Simon_new

8/12/2019 Simon_new

1/16

Cel l , Vol . 67, 701-716, November 15, 1991, Copyright 0 1991 by Cel l Press

Rasl and a Putative Guanine Nucleotide ExchangeFactor Perform Crucial Steps in Signalingby the Sevenless Protein Tyrosine KinaseMichael A. Simon, David D. L. Bowtel l , G. Steven Dodson, Todd R. Laverty,and Gerald M. RubinHoward Hughes Medical Inst i tuteand Depa rtment of Molecular and Cell BiologyUnive rsity of CaliforniaBerkeley, California 94720

SummaryWe have conducted a genet ic screen for mutat ions thatdecrease the effective ness of signaling by a proteintyrosine kinase, the product of the Drosophila melano-gaster sevenless gene. These mutat ions def ine sevengenes whose wild-type products may be required forsignaling by sevenless. Four of the seven genes alsoappear to be essential for signaling by a second proteintyrosine kinase , the product of the Ellipse gene. Theputat ive products of two of these seven genes havebeen identified. One encode s a ras protein. The otherlocus encodes a protein that is homologous to the S.cerevisiae CDC 25 protein, an activator of guanine nu-cleot ide exchange by ras proteins. These results sug-gest that the st imulation of ras protein act iv ity is a keyelemen t in the signaling by seven less and Elripse andthat this st imulat ion may be achieved by act ivat ing theexchange of GTP for bound GDP by the ras protein.IntroductionProtein tyrosine kin ases (PTK s) are importan t regulatoryproteins that control man y aspects of cellular growth , d if-ferent iat ion, and metabolism (Cant ley e t al. , 1991). Manypolypeptide hormon es are know n to regulate the metabo -lism, differentiation, and growth of target cells by bindingtransmembrane receptors that possess intracellular PTKdomains (Cross and Dexter, 1991). In addit ion, mutat ionsthat alter the normal regulation and activity of either recep -tor PTKs or the closely related cytoplasmic PTKs can gen-erate oncogen es that are capable of altering cellulargrowth control (Bishop, 1991; Hunter, 1991). Understand-ing how the PTK s influence cellular activities could provideimportan t insights into how these cellular proce sses areregulated and coordinated.

Both the ligand-bound receptor PTK s and the oncogen icPTKs show elevated PTK act iv ity relat ive to their unboundor unmutated counterparts. This increase in PTK act iv ityis presume d to be a key element in initiating the signaltransduction pathway, because m utat ions that destroyPTK activity almost invariably abolish function (for a partialexcept ion, see Henkemeyer et al. , 1990). The primary bio-chemical consequence of the increased PTK act iv ity is

* Present address: Howard Florey Insti tute of Experimental Physiologyand Medicine, Universi ty of Melbourne, Victoria, Austral ia,

generally phosphorylation of the PTK itself, but a numberof other cellular proteins are also phosphorylated on tyro-sine residues to varying extents. A number of proteinswhose content of phosphotyrosine increases after a PTKis act ivated have been characterized. Among these mole-cules are several proteins that are known or thought tobe eleme nts of signaling pathw ays and are thus potentialmediators of PTK signaling (reviewed by Cant ley et al. ,1991). These include phospholipase C-r, the ras GTPaseactivating protein (rasGAP ), a subunit of a Ptdlns 3-kinase,and the c-rafl serinelthreonine kinase. A second cons e-quence of PTK act ivat ion is the recruitment of several ofthese same cellular proteins into comp lexes with the acti-vated PTK . The precise com posit ion of these complexesis not clear, but in several case s the binding appears to bemediated by SH2 domains of the target proteins bindingto the phosphotyrosine moiet ies o f the act ivated PTK.Complex formation may act as a mechanism for local iz ingsignal transducing molecu les to their site of action.

These studies have ident ified a number of proteins thatmay play important roles in signal t ransduction by PTKs.However, the funct ional s ignif icance of these potent ialcomponents of PTK signaling pathways has been dif f icultto determine because there has been no easy way to askwhether these proteins are an essential part of the sig-naling pathway. Studies of mutat ions that af fect develop-men t in flies and nematod es have provided in vivo evi-dence for the participation of two proteins in PTK signaling.The Drosophila I(l)polehole locus, which encodes a c-raflhomo log, is required for the action o f the PTK encoded bythe torso gene (Ambro sio et al., 1989; Sprenger et al.,1989; Casanov a and Struhl, 1989). Studies of a Caenor-habditis elegans vulva1 developm ent likewise indicate thata ras protein, the product of the let-60 gene, is essentialfor the act ion of the PTK encoded by the let-23 gene (Hanand Sternberg, 1990; Beitel et al., 1990; Aroian et al.,1990).

We have attempted to study the mechanism of PTK sig-naling by f irst systematical ly ident ify ing genes whoseproduc ts ma y be essential for normal strength signalingby a particular PTK , the product of the Drosophila melano-gaster sevenles s (sev) gene, and then characterizing theproduc ts of these ca ndidate loci. Our hope is that theseexperim ents will both provide in vivo evidence for theinvolvement of suspected elements of PTK pathways andperhaps identify new signaling proteins.

The sev gene encodes a transmembrane PTK (Sev-enless) th at is mo st related to the vertebrate woos protein(Hafen et al., 1987; Basler and Hafen , 1988; Bowtell et al.,1988). The sole known funct ion of Sevenless is for thedevelopment of the Drosophila compound eye. The eyeis comp osed of approxim ately 800 repeating units calledommatidia. Each ommatidium consists of a precise arrayof eight photoreceptors and 12 accessory cel ls. In the ab-sence of funct ional Sevenless, each ommatidium lacks aparticular photorece ptor, the R7 cell. Studies of the devel-opment of f l ies mutant for sev show that the cell that would

8/12/2019 Simon_new

2/16

Cel l702

LA31 tsLA31 RI tiLA24 tsLA24 RI wt

Figure 1. The Construction of Temperature-Sensi tive sev Al lelesSi te-di rected mutagenesis was used to generate five new sev al leles,which were then placed into seti2 flies by P element-m ediated germ-l ine transformation. These mutations were based on the sequences oftwo temperature-sensitive al leles of v-src, LA 24 and LA31, and tworevertants, IA24 Rl and LA31 Rl (Fincham and Wyke, 1966). Thecentral portion of the figure shows a comparison of the amino acidsequences of portions of the Sevenless protein and pp60*m. The phe-notype of the V-SIC al leles and the changes in the ~~60~ amino acidsequence are shown below. The mutations that were induced in sevand their phenotypes are indicated above the comparison. The pheno-types of fl ies carrying the mutant sev al leles were examined by thereduced cornea1 pseudopupi l method at 16OC and 29%.

normally becom e the R7 photoreceptor cell instead devel-ops as a nonneuronal lens-secreting cone cell (Tomlinsonand Ready, 1966, 1967). Analysis of ommatidia that aremosaic for the sev mutat ion has demonstrated that therequirement for Seven less function is limited to the R7 celli tself . Thus, Sevenless is proposed to act as a receptor fora signal that determines whether the presumptive R7 cel lbecom es a photoreceptor or a nonneuronal cell.In this report w e describe a genet ic screen for mutat ionsthat decrease the effect iveness of s ignaling by Sevenless.This screen has al lowed us to ident ify seven genes, En-hancers of sevenless, that when m utated can make thephenotype of a tempera ture-sensitive sev allele more se-vere. We have molecularly characterized two of the loci.One locus, Rasl (formerly Drasl; Neuman-Silberberg etal. , 1964; B ock, 1967), encodes a ras protein. The otherlocus, Son of seven less (SOS), has been previous ly identi-fied on the basis of a dominan t allele tha t enhance s signal-ing by a mutant sev protein (Rogge et al. , 1991). We showthat SOS encodes a protein that is homologous to theCDC 25 protein, an activator of guanine nucleotide ex-change by RAS proteins in budding yeas t (Broek et al.,1967; Robinson et al. , 1967; Jones et al., 1991). Theseresults imply that an increase of ras protein act iv ity is a keyelemen t in signaling by the sev PTK and further sugge stthat this act ivat ion may be achieved by st imulating GDP-GTP exchange of the Rasl protein.ResultsEnhancers of sevenless Def ine Seven LociThe simplest way to isolate mutat ions that might af fectsignaling by the sev PTK would be to isolate new recessivemutat ions that result in the absence of the R7 cel l . This

ASevd2. + SeVB4

f + @ 22.7C : e R7 present0.0sevd2. +. sevB4 @ 24.3C 00+ + : : R7 absent

Look for mutants such that:SeVd2; * SevB4 @ 22.7C .e+ * l .l e R7 absent

0 p sevd2;; * screen for absence+ TM3 ry P[v, sev s4] of the R7 cel lFigure 2 . The Scheme for Isolating E(sev) Mutations(A) is a summary of the phenotypes of the sevB al lele at di fferenttempera tures. The black circles are a depict ion of the observed re-duced cornea1 pse udopupi l when the fl ies are reared at the indicatedtemperature. The reduced cornea1 pseudopupi l represents an opticalprojection of the posi tions of the rhabdomeres, the l ight-sensing organ-el les of the photoreceptors, from a group of neighboring omm atidia.The R7 rhabdomere is represented by the central ci rcle of the pheno-typically wi ld-type image observed when seyB4 fl ies are reared at22.7%. Each asterisk represents a mutagen-treated chromosome.(B) is a more d etai led description of the screen for E(sev) mutations.Mutagenized seyd male fl ies were m ated wi th se@*; CxD, DITMS, Sb,ry, P[ry, sep] females at 22.7%. The male and female Sb progeny(genotypes setiVY: I+; l ITM3, Sb, fy, P[ry, seva4] or sevd2/sevd2; I+;ITM3 , Sb, ry, P[ry, seve4]) were scored for the absence of R7 photore-ceptors. At this temperature, approximately 2% of the female Sb fl iesand less than 1% of the male Sb fl ies display a reduced cornea1 pseu-dopupi l lacking an R7 image in the absence of mutagenesis. Fl ies wi threduced cornea1 pseudopupi ls lacking the R7 image were backcrossedto sev? CxD, DITM3, Sb, ry, P[ry, sep] fl ies in order to determinewhich fl ies carried E(sev) m utations. The mutant progeny of this crosswere then used to balance the E(sev) m utation. The mutagen-treatedsevd flies were isogenic for both the second and thi rd chromosomes.

approach has led to the discovery of two genes: bride ofsevenless, which appears to encode the l igand for Sev-enless (Reinke and Zipursky, 1966; Hart et al. , 1990;Krlmer et al. , 1991), and seven in absent ia, which encodesa nuclear protein and is therefore unlikely to be directlyinvolved in the early steps of Seven less-mediated signal-ing (Carthew and Rubin, 1990). We chose a dif ferent strat-egy, because we suspected that much of the pathwaydownstream of the sev PTK would be shared with otherPTK s. While Sevenle ss function is entirely dispensible forthe viability of the fly, at least one other PTK is essential(Schejterand Shilo, 1 969; Price et al., 1969). Furthermore,if other PTK s have crucial roles in the developm ent of theeye, disruption of their signaling pathw ays might result insuch disorganized eyes that the presence of the R7 cel lwould be impossible to score. Mutat ions that inact ivate thecommon PTK signaling components would therefore bemissed in a genet ic screen for recessive mutat ions thatcause the absence of the R7 cel l . We therefore devised a

8/12/2019 Simon_new

3/16

Signal Transduction by the Sevenless PTK703

Figure 3. The Phenotype of E(sev) MutationsEach pan el is a photomicrograph of a sectioned eye. The fl ies were reared at 22.7%. The anterior of each eye is to the right. (A) shows the phenotypeof an eye from a female fly containing the set@ al lele in the absence of E(sev) mutations. The genotype is set@*; +/TM3, Sb, ty, P[ry, se@]. Thedark ci rcular structures in each omm atidial cluster are the rhabdomeres, the l ight-sensing organel les, of the photoreceptors. The rhabdomere ofthe R7 cel l is in the center of each cluster and is surrounded by the rhabdomeres of photoreceptors RI-R6. The rhabdomere of R6 l ies proximal lyin the retina and is not visible in these sections. The R7 cel l rhabdomere is present in al l but one ommatidial cluster. The arrow points to the exception.(6) and(C) show the phenotype of female fl ies carrying the E(~ev)3C~ or E(sev)2A m al leles, respectively. The genotypes are(B) seti? E(sev)3C9/TM3, Sb, ty, P[ry, sep] and (C) se@*; E(sev)2AWG/+; +/TM3, Sb, ry, P[ry. sep4]. In each fly, nearly al l of the omm atidial clusters are lacking an R7rhabdomere and thus an R7 photoreceptor. One exceptional omm atidium wi th an R7 cel l is seen in each panel and is indicated wi th an arrow. Ineach pan el , the most posterior row of omm atidia is approximately three rows from the posterior edge of the eye. We chose this region becausethis is the area of the eye that we assay by the reduced cornea1 pseudopupi l method. There is an apparent di fference in the quanti ty of Sevenlessactivity that is required in di fferent regions of the eye. When Sevenless activity is l imi ted, such as when fl ies carrying the se@ al lele are rearedat 22.7%, omm atidia in the anterior region of the eye general ly lack an R7 cel l , whi le more posterior clusters are sti l l phenotypical ly wi ld-type.

more sensi tive genetic screen that allows the identificationof mutat ions that reduce rather than abolish the act iv ity ofdownstream signaling proteins.Systematic studies of the effects of chromosomal dele-

tions on fly developm ent have indicated that there are veryfew loci that have observable haploid pheno types (Lind-sley et al., 1972). Other studies have sh own that, in gen-eral, the level of expression of a locus is proportional tothe gene cop y number (Muller et al., 1931). Given theseconsiderat ions, our strategy wa s to create condit ions inwhich signaling through the Sevenless pathway is sobarely adequate that even e-fold reductions in the geneact iv ity of downstream elements of the pathway might dis-rupt signaling. Under these sensitized conditions , a reces-sive loss-of-function mutation in agene encoding a compo -nent of the signaling pathw ay might yield a dominan tR7-minus phenotype even though it only inact ivates onecopy of the gene. W e refer to such mutat ions as Enhancersof sevenless.

An essential requirement for conducting such a geneticscreen for Enhancersof sevenless is the abil i ty to preciselycontrol the strength of s ignaling by Sevenless. Since wecould n ot predict the exac t level of signaling that would beideal for this genet ic screen, we constructed two tempera-ture-sens itive alleles of sev that allow fine control of signal-ing strength . The tempera ture-sensitive alleles were pro-duced by site-directed mutagenesis of the sev gene basedon the sequence of two temperature-sensit ive v-src al lelesthat change amino acid residues conserved between

v-src and sev (Fincham and Wyke, 1986). Five sev muta-tions were constructed and placed in flies using P ele-ment-me diated transforma tion (Figure 1). Two alleles,se@ and se ve4, are temperature-sensit ive.

The screen for Enhancers of sevenless was conductedas outlined in Figure 2 using the seve4 allele. The presen ceof the R7 cel l was scored by the reduced cornea1 pseu-dopupil metho d, which allow s the rapid scoring of the sevphenotype in liveflies(Francesch ini and Kirschfeld, 1971).At 22.7O C, the reduced cornea1 pseudopup ils of flies car-rying the set@ allele are wild type, while at 24.3O C, thereduced cornea1 pseudopupil is clearly lackin g the R7 cellimage. We screened for ethylmethylsulfonate-inducedmutatio ns that result in a reduced cornea1 pseud opupillacking the R7 cel l image at 22.7OC.

Approxim ately 30,000 progeny of mutagen ized flieswere screened for the absence of R7 cel ls. Twenty En-hancer o f seven less (E(sev)) mutation s were isolated andlocalized to individual chromosome s. None of these muta-t ions as heterozygotes can prevent the development of theR7 cell of a f ly that contains a sev+ gene (data not shown).Two examples of enhancement of the temperature-sensit ive sev84 phenotype by E(sev) mutat ions are shownin Figure 3. Eyes from E(sev)l+ flies generally contain lessthan 10% wild-type ommatidia, while eyes from f l ies thatdo not carry an E(sev) mutation generally contain greaterthan 90% wild-type ommatidia. However, the extent ofenhancement is variable and dependent on the exact tem-perature at which the flies are reared.

8/12/2019 Simon_new

4/16

Cel l704

E(sev)lAe3Ce0P

x 1X-l

0

Figure 4. Genetic Mapp ing of E(sev) Muta-tionsThe E(sev) mutations were mapped and testedfor possible al lel ism as described in Experi -mental Procedures. The approximate recombi-

2L

E(sev)2A nation map posi tions (*5 centimorgans) ofeach locus are shown below the arrows. TheQH map posi tions of E(sev)3A and E(sev)3B aree2Ae4G only accurate to wi thin 10 centimorgans dueeOD to their distance from the nearest marker. TheelAl ?6C E(sev)lB E(sev)3A, E(sev)3B, and E(sev)3C mutations&SF eOA define three essential genes. Animals carrying1 two mutations of the same groups rarely ora 1v 2R never survive to adul thood. TheE(sev)2A muta-2-48 Z-70 tions also define an essential gene; however,

several of the al leles retain some activity. Ani-E(sevpA

e3A E(seWBelD e6Be4A e1Ce6A e4De6D e1E3L LJ3-5

mals carrying two strong al leles of the locus,such as E(.sev)2Ati and E(sev)2AUH, die before

E(sev)SC pupation, whi le animals that carry two weakal leles, such as E(sev)2Am and E(sev)2Ae1Afre-e1B E(sev)lD quently eclose as smal l adul ts wi th extremelye2F e0Q rough and disordered eyes. The placement of

a 1 1m 3-50 3-96 3R the two X-l inked E(sev) m utations, f(sev)7Amand E(sev)lA@, into a single group is provi-sional . The f(sev)lA& mutation is viable inhemizygous males and has a weak effect on the phenotype of the se+ al lele, whi le the E(sev)7AaP mutation is tightly l inked to a recessive lethalmutation and has a strong effect on the phenotype of seve al lele. We have chosen to display these mutations as al lel ic because the two mutationsmap to the same region a nd are both capable of suppressing the phenotype of the E/p mutation. This interpretation of the data suggests that thef(sev)lAm al lele retains some residual activity. However, the possibi l ity that these two mutations affect two separate genes that each modi fysignal ing by the sep and E/p PTKs cannot be excluded.

The E(sev) mutations were then mapped by recombina-t ion to approximate chromosomal posit ions. To testwhether mutat ions that map to the same region are al lelic,we crossed f l ies carrying nearby E(sev) mutat ions to eachother. Ass uming that these genes are essential for viabil-i ty, then an animal carrying two E(sev) mutat ions that inac-tivate the sam e gene should be inviable. The results of themapping and complementat ion experiments are shown inFigure 4 (see also Table 1). The 20 E(sev) muta tions defineseven loci, which can be mutated to disrupt signaling bythe seP protein. Multiple muta nt alleles of E(sev)2A ,E(sev)3A, E(sev)3B, and E(sev)SC, have been isolated andare recessive-letha l muta tions, indicating that these lociencode essent ial funct ions. A f i f th locus, E(sev)lA, mayalso have been identified by two mutatio ns, but one allele

is viable (see Figure 4 for a discussion of this locus). Twoloci, E(sev)28 and E(.sev)3D, have each been identifiedonly once. Although the chromosomes carrying each ofthese unique m utat ions carry a recessive-lethal mutat ion,more alleles will be required to determine whether therecessive-lethal phenotypes are due to the E(sev) muta-t ions. The distribut ion of the E(sev) mutat ions among onlyseven loci suggests that we have ident if ied most of the locithat can be easily mutated to enhance the phenotype ofthe seP allele.Four of the E(sev) L oci Appear to Be Required forPhotoreceptor DevelopmentWe next wished to examine the consequences of mutat ingboth copies of the E(sev) loci for the development of the R7

Table 1. Summary of Properties of Enhancer of sevenless LociNumberof Required for Suppressor Acts PotentialMutan t Photoreceptor of in Biochemical

LOCUS Al leles Development El l ipse R7 ActivityE(sev)lA 28 Yes Yes YesE(sev)2A/Sos 7 Yes Yes Ye.9 Putative guanine nucleotide

exchange factorE(sev)2B 1 Yes YesE(sev)3A 5 ?b NoE(sev)SE 4 ?b NoE(sev)3C/Flasl 2 Yes Yes Yes GTPaseE(sev)SD 1 No NoB See Figure 4 for a discussion of this locus.b No clones were observed. These resul ts are consistent wi th a requirement for cel l growth or viabi l i ty.c Occasional omm atidia wi th extra photoreceptors are observed.d Rogge et al . (1991) have shown that a dominant al lele of this locus acts in the R7 cel l.

8/12/2019 Simon_new

5/16


cell and other photorece ptors. Since the E(Sev) mutat ionsare, with one excep tion, present on recessive-lethal chro-mosom es, we generated clones of cel ls that were homozy-gous for the E(sev) muta tions. For the loci that are repre-sented by multiple alleles, at least two alleles were tested.The analysis of these clones al lowed us to divide the E(sev)loci into three c lasse s (Figure 5 and Table 1). Mutatio ns atfour loci appear to greatly reduce the ability of a cell todevelop as either an R7 cell or as any other ph otoreceptor.This class includes E(sev)lA, E(sev)2A, E(sev)2B, and&(sev)3C. Despite the fact that these m utat ions were iden-tified owing to their ability to interfere with R7 cell develop-ment, our results suggest that they must also play a rolein some process that is common to the development of al lphotoreceptors. Homozygo us clones of two of the loci,E(sev)3A and E(sev)3B, were never or rarely seen. Thisresult is consis tent with the conclusion that these loci areessential either for cell growth or viability. One locus,E(sev)3 D, appears to be dispensable for both photorecep-tor and R7 cel l development. The homozygous E(sev)3Domm atidia are generally wild type, though o ccasiona lly anommatidium has an extra photoreceptor. However, thislocus is represented by a single m utant allele, which mayretain some act iv ity.We examined embryos homozygous for mutat ions ofeither the E(sev)2A or the E(sev)3C loci to try to determinewhich steps of development require the funct ion of theseloci. In each case the mutant embyos hatch at f requenciesequivalent to wild-type controls but die before the onset ofpupal development. This may imply that the products ofthese loci do not have imp ortant roles during embryo nicdevelopm ent. How ever, an equally likely possibility is thatthere are maternal contributions of these products that aresuff ic ient for embryonic development. Analysis of em-bryos derived from homozygous mutant germline cloneswill be required to address this issue.E(sev) Loci Function in the Developing R7 CellIf the E(sev) loci do encode proteins that participate in thetransduct ion of a signal f rom the act ivated sev PTK intothe cell, then the proteins should fu nction within the R7cell itself rather than in the cell that transm its the signal.We have tested this prediction for two of the E(sev) loci,E(sev)lA and E(sev)3C, by generating marked clones ofwild-type cells in a genetic ba ckground that contains thetempera ture-sensitive sep allele and is heterozyg ousfor the E(sev) mutat ion. The ommatidia that are outside ofthe clone lack the R7 cel l because of the enhancementof the set+ phenotype by the E(sev) mutat ion, while theomm atidia that are entirely within the clone are geneticallywild-type for the E(sev) locus and thus are pheno typicallywild-type. Since there are no absolute cell lineage relation-ships between cel ls within an ommatidium, the ommatidiathat are at the boundary of the marked clone w ill containcells that are heterozygous for the E(sev) mutation andcells that are wild-type for the E(sev) locus. If the E(sev)locus exerts i ts ef fect in the developing R7 cel l , then wewould expect R7 cel ls developing in a mosaic ommatidiumto be wild-type for the E(sev) locus. The genotype of theother photorecep tors should be irrelevant.

The results of such an analysis of mosaic ommatidia areshown in Figure 6 (see also Table 1). For the E(sev)7ABoPmutation , 40 of 41 R7 cells from phenotyp ically wild-typeomm atidia were wild-type for the E(sev)lA locus. Similarresults (34 of 35) were obtained for the E(sev)3CF muta-t ion. The two except ions are most l ikely due to the pres-ence of a small f ract ion of R7 cel ls even in the presenceof the E(sev) mutat ions (see Figure 2). These results areent irely consistent with the conclusion that the E(sev)7Aand E(sev)SC mutat ions act in the developing R7 cel l toattenuate signaling by the sev PTK. The tendency of theRl and R6 cells to be related by lineage to R7 and theincomplete penetrance of the E(sev) phenotypes makes itdifficult to absolutely eliminate the possibility that thesemutat ions also act in the Rl and R6 cel ls to block theformation of the R7 cell. How ever, the data clearly ruleout the possibi l ity that the two E(sev) mutat ions act bydecreasing the strength of signaling by the R8 cell fromwhich the bride of sevenless-m ediated signal originates(Reinke and Zipursky, 1988; Kramer et al. , 1991).Muta tions of Four f(sev) Loci Attenuate Signalingby Another PTKAn underlying assumption of our study was that dif ferentPTKs use at least some common signaling pathways. Weasked whether the E(sev) loci might encode such elementsof a common signaling pathway by test ing whether theE(sev) mutat ions could attenuate signaling by a PTK otherthan Sevenless.The P TK that we chose to test was the Ell ipse (E/p) al leleof the faint litt le ball locus (also know n as torpedo andDER ). The faint litt le ballgene encodes a PTK receptor thatis most homologous to the vertebrate EGF receptor andthe neu proteins (Livneh et al., 1985; Schejter and Shilo,1989; Price et al. , 1989). Recessive loss-of-funct ion muta-t ions of faint l i t tle bal l cause embryonic death. The muta-t ion E/p is a dominant al lele of the locus, which causesmoderate roughness of the eye when heterozygous andsevere eye roughness and reduced eye size when homo-zygous (Baker and Rubin, 1989). The basis for these phe-notypes see ms to be impaired formation of ommatidialclusters. The eyes of E/p/+ f l ies have normal ommatidia,but fewer of them tha n wild-type eyes . The E/p allele ap-pears to encode an overly active protein that is so slightlyact ivated that the act iv ity provided by the normal wild-typecopy of the gene is still a pheno typically significant fractionof the total act iv ity; while E/p/+ f l ies have rough eyes,E/p/- f l ies have wild-type eyes. The sensit iv ity of the mu-tant eye to this slight change in gene act iv ity suggestedthat i f the E(sev) mutat ions did compromise signaling bythe E/p PTK, the effect would be seen as a suppressionof the E/p phenotype.

Fl ies carrying each E(sev) mutat ion were crossed to f l iescarrying the E/p mutation in order to generate flies thatwere heterozygous for both mutat ions. These f l ies wereexamined for the roughness of their eyes. The results areillustrated in Figure 7 (see also Table 1). Mutatio ns in theE(sev)3A, E(sev)3B, and E(sev)3D loci had no effect on theE/p pheno type. Mutation s in the remaining four loci werecapable of suppressing the E/p phenotype. Three loci,

8/12/2019 Simon_new

6/16

Cel l706

Figure 5. The Requirement for the E(sev) Loci during Photoreceptor DevelopmentAnimals heterozygous for an Efsev) mutation were X-i rradiated in order to generate clones of cel ls homozygous for the E(sev) mutation. Thehomozygous Efsev) cel ls are marked by the absence of a whi te gene. The whi te gene is required for the formation of pigment granules in bothphotoreceptors and pigment cel ls. The pigment granules of the photoreceptor cel ls are apparent as dark structures near the rhabdomere of each

8/12/2019 Simon_new

7/16

;;?a1 Transduction by the Sevenless PTK

E(sev)lA, E(sevpA, and E(sev)PB, werevery effect ivesup-pressors and restored the eye to a nearly normal appear-ance. M utat ions at a fourth locus, E(sev)X, were less eff i-cient in their suppress ion, but were clearly distinguishablefrom E/p/+ flies.

These results suggest that at least four of the E(sev) lociencode proteins that some how participate in a signalingpathway that is shared by at least two PTK receptors. Thesame fou r loci are required for the formation of photorecep-tor cel ls. These results suggest that the products encodedby these four loci ma y be functionally linked, perhaps asparts of a single pathway. We therefore chose to concen-trate on the molecular cloning of two of these four loci.The E(sev)3C Locus Is the Rasl GeneThe E(sev)3C locus is represented by two mutant al leles.Recom bination mapping of the alleles placed the locus onthe right arm of the third chrom osom e near curled. Theposit ion wa s further ref ined by complementat ion mappingusing deletions for the region. Two deletions were particu-larly informative. The chromosome carrying M(3R)bylO isdeleted for the 85D8-11 to 85ElO-13 region and is unableto complement the lethal ity of either E(sev)3C mutat ion.The chromosome carrying Df(3R)by62 is deleted for theregion 85017-22 to 85Fl3-18 and rescue s the lethality ofeither allele (Kemph ues et al., 1980). These results p lacethe E(sev)JC locus in the chromosomal region 85D8-22.One g ene that is both b elieved to play a role in signaltransduct ion and that is known to map to this region isRas l, the Drosophila homolog of the c-H-rasl gene of ver-tebrates (N euman-Silberberg et al., 1984). We sough t intwo ways to test direct ly whether the E(sevpC mutat ionsmapped to Rash. Our init ial approach was to place a 12kb DNA fragment containing the Rasl gene into the germ-l ine by P element-mediated transformation and testwhether this cons truct would rescue the lethality of ani-mals carrying E(sev)SC mutat ions. The P element carryingRasl, P[w,RaslA], was capable of rescuing the lethal ityof E(sev)3CYe/E(sev)3CF f l ies. This result indicates thatthe E(sev)3C gene l ies within the 12 kb transformationfragment.

The E(sev)3C mutat ions were then localized to the Raslgene itself by sequencing the Rasl genes from the

E(sev)SC mutant chromosome s. The sequence of the en-tire coding seq uence of each allele was determined andcompared with a previously published sequence of Rasl.Both muta nt allelescontainedcha ngesthat altertheaminoacid sequenc e of the resulting Rasl protein (Figure 8). TheRasl al lele from the E(sev)3PB chromosome contains apoint mutat ion that changes glutamate-82 to a lysine. TheE(sev)3@ 2F allele contains a point mutation that change saspartate-38 to an asparagine. We conclude from theseresults that E(sev)3C is Ras7. The two mutant al leles havetherefore been renamed RaslEmK and RaslD38N .The E(sev)2n Locus Encodes a Potent ial GuanineNucleot ide Exchange FactorE(sevp4 was the second gene that we sought to ident ify.This locus is represented by seven m utant al leles. Themuta tions differ significantly in their ability to enhance thesevfu phenotype and to suppress the E/p phenotype. Whenstrong al leles, such as E(sev)2Ae2H and E(sevpA&, arecrossed to each other, v iable E(sevpA/E(sevpA progenyare not observed. However, combinat ions of weak al leles,such as E(sev)2A mo and E(sev)2Ae 1A, are generally viable,but the mutant progeny have rough eyes (data not shown).

Recom bination mapping indicated that the E(sev)u \ lo-cus is near the black gene (polytene position 34D4-8).Rogge et al. (1991) ha ve recently reported the isolation ofa dominan t allele of the gene /(2)34Ea that map s ad jacentto black (Woodruff and Ashburner, 1979; M. Ashburnerand J. Roote, personal communicat ion). This mutant al lelewas identified by its ability to suppre ss a particular muta ntal lele of sev. The locus was renamed Son of sevenless(SOS). These results suggested the possibi li ty that theE(sev)2A alleles m ight also be allelic to /(2)34Ea muta tions.Crosses between E(sev)2A al leles and two al leles of/(2)34Ea conf irmed this hypothesis (data not shown). Wewill therefore refer to the E(sev)2A locus as SOS.

The extensive genetic analysis available for the 340region sugges ted that SOS is the first essential gene distalto black. This places SOS in the vicinity of polytene band34D4. We isolated the DNA from this region by chromo-some wa lking. The start ing point for the walk was a hobotransposable elemen t inserted in polytene band 34Dl(Blackman et al. , 1989). We isolated overlapping cosmid

photoreceptor, but pigment granules are not present in every section. (A) shows a clone of whi te cel ls in the absence of E(sev) mutations. Thespacing and construction of the omm atidial cl&ters is normal. (B) shows a clone of cel ls homozygous for the E(sevJ3PW mutation. Most of theomm atidia are phenotypical ly wi ld-type, but occasionally omm atidia contain an extra photoreceptor (see arrows). The remaining panels showthe phenotypesof clonesof cel ls that are homozygousfor di fferent Efsev) mutations: (C), f(sev)lA? (D), E(sev)2A? (E), E(sev)2B? (F), E(sev)3t7e.Analysis of a number of mosaic omm atidia, as judged by the presence of at least one wi ld-type and one mutant p igment cel l , indicates that cel lshomozygous for any of these mutations are greatly underrepresented as photoreceptors. Approximately hal f of the photoreceptors from mosaicomm atidia are expected to be homozygous for whi te in the absence of E(sev) mutations. The fraction of photoreceptors from mosaic om matidiathat were homozygous for the .E(sev) mutations were as fol lows: E(sev)lA *)p, O/334; E(sevpW, 41345; E(sev)2PA, 2f411; .E(sev)3C+ls, 15i798.Simi lar resul ts were obtained for clones homozygous for the E(sev)3cZF mutation. The analysis of other al leles of the E(sev)2A locus was morecomplex. Clones of cel ls homozygous for the E(sev)2A MQ al lele were simi lar to those for the E(sev)2A an al lele. However, clones homozygous forthe f@ev)2A~D al lele were more normal. Many ommatidia lacked one or two photoreceptors, but homozygous E(sev)2Ad0 photoreceptors wereseen. Since the Efsev)2RzH and E(sev)2A MQ al leles both contain premature termination codons and produce substantial ly truncated proteins, weconclude that this is the nul l phenotype for this locus (see Figure 10). The phenotype of clones of cel ls homozygous for the E(sev)U\* muta t i onis simi lar to that observed by Rogge et al . (1991) for an al lele of this locus, cal led Son of seven/essX (see main text). When tested for lethal i ty intrans to other E(sev)2A al leles, the E(sev)2Aa0 mutation has a weaker effect than either E(sev)2AUn or E(sev)2A. Furtherm ore, the E(sev)2A-mutation is a missense mutation (see Figure 10). We conclude that E(sev)2AmD and the previously reported al lele, Son of severt/essx, retain part ialactivity.

8/12/2019 Simon_new

8/16

Cel l708

.?: o: o: e:00000080goIkkk0: 0:O:00080go 080-igure 6. The E(sev)7AP and E(sev)3PB Mutations Act in the R7 Cel l

X-i rradiation was used to generate marked clones of E(sev) cel ls in fl ies that carried the seP al lele and were heterozygous for the E(sev) mutation.The cells of the clothes were genetical ly marked by the absence of a functional whi te gene and thus lack pigment granules. If an E(sev) m utationexerts i ts effect in a particular cel l , then that cel l should tendto be whi te in mosaic omm atidia that are wi ld-type i ri construction. (A) and (6) arephotomicrographs of such clones for the (A) E(sev)lAdP and (B) E(sev)3CYB al leles. In each case, the presence of the Qsevj mutation in the cel lsoutside o f the clone blocks the abi l i ty of the seP al leleto ini tiate fly development. (C) and (D) i l lustrate the genetical ly mosaic and phenotypical lywi ld-type classes of omm atidia that were observed for E(sev)lAaP and E(sev)3PB, respectively. When a given genotypic class was observed morethan once, the number of occurrences is indicated in the upper left of the box. Open ci rcles represent whi te, E(sev) photoreceptors, whi le closedcircles represent f(sey)/+ photoreceptors. The data can be summarized as the fraction of whi te cel ls for each photoreceptor. For E(sev)7AmP, thedata are (41 total ): RI, 1U; R2, 28; R3, 27; R4, 24; R& 25 ; R6, 10; R7, 1; R8, 19. For E(.ev)3cBiB, the data are (35 total ): RI, 4; R2, 23; R3, 21;R4, 22; R5, 22; R6, 6; R7, 1; R8, 16. In,each case, the data are most consistent wi th the conclusion that the Qsev) mutation acts in the R7 cel l .Sincethe extent of enhan:em ent by the E(sev) al leles varies wi th verj l sl ight changes of the incubator temperature, we discarded clones when morethan one phenotypical ly w,l (d-type omm atidium that was not part of the clone was apparent in the sectioned region. Occasional ly, severely defectiveomm atidia were se& ci lpng the edge of the clones. These omma tidia presumably contained homozygous f(sev) cel ls that were generated by themitotic recombination.

clones representing the approximately 150 kb of DNA in This analys is indicated that at least part, if not all, of thethe 34Dl-8 region. The approxim ate position o f the SOSgene within the cloded DNA was determined by analyzing

SOS gene is contained within an 8 kb interval correspond-ing to coordinates 110 to 118 (see Figure 9). This predic-the breakpoints of several chromosomal rearrangements. tion was tested by placing a 10 kb Bglll-Xbal DNA

8/12/2019 Simon_new

9/16


Figure 7. The Effect of E(sev) Mutations on the Phenotype of the E/p MutationAl l of the eyes shown except those in (A) and (D) are from female fl ies heterozygous for the set@ mutation. (A) Scanning electron micrograph (SEM)of an eye from a female wi ld-type (Cantons) fly. (B) SEM of an eye from a fly heterozygous for the E/p mutation. The E/p mutation causes reductionin eye size and roughening of the eye surface (Baker and Rubin, 1999). (C) SEM of an eye from a fly heterozygous for both E(sev)u\tin and E/p.The presence of the E(sev) m utation strongly suppresses the phenotype of the E/p mutation. (D) to (L) show higher magni fication SEMs that i llustratethe effect on the E/p phenotype by di fferent E(sev) mutations. (D) is an SEM of an eye from a wi ld-type (Cantons) female fly. (E) is an SEM froma fly heterozygous for the E/p mutation. The fl ies in(F) to (L) were heterozygous for the E/p mutation and the indicated E(sev) m utation: (F), E(sev)2AeH;(G), f(sev)2/V; (H), E(sev)3PB ; (I), E@e v)3C? (J), E(sev)7A aP; (K), E(sev)2B? (L), E(sev)3EF. Since E/p disrupts th e norm al hexag onal arrayof the omm atidial lenses and bristles, the suppression of E/p can be scored as the return of the normal hexagonal array. E(sev)7AdP, E(sev)2Aa,&(sev)2Am, and E(sev)2ed* largely reverse the effects of the E/p mutation. The mutations f(sev)XYs (H) and E(sev)3F (I), partial ly suppressedthe E/p phenotype. Note that there are sti l l some i rregularities in the shapes of the lenses and in the placement of the bristles. Mutations at theremaining E(sev) loci had no effect (see [L] for an example).

8/12/2019 Simon_new

10/16

Cdl710

e2F: N el f?,. KtD-RhS1H-RAS i i :

D-RASI NH-RAS 6 RRGRKNYKPNRR L 189NPPbESGPGCnS s 189Figure 6. The E(sev)SC Locus Is RaslThe sequence of the Drosophi la Rasl protein (Brock, 1967) is compared wi th that of the Human p21 KU (Capon et al ., 1963). Identical residues areboxed in black. Simi lar residues are boxed in gray. The fol lowing amino acids were considered simi lar: E, D; V, L, I, M; A, G; F, Y, W; S, T; Cl , N.The mutation present in the E(sev)3pF al lele changes the GAC aspartate-36 codon to an AAC asparagine codon. The mutation present in the.E(se~)3C~ codon changes the GAG glutamate-62 codon to an AAG lysine codon. No other changes were observed.

fragme nt (coordinates 110 to 120) into the germline byP element-mediated transformation. This P element,P[w, SosA] is capable of rescuing the lethal ity of animalscarrying either strong allele, f{sev)2Ae 2H or E(sev)2A tiG,and a deletion of the S OS gene, Df(2L)bso. This indicatesthat all essential functions of SOS are contained within this10 kb region.

We isolated eye-antenna1 imaginal disc cDNA clonesderived from this region. Two classes of nonoverlappingcDNA clones were categorized. One class originates fromthe distal p ortion of this interval and from the neighboringXbal fragmen t. This class is not fully contained within the10 kb rescue fragment and was not characterized exten-sively. The second class of cDNA clones represents anmRN A transcript that spans m ost of the region implicated

by the deletion mapping. The sequences of a 5.3 kb cDNAclone and the genom ic region encoding this transcriptwere determined. The deduced structure of the transcriptis shown in Figure 9.Concep tual translation of the long open reading fra meof this t ranscript f rom the f irst in-frame methionine showsthat it is capable of encoding a protein of 1596 am ino acids(Figure 10). We determined that this open reading frameactually encode s the SOS protein by cloning this regionfrom three of the chromosomes containing SOS mutat ionsand sequencing the open reading fra mes (Figure 1 0).Each m utant al lele contains a point mutation that changesthe predicted protein seq uence. For instance , the strongalleles SOS~*~and SO&~, introdu ce termination codons ataminoacid posit ions 579 and 421, respect ively. The weak

A Df(2L) b em Df(2L) ScoRv7 (SOS - ), IOkb , C. b 85*2 (SOS +)I 1 I I I I III II I II II II I II I I 1 I I ~ II

8 8 :: sB= probe for RNA blot: -2 L lkb , 5 A h h A L A PI I I I I I =

75 11 I I.s FE F Fx $2 i3 s

Figure 9. Structure of the SOS Gene(A) A map of the region near SOS. The map of the SOS region is shown for restriction enzymes Xbal (marks below the l ine) and BamH l (marks abovethe l ine). Distal is to the left. The coordinates are given as distance from the insertion si te of the hobo[ry] element that served as the entry p ointfor the chromosomal walk. The approximate posi tions of chromosomal breakpoints were determined by genomic DNA blots and are indicated abovethe map. The sol id bar represents DNA that is deleted from the Df(2L)bwh and Df(2L)Sco kv7 chromosomes. These chromosomes are deficient forpolytene bands 34D3 to 35A4 and 34D5 to 35D5-7, respectively (M. Ashburner and J. Roote, personal communication). The broken l ine representsthe uncertainty in assigning the posi tion of the distal breakpoint. The nature of the besb2 rearrangement is not enti rely clear, but there appears tobe a breakpoint in the region indicated by the broken l ine.(B) The structure of the SOS transcript. The region (coordinates 110 to 120) indicated by the bold l ine in (A) is shown. The deduced structure of theSOS mRNA is shown above the map. The exon and intron structure was deduced from a comparison of the genomic sequence of the 6601 bp Hindl l lto Sal l region wi th the sequence of a 5.3 kb cDNA clone that was isolated from an eye imaginal disc cDNA l ibrary. The cDNA clone was incomp leteat both ends. The indicated posi tion (nucleotide 6451) of the polyadenylation si te was determined from an SOS cDNA clone derived from an embryoniccDNA l ibrary that included a poly(A) tai l . The Blend of the mRNA was determined by Sl mapping and primer extension studies and is heterogeneous.There appear to be several start si tes; the two major si tes map to the 440 to 470 region. In the diagram of the mRNA structure, protein coding regionsare indicated by the black l ine and 5 - and 3 -untranslated segments by the gray l ine. The 10 kb Bgl l l -Xba l fragment is capable of rescuing therecessive lethal i ty of SOS mutations when placed into the genome by P element-m ediated transformation.(C)Size of the SOS mRNA. Po lyadenylated RNAwas extracted from imagin al discs. Two micrograms of RNAwas fractionated, blotted, and hybridizedto =P-labeled DNA from the region indicated in (6). A single transcript of approximately 6 kb was detected. Analysis of embryonic RNA revealeda transcript of the same size (data not shown).

8/12/2019 Simon_new

11/16


SOSSOS :9SOS 137SOS 205SOS 273SOS 341

SOS 409SOS 477SOS 545SOS 613

SOS 709CDC25 1189SDC25 853BUD5 173STE6 542

SOS 776CDC25 1255SDC25 E :UD5STE6 610

MFSGPSGHAHTISYGGGIGLGTGGGGGSGGSGSGSQGGGGGIGIGGGGVAGLQDCDGYDFTKCENAARWRGLFTPSLKKVLEQVHPRVTAKEDALLYVEKLCLRLLAMLCAKPLPHSVQDVEEKVNKSFPAPIDQWALNEAKEVINSKKRKSVLPTEKVHTLLQKDVLQYKIDSSVSAFLVAVLEYISADILKMAGDYVIKIAHCEITKEDIEVVMNADRVLMDMLNQSEAHILPSPLSLPAQRASATYEETVKELIHDEKQYQRDLHMIIRVFREELVKIVSDPRELEPIFSNIMDIYEVTVTLLGSLEDVIEMSQEQSAPCVGSCFEELAEAEEFDVYKKYAYDVTSQASRDALNNLLSKPGASSLTTAGHGFRDAVKYYLPKLLLVPICHAFVYFDYIKHLKDLSSTOP (e4G)ASSQDDIESFEQVQGLLHPLHCDLEKVMASLSKERQVPVSGRVRRQLAIERTRELQMKVEHWEDKDVGQNCNEFIREDSLSKLGSGKRIWSERKVFLFDGLNVLCKANTKKQTPSAGATAYDYRLKEKYF~RRVDINDRPDSDDLKNSFELAPRMQPPIVLTAKNAQHKHDWWADLLNVITKSMLDRHLDSILQDIERKHPLRMPSPEIYKFAVPDSGDNIVLEERESAGVPM r, STOP (e2H)

SOS 641CDC25 1121f% 7:: HKWQKPSPFDSANL TDNALLQELLLSYPSTE6 493 _____---------- SSSDLFSILmHFR

SOS 844CDC25 1320SDC25 E8UD5STE6 677

SOS 910CDC25 1387SDC25 1034BUD5 377STE6 743

SOS 975CDC25 1455SDC25 22 UD5STE6 810SOS 1021 YQNQPYCLNEESTIRQFFEQLDPFNGLSDKQMSDYLYNESLRIEPRGCKTVPKFPRKWPHIPLKSPGISOS 1089 KPRRQNQTNSSSKLSNSTSSVAAAAAASSTATSIATASAPSLHASSIW DAPTAAAANAGSGTLAG EQSSOS 1157 PQHNPHAFSVFAPVIIPERNTSSWSGTPQHTRTDQNNGEVSVPAPHLPKKPGAHVWANNNSTLASASASOS 1225 RDVVFSPALPEHLPPQSLPDSNPFASDTEAPPSPLPKLVVSPRHETGNRSPFHG RMQNSPTHSTASTVSOS 1293 TLTGMSTSGGEEFCAGGFYFNSAHQGQPGAVPISPHVNVPMATNMEYRAVPPPLPPRRKERTESCADMSOS 1361 AQKRQAPDAPTLPPRDGELSPPPIPPRLNHSTGISYLRQSHGKSKEFVGNSSLLLPNTSSIMIRRNSASOS 1429 IEKRAAATSQPNQAAAG PISTTLVTVSQAVATDEPLPLPISPAASSSTTTSPLTPAM SPMSPNIPSHPSOS 1497 VESTSSSYAHQLRMRQQQQQQTHPAIYSQHHQHHATHLPHHPHQHHSNPTQSRSSPKEFFPIATSLEGSOS 1565 TPKLPPKPSLSANFYNNPDKGTMFLYPSTNEE 1596

Figure 10. The Predicted Amino Acid Sequence of the SOS ProteinA conceptual translation of the long open rea ding frame of the SOS RNA transcript is shown. The translation begins at the fi rst in-frame methion inecodon, which is approximately 370 bases from the 5 end of the transcript (posi tion 841 in the sequenced genomic DNA). The coding regions ofthree mutant al leles were sequenced. The posi tions of mutations that change the coding capaci ty of the transcript are indicated: SOS& , CAG codonto TAG; SOS&, TGG codon to TGA; SOS tiD TCC codon to TTC. The SosHo a l lele contains one addi tional change (C to T at posi tion 1473 in thegenomic sequence) that does not affect the protein sequence. The predicted SOS protein sequence is compared wi th the sequences of four relatedproteins: the S. cerevisiae CDC25 protein (Camonis et al ., 1986; Broek et al ., 1987); the S. cerevisiae SDC25 protein (Damak et al ., 1991); the S.cerevisiae BUD5 protein (Chant e t al ., 1991; Powerset al ., 1991); and the S. pombe STE Gprotein (Hugheset al ., 19gO). Residuesare boxed whereverthey are identical or simi lar to the SOS protein sequence. Identical residues are boxed in black. Simi lar residues are boxed in gray. The fol lowingamino acids were considered simi lar: E, D; V, L, I, M; A, G; F, Y, W; S, T; Q, N. The al ignments were constructed using the Genal ign program(Intelligenetics, Inc.)

allele, SOP, has a point mutation that changes serine-783 to a phenylalanine. We therefore conclude that thistranscript encode s the SOS protein.Comparison of the predicted SOS protein sequence withpreviously reported sequences sho ws considerable ho-

mology with the CDC25, SDC25, and BUD5 proteins ofSaccharomyces cerevisiae and with the STEG protein ofSchizosaccharomyces pombe (Figure 10). The SOS pro-

tein is most similar to the CDC25 and SDC25 proteins.These proteinsshare a homologous domain that is approx-imately 380 amino acids in length. For the alignmentshown, the CDC25 and SDC25 proteins are 28% and 22%identical, respe ctively, to the predicted SOS protein in thisregion.

Genet ic studies have suggested that the CDC25 proteinfunctions by activating ras protein activity (Broek et al.,

8/12/2019 Simon_new

12/16

1987; Robinson et al. , 1987). Biochemical studies havedemonstrated that the CDC2 5 and SDC25 proteins can actas guanine nucleotide exchange factors to st imulate theconversion of inact ive GDP-bound RAS2 protein to theact ive GTP-bound form (Jones et al. , 1991; Crechet et al. ,1990). These studies have further shown that a fragmentof the SDC25 protein that includes the homology to SOS issuff ic ient to catalyze nucleotide exchange by either theRASP protein or by the human c-l-bras protein. These re-sults suggest that the region of homology def ines a domainthat cataly zes guanine nucleotide exchang e by ras pro-teins and that the SOS protein may also act by catalyzingthe activation of a ras protein.DiscussionIn this study , we sought to identify genes that are requiredfor signaling by the sev PTK. Our strategy was to l imit thestrength of signaling initiated by Seven less so that thesignal was barely adequate to achieve its funct ion. Wereasoned that when the sev PTK act iv ity w as barelyabove the threshold necessary for R7 cel l formation, smallreduct ions in the abundance or act iv ity of other elementsof the pathway might lower the signal strength suff ic ient lyto cause a mutant sev phenotype. Our hope was that thissystem would be so sensit ive to the levels of componentsof the pathway that mutat ions which inact ivate only onecopy of a gene that encodes a downstream element of thepathway would result in the absence of the R7 cel l . Sincethe other copy of the gene would remain functional, theseloci, Enhance rs of seven less, could be identified even iftheir function is essential for producing a fertile fly.

Analysis of the f irst 20 E(sev) mutat ions ident if ied sevenE(sev) loci. For the two best characterized loci, SOS andRasl, the mechanism of enhancement is inact ivat ion ofone copy of an essential gene. The nature of the E(sev)mutatio ns in the other five loci has not been determined .However, for at least two loci, E(sev)3A and E(sev)36, thefrequency of E(sev) alleles is consis tent with the proposalthat these mutation s are loss-of-function alleles. Further-more, our assumption that many of the genes w hose prod-ucts are required for normal signaling would be essentialfor viability appears to be correct for at least four of theseven E(sev) loci. We do not have enough al leles of theother loci to determine whether they are essential genes .The distribution of alleles among the E@ev) loci sugg eststhat we have identified mo st, but probably not all, of theloci that can be easily m utated to enhance the se@ phe-notype.This screen was designed to ident ify genes that mightencode proteins tha t participate directly in signaling bySevenless. Mutat ions in several other classes of genesmight also be capable of producing an E(sev) phenotype.For example, mutat ions in genes whose products regulatethe expression or stability o f the tempera ture-sensitivesev protein might reduce the Sevenless-mediated signal.Another possible class of f@ev) mutat ions could affectthe abundance of the l igand that act ivates the sev protein.It is interesting to note that a mutation that inac-tivates one allele of the bride of seven less gene, the

best candidate to be that ligand (Kramer et al., 1991) doesnot show a strong enhancement of the sep phenotype inour assay (data not shown). This fai lure to enhance thesep4 phenotype might be expecte d if the ligand is pre-sented in considerably greater q uantity than is neces saryto saturate the sev protein ligand-binding sites. Some ofthe E(sev) mutations might also be expected to limit signal-ing by pathw ays that are parallel to the Seven less-mediated pathw ay and are also essential for the formationof the R7 cel l . We do not know whether such signalingpathways exist .A Role for the Rasl Protein in Seven less SignalingThe observat ion that the E(sev)SC locus is the Rasl locus,a gene whos e product is closely related to the vertebrate~21~ proteins (N euman-Silberberg et al., 1984; Broc k,1987), suggests that at least some of the signals initiatedby the act ivat ion of either the sev or E/p PTKs are trans-duced via activation of Rasl protein. Evidence for theinvolvem ent of ras proteins in signaling by other P TKscomes from several sources. The earl iest indicat ion thatRas act iv ity is necessary for the act ion of PTKs came fromexperim ents showing that the injection of monoclon al anti-bodies that inhibit Ras activity blocks the ability of onco-genie PTKs to transform cells (Smith et al. , 1988). Simi-larly, the expression of dominan t interfering ras proteinscan block the act ion of pp80- (Feig and Cooper,1988). These results are consistent with the idea that Rasmight be part of the signaling mech anism initiated byPTKs. Additional evidence for the associat ion of Ras act iv-i ty with signaling by PTKs come s from recent studies ofvulva1 developm ent in the nema tode Caenorha bditis ele-gans. A PTK, the product of the let-23 gene, is essent ialfor the decision by vulva1 precursor cells to adopt a primaryvulva1 fate (Aroian et al., 1990). M utations that reduce theactivity of a ras protein, the product of the let-60 gene , alsoprevent the vulva1 precursor cells from adopting a vulva1fate (Han and Sternberg, 1990; Beitel et al., 1990). Further-more, dominant mutat ions that act ivate the let-60 protein,or the overexpre ssion of the normal let-60 protein, causetoo many of the vulva1 precursor cel ls to comm it to thevulva1 fate and eliminate the need for the let-23 PTK .These data suggest that the function of the let-23 PTK invulva1 development is to activate the let-60 (ras) protein.How ever, there is no evidence that the let-23 and jet-60gene p roducts act in the sam e cell, a requirement for theirproposed relationship. All of these results are cons istentwith the hypothesis that activat ion of ras proteins is Con-trolled by PTK activity and is crucial for proper signalingby many PTKs. Our results are, however, also consistentwith the hypothesis that Rasl protein acts in a separatepathw ay that is merely required for the action of theSevenless-mediated pathway. Biochemical analysis wil lbe required to address w hether Seven less activation leadsto Rasl act ivat ion.A Model for Rasl Protein ActivationThe activity of a ras protein is regulated by the guaninenucleotide that is bound to the protein (reviewed by Bourneet al. , 1990, 1991). The GTP-bound state is act ive, while

8/12/2019 Simon_new

13/16


SEVENLESS

the GDP-bound state is not. The rat io of GTP:ras to GDP:ras is determined by two react ions. An act ive GTP :rasmolecule is inact ivated by the intr insic GTPase act iv ity ofthe ras protein. This inactivation reaction is greatly stimu-lated by the presence of a GTPase act ivat ing protein,rasGAP (Trahey and McCormick, 1987; Gibbs et al. ,1988). An inact ive GDP:ras molecule is act ivated by theexchange of the bound GDP molecule for a GTP molecule.The act ivat ion react ion has only been extensively charac-terized in budding yeast, where the CDC25 protein actsas a guanine nucleotide exchange factor to catalyze thereaction (Jones et al., 1991). Th e in vivo catalyst for gua-nine nucleotide exchange by Ras in metazoan organismsis not certain, although exchange-p romoting activitieshave been detected in relatively crude extra cts of cellsand, in one case , purified to near ho moge neity.

The discovery that E(sev)2A, that is, SOS, encodes aprotein that is related to the CDC25 protein suggests apossible biochemical l ink between the Sevenless andRasl act iv ity (see Figure 11). I f the similari ty between theCDC 25 protein and the SOS protein is indicative of similaractivities, then our genetic results are consis tent with amodel in which the act ivation of the sev PTK by l igandbinding stimula tes the activity of the SOS protein. The acti-vated SOS protein would then st imulate Rasl protein act iv-i ty by promoting the conversion of GDP:Rasl to GTP:Rasl. This model requires that the Rasl and SOS proteinseach perform their roles in signal transd uction in the devel-oping R7 cel l . We have demonstrated this direct ly forRasl, while Rogge et al. (1991) have show n that the domi-nant a ct ivated Soslc2 al lele acts in the R7 cel l . The mecha-nism for st imulat ion of SOS protein act iv ity by Sevenlesscould be achieved directly through tyrosine phosp horyla-tion of SOS protein by the sev PTK or indirectly. Alternately,act ivat ion by Sevenless could be indirect. I f so, then theproduc ts of the remaining E(sev) loci are the prime candi-dates for the roles of intermediaries.Our experiments cannot exclude the possibi l ity that the

Figure 11. A Mode l for the Action of the SOSand Rasl ProteinsSee text for description.

SOS protein might be a const itut ive act ivator of Rasl andthat Seven less could act solely by inhibiting the activity ofa negat ive regulator such as rasGAP. Biochemical studieswill be required to determine whether SOS protein activityis actually regulated by Seven less activity . The availabledata do, however, permit the conclusion that the level ofSOS protein activity can be a limiting step in the decisionby the presumptive R7 cel l to become an R7 cel l . Theobservat ions that the inact ivat ion of one copy of the SOSgene decreases the effect iveness of s ignaling by Sev-enless , while a dominan t allele, which presumably ele-vates SOS act iv ity, increases signaling show that smallchange s in the level of SOS function can be critical.

Only two of the seven E(sev) loci have been molecularlycharacterized. The f inding that two genes encode proteinswell-suited to participate in signaling sugg ests that thisgenet ic screen is capable of ident ify ing many componentsof the signaling pathw ay initiated by Seven less. Wha t aresome of the other part ic ipants that we might expect tofind? Since the targets forras protein action are undefined,the most interesting possibility is that one of the proteinsmight be an effector molecule that is regulated by the Raslprotein. Since ras proteins appear to participate in thesignaling by many PTK s, the strongest candidates for agene encoding a Rasl effector mo lecule are probably theremaining two E(sev) loci that also affect the Ellipse path-way. The p roducts of the E(sev) loci that af fect s ignalingby the sev PTK but not by the E/p PTK are candidates forpathw ay compon ents that differentiate the cellular re-sponse to Sevenless act ivat ion from the response to theact ivat ion of other P TKs.Experimental ProceduresGeneticsFly culture an d crosses were perform ed according to standard proce-dures. The mutagenesis screen for Enhancers ofsevenless mutationswas performed as fol lows. Male seP2 fl ies that were isogenic for thesecond and thi rd chromosomes were fed 25 mM ethylmethanesul fo-

8/12/2019 Simon_new

14/16

Cdl714

nate as described (Lewis and Bather, 1968) and mated to se@; CxD,DITM3, S b, ry, P[ry,se+] females. The Fl progeny were reared at22.7OC, an d Sb individualswere assayed by the reduced cornea1 pseu-dopupi l method for the presence of R7 photoreceptors (Franceschiniand Kirschfeld, 1971). Approximately 30,000 Sb Fl progeny werescreened. Individuals that displayed ab normal reduced cornea1 pseu-dopupi ls were crossed to s&*; CxD, DITM3, Sb, ry, P[ry,set@] fl iesand their progeny were examined for R7-minus reduced cornea1 pseu-dopupi ls. These individuals were used to map the E(sev) mutation toa chromosome and then balance the mutation. The chromosomesused for both segregation analysis and for balancing were FM7c, CyO,and TM3.

The X-l inked mutations were mappe d meiotical ly wi th the markersy, cv, f, and cer. Fema les heterozygous for the .E(sev), seti2 chromo-some and a y, cv, seti*, f, car chromo some were crossed to set@;CxD, DITMS, Sb , ry, P[ry,seve4] males. Individual vi rgin females werecol lected and scored for the E(sev) phenotype. These females werethen mated toy, cv, set@, f, car males, and their progeny were scoredfor the presence of the markers. Simi lar mapping crosses were per-formed for the second and thi rd chromosome E(sev) mutations usingsev? b, pr, c, px, sp and sevd2, t/r, St, cu, ST, e, ca flies, respectively.However, in these cases, the E(sev) phenotypes were more di fficul t toscore, an d we therefore only backcrossed fema les that were obviouslyE(sev). In al l cases, approximately 50 to 100 potential ly recombinantchromosomes were scored. We estimate an error rate of approximately2% in scoring the E(sev) phenotpye. The E(sev)24 locus was furthermappe d by the fai lure o f mutant al leles to compleme nt chromosomescarrying ei ther /(2)34,QDM7 or /(2)34EssF5 (Woodruff and Ashburner,1979; M. Ashburner and J. Roote, personal communication).HistologyFixation and sectioning (2 mm) of adul t Drosophi la eye was performedas described (Toml inson and Ready, 1987). Scanning electron micros-copy was performed as described (Kimm el e t al ., 1990).Mosaic AnalysisClones of cel ls homozygous for E(sev) mutations were generated byX-ray-induced mitotic recombination between heterozygous E(sev)+and E(sev) mutant chromosomes. Clones were visual ized by markingthe E(sev)+ chromosome wi th the cel l -autonomous w/ri fe+ gene. Cloneswere identi fied as whi te patches in otherwise pigmente d eyes. ForE(sev) mutations on the autosomes, the whi te+ gene was provided ona transposable element. The fol lowing elements were used: 2L,P[(w, ry)A-R]4-2 in 39E-F; 2R, P[(w, ry)A-RIO42 in 49D; 3L, P[w]33 in70C; 3R, P[(w, ry)D]3 in 90E (Hazelrigg et al ., 1984; Levis et al ., 1985;our unpubl ished data). For the X chromosome, the .E(sevj chromosomewas marked wi th whi te and the E(sev)+ chromosome was marked bythe normal chromosomal copy of whi te+. Al l fl ies carried one copy ofthe sev+ gene provided ei ther at i ts normal chromosomal location oron an autosome as part of a P element. Fl ies were i rradiated as fi rstinstar larvae (1 OOOR). The fl ies were reared at 25OC. The frequency ofclones was approximately 1 in 100 fl ies.

Clones of wi ld-type cel ls were generated in a simi lar man ner, exceptthat the E(sev) chromosome rather than the .E(sev)+ chromosome wasmarked by the whi te+ gene. The E(sev)+ cel ls are therefore marked aswhi te. For the E(sev)lA locus, E(sev)7AdP, sevd2 females were crossedto w, se@*; CxD, DITM 3, Sb, ry, P[ry, sev] males . For the Rasllocus, w~, se@; RaslW/TM 3 fema les were crossed to w~, sev?P[ry, se+]A males. P[ry, se@]A is a second chromosome insert. Thefl ies were i rradiated as above and reared ei ther at 22.7OC (for .E(sev)7Aexperiments) or at 23.5OC (for Rasl experiments). The di fference intemperature is due to a sl ight di fference in the level of expression fromthe two P[ry, se@] inserts. Progen y of these crosses that carried theE(sev) m utation and the P[ry, se+] insert were examined for clonesand sectioned as described above.Generation of Temperature-Sensi tive SW Al lelesSi te-di rected mutagenesis was used to generate the amino acid substi-tutions shown in Figure 1 (Sambrook et al ., 1989). The 17.5 kb DNAfragment containing the mutated sev al lele was placed in a modi fiedCarnegie 20 P-transformation vector and injected into set@, Iy em-bryos as previously described (Karess and Rubin , 1984). The 17.5 kbfragment consists of a 15 kb EcoRl fragment that has previously been

shown to be providing ful l sev function and an addi tional 2.5 kb 5 ofthe sev gene (Hafen et al ., 1987). The transformants were assayed bysectioning and by the reduced cornea1 ps eudopupi l method. The TM3,Sb, ry, P[ry, sever] chromosome was generated by transposi tion of theP[ry, sep] element.lsolatlon and Transformation of the Rash and SOS RegionsThe Rasl region was cloned from a l ibrary of Drosophi la genomic DNAcloned into the SFixl l (Stratagene, Inc.) vector, using a 700 bp Ncol-EcoRl fragment as a probe (Neuman-Si lberberg et al ., 1984). A 12 kbXhol-Notl fragment was cloned into the pW8 transformation vector(Klemenz et al ., 1987) and injected into w~, sevd2embryos as pre-viouslydescribed (Karess and Rubin, 1984). The Xhol si te l iesapproxi-mately4.5 kb 5of the ini tiating methionine of the Rasl gen e, The Notlsite is derived from the vector polylink er sequences and correspondsto a posi tion app roximately 5 kb 3of the Rasl term ination codon. TheRasl al leles from the .E(sev)3C mutant chromosomes were cloned as5.7 kb EcoRl fragments into the Zapl l (Stratagene, Inc.) vector. Themutant al leles were then cloned into pBluescript(+) or (-) for sequenc-ing. The mutant al leles cou ld be distinguished from the balancer-derived al leles on the basis of a sequence polymorphism in a noncod-ing region.The SOS region was cloned by chromosome walking using a cosmidl ibrary prepared by J. Tamkun. The starting point was the Hobo[ry+]element inserted in polytene band 34Dl (Blackman et al ., 1989). Al ibrary of genomic DNA was constructed in XFixl l and screened wi tha probe from the rosy gene. The walk was oriented and moni tored byhybridization of cosmid DNAs to wi ld-type and SOS deficiency chromo-somes. Genom ic blot analysis of deficiency chromosomes was per-formed by standard techniques using probes from the SOS region(Sambrook et al ., 1989). The SOS al leles from m utant chromosomeswere cloned into SFixl l as an approximately 18 kb Bgl l l fragment thatincludes the enti re gene. The al leles from the mutant chromosomeswere distinguished from the balancer-derived al leles on the basis ofpolymorphic Hindl l l si tes. The mutant al leleswere subcloned into plas-mid vectors for sequencing. The SOS transformation construct wasmade by inserting the IO kb Bgl l l -Xba l fragment into BarnHI-XbalcutpW8. The resul ting plasmid was injected into w~, seti2 embryos aspreviously described (Karess a nd Rubin, 1984).

The 10.3 kb Xbal fragment that contains the entire SOS gene wasused to screen an eye imaginal disc cDNA in I@10 constructed by Dr.A. Cowman. A single cDNA lsosT was then subcloned into pBlue-Script(-) for sequen cing.Transcript AnalysisPolyadenylated RNA was prepared from imagin al discs and from em-bryos, electrophoresed through a 1% agarose-formaldehyde gel , andblotted onto ni trocel lulose by standard methods (Sambrook et al .,1989). The blot was probed wi th the fragment of the sosT cDNA shownin Figure 9.

Nuclease Sl analysis and primer extension were performed exactlyas described (Sambrook et al ., 1989). The probes used for Sl analysiswere uni formly lab eled single-stranded probes corresponding to thereverse complements of nucleotides 0 to 610 and 0 to 731. The primersused correspond to nucleotides 581 to 610 and 706 to 731.DNA Sequenc ingAl l DNA sequences were determined using the dideoxy c hain termina-tion method (Sanger et al ., 1977) using Sequenase (U.S. Biochemical).Templates for the determination of the 6601 bp genomic SOS sequencewere generated by sonication of plasmid DNA and insertion of thesonicated fragments into the vector MlSm plO. The enti re sequencewas determined on both strands. The sequences of SOS cDNAs weredetermined on one strand using templates generated from plasmidscontaining EcoRl fragments inserted into the pBlueScript(-) vectorusing an Erase-a-Base ki t (Promega). Mutant al leles of SOS were se-quenced on one strand using ol igonucleotide primers spaced approxi-mately every 20 0 bp. The exons of the mutant Rasl al leles were se-quenced on one strand using ol igonucleotide primers.AcknowledgmentsWe wish to thank M. Cyert, W. Ferguson, J. Kaplan, B. Meyers, H.

8/12/2019 Simon_new

15/16


Varmus, and members of our laboratory for their helpful suggestionsregarding the manuscript. In addi tion, we wish to acknowledge theimportant contributions of J. Kaplan and I. Hariharan to the design ofthese experiments. We also thank M . Ashburner and J. Roote forinformationaboutthe34D region and for Drosophi lastrains, J. Tamkunfor providing the Drosophi la genomic cosmid l ibrary, D. Tyler for DNAsequencing, D. Hackett for microinjection of embryos, T. Cutforth fordeficiency mapp ing of the E(sev)3C locus, Jorge Santiago-Blay forscanning electron microscopy, and I. Hariharan for an RNA blot.D. D. L. B. was a C. J. Martin Fel low. M. A. S. is a Luci l le P. MarkeyScholar in Biomedica l Sciences.

The costs of publ ication of this article were defrayed in part by thepayment of page charges. This article must therefore be herebymarked advertisement in accordance wi th 18 USC Section 173 4solely to indicat e this fact.Received August 21, 1991; revised S eptember 12, 1991.

Ambrosio, L., Mahow ald, A. P., and Perrimon, N. (1989). Requirem entof the Drosophi la raf homolog ue for torso function. Nature 342, 288-291.Aroian, Ft. V., Koga, M., Mende l , J. E., Ohshima, Y., and Sternberg,P. W. (1990). The let-23 gene necessary for Caenorhabdi tis e/egansvulva1 induction encodes a tyrosine kinase of the EGF receptorsubfam-i ly. Nature 348, 893-699.Baker, N. E., and Rubin, G. M. (1989). Effect on eye development ofdominan t mutations in Drosophi la homolog ue of the EGF receptor.Nature 340 , 150-l 53.Basler, K., and Hafen, E . (1988). Control of photoreceptor cel l fate bythe sevenless protein requires a functional tyrosine kinase domain.Cel l 54, 299-311.Bei tel , G. J., Clark, S. G., and Horvi tz, H. R. (1990). Caenorhabdi tiselegans fas gene let-60 acts as a swi tch in the pathway of vulva1 induc-tion. Nature 348, 503-509.Bishop, J. M. (1991). Molecular themes in oncogenesis. Cel l 64,235-248.Blackman, R. K., Koehler, M. M. D., Grimai la, R., and Gelbart, W. M.(1989). Identi fication of a ful ly-functional hobo transposable elementand i ts use for germ-l ine transformation of Drosophi la. EMBO J. 8,211-217.Bourne, H. R.. Sanders, D.A., and McCormick, F. (1990). TheGTPasesuperfam ily: a conserved switch for diverse cell functions. Nature 348,125-132.Bourne, H. R., Sanders, D.A., and McCormick, F. (1991). TheGTPasesuperfami ly: conserved structure and molecular mechanism. Nature349, 117-127.Bowtel l , D. D., Simon , M. A., and Rubin, G. M. (1988). Nucleotidesequence and structure of the sevenless gene of Drosophi la melano.gaster. Genes. Dev. 2, 620-834.Brock, H. W. (1987). Seq uence and genomic structure of ras homo-logues Dmras85D and Dmras64B of Drosophi la melanogaster. Gene51, 129-137.Broek, D., Toda, T., Michael i , T., Levin, L., Birchmeier, C., Zol ler, M .,Powers, S.. and Wigler, M. (1987). The S. cerevisiae CDC25 geneproduct regulates the RASladenylate pathway. Cel l 48, 789-799.Camonis, J. H., Kalekine, M., Gondre, B., Garreau, H., Boy-Marcotts,E., and Jacquet, M. (1966). Characterization, cloning and sequenceanalysis of the CDC25 gene which controls the cyclic AMP level of S.cerevisiae. EMBO J. 5, 375-380.Cantley, L. C., Auger, K. R., Carpenter, C., Duckworth. B., Graziani ,A., Kapel ler, R., and Sol toff, S. (1991)Oncogenesandsignal transducl ion. Cel l 64, 281-302.Capon, D. J., Chen, E. Y., Levinson, A. D., Seeburg, P. H., and Goed-del , D. V. (1983). C omplete nucleotide sequences of the T24 humanbladder carcinoma oncogene and i ts normal homologue. Nature 392,33-37.Carthew. R. W., and Rubin. G. M. (1990). seven 1 absentia, a gene

required for speci fication of R7 cel l fate in the Drosophi la eye. Cel l 63,561-577.Casanova , J., and Struh l, G. (1989). Localized surface activity of torso,a receptor tyrosine kinase, specifies term inal body pattern in Drosoph-i la. Genes Dev. 3, 2025-2038.Chant, J., Corrado, K., Pringle, J. R., and Herskowitz. I. (1991). YeastBUD5, encoding a putative GDP-GTP exchange factor, is necessaryfor bud si te selection and interacts wi th bud formation gene SEW.Cel l 65, 1213-1224.Cross, M., and Dexter, T. M. (1991). Growth factors in development,transformation, and tumorigenesis. Cel l 64, 271-280.Crechet, J.-B., Poul let, P., Mistou, M.-Y., Parmeggiani , A., Camonis,J., Boy-Marcotte, E., Damak, F., and Jacquet, M. (1990). Enhancemen tof the GDP-GTP exchange of RAS proteins by the carboxy-terminaldoma in of SCD25. Science 248,866-868.Damak, F., Boy-Marcotte, E., Le-Roscouet, D., Gui lbaud, R., and Jac-quet, M. (1991). SDC25, a CDCZdl ike gene which contains a RAS-activating doma in and is a dispensable gene of S. cerevisiae. Mol .Cel l . Bio l . 77, 202-212.Feig, L. A., and Cooper, G. M. (1988). Inhibi tion of NIH 3T3 cel l prol i fer-ation by a mutant ras protein wi th preferential affini ty for GDP. M ol .Cel l . B iol . 8, 3235-3243.Fincham, V. J., and Wyke, J. A. (1988). Local ization of temperature-sensi tive transformation mutations and back mutations in the RousSarcoma Virus src gene. J. Virol . 58, 894-699.Franceschini , N., and Kirschfeld, K. (1971). Les phenomenes de pseu-dopi l le dans Ioei l compose de Drosophi la. Kybernetik. 9, 159-162.Gibbs, J. B., Schaber, M. D., Al lard, W. J., Sigal , I. S., and Scolnick,E. M. (1988). Pu ri fication of ras GTPase activating protein from bovinebrain. Proc. Natl . Acad. Sci . USA 85, 5026-5030.Hafen, E., Basler, K., Edstroem, J. E., and Rubin, G. M. (1987). Sev-enless, acel l -speci fic homeoticgeneof Drosophi la, encodesaputativetransmembrane receptor wi th a tyrosine kinase domain. S cience 236,55-63.Han, M., and Sternberg, P. W. (1990). let-60, a gene that specifies cel lfates during C. elegans vulva1 induction, encodes a ras protein. Cel l63, 921-931.Hart, A. C., Kramer, H., Van Vactor, D. L., Jr., Paidhunga t, M., andZipursky, S. L. (1990). Induction of cel l fate in the Drosophi la retina: thebride of sevenless protein is predicted to contain a large extracel lulardomain and seven transmembrane segments. Genes Dev. 4, 1835-1847.Hazelrigg, T., Levis, R., and Rubin, G. M. (1984). Transformation ofwhi te locus DNA in drosophi la: dosage compensation, zesre interac-tion, and posi tion effects. Cel l 36, 489-481.Henkemeyer, M., West, S . R., Gertler, F. B., and Hoffmann, F. M.(1990). A novel tyrosine kinase-independent function of Drosophi la ablcorrelates wi th proper subcel lular local ization. Cel l 63, 949-960.Hughes, D. A., Fukui , Y., and Yamam oto, M. (1990). H omologousactivators of ras in fission and budding yeast. Nature 344, 355-357.Hunter, T. (1991). Cooperation between oncogenes. Cel l 84,249-270.Jones, S., Vignais, M.-L., and Broach, J. R. (1991). TheCDC25proteinof S. cerevisiae promotes exchange of guanine nucleotides bound toras. Mol . Cel l . Biol . 17, 2641-2846.Karess, R. E., and Ru bin, G. M. (1984). Analysis of P transposableelement functions in Drosophi la. Cel l 38, 135-146.Kemphues, K. J.. Raff, E. C., Raff, R . A., and Kaufm an, T. C. (1980).Mutation in a testis-specific f3-tubul in in Drosophi la: analysis of i ts ef-fects on meiosis and map location of the gene. Cel l 21, 445-451.Kimm el, 8. E., Heberlein, U., and Rubin, G. M. (1990). The home odoma in protein rough is expressed in a subset of cel ls in the developingDrosophi la eye where i t can specify photoreceptor cel l subtype. GenesDev. 4, 712-727.Klemenz, R., Weber, U., and Gehring, W. J. (1987). The whi te geneas a marker in a new P -element vector for gene transfer in Drosophi la.Nucl . Acids Res. 75, 3947-3959.Kramer, H., Cagan, R. L., andZipursky,S. L. (1991). lnteractionofbride

8/12/2019 Simon_new

16/16

Cdl716

ofsevenless mem brane-bound l igandand thesevenless tyrosin-kinasereceptor. Nature 352, 207-212. GenBenk Accession NumberLevis, Ft., Hazelrigg, T., and Rubin, G. M. (1985). Separable c&actingcontrol elements for expression of the whi te gene of Drosophi la. EMBOJ. 4,3489-3499.

The accession number for the sequence reported in this article isM77501.

Lewis, E. B., and Bather, F. (1968). Method of feeding ethyl methane-sulphonate to Drosophi la males. Dros. Inf. Serv. 43, 193-194.Lindsley, D. L., Sandier, L., Baker, B. S., Carpenter, A. T. C., Denel l ,R. E., Hal l , J. C., Jacobs, P. A., M iklos, G. L. G., Davis, 8. K., Geth-mann, R. C., Hardy, R. H., Heasler, A., Mi l ler, S. M., Nozawa, H., Parry,D. M., and Gould-Somero, M. (1972). Segmen tal aneuploidy and thegross genetic structure of the Drosophi la genome. Genetics 77, 157-184.Livneh, E., Glazer, L., Segal , D., Schlessinger, J., and Shi lo, 8. Z.(1985). The Drosophi la EGF receptor gene homolog: conservation ofboth hormone binding and kinase domains. Cel l 40, 599-607.Mul ler, H. J., League, B. B., and Offermann, C. A. (1931). Effects ofdosage changes of sex-l inked genes, and the compensatory effectsof other gene di fferences between male and female. Anat. Rec. 51(SUPPI.), 1 IO.Neuman-Si lberberg, F. S., Schej ter, E., Hoffmann, F. M., and Shi lo,B. 2. (1984). The Drosophi la ras oncogenes: structure and nucleotidesequence. Cel l 37, 1027-1033.Powers, S., Gonzales, E., Christensen, T.. Cubert, J., and Broek, D.(1991). Fun ctional cloning of BUD & a CDC2brelated gene from S.cerevisiae that can suppress a dominant-negative RAS2 mutant. Cel l65, 1225-1231.Price, J. V., Cl i fford, R. J., and Schi ipbach, T. (1989). The maternalventral izing locus torpedo is al lel ic to faint l i l t/e bal l , an embryoniclethal , and encodes the Drosophi la EGF receptor homolog . Cel l 56,1085-l 092.Reinke, R., and Zipursky, S. L. (1988). Ce l l -cel l interaction in the dro-sophi la retina: the bride ofsevenless gene is required in photoreceptorcel l R8 for R7 cel l development. Cel l 55, 321-330.Robinson, L. C., Gibbs, J. B., Marshal l, M. S., Sigal , I. S., and Tatchel l ,K. (1987). CDC25: a component of the RAS-adenylate cyclase pathwayin Saccharomyces cefevisiae. Science 235, 1218-1221.Rogge, R. D., Karlovich, C. A., and Banerjee, U. (1991). Genetic dis-section of a neurodevelopmenta l pathway: Son of sevenless functionsdownstream of the sevenless and EGF receptor tyrosine kinases. Cel l64, 39-48.Sambrook, J., Fri tsch, E. F., and Maniatis, T. (1989). Molecular Clon-ing: A Laboratory Manual , Second Edi tion (Cold Spring Harbor, NewYork: Cold Spring Harbor Laboratory Press).Sanger, F., Nicklen, S., and Coulson, A. (1977). DNAsequencing wi thchain terminating inhibi tors. Proc. Natl . Acad. Sci . USA 74, 5463-5467.Schej ter, E. D., and Shi lo, B.-Z. (1989). The Drosophi la EGF receptorhomolog (DER) g ene is al lel ic to faint l i tt le bal l , a locus essential forembryonic development. Cel l 56, 1093-l 104.Smith, M. R., DeGudicibus, S. J., and Stacey, D. W. (1988). Require-ment for c-ras proteins during vi ral oncogene transformation. Nature320,540~543.Sprenger, F., Stevens, L. M., and Ni isslein-Volhard, C. (1989). TheDrosophi la gene torso encodes a putative receptor tyrosine kinase.Nature 338, 478-483.Toml inson, A., and Ready, D. F. (1988). Sevenless: a cel l speci fichomeotic mutation of the Drosophi la eye. Science 237, 400-402.Toml inson, A., and Ready, D. F. (1967). Cel l fate in the Drosophi laomm atidium. Dev. Biol . 123, 264-275.Trahey, M.. and McCormick, F. (1967). A cytoplasmic protein stimu-lates normal N-ras21 GTPase, but does not affect oncogenic mutants.Science 238,542-545.Woodruff, R. C., and Ashburner, M. (1979). The genetics of a smal lautosomal region of Drosophi la melanogaster containing the structuralgene for alcohol dehydrogenase. II. Lethal mutations in the region.Genetics 92, 133-149.

Documents

Simon_new