A Misexpression Screen Identifies Genes That Can Modulate RAS1 Pathway Signaling in Drosophila melanogaster

Corresponding author: Gerald M. Rubin, Howard Hughes Medical Institute, 545 Life Sciences Addition #3200, University of California, Berkeley, CA 94720-3200., gerry{at}fruitfly.BDGP.berkeley.edu (E-mail)

Communicating editor: R. S. HAWLEY

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Differentiation of the R7 photoreceptor cell is dependent on the Sevenless receptor tyrosine kinase, which activates the RAS1/mitogen-activated protein kinase signaling cascade. Kinase suppressor of Ras (KSR) functions genetically downstream of RAS1 in this signal transduction cascade. Expression of dominant-negative KSR (KDN) in the developing eye blocks RAS pathway signaling, prevents R7 cell differentiation, and causes a rough eye phenotype. To identify genes that modulate RAS signaling, we screened for genes that alter RAS1/KSR signaling efficiency when misexpressed. In this screen, we recovered three known genes, Lk6, misshapen, and Akap200. We also identified seven previously undescribed genes; one encodes a novel rel domain member of the NFAT family, and six encode novel proteins. These genes may represent new components of the RAS pathway or components of other signaling pathways that can modulate signaling by RAS. We discuss the utility of gain-of-function screens in identifying new components of signaling pathways in Drosophila.

MULTICELLULAR organisms must coordinate growth and differentiation of many different cell types throughout the course of development. To do this, individual cells must be able to recognize environmental cues, integrate multiple signals, and produce the appropriate developmental response. Cells utilize several categories of signaling molecules: transmembrane receptors that recognize extracellular cues, intracellular proteins that relay and amplify the signals, and effector molecules that convert the signals to a developmental output. The molecular mechanisms responsible for regulating these signaling events appear to have been used repeatedly in different contexts throughout all developmental stages.

The RAS1/mitogen-activated protein kinase (MAPK) pathway is one of the many evolutionarily conserved signaling modules. This pathway can respond to a number of extracellular signals including growth factors, stress, and hormones. The signaling relay consists of three protein kinases, RAF, MAPK kinase (MEK), and MAPK, which together act downstream of the small GTP-binding protein, RAS1. Signals are transduced via a protein phosphorylation cascade leading to the activation of MAPK, which then activates both cytoplasmic and nuclear targets (reviewed in SEGER and KREBS 1995 ). Although the molecules involved in this signaling cascade have been studied extensively, their interaction with other cellular targets and those effects on the course of development are not as clearly understood.

Numerous genetic screens have been conducted in Drosophila to identify novel components of the RAS1 pathway (SIMON et al. 1991 ; DICKSON et al. 1996 ; KARIM et al. 1996 ; NEUFELD et al. 1998 ; THERRIEN et al. 1998 ; REBAY et al. 2000 , and accompanying article, THERRIEN et al. 2000 , this issue). These screens were conducted in the adult eye because of the advantages it confers: it is not necessary for viability or fertility, and it is an easy tissue to screen. The eye is composed of a repeating array of 800 ommatidial clusters, each of which comprises 20 cells including 8 neuronal photoreceptor cells designated R1–R8. The eye arises from a monolayer of undifferentiated cells, the eye imaginal disc, and develops due to a concerted series of cell-cell interactions that utilize many intercellular signaling pathways.

One example is the specification of the presumptive R7 cell by the R8 photoreceptor cell (ZIPURSKY and RUBIN 1994 ). Proper differentiation of the R7 cell depends on local activation of the Sevenless receptor tyrosine kinase (RTK) by its ligand, BOSS, which is expressed on the adjacent R8 cell. The Sevenless RTK signals through the RAS1/MAPK pathway to activate downstream effectors required for specification of a neuronal cell fate. Genetic screens have identified many essential components of the RAS1/MAPK signaling cascade involved in R7 cell fate specification [SIMON et al. 1991; DICKSON et al. 1996; KARIM et al. 1996; NEUFELD et al. 1998; THERRIEN et al. 1998, 2000 (this issue); REBAY et al. 2000]. Ras1, sos, gap1, drk, Dsor1, ksr, PTP-ER, cnk, yan, and sina are all genes that were identified as essential downstream components of the Sevenless RTK pathway. The screens that led to the identification of these genes were all dominant modifier mutagenesis screens where loss-of-function mutations were identified by their ability to modify a RAS pathway-dependent rough eye phenotype.

Only one-quarter to one-third of all genes in the Drosophila genome can be mutated to an easily scored phenotype (MIKLOS and RUBIN 1996 ; ASHBURNER et al. 1999 ). For many genes, functional redundancy with related genes may prevent loss-of-function phenotypes from revealing gene function. Thus, classic loss-of-function genetic screens may miss many important components of regulatory pathways. Gain-of-function genetics have been widely used in multicopy screens in yeast (RINE et al. 1983 ; BENDER and PRINGLE 1989 ; RAMER et al. 1992 ), but have only recently come to fruition in Drosophila with the development of critical resources such as the EP and Gene Search collections (RORTH 1996 ; RORTH et al. 1998 ; TOBA et al. 1999 ). These are independent P-element lines where the P element contains a GAL4-inducible promoter that, in the presence of a tissue-specific source of GAL4, drives expression of the endogenous gene located proximal to the P element. Increasing gene dosage through overexpression may reveal gene functions that are not revealed through loss-of-function mutations.

To test the feasibility of using this gain-of-function misexpression approach as a means to identify nonessential genes that act in a signaling pathway, we took advantage of a RAS pathway-dependent rough eye phenotype caused by the expression of a dominant negative form of kinase suppressor of Ras (KSR). Many mutations in components of the RAS pathway can generate a rough eye; however, we chose to use this ksr phenotype because it is dosage sensitive and viable (see below). ksr was first identified in genetic screens in both Drosophila and Caenorhabditis elegans (KORNFELD et al. 1995 ; SUNDARAM and HAN 1995 ; THERRIEN et al. 1995 ). Genetic epistasis tests place it downstream of Ras1 and either upstream of or parallel to raf (THERRIEN et al. 1995 ). KSR facilitates signaling between RAF, MEK, and MAPK (THERRIEN et al. 1996 ). In quiescent cells, KSR is bound to MEK in the cytoplasm (DENOUEL-GALY et al. 1998 ; YU et al. 1998 ). Upon RAS1 activation, KSR translocates to the membrane where it is found in a complex with RAF (THERRIEN et al. 1998 ). Although ksr encodes a putative protein kinase, substrates have not been identified. A genetic screen for dominant modifiers of a ksr-dependent phenotype identified genes known to play a role in RAS1/MAPK signaling as interactors of KSR as well as several new loci including a putative adaptor protein connector enhancer of KSR (CNK) [THERRIEN et al. 1998, 2000 (this issue)].

The KSR kinase domain functions as a dominant-negative molecule when separated from the noncatalytic N-terminal domain (THERRIEN et al. 1996 ). Overexpression of a KSR dominant-negative (KDN) protein in a subset of cells of the developing eye under the control of the sevenless enhancer/heat shock promoter (sE-KDN) blocks RAS1-dependent photoreceptor cell differentiation. In sE-KDN ommatidia, the R7 photoreceptor cell and one or two of the other photoreceptor cells are missing. Thus most ommatidia have five to seven photoreceptor cells instead of the normal eight, resulting in a visible rough eye phenotype, which is clearly distinguishable under a dissecting microscope [THERRIEN et al. 1996, 1998, 2000 (this issue)]. The rough eye phenotype generated in this manner is sensitive to the copy number of transgenes; that is, flies carrying two copies of the transgene have more severely disrupted eyes than flies carrying only one copy. The dose sensitivity of this phenotype has provided a good background for a genetic screen designed to identify loss-of-function mutations that dominantly modify this phenotype by disruption of one of two copies of the wild-type gene [THERRIEN et al. 1998, 2000 (this issue)].

To identify new genes that can alter signaling through the RAS1 pathway in Drosophila, we present here the use of a misexpression screen to identify modifiers of a rough eye phenotype due to dominant-negative KSR. We screened the EP collection and identified several known signaling molecules as well as six novel loci that act as misexpression modifiers of this phenotype.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

EP screen:
An isogenic driver stock homozygous for both the sevenless enhancer/heat shock promoter-GAL4 (sE-GAL4) and sE-KDN [THERRIEN et al. 1998, 2000 (this issue)] P-element insertions was first generated. Female sE-KDN; sE-GAL4 flies were mated to male flies from each individual line from the EP collection (RORTH et al. 1998 ). F₁ progeny were scored under a dissecting scope for suppression or enhancement of the KDN rough eye phenotype. Potential modifiers were retested with a different isolate of the driver stock. Males from modifying lines were then crossed to CyO, P[sev-RAS1^V12]/Sco;sE-GAL4 females and curly winged F₁ progeny were again scored for enhancement or suppression of the sev-RAS1^V12 rough eye phenotype. Fly stocks were maintained at 25° on standard corn meal/molasses media.

The degree of enhancement or suppression by the misexpression lines was variable. For example, Fig 1D shows a strong suppressor of sE-KDN, yet the enhancement of sev-RAS1^V12 in Fig 1E is mild. Fig 1F shows a weak enhancement of the sE-KDN phenotype, yet this line turns out to be a very strong suppressor of sev-RAS1^V12, as seen in Fig 1G. This was not due to a difference in genetic background of the driver lines for two reasons: (1) the driver lines were outcrossed and isogenized prior to screening, and (2) the strong interactions were seen with both driver lines. This eliminates the possibility that one line generally facilitated stronger interactions than the other.

View larger version (131K): In this window In a new window Download PPT slide   Figure 1. Identification of misexpression enhancers and suppressors of KDN. Scanning electron micrographs of adult eyes of the following genotypes are shown: (A) P[sE-GAL4]/+; (B) P[sE-KDN]/+; (C) P[sev-RAS1V12]/+; (D) P[sE-KDN]/P[EP(2)2347]; P[sE-GAL4]/+; (E) P[sev-RAS1V12]/P[EP(2)2347]; P[sE-GAL4]/+; (F) P[sE-KDN]/P[EP(2)2221]; P[sE-GAL4]/+; and (G) P[sev-RAS1V12]/P[EP(2)2347]; P[sE-GAL4]/+. RAS pathway specific misexpression enhancers of KDN both make the sE-KDN eye rougher (F) as well as suppress the roughness of the sev-RAS1V12 eye (G), whereas suppressors of KDN cause the sE-KDN eye to look more like wild type (D) yet cause the sev-RAS1V12 phenotype to worsen (E).

Transgenic flies:
cDNAs were cloned into the pUAST (BRAND and PERRIMON 1993 ) vector and constructs were introduced into the germline by injection in the presence of transposase as previously described (RUBIN and SPRADLING 1982 ). UAS-msn flies were kindly provided by Jessica Treisman (SU et al. 1998 ).

Overexpression analysis:
dpp-GAL4 and Act5C-GAL4 flies were obtained from the Bloomington Stock Center, and vg-GAL4 flies were generated by Mike Brodsky.

Histology:
Adult fly eyes were prepared for thin section analysis by fixing in 2% glutaraldehyde and 2% osmium tetroxide followed by dehydration through an ethanol series as described previously (TOMLINSON et al. 1987 ). Flies were prepared for scanning electron microscopy by dehydration through an ethanol series followed by a freon series.

Sequence analysis:
DNA sequence tags flanking the P-element insertions were generated by the Berkeley Drosophila Genome Project (BDGP; http://www.fruitfly.org/) and used to BLAST search both the NCBI and BDGP databases to identify nearby genes. Expressed sequence tag (EST) clones identified by these means were sized by gel electrophoresis and the longest clones were then completely sequenced using the primer island transposon method (Applied Biosystems, Foster City, CA). Additional flanking sequence was obtained off of the other end of the P element by inverse PCR and, in some cases, plasmid rescue as described (HUANG et al. 2000 ). Full-length cDNA sequences are available in GenBank.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

A misexpression screen for modifiers of KDN:
The screen previously carried out was designed to isolate dominant modifiers of the sE-KDN phenotype [THERRIEN et al. 1998, 2000 (this issue)]. This was achieved by mutagenizing flies and identifying genes that, when mutated in one copy, could dominantly enhance or suppress the sE-KDN-dependent rough eye. These types of screens have been extremely successful and efficient in genetically dissecting signaling pathways (DICKSON et al. 1996 ; KARIM et al. 1996 ). These screens depend on a twofold difference in gene dosage to result in enhancement or suppression of a rough eye phenotype. However, since more than half of all genes in the genome do not mutate to any visible phenotype, it is probable that these previous screens have not identified all relevant interacting genes. In an effort to identify novel pathway interacting genes as well as test a new approach, we undertook a misexpression screen. The expectation is the same, that a change in dosage of an interacting molecule will affect the dosage-sensitive sE-KDN phenotype. In this study, we screened for genes that modify a rough eye phenotype when gene expression is increased rather than decreased.

The GAL4-UAS system from yeast has been used extensively to force ectopic expression of genes in flies (BRAND and PERRIMON 1993 ; BRAND et al. 1994 ). The EP collection (RORTH et al. 1998 ) is a collection of 2300 independent P-element insertion lines. The EP P element contains a GAL4-inducible promoter at the 3' end of the transgene to drive expression of the closest downstream endogenous gene when combined with a transgene expressing GAL4 under the control of a tissue-specific promoter. In this screen we used a sE-GAL4 line that expresses the yeast GAL4 activator in a subset of cell types in the developing eye (including photoreceptor cells R3, R4, R7, and cone cells). By itself, the sE-GAL4 transgene has no effect on normal eye development (Fig 1A). The sE-KDN transgene expresses the dominant-negative form of KSR in the same cell types (R3, R4, R7, and cone cells), which causes a loss of photoreceptor cells and leads to a roughened and disorganized eye (Fig 1B).

Since KDN blocks signaling efficiency through the RAS1/MAPK pathway, overexpression of genes that increase the overall output of the pathway will suppress the rough eye phenotype. Conversely, overexpression of genes that further reduce the signaling output will worsen the eye phenotype. Flies containing both the sE-KDN and sE-GAL4 transgenes were crossed to 2254 individual lines from the EP collection and F₁ progeny from each cross were screened for alterations of the KDN-dependent rough eye phenotype due to the misexpression of genes downstream of the EP element. A total of 140 EP misexpression lines that suppress or enhance the rough eye phenotype were identified from the collection (Fig 1D and Fig F). Seventy-five of these lines generate a rough eye when crossed to the sE-GAL4 driver alone in a wild-type background (A. M. BAILEY and G. M. RUBIN, unpublished results); these lines were not pursued any further because the enhancement of eye roughness likely reflects an additive effect of two unrelated phenotypes. Of the remaining 65, all exhibited wild-type eye morphology when misexpressed with sE-GAL4 in a wild-type background. These were further characterized as potential genetic interactors.

Tests to identify pathway relevant lines:
Although these lines showed no external phenotype when overexpressed in a wild-type background, there may be subtle phenotypes that were amplified by the presence of the sE-KDN transgene. Since it was already known that misexpression of 4% of the lines in this collection could generate rough eyes (A. M. BAILEY and G. M. RUBIN, unpublished results; RORTH 1996 ), we used additional criteria to determine which lines affected RAS1-dependent photoreceptor cell development. We examined the misexpression interactions in an activated RAS1 background as well as the effect on the R7 cell fate in particular. Examples of the RAS1 interaction are shown in Fig 1 and the results are summarized in Table 1.

Interaction with activated RAS1: The initial screen was performed in a background where RAS1 signaling is decreased due to the dominant-negative effect of the sE-KDN transgene resulting in fewer R7 and outer photoreceptor cells. If the RAS pathway is constitutively active in the R3, R4, R7, and cone cells, increased signaling activity results in the development of extra R7 photoreceptor cells. The RAS1^V12 cDNA encodes a protein with a glycine-to-valine substitution at position 12. This mutation prevents the hydrolysis and release of the bound GTP, which maintains RAS1 in an active form (reviewed in BOLLAG and MCCORMICK 1991 ). Overexpression of this cDNA in the eye under the control of the sevenless enhancer/promoter (sev-RAS1^V12) generates a dosage-sensitive rough eye phenotype (Fig 1C; FORTINI et al. 1992 ). The 65 putative interacting lines were crossed to a line expressing both the sev-RAS1^V12 and sE-GAL4 transgenes. F₁ progeny were screened for enhancement or suppression of the sev-RAS1^V12 phenotype (Fig 1E and Fig G). The KDN and RAS1^V12 transgenes produce their respective rough eye phenotypes by affecting photoreceptor development in opposite directions and therefore should respond oppositely to overexpression of a given interacting gene. Misexpression of genes that specifically increase RAS pathway signaling by acting downstream of RAS1 should suppress the sE-KDN phenotype but enhance the RAS1^V12 phenotype. Conversely, misexpression of lines that decrease RAS pathway signaling by acting downstream of both ksr and RAS1 should enhance the KDN phenotype and suppress the RAS1^V12 phenotype. In contrast, enhancers that roughen the eye by the addition of two unrelated phenotypes should enhance both lines. Of the 65 lines, 19 displayed the opposite interactions when misexpressed in the presence of the sE-KDN compared to the sev-RAS1^V12 transgene. The remaining 44 lines either had no visible effect on the sev-RAS1^V12 rough eye or exhibited the same phenotypic modification as with the sE-KDN transgene and were not pursued further.

The R7 photoreceptor cell fate: The RAS1/MAPK signaling cassette is involved in the cell fate specification of all eight neuronal photoreceptor cells as well as some of the nonneuronal cell types in the eye. However, this module is not always activated in a Sevenless RTK-dependent manner, which only occurs in the R7 photoreceptor. To quantify the degree of modification by each misexpression line, we focused on the R7 cell fate. We chose the R7 cell for two reasons: R7 appears most sensitive to changes in RAS1 signaling, and, whereas a rough eye may reflect changes in fates of many cell types, the number of R7 cells appears to directly reflect a change in the strength of RAS1 signaling.

The neuronal photoreceptor cells of the eye are organized in a regular, repeated trapezoidal pattern as seen in apical cross section (Fig 2A). The six outer photoreceptor cells (R1–R6) outline the trapezoid and the R7 and R8 cells lie in the center with the R7 cell situated apical to the R8 cell (which is not visible in these sections; Fig 2). In sE-KDN animals, 80% of ommatidia are missing R7 cells (Fig 2B; THERRIEN et al. 1996 ). Suppressors of this phenotype should restore the R7 cell fate. However, this change in number of R7 cells did not provide a good means by which to quantitate an enhancement of the KDN phenotype. Instead, we took advantage of the previous requirement that misexpression enhancers of sE-KDN also act as suppressors of sev-RAS1^V12. In the sev-RAS1^V12 background, 90% of ommatidia have supernumerary R7 cells (Fig 2C). There is also occasional loss of the outer photoreceptor cell types as well [KARIM et al. 1996; THERRIEN et al. 1998, 2000 (this issue)]. Suppressors of this phenotype restore the number of R7 cells to fewer than two per ommatidium in >10% of the ommatidia examined in each eye.

View larger version (123K): In this window In a new window Download PPT slide   Figure 2. Eye sections reveal suppression of R7 photoreceptor phenotype. Apical sections through eyes of the following genotypes are shown: (A) P[sE-GAL4]/+; (B) P[sE-KDN]/+; (C) P[sev-RAS1V12]/+; (D) P[sE-KDN]/+; P[sE-GAL4]/P[EP(3)1015]; and (E) P[sev-RAS1V12]/P[EP(2)2221]; P[sE-GAL4]/+. A suppressor of sE-KDN restores the R7 cell fate in most ommatidia (D); however, the suppression does not rescue the polarity defect. A suppressor of sev-RAS1V12 reduces the excess number of R7 cells to one per ommatidium (E).

Misexpression lines that appeared to suppress the sE-KDN rough eye were crossed to the sE-KDN- and sE-GAL4-expressing line. F₁ progeny were fixed, the eyes sectioned, and ommatidia examined for presence of R7 cells. Of the six potential suppressor lines, only four restored the R7 cell fate and to varying degrees (Fig 2D and data not shown). The other two lines were assumed to affect other cell types in the eye and therefore possibly to act independently of the SEV/RAS1/MAPK pathway. The 13 misexpression lines that enhance sE-KDN and suppress sev-RAS1^V12 were examined for number of R7 cells. Of these 13 lines, 9 suppress the supernumerary R7 phenotype in the sev-RAS1^V12 background (Fig 2E and data not shown) and the other 4 do not show a significant change in R7 cell number over background. Thus, 13 of 19 lines that modified the rough eye phenotype also affected R7 cell fate specification. In the remaining 6 lines, the modification of eye roughness may reflect a change in the development of nonneuronal cell types in the eye. Having more carefully examined the effect of misexpressing these 19 lines in two different genetic backgrounds, we concluded that 13 show a clear effect on the R7 cell fate and therefore likely interact with the RAS1 signaling pathway when overexpressed.

Identification of misexpressed genes:
All interacting EP lines were mapped cytologically; comparison of map positions and flanking sequences (see below; LIAO et al. 2000 ) with the genomic sequence revealed that the 13 interacting lines correspond to 12 independent genes (Table 1). The 4 genes that suppress the sE-KDN and enhance the sev-RAS1^V12 phenotypes are designated misexpression suppressors of KDN (MESK 1–4) and the 8 genes that enhance the sE-KDN and suppress the sev-RAS1^V12 phenotypes are designated misexpression suppressors of RAS1^V12 (MESR 1–8). This screen took advantage of P-element insertions in close proximity to the genes of interest. To identify the genes that were being misexpressed, genomic DNA sequence flanking the 13 P elements was obtained by inverse PCR and plasmid rescue (see MATERIALS AND METHODS) and used to search for similar sequences in the NCBI and BDGP databases. Database searches identified 3 previously described Drosophila genes and ESTs corresponding to 6 of the lines. In these six cases where only ESTs were identified, all EST clones corresponding to that particular locus were sized and the longest cDNA was sequenced completely. Flanking genomic sequence from 2 of the lines did not identify either known genes or ESTs and are not described any further in this report. One line, EP(3)3728, maps to cytological position 77A4–B1 and is located 300 nucleotides upstream of predicted gene CG13813. This 1.3-kb predicted mRNA contains one InterPro motif (PS00430), which is characteristic of the Escherichia coli tonB protein involved in shuttling substrates from the outer bacterial membrane to periplasmic space. The other line, EP(2)0980, maps to cytological position 33B3–4. There are no ESTs or predicted genes in the region of this P-element insertion. Further work is required to identify the genes misexpressed by these two insertion lines. The results are summarized in Table 1.

Verification of overexpressed genes: The open reading frame located directly downstream of the UAS promoter-containing P element is most likely the target gene expressed. To confirm that the cDNAs isolated corresponded to the misexpressed gene that modifies RAS1 signaling, these cDNAs were directly tested for their ability to modify the eye phenotypes used in the initial screen. cDNAs for 9 of these genes were individually cloned downstream of a GAL4-inducible promoter. Two independent transgenic lines expressing each cDNA were crossed to the sE-KDN; sE-GAL4 and sev-RAS1^V12; sE-GAL4 lines (with the exception of misshapen, for which only one line was tested). All nine cDNAs gave the same phenotypic modification as the corresponding original EP lines. An example is shown in Fig 3. The strength of the interaction often varied from the original EP line; this difference can be explained most easily by a difference in expression levels (Fig 3B and Fig C). Of the 10 genes identified in this screen, 3 were previously characterized based on other functions: Lk6 (KIDD and RAFF 1997 ; TREISMAN et al. 1997 ; LI et al. 1999 ), misshapen (msn), and Drosophila A kinase anchor protein 200 (Akap200). Six genes are entirely novel in sequence and extensive database searches and sequence comparisons have not revealed any defined protein domains and/or motifs or homologs in other species. These novel genes do not map to any loci from previous RAS1 screens performed in the eye [DICKSON et al. 1996; KARIM et al. 1996; NEUFELD et al. 1998; THERRIEN et al. 1998, 2000 (this issue); REBAY et al. 2000]. Twelve of the 13 EP lines identified in this screen are homozygous viable and show no adult phenotypes. Only the EP insertion in the msn gene is recessive lethal and msn was previously determined to be an essential gene (TREISMAN et al. 1997 ). Other P-element alleles may exist for 3 of the 6 novel genes isolated. The GeneScene map viewer (http://www.fruitfly.org) identified the following P-element lines to map within 5 kb of the respective locus for each of the following genes: MESK2 [l(2)k10317, l(2)01862], MESK4 [l(3)06713], and MESR4 [l(2)k15815]. It is unknown whether these lethal P-element insertions affect the same genes as the EP insertions.

View larger version (153K): In this window In a new window Download PPT slide   Figure 3. Verification that EP(3)0886 is overexpressing the Lk6 gene. Scanning electron micrographs of adult eyes of the following genotypes are shown: (A) P[EP(3)0886]/P[sE-GAL4]; (B) P[sE-KDN]/+; P[sE-GAL4]/P[EP(3)0886]; (C) P[sE-KDN]/+; P[sE-GAL4]/P[UAS-Lk6]; (D) P[sev-RAS1V12]/+; P[sE-GAL4]/+; and (E) P[sev-RAS1V12]/+; P[sE-GAL4]/P[UAS-Lk6]. Misexpression of the original P-element line in a wild-type background has no effect on normal eye development (A); however, it increases the severity of roughness of the sE-KDN rough eye phenotype (B). Misexpressing the Lk6 cDNA shows a stronger enhancement than the original EP line on the sE-KDN rough eye (C) and somewhat suppresses the sev-RAS1V12 rough eye (D and E).

Overexpression phenotype analysis: The UAS-cDNA transgenic lines were also crossed to other tissue-specific sources of GAL4 to determine if misexpression of these genes affected the development of other tissue types. GAL4 drivers used included dpp-GAL4 for expression in wing, leg, eye, and antenna; vestigial-GAL4 for expression in the wing; and actin-GAL4 for ubiquitous overexpression. Except for the examples mentioned below, misexpression of these cDNAs had no apparent effect on normal development.

When overexpressed under the control of Act5C-GAL4, the MESR1 and MESR6 crosses gave rise to 10 and 50% uneclosed F₁ pupae, respectively. The adults that eclosed appeared wild type. When crossed to the other drivers, these two lines gave rise to completely wild-type-appearing F₁ progeny. When the MESR2 cDNA was misexpressed using the vestigial-GAL4 and Actin5C-GAL4 drivers, adult wings of the progeny were reduced in size and very malformed (data not shown). This phenotype appears to be due to loss of intervein wing tissue whereas the amount of wing margin has remained constant or increased. Overexpression of MESR2 in other tissue types did not seem to affect their development.

A new Drosophila rel domain protein:
Sequence analysis of the longest EST clone corresponding to the MESR1 gene reveals that it contains a region that shares high homology with the rel family of transcription factors (Fig 4). Known Drosophila rel proteins are Dorsal, Dif, and Relish (STEWARD 1987 ; IP et al. 1993 ; DUSHAY et al. 1996 ); however, the MESR1 cDNA appears to encode a novel protein. The 4.2-kb cDNA clone contains a predicted open reading frame of 3633 bp in length encoding a predicted protein of 1211 residues. This protein contains a well-conserved rel homology domain (RHD) of 280 residues.

View larger version (81K): In this window In a new window Download PPT slide   Figure 4. A novel Drosophila rel domain-containing protein. (A) The 1211-residue predicted amino acid sequence of a novel rel protein. rel homology domain (RHD) and polyglutamine stretches are underlined and boxed, respectively. (B) Sequence alignment of the RHD from seven different rel proteins (and GenBank accession numbers): Dm, MESR1 identified in this screen (AF195496); human NFAT5 (AF200687); human NFAT1b (U43341); human NF-B p50 (M55643); Drosophila Relish (U62005); Drosophila dorsal (M23702); and Drosophila Dif (L29015). Identities shared in at least four of seven proteins are highlighted in black and similarities in light gray. The 14 residues involved in DNA contact in NFAT1b are marked above with a dot. The invariant histidine in the NFAT1 DNA recognition loop is marked above with an asterisk. The DNA recognition motif is underlined, the NFAT consensus sequence is above, and the rel consensus sequence is below.

The RHD is loosely defined by a 300-amino-acid region containing a DNA-binding domain, nuclear localization signal, and a conserved motif near the C terminus (JAIN et al. 1995 ). Alignment and sequence comparisons of the MESR1 RHD with other rel proteins show that MESR1 is most closely related to the mammalian nuclear factor of activated T-cells (NFAT) family of transcription factors. There are five distinct NFAT genes in human and most are expressed in multiple splice variants; however, the isoforms all contain the same RHD (RAO et al. 1997 ). NFAT1–4 contain a Calcineurin binding domain found only in NFAT family members whereas NFAT5 does not. NFAT5 has two polyglutamine stretches in the C-terminal half of the protein that are not found in NFAT1–4. MESR1 is most closely related to NFAT5; it is 51% identical and 63% similar to NFAT5 in the RHD while it is only 37% identical and 52% similar to the RHD of NFAT1. MESR1 shares even less sequence homology with other Drosophila rel proteins with only 38% identity to Relish, 26% identity to Dorsal, and 23% identity to Dif in the RHD. MESR1 does not share any sequence homology with the Calcineurin binding domain of NFAT1–4. However, it does contain two polyglutamine stretches >20 residues in length, one on either side of the RHD.

The crystal structure of NFAT1 interacting with AP-1 on DNA reveals 14 DNA contact residues in NFAT1, which are completely conserved among NFAT1–4 (CHEN et al. 1998 ). NFAT5 shares 11 of the 14 conserved residues (LOPEZ-RODRIGUEZ et al. 1999 ), and MESR1 shares the same 11 residues (Fig 4). Structural analysis also implicates a histidine residue in the DNA recognition loop of NFAT1–4 that confers binding specificity. However, NFAT5 and MESR1 contain an arginine instead of histidine at this site, which is also conserved in other members of the rel family including all other Drosophila rel proteins and human NF-B p50 (Fig 4 and CHEN et al. 1998 ; LOPEZ-RODRIGUEZ et al. 1999 ). Both the sequence similarity and structural similarities suggest that this novel Drosophila rel protein, MESR1, and human NFAT5 form a new class of rel proteins.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We have performed a gain-of-function modifier screen for enhancers and suppressors of the rough eye phenotype associated with overexpression of a dominant-negative form of KSR in the developing eye; this screen identifies genes that can alter signaling through the RAS1 pathway when misexexpressed. Genetic tests suggest that 12 genes from the EP collection can alter signaling efficiency through the RAS1 pathway when misexpressed in the eye. Of these 12 genes, 3 are previously cloned Drosophila genes, 1 is a new member of the rel family of proteins, 6 are completely novel, and 2 have yet to be molecularly characterized. Below we discuss the possible roles of 4 of these genes in the RAS pathway.

Drosophila Akap200:
One of the misexpression interactors, MESR2, was an insertion upstream of the Akap200 locus. DAKAP200 is Drosophila A kinase anchor protein of molecular weight 200 kd and binds the regulatory II (rII) subunit of cyclic AMP-dependent protein kinase (PKA; LI et al. 1999 ). The Akap200 gene produces two different transcripts, one that contains the binding site for RII and one where the exon encoding for the RII binding site is spliced out to generate a protein that does not interact directly with PKA (LI et al. 1999 ). Both isoforms of AKAP200 are expressed at relatively similar levels throughout development as well as in adult heads (LI et al. 1999 ).

PKA is the principal mediator of signals that activate adenylate cyclase. cAMP signals are often targeted to effectors that accumulate to discrete intracellular locations. This targeting is due to a nonuniform distribution of PKA molecules within cells. In Drosophila, PKA has been implicated in normal developmental events in all imaginal tissues through the Hedgehog signaling pathway (JIANG and STRUHL 1995 ; LI et al. 1995 ; PAN and RUBIN 1995 ) and is involved in signaling pathways that generate cell polarity, which requires it to be localized to distinct intracellular locations.

Subcellular localization of PKA occurs through association with AKAPs. AKAPs are a functionally related family of proteins that are defined by their ability to associate with PKA (RUBIN 1994 ; COLLEDGE and SCOTT 1999 ). Each AKAP contains a unique targeting domain that directs the complex to a defined intracellular location where PKA is placed proximal to both a signal generator (adenylate cyclase) as well as to potential PKA effector molecules. Coordinate binding of specific combinations of enzymes can allow such complexes to respond to distinct second messenger-mediated signals.

Studies in mammalian cells have suggested that PKA signaling via Rap1, another small molecular weight GTP-binding protein, antagonizes RAS1 signaling by competing for RAS pathway components such as B-Raf and MAPK (KITAYAMA et al. 1989 ; VOSSLER et al. 1997 ). However, more recent studies suggest no genetic interaction between Drosophila Rap1 and RAS1 (ASHA et al. 1999 ). In Drosophila, overexpression of Rap1 in a heterozygous RAS1 mutant background had no effect on photoreceptor determination, suggesting no interaction between the two gene functions. A heterozygous Rap1 mutation did not reduce the number of R7 cells in a sev-RAS1^V12 rough eye, also suggesting that the two pathways do not interact (ASHA et al. 1999 ). Although there is no direct evidence linking PKA activation to MAPK activation via Rap1, there may be a still unknown pathway by which these molecules can signal.

Akap200 was isolated as a misexpression enhancer of KDN and suppressor of RAS1^V12 in our screen. This suggests that overexpression of this AKAP decreases signaling through RAS1. Overexpression of an AKAP might cause mislocalization of PKA molecules to the plasma membrane. This could activate a signaling pathway that normally is not utilized in this cell or at this time in development. If PKA and Rap1 are involved in RAS signaling, why were they not uncovered in previous loss-of-function screens? One possibility is that mutations in either gene may not be dose sensitive and therefore be unable to dominantly modify a rough eye phenotype. Another is the possibility that overexpression of an AKAP causes abnormal targets of PKA to become activated. Whether PKA signals through Rap1 is still unclear; however, the reported effects of attenuating RAS1/MAPK signaling are supported by this study. The enhancement of the KDN rough eye phenotype could be due to the additive effects of inefficient signaling due to KDN as well as the attenuation of MAPK by mislocalized PKA. In the activated RAS1^V12 background, the attenuating effects of activated PKA due to mislocalization to the plasma membrane might reduce the amount of signaling through the pathway to suppress the RAS1-dependent rough eye phenotype.

misshapen:
Overexpression of msn in a sE-KDN background enhances the rough eye phenotype. msn encodes the Drosophila homolog of Nck interacting kinase (NIK), a member of the mammalian SPS1 subfamily of the STE20 kinase family (TREISMAN et al. 1997 ; SU et al. 1998 ). msn is an essential gene involved in dorsal closure during embryogenesis in Drosophila and mutant clones result in misshapen rhabdomeres (due to defects in polarity, malformed and missing photoreceptors) in the adult eye (TREISMAN et al. 1997 ). STE20, the founding member of the family in yeast, acts in the pheromone signaling pathway and activates the yeast MAPK module (HERSKOWITZ 1995 ). However, the upstream activators of the STE20 pathway are not known. msn, the STE20 homolog in Drosophila, acts upstream of the c-Jun amino-terminal kinase (JNK) MAPK cascade required for dorsal closure (SU et al. 1998 ) and has recently been implicated to act downstream of the Frizzled receptor in the epithelial planar polarity pathway (PARICIO et al. 1999 ). Previously published results suggest that msn, when overexpressed in the eye in an otherwise wild-type background, can generate a rough eye (PARICIO et al. 1999 ). However, in our experiments we found no effect on eye morphology using sE-GAL4 to drive either UAS-msn or EP(3)0609 and EP(3)0549, the two EP lines upstream of the msn gene (data not shown). Drosophila Jun has been implicated as a downstream target of both the JNK MAPK and RAS1/MAPK signal transduction pathways by overexpression analysis of dominant-negative mutations (KOCKEL et al. 1997 ); however, no other components of either pathway are shared. If the JNK pathway can partially compensate for the RAS1 pathway in eye development, one would expect that msn overexpression would suppress sE-KDN. The overexpression results from our screen suggest that as a misexpression suppressor of RAS1, msn decreases signaling in the pathway. msn may independently inhibit neuronal cell fate, although there is no previous evidence for this. It is possible that the JNK signaling pathway may compete with the RAS pathway for common components. Alternatively, this interaction may be tissue specific and only uncovered in the eye with overexpression. There is also the possibility that misexpression of msn causes promiscuous signaling through an independent pathway that also affects eye morphology. Although we do not see a phenotype when msn is overexpressed in the eye, this situation may sensitize the eye and nonspecifically enhance the sE-KDN phenotype.

Lk6:
Another misexpression interactor, MESR8, corresponded to two independent insertions upstream of the Lk6 gene. The Lk6 gene was originally identified biochemically as a microtubule-binding protein (KIDD and RAFF 1997 ). LK6 localizes to centrosomes in the early blastoderm embryo and appears to be expressed ubiquitously throughout development. Loss-of-function mutations in Lk6 are viable and have no visible adult phenotype (A. M. HUANG and G. M. RUBIN, unpublished results). Constitutive overexpression of Lk6 under the ubiquitin promoter causes microtubule defects in both eggs and embryos; these defects include fragmented mitotic spindles, abnormal asters, and abnormal metaphases, suggesting that LK6 may stabilize microtubules in vivo (KIDD and RAFF 1997 ). Overexpression of phenotypes in the embryo suggests the same. Overexpression of Lk6 does not appear to affect imaginal tissues (KIDD and RAFF 1997 ; A. M. HUANG and G. M. RUBIN, unpublished results).

The identification of a cytoskeleton-associated kinase that can genetically interact with the RAS pathway is consistent with studies in mammalian cells that indicate that the activation of the RAS/MAPK pathway can lead to spindle instability (SAAVEDRA et al. 1999 ). Activated RAS induces chromosome aberrations such as dicentrics, acentrics, and double minutes (DENKO et al. 1994 ). Genomic instability as measured by micronucleus formation increases tenfold in cells expressing activated RAS (SAAVEDRA et al. 1999 ). Micronucleus formation induced by activated MAPK was due largely to disruption of the mitotic spindle rather than double-strand chromosome breaks.

In our experiments the overexpression of Lk6 in the developing eye has no visible phenotype. However, when overexpressed in the sev-RAS1^V12 background, Lk6 is able to partially suppress the associated rough eye by reducing the number of supernumerary R7 cells. Since the supernumerary R7 cells in the sev-RAS^V12 background are generated postmitotically by cone cell to R7 transformations, LK6 may be playing a role in differentiation rather than proliferation in this instance. LK6 may be redundant with other kinases expressed in the eye, normally act at other stages of development, or activate normally inactive signaling pathways when overexpressed. It is possible that expressing LK6 in the sev-RAS^V12 background could have a deleterious effect on cell differentiation and cause the improperly fated photoreceptors to die.

A new rel family member in flies:
The MESR1 interactor identified in this screen is an insertion upstream of a novel gene that encodes a protein containing a rel domain. Sequence comparison of the rel domain with other known proteins of this class identifies it as a distant member of the NFAT family of transcription factors with mammalian NFAT5 its closest relative. NFAT1–4 are highly regulatable transcription factors that are sequestered in the cytoplasm of resting cells of the mammalian immune system. Upon activation of Calcineurin by intracellular calcium release, NFAT1–4 translocate to the nucleus where they bind cooperatively to DNA with the AP-1 heterodimer, FOS and JUN (RAO et al. 1997 ). The mammalian NFAT5 protein is distantly related to the others in sequence homology as well as function. NFAT5 appears to be constitutively nuclear, does not seem to interact with FOS and JUN, and does not transactivate well-characterized NFAT1–4 target promoters; its function and specificity remain unclear (LOPEZ-RODRIGUEZ et al. 1999 ).

Overexpression of this gene in the eye suppresses an activated RAS1 rough eye, implying that it acts to decrease RAS signaling efficiency. This protein may act to promote a nonneuronal cell fate or actively repress neuronal differentiation. There are four putative MAPK phosphorylation sites in the Drosophila NFAT5 homolog but it is unknown whether any of these sites are phosphorylated in vivo. No loss-of-function mutations have been identified to date. This protein may bind to RAS-dependent promoters; identification of target genes will help elucidate the wild-type function of this gene in flies.

We have identified two kinases and one kinase anchor protein that can modulate RAS signaling efficiency when overexpressed. It is possible that increased levels of these signaling proteins lead to the activation of targets, of which the novel rel protein may be one, not normally active in the developing eye. The genetic interaction we see may be the result of activation of signaling pathways not used in eye development, or represent processes found in other tissues or other stages of development.

The gain-of-function screens in animals:
Gain-of-function genetics has been a successful approach in identifying regulatory pathway components in many systems. Multicopy screens in yeast revealed drug-resistant loci (RINE et al. 1983 ), overexpression in mammalian cell culture identified MyoD as a crucial cell fate determinant (DAVIS et al. 1987 ), and overexpression in Xenopus laevis identified Noggin as a potent neural inducer (SMITH and HARLAND 1992 ). In Drosophila, overexpression has identified critical genes involved in development and oogenesis (MILAN et al. 1998 ; RORTH et al. 1998 ).

The genetic screens performed to date looking for genetic interactors with RAS pathway components have been dominant modifier screens [SIMON et al. 1991; DICKSON et al. 1996; KARIM et al. 1996; MAIXNER et al. 1998 ; NEUFELD et al. 1998; THERRIEN et al. 1998, 2000 (this issue); REBAY et al. 2000]. These screens required that a twofold reduction in gene dosage be sufficient to alter the phenotype of a genetically "sensitized" background. In some cases, these mutageneses were done to near saturation and have uncovered many RAS pathway components. However, the observation that two-thirds of all genes are not mutable to any visible phenotype suggests that the function of many genes may not be revealed by loss-of-function mutations. In addition, one concern with dominant modifier screens is that many genes that normally interact with the RAS1/MAPK pathway were not recovered because the reduction in dosage of one gene is compensated by other genes in the genome.

To identify additional genes that can modulate RAS signaling, we screened through a collection of P-element insertions that drive overexpression of otherwise wild-type genes. Six of the genes isolated in this screen have never been previously characterized. These six insertion lines are homozygous viable and show no adult phenotype (A. M. HUANG and G. M. RUBIN, unpublished observations). To exclude lines that modify the RAS-dependent rough eye due to the additive effect of two causally unrelated phenotypes, we screened for lines that have opposite effects on phenotypes due to decreased (sE-KDN) or increased (sev-RAS1^V12) signaling through the RAS pathway. Further work is required to determine the specificity of these interactions, the functions of these molecules in development, and the roles they may play in RAS signaling events.

	ACKNOWLEDGMENTS

We thank Chris Suh, Garson Tsang, Allan Wong, Martha Evans-Holm, Susan Mullaney, Alma Valeros, and Todd Laverty for technical assistance; J. Triesman for flies; and Marc Therrien, Mike Brodsky, Brian Avery, and Bobby Williams for critical reading of the manuscript. This work was supported by the Howard Hughes Medical Institute and National Institutes of Health grant HG-00750.

Manuscript received May 3, 2000; Accepted for publication July 25, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ASHA, H., N. D. DE RUITER, M. G. WANG, and I. K. HARIHARAN, 1999  The Rap1 GTPase functions as a regulator of morphogenesis in vivo. EMBO J. 18:605-615[Medline].

ASHBURNER, M., S. MISRA, J. ROOTE, S. E. LEWIS, and R. BLAZEJ et al., 1999  An exploration of the sequence of a 2.9-Mb region of the genome of Drosophila melanogaster: the Adh region. Genetics 153:179-219[Abstract/Free Full Text].

BENDER, A. and J. R. PRINGLE, 1989  Multicopy suppression of the cdc24 budding defect in yeast by CDC42 and three newly identified genes including the ras-related gene RSR1. Proc. Natl. Acad. Sci. USA 86:9976-9980[Abstract/Free Full Text].

BOLLAG, G. and F. MCCORMICK, 1991  Regulators and effectors of ras proteins. Annu. Rev. Cell Biol. 7:601-632.

BRAND, A. H. and N. PERRIMON, 1993  Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118:401-415[Abstract].

BRAND, A. H., A. S. MANOUKIAN, and N. PERRIMON, 1994  Ectopic expression in Drosophila. Methods Cell Biol. 44:635-654[Medline].

CHEN, L., J. N. GLOVER, P. G. HOGAN, A. RAO, and S. C. HARRISON, 1998  Structure of the DNA-binding domains from NFAT, Fos and Jun bound specifically to DNA. Nature 392:42-48[Medline].

COLLEDGE, M. and J. D. SCOTT, 1999  AKAPs: from structure to function. Trends Cell Biol. 9:216-221[Medline].

DAVIS, R. L., H. WEINTRAUB, and A. B. LASSAR, 1987  Expression of a single transfected cDNA converts fibroblasts to myoblasts. Cell 51:987-1000[Medline].

DENKO, N. C., A. J. GIACCIA, J. R. STRINGER, and P. J. STAMBROOK, 1994  The human Ha-ras oncogene induces genomic instability in murine fibroblasts within one cell cycle. Proc. Natl. Acad. Sci. USA 91:5124-5128[Abstract/Free Full Text].

DENOUEL-GALY, A., E. M. DOUVILLE, P. H. WARNE, C. PAPIN, and D. LAUGIER et al., 1998  Murine Ksr interacts with MEK and inhibits Ras-induced transformation. Curr. Biol. 8:46-55[Medline].

DICKSON, B. J., A. VAN DER STRATEN, M. DOMINGUEZ, and E. HAFEN, 1996  Mutations modulating Raf signaling in Drosophila eye development. Genetics 142:163-171[Abstract].

DUSHAY, M. S., B. ASLING, and D. HULTMARK, 1996  Origins of immunity: Relish, a compound Rel-like gene in the antibacterial defense of Drosophila. Proc. Natl. Acad. Sci. USA 93:10343-10347[Abstract/Free Full Text].

FORTINI, M. E., M. A. SIMON, and G. M. RUBIN, 1992  Signalling by the sevenless protein tyrosine kinase is mimicked by Ras1 activation. Nature 355:559-561. [see comments][Medline].

HERSKOWITZ, I., 1995  MAP kinase pathways in yeast: for mating and more. Cell 80:187-197[Medline].

HUANG, A. M., E. J. REHM and G. M. RUBIN, 2000 Inverse PCR, pp. 429–437 in Drosophila Protocols, edited by W. SULLIVAN, M. ASHBURNER and R. S. HAWLEY. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

IP, Y. T., M. REACH, Y. ENGSTROM, L. KADALAYIL, and H. CAI et al., 1993  Dif, a dorsal-related gene that mediates an immune response in Drosophila. Cell 75:753-763[Medline].

JAIN, J., E. BURGEON, T. M. BADALIAN, P. G. HOGAN, and A. RAO, 1995  A similar DNA-binding motif in NFAT family proteins and the Rel homology region. J. Biol. Chem. 270:4138-4145[Abstract/Free Full Text].

JIANG, J. and G. STRUHL, 1995  Protein kinase A and hedgehog signaling in Drosophila limb development. Cell 80:563-572[Medline].

KARIM, F. D., H. C. CHANG, M. THERRIEN, D. A. WASSARMAN, and T. LAVERTY et al., 1996  A screen for genes that function downstream of Ras1 during Drosophila eye development. Genetics 143:315-329[Abstract].

KIDD, D. and J. W. RAFF, 1997  LK6, a short lived protein kinase in Drosophila that can associate with microtubules and centrosomes. J. Cell Sci. 110:209-219[Abstract].

KITAYAMA, H., Y. SUGIMOTO, T. MATSUZAKI, Y. IKAWA, and M. NODA, 1989  A ras-related gene with transformation suppressor activity. Cell 56:77-84[Medline].

KOCKEL, L., J. ZEITLINGER, L. M. STASZEWSKI, M. MLODZIK, and D. BOHMANN, 1997  Jun in Drosophila development: redundant and nonredundant functions and regulation by two MAPK signal transduction pathways. Genes Dev. 11:1748-1758. [erratum: Genes Dev. 12: 447].[Abstract/Free Full Text].

KORNFELD, K., D. B. HOM, and H. R. HORVITZ, 1995  The ksr-1 gene encodes a novel protein kinase involved in Ras-mediated signaling in C. elegans. Cell 83:903-913. [see comments][Medline].

LI, W., J. T. OHLMEYER, M. E. LANE, and D. KALDERON, 1995  Function of protein kinase A in hedgehog signal transduction and Drosophila imaginal disc development. Cell 80:553-562[Medline].

LI, Z., E. A. ROSSI, J. D. HOHEISEL, D. KALDERON, and C. S. RUBIN, 1999  Generation of a novel A kinase anchor protein and a myristoylated alanine-rich C kinase substrate-like analog from a single gene. J. Biol. Chem. 274:27191-27200[Abstract/Free Full Text].

LIAO, G., E. J. REHM, and G. M. RUBIN, 2000  Insertion site preferences of the P transposable element in drosophila melanogaster. Proc. Natl. Acad. Sci. USA 97:3347-3351. [In Process Citation][Abstract/Free Full Text].

LOPEZ-RODRIGUEZ, C., J. ARAMBURU, A. S. RAKEMAN, and A. RAO, 1999  NFAT5, a constitutively nuclear NFAT protein that does not cooperate with Fos and Jun. Proc. Natl. Acad. Sci. USA 96:7214-7219[Abstract/Free Full Text].

MAIXNER, A., T. P. HECKER, Q. N. PHAN, and D. A. WASSARMAN, 1998  A screen for mutations that prevent lethality caused by expression of activated sevenless and Ras1 in the Drosophila embryo. Dev. Genet. 23:347-361[Medline].

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

MILAN, M., F. J. DIAZ-BENJUMEA, and S. M. COHEN, 1998  Beadex encodes an LMO protein that regulates Apterous LIM-homeodomain activity in Drosophila wing development: a model for LMO oncogene function. Genes Dev. 12:2912-2920[Abstract/Free Full Text].

NEUFELD, T. P., A. H. TANG, and G. M. RUBIN, 1998  A genetic screen to identify components of the sina signaling pathway in Drosophila eye development. Genetics 148:277-286[Abstract/Free Full Text].

PAN, D. and G. M. RUBIN, 1995  cAMP-dependent protein kinase and hedgehog act antagonistically in regulating decapentaplegic transcription in Drosophila imaginal discs. Cell 80:543-552[Medline].

PARICIO, N., F. FEIGUIN, M. BOUTROS, S. EATON, and M. MLODZIK, 1999  The Drosophila STE20-like kinase misshapen is required downstream of the Frizzled receptor in planar polarity signaling. EMBO J. 18:4669-4678[Medline].

RAMER, S. W., S. J. ELLEDGE, and R. W. DAVIS, 1992  Dominant genetics using a yeast genomic library under the control of a strong inducible promoter. Proc. Natl. Acad. Sci. USA 89:11589-11593[Abstract/Free Full Text].

RAO, A., C. LUO, and P. G. HOGAN, 1997  Transcription factors of the NFAT family: regulation and function. Annu. Rev. Immunol. 15:707-747[Medline].

REBAY, I., F. CHEN, F. HSIAO, P. A. KOLODZIEJ, and B. H. KUANG et al., 2000  A genetic screen for novel components of the Ras/mitogen-activated protein kinase signaling pathway that interact with the yan gene of Drosophila identifies split ends, a new RNA recognition motif-containing protein. Genetics 154:695-712[Abstract/Free Full Text].

RINE, J., W. HANSEN, E. HARDEMAN, and R. W. DAVIS, 1983  Targeted selection of recombinant clones through gene dosage effects. Proc. Natl. Acad. Sci. USA 80:6750-6754[Abstract/Free Full Text].

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

RORTH, 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].

RUBIN, C. S., 1994  A kinase anchor proteins and the intracellular targeting of signals carried by cyclic AMP. Biochim. Biophys. Acta 1224:467-479[Medline].

RUBIN, G. M. and A. C. SPRADLING, 1982  Genetic transformation of Drosophila with transposable element vectors. Science 218:348-353[Abstract/Free Full Text].

SAAVEDRA, H. I., K. FUKASAWA, C. W. CONN, and P. J. STAMBROOK, 1999  MAPK mediates RAS-induced chromosome instability. J. Biol. Chem. 274:38083-38090[Abstract/Free Full Text].

SEGER, R. and E. G. KREBS, 1995  The MAPK signaling cascade. FASEB J. 9:726-735[Abstract].

SIMON, M. A., D. D. BOWTELL, G. S. DODSON, T. R. LAVERTY, and G. M. 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].

SMITH, W. C. and R. M. HARLAND, 1992  Expression cloning of noggin, a new dorsalizing factor localized to the Spemann organizer in Xenopus embryos. Cell 70:829-840[Medline].

STEWARD, R., 1987  Dorsal, an embryonic polarity gene in Drosophila, is homologous to the vertebrate proto-oncogene, c-rel. Science 238:692-694[Abstract/Free Full Text].

SU, Y. C., J. E. TREISMAN, and E. Y. SKOLNIK, 1998  The Drosophila Ste20-related kinase misshapen is required for embryonic dorsal closure and acts through a JNK MAPK module on an evolutionarily conserved signaling pathway. Genes Dev. 12:2371-2380[Abstract/Free Full Text].

SUNDARAM, M. and M. HAN, 1995  The C. elegans ksr-1 gene encodes a novel Raf-related kinase involved in Ras-mediated signal transduction. Cell 83:889-901. [see comments][Medline].

THERRIEN, M., H. C. CHANG, N. M. SOLOMON, F. D. KARIM, and D. A. WASSARMAN et al., 1995  KSR, a novel protein kinase required for RAS signal transduction. Cell 83:879-888. [see comments][Medline].

THERRIEN, M., N. R. MICHAUD, G. M. RUBIN, and D. K. MORRISON, 1996  KSR modulates signal propagation within the MAPK cascade. Genes Dev. 10:2684-2695[Abstract/Free Full Text].

THERRIEN, M., A. M. WONG, and G. M. RUBIN, 1998  CNK, a RAF-binding multidomain protein required for RAS signaling. Cell 95:343-353[Medline].

THERRIEN, M., D. K. MORRISON, A. M. WONG, and G. M. RUBIN, 2000  A genetic screen for modifiers of a KSR-dependent rough eye phenotype in Drosophila. Genetics 156:1231-1242[Abstract/Free Full Text].

TOBA, G., T. OHSAKO, N. MIYATA, T. OHTSUKA, and K. H. SEONG et al., 1999  The gene search system. A method for efficient detection and rapid molecular identification of genes in Drosophila melanogaster. Genetics 151:725-737[Abstract/Free Full Text].

TOMLINSON, A., D. D. BOWTELL, E. HAFEN, and G. M. RUBIN, 1987  Localization of the sevenless protein, a putative receptor for positional information, in the eye imaginal disc of Drosophila. Cell 51:143-150[Medline].

TREISMAN, J. E., N. ITO, and G. M. RUBIN, 1997  misshapen encodes a protein kinase involved in cell shape control in Drosophila. Gene 186:119-125[Medline].

VOSSLER, M. R., H. YAO, R. D. YORK, M. G. PAN, and C. S. RIM et al., 1997  cAMP activates MAP kinase and Elk-1 through a B-Raf- and Rap1-dependent pathway. Cell 89:73-82[Medline].

YU, W., W. J. FANTL, G. HARROWE, and L. T. WILLIAMS, 1998  Regulation of the MAP kinase pathway by mammalian Ksr through direct interaction with MEK and ERK. Curr. Biol. 8:56-64[Medline].

ZIPURSKY, S. L. and G. M. RUBIN, 1994  Determination of neuronal cell fate: lessons from the R7 neuron of Drosophila. Annu. Rev. Neurosci. 17:373-397[Medline].