- Email: [email protected]
Receptor tyrosine phosphatases: The worm clears the picture
Dispatch
R725
Receptor tyrosine phosphatases: The worm clears the picture Kai Zinn
Recent work on the Caenorhabditis elegans clr-1 gene shows that the receptor tyrosine phosphatase that it encodes negatively regulates a receptor tyrosine kinase related to mammalian fibroblast growth factor receptors. This opens up a promising system for investigating receptor tyrosine phosphatase function. Address: Division of Biology, California Institute of Technology, Pasadena, California 91125, USA. E-mail: [email protected] Current Biology 1998, 8:R725–R726 http://biomednet.com/elecref/09609822008R0725 © Current Biology Ltd ISSN 0960-9822
Reversible protein phosphorylation is well established as a general mechanism by which diverse processes are regulated in cells. Such regulation requires not only enzymes that add a phosphate to a protein — the kinases — but also ones that will take the phosphate off — the phosphatases. Many such phosphatases have been identified, some of which have characteristics of transmembrane receptors. The receptor tyrosine phosphatases are particularly intriguing, suggesting as they do that intracellular dephosphorylation might be regulated in response to an extracellular stimulus in a similar manner to phosphorylation control by the much better known receptor tyrosine kinases. Relatively little is known about how receptor tyrosine phosphatase activities are regulated, however; we also lack a clear picture of how these phosphatases affect kinase signaling in vivo. This might soon change following recent work on the clear-1 (clr-1) gene of the nematode worm Caenorhabditis elegans [1]. C. elegans clr-1 mutants have a diverse set of defects that are caused by fluid accumulation within the pseudocoelom. Cloning and sequencing of the clr-1 gene showed it to encode a receptor tyrosine phosphatase [1]. In common with many other members of the large receptor tyrosine phosphatase family, Clr-1 has features of a cell adhesion molecule. In particular, its extracellular region includes immunoglobulin-like domains and fibronectin type III repeats, which frequently occur in adhesion molecules. Clr-1 also shares with other receptor tyrosine phosphatases the common feature of having two ‘phosphatase homology units’, D1 and D2, only the membrane proximal one (D1) of which is generally catalytically active in vitro. Although rather little is known about the physiological functions and cellular actions of receptor tyrosine phosphatases, they have been implicated in some specific control processes. Thus, receptor tyrosine phosphatases
that look like adhesion molecules control many aspects of axon guidance in Drosophila (reviewed in [2]), and may also be involved in neural adhesion and neurite outgrowth in vertebrates [3,4]. And two mammalian receptor tyrosine phosphatases, CD45 and RPTPa, can dephosphorylate and activate cytoplasmic tyrosine kinases of the Src family. Other tyrosine phosphatases must regulate signaling by receptor tyrosine kinases, but very little has been known about this and, in particular, relevant in vivo biochemical targets have not been identified [5]. Kokel et al. [1] have found that the Clr-1 receptor tyrosine phosphatase negatively regulates signaling by the Egl-15 receptor, a relative of the mammalian fibroblast growth factor (FGF) receptor tyrosine kinases. Egl-15 has a number of distinct roles in C. elegans development, the best-understood of which is in determining the migration of the sex myoblasts. Some ‘hypomorphic’ mutants with reduced egl-15 activity are defective in egg-laying — the egl phenotype — because their sex myoblasts fail to migrate anteriorly to their appropriate positions flanking the gonad [6]. Sex myoblast migration is directed by Egl15-mediated attraction towards sources of the receptor’s ligand, the FGF-like peptide Egl-17 [7]. The adaptor protein Sem-5, which links receptor tyrosine kinases to the Ras pathway, has also been shown to be required for gonad-dependent sex myoblast migration [6]. The phenotype of egl-15 mutants is quite variable: complete null mutants die as larvae, and some hypomorphic mutants have a ‘scrawny’ adult morphology that may be related to the null phenotype. A screen for suppressors of the scrawny phenotype resulted in the isolation of multiple clr-1 alleles. On a wild-type background, these mutations confer the ‘clear’ phenotype described above; egl-15, clr-1 double mutants, however, have neither scrawny nor clear phenotypes, suggesting that the two genes act antagonistically. Additional genetic studies indicated that the normal function of Clr-1 is to regulate Egl-15 signaling. The Clear phenotype would then be caused by deregulated Egl-15 activity, an inference supported by the observation that transgenic worms expressing a mutant form of Egl-15 presumed to be constitutively active — as it carries a mutation in its transmembrane domain that, in the context of other receptor tyrosine kinases, is known to cause constitutive dimerization and activation — exhibit the clear phenotype. The negative regulation of Egl-15 signaling by Clr-1 requires the phosphatase activity of the latter’s D1 domain. This was shown in two ways. First, a transgene
R726
Current Biology, Vol 8 No 20
encoding a mutant form of Clr-1 in which the essential D1 cysteine residue was replaced by serine was unable to rescue the clear phenotype of clr-1 mutants. And second, three strong clr-1 mutations alter conserved residues within the tyrosine phosphatase core consensus sequence. Transgenes encoding forms of Clr-1 with the entire D2 domain removed could, however, still rescue clr-1 mutants; in accord with results on other receptor tyrosine phosphatases, the D2 domain of Clr-1 had no detectable phosphatase activity in vitro. These results indicate that D1 phosphatase activity is required for Clr-1 function, but no specific role has been identified for the D2 domain [1]. The indications are that Egl-15 signals by a number of distinct intracellular pathways, only some of which are strongly modulated by Clr-1. For example, one class of egl-15 mutations cause sex myoblast migration defects but do not suppress the effects of clr-1 mutations, whereas those of another class do suppress the effects of clr-1 mutations but do not affect sex myoblast positioning [6]. Mutations of clr-1 have only minor effects on sex myoblast positioning in a wild-type background, and do not suppress the sex myoblast migration defect caused by some of the hypomorphic egl-15 mutations [1].
are downstream signaling proteins that link Clr-1 to the Egl-15 pathway, the genes encoding them cannot be mutated to a viable clear phenotype. The available data are consistent with, but do not prove, a model in which the Clr-1 receptor tyrosine phosphatase directly dephosphorylates a subset of the autophosphorylated tyrosines on the Egl-15 receptor tyrosine kinase, thereby downregulating adaptor-mediated signaling through certain pathways. The slight effects of clr-1 mutations on sex myoblast positioning may indicate that Clr-1 can dephosphorylate the tyrosines recognized by Sem-5 that are important for sex myoblast migration, but other tyrosine phosphatases may be primarily responsible for controlling phosphorylation at these sites. Whether this ‘direct’ model is correct will probably be established within the next few years. In any case, the work of Kokel et al. [1] is an important step forward, as it opens up a tractable system for the genetic characterization of receptor tyrosine phosphatase signaling pathways. Acknowledgements Work on development in the author’s laboratory is supported by the NIH, the Human Frontiers Science Project, and the Keck Foundation.
References
Mutations in the sem-5 gene were isolated as suppressors of clr-1, suggesting that the Sem-5 adaptor protein, which is known to be required for sex myoblast migration, is also a component of one or more of the pathways downstream of Egl-15 that are negatively regulated by Clr-1 [6]. The cytoplasmic protein Soc-2, the sequence of which includes leucine-rich repeats that might mediate protein–protein interactions, is likely to lie on one of the other downstream pathways, but its precise role is presently unknown [8]. The Egl-15 pathways that are modulated by Clr-1 are probably initiated by binding to the receptor of ligands other than Egl-17, as mutations that abolish Egl-17 production lead to sex myoblast migration defects, but do not cause the scrawny phenotype or suppress the effects of clr-1 mutations [9].
1.
The power of genetic analysis in C. elegans — particularly the ability to screen for second-site mutations that modify the effects of a given mutation — has been exploited extensively to define signaling pathways. The application of similar approaches to the Clr-1 system may well lead to the identification of other components involved in receptor tyrosine phosphatase signaling. The sequence of Clr-1 suggests that it recognizes an extracellular ligand, and the identification of ligands is one of the central problems in the receptor tyrosine phosphatase field. However, clr-1 appears to be the only gene that can be mutated alone to produce a viable clear phenotype [1], so if Clr-1 does have a ligand then either there is functional redundancy that masks the effects of mutations in the ligand’s gene, or the ligand has multiple functions and mutations in its gene confer lethality. Similarly, if there
8.
2. 3. 4.
5. 6. 7.
9.
Kokel M, Borland CZ, DeLong L, Horvitz HR, Stern MJ: clr-1 encodes a receptor tyrosine phosphatase that negatively regulates an FGF receptor signaling pathway in Caenorhabditis elegans. Genes Dev 1998, 12:1425-1437. Desai CJ, Sun Q, Zinn K: Tyrosine phosphorylation and axon guidance: of flies and mice. Curr Opin Neurobiol 1997, 7:70-74. Stoker AW, Gehrig B, Haj F, Bay B-H: Axonal localisation of the CAM-like tyrosine phosphatase CRYPa: a signaling molecule of embryonic growth cones. Development 1995, 121:1833-1844. Brady-Kalnay SM, Mourton T, Nixon JP, Pietz GE, Kinch M, Chen H, Brackenbury R, Rimm DL, Del Vecchio RL, Tonks NK: Dynamic interaction of PTPm with multiple cadherins in vivo. J Cell Biol 1998, 141:287-296. Neel BG, Tonks NK: Protein tyrosine phosphatases in signal transduction. Curr Opin Cell Biol 1997, 9:193-204. De Vore D, Horvitz HR, Stern MJ: An FGF receptor signaling pathway is required for the normal cell migrations of the sex myoblasts in C. elegans hermaphrodites. Cell 1995, 83:611-620. Burdine RD, Branda CS, Stern MJ: EGL-17(FGF) expression coordinates the attraction of the migrating sex myoblasts with vulval induction in C. elegans. Development 1998, 125:1083-1093. Selfors LM, Schutzman JL, Borland CZ, Stern MJ: soc-2 encodes a leucine-rich repeat protein implicated in growth factor receptor signaling. Proc Natl Acad Sci USA 1998, 95:6903-6908. Burdine RD, Chen EB, Kwok SF, Stern MJ: egl-17 encodes an invertebrate fibroblast growth factor family member required specifically for sex myoblast migration in Caenorhabditis elegans. Proc Natl Acad Sci USA 1997, 94:2433-2437.
R725
Receptor tyrosine phosphatases: The worm clears the picture Kai Zinn
Recent work on the Caenorhabditis elegans clr-1 gene shows that the receptor tyrosine phosphatase that it encodes negatively regulates a receptor tyrosine kinase related to mammalian fibroblast growth factor receptors. This opens up a promising system for investigating receptor tyrosine phosphatase function. Address: Division of Biology, California Institute of Technology, Pasadena, California 91125, USA. E-mail: [email protected] Current Biology 1998, 8:R725–R726 http://biomednet.com/elecref/09609822008R0725 © Current Biology Ltd ISSN 0960-9822
Reversible protein phosphorylation is well established as a general mechanism by which diverse processes are regulated in cells. Such regulation requires not only enzymes that add a phosphate to a protein — the kinases — but also ones that will take the phosphate off — the phosphatases. Many such phosphatases have been identified, some of which have characteristics of transmembrane receptors. The receptor tyrosine phosphatases are particularly intriguing, suggesting as they do that intracellular dephosphorylation might be regulated in response to an extracellular stimulus in a similar manner to phosphorylation control by the much better known receptor tyrosine kinases. Relatively little is known about how receptor tyrosine phosphatase activities are regulated, however; we also lack a clear picture of how these phosphatases affect kinase signaling in vivo. This might soon change following recent work on the clear-1 (clr-1) gene of the nematode worm Caenorhabditis elegans [1]. C. elegans clr-1 mutants have a diverse set of defects that are caused by fluid accumulation within the pseudocoelom. Cloning and sequencing of the clr-1 gene showed it to encode a receptor tyrosine phosphatase [1]. In common with many other members of the large receptor tyrosine phosphatase family, Clr-1 has features of a cell adhesion molecule. In particular, its extracellular region includes immunoglobulin-like domains and fibronectin type III repeats, which frequently occur in adhesion molecules. Clr-1 also shares with other receptor tyrosine phosphatases the common feature of having two ‘phosphatase homology units’, D1 and D2, only the membrane proximal one (D1) of which is generally catalytically active in vitro. Although rather little is known about the physiological functions and cellular actions of receptor tyrosine phosphatases, they have been implicated in some specific control processes. Thus, receptor tyrosine phosphatases
that look like adhesion molecules control many aspects of axon guidance in Drosophila (reviewed in [2]), and may also be involved in neural adhesion and neurite outgrowth in vertebrates [3,4]. And two mammalian receptor tyrosine phosphatases, CD45 and RPTPa, can dephosphorylate and activate cytoplasmic tyrosine kinases of the Src family. Other tyrosine phosphatases must regulate signaling by receptor tyrosine kinases, but very little has been known about this and, in particular, relevant in vivo biochemical targets have not been identified [5]. Kokel et al. [1] have found that the Clr-1 receptor tyrosine phosphatase negatively regulates signaling by the Egl-15 receptor, a relative of the mammalian fibroblast growth factor (FGF) receptor tyrosine kinases. Egl-15 has a number of distinct roles in C. elegans development, the best-understood of which is in determining the migration of the sex myoblasts. Some ‘hypomorphic’ mutants with reduced egl-15 activity are defective in egg-laying — the egl phenotype — because their sex myoblasts fail to migrate anteriorly to their appropriate positions flanking the gonad [6]. Sex myoblast migration is directed by Egl15-mediated attraction towards sources of the receptor’s ligand, the FGF-like peptide Egl-17 [7]. The adaptor protein Sem-5, which links receptor tyrosine kinases to the Ras pathway, has also been shown to be required for gonad-dependent sex myoblast migration [6]. The phenotype of egl-15 mutants is quite variable: complete null mutants die as larvae, and some hypomorphic mutants have a ‘scrawny’ adult morphology that may be related to the null phenotype. A screen for suppressors of the scrawny phenotype resulted in the isolation of multiple clr-1 alleles. On a wild-type background, these mutations confer the ‘clear’ phenotype described above; egl-15, clr-1 double mutants, however, have neither scrawny nor clear phenotypes, suggesting that the two genes act antagonistically. Additional genetic studies indicated that the normal function of Clr-1 is to regulate Egl-15 signaling. The Clear phenotype would then be caused by deregulated Egl-15 activity, an inference supported by the observation that transgenic worms expressing a mutant form of Egl-15 presumed to be constitutively active — as it carries a mutation in its transmembrane domain that, in the context of other receptor tyrosine kinases, is known to cause constitutive dimerization and activation — exhibit the clear phenotype. The negative regulation of Egl-15 signaling by Clr-1 requires the phosphatase activity of the latter’s D1 domain. This was shown in two ways. First, a transgene
R726
Current Biology, Vol 8 No 20
encoding a mutant form of Clr-1 in which the essential D1 cysteine residue was replaced by serine was unable to rescue the clear phenotype of clr-1 mutants. And second, three strong clr-1 mutations alter conserved residues within the tyrosine phosphatase core consensus sequence. Transgenes encoding forms of Clr-1 with the entire D2 domain removed could, however, still rescue clr-1 mutants; in accord with results on other receptor tyrosine phosphatases, the D2 domain of Clr-1 had no detectable phosphatase activity in vitro. These results indicate that D1 phosphatase activity is required for Clr-1 function, but no specific role has been identified for the D2 domain [1]. The indications are that Egl-15 signals by a number of distinct intracellular pathways, only some of which are strongly modulated by Clr-1. For example, one class of egl-15 mutations cause sex myoblast migration defects but do not suppress the effects of clr-1 mutations, whereas those of another class do suppress the effects of clr-1 mutations but do not affect sex myoblast positioning [6]. Mutations of clr-1 have only minor effects on sex myoblast positioning in a wild-type background, and do not suppress the sex myoblast migration defect caused by some of the hypomorphic egl-15 mutations [1].
are downstream signaling proteins that link Clr-1 to the Egl-15 pathway, the genes encoding them cannot be mutated to a viable clear phenotype. The available data are consistent with, but do not prove, a model in which the Clr-1 receptor tyrosine phosphatase directly dephosphorylates a subset of the autophosphorylated tyrosines on the Egl-15 receptor tyrosine kinase, thereby downregulating adaptor-mediated signaling through certain pathways. The slight effects of clr-1 mutations on sex myoblast positioning may indicate that Clr-1 can dephosphorylate the tyrosines recognized by Sem-5 that are important for sex myoblast migration, but other tyrosine phosphatases may be primarily responsible for controlling phosphorylation at these sites. Whether this ‘direct’ model is correct will probably be established within the next few years. In any case, the work of Kokel et al. [1] is an important step forward, as it opens up a tractable system for the genetic characterization of receptor tyrosine phosphatase signaling pathways. Acknowledgements Work on development in the author’s laboratory is supported by the NIH, the Human Frontiers Science Project, and the Keck Foundation.
References
Mutations in the sem-5 gene were isolated as suppressors of clr-1, suggesting that the Sem-5 adaptor protein, which is known to be required for sex myoblast migration, is also a component of one or more of the pathways downstream of Egl-15 that are negatively regulated by Clr-1 [6]. The cytoplasmic protein Soc-2, the sequence of which includes leucine-rich repeats that might mediate protein–protein interactions, is likely to lie on one of the other downstream pathways, but its precise role is presently unknown [8]. The Egl-15 pathways that are modulated by Clr-1 are probably initiated by binding to the receptor of ligands other than Egl-17, as mutations that abolish Egl-17 production lead to sex myoblast migration defects, but do not cause the scrawny phenotype or suppress the effects of clr-1 mutations [9].
1.
The power of genetic analysis in C. elegans — particularly the ability to screen for second-site mutations that modify the effects of a given mutation — has been exploited extensively to define signaling pathways. The application of similar approaches to the Clr-1 system may well lead to the identification of other components involved in receptor tyrosine phosphatase signaling. The sequence of Clr-1 suggests that it recognizes an extracellular ligand, and the identification of ligands is one of the central problems in the receptor tyrosine phosphatase field. However, clr-1 appears to be the only gene that can be mutated alone to produce a viable clear phenotype [1], so if Clr-1 does have a ligand then either there is functional redundancy that masks the effects of mutations in the ligand’s gene, or the ligand has multiple functions and mutations in its gene confer lethality. Similarly, if there
8.
2. 3. 4.
5. 6. 7.
9.
Kokel M, Borland CZ, DeLong L, Horvitz HR, Stern MJ: clr-1 encodes a receptor tyrosine phosphatase that negatively regulates an FGF receptor signaling pathway in Caenorhabditis elegans. Genes Dev 1998, 12:1425-1437. Desai CJ, Sun Q, Zinn K: Tyrosine phosphorylation and axon guidance: of flies and mice. Curr Opin Neurobiol 1997, 7:70-74. Stoker AW, Gehrig B, Haj F, Bay B-H: Axonal localisation of the CAM-like tyrosine phosphatase CRYPa: a signaling molecule of embryonic growth cones. Development 1995, 121:1833-1844. Brady-Kalnay SM, Mourton T, Nixon JP, Pietz GE, Kinch M, Chen H, Brackenbury R, Rimm DL, Del Vecchio RL, Tonks NK: Dynamic interaction of PTPm with multiple cadherins in vivo. J Cell Biol 1998, 141:287-296. Neel BG, Tonks NK: Protein tyrosine phosphatases in signal transduction. Curr Opin Cell Biol 1997, 9:193-204. De Vore D, Horvitz HR, Stern MJ: An FGF receptor signaling pathway is required for the normal cell migrations of the sex myoblasts in C. elegans hermaphrodites. Cell 1995, 83:611-620. Burdine RD, Branda CS, Stern MJ: EGL-17(FGF) expression coordinates the attraction of the migrating sex myoblasts with vulval induction in C. elegans. Development 1998, 125:1083-1093. Selfors LM, Schutzman JL, Borland CZ, Stern MJ: soc-2 encodes a leucine-rich repeat protein implicated in growth factor receptor signaling. Proc Natl Acad Sci USA 1998, 95:6903-6908. Burdine RD, Chen EB, Kwok SF, Stern MJ: egl-17 encodes an invertebrate fibroblast growth factor family member required specifically for sex myoblast migration in Caenorhabditis elegans. Proc Natl Acad Sci USA 1997, 94:2433-2437.