Each week we showcase a hot new cell signaling article from a Nature Publishing Group journal. Free full text access to the paper will be maintained for three months, after which the paper will be accessible via the Research Library.
Hedgehog (Hh) signaling occurring through the seven-transmembrane receptor Smoothened (Smo) is central to animal development. However, little is known about the mechanism of Smo regulation. Now, two different groups report that Smo is dually regulated by protein kinase A (PKA) and casein kinase I (CKI) in response to Hedgehog signaling.
PKA has an established inhibitory role in the Hh pathway through the phosphorylation of full length latent transcription factor Cubitus interrruptus (Ci-155). This targets Ci-155 for proteolytic processing to generate the repressor Ci-75. Consistent with this, overexpression of constitutively active PKA catalytic subunit (mC*) inhibits the induction of Hh target genes. However, Jia et al. observed that modest expression of mC* in Drosophila wings activated the Hh pathway, resulting in the anterior expansion of the Hh targets patched (ptc) and engrailed(en). The same was true when Apionishev et al. co-expressed mC* with Smo. Conversely, inhibition of PKA through the expression of a mutant PKA regulatory subunit (R*) in wing discs blocked Hh-induced Smo accumulation and activity, suggesting that PKA might indeed promote pathway activation.
Jia et al. used point mutants to show that Smo is sequentially phosphorylated by PKA and CKI in vitro and that blocking either activity in cells results in decreased levels of phosphorylated Smo protein. Thus, Smo is a direct target of PKA and CKI, but is this relevant to Hh signaling during development? In both studies Smo phosphorylation site mutants failed to fully rescue smo mutant phenotypes while phosphomimetic mutants resulted in constitutive Smo activity and ectopic expression of Hh target genes. Furthermore, the levels of Smo activity appeared to correlate with the levels of its phosphorylation.
While it is unclear how phosphorylation by PKA and CKI affects Smo activity, both groups concluded that it regulates Smo stability on the cell surface. Apionishev et al. observed that Smo possessing mutated PKA and CKI sites is resistant to the Ptc-dependent proteolysis that normally occurs in anterior cells, whereas Jia et al. found that phosphorylated Smo accumulates on the cell surface more readily than unphosphorylated Smo. Moreover, Hh signaling appears to promote Smo accumulation in response to phosphorylation, but whether it does so by regulating phosphorylation by PKA and CKI is still subject to further investigation. Smo hyperphosphorylation is essential for its accumulation and high-threshold Hh signaling. Hh, however, also controls Smo cell surface accumulation and pathway activation by mechanisms independent of PKA phosphorylation, thus rendering its relationship with Smo even more intricate.
article Jianhang Jia, Chao Tong, Bing Wang, Liping Luo & Jin Jiang Hedgehog signalling activity of Smoothened requires phosphorylation by protein kinase A and casein kinase I Nature, 432, 1045 – 1050
(23 December 2004); doi:10.1038/nature03179 Full Text (HTML) | PDF | Subscribe to Nature
article Sergey Apionishev, Natalya M Katanayeva, Steven A Marks, Daniel Kalderon & Andrew Tomlinson Drosophila Smoothened phosphorylation sites essential for Hedgehog signal transduction Nature Cell Biology, 7, 86 – 92
(12 December 2004); doi:10.1038/ncb1210 Full Text (HTML) | PDF | Subscribe to Nature Cell Biology
Groucho (Gro) is a transcriptional corepressor acting in concert with effectors of the Notch, Wnt and TGF- signaling pathways. Hasson et al. now show that Gro-dependent transcriptional repression is attenuated by signaling downstream of the EGF receptor (EGFR). This finding provides a biochemical explanation to previous reports that EGFR signaling antagonizes the Notch pathway.
Hasson et al. first asked whether Gro is phosphorylated in response to receptor tyrosine kinase (RTK) signaling. Transfection of cells with Gro and a constitutively active derivative of the EGFR led to increased phosphorylation of Gro. Consistent with this result, a version of Gro carrying mutations at two consensus MAP kinase (MAPK) phosphorylation sites (GroAA) showed reduced phosphorylation.
To test whether EGFR-dependent phosphorylation of Gro has an effect on transcription, the authors generated a pseudophosphorylated form of Gro by substituting the MAPK target sites with aspartic acid residues (GroDD). They misexpressed these Gro constructs in Drosophila wing imaginal discs and, as a read-out, assessed the expression of genes that are normally transcriptionally silenced by Gro. Expression of the Gro targets Distalless and Spalt was reduced, with GroAA acting as a more effective repressor, and GroDD being less effective. Thus, the potency of Gro as a corepressor of transcription depends at least in part on its EGFR-dependent phosphorylation status.
Does attenuation of Gro repressor activity by RTK signaling affect a specific phenotype in a developmental setting? The EGFR pathway promotes vein formation by overriding the vein repression activity of the Notch pathway. By misexpressing GroDD in the wing, Hasson et al. showed that they could induce extra vein tissue, consistent with the idea that constitutive activation of Gro could antagonize Notch-dependent repression of vein formation. As expected, wings expressing GroAA lost all veins. A similar phenotype was also observed during mesothoracic bristle development, suggesting that antagonism between the EGFR and Notch pathways is mediated by phosphorylation of Gro.
These results show how RTK signaling can act as a switch between transcriptional silencing and activation, and suggest that Gro has a key role in this process. As Gro mediates repression by Wnt and TGF-, RTK-mediated phosphorylation of Gro may modulate the output of these pathways as well.
Alan Packer, Senior Editor Nature Genetics
article Peleg Hasson, Nirit Egoz, Clint Winkler, Gloria Volohonsky, Songtao Jia, Tama Dinur, Talila Volk, Albert J Courey & Ze'ev Paroush EGFR signaling attenuates Groucho-dependent repression to antagonize Notch transcriptional output Nature Genetics, 37, 101 – 105
(12 December 2004); doi:10.1038/ng1486 Full Text (HTML) | PDF | Subscribe to Nature Genetics
Morphogenetic movements: Less Wnt from Kaiso
Traditionally, canonical Wnt signaling through -catenin was thought to be the major regulator of developmental decisions. But non-canonical Wnt pathways have since risen to fames, also contributing to development in several contexts. Kim et al. now show that the transcriptional repressor Kaiso and its binding partner p120-catenin form part of a novel signaling pathway that regulates non-canonical Wnt-mediated vertebrate development.
Focusing on Xenopus, Kim et al. found that depletion of Kaiso using morpholino-oligonucleotides caused severe defects in gastrulation and was specifically required to regulate movements of the dorsal marginal zone. A similar phenotype was apparent when the authors overexpressed Kaiso by injecting exogenous mRNA, suggesting that Kaiso levels should be kept within a defined range. By conducting marker analysis Kim et al. observed that expression of the non-canonical Wnt11 was upregulated in a Kaiso loss-of-function background. Consistent with a previously demonstrated role for non-canonical Wnt signaling in morphogenetic movements, overexpression of dominant negative Wnt11 or of its downstream target Dishevelled (Dsh) partially rescued the defects caused by Kaiso depletion.
Sequence analysis revealed that the Wnt11 promoter has a conserved Kaiso-specific consensus sequence. But is Wnt11 a direct target of Kaiso? The authors performed mutational analysis combined with chromatin immunoprecipitation and electrophoretic mobility shift assays to show that Kaiso binds to a sequence-specific Wnt11 promoter fragment. Importantly, in vivo depletion of Kaiso in the animal hemisphere was able to significantly increase Wnt11 levels.
P120-catenin has been implicated in vertebrate development and can interact with Kaiso. Kim et al. found that p120-catenin overexpression was sufficient to displace Kaiso from the Wnt11 promoter and that this required p120-catenin’s nuclear localization signal. In addition, the authors combined overexpression and depletion studies to demonstrate that p120-catenin also modulates Wnt11 levels in vivo; it appears that p120-catenin’s role is to relieve the Kaiso-mediated repression of the Wnt11 promoter.
Thus, Kim et al. have discovered a novel pathway whereby Kaiso and p120-catenin act to regulate non-canonical Wnt signals and modulate morphogenetic movements during embryogenesis. It is however unlikely that Kaiso is the sole regulator of non-canonical Wnt signals, as endogenous Kaiso does not seem able to completely block Wnt11 expression. Moreover, Kaiso’s known ability to also participate in methylation-dependent transcriptional repression argues for additional roles for Kaiso not necessarily related to Wnt11 or other Wnt signaling pathways.
article Si Wan Kim, Jae-Il Park, Christopher M Spring, Amy K Sater, Hong Ji, Abena A Otchere, Juliet M Daniel & Pierre D McCrea Non-canonical Wnt signals are modulated by the Kaiso transcriptional repressor and p120-catenin Nature Cell Biology , 6, 1212 – 1220
(2004); doi:10.1038/ncb1191 Full Text (HTML) | PDF | Subscribe to Nature Cell Biology
Oncogenic and oxidative stress: Seladin-1 at the crossroads
Uncontrolled Ha-Ras signaling is causally linked to cancer formation but is also associated with premature cellular senescence as it induces reactive oxygen species formation. Wu et al. have now provided insight into this apparent paradox by identifying Seladin-1 as a key regulator of Ras-induced cell cycle arrest.
While searching for oncogene-induced senescence regulators using a targeted genetic screen, the authors stumbled upon Seladin-1, the human orthologue of plant DIMINUTO/DWARF1 — a gene previously implicated in plant development and cholesterol metabolism. In the absence of Seladin-1 expression, rat fibroblasts overexpressing Ras did not exhibit the expected senescent phenotype, but formed aggressive tumors when inoculated into nude mice. A similar phenotype was observed when Seladin-1 was depleted from Ras-transformed human fibroblasts. Thus, it appears that Seladin-1 actively controls Ras-induced cancer formation and senescence.
Wu et al. next turned their attention to known mediators of Ras signaling. They observed that RNAi mediated knockdown of Seladin-1 prevented the Ras-induced accumulation of p53 and its target p21Cip1/Waf1, therefore placing Seladin-1 downstream of reactive oxygen species (ROS) signaling. Oxidative stress and Ras activation were shown to upregulate Seladin-1 expression and mediate its translocation to the nucleus. In contrast, cellular stress induced by UV radiation did not change the levels or localization of Seladin-1, suggesting that this protein has a role in regulating p53 specifically in response to oncogenic or oxidative stress.
Seladin-1 was found to not only bind p53, but also the E3 ubiquitin ligase Mdm2. Seladin-1 – Mdm2 binding was further promoted by oxidative stress. Mutational analysis confirmed the presence of two distinct binding regions on Seladin-1 for p53 and Mdm2, respectively. Seladin-1 directly competes with Mdm2 for binding to the amino terminus of p53 thus preventing Mdm2-mediated ubiquitination and degradation of p53. It remains unclear if Mdm2 activity towards other targets is regulated by Seladin-1 binding.
In light of their findings, Wu et al. propose that, following oxidative stress, Seladin-1 directly inhibits the Mdm2-dependent degradation of p53, resulting in accumulation of p53 levels and subsequent cell cycle arrest. This simple model adds another player to the duality of the Ras phenotype. Seladin-1’s new role could also be a stepping stone in the understanding of p53-independent Ras-mediated arrest observed in human cells. Seladin-1 has been implicated in Alzheimer’s disease as well as in cholesterol biosynthesis. While it remains unclear if these effects are related to p53 or Mdm2, both oxidative stress and cholesterol metabolism have been associated with ageing and neurodegenerative disorders. Whether Seladin-1 turns out to be the master regulator of these seemingly unrelated cellular functions is certainly something to watch out for.
article CHAOWEI WU, IRENE MILOSLAVSKAYA, SILVIA DEMONTIS, ROBERTA MAESTRO & KONSTANTIN GALAKTIONOV Regulation of cellular response to oncogenic and oxidative stress by Seladin-1 nature, 432, 640 – 645
(02 December 2004); doi:10.1038/nature03173 Full Text (HTML) | PDF | Subscribe to Nature
signaling pathways. Hasson et al. now show that Gro-dependent transcriptional repression is attenuated by signaling downstream of the EGF receptor (EGFR). This finding provides a biochemical explanation to previous reports that EGFR signaling antagonizes the Notch pathway.
