Portal de Periódicos da CAPES

More Surprises in the Hedgehog Signaling Pathway

2000; Cell Press; Volume: 100; Issue: 2 Linguagem: Inglês

10.1016/s0092-8674(00)81555-x

1097-4172

Andrew P. McMahon,

Developmental Biology and Gene Regulation

One of the great successes of the past decade has been the unraveling of complex signal transduction pathways that govern embryonic development. In many cases, the identification of the latest player in the process provides an insight that fits in well with the prevailing logic of the biochemical circuitry. However, in the case of the Hedgehog pathway, new discoveries often highlight how shaky our understanding of the molecular and cellular mechanisms underlying this critical signaling pathway is. The latest example comes from a recent publication in Cell, which identifies an important new player in Hedgehog signaling (3Burke R Nellen D Bellotto M Hafen E Senti K.-A Dickson B.J Basler K Cell. 1999; 99: 803-816Abstract Full Text Full Text PDF PubMed Scopus (411) Google Scholar). Hedgehog signaling has been shown to regulate an enormous variety of developmental events in the fly and vertebrate embryo and plays a central role in several cancers (for reviews on subject areas discussed below, see 6Hammerschmidt M Brook A McMahon A.P Trends Genet. 1997; 13: 14-21Abstract Full Text PDF PubMed Scopus (481) Google Scholar, 8Ingham P.W EMBO J. 1998; 17: 3505-3511Crossref PubMed Scopus (374) Google Scholar). In Drosophila, Hedgehog patterns the segment, wing, leg, eye, and regions of the fly brain either directly, or through the recruitment of other signaling factors such as Decapentaplegic (Dpp) and Wingless. In contrast to the single fly member, there are three Hedgehogs in mammals; Sonic (Shh), Desert (Dhh), and Indian (Ihh), and yet others in lower vertebrates. Shh activity at the midline patterns the overlying ventral neural tube and adjacent ventral somites, and participates in the development of left–right asymmetry. Elsewhere, Shh is a polarizing activity in the limb and regulates morphogenesis of a variety of organs including the eye, hair, and lungs. Dhh and Ihh play more restricted roles: Dhh acts in the regulation of spermatogenesis and organization of the perineurium, which ensheaths peripheral nerves, and Ihh in coordinating proliferation and maturation of chondrocytes during development of the endochondral skeleton (17St-Jacques B Hammerschmidt M McMahon A.P Genes Dev. 1999; 13: 2072-2086Crossref PubMed Scopus (1251) Google Scholar). Hedgehog signals appear to have both short- and long-range activities. In addition, they act as morphogens to induce distinct cell fates at specific concentration thresholds. Given the importance of Hedgehog signaling, there is not surprisingly a great deal of interest in how an active signal is produced, moved, received, and transduced to give an appropriate response in the target cell. It is now clear that most if not all Hedgehog signaling leads to the activation of transcriptional effectors that are zinc finger–containing proteins of the Ci/Gli family (reviewed by 1Aza-Blanc P Kornberg T.B Trends Genet. 1999; 15: 458-462Abstract Full Text Full Text PDF PubMed Scopus (144) Google Scholar). In the fly, the absence of Hedgehog signaling leads to constitutive cleavage of Ci, generating a repressor form that binds to, and inactivates, Hedgehog target genes. Hedgehog signaling inhibits this proteolytic cleavage by a phosphorylation-dependent process. As a result, full-length protein, in conjunction with appropriate partners, activates transcriptional targets (Figure 1). In vertebrates, there are three Gli genes whose precise roles in activating and repressing Hedgehog targets are not fully resolved. The first big surprise came with the initial genetic analysis of Hedgehog signaling in Drosophila. On the basis of an elegant study, Ingham and colleagues proposed a model that is now widely accepted: Hedgehog signaling leads to ligand-dependent receptor inactivation, the likely receptor being encoded by the segment polarity gene, patched (7Ingham P.W Taylor A.M Nakano Y Nature. 1991; 353: 184-187Crossref PubMed Scopus (342) Google Scholar). Biochemical and genetic analyses have confirmed this model and provided many important new insights. Hedgehog indeed binds to Patched, a multipass membrane protein. In the absence of ligand, Patched is required to inhibit the activity of Smoothened, a seven-transmembrane protein that shares homology with G protein–coupled receptors (Figure 1). Smoothened is absolutely required for all Hedgehog signaling in Drosophila. In the absence of both Patched and Hedgehog, Smoothened activity leads to constitutive activation of Hedgehog targets. Thus, Hedgehog binding to Patched represses Patched activity leading to derepression of Smoothened and apparent ligand-independent signaling by Smoothened. How then is Hedgehog binding to Patched communicated to Smoothened? The suggestion from biochemical studies is that these two proteins are in fact together in a receptor complex pointing to a direct mechanism (18Stone D.M Hynes M Armanini M Swanson T.A Gu Q Johnson R.L Scott M.P Pennica D Goddard A Phillips H et al.Nature. 1996; 384: 129-134Crossref PubMed Scopus (928) Google Scholar). However, this important conclusion is based on nonresponsive cells engineered to express high levels of Smoothened and Patched, and needs to be confirmed by direct cellular and biochemical analysis in the organism itself. A further twist comes from the fact that patched is itself a target of Hedgehog signaling. The increased levels of Patched produced in response to Hedgehog signaling leads to the sequestration of Hedgehog thereby limiting the range of Hedgehog action. This negative regulatory mechanism may also facilitate the generation of a sharp concentration gradient of Hedgehog, thereby altering the potential responses of cells within the target field. As usual, the situation is more complex in vertebrates. In mammals there are two Patched proteins, both of which bind vertebrate Hedgehogs with similar affinity (4Carpenter D Stone D.M Brush J Ryan A Armanini M Frantz G Rosenthal A de Sauvage F.L Proc. Natl. Acad. Sci. USA. 1998; 95: 13630-13634Crossref PubMed Scopus (200) Google Scholar). Patched-1 is confined to target cells and is upregulated in response to Hedgehog signaling. As expected from the Drosophila work, mice lacking Patched-1 activity constitutively activate Hedgehog response genes in target tissues (5Goodrich L.V Milenkovic L Higgins K.M Scott M.P Science. 1997; 277: 1109-1113Crossref PubMed Scopus (1366) Google Scholar). By contrast, Patched-2 is actually coexpressed with Hedgehogs and its transcription is independent of Hedgehog signaling (16St-Jacques B Dassule H.R Karavanova I Botchkarev V.A Li J Danielian P.S McMahon J.A Lewis P.M Paus R McMahon A.P Curr. Biol. 1998; 8: 1058-1068Abstract Full Text Full Text PDF PubMed Google Scholar). Mutants in Patched-2 have not been described. Several unexpected findings have come from studies on the ligands themselves. Hedgehogs undergo an autocatalytic processing that releases an active 19 kDa ligand with cholesterol covalently linked to its C terminus (Figure 1; 11Porter J.A Young K.E Beachy P.A Science. 1996; 274 (a): 255-259Crossref PubMed Scopus (1028) Google Scholar). This raises several questions. First, why cholesterol? Many other signals are lipid-modified, but so far Hedgehog is unique in undergoing cholesterol modification. Does cholesterol confer specific properties on Hedgehogs that are essential for their complex signaling activities? For example, Hedgehogs show polarized distribution within the cell, and cholesterol-rich rafts have been implicated in directed protein trafficking within the secretory pathway (reviewed in 15Simons K Ikonen E Nature. 1997; 387: 569-572Crossref PubMed Scopus (7747) Google Scholar). Indeed, recent work suggests that Drosophila Hedgehog is present in rafts (14Rietveld A Neutz S Simons K Eaton S J. Biol. Chem. 1999; 274: 12049-12054Crossref PubMed Scopus (235) Google Scholar). Thus, the cholesterol moiety could be involved in directing intracellular transport within epithelia. In addition to cholesterol, there is also evidence from cell culture studies for a degree of palmitoylation of Hedgehog (10Pepinsky R.B Zeng C Wen D Rayhorn P Baker D.P Williams K.P Bixler S.A Ambrose C.M Garber E.A Miatkowski K Taylor F.R Wang E.A Galdes A J. Biol. Chem. 1998; 273: 14037-14045Crossref PubMed Scopus (547) Google Scholar). How does cholesterol influence the movement of ligand, a key issue given the apparent long-range action of some Hedgehog signaling? In cell culture, cholesterol-modified Hedgehog remains bound to the cell surface and is not normally present in the medium, suggesting a very limited range of movement in vivo. Indeed, when an unmodified ligand is produced by simply inserting a termination codon into the primary transcript at the position where processing of its protein product would normally occur, Hedgehog targets are activated at a greatly increased distance from the source of ligand (12Porter J.A Ekker S.C Park W.J von Kessler D.P Young K.E Chen C.H Ma Y Woods A.S Cotter R.J Koonin E.V Beachy P.A Cell. 1996; 86 (b): 21-34Abstract Full Text Full Text PDF PubMed Scopus (412) Google Scholar). One caveat is that these experiments were all performed in the presence of wild-type protein. Consequently, it is unclear whether there is direct signaling by the cholesterol-deficient ligand, or a change in the distribution of the wild-type protein in the presence of the unmodified form of Hedgehog. Two interesting developments have been reported with regard to the movement issue. The first was the discovery in Drosophila of tout-velu, which encodes an enzyme involved in proteoglycan biosynthesis that is required for movement of cholesterol-modified, but not -unmodified, Hedgehog (19The I Bellaiche Y Perrimon N Mol. Cell. 1999; 4: 633-639Abstract Full Text Full Text PDF PubMed Scopus (301) Google Scholar). This finding suggests a role for proteoglycans in transcellular trafficking of cholesterol-modified Hedgehog (Figure 1). A second is the observation of long cytoplasmic processes (cytonemes) produced by cells within the Hedgehog target field. It is speculated that these play a role in the long-range transport of Hedgehog or other signaling molecules (13Ramirez-Weber F.-A Kornberg T.B Cell. 1999; 97: 599-607Abstract Full Text Full Text PDF PubMed Scopus (425) Google Scholar). If this is so, the ability of Hedgehog to attach to the membrane by its cholesterol anchor might be critical for cytoneme-mediated transport, increasing the distance over which Hedgehog acts. The role of cholesterol in Hedgehog signaling has become even more intriguing with the observation that Patched contains a region shared by a number of sterol-sensing proteins (reviewed by 9Osborne T.F Rosenfeld J.M Curr. Opin. Lipidol. 1998; 9: 137-140Crossref PubMed Scopus (40) Google Scholar). How does sterol-sensing relate to Patched activity? Cholesterol-modified and unmodified forms of Shh both appear to bind Patched with similar binding constants (10Pepinsky R.B Zeng C Wen D Rayhorn P Baker D.P Williams K.P Bixler S.A Ambrose C.M Garber E.A Miatkowski K Taylor F.R Wang E.A Galdes A J. Biol. Chem. 1998; 273: 14037-14045Crossref PubMed Scopus (547) Google Scholar). Thus, if the sterol sensing domain in Patched interacts with cholesterol on Hedgehog ligands, such an interaction is unlikely to contribute significantly to the affinity of Hedgehog for its receptor. Given the compartmentalized distribution of lipids within the cell, it is possible that this domain might modulate Patched trafficking. Alternatively, it has been suggested that sterol sensing may link Hedgehog action to the metabolic status of the cell (2Beachy P.A Cooper M.K Young K.E von Kessler D.P Park W.J Hall T.M Leahy D.J Porter J.A Cold Spring Harb. Symp. Quant. Biol. 1997; 62: 191-204Crossref PubMed Google Scholar). Clearly, this is another gray area in Hedgehog signaling. The latest surprise in the Hedgehog saga comes from a fascinating study reported in Cell last month (3Burke R Nellen D Bellotto M Hafen E Senti K.-A Dickson B.J Basler K Cell. 1999; 99: 803-816Abstract Full Text Full Text PDF PubMed Scopus (411) Google Scholar). A new screen in Drosophila for mutants in the Hedgehog pathway identified a novel gene that has been named dispatched, for reasons that will come apparent. dispatched mutants lacking both maternal and zygotic activity have a segment polarity phenotype identical to hedgehog mutants. As the reciprocal interactions between Wingless and Hedgehog signaling make it difficult to distinguish between these pathways in the segment, Burke et al. produced mutant clones in the wing disc, where their activities are distinct. The results clearly indicate a role for dispatched in Hedgehog signaling. Furthermore, dispatched is only required in cells that produce Hedgehog. hedgehog is coexpressed in the posterior compartment of most imaginal discs along with engrailed. By contrast, dispatched is expressed ubiquitously. However, expression of dispatched in the posterior compartment under engrailed regulation is sufficient to rescue most dispatched mutant phenotypes. The exception is the eye, where hedgehog expression is under noncompartmental regulation. Together these functional studies argue that in the absence of dispatched, insufficient Hedgehog signal passes to responding cells in the anterior compartment. Of course regulation in the sending cells could occur at several levels. Analysis of hedgehog transcription, processing, and sorting in imaginal discs of dispatched mutants, or within dispatched mutant clones within a wild-type disc, indicate no obvious defects in any of these processes. Interestingly, the levels of Hedgehog protein are actually elevated in hedgehog-expressing cells in dispatched mutants. Furthermore, the characteristic accumulation of Hedgehog protein within vesicles in Hedgehog-responsive cells in the anterior compartment is not observed in dispatched mutants. Thus, it appears that in the absence of Dispatched, normal levels of Hedgehog protein are produced, Hedgehog is processed and presumably cholesterol-modified, as modification is integral to processing, but Hedgehog fails to be released from posterior cells and instead accumulates in the posterior compartment. Remarkably, dispatched is predicted to encode a twelve-transmembrane protein containing a sterol-sensing domain, a distant relative of the Hedgehog receptor Patched and the recently identified Niemann-Pick Type C gene product, whose activity appears to be required for cholesterol trafficking in cells. Thus, a sterol-sensing protein is implicated in both sending and receiving Hedgehog signals. What is the function of Dispatched? The authors address this question in the broader context of the significance of cholesterol modification to Hedgehog signaling. In an interesting set of experiments, they addressed the range of action and activity of various modified and unmodified forms of Hedgehog in the presence or absence of Dispatched. First, they demonstrated a dramatically increased range of action for unmodified Hedgehog in experiments in which wild-type Hedgehog protein was absent (employing a temperature-sensitive mutant). Thus, the extended Hedgehog signaling seen previously when unmodified protein was present along with wild-type Hedgehog (12Porter J.A Ekker S.C Park W.J von Kessler D.P Young K.E Chen C.H Ma Y Woods A.S Cotter R.J Koonin E.V Beachy P.A Cell. 1996; 86 (b): 21-34Abstract Full Text Full Text PDF PubMed Scopus (412) Google Scholar) was most likely not an indirect consequence of changes in the distribution of cholesterol-modified Hedgehog. The conclusion drawn from the current studies is that Patched is unable to sequester Hedgehog effectively when Hedgehog is lacking its cholesterol modification, even though signaling, which also requires Hedgehog binding to Patched, is clearly active. What distinguishes binding from sequestration is not clear, but the data suggest that cholesterol is important. Next, the authors determined that Dispatched is not required for signaling by unmodified Hedgehog. Finally, they expressed modified hedgehog genes encoding a transmembrane and GPI-anchored form of the protein. In the wild-type wing disc, both these forms are capable of signaling, but only to their immediate neighbors. Thus, unlike cholesterol-modified Hedgehog, both of these membrane-bound ligands appear to be fixed to the surface of expressing cells. Signaling was unaltered in the absence of Dispatched. Therefore, it would appear that Hedgehog can be tethered to the membrane by a transmembrane domain or GPI anchor and is still able to signal effectively, whether or not Dispatched is present, albeit with very limited range. These experiments also indicate that not all lipid modifications behave similarly. GPI-anchored Hedgehog only acts locally even in the presence of Dispatched, suggesting some specificity in the action of Dispatched for cholesterol-tethered Hedgehog. Given these results, there are two reasonable models for Dispatched activity (Figure 1). First, Dispatched might be required for trafficking of cholesterol-modified Hedgehog through the secretory pathway so that ultimately an active form of the protein arrives at the cell surface. Here, the possible involvement of cholesterol rafts in Hedgehog transport and putative sterol-sensing properties of Dispatched are intriguing. However, the authors do not observe any clear trafficking defect, and some Hedgehog protein is on the surface of hedgehog-expressing cells in dispatched mutants. What is not clear is how much relative to wild type as the immunostaining of nonpermeabilized imaginal discs was nonquantitative. dispatched mutant discs do show some upregulation of Patched at the compartment border, suggesting that a low level of active Hedgehog signal is present. In an alternative model, cholesterol-modified Hedgehog might need to be displaced from the lipid bilayer for effective signaling. It is apparent from the naming of the gene that the authors prefer this model. However, both the transmembrane and GPI-modified proteins do signal, even though neither appears to be released from the cell surface. Thus, it seems likely that Dispatched would release cholesterol-modified Hedgehog at an earlier stage in the secretory process. In conclusion, the authors have demonstrated that there are special mechanisms unique to the handling of cholesterol-modified Hedgehog in the sending cell. Further progress is likely to require a thorough investigation of the fate of different forms of Hedgehog in the secretory pathway. Given the molecular similarities between Patched and Dispatched, it is likely that there are common principles behind sending and receiving a Hedgehog signal. Another piece in the Hedgehog puzzle has been identified. All the indications are that the final solution to this puzzle will be well worth the effort. The author regrets that, owing to severe space limitations, key pieces of work have been cited using reviews rather than the original articles.