Modulation of KSR activity in Caenorhabditis elegans by Zn ions, PAR-1 kinase and PP2A phosphatase
2003; Springer Nature; Volume: 23; Issue: 1 Linguagem: Inglês
10.1038/sj.emboj.7600025
ISSN1460-2075
AutoresJohn H. Yoder, Huira Chong, Kun‐Liang Guan, Min Han,
Tópico(s)Selenium in Biological Systems
ResumoArticle11 December 2003free access Modulation of KSR activity in Caenorhabditis elegans by Zn ions, PAR-1 kinase and PP2A phosphatase John H Yoder John H Yoder Department of Molecular, Cellular and Developmental Biology, Howard Hughes Medical Institution, University of Colorado, Boulder, CO, USA Present address: Howard Hughes Medical Institute, Laboratory of Molecular Biology, University of Wisconsin, Madison, WI 53706 USA Search for more papers by this author Huira Chong Huira Chong Department of Biological Chemistry and The Institute of Gerontology, University of Michigan, Ann Arbor, MI, USA Search for more papers by this author Kun-liang Guan Kun-liang Guan Department of Biological Chemistry and The Institute of Gerontology, University of Michigan, Ann Arbor, MI, USA Search for more papers by this author Min Han Corresponding Author Min Han Department of Molecular, Cellular and Developmental Biology, Howard Hughes Medical Institution, University of Colorado, Boulder, CO, USA Search for more papers by this author John H Yoder John H Yoder Department of Molecular, Cellular and Developmental Biology, Howard Hughes Medical Institution, University of Colorado, Boulder, CO, USA Present address: Howard Hughes Medical Institute, Laboratory of Molecular Biology, University of Wisconsin, Madison, WI 53706 USA Search for more papers by this author Huira Chong Huira Chong Department of Biological Chemistry and The Institute of Gerontology, University of Michigan, Ann Arbor, MI, USA Search for more papers by this author Kun-liang Guan Kun-liang Guan Department of Biological Chemistry and The Institute of Gerontology, University of Michigan, Ann Arbor, MI, USA Search for more papers by this author Min Han Corresponding Author Min Han Department of Molecular, Cellular and Developmental Biology, Howard Hughes Medical Institution, University of Colorado, Boulder, CO, USA Search for more papers by this author Author Information John H Yoder1,3, Huira Chong2, Kun-liang Guan2 and Min Han 1 1Department of Molecular, Cellular and Developmental Biology, Howard Hughes Medical Institution, University of Colorado, Boulder, CO, USA 2Department of Biological Chemistry and The Institute of Gerontology, University of Michigan, Ann Arbor, MI, USA 3Present address: Howard Hughes Medical Institute, Laboratory of Molecular Biology, University of Wisconsin, Madison, WI 53706 USA *Corresponding author. Department of Molecular, Cellular and Developmental Biology, Howard Hughes Medical Institution, University of Colorado, Boulder, CO 80309, USA. Tel.: +1 303 735 0375; Fax: +1 303 735 0175; E-mail: [email protected] The EMBO Journal (2004)23:111-119https://doi.org/10.1038/sj.emboj.7600025 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Vulval differentiation in Caenorhabditis elegans is controlled by a conserved signal transduction pathway mediated by Ras and a kinase cascade that includes Raf, Mek and MAPK. Activation of this cascade is positively regulated by a number of proteins such as KSR (kinase suppressor of Ras), SUR-8/SOC-2, SUR-6/PP2A-B and CDF-1. We describe the functional characterization of sur-7 and several genes that regulate signaling downstream of ras. We identified sur-7 by isolating a mutation that suppresses an activated ras allele, and showed that SUR-7 is a divergent member of the cation diffusion facilitator family of heavy metal ion transporters that is probably localized to the endoplosmic recticulum membrane and regulates cellular Zn2+ concentrations. Genetic double mutant analyses suggest that the SUR-7-mediated effect is not a general toxic response. Instead, Zn2+ ions target a specific step of the pathway, probably regulation of the scaffolding protein KSR. Biochemical analysis in mammalian cells indicates that high Zn2+ concentration causes a dramatic increase of KSR phosphorylation. Genetic analysis also indicates that PP2A phosphatase and PAR-1 kinase act downstream of Raf to positively and negatively regulate KSR activity, respectively. Introduction The small GTPase Ras is an important regulator of eukaryotic cell proliferation and differentiation. Genetic and biochemical studies have defined a well-characterized linear pathway in which Ras functions as a molecular switch for transducing extracellular signals to the nucleus via activation of a canonical mitogen-activated protein kinase (MAPK) cascade (reviewed in Vojtek and Der, 1998). Continued genetic and biochemical studies argue that multiple regulatory inputs converge on the RTK/Ras/MAP kinase cascade to ensure appropriate levels, timing and location of the signaling activity mediated by this pathway. Genetic modifier screens in both Drosophila and Caenorhabditis elegans have identified several key players in the central linear pathway. Such screens also identified both positive and negative regulators of Ras-mediated signaling, including a number of factors that appear to regulate Raf and MEK activities. For example, KSR (kinase suppressor of Ras) was identified as a positive regulator of Ras signaling in both C. elegans and Drosophila suppressor screens (Kornfeld et al, 1995; Sundaram and Han, 1995; Therrien et al, 1995). Subsequent analysis suggested that kinase activity is not required for KSR function and that it may serve as a scaffold for Raf, MEK and other proteins (Morrison, 2001). The positive regulators sur-6 and sur-8 (suppressor of ras mutations) were identified in C. elegans (Sieburth et al, 1998, 1999). SUR-6 is a regulatory B/PR55 subunit of protein phosphatase 2A while SUR-8 is a novel but conserved leucine-rich repeat protein that complexes with Ras and Raf to enhance Raf activation (Li et al, 2000). Additionally, a C. elegans cation diffusion facilitator (CDF) family member, cdf-1, was identified as a positive regulator of Ras signaling (Bruinsma et al, 2002). Presumably, cdf-1(lf) results in increased cytosolic zinc concentrations, negatively affecting the pathway downstream of let-60/Ras and upstream of mek-2. There are some conflicting genetic and biochemical data however for some positive regulators of Ras signaling. Biochemical analyses indicated that KSR acts as a scaffold protein that facilitates the activation of MEK by Raf (reviewed by Morrison, 2001; Raabe and Rapp, 2002). However, previous genetic analyses in Drosophila and C. elegans suggested that KSR functions upstream of Raf in concert with SUR-6 (Therrien et al, 1995; Sieburth et al, 1999). Here we characterize the role of a divergent CDF family member, sur-7, as a positive regulator of Ras signaling. Our analyses of interactions between sur-7 and other regulators clarify the discrepancies between genetic and biochemical studies of KSR and sur-6, and more precisely address where within the pathway intracellular zinc may exert its effect. Results sur-7(ku119) suppresses an activated ras allele Previous studies have shown that the C. elegans let-60/Ras gene plays critical roles in a number of developmental processes (reviewed by Sternberg and Han, 1998). The best understood role for let-60 is its requirement for appropriate induction of a subset of hypodermal cells to generate the vulva, the hermaphrodite egg-laying and copulatory structure (reviewed in Kornfeld, 1997; Sternberg and Han, 1998). Reduction of let-60 activity by mutations leads to a failure in vulva precursor cell (VPC) induction, while gain-of-function (gf) alleles result in VPC overinduction causing a multivulva (Muv) phenotype characterized by numerous ventral protrusions along the length of the animal. To identify genes acting downstream of let-60/Ras, screens were previously carried out in our laboratory and others to isolate alleles that suppress this Muv phenotype (e.g. Wu and Han, 1994; Kornfeld et al, 1995; Sieburth et al, 1998). One allele, ku119, was identified and mapped genetically to the right arm of LG X. In wild-type animals, three of six VPCs are induced to a vulva fate. let-60(gf) mutant animals are 76% Muv and on average 4.6 VPCs are induced. When ku119 is homozygous in a let-60(n1046) background, the Muv percentage and induction are reduced to nearly wild-type levels (Table I). In an otherwise wild-type background however, ku119 exhibits no obvious vulva underinduction defects. The locus defined by ku119 was named sur-7. Table 1. Double and triple mutant analyses of sur-7 and other genes for their roles in vulval induction Genotype %M UV (n) Average #VPC induced (n) let-60(n1046) 76 (128) 4.62 (27) let-60(n1046);sur-7(ku119) 2 (634) 3.03 (38) let-60(kuIs12) 86 (22) 4.68 (22) let-60(kuIs12);sur-7(ku119) 17 (35) 3.21(35) lin-15(n765) 91 (47) 5.37 (51) lin-15(n765) sur-7(ku119) 41 (49) 3.36 (37) lin-31(n301) 90 (49) 3.81(20) lin-31(n301);sur-7(ku119) 95 (65) 3.96 (20) lin-1(ar147) 100 (>100) 5.91 (21) lin-1(ar147);sur-7(ku119) 100 (>100) 5.85 (27) Hsp-dRaf(gf) 35 (60) 3.84 (30) Hsp-dRaf(gf);sur-7(ku119) 35 (62) 3.63 (28) Hsp-dRaf(gf)a 47 (30) 3.57 (30) Hsp-dRaf(gf);sur-6(ku123)a 47 (17) 3.75 (17) Hsp-dRaf(gf);sur-8(ku167)b ND 3.75 (28) Hsp-dRaf(gf);ksr-1(ku68)a 46 (28) 3.51(28) Hsp-dRaf(gf);mpk-1(ku1)b ND 2.97 (28) lin-45(gf) 77 (334) 4.44 (20) lin-45(gf);sur-7(ku119) 2 (435) 3.06 (20) lin-45(gf);sur-8(ku167) 70 (246) 3.81 (25) Sur-6(ku123) lin-45(gf) 0 (334) 3.00 (23) lin-45(gf);ksr-1(ku68) 1 (98) 3.00 (25) Mek-2(ku114);lin-45(gf) 0 (104) 3.00 (25) Hsp-mpk-1(gf) hsp-Dmek(gf) 97 (240) ND Hsp-mpk-1(gf) hsp-Dmek(gf);sur-7(ku119) 95 (273) ND Hsp-mpk-1(gf) hsp-Dmek(gf);sur-2(lf)c 0 (25) ND Rol-4(sc8); let-60(n1046) 86 (217) ND Par-1(b274);let-60(n1046) 100 (112) ND Sur-6(ku123);let-60(n1046) 6 (240) ND Sur-6(ku123);par-1(b274);let-60(gf) 65 (48) ND let-60(n1046);par-1(b274);sur-7(ku119) 94 (54) ND Sur-8(ku167) let-60(n1046) 4 (333) ND Sur-8(ku167) let-60(n1046);par-1(b274) 65 (111) ND let-60(n1046);ksr-1(ku68) 0 (>100) ND let-60(n1046);par-1(b274);ksr-1(ku68) 13 (23) ND Mpk-1(ku1);let-60(n1046) 0 (>100) ND Mpk-1(ku1);let-60(n1046);par-1(b274) 2 (41) ND %Muv, percentage of worms with a Multivulva phenotype examined under a dissecting scope. #VPC induced, number of vulval precursor cells induced to vulval cells examined under Nomarski optics on a compound microscope. (n), number of animals examined. ND, not determined. kuIs12 is an integrated array carrying multiple copies of wild-type let-60. For lin-15(n765) experiments, animals were grown at 18.2°C. hsp-dRaf(gf) is described in Materials and methods and Sieburth et al, (1999). lin-45(gf) is described in the text and Materials and methods. The hsp-mek/mpk-1(gf) strain has an integrated array containing gain-of-function mutant genes of both mpk-1 and d-mek (Lackner and Kim, 1998) (see Materials and methods for assay conditions. par-1 was marked with rol-4 in all lines tested. sur-8, sur-6 and ksr-1 were marked with dpy-20, unc-29 and lon-2, respectively. sur-7(ku119) and mek-2(ku114) homozygotes were identified by sequence analysis. Induction data were not collected for par-1 experiments in order to monitor embryonic phenotype to score for homozygous par-1(lf)). a Sieburth et al (1999). b Sieburth and Han (1998). c Lackner and Kim (1996). To determine if ku119 is a loss-of-function mutation, we performed RNAi against sur-7 in wild-type, let-60(n1046) and let-60(n1046);sur-7(ku119) animals (Materials and methods). RNAi against sur-7 caused no overt phenotype in wild-type or let-60(n1046);sur-7(ku119) animals and strongly suppressed the Muv phenotype to 3.2% (n=947) in let-60(n1046). As there were no obvious phenotypic differences between sur-7(ku119) and sur-7(RNAi), we conclude that sur-7(ku119) is a strong lf allele. SUR-7 cloning and classification as a CDF family member To identify the sur-7 locus, ku119 was mapped using genetic markers and single-nucleotide polymorphisms (SNPs) to an LGX map position of approximately 23.9 between the SNPs pk6169 and F38E9.1 (Materials and methods). One cosmid in this interval (F01G12) and a subclone containing the single open reading frame F01G12.2 conferred robust rescuing activity to sur-7(ku119). The predicted F01G12.2 ORF was sequenced and a G to A transition was identified at position +5 in the 5′ splice site in the third of six introns (Figure 1). We performed RT–PCR to confirm the gene structure and that the ku119 mutation results in intron 3 splicing failure. Sequence analysis of wild-type RT–PCR product confirmed that the predicted F01G12.2 gene structure is correct. Using an exon three specific primer, we identified two RT–PCR products in ku119 animals, one containing the third intron, which would result in premature translational stop, while the second corresponded to the predicted wild-type mRNA. Thus, sur-7(ku119) may not be a complete lf allele. Figure 1.SUR-7 is a cation diffusion facilitator protein. (A) Structure of sur-7 and lesion identified in sur-7(ku119). A G to A transition was identified in the 5′ splice site of the third intron. (B) Comparison between SUR-7 and a consensus for the CDF family of proteins. SUR-7 contains six transmembrane domains (shaded regions). Five of six domains align physically with those in other CDF proteins. Most CDF proteins have a histidine-rich loop with as many as 13 histidine residues between TMs IV and V. SUR-7 has a short H-rich region between TMs III and IV. (C) Alignment between SUR-7, CDF-1 and S. pombe Zhf-1. SUR-7 shares the greatest degree of identity with other CDF proteins from TM VI to its C-terminus. In this region, SUR-7 is 16% identical and 36% similar to CDF-1, and 17% identical and 32% similar to Zhf1. CDF-1 and Zhf1 are 24% identical and 44% similar over the same region. Black backgrounds represent identity and outlines represent similarity. Download figure Download PowerPoint An NCBI BLAST search using the SUR-7 protein sequence produced no significant homology. However, motif analysis using the PROSITE database suggested similarity to the CDF family of proteins (Falquet et al, 2002). This similarity was primarily recognized due to the presence of six transmembrane domains, a feature common to most CDF proteins. CDF proteins are a class of ubiquitous metal transporters that facilitate metal ion traffic across the plasma membrane and the membranes of organelles. Another C. elegans CDF family member, cdf-1, was recently identified as a positive regulator of Ras signaling (Bruinsma et al, 2002). cdf-1(lf) alleles similarly suppress let-60(n1046) Muv to nearly wild-type levels. CDF proteins are found in both prokaryotes, and eukaryotes and vary greatly in size and sequence. In general, CDF proteins possess six transmembrane domains and a histidine-rich loop between membrane spanners IV and V (Palmiter and Findley, 1995; Paulsen and Saier, 1997). This H-rich loop, thought to be important for metal ion binding, and the termini of the protein presumably face cytoplasmically. There are several examples of deviation from these family characteristics. Yeast Msc2 and human hZTL1 have 12 membrane spanners (Li and Kaplan, 2001; Cragg et al, 2002). Additionally, the termini of hZTL1 are predicted to face extracellularly and hZnt6 has no H-rich loop (Huang et al, 2002). Therefore, this is a loosely defined family of proteins, which share the one specific trait of conferring either resistance or hypersensitivity to various heavy metals such as Zn2+, Cd2+, Cu2+, Co2+ and Ni2+ when their expression is altered (Gaither and Eide, 2001). SUR-7 has no H-rich loop between membrane spanners IV and V. Instead there is a cluster of five histidine residues between spanners III and IV (Figure 1). We compared the sequences of five representative CDF proteins (yeast cot1, yeast zhf1, human Znt1, human Znt5 and C. elegans CDF-1) and found the greatest sequence similarities in their C-termini and six transmembrane domains. SUR-7 displays the greatest sequence similarity to these proteins from membrane spanner VI through the C-terminus, with sequence identities ranging from 16 to 23% (Figure 1). While SUR-7 lacks similarity to other CDF family members in the remainder of its sequence, the N-terminal half of SUR-7 and the other proteins primarily define the other five transmembrane domains. When aligned against the other five CDF proteins, all but one of the predicted SUR-7 membrane spanners aligned precisely with those of the other protein family members. sur-7(ku119) animals are sensitive to high Zn2+ concentrations Like several human and yeast CDF proteins, mutations in C. elegans cdf1 caused hypersensitivity toward increased Zn2+ concentrations (Palmiter and Findley, 1995; Bruinsma et al, 2002). To investigate a similar response by sur-7(ku119) animals, we performed a Zn2+ sensitivity assay (Bruinsma et al, 2002; Materials and methods). Expectedly, sur-7(ku119) animals were more sensitive to increasing concentrations of Zn2+ than wild type (Figure 2). The progeny of animals placed on metal-supplemented growth media showed increasing sensitivity to Zn2+ concentrations at and above 0.25 mM as assayed by their rate of developmental maturation. These differences were found to be significant at and above 1.0 mM Zn2+ (student's t-test: α 0.05). However, sur-7(ku119) animals were more tolerant to increased Zn2+ than cdf-1(n2527) controls. The difference in tolerance between cdf-1(n2527) and wild type as well as sur-7(ku119) was significant at 0.25 and 0.05 mM Zn2+ (Tukey–Kramer HSD: α 0.05); however, Zn2+ tolerance was not statistically different between sur-7(ku119) and cdf-1(n2527) at and above 1.0 mM Zn2+. While these observations may reflect functional differences between the two proteins, they suggest that like CDF-1, SUR-7 functions as a heavy metal ion transporter, and specifically, but perhaps not exclusively, mediates cytoplasmic efflux of Zn2+ ions. As further evidence for similar functions of SUR-7 and CDF-1, we found that alleles in both genes conferred statistically significant (student's t-test: α 0.05) increased tolerance—rather than heightened sensitivity as was the case for Zn2+—to Cu2+ (at 0.5 mM but not above) compared to wild-type controls, but showed little or no difference in sensitivity for other heavy metal ions tested (Co2+ and Cd2+) (Materials and methods; data not shown). CDF proteins are not thought to function as copper transporters; therefore, the increased resistance to Cu2+ levels exhibited by both sur-7(ku119) and cdf-1(n2527) may reflect physiological conditions resulting from excess Zn2+ accumulation. Nonetheless, these results support the view that SUR-7 and CDF-1 similarly affect heavy metal ion homeostasis in C. elegans. Figure 2.cdf-1(lf) and sur-7(lf) mutants display heightened toxicity toward increasing concentrations of Zn2+. Animals grown on metal supplemented media were scored for percentage reaching egg-laying maturity after 4 days (Materials and methods). Each bar represents the mean of three independent experiments and shows the calculated standard deviation. sur-7(ku119) and cdf-1(n2527) responses were found to be significantly different from wild type (but not one another) at and above 1.0 mM Zn2+ (Tukey–Kramer: α 0.05). Download figure Download PowerPoint SUR-7 is likely localized to the endoplosmic recticulum membrane We next generated a translational green fluorescent protein (GFP) fusion construct to determine the expression pattern and subcellular distribution of SUR-7 (Materials and methods). The sur-7∷GFP construct is functional since the expression of this construct rescues sur-7(ku119);let-60(n1046) (data not shown). We integrated an extrachromosomal array carrying this construct into the genome. SUR-7∷GFP was expressed in most tissues with the notable exception of intestinal cells. Under a UV-light dissecting microscope, GFP was seen to be strongly expressed in the pharynx, some tail cells and the vulva (data not shown). Under higher magnification, GFP was observed in many if not all neurons, muscles and hypodermal cells. Strikingly, SUR-7∷GFP was not localized to the plasma membrane, as was the case for a CDF-1∷GFP construct (Bruinsma et al, 2002), but was enriched at the nuclear periphery and throughout the cytoplasm in a fenestrated pattern (Figure 3A and B). We suggest that this localization is endoplasmic reticular for two reasons. First, this expression pattern is indistinguishable from that of a C. elegans YFP∷TRAM construct (data not shown; TRAM is an endoplosmic recticulum (ER) transmembrane protein) (Rolls et al, 2002). Secondly, several CDF family members are known to be ER localized (Li and Kaplan, 2001; Clemens et al, 2002). We suggest that rather than functioning to transport heavy metals extracellularly like CDF-1, SUR-7 is localized to the membrane of cellular organelles, likely the ER, and functions to sequester intracellular Zn2+ into this structure, likewise reducing cytoplasmic levels of this metal ion. The two proteins function redundantly however, as RNAi of sur-7 in the cdf-1(lf) mutant background resulted in slow growth and pleiotrophic developmental defects (data not shown). Figure 3.A functional translational SUR-7∷GFP construct is expressed in most tissues and is probably localized to the ER. (A) High-powered magnification of SUR-7∷GFP in four descendents of the vulva precursor cell P6.p and (B) a body wall muscle cell. SUR-7∷GFP is enriched at the nuclear envelope and expressed in a reticular pattern throughout the cytoplasm. Download figure Download PowerPoint sur-7, ksr-1 and sur-6 likely act on factors downstream of Raf The observations that loss-of-function mutations in sur-7, ksr-1 and sur-6 suppress activated ras suggest that these genes act on factors downstream of ras. To understand the mechanism of their functions, it is critical to know whether they also act downstream of raf. Previous genetic analyses of sur-6, sur-8 and ksr-1 suggested that each functions upstream of lin-45/Raf, as lf mutations in none of these three genes can suppress the Muv phenotype associated with an overexpressed raf gain-of-function (gf) construct (Sieburth et al, 1998, 1999) (Table I). Similarly, a Drosophila ksr lf allele could not suppress the rough eye phenotype caused by overexpression of the same Raf construct (Therrien et al, 1995). This raf(gf) construct contains a Drosophila raf transcript fused to a transmembrane domain of the Torso receptor and is under transcriptional control of a heat shock promoter (Dickson et al, 1992). Like mutations in the other positive regulators, sur-7(ku119) could not suppress the Muv phenotype associated with this construct (Table I). However, biochemical work has indicated that KSR directly interacts with both Raf and MEK and therefore ksr-1 should function genetically downstream of lin-45/Raf. These conflicting observations have been puzzling and have raised concerns about assays using this raf(gf) construct. To address this concern, we took advantage of another, likely more physiologically relevant, raf(gf) construct. The phosphorylation state of Raf is a major mechanism of its activity modulation (Brtva et al, 1995; Hu et al, 1995; Chong et al, 2001). Basal phosphorylation of putative 14-3-3 binding sites is inhibitory to Raf activity. Relief of this inhibition may be mediated by protein phosphatases 1 and 2A (Abraham et al, 2000; Jaumot and Hancock, 2001). Two consensus 14-3-3 binding sites were identified in LIN-45, homologous to those in vertebrate B-Raf, and were shown to inhibit LIN-45 activity when phosphorylated (Chong et al, 2001). Furthermore, expression of a lin-45 construct encoding alanine substitutions for these conservative serine residues caused a penetrant Muv phenotype in wild-type worms. We integrated this lin-45(gf) construct producing a strain with a strong Muv phenotype (Materials and methods and Table I). In contrast to observations using hsp-dRaf(gf), we observed that both sur-6(lf) and ksr-1(lf), but not sur-8(lf), robustly suppressed the Muv phenotype caused by the lin-45(gf) construct (Table I). These data support a model in which SUR-8 exerts its positive regulation by facilitating the interaction between Ras and Raf (Li et al, 2000) while SUR-6 and KSR-1 function downstream of Raf possibly in a common convergent regulatory event (Sieburth et al, 1999). Moreover, PP2A has been shown to associate with Raf-1 in vertebrate cell culture and potentially mediate dephosphorylation of S259 (a putative 14-3-3 binding site), thereby contributing to Raf activation (Abraham et al, 2000). Our results suggest that such potential regulation of Raf by PP2A at S259 is not mediated by the SUR-6/PR55 regulatory subunit, since sur-6(lf) can suppress LIN-45 activation caused by mimicking dephosphorylation of this residue, which is presumably targeted by PP2A. Lastly, our data argue that sur-7 exerts its influence downstream of lin-45 as sur-7(ku119) also suppresses lin-45(gf) (Table I). Our genetic analysis also indicates that intercellular zinc concentration, regulated by SUR-7 activity, likely affects this pathway upstream of MEK. sur-7(ku119) does not suppress the Muv phenotype of a previously described strain containing gain-of-function mutant genes in both mpk-1 and the Drosophila mek gene (Lackner and Kim, 1998) (Table I). The Muv phenotype of this mpk-1(gf)/Dmek(gf) strain has been shown to be epistatic to mutations in genes acting upstream of mek but suppressed by mutations in factors known to act downstream of mpk-1 (Lackner and Kim, 1998). Consistent with SUR-7 acting upstream of MEK, ku-119 also failed to suppress the Muv phenotype of loss-of-function mutations in lin-1 and lin-31, two transcription factors that are targets of MAPK in the VPCs. These observations are consistent with the study of CDF-1 in Xenopus oocytes (Bruinsma et al, 2002). Together, these results suggest that Zn ions negatively affect this signaling pathway at the level of Raf activation of MEK. Protein interaction analyses and the data we present here place SUR-6 and KSR-1 as the only other known positive regulators at this point in the pathway. sur-7 and sur-6 antagonize par-1 in regulating ksr activity The kinase C-TAK1 (Cdc-25C-associated kinase 1) has recently been shown to act as a negative regulator of Ras signaling through the MAPK cascade (Muller et al, 2001). C-TAK1 was shown to phosphorylate murine KSR and inhibit its translocation to the plasma membrane. This prevents KSR from associating with activated Raf, thereby abrogating transduction of the signal to MEK and MAPK. The C. elegans homolog of C-TAK1 is PAR-1, a kinase previously defined for its role in early embryonic cell polarity (Guo and Kemphues, 1995). No role for C. elegans par-1 in Ras-mediated VPC induction has previously been described, but mutations in par-1 have recently been shown to alter vulval morphology (Hurd and Kemphues, 2003). Thus, we investigated a role for par-1 in Ras-mediated VPC induction. par-1 lf mutant animals exhibit a 100% maternal embryonic lethal phenotype as a result of failed partitioning of early embryonic cytoplasmic determinants. We used a balanced par-1(b274) strain to generate par-1(b274);let-60(n1046) double mutants as well as triple mutants containing lf in various positive regulators. The par-1(b274);let-60(n1046) strain exhibited a 100% Muv phenotype (compared to 76% for let-60(n1046) alone) (Table I). This increase in Muv percentage suggests that PAR-1 does play a negative role in VPC induction and may function similarly to its murine homolog by regulating KSR-1. If this is true, then par-1(lf) may revert the suppression of the ras(gf) phenotype by positive regulators upstream of ksr-1 but not regulators downstream. Indeed, this is what we observed (Table I). We used sur-8 and mpk-1 as controls since genetic characterization of these genes placed them respectively upstream and downstream of lin-45/Raf. An mpk-1 partial lf allele robustly suppressed the Muv phenotype of let-60(n1046) in the presence of par-1(b274); however, sur-8(ku167) suppression was overridden and the resulting triple mutant was 65% Muv. The suggested role for par-1 also predicts that ksr-1(lf) should still suppress let-60(n1046) with par-1(b274) in the background. Again, this is what we observed. Importantly, neither sur-6(lf) nor sur-7(lf) could suppress let-60(n1046) with par-1(b274) in the background, suggesting that the loss-of-function mutant phenotype of par-1 is epistatic to that of both sur-6 and sur-7. Given that PAR-1 most likely acts upstream of KSR based on biochemical data on the mammalian homologs (Muller et al, 2001) and genetic epistatic relationships between them (Table I), this result also suggests that SUR-6 and SUR-7 act upstream of KSR-1 to regulate the signaling activity. High Zn2+ levels increase KSR phosphorylation in mammalian tissue culture cells In mammalian tissue culture cells, C-TAK1 negatively regulates KSR function by phosphorylating a serine residue at 392 (Muller et al, 2001). Such a modification results in KSR binding to 14-3-3 and being cytoplasmically sequestered. Our genetic data presented above suggest that increased Zn2+ concentrations in sur-7(ku119) animals may negatively affect signaling through the MAPK cascade by one of at least three means. Increased Zn2+ levels may enhance the activity of PAR-1/C-TAK1, reduce the activity of PP2A that is regulated by the B-subunit SUR-6, or reduce KSR activity via a mechanism that is independent of PAR-1 and SUR-6. To further address thes
Referência(s)