond to genes that this study was aimed at identifying. Further studies involving cloning are required to clarify the roles of ppo in bristle and thorax formation.

Misexpression screens have been shown to be a powerful tool in identifying genes involved in development. In this screen 11 genes already known to be required for bristle development, emc, dpp, glec, stg, tara, esg, Bx, Mef2, faf, sd, and bib, were identified. Nevertheless, some of these genes, such as tara, esg, and faf, were not known to also affect thoracic closure. We also identified 4 other genes previously characterized that had not yet been implicated in thorax formation: Dsp1, kis, ptc, and amn. Despite this, most of the pnrGal4 or apGal4 interacting lines likely represent genes that were not known to be linked to thorax formation. In 20 of these interesting EP lines, the insertions fall close to uncharacterized loci for which cDNAs or lethal lines exist (Table 1). Most of the interacting lines identifed represent novel genes, and from these we studied and characterized a new locus, ppo, whose misexpression and loss-of-function phenotypes show effects on the thorax and wing.

This result could stem from the fact that few, if any, concerted efforts have been made to catalog the genome's involvement in the formation of the thorax. This screen thus represents an entry point to a wealth of new genetic interactions. Most of what we found represents novel or hitherto unrecognized putative relationships, the interpretation of which has been greatly aided by the completion of the fly's genome sequence (ADAMS et al. 2000 ). It is now necessary to characterize loss-of-function phenotypes for many of these genes to delve into their functions and specific roles. Misexpression of genes may produce abnormalities from completely nonspecific effects in some cases, but our results demonstrate that the opposite is also true. The effects of emc, glec, stg, and tara overexpression, for example, show specific defects in structures also known to require the gene's function, and thus similar phenotypes produced by misexpression from other EP lines may include genes required for thorax development. We also screened lines like esg, which gave antisense misexpression and which yielded loss-of-function phenotypes (RORTH et al. 1998 ), akin to traditional loss-of-function alleles.

Several other misexpression screens show partial overlap with ours. We found that 33 loci identified in our screen also showed effects on external sensory organs using scaGal4 (ABDELILAH-SEYFRIED et al. 2000 ), an overlap of only 20%. This was expected, since the ABDELILAH-SEYFRIED et al. (2000) screen focused on SOP formation, a structure also present in the developing thorax. Another misexpression screen for genes required in axon guidance and synaptogenesis showed eight EP lines in common with our screen, an overlap of 5% (KRAUT et al. 2001 ). This is important if we consider that this screen studied effects on synapse formation at the neuromuscular junction in larvae. Genes detected in both screens may represent genes required for nervous system formation with effects on peripheral sensory organs. Overall, our screen, when compared to other misexpression screens using the same collection of EP lines (RORTH et al. 1998 ; ABDELILAH-SEYFRIED et al. 2000 ; KRAUT et al. 2001 ), identified a higher number of interacting lines. This result possibly reflects the fact that our attention was centered on the formation of the dorsal thorax, a complex process that involves more aspects than those covered by these other, more narrowly focused screens.

The application of genomic science and DNA microarray technology to characterize expression patterns is likely to be very useful in studying developmental processes. In striving to define the set of genes important for completion of a developmental program, this technology may be an alternative method to study formation of complex structures, such as the thorax. However, functional data are missing. Misexpression screens like the one we present here, combined with loss-of-function data, try to achieve the same goal by examining functional consequences of alterations in gene expression levels and/or domains and offer a complementary approach. Comprehensive functional screens can then be thought of as providing a phenotypical translation of genomic science studies, since we can assess phenotype aided by genome sequence data.

	ACKNOWLEDGMENTS

We gratefully acknowledge the help of Janos Szabad and Ernst Hafen in sending us the 2100 EP lines collection, and C. Carsolio, M. Jeziorski, and G. Martinez for critical reading of the manuscript. This work was supported by the Consejo Nacional de Ciencia y Tecnología (CONACYT) grant no. J27954-N to J.R.R.-E. and a Programa de Cooperacion Científica con Iberoamérica. M.T.P.-R. is a recipient of a Consejo Nacional de Ciencia y Tecnología (CONACYT) Ph.D. training grant.

Manuscript received October 11, 2001; Accepted for publication December 10, 2001.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS AND DISCUSSION LITERATURE CITED

ABDELILAH-SEYFRIED, S., Y. M. CHAN, C. ZENG, N. JUSTICE, and S. YOUNGER-SHEPHERD et al., 2000  A gain-of-function screen for genes that affect the development of the Drosophila adult external sensory organ. Genetics 155:733-752[Abstract/Free Full Text].

ADAMS, M. D., S. E. CELNIKER, R. A. HOLT, C. A. EVANS, and J. D. GOCAYNE et al., 2000  The genome sequence of Drosophila melanogaster.. Science 287:2185-2195[Abstract/Free Full Text].

AGNÈS, F., M. SUZANNE, and S. NOSELLI, 1999  The Drosophila JNK pathway controls the morphogenesis of imaginal discs during metamorphosis. Development 126:5453-5462[Abstract].

ANAND, A., J. FERNÁNDEZ, M. C. ARUNAN, S. BHOSEKAR, and A. CHOPRA et al., 1990  Drosophila enhancer-trap transposants. Gene expression in chemosensory and motor pathways and identification of mutants affected in smell and taste ability. J. Genet. 69(3):151-168.

ASHBURNER, M., P. THOMPSON, J. ROOTE, P. F. LASKO, and Y. GRAU et al., 1990  The genetics of a small autosomal region of Drosophila melanogaster containing the structural gene for alcohol dehydrogenase. VII. Characterization of the region around the snail and the cactus loci. Genetics 126:679-694[Abstract].

BLAIR, S., 1995  Compartments and appendage development in Drosophila. Bioessays 4:299-309.

CALLEJA, M., E. MORENO, S. PELAZ, and G. MORATA, 1996  Visualization of gene expression in living adult Drosophila.. Science 274:252-255[Abstract/Free Full Text].

CALLEJA, M., H. HERRANZ, C. ESTELLA, J. CASAL, and P. LAWRENCE et al., 2000  Generation of medial and lateral dorsal body domains by the pannier gene of Drosophila. Development 127:3971-3980[Abstract].

CAMPBELL, S., A. DUTTAROY, A. KATZEN, and A. CHOVNICK, 1991  Cloning and characterization of the scalloped gene of Drosophila melanogaster.. Genetics 127:367-380[Abstract].

CAMPBELL, S., M. INAMDAR, V. RODRIGUEZ, V. RAGHAVAN, and M. PALAZZOLO et al., 1992  The scalloped gene encodes a novel, evolutionarily conserved transcription factor required for sensory organ differentiation in Drosophila.. Genes Dev. 6:367-379[Abstract/Free Full Text].

CAMPUZANO, S. and J. MODOLELL, 1992  Patterning of the Drosophila nervous system: the achaete-scute gene complex. Trends Genet. 8:202-208[Medline].

CHEN, X. and J. FISHER, 2000  In vivo structure/function analysis of the Drosophila fat facets deubiquitinating enzyme gene. Genetics 156:1829-1836[Abstract/Free Full Text].

CHEN, Y. and G. STRUHL, 1998  In vivo evidence that Patched and Smoothened constitute distinct binding and transducing components of a Hedgehog receptor complex. Development 125:4943-4948[Abstract].

CHOU, T. B. and N. PERRIMON, 1996  The autosomal FLP-DFS technique for generating germline mosaics in Drosophila melanogaster.. Genetics 144:1673-1679[Abstract].

CLIFFORD, R. J. and T. SCHÜPBACH, 1989  Coordinately and differentially mutable activities of torpedo, the Drosophila melanogaster homolog of the vertebrate EGF receptor gene. Genetics 123:771-787[Abstract/Free Full Text].

COHEN, B., M. MCGUFFIN, C. PFEIFLE, D. SEGAL, and S. COHEN, 1992  apterous, a gene required for imaginal disc development in Drosophila, encodes a member of the LIM family of developmental regulatory proteins. Genes Dev. 6:715-729[Abstract/Free Full Text].

DAUBRESSE, G., R. DEURING, L. MOORE, O. PAPOULAS, and I. ZAKRAJSEK et al., 1999  The Drosophila kismet gene is related to chromatin-remodeling factors and is required for both segmentation and segment identity. Development 126:1175-1187[Abstract].

EDGAR, B., D. LEHMAN, and P. O'FARRELL, 1994  Transcriptional regulation of string (cdc25): a link between developmental programming and the cell cycle. Development 120:3131-3143[Abstract].

ELLIS, H., D. SPANN, and J. POSAKONY, 1990  extramacrochaetae, a negative regulator of sensory organ development in Drosophila, defines a new class of helix-loop-helix proteins. Cell 61:27-38[Medline].

FERNANDEZ-FUNEZ, P., M. L. NINO-ROSALES, B. DE GOUYON, W. SHE, and J. M. LUCHAK et al., 2000  Identification of genes that modify ataxin-1-induced neurodegeneration. Nature 408:101-106[Medline].

FISHER-VIZE, J., G. RUBIN, and R. LEHMANN, 1992  The fat facets gene is required for Drosophila eye and embryo development. Development 116:985-1000[Abstract].

GARRELL, J. and J. MODOLELL, 1990  The Drosophila extramacrochaetae locus, an antagonist of proneural genes that, like these genes, encodes a helix-loop-helix protein. Cell 61:39-48[Medline].

GÓMEZ-SKARMETA, J. L., R. DIEZ DEL CORRAL, E. DE LA CALLE, D. FERRÉS-MARCÓ, and J. MODOLELL, 1996  araucan and caupolican, two members of the novel Iroquois complex, encode homeoproteins that control proneural and vein-forming genes. Cell 85:95-105[Medline].

HAENLIN, M., Y. CUBADA, F. BLONDEAU, P. HEITZLER, and Y. LUTZ et al., 1997  Transcriptional activity of Pannier is regulated negatively by heterodimerization of the GATA DNA-binding domain with a cofactor encoded by the u-shaped gene of Drosophila.. Genes Dev. 11:3096-3108[Abstract/Free Full Text].

HALDER, G., P. POLACZYK, M. E. KRAUS, A. HUDSON, and J. KIM et al., 1998  The Vestigial and Scalloped proteins act together to directly regulate wing-specific gene expression in Drosophila.. Gene Dev. 12:3900-3909[Abstract/Free Full Text].

HAYASHI, S., S. HIROSE, T. METCALFE, and A. SHIRRAS, 1993  Control of imaginal cell development by the escargot gene of Drosophila.. Development 118:105-115[Abstract].

HEITZLER, P., M. HAENLIN, P. RAMAIN, M. CALLEJA, and P. SIMPSON, 1996  A genetic analysis of a pannier, a gene necessary for viability of dorsal tissues and bristle positioning in Drosophila.. Genetics 143:1271-1286[Abstract].

HUANG, A. and G. RUBIN, 2000  A misexpression screen identifies genes that can modulate RAS1 pathway signaling in Drosophila melanogaster.. Genetics 156:1219-1230[Abstract/Free Full Text].

ITO, T., M. BULGER, M. J. PAZIN, R. KOBAYASHI, and J. T. KADONAGA, 1997  ACF and ISWI-containing and ATP-utilizing chromatin assembly and remodeling factor. Cell 90:145-155[Medline].

KENNISON, J. A. and J. W. TAMKUN, 1988  Dosage-dependent modifiers of polycomb and antennapedia mutations in Drosophila.. Proc. Natl. Acad. Sci. USA 85:8136-8140[Abstract/Free Full Text].

KIMURA, K., A. USUI-ISHIHARA, and K. USUI, 1997  G2 arrest of cell cycle ensures a determination process of sensory mother cell formation in Drosophila.. Dev. Genes Evol. 207:199-202.

KINGSTON, R. E. and G. J. NARLIKAR, 1999  ATP-dependent remodeling and acetylation as regulators of chromatin fluidity. Genes Dev. 13:2339-2352[Free Full Text].

KRAUT, R., K. MENON, and K. ZINN, 2001  A gain-of-function screen for genes controlling motor axon guidance and synaptogenesis in Drosophila.. Curr. Biol. 11:417-430[Medline].

LEHMING, N., A. LE SAUX, J. SCHÜLLER, and M. PTASHNE, 1998  Chromatin components as part of a putative transcriptional repressing complex. Proc. Natl. Acad. Sci. USA 95:7322-7326[Abstract/Free Full Text].

LINDSLEY, D. L., and G. G. ZIMM, 1992 The Genome of Drosophila melanogaster. Academic Press, San Diego.

LUNDGREN, S., C. CALLAHAN, S. THOR, and J. THOMAS, 1995  Control of neuronal pathway selection by the Drosophila LIM homeodomain gene apterous.. Development 121:1769-1773[Abstract].

MANIATIS, T., 1999  A ubiquitin ligase complex essential for the NF-B, Wnt/Wingless, and Hedgehog signaling pathways. Genes Dev. 13:505-510[Free Full Text].

MARTIN-BLANCO, E., J. PASTOR-PAREJA, and A. GARCIA-BELLIDO, 2000  JNK and decapentaplegic signaling control adhesiveness and cytoskeleton dynamics during thorax closure in Drosophila.. Proc. Natl. Acad. Sci. USA 97:7888-7893[Abstract/Free Full Text].

MIKLOS, G. and G. RUBIN, 1996  The role of the genome project in determining gene function: insights from model organism. Cell 86:521-529[Medline].

MILAN, M., S. CAMPUZANO, and A. GARCIA-BELLIDO, 1996  Cell cycling and patterned cell proliferation in the wing primordium of Drosophila.. Proc. Natl. Acad. Sci. USA 93:640-645[Abstract/Free Full Text].

MILAN, M., S. CAMPUZANO, and A. GARCIA-BELLIDO, 1997  Developmental parameters of cell death in the wing disc of Drosophila.. Proc. Natl. Acad. Sci. USA 94:5691-5696[Abstract/Free Full Text].

MOSCOSO DEL PRADO, J. and A. GARCIA-BELLIDO, 1984  Genetic regulation of the achaete-scute complex of Drosophila melanogaster. Roux's Arch. Dev. Biol. 193:242-245.

MOSRIN-HUAMAN, C., L. CANAPLE, D. LOCKER, and M. DECOVILLE, 1998  DSP1 gene of Drosophila melanogaster encodes an HMG-domain protein that plays multiple roles in development. Dev. Genet. 23:324-334[Medline].

NELLEN, D., R. BURKE, G. STRUHL, and K. BASLER, 1996  Direct and long-range action of a DPP morphogen gradient. Cell 85:357-368[Medline].

NÜSSLEIN-VOLHARD, C. and E. WIESCHAUS, 1980  Mutations affecting segment number and polarity in Drosophila. Nature 287:795-801[Medline].

OLIVIER, J., T. RAABE, M. HENKEMEYER, B. DICKSON, and G. MBAMALU et al., 1993  A Drosophila SH2–SH3 adaptor protein implicated in coupling the Sevenless tyrosine kinase to an activator of Ras guanine nucleotide exchange, Sos. Cell 73:179-191[Medline].

PAPOULAS, O., S. J. MOSELEY, C. M. MCCALLUM, M. SARTE, and A. SHEARN et al., 1998  The Drosophila trithorax group proteins BRM, ASH1 and ASH2 are subunits of distinct protein complexes. Development 125:3955-3966[Abstract].

PECASSE, F., Y. BECK, C. RUÍZ, and G. RICHARDS, 2000  Krüppel-homolog, a stage-specific modulator of the prepupal ecdysone response, is essential for Drosophila metamorphosis. Dev. Biol. 221:53-67[Medline].

PHILLIPS, R., N. WARNER, and R. WHITTLE, 1999  Wingless signaling leads to asymmetric response to Decapentaplegic-dependent signaling during sense organ patterning on the notum of Drosophila melanogaster.. Dev. Biol. 207:150-162[Medline].

POSAKONY, J. W., 1994  Nature versus nurture: asymmetric cell divisions in Drosophila bristle development. Cell 76:415-418[Medline].

RAMAIN, P., P. HEITZLER, M. HAENLIN, and P. SIMPSON, 1993  pannier, a transcription regulatory gene of achaete and scute in Drosophila, encodes a zinc finger protein with homology to the vertebrate transcription factor GATA-1. Development 119:1277-1291[Abstract].

RIESGO-ESCOVAR, J. and E. HAFEN, 1997  Common and distinct roles of Dfos and Djun during Drosophila development. Science 278:669-672[Abstract/Free Full Text].

RØRTH, P., 1996  A modular misexpression screen in Drosophila detecting tissue-specific phenotypes. Proc. Natl. Acad. Sci. USA 93:12418-12422[Abstract/Free Full Text].

RØRTH, P., K. SZABO, A. BAILEY, T. LAVERTY, and J. REHM et al., 1998  Systematic gain-of-function genetics in Drosophila.. Development 125:1049-1057[Abstract].

SATO, M., T. KOJIMA, T. MICHINE, and K. SAIGO, 1999  Bar homeobox genes are latitudinal prepattern genes in the developing Drosophila notum whose expression is regulated by the concerted functions of decapentaplegic and wingless.. Development 126:1457-1466[Abstract].

SIMON, M., D. BOWTELL, G. DODSON, T. LAVERTY, and G. RUBIN, 1991  Ras1 and a putative guanine nucleotide exchange factor perform crucial steps in signaling by the sevenless protein tyrosine kinase. Cell 67:701-716[Medline].

STERN, C., 1954  Two or three bristles. Am. Scientist 42:213-247.

TOMOYASU, Y., M. NAKAMURA, and N. UENO, 1998  Role of Dpp signaling in prepattern formation of the dorsocentral mechanosensory organ in Drosophila melanogaster.. Development 125:4215-4224[Abstract].

TSUKIYAMA, T. and C. WU, 1995  Purification and properties of an ATP-dependent nucleosome remodelling factor. Cell 83:1011-1020[Medline].

VARGA-WEISZ, P. D., M. WILM, E. BONTE, K. DUMAS, and M. MANN et al., 1997  Chromatin-remodelling factor CHRAC contains the ATPases ISWI and topoisomerase II. Nature 388:598-602[Medline].

VERHEYEN, E. M., K. J. PURCELL, M. E. FORTINI, and S. ARTAVANIS-TSAKONAS, 1996  Analysis of dominant enhancers and suppressors of activated Notch in Drosophila. Genetics 144:1127-1141[Abstract].

WINICK, J., T. ABEL, M. LEONARD, M. MICHELSON, and I. CHARDON-LORIAUX, 1993  A GATA family transcription factor is expressed along the embryonic dorsoventral axis in Drosophila melanogaster.. Development 119:1055-1065[Abstract].

ZEITLINGER, J. and D. BOHMANN, 1999  Thorax closure in Drosophila: involvement of Fos and JNK pathway. Development 126:3947-3956[Abstract].

This article has been cited by other articles:

				M. Stofanko, S. Y. Kwon, and P. Badenhorst A Misexpression Screen to Identify Regulators of Drosophila Larval Hemocyte Development Genetics, September 1, 2008; 180(1): 253 - 267. [Abstract] [Full Text] [PDF]

				A. Khokhar, N. Chen, J.-P. Yuan, Y. Li, G. N. Landis, G. Beaulieu, H. Kaur, and J. Tower Conditional Switches for Extracellular Matrix Patterning in Drosophila melanogaster Genetics, March 1, 2008; 178(3): 1283 - 1293. [Abstract] [Full Text] [PDF]

				T. Zhao, T. Gu, H. C. Rice, K. L. McAdams, K. M. Roark, K. Lawson, S. A. Gauthier, K. L. Reagan, and R. S. Hewes A Drosophila Gain-of-Function Screen for Candidate Genes Involved in Steroid-Dependent Neuroendocrine Cell Remodeling Genetics, February 1, 2008; 178(2): 883 - 901. [Abstract] [Full Text] [PDF]