The Patched Receptor: Switching On/Off the Hedgehog Signalin

Posted: Sat Apr 15, 2006 3:59 pm

The Patched Receptor: Switching On/Off the Hedgehog Signaling Pathway

Luis Quijada
Ainhoa Callejo
Carlos Torroja
Isabel Guerrero
Corresponding author: Isabel Guerrero—Centro de Biología Molecular “Severo Ochoa”, Universidad Autónoma de Madrid Cantoblanco. E-28049, Madrid, Spain.

The activities of the Hedgehog (Hh) family of proteins are central to the growth and patterning of developing tissues and organs in many different organisms. Hh proteins are secreted ligands synthesized in discrete regions. The receptor of Hh is Patched (Ptc) and it is expressed in the cells close to the source of Hh. Ptc binds the ligand and transduces a signal which is modulated depending on the context and the concentration of Hh received. Hh and several molecular components of the pathway were first identified and characterized in Drosophila, providing relevant milestones to our understanding on how the Hh signal is transduced. However, important gaps in the pathway still need to be elucidated. Some of these gaps converge on the Ptc receptor and its intriguing mechanisms of Hh reception and signal transduction. Mutations of Ptc that prevail both in animal and human populations are giving some clues on crucial aspects of its function. Patients bearing mutated forms of Ptc suffer a variety of serious diseases. Molecular and cellular studies in Drosophila have given us a clue of the function of Ptc receptor such as the normal topology and/or sorting of the receptor. Thus, a widened knowledge of the function of Ptc might help to design specific therapies for these disorders. This chapter focuses on recent advances that shed some light on how Ptc may operate in the cell.

The Hedgehog Signaling Pathway

Several genes of the Hh pathway were first identified in the fly Drosophila melanogaster and later in vertebrates. Many of the names of the genes involved in the Hh signaling pathway were originally descriptive of the phenotype manifested in mutant Drosophila larvae. Wild-type larvae show a clearly segmented pattern due to alternate bands of denticles in ventral position, whereas the inter-band space is naked. During a screening for mutations that affected the segmental pattern, Nüslein-Volhard and Wieschaus described a group of mutants with alterations in the patterning of the segments. 1 Instead of the wild-type alternate belts of denticles and naked cuticle, Hh mutants showed a continuous lawn of denticles, which gave the larva a resemblance to a hedgehog (fig.1B), and Ptc mutants showed patches of denticles (fig.1C).

Figure 1. Two model systems in Drosophila melanogaster: The larva cuticle and the wing imaginal disc. A) Wild-type ventral larva cuticle showing its regular pattern of denticles. B) Hh mutant ventral cuticle. C) Ptc mutant ventral cuticle. D) Adult wing. The dashed line marks the boundary between the anterior (a) and posterior compartment (p). These compartments are already present in the wing imaginal disc, the wing primordium (E; front and lateral view) in the larva. Hh is produced in the posterior compartment (green) and signals to cells in the anterior compartment. Ptc expression is induced by Hh signaling (red). A color version of this figure is available online at http://www.Eurekah.com.

The active Hh protein is a peptide that has undergone an autocatalytical processing in which the peptide is also modified at its N- and C-termini by palmitoyl and cholesterol adducts, respectively. Hh is secreted by discrete subsets of cells within a developing organ in a process that seems mediated, at least in part, by the transmembrane protein Dispatched (Disp), which has a structure very similar to Ptc. In the wing imaginal disc of Drosophila, Hh is expressed in the cells of the posterior compartment and signals to the cells of the anterior compartment, which in turn express the Ptc receptor (fig.1D,E). A graded and short-range response to Hh signaling occurs in the Hh receiving cells (fig.1E). Ptc, a 12-transmembrane protein, is a negative receptor because it keeps the Hh pathway silenced in its unliganded state. 2 In the absence of Hh, Ptc suppresses the activity of Smoothened (Smo), a 7-transmembrane protein that is the positive modulator of the pathway. Although Hh does not bind to Smo, without Smo there is no signaling. The target cells bind Hh by Ptc, which results in the activation of Smo, and, subsequently, in the activation of transcription factors: the Cubitus interruptus protein (Ci) in Drosophila and the orthologous Gli proteins in mammals (reviewed in ref. 10). One peculiarity of the Hh signaling pathway is that Ptc is up-regulated in response to increasing amounts of Hh. 3 - 7 Therefore, Hh controls both its own activity and its own spreading expressing high levels of the receptor. 8 By this means, a morphogenetic gradient is formed in the Hh receiving cells. Theoretical analysis of model systems built to explain how a morphogen gradient is formed indicates that this feedback is indeed required to give robustness to the gradient while maintaining its range of action. 9 Thus, according to one model, it is expected that fluctuations in ligand production have minor effects on the slope of the gradient, compared to a model in which the receptor has a constant, or ligand-independent level of expression. 9

Since this chapter is particularly centered in the Ptc receptor, for a more general view on the Hh response network, the reader is referred to some other recent and exhaustive reviews. 10 - 14

Although this sequence of events is well established, many molecular mechanisms connecting these and other components of the pathway remain elusive. Focusing in the case of the Ptc receptor, several fundamental questions remain unanswered: How does the Ptc receptor recognize the Hh ligand? What happens after the receptor-ligand complex is formed? Is it internalized or does it remain in the membrane before the signal is transduced? How does the unliganded Ptc receptor repress the Smo activity? And how does the liganded Ptc derepress Smo activity?

Sequence and Functional Analysis of Ptc

A single ptc gene is present in Drosophila, whereas two are present in vertebrates (PTCH1 and PTCH2 in humans; Ptc1 and Ptc2 in mice) (reviewed in ref. 10) and several ptc-related genes are found in the nematode worm Caenorhabditis elegans. 15 Ptc shows homology to bacterial proton-driven transmembrane molecular transporters, 16 , 17 and presents a conserved a Sterol-Sensing Domain (SSD) that overlaps between transmembrane segment 4 to 6, and also two extracellular large loops (reviewed in ref. 18). In general, SSDs are thought to function as a regulatory domain involved in linking vesicle trafficking and protein localization with processes such as cholesterol homeostasis and cell signaling. 18 Concretely, the SSD of Ptc seems to mediate the intracellular trafficking of Ptc, a process that might be essential to regulate Smo activity. 19 , 20 Other functional studies have assigned particular roles to regions of the protein with no obvious similarity to conserved domains. Thus, the two extracellular large loops are required for Hh binding, 21 the cytoplasmic C-terminus has been involved in the transduction of the signal, 22 and an intracellular small loop between the SSD and the following transmembrane domain has been reported to interact with cyclin B1 to regulate cell-cycle progression in vertebrates. 23 However, despite this emerging body of data, we still do not know the full sequence of molecular events that involves the reception of the Hh ligand and how Ptc regulates Smo.

PTCH2 is a 12-transmembrane protein as well and the two large extracellular loops and an SSD are also conserved (reviewed in ref. 24). However, both amino- and carboxy-termini are different from PTCH1. Similar differences are found between Ptc1 and Ptc 2 in mice. PTCH1 and PTCH2 proteins are expressed differentially during the development of the epidermis, suggesting specific roles for each protein, 25 although the function of PTCH2 remains unclear to date. A recent study suggests that PTCH2 may act as a Hh receptor that tunes finely the signaling in various cellular environments. 26

Hedgehog Lipid Modifications and Morphogen Distribution

The lipid modifications on the Hh protein appear to regulate its activity and distribution. Hedgehog proteins undergo two sequential lipid modifications during their posttranscriptional maturation. Following cleavage of an N-terminus signal sequence upon entry into the secretory pathway, Hh proteins undergo an autoprocessive reaction that results in an internal cleavage inside the 45-KDa precursors, between glycine-cysteine residues from a conserved GCF motif. After this, a cholesterol molecule is covalently added at the newly generated C-terminus of the proteins. 27 In Drosophila, a construct of Hh lacking the cholesterol moiety is active in signaling but not appropriately restricted spatially in its signaling activity. 28 Therefore, the cholesteryl moiety restricts the spatial deployment of the signal via insertion into the lipid bilayer of the cell membrane and also functions as an essential molecular handle for a proper intracellular and extracellular trafficking and localization of the signal (reviewed in refs. 29, 30). In addition, Hh proteins are palmitoylated on a highly conserved amino-terminal cysteine residue, the first of a pentapeptide CGPGR. 31 In Drosophila this palmitoylation is a total requirement in order to produce a fully active Hh signal. Although in vitro studies in vertebrates indicated that Hh mutant forms lacking the pamitoyl adduct retained significant activity, 32 , 33 knockout mice deficient in Skn (the murine ortholog of the Drosophila Ski, which catalyzes Hh palmitoylation) showed that Hh acylation is absolutely required for Hh long-range signaling. 24

In summary, dual lipidation of Hh protein promotes membrane affinity and allows its association to sterol-rich membrane microdomains in Drosophila, and to lipid rafts in mammalian cells, that function as platforms for intracellular sorting and signal transduction. 34 The close association of Hh proteins to the plasma membrane due to the lipid modifications could provide a mechanism for restricting the range of their activities.

Reception of Hh

Biochemical studies have shown that Hh monomers can form multimeric complexes in which the hydrophobic moieties have been proposed to be sequestered in the interior of the multimer, making the complex soluble and diffusible. 35 , 36 This could be relevant to the movement of Hh through the extracellular matrix 37 and to the reception of Hh by Ptc. On one hand, the restricted diffusion of Hh through the extracellular matrix might require the formation of these multimeric complexes of lipidated Hh. On the other hand, a multimeric Hh complex might have either a higher intrinsic affinity for Ptc, or maybe is capable of eliciting greater biological responses, for instance through receptor oligomerization, or binding to coreceptor proteins. In this context, structural studies on transmembrane transporters related to Ptc determined that that proteins operate as homotrimers. 38 Interestingly, a mutant version of Ptc that internalizes inefficiently is, however, localized to intracellular vesicles when coexpressed with a dominant negative Ptc. 39 This ptc inter-allelic complementation strongly suggests that Ptc has an oligomeric structure.

Several other proteins have been identified as capable of binding to Hh, raising the question of whether they are coreceptors or adjuvants involved in the reception of Hh. Especially, recent analysis in Drosophila have illustrated the critical roles of heparan sulfate proteoglycans (HSPGs) in developmental signaling pathways (reviewed in ref. 40). These large macromolecules are found at the cell surface and form part of the extracellular matrix. A major feature of these proteins is the attachment of long unbranched chains of repeating and sulfated disaccharides to specific serine residues in their protein core. Dally and Dally-like (Dly), two Drosophila glypican members of the HSPG family, has received great attention. A specific role in Hh signaling in the embryonic epidermis has been ascribed to Dly, 41 while both Dally and Dly seem to be functionally redundant in the wing disc, 42 indicating distinct activities of these two glypicans in the embryo and the wing disc. Hh signaling might also be regulated by other HSPGs in different tissues or developmental processes. For example, Trol, the Drosophila ortholog of Perlecan, seems to form a complex with Hh, and mutations in the gene encoding it cause neuroblasts to undergo cell cycle arrest in the larval brain. 43 , 44 In addition, trol is required for the neuroblast division induced by Hh. 43

Two data favour a model in which the polyanionic branches of the HSPGs would act mainly as a molecular trap to keep the morphogens in touch with epithelium surfaces. First, the strongest loss-of-function phenotypes are achieved with mutants affecting the synthesis of sugar chains of the HSPGs; 42 and second, the fact that the sugar chains seem to be directly involved in the movement of other morphogens. 45 , 46 Hence, the HSPGs might restrict the morphogen diffusion to the environment of the extracellular matrix, and thus preventing a massive spreading to cavities such as the imaginal disc lumen. This lateral distribution of Hh would facilitate the encounter with Ptc. This scenario does not rule out the possibility of specific regulations involving Dally or Dly. In fact, in tissue culture experiments, Dally-like protein (Dlp), but not Dally was required for Hh signaling. 47 In addition, it has been shown that Dlp is specifically required for Hh signaling in the embryonic epidermis. 41 A Dlp-mediated regulation of Hh signaling might involve Dlp acting as a coreceptor, perhaps by transferring Hh to Ptc, or by forming a Hh-Dlp-Ptc interaction or stabilizing a Hh-Ptc complex (reviewed in ref. 48). In the case of the Wingless pathway (another morphogenetic signal), it has been elegantly shown that Dlp is cleaved specifically from the cell surface by Notum, a secreted enzyme, resulting in an adequate regulation of the pathway. 49

Hh Internalization and Signal Transduction

After receiving Hh, the receptor-ligand complex is internalized and the morphogen is thus depleted from the extracellular milieu. It was proposed that this internalization has a role shaping the Hh gradient. 8 , 50 This hypothesis has been reinforced by experiments that uncoupled the Hh sequestration and the Hh signaling using a Ptc mutant that is unable to sequester Hh and yet retains a normal capacity to mediate Hh signaling. 39 Furthermore, the ability to uncouple these functions suggests that Ptc-mediated internalization of Hh does not play a major role in the transduction of the signal but serves mainly to limit and control the Hh gradient. 39

The analysis of different ptc mutants (fig.2)

Figure 2. Ptc mutations. Ptc is a 12-transmembrane protein. The transmembrane domains highlighted in green correspond to the sterol-sensing domain (SSD). Mutations ptcP83, ptcN15, ptcQ67 (see ref. 51) and ptcCon (see ref. 72) affect both Hh sequestration and signaling. Mutations ptc13, ptc34 (see ref. 20), ptcS2 (see ref. 19, 20), ptc1130 (See ref. 20, 22), ptcG500V, ptcD504Y (see ref. 52) and ptcG20 (unpublished data and ref. 73) affect only signaling. In the ptc 14 mutant only the sequestration is affected.

in Drosophila have demonstrated that the two functions of Ptc, sequestration and signalling, can be genetically separated. 8 , 19 , 20 , 22 , 39 , 51 PtcS2, Ptcm 13 , Ptc 34 and Ptc1130 are defective for Hh signal transduction but sequestration properties are normal, 19 , 20 , 22 while the Ptc 14 mutant protein is defective for Hh sequestration, but normal in terms of Hh signal transduction. 39

One of the most intriguing aspects of the Hh pathway is how Ptc regulates Smo activity. The mechanism is most probably indirect, since Ptc and Smo do not need to bind or to colocalize to control signaling and, in addition, Ptc acts substoichometrically to inhibit Smo, possibly through changes in the distribution or concentration of an as yet uncharacterized small molecule (fig.3).

Figure 3. Model of Hh signal activation. 1) In the absence of Hh (left panel), Ptc is constantly recycling from the plasma membrane to vesicles inside the cells. Occasionally, these vesicles enter the proteolytic pathway. 2) At the same time, Smo is kept inactive in an intracellular compartment, unable to signal. Smo enters as well the degradative pathway under this situation. 3) Upon the reception of Hh (right panel), the receptor-ligand complex is internalized, contributing to the slope of the gradient of Hh. 4) At the same time, Ptc no longer represses Smo, which is stabilized and traffics to the plasma membrane preferentially. This plasma membrane stabilization of Smo correlates with signal activation.
52 A recent study 53 suggests a mechanism by which the ratio of unliganded to liganded Ptc determines the cellular response (reviewed in ref. 54). In this context, several alternative models are possible. For example, unliganded Ptc might import a Smo antagonist, while liganded Ptc might promote its export. Alternatively, the functional Hh receptor may comprise a multimer of Ptc proteins and the binding of Hh to one Ptc subunit in a receptor complex may block the ability of the multimer to inhibit Smo activitity. 53 , 54

Smo Regulation by Ptc

When Hh binds Ptc, Smo is somehow liberated to start the process of activating target gene transcription. The mechanism in which Ptc inhibits and Hh activates Smo protein remains unknown. Smo is transcribed in a generalized pattern not transcriptionally regulated by the Hh signal. Smo protein is posttranscriptionally regulated by phosphorylation, becoming more stable, and by moving to the cell surface after a Hh signal is received. 55 - 60 A detailed study has shown that Smo accumulates in the plasma membrane of cells in which Ptc activity is abrogated by Hh but is targeted to the degradative pathway in cells where Ptc is active. 61 Thus, Ptc could regulate Smo activity through inhibition of its accumulation in the plasma membrane, targeting instead to lysosomal compartments. In this view, Ptc might regulate Smo activity simply by modulating the levels of protein present in the cell. Alternatively, it may be that sub-cellular location of Smo is critical for its activation, the plasma membrane providing an environment for Smo to be accessible to an intracellular agonist (see fig.3 for details). 61

Ptc and Human Disease

Genetic studies in various model organisms are beginning to elucidate the factors that are likely candidates for the causes of early embryonic defects in humans (reviewed in refs. 62, 63). Thus, detailed knowledge of the Hh signaling pathway is fundamental to an understanding of vertebrate development as well as several birth defects in humans.

Since the unliganded Ptc receptor exerts a repression on the Hh pathway, mutations affecting Ptc are frequently associated to a constitutive activation of the Hh signaling pathway. This is in agreement with the fact that most of the mutations characterized to date in PTCH1 result in protein truncation. 24 The spectrum of human PTCH1 mutations also includes deletions, insertions, splice site alterations, and nonsense and missense mutations distributed throughout the gene (for an example, see Table 1). 64 - 66

PTCH1 deregulation in the epidermis is sufficient to induce Basal Cell Carcinomas (BCCs) of the skin. 30-40% of Gorlin syndrome patients (GS; also known as Nevoid basal cell carcinoma syndrome) have familial loss-of-function mutations in the PTCH1 gene. GS is an autosomal dominant disease with nearly complete penetrance and variable expressivity because mutations in PTCH1 seem to be haploinsuficient in humans (a mutation in only one of the two alleles is enough to show a phenotype). Clinically, GS patients present congenital abnormalities that includes skeletal defects (polydactyly, fused or bifid ribs), early onset of multiple BCCs and an increased rate to develop other tumors, including medulloblastomas of the cerebellum (reviewed in refs. 67-69). A limited number of mutations in PTCH1 have been linked to Holoprosencephaly (HPE). 70 HPE affects the forebrain and face to various degrees, from the most extreme lethal alobar type to milder microforms that include small midline facial defects. However, since HPE arises from loss of SHH signaling, these mutations might rather reflect an impaired ability of PTCH1 to interact with SHH, thus permanently shutting down the pathway. In fact, two out of four rare PTCH1 missense mutations that have been reported to be associated to HPE were localized in the extracellular loops of PTCH1 required for SHH binding. 70

There is a limited number of missense mutations described for PTCH1. Missense mutations do not occur as frequently as frameshift or nonsense mutations. In a clinical study on French patients, missense mutations span the second group of six transmembrane domains of the protein. 64 Unfortunately, no data are available to interpret the pathologic value of the missense mutations described in humans. 64 So far, only the molecular and cellular studies of Drosophila ptc mutations indicate an alteration of the correct topology and/or sorting of the Ptc receptor. We propose Drosophila as an ideal model system to analyze the functional alterations of the missense mutations found in the PTCH1 receptor. In fact, a recent study analyses in the fly the function of an artificially mutated form of Ptc analog to a mutated PTCH1 previously characterized in GS patients. 71 Therefore, understanding the molecular mechanisms of Hh signal reception will allow identifying potential drug targets in order to devise strategies for the treatment of BCC.

Concluding Remarks

More genes of the Hh pathway are going to be uncovered in the near future and Drosophila genetics will undoubtedly help in to resolve this issue. Hence, concerning the function of Ptc receptor, an unresolved question is how Ptc is able to repress the activity of Smo, and whether there are intermediate elements. Although other molecules have been found as possible candidates in the reception of Hh in vertebrates such as Megalin, Hip or Gas1, mutagenesis screens need to be done in Drosophila to find more proteins implicated in Hh reception. The study of their function will help to understand how several cellular and molecular processes drive and regulate the pathway. Another important unresolved issue in the Hh pathway is the role of cell polarity in the transduction of the signal, or the endocytic and exocytic routes involved. Also questions remain regarding how a morphogen gradient is formed, how the cytoskeleton affects the pathway, or how the transcription of target genes is regulated. New model systems in vertebrate genetics, cellular biology and biochemistry will help to elucidate some of these questions. Definitely, a huge biochemical and cell biology effort needs to be done in order to solve the role of this kind of proteins in Hh signaling pathway. This knowledge will undoubtedly help to design specific therapies to serious human diseases caused by mutations on genes of this signaling pathway.
Acknowledgments

We thank Cristina Guerra-Giraldez for critical reading the manuscript. LQ is supported by an I3P contract of the CSIC; AC is supported by Comunidad Autónoma de Madrid, grants (08.6/0045/2001.2 and GR/SAL/0185/2004); and CT was supported by by the Spanish D.G.I.C.Y.T, grant (BMC2002-03839). Also, an institutional grant from Fundación Ramón Areces is acknowledged.
http://www.ncbi.nlm.nih.gov/books/bv.fcgi?rid=eurekah.section.44405