Identification of Autosomal Regions Involved in Drosophila Raf Function

Corresponding author: Norbert Perrimon, Department of Genetics, Harvard Medical School, 200 Longwood Ave., Boston, MA 02115., perrimon{at}rascal.med.harvard.edu (E-mail)

Communicating editor: T. SCHÜPBACH

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Raf is an essential downstream effector of activated p21^Ras (Ras) in transducing proliferation or differentiation signals. Following binding to Ras, Raf is translocated to the plasma membrane, where it is activated by a yet unidentified "Raf activator." In an attempt to identify the Raf activator or additional molecules involved in the Raf signaling pathway, we conducted a genetic screen to identify genomic regions that are required for the biological function of Drosophila Raf (Draf). We tested a collection of chromosomal deficiencies representing 70% of the autosomal euchromatic genomic regions for their abilities to enhance the lethality associated with a hypomorphic viable allele of Draf, Draf^Su2. Of the 148 autosomal deficiencies tested, 23 behaved as dominant enhancers of Draf^Su2, causing lethality in Draf^Su2 hemizygous males. Four of these deficiencies identified genes known to be involved in the Drosophila Ras/Raf (Ras1/Draf) pathway: Ras1, rolled (rl, encoding a MAPK), 14-3-3, and bowel (bowl). Two additional deficiencies removed the Drosophila Tec and Src homologs, Tec29A and Src64B. We demonstrate that Src64B interacts genetically with Draf and that an activated form of Src64B, when overexpressed in early embryos, causes ectopic expression of the Torso (Tor) receptor tyrosine kinase-target gene tailless. In addition, we show that a mutation in Tec29A partially suppresses a gain-of-function mutation in tor. These results suggest that Tec29A and Src64B are involved in Tor signaling, raising the possibility that they function to activate Draf. Finally, we discovered a genetic interaction between Draf^Su2 and Df(3L)vin5 that revealed a novel role of Draf in limb development. We find that loss of Draf activity causes limb defects, including pattern duplications, consistent with a role for Draf in regulation of engrailed (en) expression in imaginal discs.

THE Raf serine/threonine kinase is an essential effector, downstream of Ras, in mediating the transmission of signals that control cellular proliferation, differentiation, and development (reviewed by AVRUCH et al. 1994 ; MARSHALL 1994 ; MOODIE and WOLFMAN 1994 ). In a simple linear model, Ras, Raf, MEK, and MAPK are sequentially activated following the activation of a cell surface receptor tyrosine kinase (RTK). While the signaling cascade from Raf to MAPK involves direct phosphorylation, it is not known precisely how Raf is activated following Ras activation (reviewed by MORRISON and CUTLER 1997 ). It has been proposed that the sole function of Ras in Raf activation is to translocate Raf to the plasma membrane, where Raf activation is regulated by an unknown factor (LEEVERS et al. 1994 ; STOKOE et al. 1994 ). However, from our analyses of Draf we have proposed that in addition to targeting Raf to the plasma membrane, Drosophila Ras (Ras1) has a second function and activates the "Raf activator," which in turn activates Raf (LI et al. 1998 ). Such results present a more complex picture for Raf activation and predict the presence of a hypothetical factor, the Raf activator, that is also regulated by Ras. Draf is structurally and functionally homologous to mammalian Raf-1. Human Raf-1 is 46% identical in amino acid sequence to Draf and is able to substitute for Draf for viability and signal transduction in Drosophila (AMBROSIO et al. 1989 ; CASANOVA et al. 1994 ; A. BRAND, X. LU and N. PERRIMON, unpublished data). Thus, we expect the mechanism of activation of Draf to be evolutionarily conserved.

The Ras1/Draf/MEK/MAPK signaling cassette is commonly used for signaling by a number of Drosophila RTKs, such as Torso (Tor; reviewed by DUFFY and PERRIMON 1994 ), the epidermal growth factor receptor (EGFR or DER; BRAND and PERRIMON 1994 ), Sevenless (Sev), and the fibroblast growth factor (FGF) receptor homologs Breathless (Btl; KLAMBT et al. 1992 ) and Heartless (Htl; BEIMAN et al. 1996 ; GISSELBRECHT et al. 1996 ). These RTKs are involved in various developmental decisions. The Tor pathway specifies cell fates at the embryonic termini. Btl and Htl are required for cell migration and differentiation of the embryonic tracheal system and mesoderm, respectively. The Sev pathway functions exclusively in the eye to specify R7 photoreceptor cell fate. In contrast, the EGFR has multiple functions throughout development and is required during oogenesis and embryogenesis, as well as the development of wing vein and the eye. All these Drosophila RTKs activate the Ras1/Draf/MEK/MAPK cassette to mediate signal transduction to the nucleus, and perturbation of Draf activity impedes signaling processes of these RTKs, resulting in visible phenotypes and/or lethality to the animal.

To genetically isolate the potential Raf activator, as well as new components of the Raf signaling cascade, we conducted an F₁ screen for modifiers of Draf function. Here we describe an F₁ screen to identify genomic regions on the second and third chromosomes that enhance the lethality of a mutation in Draf (Draf^Su2) that is associated with reduced activity (LU et al. 1994 ). We demonstrate that some of the deficiencies identify genes known to be involved in Draf signaling, thus validating the specificity of this screen. We also identified regions that were not described in previous screens and potentially contain new components involved in Draf signaling. Interestingly, two of these deficiencies remove Src64B and Tec29A, respectively, raising the possibility that these cytoplasmic tyrosine kinases are directly involved in Raf activation. Finally, we describe a novel role of Draf in limb patterning. We found that reduction in Draf function caused limb defects, including notching of the posterior wing and anterior pattern duplications in the posterior compartment. Such defects are similar to those caused by misregulation of engrailed (en) in the imaginal discs, suggesting a possible role for Draf in the regulation of en expression during imaginal disc development.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Stocks:
The deficiency kit stocks for second and third chromosomes were kindly provided by the Bloomington Stock Center (Bloomington, IN; http://flystocks.bio.indiana.edu/df-kit.htm). The Draf^Su2 allele is the same as Su(Draf^C110)² described in LU et al. 1994 . Transgenic flies bearing the hsp70-Src64B⁵⁴⁰ transgene are as described in KUSSICK et al. 1993 .

Genetic screens:
Virgin females homozygous for Draf^Su2 were crossed to males of each mutant or deficiency-bearing stock. Multiple broods for each cross were made when necessary to ensure scoring of >100 progeny. A mutant was considered as an enhancer of Draf^Su2 if far fewer Draf^Su2/Y; m/+ males were recovered compared to the number of Draf^Su2/Y; +/Balancer males. Similarly, a deficiency was considered to enhance Draf^Su2 if far fewer deficiency-bearing males were recovered, relative to the number of balancer-bearing Draf^Su2 males.

Examination of embryos:
To determine the effect of Src64B⁵⁴⁰ overexpression on tll expression, one copy of the hsp70-Src64B⁵⁴⁰ transgene (KUSSICK et al. 1993 ) was introduced from the father by crossing wild-type females with males homozygous for the hsp70-Src64B⁵⁴⁰ transgene. Embryos (0- to 1-hr old) were collected on agar plates and allowed to develop for an additional hour at 25°. They were heat shocked by floating the plates in a 37° water bath for 50 min, cooled to 4°, then fixed for in situ hybridization using a tll probe.

tll mRNA expression in embryos was examined by whole-mount in situ hybridization as described in LI et al. 1997 . Embryos were mounted in Euparal (Carolina Biological Supply) following in situ staining or Hoyer's mountant for cuticle preparations.

Generation of partially rescued Draf null males:
To generate Draf null males, Draf^11-29 females were crossed to males carrying an autosomal insertion of a wild-type Draf gene under the control of the heat-shock promoter (DIAZ-BENJUMEA and HAFEN 1994 ) to create a line that was Draf^11-29/FM7; hs-Draf^BT127/hs-Draf^BT127. This line could then be used to generate partially rescued Draf males, or outcrossed to a line carrying a homozygous viable en-lacZ insertion, to generate partially rescued Draf null males in which engrailed expression could be assessed. If provided with daily heat shocks from embryonic stages through late pupal stages, this transgene will rescue Draf^11-29 males to full viability and fertility.

To examine the role of Draf in the development of the imaginal discs, heat shocks were provided daily (2 hr, 37° waterbath) until the beginning of the second larval instar. Then no further heat shocks were provided until 24–36 hr after pupation, when daily heat shocks were again provided until adults eclosed. Imaginal discs were dissected from wandering third instar larvae, fixed briefly in 4% methanol-free formaldehyde, and stained with X-gal according to established procedures. To examine adult structures, legs and wings were mounted in Hoyer's mountant.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Design of a sensitized screen:
To genetically isolate the potential Raf activator or molecules involved in the Raf signaling pathway, we conducted an F₁ screen for modifiers of Draf function. Among a series of Draf mutant alleles, one particular allele, Draf^Su2, which carries two point mutations in Draf, appeared suitable for such a screen. Unlike most other Draf alleles, Draf^Su2 homozygous females or hemizygous males are viable. Draf^Su2 was isolated as a suppressor of Draf^C110, a hypomorphic mutation associated with late zygotic lethality. Draf^C110 carries a point mutation, R217L, that prevents the Draf:Ras1 interaction (MELNICK et al. 1993 ; HOU et al. 1995 ). Arg217 in Draf is equivalent to Arg89 in Raf-1, which is a critical amino acid residue lying in the first Ras-binding domain (RBD) of Raf-1 and is essential for its binding to activated Ras (FABIAN et al. 1994 ). In a screen for dominant suppressors of Draf^C110, LU et al. 1994 isolated Su2 that restores viability to Draf^C110 flies. Su2 is an intragenic point mutation [P308L] in the second RBD of Draf, also known as the cysteine-rich domain (CRD). Su2 increases the biological activity of Draf^C110 without restoring the binding of Draf^C110 with Ras1 (LI et al. 1998 ). Although flies carrying both the original Draf^C110 mutation, R217L, and the Su2 mutation, P308L, are fully viable as homozygotes (referred to as Draf^Su2), they are only 18% viable in trans with Draf^C110, and are lethal with a null allele of Draf (LU et al. 1994 ). This suggests that further reduction of Draf activity, such as that resulting from mutation of one copy of its activator or downstream signaling components, should reduce the viability of Draf^Su2 flies. Since the Draf^Su2 protein does not bind to Ras1, it should rely more on the hypothetical Raf activator for its activity and therefore be more sensitive to changes in the levels of such an activator.

Validity of the screen:
To validate the hypothesis this screen was based upon, we tested available mutant alleles of genes that are known to be involved in the Ras1/Draf signaling pathway (Table 1). Since Draf^Su2 encodes a mutant protein that has reduced Draf function, we predicted that it might be most sensitive to changes in the dosage of genes that act close to Draf in the signaling pathway. In support of this hypothesis, mutant alleles of Ras1 (SIMON et al. 1991 ), Ksr (THERRIEN et al. 1995 ), Dsor (encoding a MEK or MAPKK; TSUDA et al. 1993 ), 14-3-3, and rolled (rl, encoding a MAPK; BRUNNER et al. 1994 ) enhanced Draf^Su2, causing lethality in trans with Draf^Su2 hemizygous males (Table 1). These results indicate that Draf^Su2 can provide a sensitive background for isolating additional components required for Draf function and particularly for its activation.

The autosomal deficiencies that removed the genes rl, Ras1, and 14-3-3 enhanced the lethality of Draf^Su2 hemizygous males. In(2R)bwV^De2LCy^R carries a deficiency of the 41A–B; 42A2–3 interval and removes rl. This deficiency behaved as a strong enhancer of Draf^Su2 (Table 2), indicating that the lethality of Draf^Su2 observed in trans with In(2R)bwV^De2LCy^R is due to removal of at least rl. Df(3R)by10 deletes the 85D8–12; 85E7–F1 interval and removes Ras1. This deficiency strongly enhances Draf^Su2 lethality (Fig 1 and Table 2). Therefore, the lethality of Draf^Su2 observed in trans with Df(3R)by10 is due at least in part to haploinsufficiency of Ras1. Finally, Df(3R)DG2, which removes 89E1–F4;91B1–2, behaved as a strong enhancer of Draf^Su2. Df(3R)C4, removing 89E; 90A, did not display any enhancement of the phenotype. Two genes located in this region have been previously implicated in Draf signaling: Suppressor of Ras85D 3-6 (SR3-6; KARIM et al. 1996 ), which maps to 90D–E, and 14-3-3, which maps to 90F6–7 (CHANG and RUBIN 1997 ). A 14-3-3 null allele was tested and was found to modestly enhance Draf^Su2 (Table 1 and Table 2).

View larger version (36K): In this window In a new window Download PPT slide   Figure 1. Autosomal deficiencies screened for enhancement of DrafSu2. The four major autosomal arms are represented as solid lines. Numbers above each line represent the cytological regions. Boxes below each autosomal arm represent the approximate genomic regions removed by each deficiency tested in this study. Solid and open boxes represent the deficiencies that did and did not enhance DrafSu2, respectively. Shaded boxes indicate that the results were uninterpretable due to haploinsufficiency of the deficiency tested. The locations of possible candidate genes are indicated above the respective autosomal arms. However, they may not enhance DrafSu2 in this study. Regions where no obvious candidate genes were found are indicated by a question mark.

Autosomal regions that enhance Draf^Su2 lethality:
In this section we describe the deficiencies that enhance Draf^Su2. FLYBASE 1999 search for some deficiency regions allowed us to further identify some of the genes that interact with Draf^Su2.

21A1; 21B7–8: This cytogenetic interval is removed by Df(2L)net-PMF, which enhances Draf^Su2 lethality. No obvious candidate gene could be identified in this region.

23A1–2; 23C3–5: Df(2L)C144 removes the cytogenetic interval 23A1–2; 23C3–5 and acts as a moderate enhancer of Draf^Su2. This region contains a previously identified locus Su(Raf)2A that suppresses a rough eye phenotype generated by expression of an activated form of Draf (Draf^Tor) in the eye and therefore is thought to act positively in regulation of Draf function (DICKSON et al. 1996 ). The molecular nature of the Su(Raf)2A mutation is not known and mutants were not available for testing in the Draf^Su2 background.

35F6–12; 36A8–9: Draf^Su2 is enhanced by Df(2L)cact-255rv64, which removes 35F6–12; 36D, but not by Df(2L)H20, which deletes 36A8–9; 36E1–2, suggesting that the cytogenetic region 35F6–12; 36A8–9 contains an enhancer of Draf^Su2. The gene dachshund (dac) is mapped to 36A1–2 and is disrupted by Df(2L)cact-255rv64. dac was first isolated as a dominant suppressor of an activated EGFR mutation, Ellipse (Elp). It also enhances a partial loss-of-function mutation in EGFR during eye development, and therefore is a positive regulator of the EGFR signaling pathway (MARDON et al. 1994 ). Thus, it is possible that the enhancement of Draf^Su2 is due to removing half a dose of dac; however, this was not confirmed by using a dac allele (Table 2).

36E1–2; 37B9–C1: This cytological interval is defined by two overlapping deficiencies: Df(2L)TW137, which removes 36C2–4; 37B9–C1 and behaved as a strong enhancer of Draf^Su2, and Df(2L)H20, which removes 36A8–9; 36E1–2 and is completely viable as trans-heterozygotes in the Draf^Su2 background. The gene tailup (tup), which maps to 37A2–6, is disrupted by Df(2L)TW137. tup, also known as islet (isl), encodes a LIM domain protein (THOR and THOMAS 1997 ) that mediates Tor signaling (STRECKER et al. 1991 ). However, a tup allele tested did not show enhancement of Draf^Su2 (Table 2).

44C4–5; 44E2–4: This cytological region is removed by Df(2R)44CE, which enhanced Draf^Su2. A candidate enhancer in this region is peanut (pnut; NEUFELD and RUBIN 1994 ), encoding a Septin homolog required for cytokenesis. pnut was first identified as E(sina)8, which genetically interacts with seven in absensia (sina; CARTHEW et al. 1994 ), a gene encoding a nuclear protein required for the correct development of R7 photoreceptor cells in the Drosophila eye. pnut has been demonstrated to act positively in the Sev pathway. Loss of pnut could therefore enhance Draf^Su2. However, a pnut allele did not show enhancement of Draf^Su2 (Table 2).

46A; 46C: Df(2R)B5 removes region 46A; 46C and behaves as a weak enhancer of Draf^Su2. No previously identified components of the Ras/Raf signaling pathway are located in this region. Df(2R)B5 disrupts, among other genes, Mef2 (RANGANAYAKULU et al. 1995 ), which encodes the Drosophila homolog of the myocyte enhancing factor 2 and functions during embryonic muscle development (BOUR et al. 1995 ; TAYLOR et al. 1995 ). A loss-of-function allele of Mef2 did not show enhancement of Draf^Su2 male lethality (Table 2).

49C1–4; 49E2: This cytological interval is defined by two overlapping deficiencies Df(2R)CX1 and Df(2R) Vg-C, both of which behaved as weak enhancers of Draf^Su2. A gene disrupted by these two deficiencies is E(EGFR)B56, which dominantly enhances mutations in EGFR [PRICE et al. 1997; E(EGFR)B5 was not available for testing]. Df(2R)CX1 additionally disrupted drk, a known component of the Ras1/Draf signaling pathway (SIMON et al. 1993 ). However, a drk allele did not show enhancement (Table 2).

57D11–12; 58A1–2: This cytological region is defined by the overlapping deficiencies Df(2R)AA21, Df(2R) X58-7, and Df(2R)Pu-D17. Df(2R)AA21 and Df(2R)X58-7 remove 56F9–17; 57D11–12 and 58A1–2; 58E4–10, respectively, and did not enhance the lethality of Draf^Su2 males. However, Df(2R)Pu-D17, which deletes the 57B4; 58B region, behaved as a strong enhancer of Draf^Su2. Df(2R)Pu-D17 disrupts the EGFR gene, which maps at 57F1. However, neither of the two mutant alleles of EGFR tested were able to dominantly enhance the lethality of Draf^Su2 males (Table 2).

62F; 63D: This cytological region is removed by Df(3L) M21, which behaves as a strong enhancer of Draf^Su2. This deficiency disrupts Hsp83, which has been shown to be required for Draf signaling (VAN DER STRATEN et al. 1997 ; not available for testing).

67A2; 67D13: This cytological region is defined by Df(3L)AC1, which behaves as a weak enhancer of Draf^Su2. There are no obvious candidate genes that might be involved in Draf signaling in this region.

81F3–6; 82F5–7: Df(3R)ME15 removes this region and behaved as a strong enhancer of Draf^Su2. A candidate gene disrupted by this deficiency is huckebein (hkb), a Tor target gene involved in the specification of terminal cell fates (BRONNER and JAECKLE 1991 ). However, this was not confirmed by testing a hkb mutation (Table 2).

86E2–4; 87B1–5: This genomic region is defined by the two overlapping deficiencies Df(3R)M-Kx1, removing 86C1; 87B1–5, and Df(3R)T-32, removing 86E2–4; 87C6–7. Both deficiencies behaved as strong enhancers of Draf^Su2. A candidate gene in the region is seven up (svp), which maps to 87B4 and is deleted in both Df(3R)M-Kx1 and Df(3R)T-32. svp encodes a member of the steroid receptor gene superfamily and is required for photoreceptor cell fates during eye development (MLODZIK et al. 1990 ). Further, all the components of the Ras1/Draf pathway are required for the function of svp (BEGEMANN et al. 1995 ). However, a svp allele tested did not enhance Draf^Su2 (Table 2).

91F5; 92B3: This region is defined by deficiencies Df(3R)Cha7, removing 90F1–2; 91F5, Df(3R)Dl-BX12, removing 91F1–2; 92D3–6, and Df(3R)H-B79, removing 92B3; 92F13. Df(3R)Dl-BX12 behaved as an intermediate enhancer of Draf^Su2, while the other two did not, thus narrowing the genomic region required for Draf^Su2 function to 91F5; 92B3. A candidate gene in the region is branchless (bnl), which encodes the Drosophila FGF homolog. bnl is located at 92B2–3 and is deleted by Df(3R)Dl-BX12 (SUTHERLAND et al. 1996 ). Bnl functions as a ligand for the Btl RTK to specify the tracheal branching pattern. However, a hypomorphic P-element allele of bnl (Table 2) did not enhance Draf^Su2.

99C8; 100C: This region is defined by the two overlapping deficiencies Df(3R)B81, removing 99C8; 100F5, Df(3R)awd-KRB, removing 100C; 100D, and Df(3R)faf-BP, removing 100D; 100F5. Of these three deficiencies, only Df(3R)B81 behaved as a strong enhancer of Draf^Su2. The gene tll that maps to 100B1 was a possible candidate; however, a loss-of-function tll mutation did not enhance Draf^Su2.

Su(tor)⁸⁵ is allelic to bowl and is responsible for the enhancement of Draf^Su2 by Df(2L)sc19-8:
24C2–8; 25A5: This cytological region is defined by the overlapping deficiencies Df(2L)sc19-8, removing 24C2–8; 25C8–9, and Df(2L)sc19-4, removing 25A5; 25E5. Df(2L)sc19-8 behaved as a strong enhancer of Draf^Su2, while Df(2L) sc19-4 did not. The gene bowel (bowl) maps to 24C2–3 and is disrupted by Df(2L)sc19-8. bowl is a terminal class gene regulated by the Tor target gene tll and encodes a zinc-finger protein that mediates a subset of tll functions in specifying the hindgut and proventriculus in the posterior of the embryo (WANG and COULTER 1996 ). In trans-heterozygotes (see Table 2 and below) a bowl mutation enhanced Draf^Su2, indicating that the lethality of Draf^Su2 observed in trans with Df(2L)sc19-8 is likely due to removal of bowl.

A previous screen for suppressors of a gain-of-function tor allele identified a number of loci, referred to as Su(tor), that may encode signaling components downstream from the Tor RTK (DOYLE and BISHOP 1993 ). We surveyed the existing collection of Su(tor) alleles for their abilities to enhance Draf^Su2. We tested eight Su(tor) alleles that were available (17-29-5, 85, 293-12-3, 307-14, 321-16-5, 337-73, 341, and 404-9-1; DOYLE and BISHOP 1993 ). Among these, Su(tor)⁴⁰⁴ and Su(tor)⁸⁵ enhanced Draf^Su2 [i.e., Draf^Su2/Y; Su(tor)⁴⁰⁴/+ and Draf^Su2/Y; Su (tor)⁸⁵/+ males died]. Su(tor)⁴⁰⁴ is allelic to Ras1 (DOYLE and BISHOP 1993 ). Because Su(tor)⁸⁵ has been mapped meiotically to map position 16 on 2L, which corresponds to the 24–25 cytological region, we determined if Su (tor)⁸⁵ is the mutation in the 24C2–8; 25A5 region responsible for the enhancement of Draf^Su2. We found that Su(tor)⁸⁵ is an allele of bowl on the basis of its noncomplementation with either Df(2L)sc19-8 or the bowl^L26 allele. In addition, Su(tor)⁸⁵ is an embryonic lethal recessive mutation and the embryonic phenotypes associated with loss of Su(tor)⁸⁵ are identical to those of bowl embryos (WANG and COULTER 1996 ). These embryos are missing the denticle bands two, six, and eight.

Src64B genetically interacts with Draf and overexpression of an activated form of Src64B causes ectopic expression of the Tor target gene tll:
63F4; 64C13–15: Df(3L)GN24 removes this region and behaves as a moderate enhancer of Draf^Su2. Df(3L)GN24 removes the Src64B gene, which maps to 64B12–17, that encodes the Drosophila homolog of mammalian c-src (SIMON et al. 1985 ). The existing Src64B mutation is homozygous viable, suggesting that Src64B may be functionally redundant, or that it may not be essential for viability. Src64B¹⁷ is associated with a deletion that removes the first two exons of the Src64B transcript, and Src64B¹⁷ homozygotes produce no Src64B protein. Interestingly, Src64B¹⁷ is classified as a weak allele on the basis of the observation that animals hemizygous for Src64B¹⁷ (Src64B¹⁷ in trans with a deficiency of the region) exhibit a more severe oogenesis phenotype than Src64B¹⁷ homozygotes (DODSON et al. 1998 ). We found no enhancement of Draf^Su2 by Src64B¹⁷ in the heterozygous situation. However, we found that Draf^Su2/Y; Src64B¹⁷/Src64B¹⁷ males are semilethal, 51% (38/74) as viable as FM7/Y; Src64B¹⁷/Src64B¹⁷ siblings. Further, the Draf^Su2/Y; Src64B¹⁷/Src64B¹⁷ males that survived had eyes that were slightly small and rough (data not shown), similar to the eye phenotypes due to certain mutations in genes of the Ras/Raf pathway. Finally, when Draf^Su2/FM7; Src64B¹⁷ females were crossed to Df(3L)GN24/TM8 Sb males, no Draf^Su2/Y; Df(3L)GN24/Src64B¹⁷ progeny were recovered, while FM7/Y; Df(3L)GN24/Src64B¹⁷ males survived (N = 31), as well as females with or without the Df(3L)GN24 chromosome (N = 112). These results suggest that Src64B is a candidate gene responsible for the enhancement of Draf^Su2 located within Df(3L)GN24. There may be another gene deleted by Df(3L)GN24, which, in conjuction with Src64B¹⁷, is also required for Draf function (see DISCUSSION).

To investigate the involvement of Src64B in Draf signaling, we overexpressed an activated form of Src64B, Src64B⁵⁴⁰ (KUSSICK et al. 1993 ), in early embryos under the control of the heat-shock promoter and examined its effects on the Tor pathway. Src64B⁵⁴⁰ lacks the C-terminal negative regulatory domain and acts as a constitutively activated kinase (KUSSICK et al. 1993 ). Heat-shock induction of Src64B⁵⁴⁰ in early embryos caused dramatic defects in the cuticular structures, most notably deletions of the ventral denticle bands (Fig 2E). Such defects are reminiscent of those associated with tor gain-of-function mutations. The embryos derived from females carrying a tor gain-of-function mutation show expansion of the tll expression domains (Fig 2D) and disruption of the cuticular structures in the central region of the embryo (Fig 2C). tll is activated at the embryonic termini (PIGNONI et al. 1990 , PIGNONI et al. 1992 ) and the posterior tll expression domain has been used as a molecular "readout" for the strength of Tor as well as Draf activities (HOU et al. 1995 ; LI et al. 1997 , LI et al. 1998 ). To investigate whether the cuticular defects caused by overexpression of Src64B⁵⁴⁰ are due to expansion of tll expression domains, we examined tll mRNA levels in embryos following heat-shock induction of Src64B⁵⁴⁰. Indeed, 16% (N = 36) of appropriately aged embryos showed marked expansion of tll expression domains (Fig 2F). No control embryos showed expansion of tll expression under the same heat-shock treatment. Together, these results suggest that Src64B activation can positively influence Tor signaling, raising the possibility that it activates Draf in vivo.

View larger version (136K): In this window In a new window Download PPT slide   Figure 2. Cuticular phenotypes and tll mRNA expression patterns. (A) A wild-type larva, showing eight abdominal denticle bands, the Filzkörper materials (arrow) within the posterior spiracles, and head skeletons. (B) tll expression pattern in a wild-type embryo at the precellularization stage. The posterior domain is 15% of egg length. (C) An embryo laid by a female heterozygous for the gain-of-function mutant torY9. Note that all the ventral denticle bands are missing and the Filzkörper is slightly enlarged. (D) Expansion of tll in an embryo derived from a torY9/+ mother. (E) Embryos carrying one copy of the hsp70-Src64B540 transgene were heat shocked during early embryogenesis. Note the severely disrupted denticle bands and slightly enlarged Filzkörper materials (compare with C). (F) tll expression in embryos carrying one copy of the hsp70-Src64B540 transgene following heat-shock treatment. Note the expansion of tll expression domains similar to the one shown in D. (G) A larva homozygous for Tec29A206. Note the defective mouth parts (see also ROULIER et al. 1998 ) and shortened Filzkörper (arrow). (H) An unhatched embryo from a torY9/Tec29A206 mother crossed to Tec29A206/+ males presumably homozygous for Tec29A206. These embryos were recognized as Tec29A206 homozygotes because they had defective mouth parts and shortened Filzkörper that are identical to Tec29A206 homozygous embryos. They exhibited significantly more ventral denticle bands than those laid by torY9/+ females.

Mutation in Tec29A partially suppressed a gain-of-function mutation of tor:
28E4–7; 29B2–C1: Df(2L) TE29Aa-11 removes the cytogenetic region 28E4–7; 29B2–C1 and behaves as a weak enhancer of Draf^Su2. Tec29A, which is located in this interval, encodes the Drosophila homolog of a cytoplasmic tyrosine kinase of the Tec family. Mammalian Raf-1 can be phosphorylated and activated by the Src tyrosine kinase in vitro, but there has been no confirmation of this in vivo (reviewed by BROWN and COOPER 1996 ; MORRISON and CUTLER 1997 ). Tec proteins share homology with Src in the kinase domain and are regulated by Src (GUARNIERI et al. 1998 ; ROULIER et al. 1998 ). Although a mutant allele of Tec29A did not dominantly enhance the lethality of Draf^Su2 (Table 2), by examining the genetic interactions between Tec29A and tor, we found that Tec29A mutations partially suppress a gain-of-function allele of tor (see below).

A mutation in Tec29A, Tec29A²⁰⁶ is caused by a P-element insertion at the 5' untranslated region of Tec29A that greatly reduces Tec29A transcripts such that no Tec29A mRNA is detectable in Tec29A²⁰⁶ homozygous embryos (ROULIER et al. 1998 ). However, in contrast to Df(2L)TE29Aa-11, we found that Tec29A²⁰⁶ did not cause lethality to Draf^Su2 hemizygous males in the heterozygous situation. Embryos homozygous for Tec29A²⁰⁶ do not hatch, and their cuticles showed defects in the mouth parts and the posterior spiracles. These terminal structures are either defective or missing in embryos mutant for the Tor pathway genes. As has been previously demonstrated (ROULIER et al. 1998 ), the mouth parts of Tec29A²⁰⁶ homozygous embryos are missing several components (Fig 2G). In addition, the posterior spiracles of Tec29A²⁰⁶ homozygous embryos are shorter than in wild type (arrow in Fig 2G). These observations suggest that Tec29A is required for the differentiation of terminal structures and that it may function in the same genetic pathway as Tor and Draf. Since mutations in Tec29A disrupt the terminal structures in the embryos, we examined the relationship between Tec29A and tor. A single copy of the gain-of-function allele tor^Y9 causes expansion of the expression domains of tll and consequently disruptions of the larval cuticular structures. Embryos laid by tor^Y9 heterozygous females lack all denticle bands and have a severely disrupted head skeleton. Most of them retain intact and often enlarged posterior spiracles (Fig 2C). We found that zygotic homozygosity for Tec29A²⁰⁶ partially suppressed the tor^Y9 maternal effect phenotype. When tor^Y9/Tec29A²⁰⁶ females were crossed to Tec29A²⁰⁶/+ males, all embryos showed a significant increase in the number of ventral denticle bands, suggesting that reducing the maternal amount of Tec29A by half suppressed tor^Y9 (N > 500). About one-quarter of the embryos, presumably zygotically null for Tec29A, exhibited mostly the Tec29A²⁰⁶ mutant phenotype (Fig 2H). However, they did not completely restore all the ventral denticle bands (Fig 2H), suggesting that homozygosity of Tec29A does not completely suppress tor^Y9. We observed similar partial suppressions of the tor^Y9 phenotype when we used tor^Y9/Df(2L)TE29Aa-11 females in the above cross. These results indicate that Tec29A²⁰⁶ is epistatic to tor^Y9, consistent with a model in which Tec29A functions downstream from, or in parallel, to Tor and is minimally required for a gain-of-function Tor protein to transduce signal.

Since Tec29A is regulated by Src64B during oogenesis (GUARNIERI et al. 1998 ; ROULIER et al. 1998 ) and overexpression of Src64B⁵⁴⁰ is associated with an expansion of tll, we examined whether overexpression of Tec29A is able to increase Tor signaling. In contrast to Src64B⁵⁴⁰, overexpression of Tec29A from a heat-shock inducible transgene did not cause a significant change in tll expression in the embryo (data not shown). If the activity of Tec29A is regulated by Src64B, overexpression of a wild-type version of the gene may not have significant effects on the animal.

Draf function is required for limb development:
68D6; 69A1: This cytological region is defined by the overlapping deficiencies Df(3L)vin5, which removes 68A2; 69A1, Df(3L)vin7, which deletes 68C8–11; 69B4–5, and Df(3L)vin2, removing 67F2; 68D6. Df(3L)vin5 and Df(3L)vin7 weakly enhanced Draf^Su2 male lethality, whereas Df(3L)vin2 showed no interaction with Draf^Su2. Interestingly, about half of the Draf^Su2/Y; Df(3L)vin5/+ escaper males (10/25) exhibited wing defects in the posterior part of the wing, showing either notches along the margin (Fig 3B) or pattern duplications in 30% of animals (Fig 3C). These defects were also found in 10% of Draf^Su2/Y; Df(3L)vin7/+ males. One gene deleted by Df(3L)vin5 and Df(3L)vin7, but not by Df(3L)vin2, is brachyenteron (byn), located at 68E1–4. byn encodes a homolog of the mouse Brachyury gene, a T-related homeobox gene regulated by Tor and required for specification of the hindgut and anal pads during embryogenesis (SINGER et al. 1996 ). byn is activated by tll and repressed by hkb. Therefore it is possible that byn is responsible for the enhancement of these two overlapping deficiencies. However, a strong byn allele that we tested did not enhance Draf^Su2 in heterozygotes (Table 2), nor did the Draf^Su2/Y; byn/+ males exhibit any wing defects.

View larger version (113K): In this window In a new window Download PPT slide   Figure 3. Draf is essential for limb patterning. Adult wing defects exhibited by DrafSu2/Y; Df(3L)vin5/+ males. (A) Wild-type adult wing. (B) A wing from a DrafSu2/Y; Df(3L)vin5/+ adult male showing notching in the posterior compartment of the wing. (C) A wing from a DrafSu2/Y; Df(3L)vin5/+ adult male showing partial duplication of anterior structures in the posterior compartment. (D) A wild-type wing disc showing en-lacZ expression pattern. (E) en-lacZ expression in a wing disc with insufficient levels of Draf during the second and early third larval instars. Note expansion of en-lacZ staining beyond the normal posterior compartment and the mirror-image duplication of the anterior compartment in the posterior. (F) Mirror image duplication of anterior structures in the posterior compartment due to an insufficient amount of Draf. (G) Leg duplication in a similarly treated partially rescued Draf male.

Draf^Su2/Y; Df(3L)vin5/+ male escapers exhibit wing notches that are restricted in the posterior compartment and anterior pattern duplications at the expense of posterior pattern elements (see above and Fig 3B and Fig C). This suggests that Draf, together with a gene that is removed by Df(3L)vin5, has a function in patterning the imaginal discs.

The wing disc is divided into anterior and posterior compartments, and the cell identity in the posterior is maintained by continued expression of en (reviewed by LAWRENCE and STRUHL 1996 ), but elevated levels of En in the posterior compartment result in partial inactivation of both en and invected (inv), indicating that En has a negative autoregulatory mechanism (GUILLEN et al. 1995 ). hedgehog (hh) is expressed by cells in the posterior compartment, which by virtue of their En expression are not responsive to Hh. The Hh protein diffuses into the anterior region and, along the anteroposterior (A/P) boundary, activates decapentaplegic (dpp), which encodes a morphogen that organizes the global patterning of the wing (TABATA et al. 1992 ; BASLER and STRUHL 1994 ; ZECCA et al. 1995 ).

To determine if the cause of the posterior wing notching and anterior pattern duplications observed in the Draf^Su2/Y; Df(3L)vin5/+ male survivors could be explained by an additional reduction in Draf activity, we examined the role of Draf in patterning of the wing and leg imaginal discs using partially rescued Draf null males (see MATERIALS AND METHODS; clones of Draf null alleles cannot be recovered). When Draf is not provided during the second and third larval instars, the domain of en-lacZ expression and the overall levels of en-lacZ expression are greatly increased [ Fig 3E; this was also confirmed using antibodies directed against En (data not shown)], and an ectopic anterior compartment is induced. When allowed to reach adulthood these animals exhibited the pattern duplications predicted on the basis of en expression observed in the discs, that is, duplication of anterior pattern elements in the posterior compartment (Fig 3F). Comparable pattern duplications originating from the posterior compartment were also observed in the legs by withholding Draf during the second and third larval instars (Fig 3G).

Therefore, Draf appears to have a role in negatively regulating en expression in imaginal discs. The wing notching observed exclusively in the posterior compartment is also consistent with Draf negatively regulating en, since increased expression of En in the posterior compartment serves to partially inactivate en and inv, which is thought to be required for the determination of posterior cell fates (GUILLEN et al. 1995 ). These results, together with the observation that ectopic activation of Raf or Draf in the discs results in a marked reduction in En and Hh expression in the posterior compartment (E. NOLL, unpublished observations), suggest that Draf is essential for limb patterning. On the basis of these observations, the simplest interpretation of the posterior wing notching and duplication in the Draf^Su2/Y; Df(3L)vin5/+ males is that the gene removed by Df(3L)vin5 further reduces Draf signaling.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Previously, genetic screens using EMS-induced lesions have been conducted for modifiers of activated forms of Ras1 or Draf (DICKSON et al. 1996 ; KARIM et al. 1996 ) or for suppressors of a Draf hypomorphic allele (TSUDA et al. 1993 ; LU et al. 1994 ). These screens have proven to be very fruitful in isolating new components in the Ras/Raf pathway. However, the mechanism of Raf activation remains unclear. To isolate additional components of the Ras1/Draf pathway, possibly including a Draf activator, we conducted a screen for enhancers of a viable allele of Draf. We tested 148 autosomal deficiencies in an F₁ screen and found that 23 behaved as dominant enhancers of Draf^Su2. Some of these deficiencies remove genes or loci known to be involved in the Ras1/Draf pathway. Among the deficiencies that remove genomic regions not previously known to contain genes involved in the Ras1/Draf pathway, two deficiencies remove genes encoding Tec29A and Src64B, respectively. In addition, through this screen, we identified a novel role for Draf in limb development.

Verification of candidate genes for the identified genomic regions that enhanced Draf^Su2:
For each genomic region that enhanced Draf^Su2, we searched FlyBase and tried to identify a candidate gene that is most likely responsible for the enhancement. For most of these genes, we obtained mutant alleles and determined if they were able to enhance Draf^Su2 as heterozygotes. As shown in Table 2, in four of the candidate genes, bowl, rl, Ras1, and 14-3-3, the mutant allele behaved as the respective deficiency and dominantly enhanced Draf^Su2. The rest of the candidate mutations did not dominantly enhance Draf^Su2 lethality. There are several possible explanations for this. First, the candidate gene selected may not be responsible for the enhancement of Draf^Su2, but rather the true enhancer gene was not identified because the responsible gene has not yet been identified, or is not an obvious candidate for an enhancer. Second, it is possible that some of the alleles tested are not as strong as a deficiency and therefore are unable to enhance Draf^Su2 as the deficiency does. Third, mutations in a single candidate gene alone may not be sufficient to enhance Draf^Su2. It is possible that a second gene located in the same deficiency region has to be mutated in order to observe an enhancement of the lethality. Thus, we cannot rule out a gene as a candidate simply on the basis of the inability of a particular mutant allele to enhance Draf^Su2.

Are Src64B and Tec29A activators of Draf?
Src64B and Tec29A are removed by two deficiencies that each dominantly enhanced the lethality of Draf^Su2. They were selected as candidate genes for these two deficiencies because a survey of FlyBase for genes in the regions removed by the deficiencies did not yield other genes more likely to be involved in Draf function. We showed that the Src64B¹⁷ allele in homozygotes enhanced Draf^Su2, confirming that Src64B genetically interacts with Draf^Su2. We further showed that overexpression of an activated form of Src64B in early embryos can cause activation of the Tor target gene tll and cuticular defects similar to those caused by gain-of-function mutations in tor. These results are consistent with a role of Src64B in Tor signaling and/or Draf activation (Fig 4). We were unable to demonstrate that Tec29A could enhance Draf using an available mutant allele of Tec29A. However, we obtained indirect evidence suggesting a requirement of Tec29A in Tor signaling. First, Tec29A²⁰⁶ homozygous mutant embryos exhibit defects in the terminal structures that are specified by the Tor pathway. Specifically, they showed defective mouth parts and shortened Filzkörper, phenotypes consistent with disruption of Tor signaling. Further, we found that reducing the activity of Tec29A suppresses a gain-of-function tor allele. Most strikingly, embryos zygotically homozygous for Tec29A²⁰⁶ that are derived from tor^Y9 mothers exhibited mouth parts and Filzkörper indistinguishable from those of Tec29A²⁰⁶ embryos. Mutation of Tec29A restored most of the ventral denticle bands that would have been deleted due to tor^Y9, suggesting that Tec29A is genetically epistatic to tor. However, many of the embryos still exhibited minor disruptions in the ventral denticle bands, a defect reminiscent of weak tor gain-of-function mutations. This suggests that homozygosity for Tec29A²⁰⁶ cannot completely suppress tor^Y9. Possibly, while Tec29A may be required for Tor signaling, Tec29A²⁰⁶ may not be a null allele and therefore cannot completely suppress tor^Y9. This would be consistent with the inability of this allele to enhance Draf^Su2. Alternatively, the maternally contributed Tec29A may be able to partially mediate signaling by the mutant Tor^Y9 protein. Finally, Tec29A may not be an absolute requirement for Tor signaling, but rather functions in a separate pathway that in conjunction with Tor is required for the differentiation of terminal structures.

View larger version (26K): In this window In a new window Download PPT slide   Figure 4. Components of Draf signaling in the Tor pathway. Components of the Tor signaling pathway identified in this and other studies are listed. The relationship between adjacent components is indicated by an arrow (activation); —| (inactivation), or bracket (structural). Molecules that interacted with DrafSu2 in this study are boxed.

The likelihood that Src64B and Tec29A are involved in Draf activation is based upon data from in vitro studies of mammalian c-Src function. Src kinases can phosphorylate and activate Raf-1 in vitro, and the tyrosine residues phosphorylated by Src are important for Raf-1 activation (reviewed by MORRISON and CUTLER 1997 ; THOMAS and BRUGGE 1997 ). Tec kinases are very similar to Src kinases in the kinase domain, but lack the C-terminal regulatory tyrosine and the N-terminal myristylation site that are specific for Src family members. Tec kinases interact with and are activated by Src through phosphorylation (RAWLINGS et al. 1996 ). It has been shown in Drosophila that Tec29A is regulated by Src64B and both are required for the growth of ring canals of the egg chamber (GUARNIERI et al. 1998 ; ROULIER et al. 1998 ). Although it has not been documented that Tec can phosphorylate Raf in vivo, given the similarities in the kinase domain, it is not unreasonable to propose that Tec could do so. Finally, consistent with our results, the two genomic regions containing Src64B and Tec29A were also identified as required for the function of Corkscrew (Csw) in a similar screen for modifiers of a partial loss-of-function csw allele (L. PERKINS, personal communication).

Function of Draf in limb patterning:
The proper expression of en in the posterior compartment of imaginal discs is essential for maintaining compartmental boundaries and patterning of Drosophila limbs (TABATA et al. 1995 ). Despite much insight into the events required for Hh signaling (TABATA et al. 1992 ; BASLER and STRUHL 1994 ; ZECCA et al. 1995 ; reviewed by LAWRENCE and STRUHL 1996 ), little is known about the mechanism(s) by which en expression is controlled in the posterior compartment.

We have identified two instances where a further reduction in Draf function, due to the presence of a deficiency, results in defects in posterior pattern elements in the limbs. Draf^Su2/Y; Df(3L)vin5/+ male survivors exhibit notching only in the posterior region of the wing, and partial pattern duplications in the posterior compartment. Since no specific role for Draf has been described in the limbs, we examined what the requirements for Draf were in the imaginal discs. Since clonal analysis with null alleles is uninformative, because Draf mutant clones do not develop, we conditionally provided Draf to the developing animals in a Draf null background.

By withholding Draf during the second and early third larval instars, animals with anterior pattern element duplications in the posterior compartment were frequently observed (Fig 3F). By examining the imaginal discs of these animals, we were able to determine that when there are insufficient levels of Draf, en expression is no longer restricted to the normal posterior compartment, which suggests that Draf may act to repress/restrict En expression. Along with ectopic expression of En in the anterior compartment and increased levels of En in the posterior compartment, a new mirror image anterior compartment devoid of en expression was induced (Fig 3E). This observation is consistent with the observations of GUILLEN et al. 1995 , who found that when En was ectopically expressed, ectopic anterior pattern elements were induced. They also found that ectopic expression of En in the anterior compartment induced expression of high levels of Hh and Dpp, which were responsible for overgrowth and the duplication of anterior pattern elements. Indeed, when Hh was examined in the partially rescued Draf null males, it was found to be widely ectopically expressed (E. NOLL, unpublished observations). The posteriorly restricted wing notching observed in Draf^Su2/Y; Df(3L)vin5/+ male survivors is also consistent with a requirement for Draf in negatively regulating en, since elevated levels of En expression in the posterior compartment partially inactivate both en and inv, which are necessary for the development and terminal differentiation of posterior fates (GUILLEN et al. 1995 ; TABATA et al. 1995 ). Taken together, these observations suggest that the Df(3L)vin5 deficiency contains a gene that participates with Draf in patterning of the limbs.

Specificity of genetic screens using deficiency stocks:
One concern regarding screens for enhancement of lethality using deficiency stocks is that the resulting flies are in general less healthy as heterozygotes. Therefore, this could produce enhanced lethality in a nonspecific manner due to the fact that a deficiency usually disrupts multiple genes. In our screen we identified 23 deficiencies that behaved as enhancers of Draf^Su2, and it is possible that some of these are not true enhancers of Draf^Su2. However, we believe that at least some of them are specifically required for Draf function. Comparison of our results with those of similar screens performed by others suggests that screens with deficiency stocks can reveal specific interactors. For example, in a similar screen using deficiency stocks for enhancers of a weak allele of dpp, W. Gelbart's laboratory identified three deficiencies that maternally dominantly enhanced the lethality of a weak dpp allele (NICHOLLS and GELBART 1998 ). None of these three deficiencies were found to enhance the lethality of Draf^Su2. dpp has essential functions in multiple processes during Drosophila development that are mostly distinct from those that require Draf function, except perhaps in the embryonic endoderm induction (SZUTS et al. 1998 ). Therefore, it is not surprising that no deficiencies were found to commonly enhance both Draf and dpp mutations. In contrast, L. Perkins' laboratory performed a screen for modifiers of corkscrew (csw; PERKINS et al. 1992 ), a component of Drosophila RTK signaling that shares a common pathway, Draf. They were able to identify a total of 27 autosomal regions required for csw function (L. PERKINS, personal communication), and 12 of these were also identified in our study as enhancers of Draf^Su2. This suggests that many deficiencies do not nonspecifically increase the lethality of flies heterozygous for a mutation in an essential gene, but rather can provide a source for identifying specific interacting genes.

	ACKNOWLEDGMENTS

We thank E. Hafen, M. Bishop, J. Lengyel, S. K. Beckendorf, and the Bloomington Stock Center (Bloomington, IN) for reagents or strains. We thank L. Perkins for communicating results prior to publication. This work was supported by the Howard Hughes Medical Institute from which N.P. is an Investigator. W.L. is supported by a postdoctoral fellowship from the American Cancer Society.

Manuscript received March 23, 2000; Accepted for publication June 26, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

AMBROSIO, L., A. P. MAHOWALD, and N. PERRIMON, 1989  Requirement of the Drosophila raf homologue for torso function. Nature 342:288-291[Medline].

AVRUCH, J., X. ZHANG, and J. M. KYRIAKIS, 1994  Raf meets Ras: completing the framework of a signal transduction pathway. Trends Biochem. Sci. 19:279-283[Medline].

BASLER, K. and G. STRUHL, 1994  Compartment boundaries and the control of Drosophila limb pattern by hedgehog protein. Nature 368:208-214[Medline].

BEGEMANN, G., A. M. MICHON, L. VAN DER VOORN, R. WEPF, and M. MLODZIK, 1995  The Drosophila orphan nuclear receptor Seven-up requires the Ras pathway for its function in photoreceptor determination. Development 121:225-235[Abstract].

BEIMAN, M., B. Z. SHILO, and T. VOLK, 1996  Heartless, a Drosophila FGF receptor homolog, is essential for cell migration and establishment of several mesodermal lineages. Genes Dev. 10:2993-3002[Abstract/Free Full Text].

BOUR, B. A., M. A. O'BRIEN, W. L. LOCKWOOD, E. S. GOLDSTEIN, and R. BODMER et al., 1995  Drosophila MEF2, a transcription factor that is essential for myogenesis. Genes Dev. 9:730-741[Abstract/Free Full Text].

BRAND, A. H. and N. PERRIMON, 1994  Raf acts downstream of the EGF receptor to determine dorsoventral polarity during Drosophila oogenesis. Genes Dev. 8:629-639[Abstract/Free Full Text].

BRONNER, G. and H. JAECKLE, 1991  Control and function of terminal gap gene activity in the posterior pole region of the Drosophila embryo. Mech. Dev. 35:205-211[Medline].

BROWN, M. T. and J. A. COOPER, 1996  Regulation, substrates and functions of src. Biochim. Biophys. Acta 1287:121-149[Medline].

BRUNNER, D., N. OELLERS, J. SZABAD, W. H. BIGGS, III, and S. L. ZIPURSKY et al., 1994  A gain-of-function mutation in Drosophila MAP kinase activates multiple receptor tyrosine kinase. Cell 76:875-888[Medline].

CARTHEW, R. W., T. P. NEUFELD, and G. M. RUBIN, 1994  Identification of genes that interact with the sina gene in Drosophila eye development. Proc. Natl. Acad. Sci. USA 91:11689-11693[Abstract/Free Full Text].

CASANOVA, J., M. LLIMARGAS, S. GREENWOOD, and G. STRUHL, 1994  An oncogenic form of human raf can specificy terminal body pattern in Drosophila. Mech. Dev. 48:59-64[Medline].

CHANG, H. C. and G. M. RUBIN, 1997  14-3-3 epsilon positively regulates Ras-mediated signaling in Drosophila. Genes Dev. 11:1132-1139[Abstract/Free Full Text].

DIAZ-BENJUMEA, F. J. and E. HAFEN, 1994  The sevenless signalling cassette mediates Drosophila EGF receptor function during epidermal development. Development 120:569-578[Abstract].

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].

DODSON, G. S., D. J. GUARNIERI, and M. A. SIMON, 1998  Src64 is required for ovarian ring canal morphogenesis during Drosophila oogenesis. Development 125:2883-2892[Abstract].

DOYLE, H. J. and J. M. BISHOP, 1993  Torso, a receptor tyrosine kinase required for embryonic pattern formation, shares substrates with the sevenless and EGF-R pathways in Drosophila. Genes Dev. 7:633-646[Abstract/Free Full Text].

DUFFY, J. B. and N. PERRIMON, 1994  The Torso pathway in Drosophila: lessons on receptor protein tyrosine kinase signaling and pattern formation. Dev. Biol. 166:380-395[Medline].

EBERL, D. F., B. J. DUYF, and A. J. HILLIKER, 1993  The role of heterochromatin in the expression of a heterochromatic gene, the rolled locus of Drosophila melanogaster. Genetics 134:277-292[Abstract].

FABIAN, J. R., A. B. VOJTEK, J. A. COOPER, and D. K. MORRISON, 1994  A single amino acid change in Raf-1 inhibits Ras binding and alters Raf-1 function. Proc. Natl. Acad. Sci. USA 91:5982-5986[Abstract/Free Full Text].

FLYBASE,, 1999  The FlyBase: database of the Drosophila Genome Projects and community literature. The FlyBase Consortium. Nucleic Acids Res. 27:85-88[Abstract/Free Full Text].

GISSELBRECHT, S., J. B. SKEATH, C. Q. DOE, and A. M. MICHELSON, 1996  heartless encodes a fibroblast growth factor receptor (DFR1/DFGF-R2) involved in the directional migration of early mesodermal cells in the Drosophila embryo. Genes Dev. 10:3003-3017[Abstract/Free Full Text].

GUARNIERI, D. J., G. S. DODSON, and M. A. SIMON, 1998  SRC64 regulates the localization of a Tec-family kinase required for Drosophila ring canal growth. Mol. Cell 1:831-840[Medline].

GUILLEN, I., J. L. MULLOR, J. CAPDEVILA, E. SANCHEZ-HERRERO, and G. MORATA et al., 1995  The function of engrailed and the specification of Drosophila wing pattern. Development 121:3447-3456[Abstract].

HOU, X. S., T.-B. CHOU, M. B. MELNICK, and N. PERRIMON, 1995  The Torso receptor tyrosine kinase can activate Raf in a Ras-independent pathway. Cell 81:63-71[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].

KLAMBT, C., L. GLAZER, and B. Z. SHILO, 1992  breathless, a Drosophila FGF receptor homolog, is essential for migration of tracheal and specific midline glial cells. Genes Dev. 6:1668-1678[Abstract/Free Full Text].

KUSSICK, S. J., K. BASLER, and J. A. COOPER, 1993  Ras1-dependent signaling by ectopically-expressed Drosophila src gene product in the embryo and developing eye. Oncogene 8:2791-2803[Medline].

LAWRENCE, P. A. and G. STRUHL, 1996  Morphogens, compartments, and pattern: lessons from Drosophila? Cell 85:951-961[Medline].

LEEVERS, S. J., H. F. PATERSON, and C. J. MARSHALL, 1994  Requirement for Ras in Raf activation is overcome by targeting Raf to the plasma membrane. Nature 369:411-414[Medline].

LI, W., E. M. C. SKOULAKIS, R. L. DAVIS, and N. PERRIMON, 1997  The Drosophila 14-3-3 protein Leonardo enhances Torso signaling through D-Raf in a Ras1-dependent manner. Development 124:4163-4171[Abstract].

LI, W., M. MELNICK, and N. PERRIMON, 1998  Dual function of Ras in Raf activation. Development 125:4999-5008[Abstract].

LU, X., M. B. MELNICK, J.-C. HSU, and N. PERRIMON, 1994  Genetic and molecular analyses of mutations involved in Drosophila raf signal transduction. EMBO J. 13:2592-2599[Medline].

MARDON, G., N. M. SOLOMON, and G. M. RUBIN, 1994  dachshund encodes a nuclear protein required for normal eye and leg development in Drosophila. Development 120:3473-3486[Abstract].

MARSHALL, C. J., 1994  Map kinase kinase kinase, Map kinase kinase and Map kinase. Curr. Opin. Genet. Dev. 4:82-89[Medline].

MELNICK, M. B., L. A. PERKINS, M. LEE, L. AMBROSIO, and N. PERRIMON, 1993  Developmental and molecular characterization of mutations in the Drosophila raf serine-threonine kinase. Development 118:127-138[Abstract].

MLODZIK, M., Y. HIROMI, U. WEBER, C. S. GOODMAN, and G. M. RUBIN, 1990  The Drosophila seven-up gene, a member of the steroid receptor gene superfamily, controls photoreceptor cell fates. Cell 60:211-224[Medline].

MOODIE, S. A. and A. WOLFMAN, 1994  The 3Rs of life: Ras, Raf and growth regulation. Trends Genet. 10:44-48[Medline].

MORRISON, D. K. and R. E. J. CUTLER, 1997  The complexity of Raf-1 regulation. Curr. Opin. Cell Biol. 9:174-179[Medline].

<