Caenorhabditis elegans lin-45 raf Is Essential for Larval Viability, Fertility and the Induction of Vulval Cell Fates

Corresponding author: Kerry Kornfeld, Washington University School of Medicine, Campus Box 8103, 660 S. Euclid Ave., St. Louis, MO 63110., kornfeld{at}molecool.wustl.edu (E-mail)

Communicating editor: P. ANDERSON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The protein kinase Raf is an important signaling protein. Raf activation is initiated by an interaction with GTP-bound Ras, and Raf functions in signal transmission by phosphorylating and activating a mitogen-activated protein (MAP) kinase kinase named MEK. We identified 13 mutations in the Caenorhabditis elegans lin-45 raf gene by screening for hermaphrodites with abnormal vulval formation or germline function. Weak, intermediate, and strong loss-of-function or null mutations were isolated. The phenotype caused by the most severe mutations demonstrates that lin-45 is essential for larval viability, fertility, and the induction of vulval cell fates. The lin-45(null) phenotype is similar to the mek-2(null) and mpk-1(null) phenotypes, indicating that LIN-45, MEK-2, and MPK-1 ERK MAP kinase function in a predominantly linear signaling pathway. The lin-45 alleles include three missense mutations that affect the Ras-binding domain, three missense mutations that affect the protein kinase domain, two missense mutations that affect the C-terminal 14-3-3 binding domain, three nonsense mutations, and one small deletion. The analysis of the missense mutations indicates that Ras binding, 14-3-3-binding, and protein kinase activity are necessary for full Raf function and suggests that a 14-3-3 protein positively regulates Raf-mediated signaling during C. elegans development.

A small number of distinct signaling pathways are used reiteratively during animal development to control many different cell fate choices. These pathways have been conserved during evolution, and the current understanding of the identity and function of the signaling proteins that comprise these pathways is based on analyses of several organisms, including vertebrates, Caenorhabditis elegans, and Drosophila. One such pathway consists of seven core signaling proteins (CANTLEY et al. 1991 ; DICKSON and HAFEN 1994 ; MARSHALL 1994 ). The pathway is activated by a protein ligand such as epidermal growth factor (EGF). This ligand binds a receptor tyrosine kinase (RTK), resulting in dimerization and autophosphorylation. These phosphotyrosine residues create docking sites for proteins that contain Src homology 2 (SH2) domains, such as the SH2-SH3-SH2 adapter Grb2/SEM-5. This adapter protein interacts with a guanine nucleotide exchange factor such as Son of Sevenless (Sos). Sos activates the small GTPase Ras by catalyzing the conversion of inactive GDP-bound Ras to active, GTP-bound Ras. A crucial effector of Ras is Raf, a protein kinase that phosphorylates and activates a mitogen-activated protein (MAP) kinase kinase named MEK, which phosphorylates and activates extracellular-signal-regulated kinase (ERK), a member of the MAP kinase family. ERK phosphorylates a variety of proteins, including transcription factors, and thus is likely to be an important link between signaling proteins and proteins that mediate particular cell fates (TREISMAN 1996 ). In addition to these core signaling proteins, regulatory and effector proteins feed into and out of this pathway at multiple levels.

In the nematode worm C. elegans, an RTK/Ras/ERK signaling pathway has been analyzed most extensively during the formation of the hermaphrodite vulva (HORVITZ and STERNBERG 1991 ; GREENWALD 1997 ; KORNFELD 1997 ; STERNBERG and HAN 1998 ). In third larval stage hermaphrodites, six ventral epidermal blast cells called P3.p, P4.p, P5.p, P6.p, P7.p, and P8.p (Pn.p cells) lie along the anterior-posterior axis. Each of these Pn.p cells can adopt any of three distinct fates: the primary (1°) vulval cell fate (eight descendants), the secondary (2°) vulval cell fate (seven descendants), or the tertiary (3°) nonvulval cell fate (two descendants). The anchor cell of the somatic gonad signals P6.p to adopt the 1° fate by activating an RTK/Ras/ERK signaling pathway. P6.p signals P5.p and P7.p to adopt the 2° fate by activating lin-12, which is similar to the receptor Notch. P3.p, P4.p, and P8.p receive neither signal and adopt the 3° fate. The 22 descendants of P5.p, P6.p, and P7.p generate the vulva, a specialized epidermal structure used for egg laying. Similar pathways also control other cell fate decisions including the differentiation of the excretory duct cell, which is necessary for larval viability, and the progression of germ cells through pachytene, which is necessary for hermaphrodite fertility (CHURCH et al. 1995 ; YOCHEM et al. 1997 ).

The anchor cell and P6.p communicate using a highly conserved signal transduction pathway that includes the lin-3 ligand, which is similar to EGF; the let-23 RTK; the sem-5 adapter protein; let-341 guanine nucleotide exchange factor; let-60 Ras; lin-45 Raf; mek-2 MEK; mpk-1 ERK; and lin-1 ETS transcription factor (KORNFELD 1997 ; STERNBERG and HAN 1998 ). With the exception of lin-1, a mutation that reduces the activity of one of these genes causes all six Pn.p cells to adopt the nonvulval 3° fate, resulting in a vulvaless (Vul) phenotype. By contrast, a constitutively active form of one of these genes causes all six Pn.p cells to adopt the 1° or 2° vulval fate, resulting in a multivulva (Muv) phenotype characterized by ectopic patches of vulval tissue. The signaling pathway negatively regulates the LIN-1 ETS transcription factor, so a lin-1(lf) mutation causes a Muv phenotype. These phenotypes are dramatic, and thus the extent of vulval induction can serve as an easily visualized readout of Ras pathway activity.

Here we focus on the Raf protein kinase. Raf has been analyzed extensively using purified Raf protein and Raf expressed in vertebrate cultured cells. Prior to ligand stimulation, Raf is catalytically inactive and localized to the cytoplasm in a multiprotein complex. The initial event in Raf activation is the recruitment of Raf to the plasma membrane through a high-affinity interaction between the switch 1 region of activated Ras-GTP and the N-terminal minimal Ras-binding domain of Raf (VOJTEK et al. 1993 ; ZHANG et al. 1993 ; STOKOE et al. 1994 ; FINNEY and HERRERA 1995 ; MARAIS et al. 1995 ). Plasma membrane-localized Raf is then activated by a mechanism that has yet to be fully characterized, but it appears to depend on relieving the interaction between the N-terminal regulatory domain and the C-terminal kinase domain (STANTON et al. 1989 ; HEIDECKER et al. 1990 ). Activated Raf phosphorylates MEK on two serine residues, which greatly stimulates MEK kinase activity (ALESSI et al. 1994 ; GARDNER et al. 1994 ; ZHENG and GUAN 1994 ). Raf has been reported to interact with several other proteins that may positively or negatively modulate Raf activity. These include 14-3-3, Ksr, Hsp90, Cdc37, and a variety of protein kinases and phosphatases (MORRISON and CUTLER 1997 ). The physiological significance of many of these interactions has yet to be fully characterized.

By contrast to the large number of studies of vertebrate Raf, the C. elegans lin-45 raf gene has not been analyzed extensively. HAN et al. 1993 molecularly characterized lin-45 and demonstrated that C. elegans Raf is similar to vertebrate Raf in conserved regions (CR) 1, 2, and 3. A single mutation in the lin-45 gene, lin-45(sy96), which reduces but does not eliminate the activity of the lin-45 locus, was identified. The analysis of this mutant allele indicated that Raf is important for RTK/Ras/ERK signaling pathways at multiple times during development (HAN et al. 1993 ). Here we present an analysis of 13 additional alleles of lin-45, including weak, intermediate, and strong loss-of-function mutations. The strong loss-of-function or null alleles demonstrate that Raf is required for Ras-mediated signaling during larval development, oocyte maturation, and vulval induction. These mutations were analyzed molecularly, and eight are missense mutations that identify functionally significant residues in the Ras-binding domain, the protein kinase domain, and the C-terminal 14-3-3 binding domain. Our results demonstrate the importance of these residues and domains for Raf function and suggest that the interactions of Raf with these binding proteins are important for Raf activity.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

General methods and strains:
C. elegans strains were cultured as described by BRENNER 1974 and grown at 20° unless otherwise noted. The wild-type strain and parent of all mutant strains was N2. The following mutations cause a visible phenotype, were used to mark chromosomes, and are described by RIDDLE et al. 1997 : LGI, sup-11(n403); dpy-5(e61); LGIII, unc-79(e1068); dpy-17(e164); and LGIV, unc-24(e138); unc-5(e53); dpy-20(e1282); bli-6(sc16). Standard techniques were used to mutagenize animals and screen for mutants, separate the lin-45 mutations from the let-60, lin-15, or gon-2 unc-29 mutations, backcross newly identified mutations using N2, position mutations on the genetic map, perform complementation tests, and generate double mutants (BRENNER 1974 ).

The following mutations that affect vulval development were used: lin-15(n309) is a recessive, strong loss-of-function or null allele caused by an 13-kb deletion that removes all of lin-15A and most of lin-15B (CLARK et al. 1994 ). let-60(n1046 G13E) is a semidominant, gain-of-function allele (BEITEL et al. 1990 ). Similar mutations of vertebrate Ras result in oncogenic proteins that have reduced GTPase activity. lin-1(e1275 R175Stop) is a recessive, heat-sensitive, partial loss-of-function allele that truncates LIN-1 downstream of the ETS DNA-binding domain (BEITEL et al. 1995 ). lin-31(n1053 Trp57Stop) is a recessive, strong loss-of-function or null allele (MILLER et al. 2000 ). lin-12(n137 S872P) is a dominant gain-of-function allele that affects the extracellular domain of the LIN-12 Notch receptor (GREENWALD and SEYDOUX 1990 ). mek-2(n2516 E238K) and mek-2(n2678 D213N) are recessive, strong loss-of-function or null alleles that alter highly conserved residues in the kinase domain (KORNFELD et al. 1995A ). mpk-1(ga117) is a recessive, strong loss-of-function allele caused by a single-nucleotide change in a splice site upstream of residue 24 that eliminates detectable MPK-1 protein (LACKNER and KIM 1998 ). lin-45(sy96) is a recessive, partial loss-of-function allele caused by a single nucleotide change in a splice site upstream of residue 229 that eliminates most but not all functional lin-45 mRNA (HAN et al. 1993 ). nDf41 is a deletion of 1.5 map units of chromosome IV that fails to complement genes positioned to the left (pat-8) and right (dif-1) of lin-45 (RIDDLE et al. 1997 ). The lin-45 alleles n1924, n1925, n2018, n2506, n2510, n2520, n2523, oz166, oz178, oz201, dx19, dx84, and dx89 are described here.

Identification of the lin-45 mutations:
We previously described a screen for suppressors of the let-60(gf) Muv phenotype (LACKNER et al. 1994 ; KORNFELD et al. 1995A , KORNFELD et al. 1995B ; JACOBS et al. 1998 ; JAKUBOWSKI and KORNFELD 1999 ). In brief, we mutagenized let-60(n1046) hermaphrodites with ethyl methane sulfonate (EMS), placed 2794 F₁ self-progeny on separate petri dishes, and examined F₂ self-progeny for non-Muv animals at 22.5°. We identified 33 independently derived mutations that reduced the penetrance of the Muv phenotype from 93 to <10%, including the lin-45 alleles n2506, n2510, n2520, and n2523. In a related screen that was described previously (BEITEL et al. 1990 ), lin-8(n111); let-60(gf) hermaphrodites were mutagenized with EMS and non-Muv F₁ self-progeny were picked to separate petri dishes at 25°. Ten extragenic mutations identified in this screen met the criteria described above, including the lin-45 alleles n1924 and n1925. lin-45(n2018) was isolated in a previously described screen (BEITEL et al. 1990 ; CLARK et al. 1992 ) by mutagenizing lin-15(n765ts) hermaphrodites with EMS and examining the progeny of 38,000 F₁ animals for non-Muv animals. The lin-45 alleles oz166, oz178, and oz201 were generated by mutagenizing N2 hermaphrodites with EMS and identified in F₂ clonal screens for mutations that caused sterility. The lin-45 alleles dx19, dx84, and dx89 were generated by irradiating adult hermaphrodites with 310 nm ultraviolet light (12–18 sec of 25 J/m²/sec). dx19 was generated by mutagenizing N2 hermaphrodites and identified in an F₂ clonal screen for gonadal defects. dx89 was generated by mutagenizing gon-2(q388) unc-29(e1072) hermaphrodites and identified in an F₂ clonal screen for enhancement of the gon-2 gonadal defect. dx84 was generated by mutagenizing N2 hermaphrodites and identified by screening for deletions in the lin-45 locus using the outer primers 5'-GACATATTTTGTCAGGTAATCG-3' and 5'-GTCTAAGTGAAGAACATTCGG-3' and the inner primers 5'-TCTCAATTATTCAGGAGCTCG-3' and 5'-GAGTCAATTTTGGAAGAATTATG-3' according to the method described by DENBURG et al. 1998 .

Genetic mapping and complementation tests:
The following genetic mapping and complementation experiments support the conclusion that the identified mutations are alleles of lin-45. n1924, n1925, n2506, n2520, and n2523 displayed linkage to dpy-20 IV (data not shown); three factor-mapping experiments indicated that n1924 is to the left of unc-24 IV and n2506 is positioned between bli-6 IV and unc-24 IV (data not shown), an 0.36 map unit interval that contains lin-45 (RIDDLE et al. 1997 ). Each of these five alleles failed to complement the other four alleles for the suppression of the let-60(gf) Muv phenotype (data not shown), indicating that they represent a single complementation group. n2510 displayed linkage to dpy-20 IV and failed to complement the vulval defect caused by n2018 and lin-45(sy96) (data not shown). oz166 and oz178 were mapped between bli-6 IV and unc-24 IV (data not shown). oz166 failed to complement the lethality caused by n1924, n1925, n2520, and n2506 (Table 2) and the vulval defects caused by sy96, n2018, and dx19 (data not shown). dx19 and dx89 displayed linkage to chromosome IV (data not shown).

Determination of DNA sequences of lin-45 alleles:
For each of the lin-45 alleles, genomic DNA was derived from homozygous mutant adult hermaphrodites and was amplified by polymerase chain reaction (PCR) according to WILLIAMS et al. 1992 . lin-45 contains 12 exons (HAN et al. 1993 ). To identify the molecular lesion in the 12 alleles identified in screens for visible phenotypes, the oligonucleotide primers Raf-Fwd-Ex1 (5'-GTCACATCATCTCAAACGCC-3') and Raf-Rev-Ex4 (5'-CCAGGCAGTCGGGATGCG-3') were used to amplify a DNA fragment containing 25 bp upstream of exon 1, intron 1, exon 2, intron 2, exon 3, intron 3, and 17 bp of exon 4. Raf-Fwd-Ex3 (5'-ACAAGAATCTATTGAATTATCGG-3') and Raf-Rev-In4 (5'-CAAATGTTGGGACAACATTGG-3') were used to amplify a DNA fragment containing 51 bp of intron 3, exon 4, and 41 bp of intron 4. Raf-Fwd-In4 (5'-CTCACCTCTGCTTCAGAAAC-3') and Raf-Rev-Ex7 (5'-GGGGATTGTCCAAGAGTAGG-3') were used to amplify a DNA fragment containing 29 bp of intron 4, exon 5, intron 5, exon 6, intron 6, and 104 bp of exon 7. Raf-Fwd-Ex7 (5'-CTCGAATGAATCGTCTTCACC-3') and Raf-Rev-In8 (5'-TTTGTAGAACTGCCGGTTTGC-3') were used to amplify a DNA fragment containing 191 bp of exon 7, intron 7, exon 8, and 38 bp of intron 8. The amplified regions of exon 7 overlap by 80 bp. Raf-Fwd-In8 (5'-AAAAACGCTCAAAACTTCTCTC-3') and Raf-Rev-Ex10 (5'-CAGTATATCAATGATAGCACCC-3') were used to amplify a DNA fragment containing 28 bp of intron 8, exon 9, intron 9, and 22 bp of exon 10. Raf-Fwd-Ex9 (5'-TATAGACATATTCATGTTCAAGG-3') and Raf-Rev-In11 (5'-CCCCATAATAAATCATAGTTCTAC-3') were used to amplify a DNA fragment containing 45 bp of intron 9, exon 10, intron 10, exon 11, and 31 bp of intron 11. Raf-Fwd-In11 (5'-CATGAATAACTCCACTACACTG-3') and Raf-Rev-Ex12b (5'-GGTACATATTCGGGGAGAGACGAG-3') were used to amplify a DNA fragment containing 68 bp of intron 11, exon 12, and 58 bp downstream of exon 12.

We purified PCR-amplified DNA fragments and determined the complete sequences of both strands using the amplification primers and additional primers positioned inside the larger DNA fragments with an ABI 373A DNA sequencer (Applied Biosystems, Foster City, CA).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Isolation and molecular characterization of lin-45 raf alleles:
We identified 12 alleles of lin-45 raf by conducting several different genetic screens (see MATERIALS AND METHODS). The alleles n1924, n1925, n2506, n2510, n2520, and n2523 were identified in screens for mutations that suppressed the Muv phenotype caused by a gain-of-function (gf) mutation that constitutively activates the let-60 ras gene. The allele n2018 was identified in a screen for mutations that suppress the Muv phenotype caused by a lin-15(lf) mutation. lin-15 is a negative regulator of vulval cell fates and appears to act upstream of or parallel to let-60 ras. The alleles oz166, oz178, oz201, dx19, and dx89 were isolated in screens for mutations that cause sterility or gonadal defects.

Three approaches were used to demonstrate that these 12 mutations are alleles of lin-45. First, genetic mapping experiments were used to position the mutations in the region of chromosome IV that contains lin-45 (see MATERIALS AND METHODS). Second, complementation tests showed that these mutations failed to complement the previously characterized lin-45(sy96) allele or other alleles in this collection (see MATERIALS AND METHODS). Third, DNA from each mutant was isolated and used to determine the DNA sequence of the entire lin-45 coding region and the regions of introns close to splice sites on the basis of the gene structure characterized by HAN et al. 1993 . These experiments revealed a nucleotide change that affects the lin-45 coding region in 11 of these 12 alleles (Fig 1). We did not identify the mutation in the oz166 allele, indicating that this mutation may be outside the region that was analyzed. A thirteenth allele, dx84, was isolated by using PCR to screen for deletions in the lin-45 locus (see MATERIALS AND METHODS).

View larger version (48K): In this window In a new window Download PPT slide   Figure 1. Molecular analysis of lin-45 alleles. (A) The codon number, wild-type amino acid, and DNA sequence (HAN et al. 1993 ) are followed by the corresponding information for the mutant alleles. Of 11 nucleotide changes, 10 were GC-to-AT transitions, the characteristic mutation caused by ethyl methanesulfonate (COULONDRE and MILLER 1977 ), the mutagen used to generate all the mutations except dx19, dx89, and dx84. n1924 had an AT-to-TA transition. dx84 is an 851-bp deletion that removes 28 bp of intron 10, 412 bp of exon 11, and 411 bp of intron 11. Exon 11 encodes amino acids 601–738, so these residues will be absent from any proteins produced by the dx84 allele. The predicted LIN-45 proteins are drawn to scale. CR 1, 2, and 3 are darkly shaded. The RBD, CRD, and C-terminal 14-3-3 binding domain (14-3-3) are lightly shaded. Arrowheads indicate the positions of affected amino acids. We detected a single nucleotide change in the oz166 allele. However, the same nucleotide change was present in the strain prior to mutagenesis, and thus we conclude that this mutation does not cause the oz166 phenotype. The oz166 mutation probably lies outside the region that was sequenced. (B) Alignments of the amino acid sequences of C. elegans LIN-45 (HAN et al. 1993 ), human Raf-1 (BONNER et al. 1986 ), and Drosophila melanogaster D-Raf (NISHIDA et al. 1988 ) for portions of the Ras-binding domain, protein kinase domain, and C-terminal 14-3-3-binding domain; identical residues are shaded. Numbers indicate the position of the first amino acid in each line. Missense changes are shown. Arrows and a bar overlie residues of the Ras-binding domain that form ß-sheets (B1 and B2) and an -helix (A1), respectively (see Fig 2).

Alignments of Raf proteins from many species reveal three highly conserved regions designated CR1, CR2, and CR3. CR1 contains the minimal Ras-binding domain (RBD) and a cysteine-rich domain (CRD; Fig 1A). The alleles n2018, n1925, and n2506 contained missense mutations in the Ras-binding domain. n2018 changes a conserved proline to serine, n2506 changes a conserved arginine to tryptophan, and n1925 changes a nonconserved arginine to tryptophan (Fig 1B). The alleles dx19 and dx89 contained the same mutation, a nonsense change that is predicted to terminate the LIN-45 protein at residue 192 (Fig 1A). The allele n2510 contained a nonsense mutation that is predicted to terminate LIN-45 at residue 448 (Fig 1A). These mutant proteins lack the kinase domain. The alleles oz178, oz201, and n1924 contained missense mutations in the kinase domain. The oz178 and oz201 alleles contained the same mutation and change a highly conserved serine to asparagine (Fig 1B). The n1924 mutation changes a moderately conserved isoleucine to phenylalanine (Fig 1B). The dx84 mutation is a deletion that removes all of exon 11 and portions of introns 10 and 11 (Fig 1A). Exon 11 encodes residues 601–738; this region includes many highly conserved residues in the kinase domain. The n2520 and n2523 alleles contained the same nucleotide change, a missense mutation that changes the C-terminal 14-3-3 binding domain by converting a highly conserved serine to phenylalanine (Fig 1B).

lin-45 mutations cause larval lethality, defective vulval development, and sterility and can be arranged in an allelic series:
To determine how these mutations affect the activity of lin-45 and characterize the role of lin-45 during development, we analyzed the phenotypes of the mutants. The analysis of lin-45(sy96) by HAN et al. 1993 showed that a reduction of lin-45 activity causes partially penetrant larval lethality, sterility, and vulval defects and indicated that lin-45 is a positive regulator of Ras-mediated signaling. To carefully examine these processes, we placed a single egg on a petri dish and used a dissecting microscope to determine whether the animal died during larval development, whether the adult hermaphrodite laid eggs normally and appeared to have a normal vulva, and whether the adult hermaphrodite generated live progeny. If a mutation could be maintained as a homozygous strain, then eggs were derived from homozygous mutant hermaphrodites. If a mutation caused a fully penetrant sterile phenotype, then eggs were derived from heterozygous mutant hermaphrodites. n2520, n1924, and n1925 caused little or no lethality, gross vulval defects, or sterility (Table 1, lines 2–4). n2018 and n2506 caused significant larval lethality (76 and 86%), and many of the surviving adult hermaphrodites had an abnormal vulva (24 and 93%). However, these mutations caused only a low-penetrance sterile phenotype (1 and 3%; Table 1, lines 5 and 6). The mutations n2510, oz166, oz178, oz201, dx19, dx84, and dx89 caused similar defects; each mutation caused a partially penetrant larval lethal phenotype and fully penetrant vulvaless and sterile phenotypes (Table 1, lines 7–12, and data not shown). Thus, these mutations could not be propagated as homozygous mutants.

The finding that the mutations that cause completely penetrant Vul and sterile phenotypes cause only partially penetrant larval lethality might indicate maternal rescue of the larval lethality in homozygous mutants derived from heterozygous hermaphrodites. This type of maternal rescue has been previously documented for mek-2(null) and mpk-1(null) mutations (KORNFELD et al. 1995A ; LACKNER and KIM 1998 ). To investigate this possibility, we generated trans-heterozygotes with the oz166 allele, which caused a completely penetrant sterile phenotype, and n2520, n1924, n1925, and n2506 (Table 2). Trans-heterozygous hermaphrodites generated progeny with three different genotypes that were distinguished using marker mutations (Table 2 legend). Homozygous oz166 animals derived from trans-heterozygous hermaphrodites displayed highly penetrant or completely penetrant larval lethality (85–100%; Table 2, lines 3, 7, 10, and 15). These findings indicate that oz166 strongly affects the ability of lin-45 to function during embryonic and/or larval development and the survival of oz166/oz166 animals derived from an oz166/+ hermaphrodite is due to maternal rescue. Furthermore, these findings indicate that n2520, n1924, and n1925 affect the function of lin-45 during embryonic and/or larval development, since these mutations failed to complement the larval lethality caused by oz166.

On the basis of these findings, these alleles can be arranged in a series of increasing severity that is likely to correspond to an increasing loss of lin-45 activity; the series is the same whether larval lethality, vulval formation, or sterility are considered. n2520, n1924, and n1925 are weak alleles. These mutations do not cause highly penetrant defects in a wild-type genetic background. However, they affect vulval development, since they suppress the Muv defect caused by the let-60(gf) mutation, and they affect viability, since they fail to complement the larval lethality caused by lin-45(oz166). There is no evidence that they affect the function of lin-45 in the germline, since they complement the sterility caused by lin-45(oz166). However, these mutations may affect the germline in a subtle way that was not detected by scoring progeny production. n2018 and n2506 are intermediate-strength alleles. These mutations cause partially penetrant larval lethality and vulval defects. Like the weak alleles, these mutations do not cause significant sterility, and n2506 complemented the sterility caused by lin-45(oz166) (Table 2, line 11). The analysis of HAN et al. 1993 indicates that lin-45(sy96) is also intermediate in strength. n2510, oz166, oz178, oz201, dx19, dx84, and dx89 are strong alleles. These mutations caused completely penetrant sterile and Vul phenotypes and oz166 caused completely penetrant lethality when the maternal contribution of lin-45 was reduced.

The strong lin-45 mutations are probably null alleles:
Three types of evidence indicate that the strong mutations cause a very severe or complete loss of lin-45 activity. First, these seven mutations all cause a very similar phenotype, and they are the strongest in the allelic series. Second, the molecular analysis suggests that several of the mutant proteins have no activity; dx19 and n2510 encode truncated proteins that completely lack the kinase domain and dx84 encodes a protein lacking a large region of the kinase domain. To rigorously test this hypothesis, we compared oz166, one of the strong alleles, to nDf41, a deficiency allele that deletes the lin-45 locus as well as genes positioned right and left of lin-45 (Table 2). Comparing n2520/oz166 to n2520/nDf41 (Table 2, lines 2 and 4), n2506/oz166 to n2506/nDf41 (Table 2, lines 11, 12, 14, and 17), and oz166/oz166 to oz166/nDf41 (Table 1, line 12, and Table 2, line 18) reveals that the defects caused by nDf41 were similar to or less severe than the defects caused by oz166. These observations strongly support the hypothesis that oz166 and the other six strong alleles are null mutations. However, it is possible that oz166 retains some lin-45 function that was not measured in these assays. The finding that oz166 caused defects that were slightly more severe than those caused by nDf41 might indicate that oz166 causes some dominant negative effects or that nDf41 deletes a gene(s) that affects the Ras pathway in addition to lin-45. oz166 does not cause a strong dominant negative effect, since oz166/+ animals do not display noticeable defects (Table 1, line 13).

lin-45 raf is essential for viability, fertility, and the induction of vulval cell fates:
Because the previously characterized lin-45 (sy96) allele is a partial loss-of-function mutation, the analysis of this allele did not demonstrate whether lin-45 is essential for the processes that are affected in these mutants (HAN et al. 1993 ). We used the probable null alleles of lin-45 to address these questions. lin-45 appears to be essential for viability, since 100% of oz166 mutants derived from n2520/oz166 or n1925/oz166 hermaphrodites displayed larval lethality (Table 2, lines 3 and 10). lin-45 is likely to promote viability by regulating the differentiation of the excretory duct cell (YOCHEM et al. 1997 ). lin-45 appears to be essential for fertility, since 100% of homozygous mutants containing a strong mutation displayed sterility (Table 1). lin-45 is likely to promote fertility by regulating germ cell progression through the pachytene stage (CHURCH et al. 1995 ). One hundred percent of homozygous mutants containing a strong lin-45 allele displayed an absence of vulval formation when viewed with a dissecting microscope. To further characterize vulval development, we used Nomarski optics to examine the fate of the vulval precursor cells in L4 stage hermaphrodites. In oz166 mutants, P5.p, P6.p, and P7.p always generated two descendants that appeared to adopt the nonvulval 3° fate, resulting in a total of six descendants (Table 1, line 12). The same defect is caused by ablation of the anchor cell or severe loss-of-function mutations in mek-2 and mpk-1 (HORVITZ and STERNBERG 1991 ; KORNFELD et al. 1995A ; LACKNER and KIM 1998 ). These observations demonstrate that lin-45 is essential for vulval precursor cells to adopt a vulval fate.

lin-45 mutations exhibit complex interactions with a let-60(gf) mutation:
Six of the lin-45 mutations were identified as suppressors of the Muv phenotype caused by a let-60(gf) mutation. n1924, n1925, n2520, n2523, and n2506 can be maintained as homozygous strains and reduced the penetrance of the let-60(gf) Muv phenotype from 90 to 1% or less (Table 3). When the lin-45(n2506) mutation was separated from the let-60(gf) mutation and examined in a let-60(+) genetic background, several interesting features were apparent. Compared to the lin-45(n2506) let-60(gf) double mutants, the lin-45(n2506) single mutants displayed significantly more larval lethality (86 vs. 26%) and abnormal vulval development (93 vs. 4%) (Table 1, line 6, and Table 3, line 6). These observations suggest that the let-60(gf) mutation suppresses the larval lethal and vulvaless defects caused by lin-45(n2506) at the same time that lin-45(n2506) suppresses the multivulva defect caused by let-60(gf). The increased activity of the mutant Ras and the decreased activity of the mutant Raf appear to be balanced such that the double mutant is more like wild type than either single mutant. It is noteworthy that the lin-45(n2506) mutation affects a residue in the Ras-binding domain.

lin-45(n2510) was identified as a dominant suppressor of the let-60(gf) Muv phenotype: 28% of lin-45(n2510) let-60(gf)/+ let-60(gf) animals displayed the Muv phenotype (Table 3, line 8). lin-45(n2510) let-60(gf) homozygous mutants displayed highly penetrant larval lethality (91%) and surviving adults were sterile and vulvaless (Table 3, line 7). Interestingly, when lin-45(n2510) was separated from let-60(gf), the homozygous lin-45(n2510) mutants displayed a lower penetrance of 30% larval lethality (Table 1, line 9). These observations suggest that the let-60(gf) mutation enhances the larval lethality caused by lin-45(n2510). To determine whether this result is typical of strong lin-45 alleles, we generated a recombinant chromosome containing lin-45(oz166) and the let-60(gf) mutation. Whereas 68% of lin-45(oz166) animals displayed larval lethality (Table 1, line 12), 98% of lin-45(oz166) let-60(gf) animals displayed larval lethality (Table 3, line 9). Thus, the let-60(gf) mutation enhanced the larval lethality caused by two different strong lin-45 mutations, suggesting this is a general phenomenon.

It is surprising that the let-60(gf) mutation suppressed the larval lethality caused by lin-45(n2506) and enhanced the larval lethality caused by lin-45(n2510) and lin-45(oz166). One possible explanation is that the let-60(gf) mutation both positively and negatively affects the activity of the let-60 gene. The positive effect is evident in combination with lin-45(n2506), whereas the negative effect is evident in combination with lin-45(n2510) and lin-45(oz166).

Comparison of strong loss-of-function mutations of lin-45, mek-2, and mpk-1:
Strong loss-of-function or null mutations in the mek-2 MAP kinase kinase and mpk-1 ERK MAP kinase genes have been identified and characterized (CHURCH et al. 1995 ; KORNFELD et al. 1995A ; WU et al. 1995 ; LACKNER and KIM 1998 ). Like the lin-45(null) mutations, mek-2(null) and mpk-1(null) mutations cause a partially penetrant larval lethal phenotype and a fully penetrant sterile and vulvaless phenotype in homozygous mutants derived from heterozygous hermaphrodites. Interactions between mutations that cause a Muv phenotype and loss-of-function mutations of lin-45, mek-2, and mpk-1 have been analyzed previously (HAN et al. 1993 ; LACKNER et al. 1994 ; WU and HAN 1994 ; WU et al. 1995 ; KORNFELD et al. 1995A ; LACKNER and KIM 1998 ; TAN et al. 1998 ). However, these studies were not performed with a lin-45(null) mutation. To directly compare the role of these three genes, we analyzed vulval development in double mutants containing a mutation that causes a Muv phenotype and a strong loss-of-function mutation in lin-45, mek-2, or mpk-1.

lin-15 is a complex locus that encodes one class A and one class B synthetic multivulva gene; lin-15(n309) strongly reduces the activity of both genes and causes a fully penetrant, highly expressive Muv phenotype (CLARK et al. 1994 ). The Vul phenotype caused by strong lin-45(lf), mek-2(lf), and mpk-1(lf) mutations was fully epistatic to the lin-15(lf) Muv phenotype (Table 4, lines 1–7). Similarly, the Vul phenotype caused by these mutations was fully epistatic to the let-60(gf) Muv phenotype (Table 4, lines 8–14).

lin-1 encodes an ETS domain transcription factor that negatively regulates the 1° vulval cell fate, and loss-of-function mutations in lin-1 cause a Muv phenotype (BEITEL et al. 1995 ). The lin-1(lf) Muv phenotype was fully epistatic to the Vul phenotype caused by strong loss of function in lin-45, mek-2, and mpk-1 (Table 4, lines 15–18). lin-12 encodes a transmembrane receptor in the Notch family that promotes the 2° vulval cell fate, and it is likely that lin-12 is normally activated in P5.p and P7.p by ligand produced by the 1° vulval cell, P6.p (GREENWALD 1997 ). A gain-of-function mutation in lin-12 causes a highly penetrant Muv phenotype, since all six Pn.p cells adopt the 2° vulval fate. The lin-12 Muv phenotype was fully epistatic to the Vul phenotype caused by strong loss-of-function mutations in lin-45, mek-2, and mpk-1 (Table 4, lines 28–31).

The predicted LIN-31 protein contains a winged helix domain and is likely to function as a transcription factor (TAN et al. 1998 ; MILLER et al. 2000 ). A strong loss- of-function or null mutation of lin-31 causes a partially penetrant Muv phenotype and a partially penetrant Vul phenotype, indicating that lin-31 positively and negatively regulates vulval cell fates. The strong loss-of-function mutations lin-45(oz166) and lin-45(oz201) reduced the penetrance of the lin-31 Muv phenotype from 70% to 30 and 17%, respectively (Table 4, lines 19–23). By contrast, strong loss-of-function mek-2 and mpk-1 mutations did not reduce the penetrance of the lin-31 Muv phenotype significantly (Table 4, lines 24–27). This is the first genetic background we identified in which the lin-45(null) mutations caused a significantly different phenotype than the mek-2(null) and mpk-1(null) mutations. These findings suggest that the expression of the lin-31 Muv phenotype is partially dependent on lin-45 activity but is not dependent on mek-2 or mpk-1 activity. Therefore, lin-45 appears to have an activity that is not mediated by mek-2 and mpk-1. Furthermore, this finding suggests that the position of lin-31 in the signaling pathway may be complex, since it is not fully epistatic to lin-45.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

lin-45 is necessary for Ras-mediated signaling, and Raf, MEK, and ERK function in a predominantly linear signaling pathway:
We identified 13 mutations that reduce the activity of the lin-45 raf gene. These mutations can be arranged in an allelic series on the basis of the defects they caused in larval viability, fertility, and vulval development. Molecular and genetic analyses indicate that alleles that cause the most severe phenotype are likely to be null mutations. Previous analyses of the lin-45(sy96) mutation, which causes a partial loss of function, indicated that lin-45 is important for larval viability, fertility, and vulval development (HAN et al. 1993 ). However, these studies did not resolve whether lin-45 is essential for these processes, because the defects caused by lin-45(sy96) are partially penetrant. Our analysis of lin-45(null) alleles indicates that lin-45 is essential for larval viability, fertility, and the induction of vulval cell fates, since lin-45(null) mutations cause completely penetrant larval lethality, sterility, and a vulvaless phenotype in which P5.p–P7.p are transformed to the 3° cell fate.

The MAP kinase kinase kinase protein family consists of multiple proteins, including Raf. Similarly, the MAP kinase kinase and MAP kinase families consist of multiple proteins, including MEK and ERK, respectively (FERRELL 1996 ). The finding that each of these protein families consists of multiple members has raised the possibility that more than one MAP kinase kinase kinase might phosphorylate a single MAP kinase kinase or that more than one MAP kinase kinase might phosphorylate a single MAP kinase (GARDNER et al. 1994 ). If this were the case during C. elegans development, then a null mutation in Raf might cause a less severe phenotype than a null mutation in MEK, since another MAP kinase kinase kinase could substitute for Raf and phosphorylate MEK. Similarly, if multiple MAP kinase kinases regulate ERK, then a null mutation in MEK might be expected to cause a less severe phenotype than a null mutation in ERK. To investigate these possibilities, we compared the phenotypes caused by null mutations in lin-45 raf, mek-2 MEK, and mpk-1 ERK MAP kinase. A mutation in each gene caused a completely penetrant vulvaless defect in which P5.p–P7.p adopt the 3° nonvulval cell fate (KORNFELD et al. 1995A ; WU et al. 1995 ; LACKNER and KIM 1998 ). Mutations in each gene were completely epistatic to the Muv phenotype caused by a lin-15(lf) mutation and a let-60(gf) mutation, whereas the Muv phenotype caused by a lin-1(lf) and a lin-12(gf) mutation was completely epistatic to these Vul mutations. These results indicate that the loss of Raf activity is not less severe than the loss of MEK activity and that the loss of MEK activity is not less severe than the loss of ERK activity. Thus, our results support the model that Raf is the only physiological activator of MEK and that MEK is the only physiological activator of ERK during these processes.

We did identify one genetic background in which the lin-45(null) mutation caused a more severe phenotype than the mek-2(null) or mpk-1(null) mutation. lin-45(null) mutations partially suppressed the Muv phenotype caused by a lin-31(lf) mutation, whereas mek-2(null) and mpk-1(null) mutations did not have this effect. These results demonstrate that lin-45 activity is necessary for the full expression of the lin-31 Muv phenotype, whereas mek-2 and mpk-1 activities do not appear to be necessary. However, it is possible that residual maternal mek-2 and mpk-1 activity account for this difference. One model that can explain these findings is that lin-45 regulates a protein(s) in addition to MEK, and the regulation of this protein(s) is important for the lin-31 Muv phenotype. According to this model, lin-45 is a branchpoint in the signaling pathway. Taken together, our results indicate that lin-45, mek-2, and mpk-1 function in a predominantly linear signaling pathway and raise the possibility that a small part of lin-45 function is mediated by a protein(s) other than mek-2.

The Ras-binding domain, kinase domain, and 14-3-3-binding domain are necessary for Raf activity:
It is important to identify and characterize the functional domains of Raf. Comparisons of Raf proteins that have diverged during evolution have identified conserved domains that are likely to be functionally significant. The mechanism of action of these domains can be investigated using biochemical experiments, whereas the functional significance of the domains can be investigated using genetic analysis. The genetic analysis of vertebrate Raf can yield results that are difficult to interpret, since mutant Raf is typically overexpressed in immortalized cultured cells and these cells also contain endogenous, wild-type Raf. By contrast, the C. elegans system described here overcomes many of these limitations. The lin-45 mutations were present in both chromosomal copies, and thus mutant LIN-45 protein replaced wild-type LIN-45 in an otherwise wild-type animal. Furthermore, the use of random mutagenesis and screens for functional defects is a relatively unbiased way to identify functionally significant residues and domains. It is not completely unbiased, because chemical mutagens preferentially affect certain nucleotides and codons.

The interaction of Raf and Ras initiates Raf activation. This interaction has been characterized in a variety of binding assays (VOJTEK et al. 1993 ; ZHANG et al. 1993 ; FINNEY and HERRERA 1995 ); the most detailed information about the structural basis for the Raf-Ras interaction comes from X-ray crystallography (NASSAR et al. 1995 ). lin-45(n2506 R118W) affects a highly conserved arginine at the terminus of -helix 1 that directly contacts Ras by hydrogen bonding to aspartic acid 38 and serine 39 of Ras (Fig 2). The finding that this mutation causes a significant reduction of Raf activity in worms supports the model that this residue is important for Raf binding to Ras. Remarkably, a mutation of Drosophila Raf that changes the homologous arginine 217 to leucine was discovered as an intermediate loss-of-function allele, D-raf^C110 (MELNICK et al. 1993 ). Introduction of this substitution of arginine 89 of vertebrate Raf reduces binding to Ras (FABIAN et al. 1994 ; BLOCK et al. 1996 ). Thus, the importance of this arginine for Raf binding to Ras has been highly conserved. The lin-45(n2018 P92S) mutation, which causes a moderate loss of function, and the lin-45(n1925 R108W) mutation, which causes a weak loss of function, affect residues in the Ras-binding domain that are not predicted to interact directly with Ras according to the crystal structure (Fig 2). Proline 92 is between ß1-sheet and ß2-sheet; both ß-sheets contain residues predicted to directly contact Ras (NASSAR et al. 1995 ). Arginine 108 is in -helix 1 but positioned away from Ras. It is likely that the substitution of these residues disturbs the conformation of the Ras-binding domain and reduces the affinity of Raf and Ras.

View larger version (38K): In this window In a new window Download PPT slide   Figure 2. Structure of the Ras-binding domain of Raf-1 complexed with Rap1A showing the position of the lin-45 missense substitutions. Structure of the Ras-binding domain of human c-Raf1 (RBD, Ser51–Leu131) bound to truncated Rap1A complexed with GTP determined by X-ray crystallography (NASSAR et al. 1995 ). Rap1A is a small GTP-binding protein with an effector region identical to Ras. ß-sheets are yellow, -helixes are blue, and other regions are orange. The side chains are shown for amino acids of human Raf-1 that are homologous to the residues of LIN-45 affected by missense mutations n2018 (P92 LIN-45/P63 Raf-1), n1925 (R108 LIN-45/H79 Raf-1), and n2506 (R118 LIN-45/R89 Raf-1). The side chains of Rap1A amino acids Asp38 and Ser39 are shown, since these directly contact Arg 89 by hydrogen bonds.

The lin-45(oz178 S645N) mutation, which causes a strong loss of function, and the lin-45(n1924 I726F) mutation, which causes a weak loss of function, affect residues in the protein kinase domain. Serine 645 is a highly conserved residue, and our findings support the model that the substitution of this residue strongly reduces the kinase activity and kinase activity is essential for Raf function. Isoleucine 726 is moderately conserved—human Raf has valine at this position. Our findings support the model that this residue plays a secondary role in the function of the kinase domain. Mutations that affect highly conserved residues of the kinase domain of Drosophila Raf also cause a strong loss of function (MELNICK et al. 1993 ).

14-3-3 proteins can bind to several phosphoserine motifs, including the RSXpSXP motif (MUSLIN et al. 1996 ). Raf contains two evolutionarily conserved RSXSXP motifs: an N-terminal motif in CR2 and a C-terminal motif following the kinase domain. Mutational analysis of vertebrate Raf indicates that the N-terminal 14-3-3 binding site negatively regulates Raf activity (MICHAUD et al. 1995 ; ROMMEL et al. 1996 , ROMMEL et al. 1997 ; ROY et al. 1997 ; MCPHERSON et al. 1999 ). 14-3-3 binding to the C-terminal site has been suggested to be important for Raf function, since mutations in this motif reduce the function of Raf-1 expressed in cultured cells and the kinase activity of purified Raf-1 (THORSON et al. 1998 ; TZIVION et al. 1998 ). The lin-45(n2520 S754F) mutation changes the conserved 14-3-3 binding site from RSXSXP to RFXSXP and partially reduces the activity of lin-45 in the worm. These findings indicate that this serine is important for 14-3-3 binding and/or for the phosphorylation of this motif by a protein kinase. Consistent with the possibility that this mutation reduces 14-3-3 binding, phosphorylated peptides with substitutions of alanine for this serine display reduced binding to 14-3-3 (MUSLIN et al. 1996 ). Our findings indicate that 14-3-3 binding to this position of Raf is necessary for full Raf activity and support the model that 14-3-3 binding to the C terminus positively regulates Raf.

Our findings have an additional implication for the role of 14-3-3 proteins in Ras-mediated signaling in C. elegans. C. elegans has multiple genes encoding 14-3-3 proteins (WANG and SHAKES 1997 ). If 14-3-3 binding to Raf promotes Raf activity, then a mutant that lacks 14-3-3 protein is likely to have reduced Raf activity and reduced signal transduction. Mutations in C. elegans genes encoding 14-3-3 proteins have not been reported to display defects in Ras-mediated signaling. However, some genes have not been analyzed genetically. Genetic analysis in yeast and Drosophila has demonstrated that 14-3-3 proteins play a positive role in Ras-ERK signaling (CHANG and RUBIN 1997 ; KOCKEL et al. 1997 ; LI et al. 1997 ; ROBERTS et al. 1997 ). Our findings suggest that 14-3-3 proteins play a positive role in Ras-mediated signaling in C. elegans by promoting the activity of Raf.

	ACKNOWLEDGMENTS

We thank Greg Beitel and Scott Clark for the identification and initial characterization of n1924, n1925, and n2018. These and several other lin-45 alleles were identified in the laboratory of Bob Horvitz, and we are grateful for his support during the initiation of this project. We thank Dil Nawaz Kapadia for the identification of dx19, Andy Golden for providing the genomic DNA sequence of the lin-45 locus, and Weiyang Shi, Stacie Foglesong, and Blake Coblenz for assistance scoring lin-45 mutants. This work was supported by grants from the National Institutes of Health to K.K. (CA-84271), T.S. (GM-63310), and E.J.L. (GM-49785). K.K. is a recipient of a Burroughs Wellcome Foundation New Investigator Award in the Pharmacological Sciences and a Leukemia and Lymphoma Society Scholar Award.

Manuscript received September 10, 2001; Accepted for publication December 5, 2001.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ALESSI, D. R., Y. SAITO, D. G. CAMPBELL, P. COHEN, and G. SITHANANDAM et al., 1994  Identification of the sites in MAP kinase kinase-1 phosphorylated by p74raf-1. EMBO J. 13:1610-1619[Medline].

BEITEL, G. J., S. G. CLARK, and H. R. HORVITZ, 1990  Caenorhabditis elegans ras gene let-60 acts as a switch in the pathway of vulval induction. Nature 348:503-509[Medline].

BEITEL, G. J., S. TUCK, I. GREENWALD, and H. R. HORVITZ, 1995  The Caenorhabditis elegans gene lin-1 encodes an ETS-domain protein and defines a branch of the vulval induction pathway. Genes Dev. 9:3149-3162[Abstract/Free Full Text].

BLOCK, C., R. JANKNECHT, C. HERRMANN, N. NASSAR, and A. WITTINGHOFER, 1996  Quantitative structure-activity analysis correlating Ras/Raf interaction in vitro to Raf activation in vivo. Nat. Struct. Biol. 3:244-251[Medline].

BONNER, T. I., H. OPPERMANN, P. SEEBURG, S. B. KERBY, and M. A. GUNNELL et al., 1986  The complete coding sequence of the human raf oncogene and the corresponding structure of the c-raf-1 gene. Nucleic Acids Res. 14:1009-1015[Abstract/Free Full Text].

BRENNER, S., 1974  The genetics of Caenorhabditis elegans.. Genetics 77:71-94[Abstract/Free Full Text].

CANTLEY, L. C., K. R. AUGER, C. CARPENTER, B. DUCKWORTH, and A. GRAZIANI et al., 1991  Oncogenes and signal transduction. Cell 64:281-302[Medline].

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

CHURCH, D. L., K.-L. GUAN, and E. J. LAMBIE, 1995  Three genes of the MAP kinase cascade, mek-2, mpk-1/sur-1 and let-60 ras, are required for meiotic cell cycle progression in Caenorhabditis elegans.. Development 121:2525-2535[Abstract].

CLARK, S. G., M. J. STERN, and H. R. HORVITZ, 1992  C. elegans cell-signaling gene sem-5 encodes a protein with SH2 and SH3 domains. Nature 356:340-344[Medline].

CLARK, S. G., X. LU, and H. R. HORVITZ, 1994  The Caenorhabditis elegans locus lin-15, a negative regulator of a tyrosine kinase signaling pathway, encodes two different proteins. Genetics 137:987-999[Abstract].

COULONDRE, C. and J. H. MILLER, 1977  Genetic studies of the lac repressor. IV. Mutagenic specificity in the LacI gene of Escherichia coli. J. Mol. Biol. 117:577-606[Medline].

DENBURG, A. F., K. MCDONALD, G. MOULDER, R. BARSTEAD, and M. DRESSER et al., 1998  Meiotic recombination in C. elegans initiates by a conserved mechanism and is dispensable for homologous chromosome synapsis. Cell 94:387-398[Medline].

DICKSON, B. and E. HAFEN, 1994  Genetics of signal transduction in invertebrates. Curr. Opin. Genet. Dev. 4:64-70[Medline].

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].

FERRELL, J. E., 1996  MAP kinases in mitogenesis and development. Curr. Top. Dev. Biol. 33:1-60[Medline].

FINNEY, R. and D. HERRERA, 1995  Ras-Raf complexes: analyses of complexes formed in vivo. Methods Enzymol. 255:310-323[Medline].

GARDNER, A. M., R. R. VAILLANCOURT, C. A. LANGE-CARTER, and G. L. JOHNSON, 1994  MEK-1 phosphorylation by MEK kinase, Raf, and mitogen-activated protein kinase: analysis of phosphopeptides and regulation of activity. Mol. Biol. Cell 5:193-201[Abstract].

GREENWALD, I., 1997 Development of the vulva, pp. 519–541 in C. elegans II, edited by D. L. RIDDLE, T. BLUMENTHAL, B. J. MEYER and J. R. PRIESS. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

GREENWALD, I. and G. SEYDOUX, 1990  Analysis of gain-of-function mutations of the lin-12 gene of Caenorhabditis elegans.. Nature 346:197-199[Medline].

HAN, M., A. GOLDEN, Y. HAN, and P. W. STERNBERG, 1993  C. elegans lin-45 raf gene participates in let-60 ras-stimulated vulval differentiation. Nature 363:133-140[Medline].

HEIDECKER, G., M. HULEIHEL, J. L. CLEVELAND, W. KOLCH, and T. W. BECK et al., 1990  Mutational activation of c-raf-1 and definition of the minimal transforming sequence. Mol. Cell. Biol. 10:2503-2512[Abstract/Free Full Text].

HORVITZ, H. R. and P. W. STERNBERG, 1991  Multiple intercellular signalling systems control the development of the Caenorhabditis elegans vulva. Nature 351:535-541[Medline].

JACOBS, D., G. J. BEITEL, S. G. CLARK, H. R. HORVITZ, and K. KORNFELD, 1998  Gain-of-function mutations in the Caenorhabditis elegans lin-1 ETS gene identify a C-terminal regulatory domain phosphorylated by ERK MAP kinase. Genetics 149:1809-1822[Abstract/Free Full Text].

JAKUBOWSKI, J. and K. KORNFELD, 1999  A local, high-density, single-nucleotide polymorphism map used to clone Caenorhabditis elegans cdf-1.. Genetics 158:743-752.

KOCKEL, L., G. VORBRUGGEN, H. JACKLE, M. MLODZIK, and D. BOHMANN, 1997  Requirement for Drosophila 14–3-3 in Raf-dependent photoreceptor development. Genes Dev. 11:1140-1147[Abstract/Free Full Text].

KORNFELD, K., 1997  Vulval development in Caenorhabditis elegans.. Trends Genet. 13:55-61[Medline].

KORNFELD, K., K.-L. GUAN, and H. R. HORVITZ, 1995a  The Caenorhabditis elegans gene mek-2 is required for vulval induction and encodes a protein similar to the protein kinase MEK. Genes Dev. 9:756-768[Abstract/Free Full Text].

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

LACKNER, M. R. and S. K. KIM, 1998  Genetic analysis of the Caenorhabditis elegans MAP kinase gene mpk-1.. Genetics 150:103-117[Abstract/Free Full Text].

LACKNER, M. R., K. KORNFELD, L. M. MILLER, H. R. HORVITZ, and S. K. KIM, 1994  A MAP kinase homolog, mpk-1, is involved in ras-mediated induction of vulval cell fates in Caenorhabditis elegans.. Genes Dev. 8:160-173[Abstract/Free Full Text].

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].

MARAIS, R., Y. LIGHT, H. F. PATERSON, and C. J. MARSHALL, 1995  Ras recruits Raf-1 to the plasma membrane for activation by tyrosine phosphorylation. EMBO J. 14:3136-3145[Medline].

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

MCPHERSON, R. A., A. HARDING, S. ROY, A. LANE, and J. F. HANCOCK, 1999  Interactions of c-Raf-1 with phosphatidylserine and 14–33. Oncogene 18:3862-3869[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 protein kinase. Development 118:127-138[Abstract].

MICHAUD, N. R., J. R. FABIAN, K. D. MATHES, and D. K. MORRSION, 1995  14–3-3 is not essential for Raf-1 function: identification of Raf-1 proteins that are biologically activated in a 14–3-3 and Ras-independent manner. Mol. Cell. Biol. 15:3390-3397[Abstract].

MILLER, L. M., H. A. HESS, D. B. DOROQUEZ, and N. M. ANDREWS, 2000  Null mutations in the lin-31 gene indicate two functions during Caenorhabditis elegans vulval development. Genetics 156:1595-1602[Abstract/Free Full Text].

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

MUSLIN, A. J., J. W. TANNER, P. M. ALLEN, and A. S. SHAW, 1996  Interaction of 14–3-3 with signaling proteins is mediated by the recognition of phosphoserine. Cell 84:889-897[Medline].

NASSAR, N., G. HORN, C. HERRMANN, A. SCHERER, and F. MCCORMICK et al., 1995  The 2.2 A crystal structure of the Ras-binding domain of the serine/threonine kinase c-Raf1 in complex with Rap1A and a GTP analogue. Nature 375:554-560[Medline].

NISHIDA, Y., M. HATA, T. AYAKI, H. RYO, and M. YAMAGATA et al., 1988  Proliferation of both somatic and germ cells is affected in the Drosophila mutants of raf proto-oncogene. EMBO J. 7:775-781[Medline].

RIDDLE, D. L., T. BLUMENTHAL, B. J. MEYER and J. R. PRIESS (Editors), 1997 C. elegans II. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

ROBERTS, R. L., H.-U. MOSCH, and G. R. FINK, 1997  14–3-3 proteins are essential for RAS/MAPK cascade signaling during pseudohyphal development in S. cerevisiae. Cell 89:1055-1065[Medline].

ROMMEL, C., G. RADZIWILL, J. LOVRIC, J. NOELDEKE, and T. HEINICKE et al., 1996  Activated Ras displaces 14–3-3 protein from the amino terminus of c-Raf-1. Oncogene 12:609-619[Medline].

ROMMEL, C., G. RADZIWILL, K. MOELLING, and E. HAFEN, 1997  Negative regulation of Raf activity by binding of 14–3-3 to the amino terminus of Raf in vivo. Mech. Dev. 64:95-104[Medline].

ROY, S., A. LANE, J. YAN, R. MCPHERSON, and J. F. HANCOCK, 1997  Activity of plasma membrane-recruited Raf-1 is regulated by Ras via the Raf zinc finger. J. Biol. Chem. 272:20139-20145[Abstract/Free Full Text].

STANTON, V. P., JR., D. W. NICHOLS, A. P. LAUDANO, and G. M. COOPER, 1989  Definition of the human raf amino-terminal regulatory region by deletion mutagenesis. Mol. Cell. Biol. 9:639-647[Abstract/Free Full Text].

STERNBERG, P. W. and M. HAN, 1998  Genetics of RAS s