Mutations in the sterol-sensing domain of Patched suggest a role for vesicular trafficking in Smoothened regulation

608 Brief Communication

Mutations in the sterol-sensing domain of Patched suggest a role for vesicular trafficking in Smoothened regulation H. Strutt*, C. Thomas*, Y. Nakano, D. Stark, B. Neave, A. M. Taylor and P. W. Ingham The tumor suppressor gene patched (ptc) encodes an approximately 140 kDa polytopic transmembrane protein that binds members of the Hedgehog (Hh) family of signaling proteins and regulates the activity of Smoothened (Smo), a G protein–coupled receptor-like protein essential for Hh signal transduction. Ptc contains a sterol-sensing domain (SSD), a motif found in proteins implicated in the intracellular trafficking of cholesterol, and/or other cargoes. Cholesterol plays a critical role in Hedgehog (Hh) signaling by facilitating the regulated secretion and sequestration of the Hh protein, to which it is covalently coupled. In addition, cholesterol synthesis inhibitors block the ability of cells to respond to Hh, and this finding points to an additional requirement for the lipid in regulating downstream components of the Hh signaling pathway. Although the SSD of Ptc has been linked to both the sequestration of, and the cellular response to Hh, definitive evidence for its function has so far been lacking. Here we describe the identification and characterization of two missense mutations in the SSD of Drosophila Ptc; strikingly, while both mutations abolish Smo repression, neither affects the ability of Ptc to interact with Hh. We speculate that Ptc may control Smo activity by regulating an intracellular trafficking process dependent upon the integrity of the SSD. Address: MRC Intercellular Signalling Group, Centre for Developmental Genetics, Department of Biomedical Science, University of Sheffield, United Kingdom. * These authors contributed equally to this work.

Smo (reviewed in [1]); how Ptc exerts this repression is unclear. We therefore sought to identify and characterize mutations that inactivate key functional domains of the protein. Mutations of Ptc that disrupt its topology and/or trafficking to the plasma membrane might be expected to abolish both the Hh binding activity and the Smo repressing activity of the protein. On the other hand, lesions that specifically abolish the Smo-repressing activity of Ptc could in principle leave its Hh binding ability intact; indeed, genetic analysis suggests that the ptcS2 allele encodes just such a mutated form of the protein [2]. To discriminate between these two types of mutation, we undertook a systematic analysis of the subcellular distribution of Ptc and Hh proteins in embryos homozygous for extant and newly induced Ptc loss-of-function mutations. To this end, we raised a monoclonal antibody (mAb 5E10) against the most N-terminal portion of the Drosophila Ptc protein (see Materials and methods). In wild-type embryos, Ptc protein shows a striking subcellular distribution [3,4], the highest accumulations being found in punctate structures within the cell (see Figure 1a). Previous studies have suggested that at least some of these structures correspond to multivesicular bodies and endosomes [3]. In cells located close to those that secrete Hh, some of these Ptc-positive vesicles also label with anti-Hh ([5,6]; see Figure 1b); this finding is consistent with Ptc mediating the endocytosis of Hh [7]. Notably, however, punctate accumulations are found in all Ptcexpressing cells (Figure 1a), and this observation suggests that the protein cycles to and from the membrane independently of any interaction with Hh.

Results and discussion

In embryos homozygous for the ptcS2 allele, the distribution of mutant Ptc protein is similar, though not identical, to that seen in the wild type (Figure 1c); the protein is found predominantly in the same characteristic punctate accumulations and colocalizes with Hh in a subset of these; these results are consistent with its binding and endocytosing Hh. Only two other alleles, ptc13 and ptc34, encode proteins that similarly colocalize with Hh (Figure 1d,e). We conclude that each of these alleles specifically inactivates the Smo-repressing activity of Ptc without compromising its ability to bind and endocytose Hh. We note that the distribution of the mutant protein in ptc13 differs from that in ptcS2 and ptc34, with more diffuse cytoplasmic staining being apparent in the former.

The prevailing model of Hh signal transduction postulates that Hh acts by abrogating the repressive effect of Ptc on

By contrast, in embryos homozygous for the majority of

Correspondence: Philip W. Ingham E-mail: [email protected] Received: 7 February 2001 Revised: 13 March 2001 Accepted: 16 March 2001 Published: 17 April 2001 Current Biology 2001, 11:608–613 0960-9822/01/$ – see front matter  2001 Elsevier Science Ltd. All rights reserved.

Brief Communication 609

Figure 1

ptc mutant alleles, there is no colocalization of Ptc and Hh (Figure 1g–j), and the distribution of Hh protein appears more diffuse than in wild-type or ptc13 ptcS2 and ptc34 embryos. The levels of Ptc vary between alleles, but in each case (except ptc16) the protein is predominantly perinuclear (Figure 1f–i). Although some punctate accumulations are still apparent, these are less numerous and less intense than those seen in wild-type or ptcS2, ptc13, and ptc34 embryos. We note that in embryos homozygous for the ptc16 allele (previously characterized as a protein null [8]), diffuse cytoplasmic staining is detected by the mAb 5E10 (Figure 1j). Using SSCP and/or sequence analysis, we found that four of the mutations that abolish colocalization with Hh are associated with nonconservative substitutions of highly conserved residues; one of these, ptc14, is in the first transmembrane domain, while the other three are in the large extracellular loops (see Figure 2a). Although these loops have been implicated in Hh binding [9], we consider it most likely, on the basis of the observed distributions of the mutant proteins, that these substitutions disrupt the normal topology and/or sorting of the protein and hence its ability to bind and endocytose Hh. In the case of ptc9, the S809N substitution removes a putative glycosylation site that is conserved in all Ptc proteins, a defect that could disrupt folding of the protein and hence its transport from the ER [10]. The remainder of the mutations that eliminate Hh binding are associated with premature termination codons predicted to generate truncated forms of the protein (Figure 2b). These range from ptc16, which retains just part of the putative intracellular N-terminal domain of the protein, to ptc47, which retains both large extracellular loops required for Hh binding but lacks the C-terminal tail and the last three transmembrane domains.

Sub-cellular distribution of Ptc and Hh in wild type and ptc mutant embryos. All panels show high-magnification confocal images of segments from the abdominal regions of Stage 10 embryos stained with (a,f) mAb 5E10 (green) and propidium iodide (red) to reveal distribution of Ptc protein and nuclei respectively or (remaining panels) mAb 5E10 (green) and polyclonal anti-Hh antisera (red). (a) Wild-type embryo. Note the diffuse cytoplasmic distribution and large punctate accumulations of Ptc within each cell. (b) In wild-type embryos Ptc and Hh accumulate in characteristic punctate structures, some of which coincide (yellow) at the interface between the two cell populations. Similar punctate distributions and colocalizations are apparent in embryos homozygous for the missense mutations (c) ptc34, (d) ptcS2, and (e) ptc13. In no other case that we analyzed did we observe such colocalization of Ptc and Hh. In most cases — including all other missense mutations, exemplified by (f,g) ptc37 and

Significantly, of the three mutant forms of the protein that retain the property of colocalizing with Hh, one, ptc13, is associated with a substitution of a conserved glutamic acid in the putative cytoplasmic C-terminal tail, while the other two are associated with nonconservative substitutions in the sterol-sensing domain (SSD). This finding argues against the previously suggested roles for the SSD in mediating Hh binding [6] or in targeting the Ptc-Hh complex to the lysosome [14]. In the case of ptc34, a highly conserved glycine in the third transmembrane domain is replaced by an arginine, while in ptcS2 a highly conserved aspartic acid at the intracellular boundary of the sixth

(h) ptc9 shown here, as well as most of the deletion alleles, as exemplified by (i) ptc47, Ptc protein is predominantly perinuclear and accumulates in few if any punctate structures. The one exception is ptc16, which is predicted to encode a short fusion protein containing only the first 43 amino acids of Ptc; ( j) this shows a diffuse cytoplasmic distribution.

610 Current Biology Vol 11 No 8

Figure 2

Sequence analysis of ptc loss-of-function mutations. (a) Schematic representation of the Ptc protein showing the location and nature of the seven missense mutations identified in this study. The light shading indicates the SSD. In assessing the likely significance of any polymorphism detected by sequence analysis, we made reference to the highly related sequence of the D. virilis ptc gene [29], as well as to the various vertebrate Ptc-1 sequences available in the databases (see Supplementary material); in most cases we were also able to make a direct comparison with the parental ptc alleles, which have been maintained as balanced stocks associated in coupling with other lethals since the original mutagenesis experiments. Because no parental chromosome was available for the ptcS2 allele, we

sequenced regions of genomic DNA from this allele that encode the entire ORF. Twelve polymorphisms were identified, three of which result in amino acid substitutions; of these, two are conservative substitutions of residues that vary between species (I650S and V1129M). The only nonconservative change is caused by a G-to-A transition at nucleotide 1750 that results in the replacement of an acidic residue D by an uncharged residue N at position 584 within the SSD. (b) Graphical representation of the truncated proteins predicted to be encoded by the 11 ptc alleles associated with premature termination codons. The numbers indicate the 12 transmembrane domains (dark shading) in the full-length protein.

transmembrane domain is replaced by an asparigine residue (see the legend to Figure 2 for details of the sequence analysis). Notably, this latter substitution is identical to that identified as the cause of a gain-of-function mutation in the SREBP cleavage-activating protein (SCAP) [15]. The fact that the same substitution can generate opposite types of mutations in the two proteins could presumably reflect the way in which the SSD is deployed in each case. Consistent with this, other substitutions that cause gain of function in SCAP have been shown to result in a loss of function when they are introduced into the SSD of the NPC-1 protein [16].

are accompanied by a thickening of the veins in this region (Figure 3). A similar phenotype is seen in trans heterozygous combinations between ptctuf and each of the nonsense and most of the missense alleles that we have characterized (Figure 3). Significantly, however, ptcS2, ptc13, and ptc34, the three alleles that retain Hh binding, all exhibit much stronger wing phenotypes when in trans to ptctuf. In the case of ptcS2/ptctuf, the wings are more rotund than those of wild-type or hemizygous ptctuf flies and have distinctive mirror image duplications in the anterior-proximal region of the wing [18] (see Figure 3). Animals of the genotype ptc34/ptctuf and ptc13/ptctuf exhibit even stronger phenotypes; in both cases, the wings become almost circular with the triple-row bristles, characteristic of the anterior margin, being replaced by double-row bristles, typical of the medial-distal margin (Figure 3). These phenotypic effects are indicative of substantial ectopic derepression of the Hh pathway (data not shown); the fact that they are stronger than those seen in ptctuf hemizygotes suggests that each of these mutant forms of Ptc acts as an antimorph and competitively inhibits the reduced levels of wild-type Ptc protein expressed by the ptctuf allele.

While the effects of the different types of ptc alleles on cell patterning are indistinguishable in homozygous embryos [17] (see also Supplementary material available with this article on the internet), we found a striking heterogeneity in their effects on the patterning of the wing when the ptc alleles were in trans to the weak hypomorphic allele ptctuf. Molecular analysis of this allele has previously identified an insertion upstream of the transcription start site [8, 17], and this finding suggests that the mutant phenotype is caused by a reduction in the levels of wild-type Ptc protein. Hemizygous ptctuf animals complete embryogenesis and develop into adults that exhibit mild abnormalities in the anterior compartment of the wing. These abnormalities usually include a small bulge in the costal region and

To investigate the properties of these mutations further, we introduced base changes identical to those found in ptcS2 and ptc13 (see Figure 2a) into the full-length wild type ptc cDNA and assayed the effect of the misexpression of

Brief Communication 611

Figure 3 Phenotypic variation between different mutant ptc alleles. Wild-type (⫹) wings display a stereotypic venation pattern and are characterized by an anterior margin (top) that is decorated with stout sensory bristles. Wings of flies homozygous (tuf1) or hemizygous (G12) for ptctuf or trans-heterozygous for ptctuf and the missense mutations ptc9, ptc14, ptc17, and ptc37 (9,14,17,37) and all nonsense mutations, exemplified here by ptc15 ptc16, ptc47 ptc35, and ptc40 (15,16, 31, 35, 40) exhibit varying degrees of distortion in the costal (anterior proximal region), which is indicative of ectopic activity of the Hh pathway, and an increase in the distance between the third (L3) and fourth (L4) longitudinal veins, which is indicative of an increase in the range of the Hh protein and is thus consistent with a reduction in sequestration of the Hh protein. By contrast, wings from animals transheterozygous for ptctuf and the missense mutations ptcS2, ptc13, and ptc34 show normal separation of L3 and L4, which indicates normal levels of Hh sequestration, but display dramatic shape changes and abnormalities of the anterior margin. Such abnormalities range from mirror image duplications of the proximal half of the anterior wing (S2) to complete replacement of the stout triple row of bristles along the anterior margin by thinner bristles typical of the medial-distal margin (13, 34).

these mutated forms of Ptc in transgenic flies. Previous studies have shown that misexpression of the wild-type Ptc protein can disrupt normal wing development, either by sequestering Hh protein (when expressed in Hh secreting cells — see Figure 4) or suppressing Smo activity (when expressed in Hh responding cells – see Figure 4) [19, 20]. On the other hand, the misexpression of Ptc in cells that neither secrete nor respond to Hh has no effect on wing development or Hh target gene expression. We found that the misexpression of the mutant forms of the protein in the posterior compartment has a similar effect to wild-type Ptc (Figure 4), and this is consistent with their retaining Hh binding activity as indicated by our colocalization analysis. When misexpressed in Hh-responding cells, however, both mutant forms cause a thickening of the L3 vein (Figure 4), an effect indicative of increased Hh pathway activity. Consistent with this, the misexpression of either form in cells that do not normally receive the Hh signal results in the ectopic activation of the Hh target genes dpp and ptc (Figure 4). These effects indicate that the mutated forms of Ptc act as competitive inhibitors of the wild-type protein. Similar effects of a C-terminally truncated form of Ptc have recently been reported [21]. Conventional models of Hh signaling envisage Ptc to be

a ligand binding sub-unit of a Hh receptor that regulates the activity of a signaling subunit, the Smo protein, by inducing conformational changes in the latter (reviewed in [1]). The results of recent studies of Smo in Drosophila have, however, challenged this view and suggested instead that Ptc regulates Smo activity by promoting its posttranslational modification and/or decreasing its stability rather than by locking it into an inactive conformational state [14, 22, 23]. How and where such modifications occur is not known, but our finding that two of three antimorphic alleles of ptc are associated with lesions in the SSD indicates a critical role for this domain in the regulation of Smo. Given that other SSD-containing proteins, such as SCAP and NPC1, are known to mediate trafficking between intracellular compartments [24–26], it is tempting to speculate that Ptc may act in a similar manner by directing Smo to an intracellular compartment where it is targeted for modification/degradation. The antimorphic nature of the ptcS2 and ptc34 alleles could be explained if mutation of the SSD abolishes the putative trafficking activity of Ptc without affecting interaction with its cargo. The mutant forms of the protein would thus protect Smo from modification/degradation and leave it free to activate the downstream components of the pathway. By the same token, mutation of the C-terminal

612 Current Biology Vol 11 No 8

Figure 4

Ectopic expression of antimorphic forms of Ptc activates the Hedgehog pathway. Overexpression of the wild-type ptc cDNA in the anterior compartment of the wing (via the 71B GAL4 driver [28]) results in a significant reduction in the size of the anterior compartment and a concomitant loss of pattern elements, notably vein L3. By contrast, misexpression of the ptc13 and ptcS2 cDNAs via the same 71BGAL4 driver causes a slight increase in size of the anterior compartment and results in an expansion of L3 vein material. Ectopic expression of the wild-type ptc cDNA in the posterior compartment via the enGAL4

driver results in a reduction of the L3-L4 intervein region. Such a reduction is indicative of an attenuation of Hh signaling; an almost identical phenotype is displayed by wings in which the ptc13 or ptcS2 mutant cDNAs are driven under enGAL4 control. The 30A driver expresses GAL4 in a ring around the periphery of the presumptive wing blade (see [30]). The expression of either ptc13 or ptcS2 mutant cDNAs under 30A control results in the ectopic activation of dpplacZ and ptclacZ reporter genes.

tail ( ptc13) might also disrupt the hypothetical trafficking activity; alternatively, it could disrupt cargo interaction. Recent studies have failed to reveal an interaction between the Ptc C-terminal tail and Smo [21], and this failure favors the former possibility.

Immunohistochemistry

Materials and methods Induction and isolation of new ptc alleles Virgin males carrying an isogenic second chromosome marked with cn bw sp were treated with EMS according to standard procedures and mated to homozygous ptctuf ltd virgin females. F1 progeny were reared at 25⬚C and scored for wing abnormalities. Fourteen new alleles (designated ptc32 –ptc47) were identified in this way. The ptc44 and ptc45 alleles were isolated in a similar screen (M. Fietz and P. W. I., unpublished data) except in this case the mutagenized chromosome carried a GAL4 insertion at the ptc locus [27].

Generation of mAb 5E10 A 220 nucleotide fragment encoding the first 76 amino acids of Ptc and incorporating BamH1 and Sma1 sites at the 5⬘ and 3⬘ ends, respectively, was amplified by PCR from the ptc cDNA and cloned into the BamH1/ Sma1 sites of the expression vector pGEX3X. Induction of bacteria transformed with this construct, designated pGEXNterm, produced a band of the predicted size (34 kDa). The fusion protein was purified by preparative electrophoresis and used for the immunization of mice. After three boosts, animals producing positive sera were sacrificed, and their spleens were removed for fusion. We assayed supernatants from secreting clones by staining wild-type embryos.

Embryos were collected on apple juice/agar plates and fixed following standard protocols. To visualize Ptc and Hh proteins, we blocked fixed embryos with BSA and incubated them overnight at 4⬚C in mAb 5E10 at 1:100 dilution and in polyclonal anti-Hh antibody [4] at 1:250. After being washed, embryos were incubated with Cy3-conjugated anti-mouse IgG and Cy5-conjugated anti-rabbit IgG (Jackson) for 4 hr at room temperature. Following further extensive washing, the embryos were mounted in glycerol and observed with a Leica SP confocal microscope. Capture images were manipulated with Photoshop.

SSCP analysis A series of PCR primers (see Supplementary material) was used for the amplification of the ptc coding region from genomic DNA of flies heterozygous for each of 18 different ptc alleles. We prepared genomic DNA by homogenizing single flies in extraction buffer (10 mM Tris-HCl [pH 8.2], 1 mM EDTA, 25 mM NaCl, 200 ␮g/ml proteinase K) and incubating at 37⬚C for 30 min, followed by heat inactivation of the proteinase K (95⬚C/5 min). This DNA (1 ␮l) was used in a standard PCR reaction. The PCR product (10 ␮l) was heat denatured, cooled rapidly on ice, and run on an SSCP gel with 0.5⫻ MDE concentrate (Flowgen) according to manufacturer’s instructions. All samples were run overnight at 4⬚C, both in the presence and absence of 10% glycerol, which allowed detection of more polymorphisms. The separated DNA was detected by silver staining.

DNA Sequencing For sequencing, 1 ␮l of genomic DNA was used in a standard PCR reaction, and the PCR product was purified on Qiagen PCR purification columns (per the manufacturer’s instructions). One fifth to one eighth

Brief Communication 613

of the PCR reaction was used for sequencing with Big Dye terminator cycle sequencing kit (PE Applied Biosystems) The reaction products were analyzed with an ABI 377 sequencer.

10.

In vitro mutagenesis of ptc

11.

A full-length ptc cDNA in Bluescript was digested with BglII and XbaI to release a fragment encoding the C-terminal region of the protein, and this fragment was subcloned into Bluescript (ptcBglII-XbaI). A fragment of ptc containing the SSD-coding sequence was amplified with primers ptcSna5 and ptcBst3 and cloned into pCRIITOPO vector (Invitrogen) (ptcSna-Bst). A D584N point mutation was incorporated into a ptcSnaBstEII clone, and an E1185K point mutation was incorporated into a ptcBglII-XbaI clone with a QuikChange site-directed mutagenesis kit (Stratagene) per the manufacturer’s recommended protocol. The primers used were as follows: ptcSna5 5⬘-ACAGCTGCTACGTAAACAGTC GAGAATTGC -3⬘, ptcBst3 5⬘-TTGCCCTGGGTAACCGCATACATG CTGTAG-3⬘, D584N5 5⬘-GGCCATGATTTCGTTGAATCTACGGAG ACG-3⬘, D584N3 5⬘-CGTCTCCGTAGATTCAACGAAATCATGGCC-3⬘, E1185K5 5⬘-GACGACGATCACCAAGGAGCCGCAGTCGTG-3⬘, E1185K3 5⬘-CACGACTGCGGCTCCTTGGTGATCGTCGTC-3⬘. After sequence confirmation, the mutated fragments were digested with SnaBI and BstEII (ptcSna-Bst clone) or BglII and XbaI (ptcBglII-XbaI clone) and used to replace the corresponding wild-type fragments in the full-length cDNA. The modified cDNAs were cloned into the transformation vector pUAST [28].

12. 13.

14. 15. 16.

17. 18.

Supplementary material Two supplementary figures and a supplementary table are available with the electronic version of this article at http://images.cellpress.com/ supmat.supmatin.htm.

19.

Acknowledgements

20.

We are grateful to Monica Kibart for assistance with the mutant screens; Jeremy Quirk for establishing the SSCP protocol; A.Moir, Liz Jones, Bill Brown, and Debbie Sutton for help with the sequence analysis; David Strutt for assisting with embryo injections; and Andy Furley for helpful comments on the manuscript. The production and screening of hybridomas were performed by Dr. Wendie Norris at the Imperial Cancer Research Fund, London. mAb 5E10 supernatants were purified by Simon Smith (Sheffield Hybridomas). The confocal microscopy facility at the Centre for Developmental Genetics is funded by a grant from Yorkshire Cancer Research. This analysis was supported by a Wellcome Trust Program Grant to P. W. I.

References 1. Ingham PW: Transducing hedgehog: the story so far. EMBO J 1998, 17:3505-3511. 2. Chen Y, Struhl G: Dual roles for Patched in sequestering and transducing Hedgehog. Cell 1996, 87:553-563. 3. Capdevila J, Pariente F, Sampedro J, Alonso JL, Guerrero I: Subcellular localisation of the segment polarity protein Patched suggests an interaction with the Wingless reception complex in Drosophila embryos. Development 1994, 120:987-998. 4. Taylor AM, Nakano Y, Mohler J, Ingham PW: Contrasting distributions of Ppatched and Hedgehog proteins in the Drosophila embryo. Mech Dev 1993, 43:89-96. 5. Bellaiche Y, The I, Perrimon N: Tout-velu is a Drosophila homologue of the putative tumour suppressor EXT-1 and is needed for Hh diffusion. Nature 1998, 394:85-88. 6. Burke R, Nellen D, Bellotto M, Hafen E, Senti K, Dickson B, Basler K: Dispatched, a novel sterol-sensing domain protein dedicated to the release of cholesterol-modified Hedgehog from signaling cells. Cell 1999, 99:803-815. 7. Incardona J, Lee J, Robertson C, Enga K, Kapur R, Roelink H: Receptor-mediated endocytosis of soluble and membranetethered sonic hedgehog by patched-1. Proc Natl Acad Sci USA 2000, 97:12044-12049. 8. Capdevila J, Estrada MP, Sanchez-Herrero E, Guerrero I: The Drosophila segment polarity gene patched interacts with decapentaplegic in wing development. EMBO J 1994, 13:7182. 9. Marigo V, Davey R, Zuo Y, Cunningham J, Tabin C: Biochemical-

21.

22. 23.

24. 25.

26.

27. 28. 29. 30.

evidence that patched is the hedgehog receptor. Nature 1996, 384:176-179. Ellgaard L, Molinari M, Helenius A: Setting the standards: quality control in the secretory pathway. Science 1999, 286:18821888. Carstea E, Morris J, Coleman K, Loftus S, Zhang D, Cummings C, et al.: Niemann-Pick C1 Disease gene: homology to mediators of cholesterol homeostasis. Science 1997, 277:228231. Loftus S, Morris J, Carstea E, Gu J, Cummings C, Brown A, et al.: Murine model of Niemann-Pick C disease: mutation in a cholesterol homeostasis gene. Science 1997, 277:232-235. Beachy P, Cooper M, Young K, von Kessler D, Park W, Hall T, et al.: Multiple roles of cholesterol in Hedgehog protein biogenesis and signaling. Cold Spring Harb Symp Quant Biol 1997, 62:191-204. Denef N, Neubuser D, Perez L, Cohen S: Hedgehog induces opposite changes in turnover and subcellular localization of Patched and Smoothened. Cell 2000, 102:521-531. Hua X, Nohturfft A, Goldstein J, Brown M: Sterol resistance in CHO cells traced to point mutation in SREBP cleavageactivating protein. Cell 1996, 87:415-426. Watari H, Blanchette-Mackie E, Dwyer N, Watari M, Neufeld E, Patel S, Strauss JR: Mutations in the leucine zipper motif and sterol-sensing domain inactivate the Niemann-Pick C1 glycoprotein. J Biol Chem 1999, 274:21861-21866. Hidalgo A: Molecular cloning of patched and analysis of its role in intrasegmental patterning in D. Melanogaster. D. Phil Thesis, University of Oxford: 1989. Phillips R, Roberts I, Ingham PW, Whittle JRS: The Drosophila segment polarity gene patched is involved in a positionsignalling mechanism in imaginal discs. Development 1990, 110:105-114. Johnson RL, Grenier JK, Scott MP: Patched overexpression alters wing disc size and pattern-transcriptional and posttranscriptional effects on Hedgehog targets. Development 1995, 121:4161-4170. Alexandre C, Jacinto A, Ingham PW: Transcriptional activation of hedgehog target genes in Drosophila is mediated directly by the Cubitus interruptus protein, a member of the GLI family of zinc finger DNA-binding proteins. Genes Dev 1996, 10:2003-2013. Johnson R, Milenkovic L, Scott M: In vivo functions of the Patched protein: requirement of the C terminus for target gene inactivation but not Hedgehog sequestration. Mol Cell 2000, 6:467-478. Alcedo J, Zou Y, Noll M: Posttranscriptional regulation of smoothened is part of a self-correcting mechanism in the Hedgehog signaling system. Mol Cell 2000, 6:457-465. Ingham P, Nystedt S, Nakano Y, Brown W, Stark D, van den Heuvel M, et al.: Patched represses the Hedgehog signalling pathway by promoting modification of the Smoothened protein. Curr Biol 2000, 10:1315-1318. Liscum L, Klansek J: Niemann-Pick disease type C. Curr Opin Lipidol 1998, 9:131-135. Neufeld E, Wastney M, Patel S, Suresh S, Cooney A, Dwyer N, et al.: The Niemann-Pick C1 protein resides in a vesicular compartment linked to retrograde transport of multiple lysosomal cargo. J Biol Chem 1999, 274:9627-9635. Nohturfft A, DeBose-Boyd R, Scheek S, Goldstein J, Brown M: Sterols regulate cycling of SREBP cleavage-activating protein (SCAP) between endoplasmic reticulum and Golgi. Proc Natl Acad Sci USA 1999, 96:11235-11240. Speicher SA, Thomas U, Hinz U, Knust E: The Serrate locus of Drosophila and its role in morphogenesis of imaginal discs: control of cell proliferation. Development 1994, 120:535-544. Brand AH, Perrimon N: Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 1993, 118:401-415. Forbes AJ: Segment polarity genes in Drosophila development. D. Phil Thesis, 1992, University of Oxford. Ingham PW, Fietz MJ: Quantitative effects of hedgehog and decapentaplegic activity on the patterning of the Drosophila wing. Curr Biol 1995, 5:432-441.